Handbook of Youth Mentoring [2 ed.] 9781412980135, 9781412980142, 9781412996907, 2013000548, 9781483309811

Citation preview

Handbook of

Youth Mentoring Second Edition

The SAGE Program on Applied Developmental Science Consulting Editor Richard M. Lerner The field of Applied Developmental Science has advanced the use of cutting-edge developmental systems models of human development, fostered strength-based approaches to understanding and promoting positive development across the life span, and served as a frame for collaborations among researchers and ­practitioners, including policymakers, seeking to enhance the life chances of diverse young people, their families, and communities. The SAGE Program on Applied Developmental Science both integrates and extends this scholarship by publishing innovative and cutting-edge contributions.

Handbook of

Youth Mentoring Second Edition

Edited by

David L. DuBois University of Illinois at Chicago

Michael J. Karcher University of Texas at San Antonio

FOR INFORMATION:

Copyright  2014 by SAGE Publications, Inc.

SAGE Publications, Inc.

All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the publisher.

2455 Teller Road Thousand Oaks, California 91320 E-mail: [email protected] SAGE Publications Ltd. 1 Oliver’s Yard 55 City Road London EC1Y 1SP United Kingdom SAGE Publications India Pvt. Ltd. B 1/I 1 Mohan Cooperative Industrial Area Mathura Road, New Delhi 110 044 India SAGE Publications Asia-Pacific Pte. Ltd. 3 Church Street #10-04 Samsung Hub Singapore 049483

Printed in the United States of America Library of Congress Cataloging-in-Publication Data Handbook of youth mentoring / [edited by] David L. DuBois, University of Illinois, Michael J. Karcher, University of Texas at San Antonio. — Second edition. pages cm. — (The sage program on applied developmental science) Includes bibliographical references and index. ISBN 978-1-4129-8013-5 (hardcover : alk. paper)— ISBN 978-1-4129-8014-2 (pbk. : alk. paper)— ISBN 978-1-4129-9690-7 (xml-tei) 1. Youth—Counseling of—United States. 2. Mentoring—United States. 3. Youth development— United States. 4. Social work with youth—United States. I. DuBois, David L. II. Karcher, Michael J. HV1431.H3 2014 362.71′6—dc23   2013000548

Acquisitions Editor:  Kassie Graves

This book is printed on acid-free paper.

Editorial Assistant:  Elizabeth Luizzi Production Editor:  Stephanie Palermini Copy Editor:  Amy Marks Typesetter:  C&M Digitals (P) Ltd. Proofreader:  Susan Schon Indexer:  John Hulse Cover Designer:  Candice Harman Marketing Manager:  Erica DeLuca Permissions Editor:  Jennifer Barron

13 14 15 16 17 10 9 8 7 6 5 4 3 2 1

CONTENTS

Foreword ix Preface xi Acknowledgments xiii PART I. INTRODUCTION

1

  1. Youth Mentoring in Contemporary Perspective David L. DuBois and Michael J. Karcher

3

PART II. FRAMEWORKS AND FOUNDATIONS

15

  2. Mentoring and Positive Youth Development Richard M. Lerner, Christopher M. Napolitano, Michelle J. Boyd, Megan K. Mueller, and Kristina S. Callina

17

  3. Mentoring and Prevention Science Timothy A. Cavell and L. Christian Elledge

29

PART III. MENTORING RELATIONSHIPS

43

  4. Assessing Mentoring Relationships Michael J. Nakkula and John T. Harris

45

  5. Mentoring Activities and Interactions Michael J. Karcher and Keoki Hansen

63

  6. Goals in Mentoring Relationships Fabricio E. Balcazar and Christopher B. Keys

83

  7. Mentoring Relationships in Developmental Perspective Gil G. Noam, Tina Malti, and Michael J. Karcher

99

  8. A Biological Perspective on Positive Adult-Youth Relationships Amber L. Allison and Elizabeth A. Shirtcliff

117

  9. Social Networks and Mentoring Thomas E. Keller and Jennifer E. Blakeslee

129

PART IV. CULTURAL PERSPECTIVES

143

10. Race, Ethnicity, and Culture in Mentoring Relationships Bernadette Sánchez, Yarí Colón-Torres, Rachel Feuer, Katrina E. Roundfield, and Luciano Berardi

145

11. Gender in Mentoring Relationships Belle Liang, G. Anne Bogat, and Nicole Duffy

159

12. Social Class Nancy L. Deutsch, Edith C. Lawrence, and Angela K. Henneberger

175

13. International and Cross Cultural Aspects in Youth Mentoring Limor Goldner and Miri Scharf

189

PART V. PROGRAMS AND CONTEXTS

201

14. School-Based Mentoring Carla Herrera and Michael J. Karcher

203

15. After-School Programs Megan A. Mekinda and Barton J. Hirsch

221

16. Cross-Age Peer Mentoring Michael J. Karcher

233

17. Electronic Mentoring and Media Carmit-Noa Shpigelman

259

18. Group Mentoring Gabriel P. Kuperminc and Jessica D. Thomason

273

19. Work and Service-Learning Stephen F. Hamilton and Mary Agnes Hamilton

291

PART VI. SPECIAL POPULATIONS

301

20. Academically At-Risk Students Simon Larose and George M. Tarabulsy

303

21. Juvenile Offenders Wing Yi Chan and David B. Henry

315

22. Youth With Mental Health Needs David C. R. Kerr and Cheryl A. King

325

23. Youth in Foster Care Preston A. Britner, Kellie G. Randall, and Kym R. Ahrens

341

24. Immigrant and Refugee Youth Dina Birman and Lyn Morland

355

25. Children of Incarcerated Parents J. Mark Eddy, Jennifer Cearley, Joseph Bergen, and Jenny Stern-Carusone

369

PART VII. PRACTICE AND PROGRAMMATIC CONSIDERATIONS

383

26. Program Funding Susan G. Weinberger

385

27. Mentor Recruitment and Retention Arthur A. Stukas, E. Gil Clary, and Mark Snyder

397

28. Mentor Screening and Youth Protection Sarah E. Kremer and Becky Cooper

411

29. Mentor and Youth Matching Julia Pryce, Michael S. Kelly, and Sarah R. Guidone

427

30. Mentor Training Janis B. Kupersmidt and Jean E. Rhodes

439

31. Family Involvement Andrea Taylor

457

32. Termination and Closure of Mentoring Relationships Renée Spencer and Antoinette Basualdo-Delmonico

469

33. Program Evaluation David L. DuBois

481

34. Economic Evaluation of Mentoring Programs E. Michael Foster

499

35. Ethical Issues in Youth Mentoring Jean E. Rhodes, Belle Liang, and Renée Spencer

511

PART VIII. CONCLUSIONS

523

36. Youth Mentoring in the 21st Century: Progress and Prospects David L. DuBois and Michael J. Karcher

525

Author Index

535

Subject Index

553

About the Editors

569

About the Contributors

571

FOREWORD

When you talk about mentoring youth as much as I do, it does not take long to pick up patterns that reinforce the promise as well as the perils of the wide range of endeavors that in today’s times are mounted under the umbrella of this concept. Reactions frequently result from how engagement in mentoring is presented. When viewed as organic and magical, it is extremely popular as a panacea for so much of what faces our young people. On the flip side, when described with rigor and as requiring intentionality and investment of money and time, it loses some of its allure as a societal solution. Some of this latter reaction stems from our base desire to find the shortest and easiest distance between two points in all facets of our lives. But there is likely another ingredient at work when it comes to mentoring. We gravitate toward a beneficial behavior that is as old as Greek mythology. Yet the tradition of structured mentoring for youth, with the exception of Big Brothers Big Sisters’ century-old existence, is relatively young. It is only 22 years ago that basic standards for quality were first created through the leadership of MENTOR: The National Mentoring Partnership and the United Way of America in the form of the Elements of Effective Practice for Mentoring, which is now in its 3rd edition. When David DuBois and Michael Karcher released the first edition of the Handbook of Youth Mentoring, they provided a powerful collection of accounts that not only helped to crystalize, but also set the stage for advancement of this young field of scholarship and practice. In the wake of the accelerating pace of new theory and research about the mentoring of young people over the past several years, even a straightforward updating of the field’s core understandings would have been a challenging undertaking. As one point of proof of this acceleration, in the past two years, the first two academic centers for youth mentoring have been launched by authors of Handbook chapters, Dr. Jean Rhodes at University of Massachusetts-Boston and Dr. Tom Keller at Portland State University.

In forging this second edition, it is clear DuBois and Karcher had loftier goals in mind. Not the least of these is reflected in the careful attention that is paid in this new volume to the rapid expansion in mentoring as both a language and a tool for application in efforts to support young people on pathways to well-being and success. As editors, they have assembled an extraordinary cast of experts to look at mentoring through many lenses. The result is an invaluable resource for those immersed in navigating the practical challenges to achieving desired outcomes for young people through mentoring. At the same time, this text serves as a stimulus for reflection on questions that remain central challenges to the field’s continued development. Karcher and DuBois seek to ground the proven promise of mentoring in the known research while also addressing the reality of emerging intersections with topics as diverse as technology, funding, biology, culture, social capital, and emerging program structures. In drawing on experts from a variety of disciplines, the editors acknowledge the holistic reality of our young people’s development, and mentoring’s place in this overall context. I am reminded constantly that cultivating and acting on this understanding of how mentoring fits within the “big picture” of young people’s lives is at the heart of what it has to offer our society. For example, I spoke recently with a young man in a public housing development where another young man had been shot and killed while playing basketball. I was aware that a mentoring program had been launched in the development not long ago, so I asked him simply how he was doing with all that was going on around him. The essence of what he said was that a lot of new people he did not know were paying attention to him and other youth in his neighborhood because of what had occurred. But, the person he could talk with was his mentor, who showed up following the tragedy just like he had for the past 7 months—someone who had been and ix

x   HANDBOOK OF YOUTH MENTORING would continue to be there for him. Mentoring professionals had set the stage for this young man to have someone to turn to, someone with whom he was already in a trusted relationship before a time of crisis, and they did so based on the learnings in this text. Our focus at MENTOR and with our network of local Mentoring Partnerships and many cross-sector partners, is to galvanize a movement for quality youth mentoring. We constantly walk that tightrope between scale and quality because what we know is that mentoring helps drive greater public health, public safety, and the very things that lead to opportunity—school attendance, achievement, graduation, workforce readiness, and connection to community—when it is done with rigor and clarity of purpose. This is how to turn mentoring from a “nice to have” into a “need to have.” It must be grounded in scholarly evidence that informs practice. We must be champions for that necessary tightrope walk to drive the quality and quantity of professionally supported mentoring relationships. It is important and possible in small and large programs, with volunteers and paid staff, in rural and urban communities, and with all ages and theories of change. The second edition of the Handbook arms us with a treasure trove of information to guide us in this work and in the innovations in mentoring that lie ahead.

As I alluded to earlier, in pursuing our mission at MENTOR, we are determined to foster efficient and effective linkages with theory and research. But we know from experience that such linkages will be fruitful only to the extent that they also reflect a grounded understanding of the demands that practitioners routinely face in their day-to-day work. The first edition of the Handbook has proved to be an invaluable resource for informing and guiding the building of strong bridges between scholarship and practice. I salute the editors and contributors to this new edition for redoubling their efforts in this area. In doing so, they clearly appreciate what we have learned over the years at MENTOR and in our Mentoring Partnerships. Mentoring professionals gain rich wisdom from their practical experience while at the same time seeking accessible, implementable, and informed guidance from reliable sources in their quest for improved outcomes for young people. As we all know, when we give young people the best we’ve got, they stand up. They show up. And we are all lifted. The Handbook contributes greatly to our collective efforts to ensure that the current generation of young people is a mentored generation, ultimately leading to a better supported, more productive, and unified society. David Shapiro Chief Executive Officer, MENTOR

PREFACE

As with the first edition of the Handbook of Youth Mentoring, this volume is the result of several years of planning, writing, and editing. We embarked on this journey pleased to have had the initial edition receive considerable praise and recognition, including a national award (Social Policy Award for Best Edited Book from the Society for Research on Adolescence). Yet, we were equally aware of several opportunities for improvement upon the original volume. Chief among these was a need that we perceived to increase the relevance and utility of the Handbook for those involved in mentoring practice. To this end, we added chapters on several topics that are of direct concern in the day-to-day work of practitioners (e.g., mentor training, mentor-youth matching, mentor-youth activities, ethical issues) but that had not benefited from dedicated attention in the first edition. We made a concerted effort as editors, furthermore, to work with authors to better highlight the applied implications of the theory and research that is reviewed in each chapter. We have done so, in part, by having authors include key recommendations and guiding questions for practice summarized in tabular format within chapters for easy reference. Finally, and perhaps most significantly, we also enlisted experienced practitioners as anonymous reviewers for nearly all chapters. Their insightful observations and constructive suggestions proved invaluable for ensuring that potential implications for practice were brought into appropriate relief. The names of these reviewers can be found in this volume under Acknowledgments. Rigorous syntheses of theory and research on topics of importance to the field of youth mentoring were a distinguishing feature of the first edition of the Handbook. In preparing the current volume, we worked to retain the same high scholarly standard. As before, we started by identifying and recruiting a team of contributors whose expertise and qualifications ensured that accounts of the current state of knowledge in different topic areas would be uniformly rigorous, but also constructive, informative,

and enlightening. Many of the contributors’ names will be familiar to readers of the first edition. In most instances, these authors were asked to revisit the same topics they had addressed previously, updating and expanding on their earlier contributions. In several instances, however, we successfully enlisted authors from the first edition in tackling entirely new topics. Their contributions delve insightfully into a range of topics that have been the source of growing attention within the field (e.g., the interface of mentoring with the larger field of prevention science, issues relating to closure and termination in mentoring relationships for youth). Our goal of keeping the contents of the Handbook as current and timely as possible could not have been realized, however, without the additional contributions of numerous leading experts and scholars who are new to this edition. These authors led the way in addressing promising, but to date relatively unexplored theoretical perspectives on mentoring relationships (e.g., biological, social class), opportunities and challenges associated with mentoring specific populations of youth whose unique needs and circumstances only recently have begun to be fully appreciated (e.g., immigrant and refugee youth, youth with mental health needs), and the latest developments in our knowledge regarding emerging mentoring program structures (e.g., group, electronic). Returning or new, our contributors are owed a tremendous debt of gratitude. Not the least of their travails was the need to persevere through unexpected delays in receiving our editorial feedback, which all managed to do not only with great forbearance, but also with unwavering collegial amiability. No less appreciated is the manner in which authors clearly brought their “A game” to their contributions. We are delighted, in particular, to have such an abundance of truly novel and potentially groundbreaking ideas and recommendations woven into this edition of the Handbook. Readers will likely find many of these viewpoints challenging not only to the status quo—whether in theory, research, or practice—but xi

xii   HANDBOOK OF YOUTH MENTORING also to their own assumptions (as we did as editors). Yet, we hope that readers will agree with our assessment that the types of arguments presented, even when provocative, are essential to the growth and advancement of the field. Finally, we would be remiss not to thank our colleagues for running with (rather than away from!) all aspects—substantive and stylistic—of our editorial feedback. We hope that in at least some small measure our input and vision helped to strengthen individual contributions and that in the process we were successful in forging an overall volume that is as cohesive and “user friendly” as it is comprehensive. Other acknowledgments are due as well. We are indebted to Kassie Graves, senior acquisitions editor at Sage Publications, for her highly capable and remarkably patient and understanding stewardship of this project from inception to completion, and to Liz Luizzi and Laureen Gleason, for shepherding us so smoothly and painlessly through the manuscript assembly and production process. Further thanks are due to the academic programs and institutions with

which we continue to have the good fortune to be affiliated, the School of Public Health at the University of Illinois at Chicago and the College of Education and Human Development at the University of Texas at San Antonio, and to the William T. Grant Foundation, Office of Juvenile Justice and Delinquency Prevention (U.S. Department of Justice), Office of Minority Health (U.S. Department of Health and Human Services), and Institute of Education Sciences (U.S. Department of Education) for their generous support of our programs of research on youth mentoring. Finally, we reserve our most heartfelt appreciation for the support that we have received from our families, each of which have grown considerably since we undertook the first edition of the Handbook. To our spouses, Natalie and Sara, we offer special thanks for so generously supporting us in what turned out again to be a far more ambitious undertaking than either of us anticipated. David L. DuBois Michael J. Karcher

ACKNOWLEDGMENTS

We gratefully acknowledge the contributions of the following individuals, who each served as reviewers for one or more chapters in this volume: Ann Adalist-Estrin

Mark Gesner

Graig Meyer

Armen Babajanian

Gerald Goodman

Rachel Pillar

Katherine Balsley

Celeste Janssen

Brian P. Sales

Denise Barkhurst

Jamie Johnson

Karen Shaver

Christy Beighe-Byrne

Jolynn Kenney

Roman Sklotskiy

Kelly Belmonte

Sandra Louk LaFleur

Judith Stavisky

Angela Benke

Karen Lamothe

Jenny Stern-Carusone

Susan Climie

Tracey Lewis

Sabrina Vegnone

Hector Cortez

Tracy Luca-Huger

Judy Vredenburgh

Rebecca Farnell

Jane Marion

Marc Wheeler

Janet Forbush

Dave Marshall

Tonya Wiley

Sherry Garrett

Kris Marshall

Nicole Yohalem

xiii

With much love, to Natalie, Pookers, Checkers, and Gwyn-gwyn. You show me each day that I’m never too old to grow or to play. – D. L. D. To my wife, Sara, and my children, Reed, Jack, and Caroline, for their patience, encouragement, and support. – M. J. K.

PART I Introduction

1  YOUTH MENTORING IN CONTEMPORARY PERSPECTIVE David L. DuBois and Michael J. Karcher

Introduction As many in the mentoring field had done before us, we began the first edition of the Handbook of Youth Mentoring with an historical reference to the origins of the term mentor in Homer’s Odyssey. It is interesting, however, to contemplate that the concept of mentoring may have been an important factor in the earliest beginnings of the human species. There is evidence, for example, that early evolutionary conditions favored the development of innate, prosocial motivations to be helpful not only to one’s genetic relatives, but also to other group members (Bowles, 2006). Allo-parenting—in which individuals who are not the biological parents assume some of the caregiving responsibilities for a child—remains a salient and beneficial practice in less modernized hunter-gatherer societies today (Diamond, 2012). Interestingly, it also appears that the ancestors of modern humans may have bested Neandertals in the quest for survival in part by virtue of their longer life span, which allowed acquired knowledge to be passed on more easily from one generation to the next (Wong, 2009). Such considerations suggest that the inclination to mentor may be part and parcel of our human DNA and that the payoffs of doing so are far from unique to contemporary times or cultures. The environment over the past couple of decades is distinctive, however, in the remarkably high levels of support and interest that initiatives directed toward mentoring of youth have generated both in the United States and abroad (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). Developments within the practice and policy realms since the first edition of the Handbook give few signs that this

trend is abating (see Table 1.1). Fueling these efforts appears to be a continued belief that supportive relationships between young people and nonparental adults, whether established via programs or through more informal connections, represent assets vital for positive youth development (Scales, 2003). Although still lagging behind practice and policy, the research on youth mentoring has in many respects come of age in recent years as an established area of inquiry within several different fields such as psychology, education, human development, social work, and public health. Also important, several developments have facilitated the integration and sharing of research findings across disciplines and areas of study, helping to solidify the field’s knowledge base and spur innovative lines of investigation. Complementing the progress on these fronts, increased channels are now available to disseminate and translate findings so that they are both accessible and practically useful to those working to support mentoring of youth within diverse settings such as schools, child welfare, juvenile justice, the workplace, and afterschool programs. In this chapter, we consider recent developments in practice and research in greater detail. With these developments as a backdrop, we then provide an overview of our goals for this new edition of the Handbook. In doing so, we highlight how these goals informed our decisions about what topics were covered in the Handbook as well as the types of information that were most important for the chapters to include. As we did in the first edition, however, we first briefly consider how the mentoring of youth may best be conceptualized and defined in light of the field’s current status. 3

4  Introduction

Multilevel Conceptualization of Youth Mentoring Most often, efforts to define or conceptualize youth mentoring have focused on characteristics of the mentor and his or her relationship with the young person involved. Typically, for example, the mentor must be a nonparental adult who is not acting in a formal professional capacity (e.g., therapist) or, if younger, must be an older youth rather than a sameage peer. The mentoring relationship is usually described as one in which there is an emotional connection and in which the mentor offers guidance and others forms of support to the young person. Either implicitly or explicitly, most accounts have pointed toward a relationship that is sustained over some significant period of time rather than being merely transitory. An overview of these types of definitions can be found in our introductory chapter to the first edition of the Handbook (DuBois & Karcher, 2005). As the field continues to evolve, however, we see value in working toward a broader and more differentiated view of youth mentoring that encompasses multiple levels of conceptualization. In Table 1.1, we thus offer a conceptualization of youth mentoring that includes several distinct and progressively broader levels: activity, relationship, intervention, policy, and societal. At the level of activity, the focus is on the actors who may be engaged in mentoring interactions with a young person and the nature of those interactions. Our

conceptualization of mentoring activity emphasizes largely the same considerations as those that have factored significantly in prior definitions of mentoring (e.g., provision of guidance). Unlike most of these earlier views, however, mentoring activity is not assumed to necessarily involve a mentoring relationship. Thus, for example, a neighbor with whom a youth does not have a significant personal connection could nonetheless engage in mentoring with that young person by offering advice or encouragement. Clearly, though, even within established mentoring relationships, it is important to take into account what the mentor and youth actually do and talk about together as well as more subtle dynamics of their interactions such as how much voice the youth has in decision making within the relationship (Keller, 2005). In line with these considerations, this edition of the Handbook includes a chapter devoted exclusively to mentoring activities. We assume, furthermore, that mentoring activities are geared intentionally toward supporting one or more areas of the youth’s development. This is not meant to exclude modes of influence that are indirect (e.g., an individual making an effort to serve as a positive role model for a youth). Nor does it leave out those that are largely ineffectual or even unwittingly result in harm to the youth, such as could be the case with a mentor whose attempts at constructive guidance are experienced by the young person as invalidating or demeaning. A mentoring relationship, in turn, would be one in which mentoring activities occur within the

Table 1.1   Multilevel Conceptualization of Youth Mentoring Level

Conceptualization

Activity

Social interactions in which nonparental adults or older peers without advanced professional training provide guidance and other forms of support to youth that is intended to benefit one or more areas of their development

Relationship

Ongoing significant interpersonal ties involving youth in which mentoring activity takes place regularly

Intervention

Intentional efforts of agencies, organizations, or programs to foster mentoring activity and/or relationships for specific groups of youth or those living within particular communities

Policy

Governmental and other institutional initiatives to support mentoring activity, relationships, and/or interventions for youth

Societal

Public perceptions of youth mentoring and willingness and ability to engage in or otherwise support mentoring activities, relationships, interventions, and policies for young people

Youth Mentoring in Contemporary Perspective    5 context of significant interpersonal ties between youth and mentor(s). Defining attributes at this level would include not only a relationship that is sustained over some noteworthy period of time, but also one in which mentoring activity takes place on a regular basis. Thus, whereas a youth could have an enduring and emotionally salient tie with an important adult in his or her life, such as a grandparent, to constitute a mentoring relationship within this framework there also would need to be an ongoing pattern of support and guidance. Mentoring relationships are important to consider, in part, because they may offer effective avenues of support for youth that are not easily duplicated through more isolated interactions with a range of different persons. This may be especially true when relationships are sustained over significant portions of a young person’s development, thus providing a consistent and secure base of support and fostering levels of trust and attunement that may only be able to emerge over the course of significant time (Rhodes, 2005). As in the previous edition of the Handbook, an entire section is devoted to chapters that focus on mentoring relationships. This part of the Handbook begins with an in-depth examination of the major features of mentoring relationships for young persons and a critical review of currently available assessment tools. Other chapters consider how mentoring relationships and their impacts may be shaped by the youth’s developmental stage as well as by the intersecting characteristics of the youth and mentor along dimensions of gender, race/ ethnicity, and social class. At the next level, that of an intervention, the focus is on organizational and programmatic efforts to support mentoring for youth within particular groups or communities. These initiatives would include the wide range of programs established to cultivate mentoring relationships for youth, most often with adult or older peer volunteers (e.g., Big Brothers Big Sisters). Although reliable estimates are not available, several thousand such programs exist in the United States and they are growing in number internationally, as well. These programs occupy a central role in any account of the current landscape of the field of youth mentoring. Based on polling data collected in 2006, for example, it was estimated that formal mentoring programs were serving an estimated 2.5 million young people in the United States alone (MENTOR, 2006). At the same time, such estimates overlook the even larger numbers of youth who routinely participate in afterschool programs and other youth organizations in which mentoring is an important part of their mission and activities. The Boys & Girls Clubs of

America, for example, includes nearly 4,000 clubs serving an estimated 4.1 million young people (Boys & Girls Clubs of America, n.d.). Youth can and do develop ties with staff and volunteers in such settings that clearly would qualify as mentoring relationships within our proposed framework (Hirsch, Deutsch, & DuBois, 2011). Yet, beyond these types of deeper connections, youth-serving programs and organizations typically are structured with the goal of ensuring that youth have ample opportunity to receive guidance and support from adults (and, in some instances, also older peers) in areas relevant to program goals (e.g., arts, sports, academic enrichment). In this way, they have the potential to greatly expand the scope of a young person’s exposure to what would be referred to as mentoring activity in our proposed framework. Similar aims are routinely reflected in other initiatives that are aimed at benefiting youth such as those focused on educational reform, neighborhood improvement, or workforce preparedness. When the aim is to reach a “tipping point” of impact within an entire group or community, it may be especially important to actively cultivate these more informal avenues for young people to receive mentoring. With such considerations in mind, at this level of the framework, interventions are defined broadly. Thus, rather than being limited only to programs that are geared toward establishing formal or sustained ties with specific persons (i.e., mentoring relationships), interventions also include the wide range of efforts that may be undertaken to make guidance and support available to youth as part of their participation in different types of activities and settings (i.e., mentoring activity). Several chapters in the Handbook, such as those that examine opportunities for mentoring in the contexts of schools, after-school programs, and the workplace and service-learning, are aligned with this perspective. At the policy level, the focus is on the role of governmental and other types of institutions (e.g., philanthropic organizations) in offering support for mentoring activity, relationships, and/or interventions for youth. Funding is clearly one significant consideration in this regard. Indeed, as we discuss later in this chapter, arguably no developments over the past several years had more impact than those affecting the levels of federal funding available to support implementation of youth mentoring programs. It is important, however, to also consider how initiatives that originate within a governmental or other institutional context may shape the types of approaches to youth mentoring most likely to be put into practice. Illustratively, the Student Mentoring Program initiative of the U.S. Department of

6  Introduction Education undoubtedly contributed to the sharp rise that occurred over a similar period of time in the prevalence of mentoring programs based in schools (Wheeler, Keller, & DuBois, 2010). Ideally, from our perspective, such efforts are both informed by theory and well-grounded in findings from available research. In reality, however, we know that this is not always the case. The importance of fostering stronger linkages among theory, research, and practice was a key theme in the first edition of the Handbook and continues to be reflected in this edition. From this perspective, initiatives that are geared toward ensuring adherence to empirically supported program practices, facilitating the development and ongoing refinement of programs within an evidence-driven approach, and building stronger collaborations between practitioners and researchers all may be critically important to implement at the policy level (DuBois et al., 2011). At the societal level, one key concern is with public perceptions of youth mentoring. In the United States, the idea of mentoring young people appears to have quite broad appeal (Scales, 2003). It may be, in part, for this reason that federally funded youth mentoring initiatives have been able to originate from different points on the political spectrum and benefit from bipartisan support (Walker, 2005). A distinctive aspect of mentoring, however, is that it depends on ordinary persons (i.e., those without specialized or advanced training or formal role responsibilities focused on work with youth) giving freely of their time to support a young person’s development and having the ability to do so in ways that serve this goal. In a 2005 national poll, slightly less than one-third of adults reported having mentored a youth in some capacity during the past year (MENTOR, 2006). Equally noteworthy, however, is that of those who had not done so, a similar percentage (29%) reported that they would seriously consider mentoring. Translating these attitudes into action remains a key issue for the field, especially in view of arguments that a “mentoring gap” (MENTOR, 2006) exists in which the number of youth involved in formal, one-to-one mentoring relationships in the United States is only about 15% of the total of young people who could benefit from such mentoring. With relevance to this concern, this edition of the Handbook again includes a chapter focusing on the recruitment and retention of mentors within programs. The need to ensure that volunteers are suitable candidates for work with youth and adequately prepared for this undertaking is addressed as well in chapters devoted to issues involved with screening and training of mentors.

The foregoing discussion speaks to the range of interrelated and often challenging issues that are likely to be confronted in efforts to improve either the efficacy of youth mentoring or its scale of availability. Many, if not all, of the questions that might need to be asked in mounting such initiatives have yet to be fully answered by research. Likewise, optimal strategies for tackling more practical aspects of quality assurance and access have by no means been fully established by those working in applied roles. Yet, in the 8 years that have elapsed since the first edition of the Handbook was published, significant progress has been made on several important fronts that pertain to both research and practice. We turn now to an overview of these developments.

Recent Developments in Youth Mentoring Table 1.2 summarizes key developments in youth mentoring since the publication of the first edition of the Handbook. Because any such list will inevitably leave out important events and information, we encourage readers to view the developments as representative of those occurring during the past several years, rather than as an exhaustive list. Furthermore, although we organized developments into several different categories (research, practice and policy, connecting research with practice, funding, and international), several entries clearly could fit well under more than one category. Research Noteworthy developments in the area of research include several large-scale randomized controlled trial evaluations of school-based mentoring programs (Bernstein, Rappaport, Olsho, Hunt, & Levin, 2009; Herrera, Grossman, Kauh, Feldman, & McMaken, 2007; Karcher, 2008; for a review of these studies and their major findings, see Wheeler et al., 2010). The primary reports of these trials have already had a significant impact on practice and policy. Informed by findings from a trial that focused on the school-based mentoring program of Big Brothers Big Sisters of America (BBBSA) as implemented by several of its affiliate agencies (see Herrera et al., 2007, and Herrera, Kauh, Cooney, Grossman, & McMaken, 2008), the organization recommended that its affiliates consider adopting a range of practices aimed at enhancing the effectiveness of the program in areas such as volunteer training, mentoring relationship duration, and support for relationships during the summer months. These

Youth Mentoring in Contemporary Perspective    7 Table 1.2   Recent Developments in Youth Mentoring Research 2005

Special issue of The Journal of Primary Prevention: Mentoring With Children and Youth

2006

Special issue of Journal of Community Psychology: Youth Mentoring: Bridging Science With Practice

2007

Big Brothers Big Sisters School-Based Mentoring Impact Study

2007

The Blackwell Handbook of Mentoring: A Multiple Perspectives Approach

2007

Understanding and Facilitating the Youth Mentoring Movement (Social Policy Report of the Society for Research on Child Development)

2008

Study of Mentoring in the Learning Environment (SMILE)

2009

Impact evaluation of the U.S. Department of Education’s Student Mentoring Program

2010

Establishment of Center for Interdisciplinary Mentoring Research (Portland State University)

2011

Special issue of New Directions for Youth Development: Play, Talk, Learn: Promising Practices in Youth Mentoring

2011

Meta-analysis of evaluations of youth mentoring program effectiveness, 1999–2010

Practice and Policy 2006

Formation of Federal Mentoring Council

2006

National Agenda for Action: How to Close America’s Mentoring Gap report from MENTOR: The National Mentoring Partnership

2007

Mentoring Children of Prisoners Voucher Demonstration Program (“Caregivers Choice”)

2009

Elements of Effective Practice for Mentoring, 3rd edition

2010

Discontinuation of the SafetyNET program for criminal background checks

2011

First annual National Mentoring Summit

Connecting Research and Practice 2006

Formation of Mentoring Research and Policy Council of MENTOR: The National Mentoring Partnership and Research Advisory Council of Big Brothers Big Sisters of America

2006

Initiation of listserv: Youth Mentoring Research and Practice

2007

Research in Action Series (MENTOR: The National Mentoring Partnership)

2007

First annual Summer Institute on Youth Mentoring (Portland State University)

2011

Establishment of Center for Evidence-Based Mentoring (University of Massachusetts Boston)

(Continued)

8  Introduction Table 1.2  (Continued) Funding 2008

Office of Juvenile Justice and Delinquency Prevention (OJJDP): National Mentoring Programs

2008 and subsequent years

OJJDP: Funding directed toward mentoring of a variety of specific youth populations, including tribal and Latino youth, ex-juvenile offenders, youth in foster care, and child victims of commercial sexual exploitation

2009

Elimination of funding for U.S. Department of Education’s Student Mentoring Program

2009

OJJDP: Mentoring Initiative for System Involved Youth and Strategic Enhancement to Mentoring Programs

2010

OJJDP: Initiation of Mentoring Best Practices Research funding program

2011

Elimination of funding for U.S. Department of Health and Human Services’ Mentoring Children of Prisoners Program

2012

OJJDP: Mentoring Enhancement Demonstration Program

International 2005

Formation of Australian Youth Mentoring Network and New Zealand Youth Mentoring Network

2007

Initiation of longitudinal study of the Big Brothers Big Sisters of Canada CommunityBased Mentoring Program

2007

Big Brothers Big Sisters International establishes new global standards and qualifies 12 country programs as affiliates

2010

Evaluation of the impact of the Big Brothers Big Sisters Mentoring Program in Ireland

Note: The developments listed are those that have taken place since publication of the first edition of the Handbook of Youth Mentoring in 2005. recommendations were incorporated into a revised school-based program model that was piloted and evaluated in 23 BBBSA agencies (Big Brothers Big Sisters of America, 2009). Efforts to further mine the data from these studies have already yielded important insights regarding issues such as the characteristics of youth who are most likely to benefit from mentoring program participation in the school context (Schwartz, Rhodes, Chan, & Herrera, 2011), the interaction of youth and mentor characteristics in predicting programmatic impact (Karcher, Davidson, Rhodes, & Herrera, 2010), and the advisability of re-matching youth with new mentors when a relationship ends part way through the school year (Grossman, Chan, Schwartz, & Rhodes, 2012). Further noteworthy research has shed important light on the effectiveness of mentoring programs for

specific populations of youth, such as youth with mental health concerns (e.g., King et al., 2009) and those with incarcerated parents (ICF International, 2011), as well as on the contributions that natural (non-program-assigned) mentoring relationships of youth—with teachers (Ahrens, DuBois, Lozano, & Richardson, 2010; Erickson, McDonald & Elder, 2009; Gastic & Johnson, 2009), after-school program staff (Hirsch et al., 2011), work supervisors (e.g., Vazsonyi & Snider, 2008), and nonparental adults in general (e.g., Hurd & Zimmerman, 2010)—can make to their development. Still other investigations have begun to scratch the surface of topics that previously received little or no exploration, such as internetbased formats for mentoring (Rhodes, Spencer, Saito, & Sipe, 2006) and the potential interplay of genes and environment in shaping youth outcomes in

Youth Mentoring in Contemporary Perspective    9 response to mentoring (Shanahan, Erickson, Vaisey, & Smolen, 2008). Coincident with growth in the field’s knowledge base, a significant network of scholars conducting research on youth mentoring has emerged (Blakeslee & Keller, in press). Such networks, sometimes referred to as “invisible colleges” (Cooper, 2010), are important to the development of any field of study, both for the fruitful collaborations they are likely to spawn and for the ways in which they can be expected to facilitate sharing of methods and findings across researchers. Other similarly valuable developments have included the publication of several special issues of academic journals that feature research on youth mentoring (Britner, 2005; DuBois & Rhodes, 2006; Karcher & Nakkula, 2010) and the launch of a new center dedicated to fostering multidisciplinary research on mentoring (Center for Interdisciplinary Mentoring Research, 2012). Also noteworthy are several published syntheses of research on youth mentoring. These include a recent comprehensive meta-analysis of over 70 evaluations of youth mentoring program effectiveness published over the preceding decade (1999– 2010; DuBois et al., 2011). Several additional meta-analyses or systematic reviews have focused on the effectiveness of mentoring programs for specific youth populations (Jolliffe & Farrington, 2007), program models (Wheeler et al., 2010; Wood & Mayo-Wilson, 2012), and outcomes (Thomas, Lorenzetti, & Spragins, 2011; Tolan, Henry, Schoeny, & Bass, 2008) and on the outcomes associated with mentoring relationships for youth in comparison to those experienced by older age groups such as college students and adults in the workplace (Eby, Allen, Evans, Ng, & DuBois, 2008). One useful purpose of these syntheses has been to consolidate research in different areas of the literature. Equally, if not more important, however, is the manner in which these reviews have been able to reveal trends across studies that would have been difficult, if not impossible, to glean from the results of individual investigations. The DuBois et al. (2011) meta-analysis, for example, found that effectiveness was significantly greater when programs emphasized similarity of interests in the pairing of mentors and youth and when programs were structured to support mentors in assuming teaching or advocacy roles with youth. Also prominent in the conclusions of all reviews have been identified gaps in the knowledge base needed to inform youth mentoring initiatives. These gaps encompass critical questions such as the types of mentor interactions with youth that are most likely to prove beneficial, the effectiveness of newer

program models (e.g., e-mentoring), the extent to which mentoring can be helpful in addressing particular outcomes (e.g., obesity prevention), and the long-term effects of mentoring relationships on youth as they transition into adulthood (e.g., high school graduation and employment). Practice and Policy Notable developments also have occurred in the areas of practice and policy. Among the most noteworthy in the governmental realm is the formation of the Federal Mentoring Council. This group, established in 2006, is comprised of representatives from a range of federal government agencies and tasked with ensuring effective and efficient coordination of federal investments in mentoring activities (Federal Mentoring Council, 2012). A troubling development at the federal level was the discontinuation of the SafetyNET program, a low-cost criminal background check pilot program established in 2003 for use by mentoring organizations across the country as part of the PROTECT Act (MENTOR, 2010), with a comparable system yet to replace it. At the same time, MENTOR: The National Mentoring Partnership, the most prominent advocacy organization within the field, issued a report detailing an action agenda for addressing the previously discussed mentoring gap in the United States (MENTOR, 2006). This organization also issued the third edition of its Elements of Effective Practice for Mentoring (MENTOR, 2009), which for the first time included specific operational standards in areas such as mentor screening and relationship monitoring and support. In 2011, MENTOR, in partnership with several other private and governmental organizations, including BBBSA and the Corporation for National and Community Service, organized and coordinated an inaugural National Mentoring Summit (MENTOR, n.d.a.). This event, which brought together major youth mentoring organizations, along with government, civic, research, and corporate leaders, has been continued on an annual basis and fast become a major gathering point for the field. Connecting Research and Practice Further significant initiatives have focused on connecting research and practice in the area of youth mentoring. These include establishment of advisory committees that are comprised in significant part or entirely of researchers to inform the work of national mentoring programs and advocacy organizations (i.e., BBBSA and MENTOR); a new university-based

10  Introduction center dedicated to facilitating evidence-based practice and policy in the area of youth mentoring (Center for Evidence-Based Mentoring, 2011); and the launch of publications focused on translating research findings into practice, such as the Research in Action Series of MENTOR (n.d.b.). Also noteworthy are several initiatives geared toward facilitation of direct communication and exchange of ideas between those in research and applied roles within the field. These include the Summer Institute on Youth Mentoring (Portland State University, 2012), which was begun in 2007 at Portland State University. The Institute, which occurs annually, brings together mentoring researchers and experienced practitioners from the United States and abroad for an intensive week-long seminar focused on the latest developments in theory and research on youth mentoring. A listserv dedicated to facilitating sharing related to youth mentoring research and practice has also been established and grown to include over 600 members (L-Soft, 2012). These latter developments are well-aligned with research that indicates direct dialogue between researchers and practitioners is one of the most effective strategies for facilitating use of research evidence (Nutley, Walter, & Davies, 2007). Funding Developments regarding the availability of funding to support the areas of research, practice, and the research-practice linkage for youth mentoring have been mixed. Of significant note is the discontinuation of two major streams of federal funding in the United States for mentoring programs within schools through the U.S. Department of Education and for those serving children with incarcerated parents through the U.S. Department of Health and Human Services (Fernandes-Alcantara, 2012). Questions have been raised regarding the extent to which both the design and evaluation of these funding initiatives were adequately informed by available research evidence (see Fernandes-Alcantara, 2012; Wheeler et al., 2010). In contrast, those sponsored in recent years by the Office of Juvenile Justice and Delinquency Prevention are noteworthy for their explicit use of current research to inform parameters for practice funding (e.g., Strategic Enhancement to Mentoring Programs) as well as for the use of rigorous methods to evaluate new practices (e.g., Mentoring Enhancement Demonstration Program). Notably lacking, too, in the area of funding is a major philanthropic commitment to support effective and wide-scale mentoring of youth either within the United States or globally on the order of those that have been directed toward other priorities

(see, e.g., initiatives sponsored by the Bill & Melinda Gates Foundation, 2012). Overall, although the current levels and types of investments to support youth mentoring from governmental and private sources are noteworthy, they appear to fall short of the foundation likely needed to facilitate major advances in the areas of either research or practice within the near horizon. International Significant developments also have taken place internationally. These include ambitious multisite investigations of mentoring relationships and programs in non-U.S. contexts. In Canada, for example, researchers are conducting a multiyear investigation of nearly a thousand youth served by 15 affiliate agencies of Big Brothers Big Sisters of Canada and their mentors (Big Brothers Big Sisters of Canada, 2011). The past several years also have seen the cultivation of new and stronger networks for supporting youth mentoring practice in diverse settings globally. Illustratively, Big Brothers Big Sisters International (BBBSI) (n.d.) now includes countrywide affiliates in most continents. It still appears to be the case, however, that organized youth mentoring initiatives have yet to take hold abroad to the same extent that has occurred in the United States. All international affiliates of BBBSI, for example, served an estimated 27,000 youth in 2009, whereas BBBSA affiliates served nearly ten times this number in the United States (BBBSI, n.d.). Nevertheless, as major geopolitical developments continue to unfold in different parts of the world, this may provide new and important opportunities for formal mentoring support to be made available to much larger and more diverse populations of youth.

Connecting the Handbook to Contemporary Trends in Youth Mentoring The contemporary trends in youth mentoring that we have summarized were an important point of reference in determining the content of this second edition of the Handbook. With regard to research and advancing the field’s knowledge base, we elected to include new chapters on several topics that have been the subject of relatively limited investigation. We did so with the hope that highlighting the needs and opportunities for research on such topics would serve as a catalyst for improving the base of available evidence in those areas.

Youth Mentoring in Contemporary Perspective    11 Among the chapters added with this aim in mind are examinations of mentoring relationships from the varying perspectives of biology, goal setting and goal pursuit, social networks, and social class, groups as a format for mentoring programs, and the potential of mentoring to make a positive difference in the lives of youth with mental health needs, those with incarcerated parents, and those who are recent immigrants or refugees. We also asked all contributors to make a concerted effort to comprehensively identify available research relevant to the topics of their chapters through the use of a well-delineated and multifaceted literature search strategy (Cooper, 2010). These strategies, summarized briefly in each chapter, were intended to ensure that coverage of each topic would effectively represent the current state of knowledge in different areas of the field. Contributors for topics addressed in the first edition were encouraged, furthermore, to focus their reviews on research not included in the previous version of the same chapter in order to build on this earlier work. With respect to practice, one of our aims was to provide enhanced consideration of mentoring of young people as it occurs within the broader landscape of youth-serving programs and organizations. As discussed in this chapter, we see greater attention to opportunities for mentoring within these contexts as likely to be essential when the aim is to have young persons in an entire community or population experience improved outcomes. To this end, the Handbook begins with foundational chapters that examine in-depth the interface of mentoring with the much larger and encompassing domains of positive youth development and prevention science. At the same time, we endeavored to make the Handbook more useful to those in applied roles who are engaged in the design, implementation, and evaluation of formal mentoring programs for youth. An entire section of the Handbook now addresses the concerns that routinely arise in the day-to-day operations of youth mentoring programs. The chapters comprising this section address many topics that did not receive similarly dedicated consideration in the first edition of the Handbook. These include program funding, mentor screening and youth protection, mentor and youth matching, family involvement, relationship termination and closure, and ethical issues. Complementing these chapters are updated treatments of mentor recruitment and retention, program evaluation, and issues involved with assessing the benefits of mentoring from an economic perspective. To further address applied concerns, chapters within this edition of the Handbook include summary “checklists” of recommendations for practice. These

checklists are derived as much as possible from available empirical evidence and thus are intended to aid readers in making fruitful linkages between research and practice. We also had at least one experienced practitioner anonymously review each chapter. This feedback helped to ensure that chapters addressed issues of key concern to those in applied roles. In an effort to support growing attention to mentoring of youth within a global context, this edition of the Handbook includes a chapter focused on advancing understanding of mentoring relationships from an international and cross-cultural perspective. Clearly, though, as we have discussed, progress on this front is likely to be tied intricately to a range of other considerations, including funding availability for youth mentoring initiatives worldwide and geopolitical developments occurring outside of the field. We are optimistic that the field will be able to build on its recent accomplishments to more fully advance the efficaciousness and accessibility of mentoring opportunities for young persons. Collectively, as we discuss in our concluding chapter, the Handbook’s contents offer a strong foundation for facilitating future progress toward these goals.

References Ahrens, K., DuBois, D. L., Lozano, P., & Richardson, L. P. (2010). Naturally acquired mentoring relationships and young adult outcomes among adolescents with learning disabilities. Learning Disabilities Research & Practice, 25(4), 207–216. Bernstein, L., Rappaport, C. D., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S. Department of Education’s Student Mentoring Program: Final report. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education, U.S. Department of Education. Big Brothers Big Sisters of America. (2009). Enhanced school-based mentoring preliminary findings and high school Bigs. Unpublished document, Big Brothers Big Sisters of America. Big Brothers Big Sisters of Canada. (2011). Big funding for research project. Retrieved from http://www .bigbrothersbigsisters.ca/en/home/newsevents/ NewsArchive/BigFundingForResearchProject.aspx Big Brothers Big Sisters International. (n.d.). Retrieved from http://www.bbbsi.org/ Bill & Melinda Gates Foundation. (2012). About the Foundation. Retrieved from http://www.gatesfoun dation.org/about/Pages/overview.aspx Blakeslee, J. E., & Keller, T. E. (in press). Building the youth mentoring knowledge base: Publishing trends and co-authorship networks. Journal of Community Psychology.

12  Introduction Bowles, J. (2006). Group competition, reproductive leveling, and the evolution of human altruism. Science, 314(5805), 1569–1572. Boys & Girls Clubs of America. (n.d.). Why Boys & Girls Clubs? Retrieved from http://www.bgca.org/who weare/Pages/WhoWeAre.aspx Britner, P. A. (Ed.). (2005). Mentoring with children and youth [Special issue]. Journal of Primary Prevention, 26(2). Center for Evidence-Based Mentoring. (2011). Welcome to The Center for Evidence-Based Mentoring. Retrieved from http://umbmentoring.org/index.html Center for Interdisciplinary Mentoring Research. (2012). The PSU Center for Interdisciplinary Mentoring Research. Retrieved from http://www.pdx.edu/men toring-research/the-psu-center-for-interdisciplinarymentoring-research Cooper, H. (2010). Research synthesis and meta-analysis: A step-by-step approach (4th ed.). Thousand Oaks, CA: Sage. Diamond, J. (2012). The world until yesterday. New York: Viking Press. DuBois, D. L., & Karcher, M. J. (2005). Youth mentoring: Theory, research, and practice. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 2–11). Thousand Oaks, CA: Sage. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., & Rhodes, J. E. (Eds.). (2006). Youth mentoring: Bridging science with Practice [Special Issue]. Journal of Community Psychology, 34(6). Eby, L. T., Allen, T., Evans, S., Ng, T., & DuBois, D. (2008). Does mentoring matter? A multidisciplinary metaanalysis comparing mentored and non-mentored individuals. Journal of Vocational Behavior, 72, 254–267. Erickson, L. D., McDonald, S., & Elder Jr., G. H. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367. Federal Mentoring Council. (2012). Retrieved from http:// www.federalmentoringcouncil.gov/ Fernandes-Alcantara, A. L. (2012). Vulnerable youth: Federal mentoring programs and issues. Congres­ sional Research Services Report for Congress. Retrieved from http://www.fas.org/sgp/crs/misc/ RL34306.pdf Gastic, B., & Johnson, D. (2009). Teacher-mentors and the educational resilience of sexual minority youth. Journal of Gay & Lesbian Social Services: Issues in Practice, Policy & Research, 21, 219-231. Grossman, J. B., Chan, C. S., Schwartz, S. E. O., & Rhodes, J. E. (2012). The test of time in schoolbased mentoring: The role of relationship duration and re-matching on academic outcomes. American Journal of Community Psychology, 49, 43–54. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., & McMaken, J. (with Jucovy, L. Z.). (2007). Making

a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Herrera, C., Kauh, T. J., Cooney, S. B., Grossman, J. B., & McMaken, J. (2008). High school students as mentors: Findings from the Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Hirsch, B. J., Deutsch, N. L., & DuBois, D. L. (2011). After-school centers and youth development: Case studies of success and failure. New York: Cambridge University Press. Hurd, N., & Zimmerman, M. (2010). Natural mentors, mental health, and risk behaviors: A longitudinal analysis of African American adolescents transitioning into adulthood. American Journal of Community Psychology, 46, 36–48. ICF International (2011). Mentoring children affected by incarceration: An evaluation of the Texas Amachi Program. Washington, DC: Office of Juvenile Justice and Delinquency Prevention. Jolliffe, D., & Farrington, D. P. (2007). A rapid evidence assessment of the impact of mentoring on re-offending: A summary. Cambridge, UK: Cambridge University (Home Office Online Report 11/07). Retrieved from http://www.youthmentoring.org.nz/content/docs/ Home_Office_Impact_of_mentoring.pdf Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9, 99–113. Karcher, M. J., Davidson, A., Rhodes, J. E., & Herrera, C. (2010). Pygmalion in the program: The role of teenage mentors’ attitudes in shaping their mentees’ outcomes. Applied Developmental Science, 14(4), 212–227. Karcher, M. J., & Nakkula, M. J. (Eds.). (2010). Play, talk, learn: Promising practices in youth mentoring [Special issue]. New Directions for Youth Development, 126. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. (pp. 82–99). Thousand Oaks, CA: Sage. King, C. A., Klaus, N., Kramer, A., Venkataraman, S., Quinlan, P., & Gillespie, B. (2009). The youthnominated support team-version II for suicidal adolescents: A randomized controlled intervention trial. Journal of Consulting and Clinical Psychology, 77, 880–893. L-Soft. (2012). [email protected]. EDU: A forum for communication among researchers and practitioners concerning mentoring of youth. Retrieved from http://www.lsoft.com/scripts/wl.exe?SL 1=YOUTHMENTORING&H=LISTSERV.UIC.EDU MENTOR. (2006). Mentoring in America 2005: A snapshot of the current state of mentoring. Alexandria, VA: Author. MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author.

Youth Mentoring in Contemporary Perspective    13 MENTOR. (2010). Overview of the SafetyNET criminal background check pilot fact sheet for mentoring organizations. Washington, DC: Author. Retrieved from http://www.mentoring.org/downloads/mentor ing_1281.pdf MENTOR (n.d.a.). 2011 National Mentoring Summit. http:// www.mentoring.org/downloads/mentoring_1281.pdf MENTOR (n.d.b.). Research in Action Series. Retrieved from http://www.mentoring.org/news_and_research/ research_and_studies/research_in_action/ Nutley, S., Walter, I., & Davies, H. (2007). Using evidence: How research can inform public services. Bristol, UK: Policy Press. Portland State University. (2012). Summer Institute on Youth Mentoring. Retrieved from http://pdx.edu/ youth-mentoring/ Rhodes, J. E. (2005). A Model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. (pp. 30–43). Thousand Oaks, CA: Sage. Rhodes, J. E., Spencer, R., Saito, R. N., & Sipe, C. L. (2006). Online mentoring: The promise and challenges of an emerging approach to youth development. The Journal of Primary Prevention, 27, 497–513. Scales, P. C. (2003). Other people’s kids: Social expectations and American adults’ involvement with children and adolescents. New York: Kluwer Academic/ Plenum Publishers. Schwartz, S., Rhodes, J. E., Chan, C., & Herrera, C. (2011). The impact of school-based mentoring on youth with different relational profiles. Develop­ mental Psychology, 47, 450–462.

Shanahan, M., Erickson, L., Vaisey, S., & Smolen, A. (2008). Helping relationships and genetic propensities: A combinatoric study of DRD2, mentoring, and educational continuation. Twin Research and Human Genetics, 10, 285–298. Thomas, R. E., Lorenzetti, D., & Spragins, W. (2011). Mentoring adolescents to prevent drug and alcohol use (Review). Cochrane Database of Systematic Reviews, Issue 11. Art. No.: CD007381. Tolan, P., Henry, D., Schoeny, M., & Bass, A. (2008). Mentoring interventions to affect juvenile delinquency and associated problems. Campbell Systematic Reviews, 16. Retrieved from http://www.campbell collaboration.org/lib/download/238/ Vazsonyi, A. T., & Snider, J. B. (2008). Mentoring, competencies, and adjustment in adolescents: American part-time employment and European apprenticeships. International Journal of Behavioral Development, 32, 46–55. Walker, G. (2005). Youth mentoring and public policy. In D. L. DuBois & M. L. Karcher (Eds.), Handbook of youth mentoring (pp. 510–524). Thousand Oaks, CA: Sage. Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials of schoolbased mentoring: Making sense of mixed findings. Social Policy Report, 24(3). Wong, K. (2009). Twilight of the Neandertals. Scientific American, 301(2), 32–37. Wood, S., & Mayo-Wilson, E. (2012). School-based mentoring for adolescents: A systematic review and meta-analysis. Research on Social Work Practice, 1–13.

PART II Frameworks and Foundations

2  MENTORING AND POSITIVE YOUTH DEVELOPMENT Richard M. Lerner, Christopher M. Napolitano, Michelle J. Boyd, Megan K. Mueller, and Kristina S. Callina1

Introduction This chapter presents the theoretical and empirical foundations of a strength-based conception of youth, termed the positive youth development (PYD) perspective, and explains how mentoring may provide an essential resource in promoting PYD. The PYD perspective suggests that, because of their enormous capacity to change their behavioral and psychological characteristics (a capacity termed plasticity), all youth possess strengths. This capacity to change is linked to the development of new cognitive abilities, interests, behavioral skills, and social relationships. The key hypothesis tested in the PYD perspective is that if the strengths of youth can be aligned with the resources for positive growth found in families, schools, and communities—for instance, the capacities of adults to provide for young people a nurturing, positive milieu in which their strengths may be enhanced and positively directed (DuBois & Rhodes, 2006; Karcher, 2005; Karcher, Davis, & Powell, 2002; Rhodes & DuBois, 2008; Rhodes & Lowe, 2009)—then young people’s healthy development may be optimized (R. M. Lerner, 2004). These healthy development-promoting resources are termed developmental assets in the PYD literature (Benson, Scales, Hamilton, & Sesma, 2006). Developmental assets may take many forms (Theokas & Lerner, 2006; Urban, Lewin-Bizan, & Lerner, 2009). However, current research from the 4-H Study of

PYD (R. M. Lerner et al., 2005; Phelps et al., 2007), which we discuss later in the chapter, suggests that the adults in an adolescent’s life may be the most important developmental asset for positive growth (Theokas & Lerner, 2006). This link between PYD and people as developmental assets suggests that mentoring may be a key resource in the lives of youth. We define effective mentoring as sustained, high-quality, and skillbuilding relationships between youth and adults (e.g., see DuBois & Rhodes, 2006; Karcher, 2005; R. M. Lerner, 2004; Rhodes & DuBois, 2008; Rhodes & Lowe, 2009). As we review the literature pertinent to the PYD perspective, and discuss what is known about mentoring in the promotion of PYD, we specify the individual and contextual conditions that make mentoring a positive resource in the lives of young people. This discussion also identifies unanswered questions about the PYD-mentoring relationship. Accordingly, this chapter also notes directions for future research and the potential implications of this research for applications to policy and programs.

Why Do We Need a Positive Youth Development (PYD) Perspective? All too often in the United States we discuss positive development in regard to the absence of negative or undesirable behaviors. Typically, such

1

The preparation of this chapter was supported in part by grants from the National 4-H Council, the John Templeton Foundation, the Thrive Foundation for Youth, and the National Science Foundation (Grant A150001 NSZ004).

17

18  Frameworks and Foundations descriptions are founded on the assumption that youth are “broken” or in danger of becoming “broken” (Benson, 2003), and thus we regard young people as “problems to be managed” (Roth, BrooksGunn, Murray, & Foster, 1998). As such, when we describe a successful young person we speak about a youth whose problems have been managed or are, at best, absent. We might say, then, that a youth who is manifesting behavior indicative of positive development is someone who is not taking drugs or using alcohol, is not engaging in unsafe sex, and is not participating in crime or violence. Benson (2003) explains that the focus in Americans’ discussions of youth on their problems and the use by Americans of a vocabulary that stresses the risks and dangers of young people occurs because we have a culture dominated by deficit and risk thinking, by pathology and its symptoms. This shapes our research, our policy, our practice. It fuels the creation of elaborate and expensive service and program delivery infrastructures, creates a dependence on professional experts, encourages an ethos of fear, and by consequence, derogates, ignores and interferes with the natural and inherent capacity of communities to be community. (p. 25) The deficit model of youth that shapes our vocabulary about the behaviors prototypic of young people results, then, in an orientation in America to discuss positive youth development as the absence of negative behaviors. Unfortunately, even in programs purportedly focused on positive youth development, a predominant emphasis in the youth development field continues to be a reliance on this deficit model of youth and, as such, on defining positive youth development as the absence of adolescent problem behaviors. For instance, Catalano, Berglund, Ryan, Lonczak, and Hawkins (1999) noted that “currently, problem behaviors are tracked more often than positive ones and, while an increasing number of positive youth development interventions are choosing to measure both, this is still far from being the standard in the field” (p. vi). The absence of an accepted vocabulary for the discussion of positive youth development has been, then, a key obstacle to evaluating the effectiveness of programs or policies aimed at promoting such change. People do not measure what they cannot name, and they often do not name what they cannot measure (T. Gore, personal communication, December 13, 2002). In short, characterizations of young people as problems to be managed or as primarily people in

need of fixing reflect both a deficit approach to human development and a belief that some shortcoming of character or personality leads youth to become involved in risky or negative behaviors. Given the presence of such a deficit, the appropriate and humane actions to take in regard to young people are to prevent the actualization of the inevitable problems they will encounter. Indeed, policymakers and practitioners are pleased when their actions are associated with the reduction of such problem behaviors as teenage pregnancy and parenting, substance use and abuse, school failure and dropout, and delinquency and violence. Everyone should, of course, be pleased when such behaviors diminish. However, it is dispiriting for a young person to learn that he or she is regarded by adults as someone who is likely to be a problem for others as well as for himself or herself. It is discouraging for a young person to try to make a positive life when he or she is confronted by the suspicion of substance abuse, sexual promiscuity, and a lack of commitment to supporting the laws of society. What sort of message are we sending our children when we speak of them as inevitably destined for trouble unless we take preventive steps? How do such messages affect the self-esteem of young people, and what is the impact of such messages on their spirit and motivation? Some words for describing positive behaviors about youth exist, for example, pertaining to academic achievement and activities relating to current or potentially successful entrepreneurship. Nevertheless, the vocabulary for depicting youth as “resources to be developed” (Roth et al., 1998) has not been as rich or nuanced as the one available for depicting the problematic propensities of young people. As a society, we must do a better job of talking about the positive attributes of young people. We believe that we must talk to our youth about what they should and can become, and not only about what they must avoid being. We should then act on our statements, and work with young people to promote their thriving. In the context of nurturing and healthy adult-youth relationships, we need to offer young people the opportunities to learn and use the skills involved in participating actively in their communities and in making productive and positive contributions to themselves and their families and society. These “oughts” for social change for youth represent a formidable challenge involving nothing short of thorough systems change in the United States. The challenge is to provide for Americans a new vision and vocabulary about youth. This challenge is being met by a historically unique and

Mentoring and Positive Youth Development    19 significant convergence of efforts by scholars, practitioners, policymakers, and youth and families. In these early years of the 21st century, a new, positive, and strength-based vision and vocabulary for discussing America’s young people is beginning to emerge. This perspective has many scholarly roots. One was the emphasis in comparative psychology and evolutionary biology on the plasticity of developmental change (e.g., Gottlieb, 1997; Schneirla, 1957; Tobach & Greenberg, 1984). A second, related source was found in several theoretical models linked to what are termed relational developmental systems theories (Overton, 2012); these are theories that focus on the mutually influential relations between people and their contexts (e.g., their families, schools, or communities). These theories (e.g., by Baltes, Lindenberger, & Staudinger, 2006; Bronfenbrenner, 2005; Elder, 1998) were a source of research aimed at optimizing individual and group changes. In the 1990s, this new vision and vocabulary about youth was propelled by the increasingly collaborative contributions of scholars (e.g., Benson, 2003; Damon, 2004; Villarruel, Perkins, Borden, & Keith, 2003), practitioners (e.g., Pittman, Irby, & Ferber, 2001), and policymakers (e.g., Cummings, 2003; Gore, 2003). Their collective efforts enabled young people to be increasingly seen within numerous sectors of American society as resources to be developed (Roth & Brooks-Gunn, 2003a, 2003b). The new vocabulary about positive youth development emphasizes the strengths present within all young people and involves concepts such as developmental assets (Benson, 2003), noble purpose (Damon, 2008), civic engagement (e.g., Sherrod, Flanagan, & Youniss, 2002), community

youth development (e.g., Eccles & Gootman, 2002; Villarruel et al., 2003), well-being (Bornstein, Davidson, Keyes, Moore, & the Center for Child Well-being, 2003), and thriving (Benson, 2008). These concepts are predicated on the ideas that every young person has the potential for successful, healthy development and that all youth possess the capacity for positive development.

The PYD Perspective These ideas about the strengths of youth converged in the formulation of the PYD perspective. The perspective is framed by two hypotheses (e.g., see J. V. Lerner, Phelps, Forman, & Bowers, 2009; R. M. Lerner, 2009). Each hypothesis is associated with two subsidiary hypotheses. The first hypothesis pertains to the measurement of PYD. The second focuses on the relations between individuals and contexts that (within relational developmental systems models) provide the basis of human development. Hypothesis 1. PYD Is Comprised of Five Cs Based both on the experiences of practitioners and on reviews of the adolescent development literature (Eccles & Gootman, 2002; R. M. Lerner, 2004; Roth & Brooks-Gunn, 2003a, 2003b), “Five Cs”—competence, confidence, connection, character, and caring—were hypothesized as a way of conceptualizing PYD (and of integrating all the separate indicators of it, such as academic achievement or self-esteem). The definitions of these Cs are presented in Table 2.1.

Table 2.1   Definitions of the Five Cs of Positive Youth Development The Five Cs of Positive Youth Development “C”

Definition

Competence

A positive view of one’s actions in specific areas, including social, academic, cognitive, health, and vocational

Confidence

An internal sense of overall positive self-worth and self-efficacy

Connection

Positive bonds with people and institutions that are reflected in exchanges between the individual and his or her peers, family, school, and community in which both parties contribute to the relationship

Character

Respect for societal and cultural norms, possession of standards for correct behaviors, a sense of right and wrong (morality), and integrity

Caring/Compassion

A sense of sympathy and empathy for others

20  Frameworks and Foundations These Five Cs were linked to the positive outcomes of youth development programs reported by Roth and Brooks-Gunn (2003a, 2003b). In addition, these “Cs” are prominent terms used by ­practitioners, adolescents involved in youth development programs, and the parents of these adolescents in describing the characteristics of a “thriving youth” (King et al., 2005). Program staff can help mentors to foster these Cs among their mentees in a variety of ways (see Table 2.2). A hypothesis subsidiary to the postulation of the Five Cs as a means to measure PYD is that, when a young person manifests the Cs across time, he or she will be on a path marked by contributions to self (e.g., maintaining one’s health and one’s ability therefore to remain an active agent in one’s own development) and to family, community, and the institutions of civil society (e.g., families, neighborhoods, schools, religious groups; Elshtain, 1999; R. M. Lerner, 2004). Simply, when the Five Cs develop, they lead to the development of youth contribution, the “sixth C” of PYD. The contributing person keeps herself healthy and fit, so as not to be an unnecessary liability to or an unnecessary user of the resources of others and, as well, helps family members without any coercion, assists neighbors without any compensation for doing so, and helps keep the institutions of civil society strong by, for instance, volunteering to help others (e.g., through food or clothing drives) and acting to support the institutions of democracy (e.g., by working to enhance voter registration, by supporting political debate, and by voting). An adult engaging in such integrated contributions is a person manifesting adaptive developmental regulations (Brandtstädter, 2006). A second subsidiary hypothesis is that there should be an inverse relation across development between PYD (e.g., the Five Cs) and behaviors indicative of risk behaviors or internalizing and externalizing problems (e.g., delinquency, substance use, depression, aggression). That is, this hypothesis suggests that as evidence for positive behavior increases, there should be fewer indications of problematic behaviors. Simply, the idea is that increases in good things are associated with decreases in what is bad. This idea was forwarded in particular by Pittman and her colleagues (e.g., Pittman et al., 2001) in regard to applications of developmental science to policies and programs. In essence, the hypothesis is that the best means to prevent problems associated with adolescent behavior and development (e.g., depression, aggression, drug use and abuse, or unsafe sexual behavior) is to promote positive development.

The Status of Empirical Support for Hypothesis 1 Findings from a national longitudinal investigation, the 4-H Study of PYD, provide some support for the ideas asserted in Hypothesis 1 (e.g., R. M. Lerner et al., 2005; see also Bowers et al., 2010; Jeličić, Bobek, Phelps, Lerner, & Lerner, 2007; Phelps et al., 2009; and Theokas et al., 2005, for other discussions of these data). The study, which currently includes about 7,000 youth from 42 states, involves longitudinal assessment of adolescents beginning in grade 5 (at about 10 years of age) and is currently designed to follow youth through grade 12 (Phelps et al., 2009; Urban et al., 2009). The study provides evidence for the existence of the Five Cs of PYD, for the existence of the “sixth C” of contribution, and for positive relations among these Cs (R. M. Lerner et al., 2005). Indeed, PYD in an earlier grade predicts contribution in subsequent grades (e.g., Jeličić et al., 2007). Similarly, Karcher (2009) found that participation in programs that promote contribution, such as peer mentoring, was associated with increases in academic self-esteem and connectedness. Findings from the 4-H Study also show inverse relations between the Cs and the risk/problem behaviors discussed earlier (Jeličić et al., 2007), although this relationship is more nuanced than originally hypothesized. For example, Phelps et al. (2007) and Zimmerman, Phelps, and Lerner (2008) found that PYD and risk/problem behaviors follow different trajectories over time; that is, the patterns of change associated with these outcomes differ among individuals. Whereas some youth show inverse relations between trajectories of PYD and risk/problem behaviors, other youth show increases in both dimensions and still others show decreases in both. These findings, that youth have diverse combinations of trajectories of positive and problematic behaviors, indicate that both prevention and promotion efforts must be pursued in terms of the policy and programs directed at youth. For instance, within the 4-H Study data set, even among youth who have a history of risk/problem behaviors, including bullying, delinquency, or substance use, there may be substantial evidence of PYD and, as well, contribution. Accordingly, there may be strengths—and the basis for positive change—among even those youth who show trajectories of marked risk/problem behaviors. Hypothesis 2. Youth-Context Alignment Promotes PYD Based on the idea that the potential for systematic change within a person across life (i.e., for

Mentoring and Positive Youth Development    21 plasticity) represents a fundamental strength of human development, the hypothesis was generated that, if the strengths of youth are aligned with resources for healthy growth present in the key contexts of adolescent development—the home, the school, and the community—then enhancements in positive functioning at any one point in time (i.e., well-being; R. M. Lerner, 2004) may occur. In turn, the systematic promotion of positive development will occur across time (i.e., thriving; e.g., R. M. Lerner, 2004; R. M. Lerner et al., 2005). A subsidiary hypothesis is that there exist, across the key settings of youth development (i.e., families, schools, and communities), developmental assets that support the growth of a healthy youth (Benson et al., 2006). There is broad agreement among researchers and practitioners in the youth development field that the concept of developmental assets is important for understanding what needs to be marshaled in homes, classrooms, and community-based programs to foster PYD (Benson et al., 2006; R. M. Lerner, 2007). In fact, a key impetus for the interest in the PYD perspective among both researchers and youth program practitioners lies in ascertaining the nature of the resources for positive development that are present in youth programs. Examples are the thousands of after-school programs delivered either by large, national organizations, such as 4-H, Boys & Girls Clubs, scouting, Big Brothers Big Sisters, YMCA, or Girls, Inc., or by local organizations. The focus on youth programs is important not only for practitioners in the field of youth development, however. In addition, the interest in exploring youth development programs as a source of developmental assets for youth derives from theoretical interest in the role of the macro-level systems effects of the ecology of human development on the course of healthy change in adolescence (Bronfenbrenner & Morris, 2006); interest derives as well from policymakers and advocates, who believe that at this point in the history of the United States communitylevel efforts are needed to promote positive development among youth (e.g., Cummings, 2003; Gore, 2003; Pittman et al., 2001). The Status of Empirical Support for Hypothesis 2 Once again, findings from the 4-H Study of PYD lend empirical support to the ideas associated with Hypothesis 2 (e.g., see also J. V. Lerner et al., 2009; R. M. Lerner, 2009; R. M. Lerner et al., 2005; Phelps et al., 2007; and Urban et al., 2009, 2010, for other discussions of these findings). For example, Theokas and Lerner (2006) and Urban et al. (2009)

found that greater ecological assets (e.g., mentors, opportunities for learning and recreation) were positively related to PYD and negatively related to problem/risk behaviors, such as depression. In addition, in all settings involving youth (families, schools, and the community), the assets most associated with high levels of PYD and contribution and with low levels of risk and problem behaviors were people. Key here are large quantities of quality time with parents; access to a competent, caring teacher; or positive and sustained relations with a mentor. Zarrett et al. (2009) explored the association between patterns of involvement in communitybased programs and PYD. Findings indicated that participation in youth development programs— which are indicated by the presence of a positive and sustained relationship with a mentor; life-skills building activities; and opportunities for youth participation in, and leadership of, valued family, school, and community activities (what R. M. Lerner, 2004, terms the “Big 3”)—was related to PYD and youth contribution, even after controlling for the total time youth spent in other out-of-school time activities, such as sports (Zarrett et al., 2009). This research underscores the importance of mentors in communitybased, youth development programs (see also Larson, 2006; Theokas & Lerner, 2006). In all these programs, mentors—caring, capable, and committed adults invested in the lives of young people—are a key developmental asset associated with higher levels of PYD and, as well, with lower levels of problem/risk behaviors (e.g., substance use or bullying) (Theokas & Lerner, 2006).

Promoting the PYD-Mentoring Linkage Mentoring clearly has the potential to constitute a set of relations between youth and their social world that enhances adolescents’ life skills, provides opportunities for their making valued contributions to self and society, and promotes PYD. To foster the characteristics of both PYD and contribution, mentors must be certain that their interactions with mentees reflect what has been noted to be the Big 3 features of effective youth development programs. Through the enactment of these features of effective youth development programs, mentors can ensure that youth gain access to the key ecological assets present in their communities and, as such, develop the Cs of PYD (see Table 2.2). While not discounting the importance of natural mentoring relationships (DuBois & Silverthorn, 2005a, 2005b; Erikson, McDonald, & Elder, 2009),

22  Frameworks and Foundations mentoring that occurs within the context of youth development programs may be particularly beneficial in the promotion of PYD. For instance, effective, high-quality and enduring mentoring is associated with the capacity for youth to engage in high-quality social relationships, to have greater academic achievement, school engagement, and school adjustment, and to view their futures more positively (Rhodes, Spencer, Keller, Liang, & Noam, 2006). The role of positive adult-youth relationships has been underscored as well by Rhodes (2002; Rhodes & Roffman, 2003). Focusing on volunteer mentoring relationships, for instance, Rhodes and Roffman (2003) noted that these nonparental “relationships can positively influence a range of outcomes, including improvements in peer and parental relationships, academic achievement, and self-concept; lower recidivism rates among juvenile delinquents; and reductions in substance abuse” (p. 227). However, Rhodes and Roffman (2003) also noted that there is a developmental course to these effects of volunteer mentoring on youth. When young people are in relationships that last a year or longer, they are most likely to experience improvements in academic, psychological, social, and behavioral characteristics. By contrast, when youth are in relationships that last for only 6–12 months, fewer positive outcomes of mentoring are evident. When young people are in mentoring relationships that end relatively quickly, it appears that mentoring may actually be detrimental. Decrements in positive functioning have been reported in such circumstances (Rhodes, 2002; Rhodes & Roffman, 2003). Features of Youth Development Programs Linking Mentoring and PYD Thus, mentoring has the potential to promote PYD, and youth development programs involve the presence of sustained and effective mentoring (e.g., R. M. Lerner, 2004; Rhodes & Lowe, 2009). As such, it is helpful to discuss in more detail how such programs may enable mentoring to guide adolescents toward enhanced PYD. Examples of youth development programs are 4-H, Boys & Girls Clubs, Boy Scouts/Girl Scouts, Big Brothers Big Sisters, YMCA, or Girls, Inc. For instance, the 4-H website indicates that the vision of 4-H is “A world in which youth and adults learn, grow, and work together as catalysts for positive change” (4-H, 2010). Similarly, the Boys & Girls Clubs website indicates that their “club programs and services promote and enhance the development of boys and

girls by instilling a sense of competence, usefulness, belonging, and influence” (Boys & Girls Clubs of America, 2010). The Big Brothers Big Sisters program has a direct focus on fostering positive relationships, stating that their mission is “expanding horizons through the power of one-toone friendships” (Big Brothers Big Sisters of America, 2009). In essence, then, it appears that youth development programs involve mentors imbuing in youth key life skills and, in turn, affording youth the opportunity to use these skills in valued family, school, or community activities. This central role of mentoring in youth development programs has been identified by other scholars of PYD. Catalano et al. (1999) note that effective youth development programs “targeted healthy bonds between youth and adults, increased opportunities for youth participation in positive social activities, . . . [involved] recognition and reinforcement for that participation” (p. vi) and often used skills training as a youth competency strategy. These characteristics of effective youth development programs are similar to those identified by Roth and Brooks-Gunn (2003b), who noted that such programs transcend an exclusive focus on the prevention of health-compromising behaviors to include attempts to inculcate behaviors that stress youth competencies and abilities through “increasing participants’ exposure to supportive and empowering environments where activities create multiple opportunities for a range of skill-building and horizon-broadening experiences” (p. 94). In addition, Roth and Brooks-Gunn (2003b) indicated that the activities found in these programs offered both “formal and informal opportunities for youth to nurture their interests and talents, practice new skills, and gain a sense of personal and group recognition. Regardless of the specific activity, the emphasis lies in providing real challenges and active participation” (p. 204). In this regard, Roth and Brooks-Gunn (2003a) note that when these activities are coupled with an atmosphere that creates an atmosphere of hope for a positive future among youth, when the program “conveys the adults’ beliefs in youth as resources to be developed rather than as problems to be managed” (p. 204), then the goals of promoting PYD are likely to be reached. In other words, when activities that integrate skill-building opportunities and active participation occur in the presence of positive and supportive adult-youth relations, positive development will occur. Blum (2003) agrees. He notes that effective youth programs offer to youths activities through

Mentoring and Positive Youth Development    23 which relationships with caring adults are formed, relations that elicit hope in young people. When these programs provide as well the opportunity for youth to participate in community development activities, PYD occurs (Blum, 2003). Implications for Future PYD-Mentoring Research and Applications Given the theory and research in support of the role of mentoring in promoting PYD, especially when mentoring is involved with the other two components of the Big 3 facets of youth development programming (i.e., life-skills-building activities and opportunities and youth participation in and leadership of valued activities), it is incumbent for future research and applications of such data to ascertain how we can improve access to effective mentoring relationships. It would be useful for future researchers to identify constraints on access to mentors that exist in diverse communities and to test ideas both for increasing the number of trained mentors in youth development programs and for finding ways that other community institutions can provide resources so that trained mentors can be a more ubiquitous part of youth programs. A key question here is “what facets of mentoring will be most effective for what sorts of programs?” Although the Big 3 point to three general characteristics of youth programs and place great reliance on effective mentoring, more nuanced information is required about what skill sets are needed by mentors to deliver the substance of programs.

This applied research should be longitudinal in focus. A key multifaceted question here would be: “What characteristics of mentors in what sorts of programs will be most effective to promote what facets of positive youth development in what sorts of youth of what ages in what communities?” Such fine-tuned knowledge is necessary to maximize opportunities for diverse youth in diverse communities to have access to mentoring that will, in turn, maximize their chances for healthy, positive development. In sum, the PYD perspective provides evidence that all young people possess strengths and that thriving among adolescents may be promoted when youth strengths are aligned with the developmental assets for healthy development present in their communities. Mentors constitute such resources. They have available to them a rich set of opportunities to promote the Cs of PYD among diverse youth. In turn, mentoring clearly has the potential to constitute a set of relations between youth and their social world that enhances their life skills, provides opportunities for their making valued contributions to self and society, and promotes PYD. To foster the characteristics of PYD and contribution, mentors must be certain that their relations with mentees reflect the Big 3. Through the enactment of the features of effective youth development programs, mentors can ensure that youth gain access to the key ecological assets present in their communities and, as such, thrive across their adolescence. Table 2.2, developed from R. M. Lerner, Brittian, and Fay (2007), summarizes some of the actions that mentors may use to promote PYD.

Table 2.2   Checklist for Practitioners

“C” of PYD Competence

Actions Practitioners Might Coach Mentors to Take to Help Develop Each of Their Mentees’ Cs of Positive Youth Development Find things your mentee likes and support these passions and activities without taking over. Find things that your mentee does well and encourage him to pursue interests, activities, or hobbies that emphasize these skills. Help your mentee see that the skills she has are portable, that they can be transferred into other areas where she feels not so skilled. Actively involve your mentee in making decisions that impact the completion of family tasks. Turn mistakes—whether trivial or serious—into teachable moments.

(Continued)

24  Frameworks and Foundations Table 2.2  (Continued)

“C” of PYD Confidence

Actions Practitioners Might Coach Mentors to Take to Help Develop Each of Their Mentees’ Cs of Positive Youth Development Make sure your mentee has a convoy of support so that he feels loved and valued every day and everywhere. Share your own life woes and lapses in confidence and ask your mentee for help when you can. Be especially attentive to obstacles that may challenge your female mentee’s confidence: Confidence is likely to dip more for girls than for boys during early and middle adolescence. Increase youth social capital by connecting her to institutions and people to whom she might not otherwise have access.

Connection

Respect your mentee’s privacy but appreciate that privacy can be perilous. Be respectful but vigilant. Create opportunities in your community so your mentee feels her voice is being heard. All youth want to feel that they matter.

Character

If you don’t approve of a friend, a relationship, or an activity, speak out! Let your mentee know your values and explain why some behaviors aren’t acceptable. Make sure your actions align with your words—you are, after all, a key model for your mentee. Keep a sense of perspective—and sometimes a sense of humor—about minor infractions in behavior or deviations in character. Provide opportunities for your mentee to make his own decisions, and, when you give him this opportunity, live with the decisions he makes.

Caring

Keep in mind that times when our mentees treat us as if we’re disposable may be when they need us the most. Hang back, wait for an opening to talk, and respond. Caring is contagious: caring mentors help develop caring youth. Model caring in your interactions with your mentee and in your community. Encourage mentees to join school boards, civic organizations, or faith-based institutions to promote caring and social justice in the world around them.

Contribution

Encourage your mentee to participate in causes that align with her interests. Encourage people and institutions to welcome youth participation. Help youth marshal the resources they need so that their contributing efforts have a good chance of succeeding. Don’t overprotect your mentees from failure; they need to understand that even the most worthwhile efforts sometimes meet with disappointment.

Conclusions Effective mentoring—sustained, high-quality relationships with youth—serves as an important, and perhaps even necessary, developmental asset in the lives of youth. Practitioners engaged in such mentoring and, as well, providing youth with opportunities

to build life skills and to undertake leadership of valued community activities, can facilitate the development among young people of the several key indicators of positive youth development: the Five Cs of competence, confidence, connection, character, and caring, as well as the “sixth C” of contribution to self and society. Within the context

Mentoring and Positive Youth Development    25 of the Big 3 features of effective youth development programs, practitioners have available to them a rich array of actions they may take to foster the development of these features of positive development among adolescents.

References 4-H USA. (2010). Retrieved from http://www.4-h.org Baltes, P. B., Lindenberger, U., & Staudinger, U. M. (2006). Life span theory in developmental psychology. In R. M. Lerner & W. Damon (Eds.), Handbook of child psychology: Vol. 1. Theoretical models of human development (6th ed., pp. 569–664). Hoboken, NJ: Wiley. Benson, P. L. (2003). Developmental assets and assetbuilding community: Conceptual and empirical foundations. In R. M. Lerner & P. L. Benson (Eds.), Developmental assets and asset-building communities: Implications for research, policy, and practice (pp. 19–43). New York: Kluwer Academic/Plenum. Benson, P. (2008). Sparks: How parents can help ignite the hidden strengths of teenagers. San Francisco: Jossey-Bass. Benson, P. L., Scales, P. C., Hamilton, S. F., & Sesma, A., Jr. (2006). Positive youth development: Theory, research, and applications. In W. Damon & R. M. Lerner (Editors-in-chief) and R. M. Lerner (Ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (6th ed., pp. 894–941). Hoboken, NJ: Wiley. Big Brothers Big Sisters of America. (2009). Retrieved from http://www.bbbs.org Blum, R. W. (2003). Positive youth development: A strategy for improving adolescent health. In F. Jacobs, D. Wertlieb, & R. M. Lerner (Eds.). Enhancing the life chances of youth and families: Public service systems and public policy perspectives. Vol. 2 of Handbook of applied developmental science: Promoting positive child, adolescent, and family development through research, policies, and programs (pp. 237–252). Thousand Oaks, CA: Sage. Bornstein, H. H., Davidson, L., Keyes, C. M., Moore, K., & the Center for Child Well-being (Eds.). (2003). Well-being: Positive development across the life course. Mahwah, NJ: Erlbaum. Bowers, E. P., Li, Y., Kiely, M. K., Brittian, A., Lerner, J. V., & Lerner, R. M. (2010). The Five Cs model of positive youth development: A longitudinal analysis of confirmatory factor structure and measurement invariance. Journal of Youth and Adolescence, 39, 720–735. Boys & Girls Clubs of America. (2010). Retrieved from www.bgca.org Brandtstädter, J. (2006). Action perspectives on human development. In W. Damon & R. M. Lerner (Editors-in-chief) and R. M. Lerner (Ed.), Handbook

of child psychology: Vol. 1. Theoretical models of human development (6th ed., pp. 561–568). Hoboken, NJ: Wiley. Bronfenbrenner, U. (2005). Making human beings human: Bioecological perspectives on human development. Thousand Oaks, CA: Sage. Bronfenbrenner, U., & Morris, P. A. (2006). The bioecological model of human development. In W. Damon & R. M. Lerner (Editors-in-chief) and R. M. Lerner (Ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (6th ed., pp. 793-828). Hoboken, NJ: Wiley. Catalano, R. F., Berglund, M. L., Ryan, J. A. M., Lonczak, H. S., & Hawkins, J. D. (1999). Positive youth development in the United States: Research findings on evaluations of youth development programs. Washington, DC: U.S. Department of Health and Human Services. Cummings, E. (2003). Foreword. In D. Wertlieb, F. Jacobs, & R. M. Lerner (Eds.), Promoting positive youth and family development: Community systems, citizenship, and civil society. Vol. 3. Handbook of applied developmental science: Promoting positive child, adolescent, and family development through research, policies, and programs (pp. ix–xi). Thousand Oaks, CA: Sage. Damon, W. (2004). What is positive youth development? Annals of the American Academy of Political and Social Science, 591, 13–24. Damon, W. (2008). The path to purpose: Helping our children find their calling in life. New York: Free Press. DuBois, D. L., & Rhodes, J. E. (2006). Youth mentoring: Bridging science with practice. Journal of Community Psychology, 34, 547–565. DuBois, D. L., & Silverthorn, N. (2005a). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92. DuBois, D. L., & Silverthorn, N. (2005b). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524. Eccles, J., & Gootman, J. A. (Eds.) (2002). Community programs to promote youth development. Washington, DC: National Academies Press. Elder, G. H. (1998). The life course and human development. In W. Damon (Series Ed.) & R. M. Lerner (Vol. Ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (5th ed., pp. 939–991). New York: Wiley. Elshtain, J. B. (1999). A call to society. Society, 36(5), 11–19. Erikson, L. D., McDonald, S., & Elder, G. H., Jr. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367. Gore, A. (2003). Foreword. In R. M. Lerner & P. L. Benson (Eds.), Developmental assets and asset-building communities: Implications for research, policy, and practice (pp. xi–xii). Norwell, MA: Kluwer.

26  Frameworks and Foundations Gottlieb, G. (1997). Synthesizing nature-nurture: Prenatal roots of instinctive behavior. Mahwah, NJ: Erlbaum. Jeličić, H., Bobek, D., Phelps, E. D. Lerner, J. V., Lerner, R. M. (2007). Using positive youth development to predict contribution and risk behaviors in early adolescence: Findings from the first two waves of the 4-H Study of Positive Youth Development. International Journal of Behavioral Development, 31, 263–273. Karcher, M. J. (2005). The effects of developmental mentoring and high school mentors’ attendance on their younger mentees’ self-esteem, social skills, and connectedness. Psychology in the Schools, 42, 65–77. Karcher, M. J. (2009). Increases in academic connectedness and self-esteem among high school students who serve as cross-age peer mentors. Professional School Counseling, 12, 292–299. Karcher, M. J., Davis, C., & Powell, B. (2002). The effects of developmental mentoring on connectedness and academic achievement. School Community Journal, 12(2), 35–50. King, P. E., Dowling, E. M., Mueller, R. A., White, K., Schultz, W., Osborn, P., . . . Scales, P. C. (2005). Thriving in adolescence: The voices of youth-serving practitioners, parents, and early and late adolescents. Journal of Early Adolescence, 25, 94–112. Larson, R. (2006). Positive youth development, willful adolescents, and mentoring. Journal of Community Psychology, 34, 677–686. Lerner, J. V., Phelps, E., Forman, Y. E., & Bowers, E. (2009). Positive youth development. In R. M. Lerner, & L. Steinberg (Eds.), Handbook of adolescent psychology: Vol. 1. Individual bases of adolescent development (3rd ed., pp. 524–558). Hoboken, NJ: Wiley. Lerner, R. M. (2004). Liberty: Thriving and civic engagement among American youth. Thousand Oaks, CA: Sage. Lerner, R. M. (2007). The good teen: Rescuing adolescents from the myths of the storm and stress years. New York: Crown Publishing Group. Lerner, R. M. (2009). The positive youth development perspective: Theoretical and empirical bases of a strength-based approach to adolescent development. In C. R. Snyder and S. J. Lopez (Eds.), Oxford handbook of positive psychology (2nd ed., pp. 149–163). Oxford, UK: Oxford University Press. Lerner, R. M., Brittian, A., & Fay, K. (2007). Mentoring: A key resource for promoting positive youth development. In J. E. Rhodes (Ed.), Research in Action Series (Issue 1, pp. 1–15). Alexandria, VA: MENTOR. Lerner, R. M., Lerner, J. V., Almerigi, J., Theokas, C., Phelps, E., Gestsdóttir, S., . . . von Eye, A. (2005). Positive youth development, participation in community youth development programs, and community contributions of fifth grade adolescents: Findings from the first wave of the 4-H Study of Positive Youth Development. Journal of Early Adolescence, 25, 17–71. Overton, W. F. (2012). Evolving scientific paradigms: Retrospective and prospective. In L. A'bate (Ed.),

Paradigms in Theory Construction, pp. 31–68. New York, NY: Springer. Phelps, E., Balsano, A., Fay, K., Peltz, J., Zimmerman, S., Lerner, R. M., & Lerner, J. V. (2007). Nuances in early adolescent development trajectories of positive and of problematic/risk behaviors: Findings from the 4-H Study of Positive Youth Development. Child and Adolescent Clinics of North America, 16, 473–496. Phelps, E., Zimmerman, S., Warren, A. E. A., Jeličić, H., von Eye, A., & Lerner, R. M. (2009). The structure and developmental course of positive youth development (PYD) in early adolescence: Implications for theory and practice. Journal of Applied Develop­ mental Psychology, 30, 571–584. Pittman, K., Irby, M., & Ferber, T. (2001). Unfinished business: Further reflections on a decade of promoting youth development. In Public/Private Ventures (Ed.), Youth development: Issues, challenges, and directions (pp. 17–64). Philadelphia: Public/Private Ventures. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E., & DuBois, D. L. (2008). Mentoring relationships and programs for youth. Current Directions in Psychological Science, 17, 254–258. Rhodes, J. E., & Lowe, S. R. (2009). Mentoring in adolescence. In R. M. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology: Vol. 2. Contextual influences on adolescent development (3rd ed., pp. 152–190). Hoboken, NJ: Wiley. Rhodes, J. E., & Roffman, J. G. (2003). Nonparental adults as asset builders in the lives of youth. In R. M. Lerner & P. L. Benson (Eds.), Developmental assets and asset-building communities: Implications for research, policy, and practice (pp. 195–209). New York: Kluwer Academic/Plenum Publishers. Rhodes, J. E., Spencer, R., Keller, T. E., Liang, B., & Noam, G. (2006). A model for the influence of mentoring relationships on youth development. Journal of Community Psychology, 34, 691–707. Roth, J. L., & Brooks-Gunn, J. (2003a). What is a youth development program? Identification and defining principles. In F. Jacobs, D. Wertlieb, & R. M. Lerner (Eds.), Enhancing the life chances of youth and families: Public service systems and public policy perspectives. Vol. 2. Handbook of applied developmental science: Promoting positive child, adolescent, and family development through research, policies, and programs (pp. 197–223). Thousand Oaks, CA: Sage. Roth, J. L., & Brooks-Gunn, J. (2003b). What exactly is a youth development program? Answers from research and practice. Applied Developmental Science, 7, 94–111. Roth, J., Brooks-Gunn, J., Murray, L., & Foster, W. (1998). Promoting healthy adolescents: Synthesis of youth development program evaluations. Journal of Research on Adolescence, 8, 423–459. Schneirla, T. C. (1957). The concept of development in comparative psychology. In D. B. Harris (Ed.), The concept of development: An issue in the study of human behavior (pp. 78–108). Minneapolis: University of Minnesota Press.

Mentoring and Positive Youth Development    27 Sherrod, L., Flanagan, C., & Youniss, J. (Eds.). (2002). Growing into citizenship: Multiple pathways and diverse influences [Special issue]. Applied Developmental Science, 6(4). Theokas, C., Almerigi, J. B., Lerner, R. M., Dowling, E. M., Benson, P. L., Scales, P. C., . . . , von Eye, A. (2005). Conceptualizing and modeling individual and ecological asset components of thriving in early adolescence. Journal of Early Adolescence, 25, 113–143. Theokas, C., & Lerner, R. M. (2006). Observed ecological assets in families, schools, and neighborhoods: Conceptualization, measurement, and relations with positive and negative developmental outcomes. Applied Developmental Science, 10, 61–74. Tobach, E., & Greenberg, G., (1984). The significance of T. C. Schneirla’s contribution to the concept of levels of integration. In G. Greenberg & E. Tobach (Eds.), Behavioral evolution and integrative levels (pp. 1–7). Hillsdale, NJ: Erlbaum. Urban, J. B., Lewin-Bizan, S., & Lerner, R. M. (2009). The role of neighborhood ecological assets and activity involvement in youth developmental outcomes:

Differential impacts of asset poor and asset rich neighborhoods. Journal of Applied Developmental Psychology, 30, 601–614. Urban, J. B., Lewin-Bizan, S., & Lerner, R. M. (2010). The role of intentional self regulation, lower neighborhood assets, and activity involvement in youth developmental outcomes. Journal of Youth and Adolescence, 39, 783–800. Villarruel, F. A., Perkins, D. F., Borden, L. M., & Keith, J. G. (Eds.). (2003). Community youth development: Programs, policies, and practices. Thousand Oaks, CA: Sage. Zarrett, N., Fay, K., Carrano, J., Li, Y., Phelps, E., & Lerner, R. M. (2009). More than child’s play: Variable- and pattern-centered approaches for examining effects of sports participation on youth development. Developmental Psychology, 45, 368–382. Zimmerman, S., Phelps, E., & Lerner, R. M. (2008). Positive and negative developmental trajectories in U.S. adolescents: Where the PYD perspective meets the deficit model. Research in Human Development, 5,153–165.

3  MENTORING AND PREVENTION SCIENCE Timothy A. Cavell and L. Christian Elledge

Introduction This chapter considers youth mentoring through the lens of prevention science. The notion that mentoring is a form of prevention is held tacitly but seldom explicated or systematically examined. We review basic terms and tenets of prevention science and provide examples related to the field of youth mentoring. We consider the challenge of advancing youth mentoring as an evidenced-based prevention strategy and suggest that a broader and more elastic definition of mentoring is needed to meet that challenge. We propose recasting the mentoring relationship as a context for delivering prevention-oriented activities and experiences and not as the essential mechanism of change. We argue that this definition captures more accurately the evolving nature of youth mentoring and offers greater flexibility when designing new mentoring programs.

Prevention Science: An Overview What Is Prevention Science? The term prevention is used here to refer to practices designed to reduce the likelihood that developmental vulnerabilities or negative health conditions will compromise the health and wellbeing of developing youth. Models of prevention can be applied to a wide range of conditions, including those that disrupt educational or occupational attainment and those that involve antisocial behavior. Prevention science has been defined as the identification and prevalence of disorder, the study of etiology and human development, the design of

carefully controlled intervention trials addressing causal and mediating factors, the design and planning of field trials that attempt to address fully the essential process underlying disorder, and the implementation and evaluation of tested prevention programs in the community. (Coie, MillerJohnson, & Bagwell, 2000, p. 96) Adhering to the tenets of prevention science (Coie et al., 2000; Coie et al., 1993) and aspiring to its standards of evidence (Flay et al., 2005) offers the best avenue for advancing youth mentoring as an evidence-based intervention for youth at risk. Prevention is now playing a larger role in matters of public health, education, and child welfare, due in part to rising costs and the inefficiency of traditional service models (Coie et al., 2000). The cost of health care in the United States was 17.6% of the nation’s gross domestic product in 2009 and is expected to grow at an average rate of 6.1% per year over the next decade (Centers for Medicare and Medicaid Services, 2009). One strategy for managing spiraling costs is to adopt a posture of prevention that seeks to reduce the incidence of negative developmental outcomes and promote healthy, prosocial development (Munoz, Mrazek, & Haggerty, 1996). The fundamental goal of prevention is eliminating the cause of dysfunction or deflecting the processes that give rise to dysfunction (Coie et al., 1993). This latter point is important as many conditions are multiply determined and cannot be traced to a single cause (Cicchetti & Cohen, 1995). Prevention researchers assume that (a) causal or maintaining factors are identifiable, (b) some factors are malleable and can serve as targets of prevention, and (3) eliminating causal factors or altering maintenance processes can alter the trajectory of developmental outcomes (Coie et al., 29

30  FRAMEWORKS AND FOUNDATIONS 2000; Coie et al., 1993; Kazdin, Kraemer, Kessler, Kupfer, & Offord, 1997). What Are Key Terms and Concepts in Prevention Science? The Institute of Medicine has identified three subcategories of prevention: universal, selective, and indicated (Mrazek & Haggerty, 1994). Universal prevention is provided to all members of a population, regardless of their health or risk status. Selective prevention focuses on population subgroups that evince certain environmental (e.g., poverty) or personal (e.g., depression) characteristics that increase their risk for specific conditions. Indicated prevention is for individuals showing signs of specific disorders or negative outcomes but who do not yet meet full criteria (Coie et al., 2000). The concept and practice of prevention are not new, but the science of prevention is an emerging discipline. In addition to regarding these distinctions as important, prevention science assumes that research on risk and protective factors that affect the development or health-related behavior of targeted youth offers a useful foundation for designing potentially effective prevention programs (Chassin, Presson, & Sherman, 1985). A risk factor is any characteristic, internal (e.g., biological, genetic) or external (e.g., family, peer group), that precedes the onset of a negative outcome and that is associated with an increased probability of that outcome (Coie et al., 2000; Mrazek & Haggerty, 1994). Protective factors are characteristics that buffer individuals from negative outcomes or increase the likelihood of adaptive development (Coie et al., 2000; Kazdin et al., 1997; Mrazek & Haggerty, 1994). Of particular interest are modifiable risk and protective factors; prevention is not possible unless factors that predict negative outcomes are amenable to change (Coie et al., 1993). Risk factors often have complex relations to health-related outcomes (Coie et al., 1993; Kazdin et al., 1997; Rutter, 2009). It is rare to find health outcomes predicted by a single risk indicator. More common is for multiple risk factors to predate and contribute to a given negative outcome (Coie et al., 2000; Coie et al., 1993). The term equifinality refers to situations in which individuals may have divergent risk histories but nonetheless share susceptibility to a common negative outcome. It is also rare for risk factors to show specificity in predicting negative outcomes; instead risk factors are typically generic in nature, placing individuals at risk for diverse negative outcomes. The tendency for myriad outcomes to arise from a single risk or protective factor is referred to as multifinality (Cicchetti &

Cohen, 1995; Coie et al., 2000). The extent to which a risk factor contributes to negative outcomes can also fluctuate over time. Some risk and protective factors are developmentally bound, which means their impact occurs only during a particular period in the life span (e.g., infancy). As youth accumulate additional risk factors, the odds of incurring a negative outcome may accelerate rather than increase linearly. Illustratively, moving from three risk factors to four may be associated with a greater increased susceptibility to a negative outcome than moving from, say, one risk factor to two. In many instances negative outcomes are likely to be overdetermined, shaped by recurring, multiple, and overlapping factors, all of which makes prevention very difficult. Consider, for example, factors implicated in the etiology and maintenance of conduct disorder, a diagnosis often given to delinquent youth (American Psychiatric Association, 2000). Contributing factors include biogenetic variables (e.g., selfregulation); conflict with parents and teachers; influences from school, community, and culture; behavioral contingencies that govern specific contexts; and cognitive and emotional processes that affect social development (see Dodge & Pettit, 2003). The prevention research cycle offers a useful framework for understanding the goals and strategies of prevention science (DuBois, Doolittle, Yates, Silverthorn, & Tebes, 2006; Mrazek & Haggerty, 1994). The cycle involves a systematic progression from very basic studies on the epidemiology or etiology of negative outcomes to studies that test whether preventive interventions are effectively disseminated and adopted. In a discussion focused on applications to mentoring, DuBois et al. (2006) drew on a framework from the National Institute of Mental Health that distinguishes three phases in the research cycle. Pre-intervention research addresses basic questions about the prevalence, origins, and course of negative developmental outcomes. Also examined here is the role of specific risk and protective factors, including predictive utility, their role in the onset or maintenance of specific outcomes, and the likely period of influence during development. Preintervention studies can also address whether risk and protective factors are modifiable. Sorely needed are studies that test whether mentoring interventions reliably affect change in targeted risk or protective factors. For example, if academic self-efficacy was a robust predictor of academic achievement in disadvantaged youth, pre-intervention researchers might examine, under tightly controlled conditions, strategies mentors could use to enhance youths’ sense of academic efficacy (Hansen & Wänke, 2009). Findings from such studies could then be

Mentoring and Prevention Science   31 used to develop and test mentoring interventions designed to promote the academic self-efficacy of youth at risk for academic failure. Preventive intervention research addresses whether mentoring works, for whom, and for what reasons. The typical sequence is from efficacy trials conducted under highly controlled conditions to effectiveness trials that take place in real-word settings (Flay, 1986). An efficacious intervention is one that leads to reliable and meaningful differences in measured outcomes under optimal conditions (e.g., ample resources, well-trained staff). Effectiveness trials, on the other hand, examine outcomes in applied service contexts where there may be fewer dedicated resources to support implementation of the intervention and more competing demands on staff. Both efficacy and effectiveness trials can be used to address two other key questions: How does the intervention work, and for whom is it effective? Studies examining the active ingredients of an intervention are termed mediational studies because the goal is to test whether hypothesized mechanisms account for the intervention outcomes. For example, Rhodes, Grossman, and Resch (2000) found evidence that positive outcomes for youth in community-based mentoring matches were partially mediated through improved relationships with parents. The notion that mentoring exerts influence through improved parental relationships was not an a priori hypothesis built into the design of the mentoring program. Thus, from a prevention science perspective, it is important to replicate this finding and to conduct research clarifying more precisely how and to what extent mentoring leads to improved parent-child relationships. The question of who benefits or is harmed from an intervention is answered through moderator studies. The term moderator refers to the possibility that intervention outcomes differ depending on the population targeted. For example, Karcher (2008) found that school-based mentoring was effective for elementary school boys and high school girls but ineffective for older boys and younger girls. Prevention service systems research focuses on strategies for and obstacles to disseminating prevention programs that have documented efficacy and effectiveness. An important challenge is how evidence-based prevention practices can be promoted so that there is widespread adoption, utilization, and sustainability on the “front lines.” Successful dissemination of evidence-based prevention programs requires careful attention to site selection (readiness and capacity to implement the intervention), training (budget, personnel, and strategies), technical assistance (support and supervision), implementation fidelity (monitoring adherence to the implementation model,

intervention integrity and quality), scalability (easily adapted for wider distribution), and sustainability (for a review, see Elliot & Mihalic, 2004). Recently, the Society for Prevention Research adopted standards to guide evaluation of evidence from studies of preventive interventions (Flay et al., 2005). The standards set a high threshold for establishing the efficacy and effectiveness of preventive interventions such as youth mentoring. Efficacious interventions are those tested in at least two independent trials using a number of methodological controls (e.g., rigorous research design, clearly defined samples, psychometrically sound measures, appropriate statistical approaches) and yielding favorable outcomes that are reliable and sustained. Effective interventions meet those criteria and (a) offer intervention manuals, training, and technical support, (b) are tested under real-world conditions with adequate dosage and quality of implementation, and (c) demonstrate meaningful outcomes (i.e., the outcomes have practical implications). Interventions are ready for broad dissemination when all of these standards are met and the intervention is scalable, has clear information on costs, and can be monitored and evaluated once implemented. Later, we revisit the topic of the prevention research cycle and consider how recent mentoring investigations fit within that cycle. What Are the Core Tenets of Prevention Science? In this section we review the basic tenets of prevention science as described by Coie and colleagues (Coie et al., 2000; Coie et al., 1993). These seven tenets can guide researchers and practitioners interested in advancing youth mentoring as an evidence-based prevention strategy. Tenet 1: Prevention Begins With Current Knowledge of Fundamental Causal Processes. Successful prevention of negative developmental outcomes requires a thorough understanding of the risk and protective factors that affect those outcomes and conditions (Chassin et al., 1985). Without that understanding, those who develop prevention programs cannot adequately specify the risk or protective factors targeted by the intervention. Also needed is a conceptual or theoretical basis for expecting the intervention to lead to changes in specific risk or protective factors. For example, what contextual factors or developmental processes are implicated, and how precisely will the intervention lead to changes in those factors or process outcomes? A good example of how this tenet applies to youth mentoring is a program designed specifically for youth in foster care (Taussig,

32  FRAMEWORKS AND FOUNDATIONS Culhane, & Hettleman, 2007). Taussig and colleagues framed their underlying model as one in which maltreatment and foster care interact with “attachment, temperament, physiological, and neurological variables to result in psychological, social, and behavioral functioning that often deviates from normative development. Impairment in these areas is hypothesized to lead to mental health problems, risk behaviors, low levels of competence, and poor quality of life” (p. 115). Basic research on factors that contribute to negative outcomes is the necessary first step in designing effective prevention programs, but because of its unique history (see below), youth mentoring bypassed this step. It could be said there is little need to specify an underlying model of risk if mentoring yields multiple benefits to a wide range of youth (for evidence of such benefits, see DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). A counter-argument, though, is that youth mentoring will continue to yield outcomes of modest magnitude (see, again, DuBois et al., 2011) unless greater attention is given to this beginning step of model specification. Tenet 2: Prevention Models Should Include Dynamic Developmental Processes, Multiple Ecological Systems, and an Appreciation for the Reduction of Risk and the Promotion of Positive Youth Development. Developmental processes that lead to negative outcomes are typically bidirectional, dynamic, and influenced by individual, social, and environmental factors (Cicchetti & Cohen, 1995; Coie et al., 2000; Coie et al., 1993; Kazdin et al., 1997). Some youth experience a protracted cascade of negative processes (Burt, Obradović, Long, & Masten, 2008; Dodge et al., 2008), including impaired self-regulation that hinders emotion regulation, interpersonal competence, and academic achievement; and environments that are economically disadvantaged, emotionally harsh, or ridden with crime and violence. For these youth, risk factors can accumulate and compound with each passing year. For example, coercive interactions with family members can lead to troubled behavior with peers and teachers, which can occasion academic failure and a tendency to associate with deviant peers, which sets the stage for teen years marked by risky behavior, substance use, or delinquency. Over time, this trajectory becomes increasingly pernicious and difficult to change (Dodge et al., 2008). Prevention models that recognize the complex nature of developmental processes should have greater success at reducing or delaying negative outcomes than those that focus on single indicators of risk or resilience. This does not mean that preventive interventions

must target all factors thought to contribute to negative outcomes; rather, it implies that it is important to consider when, how, and to what extent intervention will alter the processes targeted. Consideration should also be given to whether interventions can override factors not targeted. Prevention models also need to recognize the contribution of various ecological systems to the onset and maintenance of negative outcomes. Some risk and protective factors are linked to specific contexts or social systems (e.g., harsh parenting), and the influence of these processes can fluctuate depending on the developmental period (Coie et al., 1993). Interventions that target a single system at one time point can have limited impact if other systems are unchanged or if the system is not salient for that developmental period (Cicchetti & Cohen, 1995). Ideally, prevention programs address concerns across systems and with an appreciation for the sequencing of risk and protective factors. Applied to youth mentoring, this could mean that mentoring is but one component in a multifaceted intervention (Kuperminc et al., 2005) or that mentoring needs to change its focus, setting, or structure depending on the age, gender, and social and emotional needs of targeted youth (e.g., Karcher, 2008). Tenet 3: Prevention Efforts Specify the Population at Risk and the Methods for Identifying and Reaching Members of That Population. Prevention models make explicit the population for whom intervention is intended. Universal prevention programs target entire populations, which simplifies the process of identifying participating youth but means that resources are expended for some youth not at risk. Selective and indicated prevention programs require procedures for accurately identifying those youth who warrant intervention (Springer & Phillips, 2006). For example, a recent study of school-based mentoring as a selective prevention for bullied children relied on selfand teacher-report scales to identify children to be mentored (Elledge, Cavell, Ogle, & Newgent, 2010). Some interventions carry the risk of harm, so there is the added task of identifying youth for whom interventions are a poor fit (e.g., Karcher, 2008). Potential participants who are a good fit need to be recruited. Innovative examples of this approach include mentoring programs that recruit students enrolled in specialized schools (e.g., Hanlon, Simon, O'Grady, Carswell, & Callaman, 2009; Shpigelman, Reiter, & Weiss, 2009) and programs based in child mental health centers (Jent & Niec, 2006). Tenet 4: When Possible, Prevention Addresses Risk Factors Before They Stabilize as Predictors

Mentoring and Prevention Science   33 of Dysfunction and Become Less Amenable to Change. Some risk factors predict dysfunction throughout the lifespan; other risk factors are developmentally bound, conferring risk only during a specific developmental period (e.g., infancy). Ideally, prevention programs target risk factors before they stabilize or generalize to other contexts such as home and school (Coie et al., 2000; Coie et al., 1993). For example, Smith (2011) used high school mentors to promote healthy eating among third and fourth graders—before the onset of ­teenage obesity. Tenet 5: Effective Prevention Targets Malleable Risk and Protective Factors Implicated in the Model of Risk. A further assumption of prevention science is that effective prevention targets those factors in a model of risk that are amenable to change. From this perspective, gender, ethnicity, geographic location, and socioeconomic status might predict the onset or course of negative outcomes but may not be suitable targets for psychosocial interventions. The term fixed markers is sometimes used to describe factors associated with dysfunction but not considered malleable (Kazdin et al., 1997). Such markers are important as they stimulate the search for intervening variables linking the marker with important outcomes, thus illuminating targets for intervention. A good example is the marker of parental divorce; mentoring is unlikely to help parents’ marital difficulties but could buffer children from the risks that come with post-divorce conflict. Tenet 6: When Possible, Prevention Focuses on “Generic” Risk and Protective Factors. When possible, prevention science advocates that prevention programmers should take advantage of multifinality, the process whereby several negative developmental outcomes share a common causal pathway. Prevention efforts that effectively target generic risk or protective factors (i.e., factors that predict a diverse set of outcomes) could, in theory, ameliorate a number of negative outcomes simultaneously. When one considers the range of youth who are mentored or the evidence that benefits of mentoring interventions often extend across multiple domains of outcomes, it seems clear that youth mentoring has the potential to address generic risk and protective factors (DuBois et al., 2011). The process of equifinality, by contrast, challenges prevention programmers. If multiple pathways lead to a single negative developmental outcome, then separate and distinct prevention efforts might be needed for each pathway. Consider youth with conduct disorder. For some, symptoms begin in

childhood and persist into adulthood; for others, conduct disorder is limited to adolescence. Despite the common label, the first group is at much greater risk for maladjustment (Moffitt, 1993), and, therefore, mentoring effective for one group might be ineffective for the other group. Tenet 7: Prevention Trials Are Guided Initially by Developmental Theory but Yield Results That Inform and Revise That Theory. Prevention trials are guided by basic research on the epidemiology and etiology of various negative developmental outcomes. But prevention trials also test models of risk through interventions designed to influence proposed causal mechanisms. If interventions successfully change the mechanisms but the risk trajectory remains unchanged, the status of the causal mechanisms is questioned and the underlying theory is revised. When interventions alter causal mechanisms and reduce the incidence of negative outcomes, researchers have greater confidence in their underlying model. So, for example, if it is theorized that poor academic performance is due (at least in significant part) to a lack of school belonging, then a school-based mentoring intervention that enhances school belonging offers a valuable test of this underlying model.

Youth Mentoring as an Evidence-Based Prevention Tool Developments in Youth Mentoring: A Prevention Science Perspective Our premise is that youth mentoring is a promising, but underdeveloped form of prevention. One step in advancing the science and practice of mentoring is to appreciate the reasons it may be underdeveloped. Historically, the goal of youth mentoring was that of charity—doing good for those who were less fortunate. Charitable organizations viewed it as worthwhile to provide adult role models to disadvantaged youth thought to have none (Baker & Maguire, 2005). The idea that adult mentors would help (and not harm) disadvantaged youth was considered self-evident, with the presumption that interactions between mentors and mentees would help youth become law-abiding citizens. It is interesting to consider this notion in light of what is known today about (a) the complex and multiply determined nature of child antisocial behavior (Cavell, Hymel, Malcolm, & Seay, 2007; Dodge et al., 2008) as well as (b) the paradoxical harm that can come from well-intentioned youth intervention programs

34  FRAMEWORKS AND FOUNDATIONS (Dishion, McCord, & Poulin, 1999), including those involving mentoring (Grossman & Rhodes, 2002; McCord, 2003). The goal of providing positive role models to at-risk youth was not without scholarly support. Classic theorists such as Durkheim, Weber, and Rousseau wrote about the costs of societies lacking control over, opportunities for, and connections to its more vulnerable members. A more contemporary argument for youth mentoring is Hirschi’s (1969) social control theory, which posited that delinquency is curtailed when youth have strong connections to prosocial individuals and institutions. Early studies examining the interplay between risk and resilience (Garmezy & Masten, 1986; Werner, 1995) added further to the foundational assumptions of youth mentoring. Especially appealing was Werner’s (1986) work suggesting that the presence of at least one supportive adult was a critical feature in the lives of resilient youth. In 1995, the field took a leap forward with the release of the Public/Private Ventures (P/PV) randomized controlled study of the Big Brothers Big Sisters (BBBS) community-based mentoring program (Tierney, Grossman, & Resch, 1995). At the time, there was little recognized scientific evidence that mentoring benefited youth or that funding to support mentoring was money well spent (Boyle, 2007). P/PV’s study yielded supportive findings that became the source of oft-repeated claims for the “proven” effects of youth mentoring (e.g., mentored youth were 46% less likely to start using drugs, 27% less likely to start using alcohol, and 33% less likely to hit someone). Tierney et al. (1995) described the findings as “surprisingly robust” (p. 31) and called for expanding and increased funding of mentoring programs. Grossman and Tierney (1998), in the version of the study published in a peer-reviewed journal, characterized the findings as “solid evidence” that mentoring “has many positive and socially important effects on the lives of its young participants” (p. 403). Not surprisingly, the “impact study,” as it is often called, had its own impact, enhancing public perceptions of mentoring, accelerating growth in mentoring, and aiding efforts to increase funding for and government involvement in mentoring (Rhodes & DuBois, 2006). Less attention was given to Tierney and colleagues’ (1995) emphasis on practice guidelines used by BBBS agencies or concern over potential harm from mentoring (Rhodes, Liang, & Spencer, 2009). Enthusiasm generated by the impact study led to repeated calls to close “the mentoring gap”— said to be the difference between the number of youth who benefit from formal mentoring and the number still in need of a mentor, numbers that

recently were estimated at 3 million and 15 million, respectively (see MENTOR, 2006b). Efforts to close this gap used findings from the P/PV study to make bold claims such as the following: “Youth development experts now agree that mentoring is a critical element in any child’s social, emotional, and cognitive development” (MENTOR, 2006a, p. 1). The policy-driven nature of the P/PV impact study launched a period when, as assessed by many observers, the practice of youth mentoring rapidly outpaced its empirical support (Boyle, 2007). Rhodes and DuBois (2006), for example, noted that findings from this single study were used to make important policy and funding decisions, even though findings were actually “modest and somewhat nuanced” (p. 9). Illustratively, they noted that the size of significant impacts were relatively small, that there were no significant effects on many important outcomes (e.g., stealing, smoking), and that there were several demographic subgroup differences in the findings (Rhodes & DuBois, 2006). Also important, modest impacts from youth mentoring were also reported by DuBois and colleagues (DuBois, Holloway, Valentine, & Cooper, 2002) in their landmark meta-analysis. Commenting on the policy implications of extant research more than a decade after the impact study, Rhodes and DuBois (2006) called—not for a fivefold increase in mentor matches—but “for careful adherence to evidencebased practice and measured expansion of new program models” (p. 9). There are signs that some organizations such as BBBS have followed suit, emphasizing the quality of its matches over the number of matches (see Big Brothers Big Sisters of America, n.d.). The idea that youth mentoring is underdeveloped was made obvious with the publication of the U.S. Department of Education’s report on the randomized trial of its Student Mentoring Program (SMP; Bernstein, Rappaport, Olsho, Hunt, & Levin, 2009), an initiative that represented a major effort to grow school-based mentoring. Investigators reached the sobering conclusion that SMP “did not lead to statistically significant impacts” (p. xx) on academics, interpersonal relationships, or high-risk behavior. This study was then cited in the decision to cut Department of Education funding for school-based mentoring (Boyle, 2009; Wheeler, Keller, & DuBois, 2010). The SMP study should be considered in light of two other, large-scale school-based mentoring studies (e.g., Herrera, Grossman, Kauh, Feldman, & McMaken, 2007; Karcher, 2008). Wheeler et al. (2010) argued that all three studies yielded modest effects on truancy and school misconduct but less impact on academics and other

Mentoring and Prevention Science   35 areas of adjustment (for a more in-depth review of recent research on school-based mentoring, see Herrera and Karcher, this volume, Chapter 14). In many ways, the SMP study was a perfect storm of misguided policy-driven trends and program limitations (Boyle, 2009). It could be argued that it was less a test of school-based mentoring and more a test of a funding initiative borne of No Child Left Behind legislation, unwarranted optimism following the P/PV impact study, and recurring pleas to close the mentoring gap (Boyle, 2007; Rhodes & DuBois, 2006; Wheeler et al., 2010). To grow youth mentoring, the field emphasized school-based mentoring and its perceived practical advantages. But there was “no clear road map for how to scale up this intervention” (Rhodes, 2008, p., 41), a point perhaps reflected in findings of the SMP evaluation. Instead, the presumption appears to have been that mentoring has proven effects. Little attention was given to who was mentored, for how long, with what support, or for what purposes (Rhodes, 2008). From the standpoint of prevention science and the research cycle discussed previously in this chapter, we would argue that the SMP study is best regarded as a poorly conducted effectiveness trial (DuBois et al., 2006) conducted long before the field was ready and before we knew enough about how mentoring works and for whom. Youth mentoring is now in direct competition with other youth-serving programs for public sentiment and financial support (Boyle, 2009; Durlak, Weissberg, & Pachan, 2010; Rhodes, 2008). Other youth-serving programs, especially school-based interventions, have been in the prevention science “pipeline” for over a quarter-century (Botvin & Griffin, 2010; Coie et al., 2000). To compete, mentoring will need to establish itself as a viable, evidence-based prevention strategy and, in our view and that of others (e.g., Rhodes, 2008), that can happen only if the field earnestly embraces the difficult work of prevention science: If we are to champion this intervention strategy, we must be prepared to grapple with its complexities— even at the risk of learning that commonly deployed programs and practices do not always improve youth outcomes. To this end, prevention researchers have a central role to play. (Rhodes, 2008, p. 41) If the Goal Is Prevention, How Should We Define Youth Mentoring? If youth mentoring is to be an evidence-based prevention strategy, what are key obstacles and next

steps for researchers and practitioners? In our view, a critical issue is how to define mentoring. Despite its long history and wide appeal, there are questions about the essential features of mentoring (Cavell, Elledge, Malcolm, Faith, & Hughes, 2009) and debate about whether emerging programs are doing mentoring (Rhodes & DuBois, 2008). Absent a workable definition, we cannot realize the potential of youth mentoring and we will miss opportunities to promote mentoring as an evidence-based prevention tool (Cavell et al., 2009; Cavell & Smith, 2005). At the center of this definitional debate is the role of the relationship. Mentoring-as-Relationship. It is not uncommon for scholars to view the relationship as the sine qua non of youth mentoring: If no relationship, then no mentoring (e.g., Rhodes, 2005). Indeed, scholars sometimes refer, not to mentoring or to mentoring programs, but to mentoring relationships when naming this form of intervention (e.g., Rhodes & DuBois, 2008). A related assumption is that the quality of the relationship—its strength and duration— determines if youth actually benefit from mentoring experiences (Parra, DuBois, Neville, Pugh-Lilly, & Povinelli, 2002; Rhodes, Spencer, Keller, Liang, & Noam, 2006). A common assumption is that mentoring is not effective without strong, close matches: “only then may youth derive significant benefits” (Rhodes & DuBois, 2008, p. 255). Theoretical models explicitly endorse a mentoring-as-relationship perspective, emphasizing the importance of one-onone, face-to-face visits that lead to mutuality, empathy, and trust (Rhodes & DuBois, 2008; Rhodes et al., 2006). Newer adaptations (e.g., e-mentoring) that diverge from this ideal are looked at askance: “mentoring programs have moved in a direction that is in danger of trivializing what research indicates is at the very heart of their intervention: a caring adult–youth relationship” (Rhodes & DuBois, 2008, p. 257). Theoretical models based on the mentoring-as-relationship perspective seldom view mentoring as the sole mechanism of change (e.g., Rhodes, 2005). More common is to view the mentormentee bond as catalyzing other processes that occur “downstream” from the mentoring relationship. These other pathways include relationship-induced growth in social or cognitive development and gains in youths’ connections with significant others, such as parents (Rhodes et al., 2000; Rhodes et al., 2006). Still, the relationship is considered a prerequisite. The central role of the relationship was evident in the definition offered by DuBois et al. (2011) in their recent meta-analysis: “A program or intervention that is intended to promote positive youth outcomes

36  FRAMEWORKS AND FOUNDATIONS via relationships between young persons (18-yearsold and younger) and specific nonparental adults (or older youth) who are acting in a nonprofessional helping capacity” (p. 66). Interestingly, this metaanalysis did not identify factors related to match strength (e.g., emotional support, strong bond) or duration (e.g., overall length) as significant moderators of program effectiveness. This could be due to reduced variability in the quality of mentoring relationships as programs implement best practices: If all mentoring relationships are positive, there can be no statistical relation with youth outcomes. DuBois and colleagues (2011) hinted at this possibility: “Unevenness in the quality of mentoring programs may have become less pronounced” (p. 76). But they also noted that “translation of available evidence into policy and practice has been a slow and still-evolving process” (p. 75). There are studies examining the link between mentor relationship quality and positive youth outcomes (e.g., Rhodes, 2008), but underappreciated, it seems, is the fact that published findings tend to paint a nuanced or qualified picture of this link (Cavell & Smith, 2005). We see two trends in the findings. First, conflict and early termination— indicators of negative or poor relationships—are often better predictors of youth outcomes than are positive relationship indicators (e.g., Cavell et al., 2009; Rhodes, Reddy, Roffman, & Grossman, 2005). Second, findings that support the link between relationship quality and youth outcomes are based almost exclusively on correlational data. The first trend has important implications for how programs assess and promote match quality (see Spencer, 2007), but the second is a serious challenge to claims that better match quality causes better youth outcomes. Testing the causal link between match quality and youth outcomes would require manipulating the strength or length of the mentoring match in the absence of other potentially confounding manipulations and randomly assigning youth to different strength or length conditions (DuBois et al., 2006). To date, to our knowledge, this sort of study has never been done. Researchers have compared groups who experienced different levels of match quality or match length (e.g., Grossman & Rhodes, 2002; Grossman, Chan, Schwartz, & Rhodes, 2012), but groups were formed after the fact and match quality or length was not manipulated. Grossman and Rhodes (2002) used this approach and found that matches lasting 6 months or less had more negative outcomes compared to matches lasting a year or more. A strong effort was made to control for potential statistical confounds using instrumental variables analysis, but match

length was not manipulated, so one cannot infer a causal link between match length and youth outcomes with the level of confidence that would be possible with such a design. Among the competing hypotheses for the findings is one suggested by Grossman and Rhodes: Mentees in longer matches “were particularly well-adjusted” (p. 208). Thus better outcomes could have more to do with the social competence of mentees than the quality of their mentoring relationships. In a study that manipulated several program variables, including but not limited to those expected to influence mentoring relationship quality and duration (Cavell et al., 2009; Hughes, Cavell, & Meehan, 2001; Hughes, Cavell, Meehan, Zhang, & Collie, 2005), highly aggressive children were randomly assigned to a multicomponent, communitybased mentoring program or to a stand-alone, school-based mentoring program. The latter was meant to serve as a control condition; it was designed intentionally to diminish the strength and length of the mentoring relationship. Both programs lasted for three semesters (spring, fall, spring), but children in the school-based program were paired with a different mentor each semester, mentors were minimally trained, all visits were during school lunch, and mentors sat with mentees and their lunch mates. In the community-based program, children were mentored by the same volunteer all three semesters and mentors were extensively trained and supervised. Mentors in both programs were college students earning course credit, so very few matches ended early. As expected, children in the schoolbased program rated match quality less positively than did children in the community-based program (Cavell et al., 2009). Unexpectedly, teacher-rated outcomes at 1- and 2-year follow-ups revealed that children in three, shorter school-based matches had more favorable outcomes than children in the ­longer, more positive matches (Hughes et al., 2001). These findings were not predicted a priori and there were other, key differences between the two programs (see Cavell et al., 2009), so this study is a poor test of match strength and match length. The fact remains, however, that methodologically sound, experimental investigations of relationship quality in youth mentoring are sorely needed. Until researchers find more compelling evidence for a causal relation between relationship quality and key outcomes, the mentoring-as-relationship perspective is a working hypothesis or guiding notion about how mentoring might work. As such, we believe there is both practical and scientific value in considering alternative hypotheses to the question of how mentoring works.

Mentoring and Prevention Science   37 Mentoring-as-Context. We propose an alternative (but not mutually exclusive) perspective that challenges commonly held notions about youth mentoring. We hypothesize that mentoring is best viewed, not as a relationship, but as a context created to benefit youth (in the case of formal mentoring). We offer the following working definition: Youth mentoring is the practice of using program-sponsored relationships between identified youth and older volunteers (or paraprofessionals) as a context for prevention-focused activities and experiences. This definition retains the need for some kind of mentormentee relationship but a close mentor-mentee bond is not an essential feature. Rather, youth mentoring is a context for providing activities or experiences that serve specific prevention goals. The mentoring relationship could be strong and lasting, but it could also be limited in strength and length. Interestingly, the definition proposed by DuBois et al. (2011) would also satisfy these requirements but does not make explicit an emphasis on prevention-focused activities and experiences. Some mentoring programs might promote matches in which the central experience is a strong lasting bond; others might pursue shorter matches in which the emphasis is on engaging activities that enhance reading or math skills. From a mentoring-as-context perspective, both strategies could be “effective” but the outcomes would differ, as would the pathways and mechanisms of change. The difference between mentoring as a context and mentoring as a relationship is akin to the distinction between mentoring relationships and mentoring programs. DuBois et al. (2011) noted this when presenting a theoretical model of mentoring relationships, which they took care to point out meant the potential implications for effective programs must be inferred. A similar distinction could be made between mentoring matches and mentoring relationships, based on the assumption that not all matches will meet a certain threshold of relationship quality. A mentoring-as-context perspective offers several potential advantages to researchers and practitioners striving to advance youth mentoring as evidence-based prevention. The most obvious of these is that program success and failure is not assumed to be tied to the quality of the relationship. Both mentees (Schwartz, Rhodes, Chan, & Herrera, 2011) and mentors (Faith, Fiala, Cavell, & Hughes, 2011) bring to the match a range of relationship tendencies and talents, and available data (e.g., Grossman & Rhodes, 2002) indicate that a nontrivial proportion of matches will not be supportive or lasting. It seems wise to design mentoring programs that accommodate these less-than-ideal relationships. A mentoring-as-context

perspective can better accommodate recent adaptations in mentoring while retaining the basic requirement of a relationship context. Programs that claim to provide mentoring or that label as “mentors” adults who deliver an intervention (e.g., Wyman et al., 2010) would have to show that what is delivered occurs in the context of a definable relationship in which youth choose to participate, interactions are recurring and reciprocal, and match length and frequency of contact are not trivial. A mentoring-as-context perspective allows for wide variability in how mentoring programs are designed and the purposes they serve. Some programs could be curriculum driven, short term, and site based; other programs could be nonprescriptive, long term, and community based. Programs could be molded to fit a certain population and structured with specific goals in mind. Mentoring could be used to promote school bonding, to reduce teen substance abuse, to help children of divorced parents, to increase peer acceptance of children with Asperger syndrome, or to help young adults transition out of foster care. Mentoring could occur face-to-face or via the internet, mentors could be senior citizens or seniors in high school, and matches could last 5 years or 5 months. In short, programs could be vastly different and yet still constitute mentoring and prove to be beneficial to the youth served. When a close, lasting bond is no longer a prerequisite, programs have the flexibility to adapt mentoring to meet diverse goals and achieve important prevention goals. A context-oriented view of mentoring allows for the infusion of greater intentionality, explicit goals, and definable structure. The field has struggled with how to foster mentoring relationships that are not overly structured and prescriptive nor completely lacking structure and definition (Morrow & Styles, 1995). Recent efforts to reconcile this apparent dichotomy (Karcher & Nakkula, 2010) are in line with a mentoring-as-context perspective. Structure and prescription are neither inherently good nor bad; rather, both are to be examined against the stated purpose and goals of the program. Finally, a contextual view of mentoring does not preclude the unpredictable, spontaneous aspects of mentor-mentee relationships. Sometimes the unexpected is negative and damaging, but most mentoring programs witness positive, growth-promoting moments. Some BBBS agencies refer to the “magic” of mentoring, which captures well the challenge of trying to predict the course, nature, and outcomes of a given mentor-mentee match. Dropping the requirement of a close, lasting bond does not mean losing that magic; it simply recognizes that such moments cannot be prescribed. Mentor-mentee

38  FRAMEWORKS AND FOUNDATIONS matches vary in the relationships they form and the experiences they have, but even mentoring that is highly structured and short term can have its share of magical moments and productive relationships (Cavell et al., 2009). Prevention Science and the Future of Youth Mentoring Research on Youth Mentoring. The science of youth mentoring in our estimation is lacking when examined in light of the tenets of prevention science and the prevention research cycle (DuBois et al., 2006). Research is ongoing but still limited in its volume, its theoretical grounding, and its methodological rigor. Lacking are prevention trials based on wellspecified theoretical models, that rely on careful assessment using multiple sources and methods, that employ randomized controlled designs, and that test for potential mediators and moderators of outcomes. One of the more glaring gaps is the lack of research on long-term outcomes (DuBois et al., 2011), especially given that prevention sometimes produces delayed effects that emerge later in time (Coie et al., 1993). Investigations have typically emphasized the latter stages of the research cycle. Known as Type 2 translational research investigations, these studies focus on real-world impacts and program implementation, adoption, and sustainability (Elliot & Mihalic, 2004; O’Connell, Boat, & Warner, 2009). With the prevention science paradigm, Type 2 studies are designed for programs that have strongly documented efficacy and support for the explicit model of change. Few, if any, youth mentoring programs currently meet that threshold (DuBois et al., 2011). Even mentoring programs listed as “blue print” or “model” programs (Rhodes & DuBois, 2008) lack the level of empirical support needed to meet the rigorous standards of evidence offered by Flay and colleagues (Flay et al., 2005). Particularly lacking are investigations that fit the early stages of the research cycle. Such studies, referred to as Type 1 translational research, have the goals of transferring basic science into clinical research and using findings from intervention research to inform basic science (O’Connell et al., 2009). Type 1 research is critical in the framework of prevention science (Coie et al., 1993) and, in our view, a necessary pathway to performing what we term the other “match” in youth mentoring. The term match usually refers to the quality or length of the mentoring relationship or to the fit between mentor and mentee characteristics or interests. But this is a different kind of mentoring match. It is the match between the underlying model of risk or

resilience for a targeted population and the explicit model of change that drives and structures the mentoring program (Cavell & Smith, 2005; Chassin et al., 1985; Hughes, 2000). If a strong mentoring relationship is considered a mechanism of change, then the parameters of those relationships and the factors that promote or detract from those relationships have to be defined. Also needed are studies that identify specific processes by which mentoring relationships lead to positive outcomes (e.g., Rhodes et al., 2000). This other match is especially important when mentoring is viewed as a context for prevention-oriented activities and experiences. A good match would mean the specific goals and features of mentoring are closely linked to an empirically based model of the negative outcomes being prevented. What factors determine those outcomes and what mechanisms in mentoring will alter those factors? Do outcome studies support the change model, or is there a need to revise the change model in line with the underlying model of risk and resilience? The Practice of Youth Mentoring. If the arguments we are advancing were to receive wide acceptance, the field of youth mentoring would be a late arrival to the epistemology of prevention science, having relied on compelling testimonials and oft-repeated findings from P/PV’s impact study (Tierney et al., 1995) to make the case for mentoring. But there is tremendous promise and wide appeal associated with mentoring. The potential for evidence-based programs to be adopted, sustained, and replicated (Elliot & Mihalic, 2004) is significant given the level of support and enthusiasm that often surrounds youth mentoring. If the field can combine efforts to “grow mentoring” with specifying and improving mentoring, we see the stage as set for rapid dissemination and scalability. With notable exceptions (e.g., Spoth & Greenberg, 2011), most prevention programs lack the infrastructure and market appeal needed for wide dissemination, whereas youth mentoring is broadly dispersed and readily adopted. It stands ready to deliver a range of prevention-oriented experiences and activities. This singular advantage bodes well for the practice of mentoring. In discussing practice implications of findings from their recent meta-analysis, DuBois et al. (2011) identified three ways to increase the return on investments made in youth mentoring: (a) ensure adherence to core practices, (b) use the best evidence to strengthen and revise programs, and (c) foster stronger collaborations between practitioners and researchers. We would add a few of our own (see Table 3.1).

Mentoring and Prevention Science   39 Table 3.1   Checklist for Practitioners Topic

Recommendations

Epistemology

Embrace the epistemology of prevention science and accept the limits of youth mentoring as a preventive intervention, as currently practiced and researched.

Mentoring as context

Recognize that mentoring is a context for a potentially wide range of preventionfocused activities and experiences, including the experience of a close, lasting bond between mentor and mentee.

Specify the level of prevention

Specify the level of prevention (universal, selected, indicated), the population targeted, and the strategies for identifying that population.

Attend to the other “match”

Develop mentoring programs in which the hypothesized mechanisms of mentoring are a good match for the underlying model of risk/protection that affects the target population.

Measure proximal changes

Measure the impact of mentoring on proximal mechanisms of change as well as more distal outcomes.

We first recommend that practitioners embrace fully the epistemology of prevention science. Mentoring has many passionate champions, and thousands of youth have benefited from the support of these champions. But “evidence-based” is now the gold standard for funding decisions about youthserving programs, and programs that adhere closely to the tenets of prevention science will have the most compelling evidence. Therefore, youth mentoring needs champions who see the value of rigorous research and push hard for funding that supports a prevention science agenda. There remain serious questions about the merits of youth mentoring, and research is more than an exercise in “proving what we already know.” Practitioners should know and accept current limitations of youth mentoring as a prevention strategy. They can take pride in what has been accomplished, but pride should not cloud objectivity when judging the potential impact and likely shortfalls of mentoring. Practitioners should be especially cautious when making judgments about which youth “need” a mentor and which programs warrant continued funding and support. Available research (e.g., DuBois et al., 2011) should be used to make informed decisions about which youth and which programs are a good fit for this particular form of prevention. We encourage practitioners to view mentoring as a context for providing prevention-focused activities and experiences. A close, enduring relationship is the “holy grail” of youth mentoring, but the reality is that a significant proportion of matches will not meet that goal

(e.g., Grossman & Rhodes, 2002). A proactive response to that reality is to design mentoring programs that are not wholly dependent on the relationship capacities of mentors and mentees (e.g., Schwartz et al., 2011). These programs might be more limited in scope, more structured in their delivery, and briefer in duration, but such programs already exist and growing evidence supports their use. In this regard, DuBois et al. (2011) argued that several of the programs included in their recent meta-analysis are exemplars of the efforts needed “to scale up and disseminate mentoring interventions for youth without significantly compromising their quality or effectiveness” (p. 79). Yet, when one considers characteristics of the programs that they described as illustrative of those included in their review (see Table 3 of their review), the average expected match length was less than 9 months and most of the programs had narrow but well-specified goals. When mentoring is cast, first and foremost, as a relationship, it is possible to produce a wide range of benefits to a broad spectrum of youth, and findings reported by DuBois et al. (2011) support that possibility. But when mentoring is structured as a context for prevention-focused activities and experiences, the scope of potential benefits may narrow in line with the specific focus of the program. Practitioners might have to choose between the precision of pursuing narrow outcomes or the hope of affecting wide change via strong relationships. This speaks to the heart of mentor programming: What is the mission? Our next recommendation concerns the other match in youth mentoring. As discussed

40  FRAMEWORKS AND FOUNDATIONS ­previously, effective prevention programs are typically designed via a process in which factors that contribute to or limit negative outcomes are used to create promising interventions. Outcome studies are then used to test the underlying causal model and modify as needed the proposed intervention. From a prevention science perspective, the process of matching models of risk/protection with models of change is essential if practitioners are to enhance outcomes or develop new programs. This process assumes, of course, an adequate research base. We invite practitioners to press researchers to shed greater light on how mentoring works and for whom. Absent that knowledge, this other match in youth mentoring becomes mere guesswork. Finally, we encourage practitioners to routinely measure proximal changes in youth mentoring. If there is an orderly process to how mentoring works, then it is wise for practitioners to specify and measure steps along the way. The practice of tracking success in achieving proximal change targets is both good business and good science.

Conclusion In this chapter, we tried to make the case that youth mentoring is at an important crossroads. Modest outcomes are all too common and support for funding youth mentoring programs faces an uncertain future as the number of evidence-based programs for youth increases. But because of its popularity and wide distribution, youth mentoring has the potential to be brought to scale and rapidly disseminated. In our view, this will not happen unless mentoring researchers and practitioners embrace the epistemology of prevention science. A potential obstacle is a definition of mentoring that places undue emphasis on a close, lasting bond between mentor and mentee. We offered a more versatile, mentoring-as-context perspective that can serve as an important foundation for growing the science of youth mentoring.

References American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders—Text revision (4th ed.). Washington, DC: Author. Baker, D., & Maguire, C. P. (2005). Mentoring in historical perspective. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 14–29). Thousand Oaks, CA: Sage. Bernstein, L., Rappaport, C. D., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S.

Department of Education’s Student Mentoring Program: Final report. Washington, DC: U.S. Department of Education. Big Brothers Big Sisters of America (n.d.). Retrieved from http://www.bbbs.org/site/c.9iILI3NGKhK6F/ b.6065577/apps/s/content.asp?ct=8386019 Botvin, G., & Griffin, K. W. (2010). Advances in the science and practice of prevention: Targeting individuallevel etiological factors and the challenge of going to scale. In L. Scheier (Ed.), Handbook of drug use etiology: Theory, methods, and empirical findings (pp. 631–650). Washington, DC: American Psycho­ logical Association. Boyle, P. (2007, January 1). The study that ignited (or diluted) mentoring: How the “Big Brothers” report became the most influential, most useful and most misused evaluation in youth work. Youth Today. Retrieved from http://www.youthtoday.org Boyle, P. (2009, May 8). Obama budget cuts some youth programs, grows others: Mentoring, abstinence ed and juvenile justice lose, AmeriCorps wins, after school gets a tie. Youth Today. Retrieved from http:// www.youthtoday.org Burt, K. B., Obradović, J., Long, J. D., & Masten, A. S. (2008). The interplay of social competence and psychopathology over 20 years: Testing transactional and cascade models. Child Development, 79, 359–374. Cavell, T. A., Elledge, L. C., Malcolm, K. T., Faith, M. A., & Hughes, J. N. (2009). Relationship quality and the mentoring of aggressive, high-risk children. Journal of Clinical Child and Adolescent Psychology, 38, 185–198. Cavell, T. A., Hymel, S., Malcolm, K., & Seay, A. (2007). Socialization and the development of antisocial behavior: Models and interventions. In J. E. Grusec & P. D. Hastings (Eds.), Handbook of socialization (pp. 42–67). New York: Guilford Press. Cavell, T. A., & Smith, A. (2005). Mentoring children. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 160–176). Thousand Oaks, CA: Sage. Centers for Medicare and Medicaid Services. (2009). National health expenditure data: NHE fact sheet. Retrieved from http://www.cms.gov/NationalHealth ExpendData/25_NHE_Fact_Sheet.asp Chassin, L. A., Presson, C. C., & Sherman, S. J. (1985). Stepping backward in order to step forward: An acquisition-oriented approach to primary prevention. Journal of Consulting and Clinical Psychology, 53, 612–622. Cicchetti, D., & Cohen, D. J. (1995). Perspectives in developmental psychopathology. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathology (pp. 2–20). New York: Wiley. Coie, J. D., Miller-Johnson, S., & Bagwell, C. (2000). Prevention science. In A. J. Sameroff, M. Lewis, & S. M. Miller (Eds.), Handbook of developmental psychopathology (pp. 93–112). New York: Plenum Press. Coie, J. D., Watt, N. F., West, S. G., Hawkins, D., Asarnow, J. R., Markman, H. J., . . . Long, B. (1993).

Mentoring and Prevention Science   41 The science of prevention: A conceptual framework and some directions for a national research program. American Psychologist, 48, 1013–1022. Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: Peer groups and problem behavior. American Psychologist, 54, 755–764. Dodge, K. A., Greenberg, M. T., Malone, P. S., & Conduct Problems Prevention Research Group. (2008). Testing an idealized dynamic cascade model of the development of serious violence in adolescence. Child Development, 79, 1907–1927. Dodge, K. A., & Pettit, G. S. (2003). A biopsychosocial model of the development of chronic conduct problems in adolescence. Developmental Psychology, 39, 349–371. DuBois, D. L., Doolittle, F., Yates, B. T., Silverthorn, N., & Tebes, J. K. (2006). Research methodology and youth mentoring. Journal of Community Psychology, 34, 657–676. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Durlak, J. A., Weissberg, R. P., & Pachan, M. (2010). A meta-analysis of after-school programs that seek to promote personal and social skills in children and adolescents. American Journal of Community Psychology, 45, 294–309. Elledge, L. C., Cavell, T. A., Ogle, N. T., & Newgent, R. A. (2010). School-based mentoring as selective prevention for bullied children: A preliminary test. Journal of Primary Prevention, 31, 171–187. Elliot, D. S., & Mihalic, S. (2004). Issues in disseminating and replicating effective prevention programs. Prevention Science, 5, 47–53. Faith, M., Fiala, S. E., Cavell, T. A., & Hughes, J. N. (2011). Mentoring highly aggressive children: ­Pre-post changes in mentors’ attitudes, personality, and attachment tendencies. Journal of Primary Prevention, 32, 253–270. Flay, B. R. (1986). Efficacy and effectiveness trials (and other phases of research) in the development of health promotion programs. Preventive Medicine, 15, 451–474. Flay, B. R., Biglan, A., Boruch, R. F., Castro, F. G., Gottfredson, D., Kellam, S., . . . Ji, P. (2005). Standards of Evidence: Criteria for efficacy, effectiveness, and dissemination. Prevention Science, 6, 151–175. Garmezy, N., & Masten, A. S. (1986). Stress, competence, and resilience: Common frontiers for therapist and psychopathologists. Behavior Therapy, 17, 500–521. Grossman, J. B., Chan, C. S., Schwartz, S. E. O., & Rhodes, J. E. (2012). The test of time in schoolbased mentoring: The role of relationship duration

and re-matching on academic outcomes. American Journal of Community Psychology, 49, 43–54. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring programs. American Journal of Community Psychology, 30, 199–219. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work? An impact study of the Big Brothers Big Sisters program. Evaluation Review, 22, 403–426. Hanlon, T. E., Simon, B. D., O’Grady, K. E., Carswell, S. B., & Callaman, J. M. (2009). The effectiveness of an after-school program targeting urban AfricanAmerican youth. Education and Urban Society, 42, 96–118. Hansen, J., & Wänke, M. (2009). Think of capable others and you can make it! Self-efficacy mediates the effect of stereotype activation on behavior. Social Cognition, 27, 76–88. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., & McMaken, J. (2007). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Hirschi, R. T. (1969). Causes of delinquency. Berkeley: University of California Press. Hughes, J. N. (2000). The essential role of theory in the science of teaching children: Beyond empirically supported treatments. Journal of School Psychology, 38, 301–330. Hughes, J. N., Cavell, T. A., & Meehan, B. T. (2001). Preventing substance abuse in aggressive children. Final report of R01-DA10037 submitted to the National Institute on Drug Abuse. Hughes, J. N., Cavell, T. A., Meehan, B., Zhang, D., & Collie, C. (2005). Adverse school context moderates the outcomes of two selective intervention programs for aggressive children. Journal of Consulting and Clinical Psychology, 73, 731–736. Jent, J. F., & Niec, L. N. (2006). Mentoring youth with psychiatric disorders: The impact on child and parent functioning. Child & Family Behavior Therapy, 28, 43–58. Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9, 99–113. Karcher, M. J., & Nakkula, M. J. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. In G. G. Noam, M. J. Karcher, & M. J. Nakkula (Eds.), New directions for youth development: Theory, practice, and research. The structure of mentoring interactions (pp. 13–32). Hoboken, NJ: Wiley. Kazdin, A. E., Kraemer, H. C., Kessler, R. C., Kupfer, D. J., & Offord, D. R. (1997). Contributions of risk-factor research to developmental psychopathology. Clinical Psychology Review, 17, 375–406. Kuperminc, G., Emshoff, J. G., Reiner, M. N., Secrest, L. A., Niolon, P. H., & Foster, J. D. (2005). Integration of mentoring with other programs and services. In

42  FRAMEWORKS AND FOUNDATIONS D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 314–339). Thousand Oaks, CA: Sage. McCord, J. (2003). Cures that harm: Unanticipated outcomes of crime prevention programs. Annals of the Academy of Political and Social Sciences, 587, 16–30. MENTOR. (2006a). Mentoring in America 2005: A snapshot of the current state of mentoring. Alexandria, VA: Author. Retrieved from http://www.mentoring .org/downloads/mentoring_523.pdf MENTOR. (2006b). The national agenda for action: How to close America’s mentoring gap. Alexandria, VA: Author. Retrieved from http://www.michigan.gov/ documents/nationalagenda_157944_7.pdf Moffitt, T. E. (1993). Adolescence-limited and life-coursepersistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674–701. Morrow, K. V., & Styles, M. B. (1995). Building relationships with youth in program settings: A study of Big Brothers/Big Sisters. Philadelphia: Public/Private Ventures. Mrazek, P. J., & Haggerty, R. J. (Eds.). (1994). Reducing risks for mental disorders: Frontiers for preventative intervention research. Washington, DC: National Academies Press. Munoz, R. F., Mrazek, P. J., & Haggerty, R. J. (1996). Institute of Medicine report on the prevention of mental disorders. American Psychologist, 51, 1116–1122. O’Connell, M. E., Boat, T., & Warner, K. E. (2009). Preventing mental, emotional, and behavioral disorders among young people: Progress and possibilities. Washington, DC: National Academies Press. Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43). Thousand Oaks, CA: Sage. Rhodes, J. (2008). Improving youth mentoring through research-based practice. American Journal of Community Psychology, 41, 35–42. Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20(3). Rhodes, J., & DuBois, D. (2008). Mentoring relationships and programs for youth. Current Directions in Psychological Science, 17, 254–258. Rhodes, J. E., Grossman, J. B., Resch, N. L. (2000). Agents of change: Pathways through which mentoring relationships influence adolescents’ academic adjustment. Child Development, 71, 1662–1671. Rhodes, J., Liang, B., & Spencer, R. (2009). First do no harm: Ethical principles for youth mentoring relationships. Professional Psychology: Research and Practice, 40, 452–458. Rhodes, J., Reddy, R., Roffman, J., & Grossman, J. (2005). Promoting successful youth mentoring

relationships: A preliminary screening questionnaire. Journal of Primary Prevention, 26, 147–168. Rhodes, J., Spencer, R., Keller, T., Liang, B., & Noam, G. (2006). A model for the influence of mentoring relationships on youth development. Journal of Community Psychology, 34, 691–701. Rutter, M. (2009). Understanding and testing risk mechanisms for mental disorders. Journal of Child Psychology and Psychiatry, 50, 44–52. Schwartz, S. E., Rhodes, J. E., Chan, C. S., & Herrera, C. (2011). The impact of school-based mentoring on youths with different relational profiles. Developmental Psychology, 47, 450–462. Shpigelman, C. N., Reiter, S., & Weiss, P. L. (2009). A conceptual framework for electronic socio-emotional social support for people with special needs. International Journal of Rehabilitation Research, 32, 301–308. Smith, L. H. (2011). Piloting the use of teen mentors to promote a healthy diet and physical activity among children in Appalachia. Journal for Specialists in Pediatric Nursing, 16, 16–26. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Spoth, R., & Greenberg, M. (2011). Impact challenges in community science-with-practice: Lessons from PROSPER on transformative practitioner-scientist partnerships and prevention infrastructure development. American Journal of Community Psychology, 48, 106–119. Springer, J. F., & Phillips, J. (2006). The Institute of Medicine framework and its implication for the advancement of prevention policy, programs, and practices. Retrieved from http://www.ca-sdfsc.org/ docs/resources/SDFSC_IOM_Policy.pdf Taussig, H. N., Culhane, S. E., & Hettleman, D. (2007). Fostering Healthy Futures: An innovative preventive intervention for preadolescent youth in out-of-home care. Child Welfare. 86(5), 113–131. Tierney, J. P., Grossman, J. B., & Resch, N. L. (1995). Making a difference. An impact study of Big Brothers/Big Sisters. Philadelphia: Public/Private Ventures. Werner, E. E. (1986). Resilient offspring of alcoholics: A longitudinal study from birth to age 18. Journal of Studies on Alcohol, 47, 34–40. Werner, E. E. (1995). Resilience in development. Current Directions in Psychological Science, 4, 81–85. Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials of schoolbased mentoring: Making sense of mixed findings. SRCD Social Policy Report, 24(3). Retrieved from http://www.srcd.org Wyman, P. A., Cross, W., Hendricks, B. C., Yu, Q., Tu, X., & Eberly, S. (2010). Intervention to strengthen emotional self-regulation in children with emerging mental health problems: Proximal impact on school behavior. Journal of Abnormal Child Psychology, 38, 707–720.

PART III Mentoring Relationships

4  ASSESSING MENTORING RELATIONSHIPS Michael J. Nakkula and John T. Harris1

Introduction In 2005, we contributed a chapter on the assessment of mentoring relationship quality (MRQ) to the first edition of the Handbook of Youth Mentoring (Nakkula & Harris, 2005). Since that time, this area of the youth mentoring field has benefited from a number of key developments, with three areas of growth being particularly notable. First, basic research has linked participants’ perceptions about the quality of their matches with other aspects of match quality such as how mentors and mentees focus their activities, how they choose what to do, and how mentors perceive the training and support provided by their respective programs. Second, because instrumental or goal-oriented approaches in youth mentoring have become more common, assessments of instrumental match quality have gained more attention. And, third, just as there have been developments in our understanding of the relationship between how a match is conducted and how it is experienced, progress has been made in our understanding of the relationship between match quality and outcomes predicted by it. This chapter, which updates our earlier contribution, begins with an overview of a theoretical framework for approaching assessment of MRQ. We next review available instruments for assessing MRQ from the perspective of this framework as well as the extant evidence of their psychometric properties. Following this review, we present practical guidelines and recommendations for how MRQ assessment can be used more fully to enhance

understanding and promotion of quality mentoring experiences for youth.

Theory We subscribe to the notion that match quality influences the nature and degree of outcomes experienced by mentored youth (Rhodes, 2005). For the purposes of this chapter, we define match quality as the characteristics of relationships between adults and youth that are specific to the mentoring experience and thought to directly and substantially influence the mentee’s outcomes (growth and development). A wealth of well-established instruments exist to measure factors such as preexisting characteristics of mentees (e.g., risk status and personality traits) that might influence outcomes; and strong instruments exist to measure the outcomes themselves (e.g., academic improvement and gains in self-confidence). What remains murky is how best to assess match quality across different mentoring models and how to manipulate specific facets of match quality to influence outcomes for different types of youth and in different environments. Therefore, advancements in the assessment of MRQ should help those involved with mentoring to understand what match characteristics are most likely to yield desired outcomes, and to assess active matches to ensure that they facilitate those benefits. We believe that understanding and assessing match quality is the key to making youth mentoring as efficient and effective as it can be.

1

The authors thank the mentoring program participants who informed this work, with particular gratitude to Ms. Kathleen Murphy and the staff of Yavapai Big Brothers Big Sisters in Prescott, Arizona, for invaluable collegiality and inspiration.

45

46  Mentoring Relationships Over the relatively brief history of MRQ assessment, three broad types of match characteristics have emerged as meaningful indicators of match quality (Nakkula & Harris, 2010; see also Nakkula & Harris, 2005): •• Internal match quality reflects the way participants feel about what is done in the match and about each other. We organize subjective indicators of internal match quality into three broad characteristics that are influenced and experienced directly by the mentor and mentee in ways that strengthen (or weaken) the match: compatibility, the goodness of fit between mentor and mentee related to traits and preferences; competence related to mentoring-specific approaches, traits, and skills that nurture the match and facilitate the mentee’s growth; and, satisfaction with the relationship and benefits the mentee is perceived to derive from it. The most salient objective indicator of internal match quality is dosage, the frequency and duration of their meetings. •• Match structure reflects what mentors and mentees want to do in the match, the ways in which they decide what to do (who decides and how), and what actually gets done. Match structure is that part of MRQ that allows for the linking of subjective and objective assessments of quality with the characteristics of different types of matches. It has three main components: purposes and goal setting, which reflect mentor and mentee values and what they want to do or accomplish in the match; authorship, which refers to the way they negotiate how to act on those purposes and goals; and focuses, which reflect the way they experience the nature of match activities (Karcher & Nakkula, 2010). •• External match quality reflects how well the participants are supported by influences on the relationship that are not within their direct control, such as programmatic support, parental engagement, and circumstances or logistics associated with match outings such as scheduling, money, and travel constraints. Aspects of these broad characteristics often are considered more specifically as relational, pertaining to how the mentor and mentee feel about each other and their relationship, or instrumental, which is oriented toward the mentee’s growth. We label

indicators as predominantly relational or instrumental, but view each distinct MRQ indicator as occupying a space of its own along a continuum from extremely relational (e.g., closeness) through extremely instrumental (e.g., perceived support). Our research indicates that it is useful to assess indicators of relational and instrumental quality that reflect this full continuum (Nakkula & Harris, 2010). The vast majority of efforts to measure MRQ have focused on internal characteristics of the match, especially those related to relational match quality. However, questions regarding match structure and external match quality also have become increasingly central to understanding successful mentoring relationships (Larose, Cyrene, Garceau, Brodeur, & Tarabulsy, 2010). Match quality also may be measured through subjective and objective indicators, and it appears to be important to measure both positive and negative perspectives of MRQ (Rhodes, Reddy, Roffman, & Grossman, 2005). The tools reviewed in this chapter draw on different theoretical perspectives to inform the way they measure the aspects of MRQ on which they focus. We relate those perspectives to the framework of MRQ as it is described above (with framework constructs italicized the first time they are introduced). A broad summary of this framework is presented in Table 4.1, and Table 4.2 summarizes the aspects of the framework measured by each instrument. Relational Match Quality The most clearly demonstrated relationships between match quality and youth outcomes have involved indicators of relational match quality, which refers to indications of how a mentor and mentee feel about each other and the way they relate. Most instruments reviewed in this chapter focus on relational match quality, with heavy emphasis on relational satisfaction, of which feelings of closeness is a central component. Fewer instruments focus on the measurement of characteristics we see as helping to build and shape closeness: relational compatibility of personal traits and preferences and relational competence that reflects approaches, traits, and skills that help mentors and mentees build the match. Subjective Indicators. Relational compatibility is thought to be one of the foundations of a strong relationship (Herrera, Sipe, McClanahan, Arbreton, & Pepper, 2000). To make compatible matches, programs consider personal traits such as empathy, physical capacity, and outgoingness. These traits, along with

Assessing Mentoring Relationships   47 Table 4.1   Mentoring Relationship Quality: Overview of Constructs

Internal Match Quality

Match Structure

External Match Quality Environmental

Relational

Instrumental

Relational and Instrumental

Subjective

Subjective

Subjective

•• Compatibility  Traits  Preferences

• Compatibility

•• Competence

•• Competence

•• Purpose/goal orientation (goal setting) •• Focus of match activities •• Authorship

  

Approach Traits Skills

•• Satisfaction Precursors to closeness  Connectedness  Closeness  Intimacy 

 

Traits Preferences

Approach Traits  Skills  

Both Subjective & Objective •• Programmatic support •• Parents’ or guardians’ engagement •• Support networks •• Logistical challenges

•• Satisfaction 

Perceived support

Objective

Objective

Objective

•• Meeting frequency & intensity over time •• Longevity of the relationship •• Durability/ resilience

•• Received support

•• Activities engaged in •• Goals set

Note: The constructs in this table are presented in a positive valence. It should be noted, however, that research suggests that in some cases it is most useful to focus measurement on aspects of the construct with negative valence (e.g., rather than a sense of connectedness, feelings of disappointment or dissatisfaction).

preferences such as interests and preferred focuses for the match can influence the way a match develops (Herrera et al., 2000). Whatever the level of relational compatibility, building a strong relationship requires a certain level of relational competence on the part of both the mentor and, to a lesser extent, the mentee. The best-documented form of mentoring-specific competence entails participants’ relational approach—the way each goes about trying to develop the match. The mentor’s positive regard is a key aspect of mentors’ relational approach that has been associated with positive mentee outcomes (DuBois, Neville, Parra, & Pugh-Lilly,

2002). Mentors’ youth-centeredness/prescription also has long been viewed as an important influence on MRQ (Morrow & Styles, 1995). Mentees have less responsibility for nurturing the relationship; nonetheless, it seems clear that the way mentees engage with their match can influence how well it develops. Indicators of mentee’s approach that have been linked with variation in match quality include active engagement with the match (Rhodes, 2002) and distancing behavior (Karcher, Nakkula, & Harris, 2005). We identified two additional competence indicators drawn from mentoring research and practice: basic relational competence and risk-related

48  Mentoring Relationships relational competence. No published match quality survey currently offers a scale measuring basic relational competence. To assess it we recommend measuring traits exhibited in the matches (e.g., patience and empathy) and skills (e.g., communication and relationship management) that both the mentoring and psychotherapy research indicate might be most salient to building strong helping relationships (Norcross, 2011; Spencer, 2006). Does the mentor feel sufficiently possessed of these traits and skills to be able to use them competently within their match? Does the mentee perceive that the mentor demonstrates them effectively? Although the onus for sustaining the match falls most heavily on the mentor, the resilience literature clearly indicates that a mentee who is perceived to exhibit certain traits (e.g., friendliness, talented, strong character) and skills (e.g., verbal ability, self-management) might be expected to be more successful at recruiting an adult’s personal investment and developing a strong bond with them—whether in a natural or formal mentoring relationship (Benard, 2004). We also consider it important to measure an additional competence, which we call risk related, because so many mentored youth contend with personal and developmental challenges that require capacities in a mentor that transcend simple relationship-building. Mentors’ perceptions that their mentees face severe challenges may diminish match quality (Blocker, 1993; Slicker & Palmer, 1993). To assess risk-related competence, it would be helpful to measure mentors’ perceptions about the mentee’s behaviors and academic functioning and their feelings of capacity and preparedness to handle those challenges in a way that allows the relationship to grow. A reciprocal assessment of the mentor’s facility could be measured from the mentee’s perspective. We view empathy, trust, and respect as key precursors to closeness that allow closeness to develop over time. Even though these factors frequently are understood as part of closeness itself, they can also exist apart from it. Spencer (2006) found empathy to be central to mentors’ capacity to adjust their mentoring goals. Trust is featured as a positive aspect of MRQ and, when broken, is found to be a predictor of premature match termination (Rhodes et al., 2005). Showing respect for one’s mentee and his or her point of view is part of the collaborative mentoring approach that has been widely endorsed by researchers and practitioners alike (Karcher & Nakkula, 2010). Relational satisfaction is the most common focus of current MRQ assessments and is associated with some of the strongest research on how match

quality relates to outcomes. The central facet of relational satisfaction, closeness, is marked by connectedness, appropriate sharing, and reciprocal feelings of care. In both mentors and mentees, closeness has long been a key organizer for the understanding and assessment of mentoring relationships (Liang, Spencer, Brogan, & Corral, 2008; Rhodes, 2002). Further, it is widely accepted that relationships based in these experiences are likely to last longer and lead to better developmental outcomes (Deutsch & Spencer, 2009; Rhodes & DuBois, 2008). Distance and dissatisfaction are common measurement focuses of closeness from a negative perspective. Particularly for mentors, a sense of trajectory is another important facet of relational satisfaction, since frustration can result if it seems that the relationship is not growing and developing in ways that signal the likelihood of positive outcomes for the mentee (Karcher, 2008). Numerous theoretical perspectives are brought to bear on the measurement of relational satisfaction. Connectedness refers to the manner and extent to which mentors and mentees feel aligned with or related to one another (Karcher, 2008). Two instruments reviewed in the research section below draw from literature on healing components of effective therapeutic relationships to inform their measurement of participants’ perceptions about precursors to closeness and closeness with each other (Cavell, Elledge, Malcolm, Faith, & Hughes, 2009; Zand et al., 2009). That literature suggests that a close working alliance is important to enhancing psychological growth and wellness. Building strong mentoring alliances also has been described as a process of bonding (Herrera et al., 2000); if the bond does not develop adequately, matches are more likely to fail or end prematurely (Rhodes et al., 2005). To feel known and genuinely accepted is a hallmark of relational health and important to helping relationships (Liang et al., 2002). The counseling and therapy literature has long referred to this experience as authenticity. When one experiences authentic engagement (genuine versus superficial or “phony”) from another, it is easier to bring one’s “real self” into that relationship and to grow more fully within it (Spencer, 2006). Such authentic relatedness contributes to the experience of intimacy—the feeling of being deeply understood and accepted within ongoing close relationships (Liang et al., 2002). Objective Indicators. A number of objective indicators of MRQ have been associated with experiences of relational match quality, including meeting frequency (Sipe, 1998), meeting intensity (DuBois, Holloway, Valentine, & Cooper, 2002), and match

Assessing Mentoring Relationships   49 duration or longevity (Grossman & Rhodes, 2002). When matches meet at least twice per month on a consistent basis, relational match quality has an opportunity to solidify. Meeting intensity refers to the length of time spent together per meeting. With respect to duration, matches that last for a year or more generally are marked by greater levels of trust and closeness (Rhodes et al., 2005). Durability/Resilience refers to a particular type of match duration—that which is marked by overcoming obstacles, such as difficulties finding meeting times or relationship challenges that might have threatened the match. Relationships that grow or become resilient in the face of challenges have been found to offer particular benefits to youth and adults (Masten, 2009). Instrumental Match Quality As Karcher and Nakkula (2010) noted in their depiction of an array of mentoring approaches, match satisfaction does not pertain solely to feelings about how the mentor and mentee feel about each other. Rather, it comprises approaches ranging from those that are purely relational or fun-oriented to those that are serious and goal-focused. Satisfaction can be as contingent on instrumental aspects of the match as it is on relational experiences. Accordingly, assessments of MRQ have begun to focus more thoroughly on the full range of instrumental match quality. Subjective Indicators. Paralleling the concept of relational compatibility discussed above, instrumental compatibility reflects the goodness of fit between a mentor’s and a mentee’s instrumental (growthoriented) goals and values in the match. What does the mentor want the mentee to get out of the match? Is that congruent with what the mentee wants to get out of it? How intensely does each want to strive for those outcomes? Lower levels of instrumental compatibility may result in negative dynamics that have been associated with prescriptive mentoring (Morrow & Styles, 1995) or with dissatisfaction associated with insufficient instrumental support (Nakkula & Harris, 2010). Our conception of instrumental competence also parallels a concept pertaining to relational match quality. Mentoring-specific competence is required to translate instrumental compatibility into real and perceived benefits for the mentee. Again, we organize that competence according to the perceptions of approaches, traits, and skills. Youth-centeredness/prescription continues to figure as a key to mentors’ approach, but pertains here to issues such as how and when to incorporate

academics or other growth-oriented focuses into the match (Karcher & Nakkula, 2010). For mentees, initiating goal-oriented activity, especially active support seeking, can inspire mentors to feel more invested in the match (Karcher, 2008). Instrumentally competent mentors should act as role models and sources of inspiration, as these qualities are thought to engage mentees more effectively with growth opportunities (Blechman, 1992; Rhodes, 2002). Skills related to setting appropriate expectations and effective communication may help participants identify instrumental focuses and work to achieve them. Perceiving that the mentee is benefiting from the match promotes instrumental satisfaction. The importance of instrumental satisfaction can vary across mentees, but has been found to grow increasingly important as a function of mentees’ age (Larose et al., 2010). Investment in instrumental satisfaction tends to be more universal among mentors, with the key added element of degree: mentors want to feel that their time and effort has been useful, so their satisfaction tends to hinge on the question of whether the mentee has benefited enough. When mentors feel that they are making a meaningful difference in their mentees’ lives, they may be more likely to remain invested in the match for a longer period of time (Blocker, 1993). Objective Indicators. Received support reflects the objective or externally observable benefits for the mentee that are facilitated by the match. Differentiating received support from perceived support necessarily involves some subjective evaluation. What is the threshold for identifying something as support? No research currently provides clear guidance, so questions of scope and value remain to be resolved. For now, it may be most useful to think of received support in broad, categorical terms, so we recommend measuring the type and frequency of support. Types of received support might be defined in part by assessments of what the mentee needs and could include a broad range of categories: efforts at advocacy (e.g., with a teacher or parent); discussing personal challenges the mentee faces; doing schoolwork together; exposing the mentee to valuable experiences and opportunities (e.g., library, museum, workplace); helping a mentee achieve something (e.g., win a race or get a job); or simply identifying concrete goals for the match. Match Structure Whereas assessment of internal match quality focuses on the question of how participants feel

50  Mentoring Relationships about their matches and each other, assessment of match structure focuses on what they actually do and why. Although research has yet to demonstrate with great specificity what type of structure might best promote particular outcomes, it is hard to argue that what mentors and mentees do does not influence outcomes from the match. Further, the growth in mentoring approaches renders it important that we begin linking MRQ with specific types of matches; with continued research, progressively refined conceptions of MRQ may be developed for different match types. Generic assessments of quality may provide too little information on what is actually being experienced and assessed in the match. Subjective Indicators. Drawing on emerging conceptions of match structure (Karcher & Nakkula, 2010; Nakkula & Harris, 2010), we define purpose as what participants want to do or accomplish in the match, including the degree to which they value relational and instrumental activities. To date, no instrument measures purpose from the mentee’s perspective, which makes some sense given the greater agency the mentor generally has in dictating what activities are done in a match and the fact that mentees frequently are not able to articulate effectively what they want from the match. We define focus as the amount of time participants perceive is spent on relational and instrumental activities in the match (Karcher & Nakkula, 2010; Nakkula & Harris, 2010). As with purpose, it is important that assessment of focuses span the full range of relational and instrumental activities because it appears that the relative emphasis on each may be the most useful indicator of match quality. To date, no instrument measures focuses from the mentor’s perspective. How decisions get made about what to do in the match and who makes them has a great deal to do with both relational and instrumental match quality. Karcher and Nakkula (2010) used the notion of authorship to capture the various ways in which decisions get made within mentoring relationships. Unilateral mentor-initiated decision making tends to be experienced as prescriptive: the mentor prescribes a course of action that he or she deems helpful to the mentee. Such approaches have long shown negative outcomes (DuBois, Holloway, Valentine, & Cooper, 2002; Rhodes, 2002). Unilateral mentee-initiated decision making can be deleterious as well (Karcher, Herrera, & Hansen, 2010), leaving mentors feeling a lack of efficacy or satisfaction with their work, particularly when the choices their mentees make differ

markedly from their own perspectives on what would be helpful. Decision making is believed to be most healthy when it is done collaboratively, with the partners sharing ideas and talking through their respective interests (Karcher & Nakkula, 2010). Measurement of authorship, then, should assess participants’ perspectives on what is done in the process of making decisions, including how they are actually made. Objective Indicators. Activities engaged refers to the specific types of activities selected by the mentor and mentee for the match. Such activities can be documented objectively to help determine the match focus and the mentor’s purpose for organizing it in particular ways (Nakkula & Harris, 2010). Goals set within the match also can be captured objectively and used to help determine the shape and direction of the match’s development (see Balcazar and Keys, this volume, Chapter 6). External Match Quality Internal match quality and match structure are influenced by how external qualities that lie beyond the participants’ control either support or interfere with the match. In formal mentoring, programmatic support encompasses both the philosophies and practices of programs. These are core external factors as they shape relationship expectations and provide training and support for the match (Rhodes & DuBois, 2008). Key programmatic indicators include effective supervision and provision of structured activities for participants. Parents or guardians also can influence MRQ through support they provide and barriers they create. A positive reciprocal bond between the mentoring relationship and family or caregiving relationships can be important to developing and maintaining optimal MRQ. In addition to family supports or barriers for the match itself, the mentor’s and mentee’s personal support networks can have an impact on the match. If the mentor’s personal or professional life is not conducive to sustaining the match, even matches with high levels of internal match quality may be at risk for premature termination. Logistics related to circumstances like monetary issues, work or school commitments, and the environment in which the mentoring takes place can influence what participants can do with each other and how often. For community-based matches, scheduling and travel can be major issues. In school-based or other sitebased mentoring approaches, the environment in which the activities take place heavily influences the match.

Assessing Mentoring Relationships   51

Research In this section we review instruments that have informed and resulted from the theoretical foundation presented above. We include instruments that have substantial amounts of research support, prioritizing those with published validity evidence and those that illustrate important developments in the field. To maximize narrative consideration of each instrument, specifics related to structure (number of items and scales), constructs measured, internal consistency, and validity evidence are summarized in Table 4.2. Youth-Mentor Relationship Questionnaire (mentee only; 15 items/4 subscales) It seems most appropriate to begin this review of MRQ assessment with the Youth-Mentor Relationship Questionnaire (YMRQ), the instrument used in the Public/Private Ventures landmark study of Big Brothers Big Sisters (Grossman & Tierney, 1998). Although earlier versions of the instrument proved useful in predicting a host of youth outcomes, the current, more refined version (Rhodes et al., 2005) is stronger empirically and is used widely to assess match quality. Because the authors’ prior research showed that the absence of negative experiences was a stronger predictor of match duration than the presence of positive ones, three of the instrument’s four subscales assess negative aspects of MRQ: dissatisfaction with mentors (not dissatisfied), unhappiness during meetings (not unhappy), experience of breached trust in the relationship (trust not broken). The positively framed helped to cope captures the degree to which mentees feel supported by the mentor in addressing their challenges or concerns. Rhodes et al. (2005) found that YMRQ scores indicating a relative lack of negative relationship experiences in the match predicted match retention for one year and beyond. In addition to predicting match continuity, trust not broken and not dissatisfied predicted differences in general self-esteem and academic performance in the expected directions. The helped to cope subscale was inversely related with differences in self-esteem, which the authors suggest may signify either that the mentee is in greater need of support due to life stressors and therefore suffering declines in self worth, or that mentors who overextend support may inadvertently give the message that the mentee is particularly vulnerable. A central contribution of the YMRQ has been its emphasis on detecting negative match experiences,

which can serve as red flags for troubled matches. The authors underscore the credo of “do no harm” as a first priority for mentoring programs. Its prediction of match duration and improvements in important youth outcomes makes the YMRQ a useful instrument for assessing mentoring processes and predicting mentoring outcomes. The primary weaknesses of the YMRQ are that it covers a fairly narrow range of MRQ indicators and that it offers a fairly restricted focus on negative aspects of match quality, which limits its utility as an assessment and feedback tool for what is going well in matches. Mentor-Youth Alliance Scale (mentee only, 10 items/2 scales) The Mentor-Youth Alliance Scale (MYAS) was developed to study the relationship between mentee perceptions of their alliance with their mentors and the development of mentee competencies in different domains (Zand et al., 2009). The MYAS includes two five-item subscales, labeled caring and acceptance. Caring reflects mentees’ perspectives on how much the mentor and mentee care about each other or share a positive emotional bond. Acceptance measures how much mentees feel comfortable talking and generally being with their mentors. Confirmatory factor analyses suggest that the two subscales are not distinct enough to be used separately, but rather are part of a larger alliance construct that the authors term the relational bond. Among a sample of 276 diverse youth, ranging in age from 9 to 19, the MYAS was found to correlate moderately with scores for attachment bonds with adults, as assessed by the Adult Relationships Scale (Zand et al., 2009). Higher scores on the MYAS also predicted positive youth outcomes related to relationships with adults, family bonding, school bonding, and life skills (Thomson & Zand, 2010; Zand et al., 2009). This published convergent and predictive validity evidence is a strength of the MYAS, as is the clear focus on the mentoring alliance. The short, focused, user-friendly nature of the MYAS also represents a limitation: the scale is unidimensional and provides information only on the relational bond. Mentor Alliance Scale (MAS; mentee and mentor versions, each with 13 items/1 scale) Cavell and colleagues (Cavell & Hughes, 2000; Cavell et al., 2009) created mentor and mentee versions of a Mentoring Alliance Scale (MAS), which are adaptations of the Therapeutic Alliance Scales for children and therapists, respectively

52

(Cavell et al., 2009)

Network of Relationships (NRI), mentor version

(Cavell et al., 2009)

Network of Relationships (NRI), mentee version

(Cavell et al., 2009)

Mentor Alliance Scale (MAS), mentor version

(Cavell et al., 2009)

Mentor Alliance Scale (MAS), mentee version

(Zand et al., 2009)

Mentor-Youth Alliance Scale (MYAS)d

(Rhodes et al., 2005)

Youth-Mentor Relationship Questionnaire

Reviewed Scale/ Author(s)

Mentors of primary-school children

Mentees in primary school

Mentors of primary-school children

Mentees in primary school

Mentees from ages 9–19

Mentees in primary and secondary school

Intended Respondents .74 .81 .85 .81

Not dissatisfied (3) Helped to cope (3) Not unhappy (6) Trust not broken (6)

.91–.94 .83–.87

Conflict (6)

.59–.87

Relationship conflict (6)

Support (6)

.88–.92

.81–.90

.75–.84

Relationship support (5)

Relationship alliance/bond (13)

Relationship alliance/bond (13)

.85–.92

N/A

Overall scale (15)c

Relationship alliance/bond (10)

Internal Consistency

Named Factors Measured (Number of Items)

Table 4.2   Characteristics of Instruments for Assessing Mentoring Relationship Quality

Discriminant

Concurrent

Predictive (B)

Discriminant

Predictive (B)

Discriminant

Concurrent

Discriminant

Concurrent

Concurrent

Predictive (A, B, C)

Predictive (A, D)

Predictive (D)

Predictive (A, D)

Predictive

Validity Evidencea

Internal (R, S)

Internal (R, S)

Internal (R, S)

Internal (R, S)

Internal (R, S)

Internal (R, I, S)

Aspect of MRQ Assessedb

53

(Harris & Nakkula, 2003)

Match Characteristics Questionnaire, v. 2.0e

Mentors of youth in primary and secondary schools

Mentees in primary and secondary schools

Youth Mentoring Survey (YMS)

(Harris & Nakkula, 2003b)

Intended Respondents

Reviewed Scale/ Author(s)

.61 N/A .76 .85 .79

Prescription (3) Overall focus (20) Fun (7) Sharing (7) Growth (6)

.94 .76 .62 .82 .78 .86 .86 .87 N/A

Overall internal match quality (31) Compatibility (6) Risk-related competence (3) Closeness (4) Not distant (6) Satisfaction (5) Nonacademic support seeking (5) Academic support seeking (2) Overall purposes (20, plus 3 written/ ranked)

N/A

.77

Instrumental (7)

Overall survey (66, plus 3 written/ ranked)

.85

.87

Overall internal match quality (23) Relational (13)

N/A

Internal Consistency

Overall scale (43)

Named Factors Measured (Number of Items)

(Continued)

Structural (R, I, S)

External (R, I, S) Convergent

Internal (R, I, S) Concurrent

Structural (R, I, S)

Internal (R, I, S)

Aspect of MRQ Assessedb

Predictive

Convergent

Concurrent

Predictive

Validity Evidencea

54

(BBBSA, 2008)

Strength of Relationship (SoR) Measure, mentor version

(BBBSA, 2008)

Strength of Relationship (SoR) Measure, mentee version

Reviewed Scale/ Author(s)

Table 4.2  (Continued)

Mentors of youth in primary and secondary schools

Mentees in primary and secondary schools

Intended Respondents

.76 .73 .65 .70 N/A N/A

Overall scale (15) Connectedness (5) Frustration (4) Confidence (4) Closeness (1) Decision-making (1)

N/A

.53

Friends/family support (5)

N/A

.66

Interference (6)

Closeness (1)

.52

Parent engagement (3)

Importance (1)

.78

Program support (3)

N/A

.63

Overall external match quality (17)

.68

N/A

Main purposes (listed/ranked)

Safety (1)

.78

Education (4)

Disappointment (4)

.76

Future outlook (4)

.70

.78

Character development (4)

.73

.68

Sharing (4)

Coping (3)

.77

Fun/hanging out (4)

Overall scale (10)

Internal Consistency

Named Factors Measured (Number of Items)

Predictive

Concurrent

Predictive

Concurrent

Validity Evidencea

Internal (R, S)

Internal (R, S)

Aspect of MRQ Assessedb

55

Mentees in primary and secondary schools

Middle and high school mentees

Female college-age mentees

Intended Respondents

Dissatisfaction (6)

Emotional engagement (8)

Youth-centered relationship (5)

Overall scale (19) N/A

.79–.84

.77

Authenticity (4) Natural mentoring relational health (6)

.72

.72

Engagement (3) Empowerment/zest (4)

.86

Internal Consistency

RHI Mentoring subscale (11)

Named Factors Measured (Number of Items)

Predictive (C, D)

Validity Evidencea

Internal (R, S)

Internal (R, S)

Internal (R, I, S)

Aspect of MRQ Assessedb

Only the overall scale was reported in most recent publications. Caring and acceptance scales were reported in initial publication.

There is some overlap between v. 1.0 and v. 2.0 of the MCQ. However, only v. 2.0 is presented here as it contains additional/revised scales/items and has been administered much more extensively.

e

d

Three items load on two factors.

c

Letters associated with measured types of MRQ are as follow: R, relational; I, instrumental; S, subjective; and O, objective. These letters are components of the constructs reflected presented in Table 4.1, which were used to organize our review of the instruments for assessing mentoring relationship quality.

b

a For the purposes of this chapter, evidence of predictive validity is considered in relation to outcomes in four categories: (A) academics, including attitudes and performance; (B) behavior, including delinquent, prosocial, and life skills; (C) connections, including relationships and affiliations; and (D) dispositions, including feelings and attitudes.

(Public/Private Ventures, 2002)

Youth Survey

(Liang et al., 2010)

Relational Health Indices–Youth (RHI-Y)

(Liang et al., 2002)

Relational Health Indices (RHI-M)

Reviewed Scale/ Author(s)

56  Mentoring Relationships (Shirk & Saiz, 1992). For both the mentor and mentee versions, the MAS consists of a single scale assessing the quality of the mentoring alliance. The authors have used the scale to assess mentoring alliances for elementary school children (grades 2 through 5) of varying racial and ethnic backgrounds and their mentors, and to compare the alliances across different mentoring models. Evidence of the scales’ construct validity (i.e., expected patterns of association with measures of other constructs) is found in a study that showed children’s MAS scores for mentors and mentees were predictably higher for participants in matches with extensive amounts of training and support for mentors (Cavell et al., 2009). In the same study, evidence of convergent validity was apparent in the moderate-to-strong correlation of child and mentor MAS scores with those of support scores from the Network of Relationships Inventory (NRI; see review below). Along with the available validity evidence, the parallel mentor and mentee versions of the scales is a clear strength of the MAS. As with the MYAS, the focused survey provides information only on relational match quality and does not address instrumental match quality. Network of Relationships Inventory (mentee: 11 items/2 scales; mentor: 11 items/2 scales) To complement their mentor alliance assessments, Cavell and Hughes (2000) created modifications of the children’s Network of Relationships Inventory (NRI; Furman & Buhrmester, 1985) to derive two types of scores for both mentees and mentors: mentor relationship support (NRI-support) and conflict (NRI-conflict) scores. For each adultchild relationship listed in the NRI (for example, relationships with parents and teachers), children are asked to rate the level of social support they receive from adults. Conflict scores reflect the degree of emotional upset in the respective relationship, as well as open expressions of conflict (e.g., arguing). The modifications made by Cavell and Hughes result in support and conflict scores for the mentoring relationship only, with the Mentor-NRI created as a parallel version of the Child-NRI. The MentorNRI asks mentors to rate the degree of support and conflict present in their mentoring relationship. Validity evidence for the Mentor-NRI is reflected in a significant inverse correlation between mentors’ relational conflict and support scores, although support was assessed by a composite of the MAS (described above) and NRI-support scores due to the strong correlation among these scores. Evidence of predictive validity is suggested by the

child NRI-conflict scores accounting for substantial variance in teacher-rated externalizing problems at posttreatment, 1-year follow-up, and 2-year followup. Mentor NRI-conflict scores predicted teacherrated externalizing problems more modestly, with a marginal contribution at posttreatment and a modest one at 1-year follow-up. The focus on both support and conflict from both the mentor and mentee perspectives are strengths of the NRI. However, like most other available surveys, the NRI focuses solely on internal, relational MRQ. The survey must be completed through an interview, which may limit its utility within some contexts. Youth Mentoring Survey (mentee only: 43 items/ 6 scales) and Match Characteristics Questionnaire (mentor only: 66 items/15 scales) Harris and Nakkula (2003a, 2003b) developed the Youth Mentoring Survey (YMS) and the Match Characteristics Questionnaire (MCQ) as complementary assessments of match characteristics. Both surveys capture the relationship between internal match quality and match structure; the mentor survey also measures external match quality or support. Both surveys measure distinct aspects of these characteristics along a continuum from relational quality to instrumentality. The YMS yields three scores for internal match quality. The relational quality subscale measures how mentees experience interpersonal aspects of the relationship. The instrumental quality subscale assesses mentee’s perceptions about seeking and perceiving support from the mentor. The prescription subscale reflects mentee dissatisfaction with mentors’ efforts to impose a unilateral agenda. The YMS addresses match structure by assessing mentees’ perceptions of how much time is devoted to interactions with three focuses (in order from most relational to most instrumental): having fun and simply hanging out together (fun); talking about concerns and experiences (sharing); and improving mentees’ attitudes, behavior, or academics (growth). The MCQ measures internal match quality with seven scales reflecting mentors’ experiences of varyingly relational and instrumental aspects of the match: relational compatibility; closeness; capacity to handle their mentee’s issues (risk-related competence); awkwardness or disconnection (not distant); satisfaction with the relationship’s trajectory and how the match benefits the mentee (satisfaction); mentees’ openness to personal support (nonacademic support-seeking); and mentees’ openness to academic support (academic support-seeking). The MCQ measures structure as mentors’ purposes,

Assessing Mentoring Relationships   57 assessing their valuation of five types of interactions along a relational/instrumental continuum: having fun and just hanging out (fun); talking about concerns or experiences and sharing emotionally (sharing); improving mentees’ goal setting and future orientation (future outlook); improving mentees’ honesty, responsibility, and treatment of others (character); and improving mentees’ academic performance (education). The MCQ measures external match quality with three subscales: program support; the mentees’ parent engagement; and mentors’ experience of logistical and personal interference with the match. Each scale on the YMS and the MCQ is supported by a range of validity evidence, though much of it remains unpublished. Rather than specify evidence related to each scale, we highlight broader points that are indicative of their interrelated importance. Predictive validity is indicated by the scales’ correlation with match duration. Preliminary findings also indicate that a combination of internal match quality and structure scales from both instruments predict mentees’ postmatch academic performance. Convergent validity is indicated through the intuitive pattern of correlations between each survey’s internal match quality and structure scales at different points along the relational/instrumental continuum (Nakkula & Harris, 2010). Specifically, scales designed to assess more relational aspects of the match are correlated more highly with one another than with scales designed to assess more instrumental aspects of the match. Convergent validity also is indicated through correlations between the corresponding YMS and MCQ scales, and by correlations between scales from those surveys and related aspects of MRQ as assessed by the Mentor Mattering Survey and the Perceived Competence surveys (Harris & Nakkula, 2010). One reason we believe it is important to measure a full range of MRQ indicators is that findings indicate that combinations of scales from both surveys— reflecting internal match quality, structure, and external match quality—consistently yield stronger predictive power than scales measuring only one aspect of MRQ (Nakkula & Harris, 2010). Both the YMS and the MCQ are complementary mentee- and mentor-reported surveys that measure a full range of match characteristics that are considered to indicate effective mentoring and enhance mentee outcomes. While most available instruments measure general ends of the relational/ instrumental continuum (usually, specific to internal relational match quality), the YMS and MCQ are designed to assess indicators reflecting the full span of that continuum. The two surveys also provide the

only measures of structure in mentoring matches that are supported by published validity evidence. The MCQ provides the only measure of external match quality. The YMS and the MCQ have a number of limitations. First, much of the current validity evidence is presented only in unpublished evaluation reports. Second, the scales are supported by exploratory factor analyses, but have not yet been tested with confirmatory factor analyses. Third, little is known about the surveys’ prediction of nonacademic outcomes. Finally, the surveys’ comprehensive MRQ assessment renders both surveys longer than the others reviewed here. Strength of Relationship Measure (Littles: 10 items/5 scales; Bigs: 14 items/4 scales) The Strength of Relationship Measure (SoR) was developed for and introduced to Big Brothers Big Sisters of America (BBBSA) in 2008 to help assess, monitor, and improve the quality of their matches across the organization (DuBois, 2009). The Little SoR yields an overall strength-of-relationship score comprised of five components: closeness, mentees’ management of conflict in their matches (coping), mentees’ disappointment in their matches (disappointment), how safe mentees feel with their mentors (safety), and how much mentors and mentees value their matches (importance). Only coping and disappointment are assessed with multiple items. Predictive validity for the Little SoR is indicated by correlations between the component items/scales and 12-month match retention for both community-based and school-based matches. It is noteworthy that even the single-item subscales predicted match retention at 1 year. The Big SoR yields an overall mentor-perceived strength-of-relationship score, with four scales/ items reflecting mentors’ experience of the match: positive bond (connectedness), frustration with match difficulties, confidence they are having a positive impact, and closeness. As with the Little SoR, closeness is measured by one item. Predictive validity for the Big SoR is similar to that supporting the Little SoR, with scale scores correlating with 1-year retention. Findings were consistent across mentee age and gender. The surveys’ use of individual items to measure constructs is an important contribution, as it is easier to collect data with briefer surveys. However, the instruments lack published validity evidence, and careful investigation is warranted to determine the most appropriate applications of such parsimonious assessment.

58  Mentoring Relationships Relational Health Indices–Young Adult and Youth Versions (6 items, 1 scale) In our 2005 chapter we provided an overview of the Relational Health Indices (RHI), developed by Liang and her colleagues (2002) to assess the status of relational health among college women in three domains: peers, communities, and natural mentoring relationships. Our review summarized the validity evidence in support of three subscales comprising the RHI-Mentor scale: engagement, empowerment/zest, and authenticity. Each of these subscales was found to hold important implications for understanding relational health within natural mentoring relationships for college women. Please refer to our 2005 chapter of HYM for a complete summary of the RHI. Since 2005, Liang and colleagues (Liang, Tracy, Kenny, Brogan, & Gatha, 2010) created a modified short form of the RHI for youth. The validation study for the RHI-Youth version was conducted with the intent of examining relational health among adolescents transitioning to different levels of education. Confirmatory factor analyses resulted in a six-item RHI-Y-Mentor scale, which captures relational health (aspects of engagement, closeness, and authenticity) in students’ natural mentoring relationships. Friend and Community scales were part of the larger instrument, thereby allowing relational health in natural mentoring relationships to be compared with other contexts. In addition to the construct validity support derived from confirmatory factor analytic methods, the authors found that boys rate relational health lower than girls within their natural mentoring relationships, which is consistent with other findings. Further construct validity evidence was found in the positive correlations between RHI-YMentor scores and social support, school engagement, and reduced perceived stress. The authors acknowledge potential self-selection biases based on the voluntary sampling strategy used for the validation study, as well as limitations in the range of sampling, with only 6th- and 9th-grade students participating. Summary and Recommendations for Future Research The assessments of MRQ presented above mark growth for the youth mentoring field in a number of ways, and also reflect continuing limitations. More instruments are supported with validity evidence than was the case in the 2005 edition of this chapter. However, that evidence focuses almost entirely on a limited scope of MRQ. The majority of the reviewed

instruments emphasize the measurement of relational match quality, or the closeness of the mentoring alliance or bond. Given the centrality of relational closeness to the mentoring field, it seems healthy for the field to have multiple alliance-based instruments from which to choose. Instrumental match quality continues to receive limited focus, although there is progress in this area. Given the increase in instrumentally oriented mentoring programs, the assessment of satisfaction with the goal-oriented component of mentoring is clearly needed. Although research and theory point to reasons for measuring match structure or external match quality, there remains only one option for measuring these constructs. It is important to underscore the importance of MRQ assessment from both mentee and mentor perspectives; we see a modest increase in this area since 2005. The use of MRQ assessment to predict youth outcomes is another area of positive development for the field. A limitation, however, lies in the relative lack of randomassignment and longitudinal studies designed to explore differences in MRQ associated with different models of mentoring and with outcomes promoted by those models. We view this as a critical area of focus for future work in the field.

Practice Given the theory and research reviewed above, how should mentoring relationships be assessed? What should be measured? How much assessment is enough? When and how should MRQ instruments be administered? How should findings about MRQ be used? We believe the answers to these questions depend on who will use the assessments and for what purposes. In this section, however, we do provide general guidelines regarding recommended approaches to assessing MRQ. These guidelines are summarized in Table 4.3. What to Measure and With Whom? If one can only measure a single aspect of subjective match quality, it currently should be internal match quality. It is commonly viewed as a key moderator of the outcomes mentoring is intended to achieve. As the aspect of quality we know the most about, it is an important indicator, especially for practitioners seeking only to assess whether matches appear to be on track to produce results for mentees. For such purposes, indications of relational competence and satisfaction have been shown to be particularly relevant and there is

Assessing Mentoring Relationships   59 Table 4.3   Checklist for Practitioners Topic

Recommendations

What to measure

Relational-experiential match quality: Relational satisfaction, competence, and dosage are central. A broad array of options are available for measuring relational match quality and the greatest research support for doing so. Instrumental match quality: Instrumental satisfaction and competence are central. There is increasing support for measuring it. Match structure: Less is known about the relative value of measuring match structure. Preliminary research indicates it may be quite important. External support: Programmatic support is central. Other forms of support may vary in importance by mentoring model.

How and when to measure

Internal assessment: Make MRQ assessment a programmatic value, not an add-on. Integrate it into existing infrastructure/processes. Measure matches at similar intervals of match-length. Recognize that nonresponding matches are more likely to be disengaged or negative. When to measure: Assess match quality across time when possible. Assessments of compatibility, structure, and external match quality may be done early in the match. First assessments of satisfaction can be done at approximately 3 months into the match and, ideally, every 6–9 months afterward.

Interpreting and using results

Practitioners: Interpret results carefully. Use standard deviations as indicators of “normal” ranges and focus on extreme scores. Consider survey scores in the context of other insight on match quality. Researchers: Seek to demonstrate how all aspects of MRQ interact with preexisting factors to promote outcomes. Compare utility of all MRQ indicators to identify what is most useful to measure.

Funding and policy priorities

Funders and policymakers should emphasize accountability in establishing and supporting high-quality mentoring relationships for youth. This should include a focus on regular assessment of MRQ and utilization of these data for relationship support and ongoing efforts directed toward program strengthening. Support research to demonstrate how all aspects of MRQ interact to shape mentee outcomes.

increasing and intuitive support for measuring instrumental satisfaction (Nakkula & Harris, 2010). It also is important to measure objective relationship quality, as dosage is an important moderator of mentee outcomes. The theory and research presented here also emphasize the value of assessing match quality from both mentees’ and mentors’ perspectives. The most effective measurement likely will address participants’ negative as well as positive perceptions of their matches. Theory and research increasingly indicate that an understanding of how mentoring works is incomplete if it does not account for the way internal match quality, match structure, and external match quality interact with each other to promote outcomes (Karcher

& Nakkula, 2010). Because mentoring-related outcomes appear to be moderated by all of these relationship qualities, research is needed to clearly demonstrate, specifically, what type and level of quality are most important and how they may consistently be achieved—across models and populations of served youth, and specific to particular outcomes of interest. Were such research available, it could help move effective matchmaking and support out of the realm of art and into the realm of science, raising average match quality within and across mentoring programs. Such knowledge could support the development of clear procedures and formal standards. However, simply having the knowledge available likely would make a meaningful difference in itself.

60  Mentoring Relationships Administering Surveys

Funding and Policy Priorities

Match quality surveys can support efforts at programmatic quality control or rigorous evaluation research and be effectively administered by practitioners or researchers. The ones reviewed here are free with the authors’ permission (along with scoring) and simple to administer. However, both practitioners and researchers should be aware that response rates and respondents’ candor will be influenced by the degree to which those things are instilled as integral programmatic values rather than add-ons to the mentoring. Wherever possible, measurement of match quality should be integrated into existing program processes. Some indicators may be usefully measured early in the match. Compatibility and competence may be evident before a strong bond has formed. Indications of match structure and external match quality also are evident from the outset. The first assessment of satisfaction should occur no sooner than 3–4 months into the match, after the initial honeymoon period during which participants may be “on their best behavior” but before the match has had time to fully develop and early enough to address potential problems before they become severe or entrenched. A subsequent assessment should take place at 9–12 months into the match (or at the end of the school year), after the relationship has had time to mature. Finally, we recommend measuring match quality at similar points in every match within a given program to maximize comparability between matches.

If match quality moderates outcome, the most important thing a mentoring program can do is make and support high-quality matches. Yet, funding continues to be tied more substantially to the quantity of matches made and to outcomes produced. We see this focus perpetuating a vicious cycle of underperformance in the service of programmatic survival. Funding and policy that overemphasize outcomes are like a coach who overemphasizes winning; in both cases, the ultimate goal is important but the achievement of it depends on effective execution. We contend that a primary focus on appropriate execution of practices and, with specific relevance to our discussion in this chapter, their immediate consequences for mentoring relationship quality, is most appropriate in many, if not most, instances. Accordingly, we encourage funders and policymakers to hold programs accountable not only for youth outcomes but also for establishing high-quality mentoring relationships and implementing with fidelity practices that are known to foster them. Researchers should be supported in efforts to make those dynamics clear.

Interpreting and Using Results Those using the surveys, and especially practitioners, should interpret findings with careful attention to the many factors that may influence them. It is important to compare like matches, especially in relation to age and mentoring model. Disaffected and loosely engaged participants tend to be the least likely to respond, so findings from administrations that do not assess all matches may be skewed positively. In the absence of norms (which exist for no current survey), it is important to evaluate scores in the context of standard deviations from the mean. Given sufficient sample size, these may be calculated from the target population; statistics related to larger, more diverse populations may be sought from the surveys’ authors. Especially for small groups of respondents, it is important to avoid attending to small differences from the norm. Extremely low scores and refusal to participate in surveying can be useful red flags that merit timely follow-up.

Conclusion After enjoying a period during which mentoring received great attention and support, researchers and practitioners have grappled recently with reports that mentoring produces only modest effect sizes. This chapter summarizes encouraging evidence that assessing match characteristics can inform efforts to raise the quality of the average mentoring match, and with it the magnitude of the outcomes they produce. The assessment of match characteristics has progressed notably since we wrote about it in the first edition of the Handbook. A greater selection of surveys is available, and they are supported by a broader array of more meaningful validity evidence, especially related to scales’ prediction of retention and mentee outcomes. The surveys we reviewed in this chapter may be used to effectively measure internal match quality, structure, and external match quality. However, many questions remain. It will be important for future assessment of MRQ to demonstrate more thoroughly how mentoring works so that outcomes may be maximized. To do that, research must answer questions such as: What characteristics are most broadly predictive of desirable match quality and outcomes? Some evidence suggests that characteristics other than internal match

Assessing Mentoring Relationships   61 quality may be most broadly useful. Researchers also will need to investigate what constitutes optimal match quality for matches designed to promote very specific outcomes, to serve particular types of youth, or to implement particular mentoring models. Finally, research is needed to determine which items or scales yield the broadest insight, which are most effective for predicting particular outcomes, and which are essentially redundant. As research establishes which characteristics to measure and how to measure them, practitioners may make increasingly informed decisions about which to use. The current research already offers insight on how they can make, train, supervise, and support matches that can produce outcomes with larger effect sizes. Both the breadth and specificity of that insight should increase substantially over the next 5–10 years. Once norms are established, practitioners will be even more empowered to monitor themselves and produce better results. Funders and policymakers can speed these processes by supporting rigorous research and programs that incorporate assessment of match quality as an integral component of their programmatic structure and processes, and they can hold ­ practitioners accountable for using what is known about enhancing match quality for the types of children they serve and the types of mentoring they provide.

References Benard, B. (2004). Resiliency: What have we learned? San Francisco: WestEd Press. Big Brothers Big Sisters of America. (2008). Strength of relationship surveys for Littles and Bigs. Unpublished instruments. Blechman, E. A. (1992). Mentors for high-risk minority youth: From effective communication to bicultural competence. Journal of Clinical Child Psychology, 21, 160–169. Blocker, K. M. (1993). Factors in sustaining adult voluntary mentoring relationships with at-risk youth. Dissertation Abstracts International, 54(6-A). Cavell, T., Elledge, C., Malcolm, K. T., Faith, M. A., & Hughes, J. N. (2009). Relationship quality and the mentoring of aggressive, high-risk children. Journal of Clinical Child & Adolescent Psychology, 38, 185–198. Cavell, T., & Hughes, J. N. (2000). Secondary prevention as context for assessing change processes in aggressive children. Journal of School Psychology, 38, 199-236. Deutsch, N., & Spencer, R. (2009). Capturing the magic: Assessing the quality of youth mentoring relationships. New Directions in Youth Development: Theory, Practice and Research, 121, 47–70.

DuBois, D. L. (2009). Analyses of Strength of Relationship (SoR) survey data. Unpublished report for Big Brothers Big Sisters of America, Philadephia, PA. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. New Directions for Youth Development, 93, 21–57. Furman, W., & Buhrmester, D. (1985). Children’s perceptions of the personal relationships in their social networks. Developmental Psychology, 21, 1016–1024. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring programs. American Journal of Community Psychology, 30, 199–219. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work: An impact study of the Big Brothers/Big Sisters program. Evaluation Review, 22, 403–426. Harris, J. T., & Nakkula, M. N. (2003a). Match Characteristics Questionnaire, v. 2.0 (MCQ). Unpublished instrument. Fairfax, VA: Applied Research Consulting. Harris, J. T., & Nakkula, M. N. (2003b). Youth Mentoring Survey (YMS). Unpublished instrument. Fairfax, VA: Applied Research Consulting. Harris, J. T., & Nakkula, M. N. (2010). Unpublished evaluation report. Fairfax, VA: Applied Research Consulting. Herrera, C. L., Sipe, W. S., McClanahan, W. S., Arbreton, A. J., & Pepper, S. K. (2000). Mentoring school-age children: Relationship development in communitybased and school-based programs. Philadelphia: Public/Private Ventures. Arlington, VA: The National Mentoring Partnership. Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized study of the effectiveness of school-based mentoring. Prevention Science, 9, 99–113. Karcher, M. J., Herrera, C. L., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. New Directions for Youth Development, 2010 126, 51–69. Karcher, M. J., & Nakkula, M. J. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. New Directions for Youth Development, 2010 126, 13–32. Karcher, M. J., Nakkula, M. N., & Harris, J. T. (2005). Developmental mentoring match characteristics: Correspondence between mentors’ and mentees’ assessments of relationship quality. Journal of Primary Prevention, 26, 93–110. Larose, S., Cyrene, D., Garceau, O., Brodeur, P., & Tarabulsy, G. M. (2010). The structure of effective academic mentoring in late adolescence. New Directions for Youth Development, 2010 126, 123–140. Liang, B., Spencer, R., Brogan, D., & Corral, M. (2008). Mentoring relationships from early adolescence

62  Mentoring Relationships through emerging adulthood: A qualitative study. Journal of Vocational Behavior, 72, 168–182. Liang, B., Tracy, A. J., Kenny, M. E., Brogan, D., & Gatha, R. (2010). The Relational Health Indices for Youth: An examination of reliability and validity aspects. Measurement and Evaluation in Counseling and Development, 42(4), 255–274. Liang, B., Tracy, A., Taylor, C., Williams, L. M., Jordan, J. V., & Baker Miller, J. (2002). The Relational Health Indices: A study of women’s relationships. Psychology of Women Quarterly, 26, 25–35. Masten, A. S. (2009). Ordinary Magic: Lessons from research on resilience in human development. Education Canada, 49(3), 28–32. Morrow, K. V., & Styles, M. B. (1995). Building relationships with youth in program settings: A study of Big Brothers/Big Sisters. Philadelphia: Public/Private Ventures. Nakkula, M. J., & Harris, J. T. (2005). Assessment of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. Thousand Oaks, CA: Sage. Nakkula, M. J., & Harris, J. T. (2010). Beyond the dichotomy of work and play: Measuring the thorough interrelatedness of structure and quality in youth mentoring relationships. New Directions for Youth Development, 2010 126, 71–87. Norcross, J. C. (Ed.) (2011). Psychotherapy relationships that work: Evidence-based responsiveness. New York: Oxford University Press. Public/Private Ventures. (2002). Technical assistance packet #8: Measuring the quality of mentor-youth relationships: A tool for mentoring programs. Portland, OR: Northwest Regional Educational Laboratory. http://www.nwrel.org/mentoring/packets.html

Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. Thousand Oaks, CA: Sage. Rhodes, J. E., & DuBois, D. L. (2008). Mentoring relationships and programs for youth. Current Directions in Psychological Science, 17, 254–258. Rhodes, J. E., Reddy, R., Roffman, J., & Grossman, J. B. (2005). Promoting successful youth mentoring relationships: A preliminary screening questionnaire. The Journal of Primary Prevention, 26, 147–167. Shirk, S. R., & Saiz, C. C. (1992). Clinical, empirical, and developmental perspectives on the therapeutic relationship in child psychotherapy. Development and Psychopathology, 4, 713–728. Sipe, C. L. (1998). Mentoring adolescents: What have we learned? In J. B. Grossman (Ed.), Contemporary issues in mentoring. Philadelphia: Public/Private Ventures. Slicker, E. K., & Palmer, D. J. (1993). Mentoring at-risk high school students: Evaluation of a school-based program. The School Counselor, 40, 327–334. Spencer, R. (2006). Understanding the mentoring process between adolescents and adults. Youth & Society, 37, 287–315. Thomson, N., & Zand, D. H. (2010). Mentees’ perceptions of their interpersonal relationships: The role of the mentor-youth bond. Youth & Society, 41, 434–445. Zand, D. H., Thomson, N., Cervantes, R., Espiritu, R., Klagholz, D., LaBlanc, L., & Taylor, A. (2009). The mentor-youth alliance: The role of mentoring relationships in promoting youth competence. Journal of Adolescence, 32, 1–17.

5  MENTORING ACTIVITIES AND INTERACTIONS Michael J. Karcher and Keoki Hansen

Introduction There is perhaps no more common experience than the mentor of a youth asking himself or herself, “When I see my mentee today, what should we do together?” The answer to this question is likely shaped by obvious characteristics of the mentor and the mentee, such as age and gender, but also by where they meet. A male mentor with a male mentee meeting in the community is more likely to engage in sports than might a female pair meeting in the same context; yet matches of both sexes are more likely to engage in sports in a communitybased mentoring relationship than in a workplace or school context. So context and mentor/mentee characteristics play a central role in determining what pairs choose to do together. Less obvious personal characteristics as well as program-specific goals, however, also play an important role in shaping the nature of the chosen activities or focus of mentoring interactions. Where programs prescribe activities, it is easy to link activities to outcomes. But where personal choice enters the equation, youth and mentor characteristics may cloud a clear interpretation of both what leads to a given choice of discussion or activity topic and how any particular activity or mentoring topic may be associated with outcomes. Matches are often created based on common interests, and (regardless of whether this information is ever shared with the mentor and mentee specifically) matches are likely to gravitate to those activities in which they both share an interest. In addition, characteristics of the youth, such as his or her being at risk for flunking, struggling with peer relationships, and engaging in risky behavior,

may lead a mentor to stray toward discussions or activities intended to help the youth. Such conversations, however, may instead “cause” the negative outcomes they were intended to prevent (but which were already in the making). For example, as we discuss later, the use of academically focused interactions in schools are more likely to occur when mentees are struggling academically. In such a case, would research that correlates “helping with homework” and improvement in end-of-quarter grades (or attitudes toward school) suggest the improvement was caused by the help with homework? We state this problem up front because this chapter reports on a lot of research, virtually all of which is quasi-experimental or simply correlational at best, and from which no indisputable links between activities and outcomes can be made. For example, although some research studies we review report negative correlations between helping with homework and grades, correlation does not imply causation. We do not know if the homework help is causing decreased grades, or whether decreasing grades over the year led to mentors helping more with homework. Similarly, we do not know whether both the decrease in grades and the increase in homework help could be caused by a third factor. So please read this chapter with a critical eye. Program participants often have unstated or implicit assumptions, expectations, or beliefs that influence the activities chosen or the interactions that occur during a given meeting. For example, the elementary-aged mentee may view the relationship as a context for fun, while the mentor views the relationship as an instrument to help the child succeed in school. Conversely, the mentor

63

64  MENTORING RELATIONSHIPS may expect to focus on building a relationship and having fun in the match, while the teenage mentee may believe they should focus on completing tasks and getting something done (Spencer, 2006). The mentor and mentee also may bring to the match beliefs—about the nature of adult-youth relationships and about who has authorship in adult-youth interactions—that have as their nexus the individual’s own experience in families, schools, and other intimate vertical (imbalanced in power) relationships (see Keller, 2005). We believe programs will be more effective when they address these influences on activities and conversation choices in training, setting program goals, developing recruitment advertisements, matching adults with youth, and providing match support. We wrote this chapter to help program staff and mentors think more clearly about the ways in which less obvious characteristics that mentors bring to their relationships can shape how decisions get made about what matches should do together. In the first section, we present a series of theoretical lessons and principles, born both from the larger fields of psychology as well as out of focused studies of youth mentoring relationships, to set the stage for a framework that we then ­present. This framework interrelates three important dimensions of mentoring interactions—focus, purpose, and authorship—in ways that explain the two predominant styles of effective mentoring relationships found in the youth mentoring literature: the developmental and instrumental relationship styles (see Karcher & Nakkula, 2010). In the second and third sections of the chapter, we summarize mentoring research and highlight youth mentoring program practices to illustrate the utility of this framework. We examine the empirical support for the three dimensions in prior mentoring research and describe how various programs prescribe certain forms of structure to achieve their objectives in ways that are consistent with the framework. One goal we have is to consider the role of context (e.g., community vs. school setting), mentees’ and mentors’ genders, and age of mentee to generate testable hypotheses for what types of activities are best for what kinds of youth under what circumstances. Finally, we address the framework’s limitations in its ability to fully explain the diversity of influential youth mentoring research findings and program practices. These limitations suggest the frontiers along which additional work is needed to better capture what we know about how mentoring activities and interactions scaffold and buttress effective youth mentoring relationships.

Theory More than 10 years ago, DuBois, Holloway, Valentine, and Cooper (2002) identified in their meta-analysis of youth mentoring program effectiveness that providing structured activities is a mentoring best practice. Yet to date there has not been any one source covering the vast literature on youth mentoring activities and interactions. This chapter attempts to provide such a resource. In this section, we introduce a few theoretical principles from outside the youth mentoring literature and a new integrative framework specific to youth mentoring. We include theories outside the field of youth mentoring to illustrate the parallels between youth mentoring interactions and interactions in other domains, such as psychotherapy or youth friendship. When possible we provide models and findings for both structured relationships as well as spontaneous or naturally forming ones to consider whether the framework developed in this section applies equally well to natural as well as programbased mentoring relationships. Understanding Help-Intended Communication: We’re Not There Yet We begin this section by suggesting where the field of youth mentoring is not at this point in time regarding our understanding of the kinds of youth mentoring interactions that are most helpful. Unfortunately, we’ve forgotten the lessons provided by the first (and, to date, best) study of the nature of mentoring interactions on program outcomes: Goodman’s study, Companionship Therapy: Studies in Structured Intimacy, published more than 40 years ago. When Goodman (1972) published his findings, there was no field of youth mentoring research to build upon his findings or to share lessons learned with programs to improve mentoring practice. For example, his study preceded by 25 years Grossman and Tierney’s (1998) study of the Big Brothers Big Sisters program, which many view as the first rigorous study of youth mentoring. One of the products from Goodman’s research was a detailed analysis of the types of help-intended communications that were most and least beneficial in youth mentoring. The six-type framework covered the major pieces of language any helper could use. It has been used extensively for training and research on help-intended communication around the world, but because there was no youth mentoring field to receive Goodman’s findings, his work was published in psychotherapy journals and lives on to this day. Goodman and Dooley (1976) wrote

Mentoring Activities and Interactions   65 “A Framework for Help-Intended Communication” for paraprofessionals (i.e., mentors and other helping volunteers). Their framework has since been refined and extended by multiple researchers. Goodman’s framework detailed the uses and benefits of the following six response modes or “talk tools” that paraprofessionals use with their clients (i.e., mentees): questions, advisement, silence, explaining (interpretation), empathy (reflection), and self-disclosure. It is unfortunate that no mentor training (of which we are aware) includes information on these talk tools or help-intended communication responses that could provide important helping skills to mentors for use in their interactions with mentees. Field-tested analyses of these micro-level communication techniques could advance the mentoring field greatly. Mentoring as an Activity Setting Mentoring interactions can be viewed from a much more macro- or meta-perspective as well. Another mentoring research pioneer, Cliff O’Donnell (see Fo & O’Donnell, 1974; O’Donnell & Williams, 2013), introduced the concept of activity settings to help explain the ways in which community interventions work (O’Donnell & Tharp, 2012). In activity setting theory, “the context is integral to the nature and duration of the activity and provides purpose, resources, and constraints. These units of contextualized human activity are referred to as ‘activity settings’” (O’Donnell, Tharp, & Wilson, 1993, p. 504). Key elements of activity settings are the people in the activity, their social positions, the physical environment in which the activity occurs (e.g., school or community), time involved, funds and symbols associated with what is happening, or what purpose(s) the interactions serve. Collaboration in activity settings as a development catalyst. O’Donnell et al. (1993) suggest that shared decision making and collaboration are central to relationship maintenance in activity settings and that it fosters development and growth among participants. “When there is a common goal or product, people are said to be engaged in a joint activity. Activity settings in which people do not share a common goal or in which they are always in conflict are likely to be disbanded. To be productive, joint activity requires some cooperative interaction, which facilitates learning, relationships, and individual, family, and community development” (p. 505). O’Donnell calls this joint productive activity (JPA). Reciprocity, mutuality, and intersubjectivity. One essential element of JPA is reciprocal participation,

meaning each person assists and is assisted during the activity, and through the interaction, individual and group (dyadic) competencies emerge. According to O’Donnell et al. (1993), a common result of this synthesis of competencies is the development of unique approaches to an activity or a problem, and “when this occurs, the cognitive development of each participant is facilitated (Bronfenbrenner, 1979)” (p. 505). They suggest it also fosters an interdependence that is motivating, heightens productivity, and ultimately yields what they call intersubjectivity: Intersubjectivity refers to the similar basic ways that a group of people think, experience, and describe the world. Intersubjectivity develops through meaningful and often intense discourse over time (dialogism). To the degree that intersubjectivity is present, values and goals are more alike, more cooperation is possible, and greater harmony and partnership exist at some level. This intersubjective dimension of joint activity serves as a reward to its members and motivates their continued participation. (p. 506) Our question, then, is whether these elements of activity setting occur within youth mentoring and, if so, what role do the processes of reciprocity (both partners contribute, direct, and self-disclose), mutuality (both partners benefit, partly from knowing they benefit one another), and intersubjectivity (a sense of similarity and kinship that results from JPA and works to foster relationship maintenance) play? To assess the utility of activity setting theory, we later look for the presence of reciprocity, mutuality, and intersubjectivity in youth mentoring research. Advocacy: Going beyond the match (old wine, new bottles). Finally, a new horizon in research on mentoring activities has emerged, and this horizon is at the edge or beyond the immediate interactions of mentors and mentees. Emerging research suggests that when mentors advocate for the youth, by helping the youth communicate with adults in other contexts; help youth seek out or secure employment or educational opportunities; or otherwise act on the youth’s behalf outside of the match, the youth may improve more than when such efforts at advocacy do not occur. This is not surprising. Consider Mentor’s work with Telemachus—in the initial story of mentoring—and how Mentor helps the youth set sail on a voyage to find his father. The way he helps Telemachus is by looking outside the relationship at what the boy needed. But, as we discuss in this

66  MENTORING RELATIONSHIPS chapter, most research on youth mentoring activities has looked at what happens just within the match. Hierarchy of Interventions We can talk about “what happens” in mentoring relationships at many levels. To better organize the variety of happenings, we present a framework for thinking about modes of responding by mentors to youth. It is based on Wampold’s (2001) hierarchy for understanding the elements of psychotherapies. This hierarchy of interventions in youth mentoring includes program approaches or mentor styles, strategies, techniques, and mentoring micro-skills (see Table 5.1). At the top of the hierarchy are general mentoring approaches or mentor styles, like the “developmental style” or “instrumental style” described later. These reflect “theories of mentoring.” Below mentoring approaches and mentor styles are the

strategies that define them. Under strategies are the techniques that constitute the strategies. These are more concrete (and observable) rather than abstract (and inferred), such as “played basketball,” “went to the museum,” “helped with homework,” “talked about family.” Within any of these techniques will be a range of mentoring micro-skills (e.g., questioning, reflection, disclosure). Each theory or style reflects a set of essential strategies, most of which are not exclusive or unique to any one style—they are incidental strategies added by the mentor or youth. The strategy of giving relationship-building priority early in the match is essential to the developmental approach and is unique because it is proscribed in the instrumental style—if it happens, the approach is no longer instrumental by definition. Some strategies are common across mentoring approaches and are essential to each. For example, the strategy of fostering youth choice or being “youth focused” is

Table 5.1   Hierarchy of Interventions: Youth Mentoring Styles, Strategies, Techniques, and Acts

Hierarchy of Response Modes for Mentoring Interactions Program Approach

Unique (U) and Essential (E) to a Specific Approach or Mentoring Style

Common Factors That Are Either Essential to (E) or Incidental for (I) Most Effective Mentoring Approach or Mentor Style

or Mentor Style

Developmental mentor style (Morrow & Styles, 1995)

Instrumental mentor style (Hamilton & Hamilton, 1992)

Most approaches or mentor style with empirical support

Mentoring Strategies

Relational focus initially in match (U);

Goal-directed focus initially in match (U);

Youth-centered or collaborative approach (E);

Playful purpose (E); Problem/goal focus emerges over time after relationship is consolidated (U)

Future-oriented purpose (E); Relationship takes on increasing importance over time (U)

Fostering reciprocity, mutuality, and intersubjectivity (I);

Discuss matters related to youth’s family and friends (E); Express concerns and empathy for the youth (E)

Teach skills, or set goals (E); Discuss how mentoring activities relate to future goals (E)

Teach skills, or set goals (I), Express concerns and empathy for the youth (I)

Silence (E); Reflection (E)

Advise (E); Question (E)

Self-Disclosure (I); Interpretation (I)

Mentor Techniques

Mentor Micro-skills (“Talking Tools”)

Encourage youth (E)

Discuss future goals (I); Game play (I)

Note: Consistent with Wampold’s (2001) characterization of psychotherapies as having common curative factors and specific factors (unique to a given approach), here we list what elements are essential (E) in an approach, which are unique (U; found only in one or a few approaches), or incidental (I, acceptable but not necessary). (See also Frank & Frank, 1991.)

Mentoring Activities and Interactions   67 common to both the developmental and instrumental approaches and is essential to both. Fostering mutuality and reciprocity and engaging in advocacy are strategies that are common across developmental, instrumental, and other mentoring approaches, but they are incidental (i.e., not essential to either nor proscribed). At the bottom of the hierarchy in Table 5.1 are mentoring micro-skills. Goodman’s talk tools (1972; Goodman & Dooley, 1976, p. 106) reflect such micro-skills, the most basic units of communication. Goodman and Dooley suggest that these skills are easier to teach, train, observe, and rate than language acts reflecting intentions. (Intentions tend to inform strategies that include techniques, both of which we assign to the middle levels in the hierarchy. From this point on, we do not discuss language acts because the research at this level of detail in youth mentoring started and ended with Goodman’s (1972) work. Instead, we focus on approaches/styles, strategies, and techniques that have been associated with particular outcomes. As we discuss research that examined elements of this hierarchy, keep in mind that any given technique or strategy may be common to multiple effective mentoring approaches or styles. This concept is not new, but rather is consistent with a long line of psychotherapy research. Common Factors vs. Specific Ingredients: Psychotherapeutic Parallels It would be difficult to argue successfully that how we came to view youth mentoring relationships— both their form and function—was unrelated to how we have come to view psychotherapy in the 20th century. Mentoring relationships, like therapy relationships, include a helper with heightened status, an interpersonal exchange of some sort, and a meeting that typically occurs for a prescribed time period and often in a specific context (e.g., increasingly, site-based mentoring relationships meet for 1 hour, as do psychotherapy sessions). In the field of psychotherapy, two primary theoretical models have emerged that may be appropriately described as “meta-theories” (i.e., theories about theories, see Wampold, 2001). In the medical model meta-theory, specific problems are identified for a given client and the appropriate remedial action is prescribed. For example, if an individual’s depression appears to reflect the presence of distorted (say, overly exaggerated) fears or beliefs, then cognitive therapy is prescribed to address these symptoms. In the absence of any specific (theoretically relevant) cognitive or emotional symptoms, psychotropic

medications might be prescribed. In the medical model, two concepts are central. First, the problem is diagnosed. Second, based on a theoretical model or assumption, a specific remedy is prescribed. The quality of the therapeutic relationship is of minimal importance, typically relevant only to the degree to which it facilitates or impedes the implementation of the prescribed remedy or procedures. The alternative to the medical model metatheory is the common factors meta-theory of psychotherapy effectiveness, which others have already brought to bear to explain mentoring relationships (Frank & Frank, 1991; Spencer & Rhodes, 2005). In this approach, the essential ingredients of the medical model (accurate diagnosis and the appropriateness of the remedial procedure) are viewed as secondary to the nature of the interpersonal relationship that forms between therapist and client. A meta-analysis of psychotherapy studies conducted by Wampold (2001) revealed that the choice of remedial procedure (i.e., the psychotherapy approach chosen) made virtually no difference once relationship, therapist, and client factors were taken into account. Most important were the strength of the therapeutic relationship (alliance), the therapist’s belief in his or her theoretical model (allegiance), the therapist’s ability to apply the model (adherence), and the client’s hopefulness about the possibility of change. The extension to mentoring might be to suggest that it matters less whether a program is goal focused, relational, educational, or recreational. What matters most is the mentormentee relationship, mentors’ consistency and communication skills, and the mentees’ belief in the value of mentoring. Rogers’s Client-Centered Approach: Conditional and Unconditional Support. The exemplar of the common factors approach to psychotherapy is Rogers’s clientcentered approach. Rogers (1957) proposed three necessary and essential conditions for psychotherapeutic change. These include the therapist’s empathy, genuineness, and unconditional positive regard. A fuller explanation of this approach and its relevance to youth mentoring can be found in Spencer and Rhodes (2005). What we highlight here is how these three elements relate to mentoring interactions. Unconditional positive regard is the consistent responding by one individual to another in a way that implies that regardless of what you say or do, you are valuable, worthwhile, and important. In the case of mentoring, a youth might have broken the law, gotten into a fight, or failed a class, but ultimately that youth remains important and worthy of affirmation nevertheless. Such unconditional affirmation does not need to deny the failings or

68  MENTORING RELATIONSHIPS misbehavior of the child, but rather it places them as secondary to the goodness of the youth. Communicating conditional positive regard occurs when mentors focus primarily on the youth’s successful completion of a task (e.g., class grade, completion of a project, demonstration of a specific skill). If youth feel their performance is what the mentor cares most about, or if youth feel that their mentors are most concerned with completing a task or getting resolution to a youth’s problem, then the problem, not the child becomes the focus of attention. For youth to feel affirmed, they feel compelled to demonstrate competency of some kind. In this way, the establishment of goals or the need for the youth to demonstrate mastery of a goal can be antithetical to the provision of unconditional positive regard and, thus, counterproductive. Problem-Behavior Theory: Conventional vs. Uncon­­­­­­­­­­­­ ventional Interactions. Adults and youth bring in different views of the world, both because of developmental differences (see Noam et al., this volume, Chapter 7) and because the adult represents and typically espouses the conventions of adult society and culture, whereas youth typically embrace the more playful prerogative of youth and peer culture. As a result, what often happens is that competing purposes for the pair’s interactions emerge. In a pair with a middle-aged mentor and a preteen mentee, it’s not uncommon for the mentor to want to engage in an activity intended to prepare the youth for his or her future. Conversely, the child may simply want to play and have fun now. Attention to the distinction between playful and serious interactions has a long history in the literature on problem behaviors. Jessor and Jessor’s (1977) problem-behavior theory suggests that youth trapped in a cycle of delinquent behavior tend to underemphasize the importance of conventional activities, such as those that meet the expectations adults hold for youth and that serve a society-maintaining function. Conversely, unconventional activities, while often destructive in nature (e.g., substance use, delinquent misbehavior) reflect activities and commitments that are driven by the immediate needs of youth to have fun in the moment, to affirm one another’s status, and to demonstrate each other’s social competence in the present. However, being playful does not mean being delinquent, and having fun need not be destructive in nature. In fact, Erik Erikson wrote in the later years of his life that the youth who can help an adult relearn to play can help that adult achieve psychological integration and wholeness. So, in this sense, being playful can serve an important purpose for adults.

The TEAM Framework: Integrating Focus, Purpose, and Authorship In this chapter, we focus on the mentoring strategies and techniques reflected in the Theoretically Evolving Activities in Mentoring (TEAM) framework that we use to organize the rest of the chapter. The TEAM framework integrates Rogers’s conditional and unconditional support, Jessor’s problem-behavior theories, and Benson’s “Voice and Choice” approaches by revealing their interrelationships with regard to the strategies of focus, purpose, and authorship. Focus. Since the term mentor was first used, in the story of the Odyssey, there have been two primary manifestations of mentoring—through relational and goal-directed interactions. When King Odysseus left his homeland to engage in the Trojan War, he asked his friend and confidant Mentor to watch over his dominion. Retellings of this story have suggested the king gave specific instructions to provide guidance and structure to his son, Telemachus. The adaption of this story has likely come about as a result of the need for us to think about the roles associated with youth development, as if Odysseus saw Mentor as someone who could provide the paternal functions of guidance and structure in his absence. A close read of the story, however, reveals that Mentor’s most important work was to nurture and support Telemachus. (Of course, these functions were provided by the Goddess Athena in the guise of Mentor, implying these reflect a maternal function; see Karcher & Herrera, 2012.) So from the very beginning we are left to wonder whether a mentor’s role is more to nurture the youth through a close, supportive relationship or to provide guidance, instruction, and direction through effective apprenticeship. In more recent developments of programbased mentoring in the latter half of the 20th century, two different program types emerged. Some programs were more structured and goal oriented, reflecting the mentee-as-apprentice model (e.g., Hamilton & Hamilton, 1992), whereas others were far more relationship based, in which the supportive functions of an older sibling or friend were emulated (e.g., Big Brothers Big Sisters; Morrow & Styles, 1995). In the Big Brothers mentoring movement, the founders viewed the formation of a close relationship as pivotal and primary, while other forms of material support, training, and even advocacy were secondary to the mentoring enterprise. “In all this work of the Big Brothers, personal relationship has

Mentoring Activities and Interactions   69 been the kingpin. Its whole success has been due to the personal equation as expressed between the man and the boy” (Coulter, 1913, p. 268). This observation and its veracity were confirmed later by Morrow and Styles (1995), whose research is described below. A mentor whose focus is relational is one who conveys an enduring interest in the mentee—interest in what is happening in the youth’s life, in what the youth values, and in spending time with the youth. Having fun; asking about a mentee’s family and friends; exploring a mentee’s interests, hobbies, and competencies all convey an unconditional affirmation of the youth. This approach says to the youth, in effect, “regardless of what you do or how successful you are in school, I will find you interesting, of value, and a pleasure to spend time with.” This is consistent with Rogers’s unconditional positive regard approach. A mentor whose focus is goal directed typically views the accomplishment of a goal, the completion of a task, or the achievement of a skill as the primary reason the relationship has been formed. This belief may result from the program’s prescribed use of a curriculum, the suggestion by other stakeholders in the child’s achievement (parents, teachers), or the mentor’s own beliefs about what youth need to be successful. Goal-directed interactions are not inherently bad, but they do increase the likelihood the mentor will act in prescriptive, unilateral, top-down ways that hinder rather than help relationship formation. In fact, it is so important that mentors approach goals effectively that this Handbook includes an entire chapter covering the topic. Purpose. The purpose of an interaction reflects the needs of the individuals involved and the reasons why each person thinks he or she is interacting. Purpose takes on two dimensions, a temporal one (now or future) and a values one (adult, societal conventions or playful, fun, youthful). A conventional interaction serves to help the youth to become better able to make a useful contribution to society. Typically, this means the interaction serves a future purpose, when the youth enters society as an adult. Conversely, if the purpose is simply to have fun and enjoy being together, then the purpose has a more playful (youthful) orientation and is more immediate—it does not specifically serve some behavior, attitude, or skill that will be useful in the future.

Note that a conventional purpose and a goaloriented focus are not always synonymous, although they often accompany one another such as when the goals to be achieved are related to learning to write a resume, succeed academically, or interact more maturely in relationships. By contrast, learning to shoot basketballs is goal focused but with a playful purpose.1 Over the course of a relationship, and perhaps even during every youth mentoring meeting, a balance of both types of activities—future-oriented, adult-purpose-serving activities and present-oriented, youth-purpose-serving, playful activities—seems most likely to leave both parties feeling heard and validated. It also conveys that this relationship is neither just to have fun nor just to learn something, but a balance of both. Authorship. The degree to which each member of the mentoring relationship guides the evolving narrative of their story or time together conveys the authorship of the interactions. When youth make decisions— acting as authors—empowerment occurs. When both collaborate, connection deepens. When mentors alone drive the decision making, mentees can feel invalidated, interchangeable, along for the ride. The sense of feeling validated is powerful. Therefore, the youth’s feeling of being an equal partner in the relationship is critical to cultivate, regardless of whether the relationship has a relational or goal-directed focus, and a playful or conventional purpose. The instrumental and developmental styles, which each began with a very different focus, were highly effective in part because in both styles the mentors were youth centered. In both cases, mentors focused on the expressed (and sometimes not expressed) interests, concerns, desires, and suggestions of the youth. In the emergent field of positive youth development, this approach has come to be called giving the youth “voice and choice” (Lerner et al., this volume, Chapter 2). By putting the youth at the center in all decisions made about what the match will do, the youth is likely to be more invested in the activity, making it more meaningful or, in the language of common factors, instilling hope in mentoring. We find useful Selman and Schultz’s (1990) definition of collaboration as the hallmark of effective mentoring interactions. As one mode of authorship, collaboration must be distinguished from impulsive, unilateral, and cooperative interactions.

1 As youth age, their personal concerns gravitate toward conventional purposes—they want to graduate and get a job, for example—while the adult mentor may want to have fun in the match. In this way, conventional vs. playful is not synonymous with adult oriented vs. youth oriented.

70  MENTORING RELATIONSHIPS When either person asserts his or her point of view and determines the course of action, the interaction is unilateral (such as militarily, when one country attacks another unilaterally). The deliberate, often verbalized intentions of unilateral actions differentiate them from impulsive interactions, which lack any thought at all. More mature still is when the needs, interests, and desires (i.e., perspectives) of two people are coordinated through cooperation (e.g., turn taking). But when each considers the interests of the other and tries to flex or accommodate so that they can both find meaning, satisfaction, and purpose in their interactions, then they have collaborated and created something altogether new and unique to their relationship. Mentoring relationships that collaborate—where each participant is satisfied, feels heard, and is pleased with the outcome of the negotiation—empower and strengthen the relationship and convey an affirmation of the other person and his or her perspective. Whereas relationships that are unilateral or coordinated (e.g., use turn taking), they may not be as effective as ones that are collaboratively guided.

Research Using the TEAM framework to organize this section, we examined youth mentoring research in terms of the five key elements (relational vs. goal-directed focus; playful vs. future-oriented [conventional] purpose; and collaboratively made decisions). We conducted literature searches using PsycINFO (including the ProQuest Dissertations Database) unrestricted by participant ages, publication dates, or document type. Each search looked for the term mentor in the title, and either youth or children in the abstract. Then additional terms in abstracts were sought for each of the five elements of the framework (the number of unique references identified is noted in brackets): relational/ship or psychological/social [14]; goal or directive [20]; play or future [15]; collaborate/tion or interaction [6]; discussion or activity/ies [146 total; kept 34 scholarly articles]. This approach yielded 57 unique articles. Of these, 7 were excluded as referring to other forms of mentoring (e.g., new teachers), and of the remaining 50 that were reviewed, 11 were dissertations. In addition, regular searches of new youth mentoring articles in other databases also were conducted between 2009 and 2011, which yielded an additional 54 relevant articles. In our review of the research, we identify whether the outcomes reported in a study, or the

outcomes we choose to highlight, reflect proximal outcomes (e.g., relationship quality, duration of match)—those presumed to lead to or enable distal outcomes—or the actual distal outcomes that programs aim to affect and that program stakeholders tend to care more about (e.g., grades, attendance, behavior, self-esteem; Karcher, Kuperminc, Portwood, Sipe, & Taylor, 2006). Some research reports present associations between activities and key outcomes, like grades; others report associations with those proximal outcomes thought to lead to these distal outcomes. Developmental vs. Instrumental Mentoring Styles Developmental Style. Many think the mentoring research began with the Public/Private Ventures (P/ PV) study of the Big Brothers Big Sisters (BBBS) program, though we’ve found research that predates this report by 25 years. Morrow and Styles’s (1995) research, which was the qualitative arm of P/PV’s dual qualitative-quantitative approach to studying the BBBS program, is one of the best reports on the importance of relational interactions in establishing strong mentoring relationships. They collected observations and interviews with 82 communitybased matches over 9 months and coined the term “developmental mentors” to convey a style of relating with youth that focused on relationship formation, and that became the hallmark approach espoused by BBBS worldwide. We use the Morrow and Styles study to illustrate the difference between a mentoring approach or mentor style and specific strategies or techniques. Many people have written about the “developmental style” as if it was a single strategy—focusing on the relationship. But, in fact, it is a true example of a mentoring style in that it includes at least three essential strategies. These strategies are forming a relationship (through the techniques of talking about family, friends, and other concerns), taking a youth-focused approach (allowing the youth to guide or co-lead), and changing over time from largely relational to more goal directed once the relationship is formed: These relationships were given the label “developmental” because the adult partner in the match focused on providing youth with a comfort zone in which to address a broad range of developmental tasks—such as building emotional wellbeing, developing social skills, or gaining straightforward exposure to a range of recreational and cultural activities. Developmental volunteers responded flexibly to their youth, adjusting to any preconceived notions as to the

Mentoring Activities and Interactions   71 reality, circumstances, and needs of their younger partner. Furthermore, these volunteers intentionally incorporated youth into decision making about the relationship, allowing them to help choose activities and have a voice in determining whether and when the adult would provide advice and guidance. (Morrow & Styles, 1995, p. 19) The preceding excerpt reveals two essential strategies of the developmental style—taking a relational focus and coauthoring their story, with youth’s needs given priority. The excerpt that follows reveals the strategy of establishing a rapport and relationship with the youth, allowing problems to emerge, and then shifting from a relational to a goal-directed, problem-solving focus: [A]fter relatively extended and pacific periods primarily devoted to relationship-building—that is, to establishing trust and partnership, and enjoying activities—the majority of youth in developmental relationships began to demonstrate a pattern of independent help-seeking in which they voluntarily divulged such difficulties as poor grades or family strife . . . once their relationships were crystallized, nearly three-quarters of the developmental volunteers were successful in involving youth in conversations or activities that targeted such key areas of youth development as academic performance and classroom behavior. (Morrow & Styles, 1995, p. 20) This style, present among matches that lasted longer and reported greater satisfaction, was contrasted by the authors with a style they called prescriptive. The prescriptive style included among its strategies a problem focus from the beginning and a mentor-driven approach that was top-down, lacking collaboration. Unfortunately, this style has been confused with the strategies of being goal directed, overly serious, and nonrelational. Fortunately, there are examples of effective programs that include a goaldirected focus, which is important because older youth often prefer such approaches (Noam et al., this volume, Chapter 7, elaborate this point). Instrumental Style. In their study of a workplace apprenticeship program, Hamilton and Hamilton (1992) also used a combination of qualitative observations, interviews, and outcome data to identify more and less effective mentoring styles. Unlike the BBBS study by Morrow and Styles (1995), which focused on children in elementary and middle school, Hamilton and Hamilton’s program

provided apprenticeship mentoring to older youth that included, by necessity and design, a goal-directed focus. But what they called the “instrumental style” also included two other strategies, including a youthcentered or coauthorship strategy and a shift in focus over time from being goal directed (skill development focused) to more relational in nature. In contrast to the BBBS study with younger children, Hamilton and Hamilton’s work with teens found that those mentors who saw their primary purpose as developing a relationship with their mentees at the start of their relationship were least likely to meet regularly, whereas “the mentors who seemed best able to overcome the frustrations of their task were those who combined the aims of developing competence and developing character” (1992, p. 548). It is for this reason, they suggest, that for high-school-aged youth, mentoring is more appealing and more effective when “it occurs in the context of joint goal-directed activity” and when “the relationship develops around shared goals and actions more than purely social interaction” (2005, pp. 352–353). In terms of the TEAM framework, as shown in Table 5.2, we see that the developmental style is reflected by the middle position of the second row, whereas the instrumental style reflects the middle position on the third row. The middle position reflects the collaborative, youth-focused approach taken. The developmental style starts on the second row (more relational focus) and moves to row three (more goal focused over time), while the instrumental style does the opposite. The Importance of Balance in Interaction Focus, Purpose, and Authorship Two other studies, one in school and one in community settings, reveal the importance of a mentoring style that balances the strategies of focus, purpose, and authorship. The 1998 BBBS impact study by Grossman and Tierney, which followed Morrow and Styles’s qualitative studies, included several measures of mentoring interactions to allow researchers to further study interaction styles in the BBBS community-based mentoring program. Langhout, Rhodes, and Osborne (2004), in their secondary data analyses of the 1,138 youth (mean age of 12, or 6th grade), identified four distinct types of relationships: moderate, unconditionally supportive, active, and low-key. They found the most effective type of relationships were those labeled moderate, which included the strategy of “moderate levels of structured conversations around goals” and involved engaging in “slightly fewer activities with their

72  MENTORING RELATIONSHIPS Table 5.2   TEAM Framework Presented in Terms of Strategies (and Techniques) and Roles TALKING

PLAYING Banter (Engage in joke telling, silly play, goofy unstructured games, or other nonrelational [impersonal] but light, funny, or entertaining interactions or conversations.)

Behavior (Discuss youth’s behaviors that were related to problems with peers, teachers, adults, specifically misbehavior. Some attention paid to relationships but largely the mentor’s view of what or where the problems are.)

Casual conversation (Discuss sports, weekend activities, holiday plans, fun things to do in the community, neighborhood. Omits talk of very personal matters, but both talk about relationship with parents, friends, and the like.)

(Acts like preacher or principal)

(Acts like a peer)

Conversation on social issues (Mentor discusses news, poverty, local events, religious or cultural issues that relate to the youth. Mentor tries to foster attitudes that will help the youth. May use prevention activities to reach goals.)

Listening and learning (Discuss mentee’s hobbies, interests, feelings. Mentee shares personal and important information while mentor listens, is empathic, affirms, and sometimes volunteers own similar personal experiences.)

Creative activities (activity) (Engage in relationshipstrengthening conversations while doing youth-suggested or fun tasks like sports, board game play, drawing, arts and crafts, reading and writing for fun, etc.)

(Acts like a role model or counselor)

(Acts like a developmental mentor)

(Acts like a playmate)

Academics (Discuss grades, school, testing, etc. or engages in preplanned academic support activities like reading or tutoring on a topic not initially proposed by mentee.)

Future focus (Youth-initiated discussion about college, careers, jobs, goals, dreams, etc., or shared problem solving that may include mentor advocating for youth outside immediate match.)

Indoor and outdoor games (activity) (Play board games, cards, chess, Uno, checkers, computer games, or puzzles inside, or catch, basketball, or soccer outside.)

(Acts like a teacher or tutor)

(Acts like a comedian)

(Acts like a teammate)

(Acts like an instrumental mentor) Attendance and “stay-in-school” discussion (Conversations initiated by mentor about topics of concern mostly to adults.) (Acts like a vice principal or parole officer)

Learning, school, or job skills (Future-oriented, youth-initiated learning such as in job, computer, reading, or writing skills; taught by and practiced with mentor.)

Sports, athletic, or outdoor game (activity) (Mentor teaches or coaches basketball, soccer, tennis, etc., or other in-the-moment activity of interest to youth.)

(Acts like a master journeyperson)

(Acts like a coach)

LEARNING

mentors than did other groups” (p. 299). Youth matched with unconditionally supportive mentors (what the Rogerian model described earlier would suggest) did not see benefits and reported increased parental alienation. Similarly, Keller and Pryce (2012) examined mentoring relationships in a BBBS school-based mentoring program using both qualitative and

DOING

quantitative data involving 26 matched mentee/ mentor pairs. They identified four distinct types of mentors: teaching assistant (main strategy: tutoring), friend (main strategy: engaging), sage (main strategy: counseling), and acquaintance (main strategy: floundering). Only the latter style of relationship, the acquaintance, was unsuccessful. In terms of the TEAM framework, this style was

Mentoring Activities and Interactions   73 impulsively or unilaterally (not collaboratively) “authored” and lacked both a goal and relational focus. Goal-directed matches in schools are not uncommon, as Keller and Pryce (2012) illustrated. The teaching assistant mentors were in a school program that encouraged mentors to provide academic support (tutoring strategy) and saw this as the mentors’ critical role. These mentors focused on providing academic help, and their relationships were goal directed in nature. Difficulties in some of these relationships emerged when the expectations of the mentee (e.g., for a more relational focus, fun purpose, or collaborative interactions) did not match the mentor’s academic focus. However, some mentees appreciated the homework and tutoring assistance. Our guess, however, is that these mentors (who fall in column one, row three of Table 5.2) were not ultimately viewed as mentors, per se, by youth unless they included a relational focus and become more reciprocal or collaborative in their decision making over time (i.e., moved up a row and to the center column in Table 5.2). The friend mentors were relational in focus and directed their energy toward building a relationship and connecting with their mentees. Some achieved a more balanced focus by doing some homework or tutoring, or by assisting the youth with other problems, but the majority of the time was spent on relational activities (the developmental or playmate style in row two, columns two and three). Sage/counseling mentors were similar to the friend category; however, they were the adults in the relationship and were more likely to offer advice to their mentees, and their mentees were more likely to share verbally. Youth matched with the sage/counseling mentors reported the strongest relationships, along with the strongest outcomes. However, note that the study was small and only 6 of the 26 mentors were classified as sage mentors. These similar findings underscore the importance of balance—a strong relationship is important, but youth also need guidance and structure for the relationship to be effective in terms of outcomes deemed important to funders and communities. Focus: Relational, Goal Directed, or Balanced Karcher, Herrera, and Hansen (2010) tested the relative contributions of goal-directed and relational

2

activities on relationship quality in school-based mentoring using data from 568 mentees who participated in Herrera et al.’s (2007) study of the BBBS school-based mentoring program. The activity data were drawn from mentors’ self-reported mentoring activities and conversations at end of year, using the SMILE log (Karcher, 2007), which measured each of the types of mentoring techniques illustrated in the 12 cells in Table 5.2. The authors found that both relational and goal-directed conversations were positively associated with mentor-reported relationship quality; however, the relationship between relational conversations and relationship quality was three times larger than for goal-directed conversations, suggesting relational interactions made a bigger contribution to mentor-reported relationship quality than did goal-directed interactions but that both were important.2 Use of the same activity log within the Communities in Schools (CIS) mentoring program with 456 Latino youth revealed similar findings about the usefulness of relational interactions, but also suggested negative effects of academic-goal-related conversations and a developmental trend suggesting such goal-directed activity may be more likely to occur in conversations with older youth and for boys (Karcher, 2004). In another report from the CIS study (Karcher, 2007), 224 mentors completed the activity log weekly, and midway through the spring of the first year of the study the mentees completed a relationship quality scale, which included the scale “Feels valued by the mentor.” The CIS study, unlike other recent school-based mentoring studies, which focused on younger youth, included a sample with over half of the mentees in high school. Karcher reported an increase in goal-oriented mentoring activities from elementary to middle school and high school, and a concurrent decrease in mentees’ reports of feeling valued. Boys, who received more goal-directed activities, particularly in high school, reported feeling less valued than did girls at all age groups. Despite the evidence of the importance of a relational focus in establishing strong relationships in schools and community-based matches, there has been a move toward greater inclusion of structure and goal-directed activities in youth mentoring programs. This shift may be partly the result of DuBois, Holloway et al.’s (2002) meta-analytic

This study focused on goal-directed and relational interactions but cannot speak to the relative utility of the developmental and instrumental approaches because we cannot tell whether relational or goal-directed interactions predominated early in the relationships. Without assessing this strategy, we cannot suggest that this research supports the use of the developmental style in schools. To do so would require evidence that relational interactions came before goal-directed activities.

74  MENTORING RELATIONSHIPS finding that programs that included structured activities had an increased overall effect size of .22 (compared to .11 for programs without structured activities). For example, at least three programs targeting girls in particular have incorporated activities aimed at helping girls address the demands of peer relationships, think about career opportunities, and foster “voice” and leadership skills—all within the context of (or in conjunction with) a close, ongoing relationship with an older female mentor, such as through Cool Girls, Inc. (Kuperminc, Thomason, DiMeo, & Broomfield-Massey, 2011) and GirlPOWER! (DuBois et al., 2008), which we describe in the Practice section. One possibility is that practice evidence may have led to the creation of goal-focused mentoring programs for girls, who tend to be more comfortable with relationshipfocused interventions but who may benefit most from goal-setting interventions. Purpose: Playful Doing vs. Serious Talking The activity log used in these two studies was built on the early work of Parra, DuBois, Neville, Pugh-Lilly, and Povinelli (2002), who found that activities were more predictive of positive outcomes than were conversations among young mentees. The authors collected data on 50 matches monthly for a year in a community-based BBBS program. These were young mentees (7–14 years; average, 4th grade), suggesting that the developmental style (which Morrow and Styles also found with young mentees) should be best. Parra et al. (2002) assessed four conversation foci (behavior, relationships, casual conversation, social issues) and three activities (sports, recreational, educational/ cultural). From the perspective of the mentees, the best predictor of the relationship continuing and of youth benefiting from the match was playing sports. Both mentors’ and mentees’ ratings of how often they engaged in the other two activities (recreational, educational/cultural) predicted youth’s reports of benefiting from the match. All but one of the correlations between discussion topics and youth outcomes were not statistically significant, though mentors felt that discussing relationships predicted youth benefiting from the relationship. Thus, some evidence suggests that a relational focus best predicted outcomes. By far the strongest finding in this study was that for young mentees, “doing” (bottom right cell in Table 5.2) beats “talking” (top left cell). Hansen (2005) examined 201 peer mentoring relationships as part of a larger study of effective school-based mentoring practices. The mean age of

mentees was 10 years (78% were White). Hansen found that 64.2% of high school matches in programs that focused primarily on relational activities continued on to the following school year, compared to only 18.5% of matches that focused on goal-directed activities. This same pattern was not found for adult mentors. The expectations of engaging in relational activities, for both the mentors and the mentees in peer mentoring programs, may be particularly strong. When this expectation is not met, matches may become disengaged. Other research on activities and discussions looked at the interaction of the two and revealed that for many youth, doing is an important element of effectively being relational. Hansen and Corlett (2007) examined weekly activity reports by mentors in a BBBS school-based program in which adults met weekly with children for about an hour. The study involved 324 youth (half were White) with a mean age of 10.7 years (i.e., 4th graders). The mean age of the mentors was 23 years, 70% were female, and 77.4% were White. Six measures were used to measure relationship quality, including positive feelings and negative feelings rated by both the Bigs and the Littles, premature closures (closing before the end of the school year), and match continuation (matches that continued into the following school year). Their goal was to assess the correlation between these indicators of relationship quality and activities (talking and doing something). One of the activity log record options for mentors to mark was “just talked.” Mentors who reported that they just talked as an activity and did not also report engaging in a specific activity had weaker mentoring relationships compared to mentors who reported engaging in specific activities while they talked. Hansen and Corlett (2007) then examined how relational and goal-directed activities related to relationship quality. Relational activities, such as playing games and working on craft projects, were more strongly associated with match relationship quality than were those that were goal directed in nature, such as spending more than half of their time on tutoring or working on homework. Hansen and Corlett (2007) found the correlation between relational activities and match quality was moderated by gender and age. When female mentees engaged in crafts, it was linked to a stronger relationship, whereas male mentees engaging in craft projects had weaker reports of relationship quality. For elementary-age children, one-on-one games (involving just the mentee and mentor) and crafts led to stronger relationships, whereas cross-match activities (in which more than one mentee/mentor pair was involved) led to weaker relationships. In contrast,

Mentoring Activities and Interactions   75 cross-match activities, especially crafts, led to stronger relationships for middle school children. Finally, consistent with the balanced approach hypothesis, Hansen and Corlett (2007) found that a balance of the types of activities moderated the activity–match quality relationship. Matches that spent more than half their time playing games had weaker relationships than matches that spent about half their time or less playing games. Here we find that overuse of activities may be too much of a good thing, possibly precluding relationship formation. This supports the finding by Langhout et al. (2004) and Keller and Pryce (2012) that mentors who balanced relational and goal-directed activities were the most effective. Too little research has linked distal outcomes with activities serving different purposes. It is possible that fun activities predict relationship quality (or perceived benefits) but do not similarly predict distal outcomes such as grades, behavior changes, and conventional attitudes. For example, with a group of elementary school children whose age would predict the need for interactions with a playful purpose, Wyman et al. (2010) used “schoolbased mentors” to provide structured lessons and role modeling in social and emotional skills through the Rochester Resilience Project. Their results revealed the usefulness of the more conventional approach of role modeling mentors. Over 14 weeks, mentors of 226 students in kindergarten to 3rd grade used a conventional, goal-directed curriculum to teach social and emotional regulation skills aimed at improving peer relationships. Using multilevel modeling to account for dependencies in the data across schools, the effect sizes ranged from .31 to .47, nearly double the impact of most other school-based mentoring programs (see Wheeler, Keller, & DuBois, 2010). Thus, this project enlisted mentors in role modeling conventionally oriented skills and lessons, downplaying the relationally focused interactions, and yielded larger impacts than many other programs. Some observers may argue, however, that this was not really a mentoring program, and that larger effect sizes often accompany focused interventions whose outcomes are targeted directly by the intervention. Depending on the context, the youth’s age, gender, and skills (academic and social), and the purpose of the program, a goal-oriented, conventionally directed program may be most appropriate, while for another constellation of context, youth, and program characteristics, a more relational, fun, and activity-based program may be most appropriate. The evidence seems to suggest that programs that achieve a balance of these characteristics do

best. As Nakkula and Harris (2010) found in their study of relationship quality and self-reported activity focus data from a BBBS program, where the focus is fun, the relationship-based match quality will be strongest, while programs with a focus and purpose that are more goal directed and conventional are strongly correlated with youth satisfaction with the match’s ability to help them achieve their goals. The question really boils down to, what is it that the youth wants and needs? As in all relationships, mentoring appears to have a natural inclination toward playful, fun interactions, and such interactions appear to be relationship enhancing. Pedersen, Woolum, Gagne, and Coleman (2009) conducted a 10-year longitudinal study with 639 mentored children aged 4–18 years (50% White) from primarily low-income singleparent families. At the beginning, the matches were asked to focus their match on one of three different areas: healthy lifestyles (camping, biking, etc.), academic enrichment (library, homework, etc.), or fun and friendship (movies, pizza, etc.). During the match relationship, mentors completed activity logs that were then coded for the three different focus areas. Pedersen et al. found that the matches generally focused more of their time on their assigned areas than on the other two areas. But across all matches, when they deviated from their assigned activity area it was most often to engage in fun and friendship activities. Additionally, matches that focused on fun and friendship spent more time together than did those matches focused on the other two areas. Authorship as Collaboration In another study, only the matches with high school mentors (N = 212) were examined to test the associations between decisions about activity choice and relationship quality (Karcher et al., 2010). Collaborative choices were those in which the mentor and the mentee chose together; unilateral choices were made by staff, teachers, mentors, or mentees alone. Youth were given three relationship quality measures: youth-centeredness, youth emotional engagement, and youth dissatisfaction. Mentors completed the Relationship Quality Scale, Mentee Support-Seeking scale, and an Activity Log on which they indicted the selection approach (unilateral selection by mentor, mentee, or staff, or the collaborative approach). The majority of unilateral decisions in matches were made by mentees (n = 47) and program staff (n = 49); very few were made by teachers (n = 3) or mentors alone (n = 9). Therefore, only the “mentee chose” and “program staff chose”

76  MENTORING RELATIONSHIPS categories were used in the analysis. Collaborative matches (n = 87) had higher relationship quality than unilateral matches (n = 96) on three of the five scales. Greater dissatisfaction was reported by youth in both of the unilateral matches (mentee or staff deciding). Collaborative matches had significantly higher relationship quality assessed by mentor report, and mentors in these matches also felt “mentees more actively sought out their support.” It’s Finally Time to Consider the Lessons of Goodman’s Companionship Therapy Study In 1972, Gerald Goodman published the findings from his study of mentoring, which he called “Companionship Therapy.” In his study, 100 5thand 6th-grade boys met twice a week for 3- to 4-hour visits over an 8-month period (average length of relationships was 141 hours). Many of the lessons learned about the impact of mentoring “companionships” foreshadowed debates and research only now being conducted. As just a few examples, more troubled boys gained more than less troubled boys, and Black youth benefited more than White. Outgoing mentors—those who were assessed as more open and less guarded—and those who sought out more training had bigger impacts on their mentees, partly, Goodman surmised, because these mentors used more self-disclosure (mentoring talk tool). Goodman also identified the characteristics of mentors that suited some youth best. The four dyad types he studied varied in terms of mentor and mentee outgoing versus shyness. Goodman found mentors who were more outgoing, less quiet, and less rigid had mentees who benefited more from the program (Goodman, 1972, p. 241). “Outgoing boys with quiet counselors gain considerably less, but the group lowest in improvement was the double quiet sample” (p. 198). All of these lessons seem to put us right where we are now as a field—ready to understand for whom mentoring works best, with what mentors, in what context, and for what outcomes. Most relevant for this chapter, however, is research on the mentoring process—those activities and conversations that occur over time in more and less successful matches. After each session, the mentors (called activity counselors in the Goodman study) completed a Visit Report form. This not only tallied the activities and conversation topics that occurred (as in Hansen, 2005; Karcher, 2007; and Parra et al., 2002), but also assessed i­nterpersonal closeness. The mentees/“boys” and their parents provided similar information at the end of the companionship. Goodman examined patterns in closeness and interpersonal feelings at early, middle, and

late in the relationships. He found that match attrition was highest (a) with quiet mentors, (b) when the range of activities the match engaged in was more restricted, and (c) when less personal activities (e.g., movie watching) occurred frequently and more personal activities (e.g., sharing a meal) occurred infrequently. Goodman’s findings regarding collaboration, self-disclosure, and intimacy are perhaps most relevant to our framework and this chapter. Although Goodman’s approach was client centered and the program encouraged collaboration, evidence of the importance of collaboration, particularly early in the match was pronounced. Early phases of the relationship focused heavily on deciding what to do, and so cultivating collaboration early may be critical. Conversely, Goodman found that intimacy took six or more weeks to establish; before that time, open conversations by youth about their personal lives were minimal. For these mostly fatherless boys, conversation about their feelings about their father was less helpful than conversations about their current interactions with their mothers. Regardless, key to the youth’s openness was the counselors’ self-disclosure. Goodman learned “disclosure begets disclosure” (personal communication). He also found mentors were more open to and honest in their self-disclosure when they had received training on how to perform this mentoring talk tool. Additional Mentoring Strategies: Advocacy, Empowerment, Mutuality, and Reciprocity Certainly other interaction elements have emerged in the research literature that are not covered explicitly in the TEAM framework but that deserve attention. These elements deserve attention, given the critical roles researchers have found that they play in youth mentoring. One of these, advocacy, goes beyond the mentoring relationship and sometimes happens outside of the dyadic relationship. As such, one might not consider it to be a mentoring interaction, but rather a mentor’s intervention beyond the context of the immediate relationship. Another function, mutuality or reciprocity, reflects variations on the purpose and authorship dimensions of the TEAM framework and deserves some attention, even though to date little is known about the role of these constructs in youth mentoring outcomes Advocacy (and Empowerment). DuBois, Portillo, Rhodes, Silverthorn, and Valentine’s (2011) second and most recent meta-analysis reported fewer

Mentoring Activities and Interactions   77 statistically significant program practices than were reported in the first meta-analysis. This change may be a function of the increasing use of best practices by most programs. One practice not examined in DuBois, Holloway et al.’s (2002) meta-analysis is advocacy, one definition of which is the mentor’s efforts to help the youth bridge important worlds (e.g., school to work, family to school, friends to work; see also Cooper, 2011, for examples). In the 2011 meta-analysis, DuBois et al. found that programs that included an advocacy/teaching role had larger effects. This finding is worthy of further study: If mentors who systematically try to intervene and assist youth outside the matches have better impacts, then mentors should be trained in how to do this. A similar finding emerged in the qualitative research conducted by Ahrens et al. (2011) with 23 adults who were formerly children in foster care. One of the facilitators of an initial connection was the mentor’s willingness to go beyond the prescribed relationship, such as by helping the youth to find a job or a place to stay or by assisting through making calls to people outside the relationship. Several programs, like Blue Ribbon Advocates and Youth Advocate Program, explicitly engage in advocacy. In other programs it sometimes happens naturally. As one example, a recent article described the efforts of a Big Brothers Big Sisters mentor to help raise funds for her Little’s medical and educational needs (Davis, 2012). Yet the research literature has only recently identified the importance of advocacy in mentoring. This is probably because advocacy takes a more ecological perspective that goes beyond the dyadic or relational one usually associated with mentoring. Yet it should not be a new contribution to the literature. Decades-old programs for youth involved in the criminal justice system have long understood the importance of mentors providing strategic interventions to help youth of greatest need, such as the Youth Advocate Programs, Inc. It is worth noting, however, that in the DuBois et al. (2011) meta-analysis, the functions of teaching and advocacy were confounded (they were assigned to the same variable). As a result, the effects may really be a function of teaching rather than advocacy, or it may be that when programs teach certain skills, they use outcomes of those skills as measures of program impact (which can increase the size of these specific program effects). Had such programs been evaluated using a broader range of outcomes, as are typically used in evaluations, the overall effect size could have been considerably smaller.

It may be that research that has looked at the strategy of empowerment reflects a form of advocacy in the shape of direction and encouragement provided by mentor to youth to move beyond a comfort zone and into new relationships, work or educational contexts, or activities that could help a child achieve his or her goals. Some would argue that empowerment is exactly what good teaching is. More work should be done on this strategy, but we may use as a starting point Liang, Spencer, Brogan, and Corral’s (2008) finding that role modeling was a critical element of effective natural mentoring relationships. This may serve as one mode of empowerment. In their qualitative study of mentoring relationships among youth from middle school to early college age, they observed, however, that the forms of empowerment that youth felt were most important differed between childhood and late adolescence. Older youth were more open to receiving direction from the mentor than were younger youth, who appreciated general support more. Mutuality (Reciprocity). Liang et al. (2008) also found that the kinds of reciprocity and mutuality youth found most helpful differed developmentally. Older youth were more concerned about establishing reciprocity in the relationship than were children or early adolescents, perhaps because the older adolescents were more interested in looking like an adult to the mentor than in looking like a child. This may help explain differences in the value of role modeling by mentors for mentees of different ages. From Goodman’s (1972) early work identifying the importance of self-disclosure in helping relationships to more recent studies, we also find that reciprocity—the perception that both mentor and youth benefit and get their needs met through the mentoring relationship—is important. In Ahrens et al.’s (2011) qualitative study of graduates of the foster care system, mentors who sincerely disclosed their own experience and struggles with their mentees were better able to establish this initial connection.

Practice In this section, we draw attention to programs that employ practices in the areas of focus, purpose, and authorship that seem noteworthy, and we make suggestions for practitioners based on the theory and research described above, which we summarize in Table 5.3.

78  MENTORING RELATIONSHIPS Putting the TEAM Framework Into Action: Integrating Focus, Purpose, and Authorship Focus: Relational, Goal Directed, or Both. The decision about what a program should focus on— relational or goal-directed interactions—is not an either-or proposition. We should not ask how much, but when? Most effective programs seem to balance both types of interactions, which reflects the balance hypothesis of the TEAM framework. When each type is emphasized may be the first critical question to ask. When a program gives initial priority to goal-directed interactions, out of which relationships are cultivated, the program is more instrumental in nature. When the program is based primarily on relational interactions, but joint goaldirected activities are encouraged once the relationship has been established, then the program is more developmental. Being goal directed or relational is a strategy, but programs want to have an approach and mentors want to have a style that brings in multiple strategies to create their brand. Some programs are heavily relational, and this appears to be their strength. Big Brothers Big Sisters is heavily relational, particularly when the mentoring is in community settings, but as Morrow and Styles (1995) showed us, BBBS couples this strategy with a youth-focused authorship strategy and an allowance for problems to become the focus of the match after rapport is established. Cavell’s Lunch Buddy program (Cavell & Henrie, 2010), in which college mentors with little training meet twice weekly during lunch with elementary youth labeled by their school as aggressive, is heavily relational. The college mentors are just there to be a friend, and the mentor for the youth changes each semester. Although the program is relationally focused, what mentors discuss during lunch may take on a variety of focus, purpose, and authorship strategies. One of the program’s additional strategies is for the mentor to engage in conversations in the present with the youth and his or her peers with the purpose of strengthening the youth’s peer relationships and status. Conversely, Drexler, Borrmann, and MullerKohlenberg (2012) describe a program, Balu und Du (Baloo and You), which has a strong goaldirected focus. Based on the characters from Disney’s Jungle Book movie, this highly structured program trains and supports its mentors so that they can be more like Baloo (the friendly bear) than Bagheera (the prescriptive panther). Doing so requires a careful balance between fun and safety, relational and goal-focused interactions that are all the while collaborative (Karcher, 2009).

Cool Girls, Inc. (Kuperminc et al., 2011) is an after-school program that works to develop a positive self-concept, an academic orientation, a future orientation, and healthy behaviors in girls. The program is goal directed, using a specific curriculum focused on these areas. In addition, after 1 year of program participation, the girls are eligible to get a mentor (Cool Sister)—giving the program a more relational focus following the period of goaldirected activities, making this an excellent example of the instrumental approach. The after-school program was found to be effective in increasing scholastic competence, hope for the future, and physical activity. Girls with mentors had additional outcomes in the areas of social acceptance and body image and were more than four times more likely to report they would avoid drug use in the future. Purpose: The Future and the Fun. We have tried to illustrate that considerable empirical evidence that mentoring relationships, like all relationships, thrive when fun takes place but also when a serious purpose connects the relationship in the present with the youth’s success in the future. Balancing fun and future is a challenge, but several programs have tried to do this using a structured approach that integrates both systematically. GirlPOWER! (DuBois et al., 2008, Pryce, Silverthorn, Sanchez, & DuBois, 2010) is a program that builds on the foundation of the BBBS program (e.g., infrastructure) but uses a curriculum that balances goal-directed and relational activities. The GirlPOWER! program provides 3-hour monthly workshops for girls focused on five different topics: pride, opportunities (for learning), women-in-themaking, energy and effort, and relationships. The authors found some evidence to suggest the program could be beneficial for the girls beyond the typical BBBS program, but they also found they needed to integrate relational activities because mentees felt less satisfied with the program when it felt too prescriptive. The Take Stock in Children (n.d.) school-based mentoring program in Florida created a toolkit for mentoring in high schools. Their High School Mentor Toolkit provides rich sections with tips, ideas, and activities for supporting academic success through mentoring, fostering college readiness and career exploration, as well as conducting self-assessments. These activities would not work in elementary or middle school, but they are critically important topics for teens and in this way provide a way to address the potential problem of an overly relationally focused approach to mentoring in high school

Mentoring Activities and Interactions   79 that can be come too laissez faire and counterproductive (see Karcher, 2008). Of particular interest vis-à-vis the TEAM framework is not only that this toolkit provides guidance on how to be effective in using goal-directed activities, but early in the second section, the toolkit covers the importance of serving as an advocate for the mentee. In this way, this toolkit provides a needed and unique tool for mentors. Authorship: Collaboration. When matches meet, they have to decide what they will do. Someone initiates the conversation, but at that point, the course of the conversation often gets shifted according to differences in beliefs about how adults and children should relate (based on cultural, gender, or developmental differences), different understandings of the meaning and purpose of mentoring, and different overall interests and comfort levels in engaging in specific activities or given activities in a specific context (e.g., it might be okay from the youth’s perspective to get help on homework, but not in front of his or her peers at school). Who is writing “their story” is a question we have regularly suggested program staff ask themselves, their mentors, and their mentees. That seems to make sense to them. They know that if the story of the match (as told by mentor, youth, or parent) seems to be all about the mentor’s needs, goals, values, beliefs, or interests—such that any other child could easily be inserted into the story—then the mentee will soon become disconnected in the relationship. Because of the need for validation and the importance of reciprocity, the same happens to mentors if the story is all about the mentee or the program staff (e.g., the program’s curriculum).

What happens in the match should be a unique combination of each party’s needs, goals, values, beliefs, or interests. When that happens, it is collaboration, and it is good. Friends for Youth’s One + 1: Friendship Journal (Kremer, 2005) provides a unique way for mentors and mentees to document their relationship. This may be particularly useful for mentors of older mentees. It provides useful ideas about activities and discussions that matches can engage in across the first 12 months of their match. But most important, it provides regular prompts for the match to discuss and write about their evolving friendship. It is a graphically rich and visually appealing spiral-bound journal that may be especially appealing to teens. It provides an attractive keepsake journal that helps the mentor and mentee truly co-author their story. The Cross-Age Mentoring Program (CAMP; Karcher, 2008) goes so far as to structure its peer mentoring interactions so that in every meeting the match works their way to some collaborative activity. Regardless of the content of the curriculum activity being used, the day ends with an integration of the mentor’s and mentee’s uniqueness in a way that results in something distinctively theirs, if only it is “their time” playing something they both enjoy doing together. But daily and quarterly reflections on the relationship, how they feel about the match, and what they have done that each enjoys or dislikes, also helps CAMP to foster collaboration. Finally, one curriculum activity, lasting 4–5 weeks, uses Selman and Schultz’s (1990) developmental model and negotiation examples to teach the process of collaboration and perspective taking as a third way to foster mutual decision making in the match.

Table 5.3   Checklist for Practitioners Topic

Recommendations

Focus

Determine what the primary program focus strategy is, and whether this matches with the developmental needs of the children served. Identify specific (observable) techniques that exemplify that strategy, and coach mentors in the use of these techniques and specific talking tools (e.g., questioning, advising, silence) that match this focus.

Purpose

Determine where fun takes place in the program and how it is cultivated, and make sure this information is conveyed to mentors in training. Help mentors understand the ways in which serious, future-focused, skills-oriented activities are related to the youth’s life and how to sell their value to the youth before using them. (Continued)

80  MENTORING RELATIONSHIPS Table 5.3  (Continued) Topic

Recommendations

Authorship

Help mentors understand the nature and value of collaboration. Help program staff understand the importance of mentors and mentees coauthoring their relationship. Help program staff identify signs that coauthorship or collaboration is not happening in a match and how to cultivate it.

Hierarchy of interventions

Train mentors in the value of specific techniques and strategies that make up different mentoring approaches, and consider mentors’ skills, mentees’ needs, as well as the program’s espoused approaches to provide tailored training. Orient mentees to the approach(es) encouraged by the program so that they know what to expect and are more in tune with what their mentors expect to do. Train mentors in micro-skills or talking tools and show mentors how these different types of speech acts relate to program techniques and approaches.

Conclusion In this chapter we focused, not on the micro-skills or “talk tools” employed by mentors in moment-tomoment interactions, but on the macro-interventions such as whether matches played games or were engaged in more serious activities, whether they worked on goals or built a relationship, whether they were active and collaborative, or directive and sedentary. There is benefit in a micro-level of analysis, to be sure, but virtually no research addresses this issue, so it could not be a focus of this chapter. We implore researchers and practitioners to try to build in the lessons Goodman offers us about the importance of mentor’s micro-skills, too. Our goal was to present a framework of strategies that can be used to understand different mentoring approaches that programs take (e.g., Recreational Mentoring and Lunch Buddy) and different mentoring styles (e.g., developmental or sage style). With a framework for breaking down mentoring approaches and styles, program staff can more effectively design programs and training materials. By identifying the techniques used for each strategy that reflects a program’s chosen approach, program staff can determine better what interventions are essential, unique, common, or proscribed. These efforts should help to clarify mentors’ job descriptions and should help program staff to know what skills their mentors need to possess or acquire to be effective in their program. Ultimately, this should help mentors have greater confidence in what they are doing, boosting their allegiance to the program’s approach and adherence to correct strategies and techniques. It should also

provide a more consistent experience for the mentee, whose better understanding of mentoring, resulting from the mentor’s consistency and confidence, could help the youth to forge a stronger working alliance and cultivate his or her hope that this intervention and relationship is worthwhile. All of these are likely the mentoring common factors that are responsible for impactful matches.

References Ahrens, K. R., DuBois, D. L., Garrison, M., Renee, S. C., Richardson, L. P., & Lozano, P. (2011). Qualitative exploration of relationships with important nonparental adults in the lives of youth in foster care. Children and Youth Services Review 33, 1012–1023. Cavell, T. A., & Henrie, J. L. (2010). Deconstructing serendipity: Focus, purpose, and authorship in Lunch Buddy Mentoring. New Directions for Youth Development, 2010 126, 107–121. Cooper, C. R. (2011). Bridging multiple worlds: Cultures, identities, and pathways to college. New York: Oxford University Press. Coulter, E. K. (1913). The children in the shadow. New York: McBride, Nast, & Company. Davis, V. T. (2012, June 5). Big Sister comes to aid of injured little brother. San Antonio Express News. Retrieved from www.mysanantonio.com/news/local _news/article/Big-Sister-comes-to-aid-of-injuredlittle-brother-3608964.php Drexler, S., Borrmann, B., & Muller-Kohlenberg, H. (2012). Learning life skills strengthening basic competencies and health-related quality of life of socially disadvantaged elementary school children through the mentoring program “Balu und Du” (“Baloo and you”). Journal of Public Health, 20(2), 141–149.

Mentoring Activities and Interactions   81 DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30(2), 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 11, 59–91. DuBois, D. L., Silverthorn, N., Pryce, J., Reeves, E., Sanchez, B., Silva, A., . . .  Takehara, J. (2008). Mentorship: The GirlPOWER! program. In C. LeCroy, Craig Winston, & J. E. Mann (Eds.), Handbook of prevention and intervention programs for adolescent girls (pp. 326–336). Hoboken, NJ: Wiley. Fo, W. S., & O’Donnell, C. R. (1974). The Buddy System: Relationship and contingency conditions in a community intervention program for youth with nonprofessionals as behavior change agents. Journal of Consulting and Clinical Psychology, 42(2), 163–169. Frank, J. D., & Frank, J. B. (1991). Persuasion and healing. Baltimore, MD: Johns Hopkins Press. Goodman, G. (1972). Companionship therapy: Studies in structured intimacy. San Francisco: Jossey-Bass. Goodman, G., & Dooley, D. (1976). A framework for helpintended communication. Psychotherapy: Theory, Research & Practice, 13(2), 106–117. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work? An impact study of the Big Brothers Big Sisters program. Evaluation Review, 22(3), 403–426. Hamilton, M. A., & Hamilton, S. F. (2005). Work and service-learning. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 348–363). Thousand Oaks, CA: Sage. Hamilton, S. F., & Hamilton, M. A. (1992). Mentoring programs: Promise and paradox. Phi Delta Kappan (73), 546–550. Hansen, K. (2005). School-based mentoring study phase 1 report 2003–2004 school year: Effective practices. Philadelphia: Big Brothers Big Sisters of America. Hansen, K., & Corlett, J. (2007). School-based mentoring match activities and relationship quality. Philadelphia: Big Brothers Big Sisters of America. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/ Private Ventures. Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychological development: A longitudinal study of youth. New York: Academic Press. Karcher, M. J. (2004). The motivations of Hispanic mentors and the activities they use in their mentoring of Hispanic youth. Latino Research Initiative, University of Nebraska-Lincoln. Karcher, M. J. (2007). The importance of match activities on mentoring relationships. National teleconference conducted by the Mentoring Resource Center. Retrieved from http://www.michaelkarcher.com/ProfessorKarcher/ Media.html Karcher, M. J. (2008). The Cross-Age Mentoring Program (CAMP): A developmental intervention for ­promoting

students’ connectedness across grade levels. Professional School Counseling, 12, 137–143. Karcher, M. J. (June 17, 2009). Neither Baloo nor Bagheera alone could get Mowgli safely to the man village: Helping mentors decide what to do with their mentees. Presented at the 4th National SchoolBased Mentoring Conference, Kansas City, Missouri. Karcher, M. J., & Herrera, C. (2012). “Someone’s gotta be crazy about the kid”: Athena’s prescription for school-based mentoring. Unpublished manuscript. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. New Directions for Youth Development, 2010 126, 51–69. Karcher, M. J., Kuperminc, G. P., Portwood, S. G., Sipe, C. L., & Taylor, A. S. (2006). Mentoring programs: A framework to inform program development, research, and evaluation. Journal of Community Psychology, 34(6), 709–725. Karcher, M. J., & Nakkula, M. J. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. New Directions for Youth Development, 2010 126, 13–32. Keller, T. E. (2005). A systemic model of the youth mentoring intervention. Journal of Primary Prevention, 26(2), 169–188. Keller, T. E., & Pryce, J. M. (2012). Different roles and different results: How activity orientations correspond to relationship quality and student outcomes in school-based mentoring. Journal of Primary Prevention, 33(1), 47–64. Kremer, S. E. (2005). One + 1: Friendship journal. Redwood City, CA: Friends for Youth. Kremer, S.E. (2007). Mentoring Journal. Redwood City, CA: Friends for Youth, Inc. Kuperminc, G., Thomason, J., DiMeo, M., & BroomfieldMassey, K. (2011). Cool Girls, Inc.: Promoting the positive development of urban preadolescent and early adolescent girls. Journal of Primary Prevention, 32, 171–183. Langhout, R. D., Rhodes, J. E., & Osborne, L. N. (2004). An exploratory study of youth mentoring in an urban context: Adolescents’ perceptions of relationship styles. Journal of Youth and Adolescence, 33(4), 293–306. Liang, B., Spencer, R., Brogan, D., & Corral, M. (2008). Mentoring relationships from early adolescence through emerging adulthood: A qualitative analysis. Journal of Vocational Behavior, 72, 168–182. Morrow, K. V., & Styles, M. B. (1995). Building relationships with youth in program settings: A study of Big Brothers/Big Sisters. Philadelphia: Public/Private Ventures. Nakkula, M. J., & Harris, J. T. (2010). Beyond the dichotomy of work and fun: Measuring the thorough interrelatedness of structure and quality in youth mentoring relationships. New Directions for Youth Development, 2010 126, 71–87. O’Donnell, C. R., & Tharp, R. G. (2012). Integrating cultural community psychology: Activity settings and the shared meanings of intersubjectivity. American Journal of Community Psychology, 49, 22–30.

82  MENTORING RELATIONSHIPS O’Donnell, C. R., Tharp, R. G., & Wilson, K. (1993). Activity settings as the unit of analysis: A theoretical basis for community intervention and development. American Journal of Community Psychology, 21, 501–520. O’Donnell, C. R., & Williams, I. (2013). The Buddy System: A 35 year follow-up on criminal offenses. Clinical Psychological Science, 1, 54-66. Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30(4), 367–388. Pedersen, P. J., Woolum, S., Gagne, B., & Coleman, M. (2009). Beyond the norm: Extraordinary relationships in youth mentoring. Children and Youth Services Review, 31, 1307–1313. Pryce, J. M., Silverthorn, N., Sanchez, B., & DuBois, D. L. (2010). GirlPOWER! Strengthening mentoring relationships through a structured, gender-specific program. New Directions for Youth Development, 2010 126, 89–105. Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality change. Journal of Consulting Psychology, 21, 95–103.

Selman, R. L., & Schultz, L. H. (1990). Making a friend in youth: Developmental theory and pair therapy. Chicago: University of Chicago Press. Spencer, R. (2006). Understanding the mentoring process between adolescents and adults. Youth & Society, 37, 287–315. Spencer, R., & Rhodes, J. E. (2005). A counseling and psychotherapy perspective on mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. (pp. 118–132). Thousand Oaks, CA: Sage. Take Stock in Children. (n.d.). The high school mentor toolkit. Tampa, FL: Helios Education Foundation. Wampold, B. E. (2001). The great psychotherapy debate: Models, methods, and findings. Mahwah, NJ: Erlbaum. Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials of schoolbased mentoring: Making sense of mixed findings. Social Policy Report, 24(3), 1–21. Wyman, P. A., Cross, W., Hendricks Brown, C., Yu, Q., Tu, X., & Eberly, S. (2010). Intervention to strengthen emotional self-regulation in children with emerging mental health problems: Proximal impact on school behavior. Journal of Abnormal Child Psychology, 38, 707–720.

6  GOALS IN MENTORING RELATIONSHIPS Fabricio E. Balcazar and Christopher B. Keys

Introduction This chapter examines the conceptual and practical implications of goal setting and goal pursuit in the context of mentoring relationships for youth. The role of goals has been an area of both interest and controversy in the field, given the complex nature of mentoring relationships for young people, the differing stages of development of the relationship over time, and the varying characteristics of the mentors and mentees engaged in those relationships (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). More generally, too, as Rhodes (2005) pointed out, the fundamental question of how mentoring works has been often overshadowed by programmatic concerns about implementation. Thus, the role of goals in mentoring has at times been overlooked. The conceptual framework that Rhodes (2005) proposed assumes as a starting point a strong interpersonal connection characterized by mutuality, trust, and empathy. This interpersonal connection has the potential through various relational processes to influence the social-emotional, cognitive, and identity development of the youth being mentored. In this chapter, we argue that, with some caveats and constraints, goal setting and goal pursuit as part of the context of the mentoring relationship can have a positive mediating impact on these types of processes and their outcomes that are associated with the relationship. Keller (2005) noted, “the defining feature of youth mentoring is the personal relationship established between a young person and a caring, competent individual who offers companionship, support, and guidance” (p. 82). From this perspective, rather than simply whether or not goals

should be a part of mentoring relationships for youth, key issues become more nuanced concerns of how and when goal setting and support for goal attainment should occur—for example, degree of specificity, timing, and balance between encouragement for goal setting and pursuit and other relational considerations and objectives. Not all mentoring relationships look the same, follow the same pathway, or attain the same level of strength (Sipe, 2005). In the context of mentoring programs, goal setting and goal pursuit may be considered part of the infrastructure of some mentoring programs, with different programs having high, medium, or low levels of infrastructure (Sipe, 2005). More generally, too, as we discuss, understanding how goal setting and goal pursuit can impact the mentoring relationship requires a closer examination of how mentoring relationships evolve. Developing such an understanding calls for a review of the theories and research literature regarding goal setting and goal pursuit support in mentoring relationships with youth as well as a broader consideration of approaches to goal orientation and attainment and their potential relevance to youth mentoring. After addressing these topics, we provide an overview of practical implications for incorporation of goal setting and goal pursuit into mentoring program development and implementation.

Theory In this section, we begin by considering models and research pertaining to stages of development and relationship quality in mentoring for youth

83

84  MENTORING RELATIONSHIPS with an emphasis on their implications for setting and pursuing goals in relationships. Following a brief integrative summary of these implications, we provide an overview of an approach that seems especially promising for integration of goal-related activities into youth mentoring: goal attainment scaling. The Stages of Development of Mentoring Relationships for Youth Keller (2005) proposed a model for the stages of development of youth mentoring relationships that includes five potential periods: contemplation (which refers to the initial stage of anticipating and preparing for the relationship), initiation (beginning the relationship and becoming acquainted), growth and maintenance (meeting regularly and establishing patterns of interaction), decline and dissolution (addressing challenges to the relationship or ending it), and redefinition (negotiating terms of future contact or restarting the relationship). These stages are relevant to identifying the appropriate time to set goals and provide support for goal pursuit. In cases in which the goals can help frame the interaction and give it direction, the mentor may need to learn how to prepare the youth to set goals during the initial contemplation phase. The actual process of setting goals may take place during the initiation phase—except in cases in which building trust requires more time and effort. Support for goal pursuit can continue during the maintenance phase, with the potential introduction of new goals as needed when circumstances change or when a goal is attained and the youth decides to pursue another goal. Finally, in some cases, the redefinition phase may lead to more goal setting and the potential reinitiation of support for goal pursuit. Mentoring Relationship Quality Nakkula and Harris (2005 and this volume, Chapter 4) offered a useful framework for understanding the factors that can influence the quality of youth mentoring relationships. Their framework distinguishes between internal and external indicators of relationship quality. Internal factors can be relational, which represent characteristics of the relationship itself, including both subjective (e.g., compatibility, mentor’s approach, youth engagement, closeness) and objective (e.g., frequency of meetings, intensity, duration, resiliency of the relationship) factors. Also important, the internal portion of Nakkula and

­arris’s framework also includes goal-oriented H instrumental factors. These include both subjective (e.g., nature of the goals and degree of emphasis on pursuing goals, mentor and mentee’s instrumental orientation, perceived support for goal pursuit) and objective (e.g., degree of support received for goal attainment, including goal-directed activities and advocacy) components. External factors in their framework include things like programmatic support (e.g., supervision, training, structured activities), parent engagement, and support networks. Nakkula and Harris (2005) pointed out that “from an instrumental perspective, a high-quality mentoring relationship is marked by the mentee’s perception that he or she is being supported in accomplishing particular goals and by the mentor’s perception that he or she is being helpful in this effort” (p. 104). Deutsch and Spencer (2009) identified several factors to consider when assessing the quality of mentoring relationships. These include the duration of the relationship, the frequency and consistency of the contact between the mentor and the youth, the quality of the connection that forms, and the mentor’s approach to the relationship (i.e., how the mentor thinks about, engages with, and responds to the youth). They noted that some mentors take a youth-driven (also called developmental) approach, whereas others are more prescriptive with their mentees. Prescriptive mentors have been criticized because they often emphasize their own goals for the youth, which may be unrealistic or not developmentally appropriate. Such findings underscore that the process of setting goals in mentoring relationships with youth needs to be carried out carefully and with attention to the youth’s desires, abilities, and interests and with recognition of his or her context. Role of Goal Setting and Goal Pursuit in Mentoring Relationships for Youth To summarize, goal pursuit in theory “infuses the course of development with a sense of future orientation and self-direction” (Brunstein, Schultheiss, & Maier, 1999, p. 170). In accordance with this perspective, self-determined goal pursuit in favorable environmental circumstances (i.e., those in which goal-related training and support are available—as in the case of many youth mentoring programs) is related to successful goal attainment and positive lifespan adaptations (Brunstein et al., 1999). With this understanding, discussions of goal setting and goal pursuit by mentoring theorists highlight their potential to constructively focus

Goals in Mentoring Relationships   85 participants’ energies in a positive mentoring relationship. However, cautionary concerns have been emphasized equally, in particular because of the risk of mentors becoming overly directive or task focused in their interactions with youth in the context of working on goals. The extent to which goal setting is collaborative and, more specifically, who makes the decision about the goals to pursue, may be especially important (Karcher, Herrera, & Hansen, 2010). When the youth decides or structures the activities or direction of the match, the mentor may become dissatisfied. When the mentor makes the decision, the youth can become dissatisfied or not be invested. When program staff make the decision, both the mentor and the youth can become dissatisfied. But when youth and mentor make the decision together, they may experience a higher quality of relationship and greater sense of accomplishment (Karcher et al., 2010). In related research, Langhout, Rhodes, and Osborne (2004) found that outcomes for youth in the Big Brothers Big Sisters community-based mentoring program were most favorable when youth reported experiencing both structure and support from their mentors. We would thus hypothesize that, theoretically, helping youth to set and work toward goals that are important to their development can be of significant benefit, especially if the goals are agreed upon by mentor and youth in accordance with a youthcentered approach (Keller, 2005). As mentioned earlier, goal setting can be introduced in mentoring relationships as a potentially important contributor to the structure of the relationship under appropriate circumstances. Garate-Serafini, Balcazar, Keys, and Weitlauf (2001) proposed a framework for assessing the effectiveness of mentoring programs for marginalized youth that included four components: (a) goal attainment, (b) specific competency skills, (c) changes in the composition of the youth’s social support network, and (d) motivation and self-efficacy. These domains provide a framework for assessing youth’s abilities to formulate and pursue personal goals, learn skills related to specific needs, increase their social support networks and increase their beliefs in their power to affect change in their lives. This framework was used to introduce mentoring to lowincome African American and Latino/a youth with learning disabilities as a means of achieving personal, educational, and/or vocational goals (TaylorRitzler, Balcazar, Keys, Hayes, & Garate-Serifini, 2001). In addition, participating youth received training designed to help them realize how the

mentoring relationship can be a means for pursuing and attaining relevant life goals. On the other hand, we acknowledge that in some cases goals may make the mentoring matches seem artificial and prescriptive. Such relationships may appear to be directive like coaching and tutoring rather than mentoring. So as noted before, the way the goal-setting and goalpursuit process is introduced in the mentoring relationship, the timing of the introduction, and the skillfulness of the mentor in allowing the youth to take the lead in the goal-setting and goalpursuit process are theoretically critical for obtaining desired results. Goal Attainment Scaling Kiresuk and Sherman (1968) first introduced goal attainment scaling (GAS) to evaluate mental health services for clients of community mental health centers. Goal attainment scaling is a structured approach to recording goal achievement (Bovend’Eerdt, Botell, & Wade, 2008). Rather than simply stating that a goal has or has not been achieved, attainment scaling recognizes that sometimes achievement exceeds expectations, sometimes achievement is less than expected but nonetheless progress has been made toward the goal, and sometimes there may be no progress toward a goal or gains made toward the goal may even be lost (known as “deterioration”). Essentially, GAS involves four steps (see Marson, Wei, & Wasserman, 2009) that can be considered in relation to youth mentoring. First, in conjunction with the youth, the mentor would identify critical issues that reflect the youth’s goals. There should be flexibility in identifying goals and expectations (i.e., allowing the youth to make changes over the course of the relationship as circumstances may change). In the second step, based on the mentor’s impression, each goal would be given a weight— the more difficult the goal, the greater the weight. In the third step, the youth and the mentor would come to consensus on an expected outcome (the Kiresuk & Sherman [1968] original scale for possible outcomes had five levels: much less than expected outcome [–2], less than expected outcome [–1], expected outcome [0], more than expected outcome [+1], and much more than expected outcome [+2]). Then the mentor and the mentee would operationalize progress toward the outcomes for each goal (e.g., specifying what constitutes attaining less than desired success for each goal). In the fourth and final step, after a period of time spent

86  MENTORING RELATIONSHIPS supporting the youth in working toward a goal, the mentor would create a composite score for each goal (see original formula in footnote)1. This assessment, which could be repeated on multiple occasions to track progress toward a goal over time, could be carried out in collaboration with the youth and/or with facilitation from program staff as appropriate to circumstances. Goal attainment scaling has been used widely with respectable internal consistency coefficients, good concurrent validity, and strong interrater reliability (Emmerson & Neely, 1988; Garwick, 1974; Kiresuk, Smith, & Cardillo, 1994; Marson et al., 2009; Tennant, 2007). Turner-Stokes (2009) argued that GAS offers a number of potential advantages: First, there is emerging evidence that goals are more likely to be achieved if participants are involved in setting them (Bovend’Eerdt et al., 2008; Malec, 1999). Second, there is growing evidence that GAS is a good measure of outcome, being at least as sensitive to change (and probably more so) when compared with standard measures (McDougall & Wright, 2009; Steenbeek, Gorter, Ketelaar, Galama, & Lindeman, 2011; Turner-Stokes & Williams, 2010). The literature on GAS encompasses a range of approaches to scaling and measuring the achievement of goals, ranging from simple recording of goals achieved, partially achieved, or not achieved to rating scales of up to seven points (TurnerStokes, 2009). In the next section we review the history of GAS and consider its potential benefits in mentoring relationships for youth.

funded by the National Institute of Mental Health and proposed to determine the best way to assign patients to treatments and therapists and, at the same time, to test GAS for its feasibility. Kiresuk pointed out that, at the time, the burden of proof of mental health treatment effectiveness fell entirely on the recipient of the treatment rather than the provider. It was the patient who was deemed to be “no longer able to profit from therapy,” whereas the individual efficacy of therapists remained often unchallenged. These circumstances served as the impetus for developing an evaluation procedure that could allow for meaningful comparison of the different therapeutic outcomes of individual therapists. Most important, it became apparent to the development team that the criteria for program success had to be turned over to the participants. However, the process was not easy and the conflict with the entrenched power of physicians and hospital administrators was intense. Kiresuk also acknowledged that “some of the early features of GAS, that were to become matters of dispute, originated in an effort to anticipate every conceivable objection that had been encountered in reviewing and pretesting existing measures” (p. 147). In parallel, the dissemination activities were disorganized at best, as Kiresuk admitted, and perhaps in some cases the process was used without being fully tested and developed. This understandable yet less than optimal approach produced conflicting results in several applications and confusion about what GAS is and what its strengths and weaknesses are.

Historical Background of GAS. Kiresuk (1994) provided a detailed overview of the history of GAS. In brief, the approach originated as a reaction to the oppressive conditions experienced by institutionalized mental health patients in the 1950s and 1960s. Moreover, the dominance of psychiatry in the diagnosis and treatment of mental health illnesses, the emergence of community-based treatments and intervention programs, and the need to assess the relative effectiveness of various psychotherapeutic approaches that had long been in theoretical confrontation with each other, all contributed to the need for a new, more person-centered approach. The original study (Kiresuk & Sherman, 1968) was

Application of Goal Attainment Scaling to Youth Mentoring. Kiresuk and Choate (1994) pointed out that “a distinguishing feature of GAS is its adaptability to a wide-variety of settings and a corresponding diversity of methods and purposes of application” (p. 61). The authors provided multiple examples of GAS applications in the fields of education, medicine, vocational rehabilitation, mental health, social services, and counseling, including sample follow-up guides that were used to assess goal attainment from specific participants in the research studies (see also Hurn, Kneebone, & Cropley, 2006, for a systematic review of GAS in rehabilitation). Kiresuk and Choate acknowledged

1

The original formula for GAS developed by Kiresuk & Sherman (1968) was as follows:

GAS = 50 +

10 ∑ (Wi X i ) (.7 ∑ Wi 2 + .3 (∑ Wi ) 2 )

,

where Wi is the weight given to the goal and Xi is the numeric value of accomplishment.

Goals in Mentoring Relationships   87 that new and improved formats of GAS are needed to facilitate the process of goal setting. In particular, existing formats (referring to the scale and criteria for scoring) need to be adapted to meet the needs of multiple service delivery settings. Illustratively, Roach and Elliot (2005) argued that GAS provides an individualized, criterionreferenced approach to describing changes in the performance of students in school settings and can be very useful in documenting changes involving academic and social behaviors in these contexts. By using numerical ratings for each of the five descriptive levels of functioning (using the previously described Kiresuk scale), the rater can provide a daily or weekly report of student progress. These criterion-referenced reports can accompany direct indicators of progress (e.g., direct observations, curriculum-based measurement) and other indicators such as work samples, student self-reports, or reports from parents or other individuals. The preceding considerations suggest that GAS will prove most beneficial within youth mentoring relationships and programs when its format and procedures are adapted to the regularities, priorities, and norms of particular application contexts. For example, given the research-supported emphasis in many programs on having mentors maintain a collaborative stance in their work with youth (Keller, 2005), it may be especially important for youth mentees to have active input and participation at all phases of the process, including assessments of goal attainment. Likewise, given that mentors in programs most often are volunteers with little or no specialized prior training relevant to their roles, the extent to which GAS is well implemented and yields desired results may depend on the adequacy of the training and ongoing support that the program makes available. It should be noted, however, that the one-to-one format of mentoring relationships in most programs offers ready-made opportunities for the types of one-on-one assessment, guidance, and support that are likely to facilitate effective applications of GAS. This section has reviewed some of the conceptual issues associated with goal formulation and GAS. We now turn to an examination of research that bears on application of goal setting and goal pursuit in the context of youth mentoring relationships.

Research To facilitate our review of research on goal setting and goal pursuit, we conducted a literature review

of several databases (i.e., PsycINFO [137 records], PubMed [146 records], and the Social Sciences Citation Index [102 records]), using keywords like mentor, youth, goal setting, goal attainment, and GAS. We found a total of eight articles that focused on school-based mentoring programs, two articles that focused on mentoring in community-based programs, and one article that reported a recent meta-analysis of youth mentoring program evaluations in which findings relevant to goal-related activities in mentoring were reported. We also identified16 articles that described a variety of examples of GAS applications for behavior change with children and youth. This last group of research studies is not included in this review because the studies do not include mentoring. However, as a group, their findings point toward the benefits of GAS for behavioral improvement in working with children and adolescents (see, e.g., Steenbeek, Ketelaar, Galama, & Gorter, 2007; Vansteenkiste, Simons, Lens, Soenens, & Matos 2005). School-Based Mentoring Portwood, Ayers, Kinnison, Waris, and Wise (2005) conducted an evaluation of YouthFriends, a school-based mentoring program that used community volunteers who helped participants set goals in relevant areas. The study employed a quasiexperimental, pretest-posttest, comparison group design. Students (N = 170) across five school districts from low-income minority communities provided data on eight kinds of dependent variables: drug and alcohol use (from older youth only), drug and alcohol attitudes and beliefs, attitudes toward school, academic performance, school and community connectedness, attitudes toward self as well as goal-setting skills, attitudes toward adults, and attitudes toward the future. At posttest, there was a statistically significant difference favoring YouthFriends participants over the students in the comparison group on sense of school membership. In addition, for at-risk students (those who had low scores on the outcome measure at baseline), results indicated a statistically significant improvement unique to YouthFriends participants on school and community connectedness and goal setting. Analyses of students’ academic performance among those with low grades at baseline also indicated significantly greater gains for YouthFriends participants relative to those in the comparison group. On the other hand, the analyses revealed no significant differences between the two groups in regard to drug and alcohol use, attitudes toward drug use, commitment not to use drugs, or students’ school disciplinary records. The authors

88  MENTORING RELATIONSHIPS concluded that, consistent with earlier efforts to establish that mentoring has a positive impact on young people, this evaluation produced mixed results. A significant limitation of this evaluation was the researchers’ inability to document the extent of individual participants’ involvement with their mentor over the course of the evaluation year. Nevertheless, the increase in goal-setting behaviors among participant youth is noteworthy and suggests its potential as a teachable skill in the context of mentoring relationships. Although not examined by the researchers, it would have been informative to investigate gains in goal-setting behaviors as a mediator of improvements in other areas (e.g., academic performance). A similar study conducted by LoSciuto, Rajala, Townsend, and Taylor (1996) evaluated the effectiveness of Across Ages, an intergenerational mentoring program of drug prevention for high-risk middle school students (6th graders). The project involved older adults (aged 55 years or more) as mentors, who met for an average of 4 hours per week with the youth. Mentors were supported in helping children to set personal goals and to develop awareness, self-confidence, and skills they could use to resist drugs and overcome life obstacles. Students also participated in community service activities that benefit frail elders and received a classroom-based, life skills curriculum. The results indicated favorable program effects on students’ knowledge, attitudes, and behaviors regarding substance abuse. Findings further suggested that the degree of involvement of the mentors was a critical moderating factor: Youth who had less engaged mentors demonstrating nonsignificant differences in outcomes relative to control group youth. These research findings indicate a potential for positive effects when goal setting is involved in mentoring, but due to the multifaceted nature of the mentoring and other program components, the contribution of activities in this area to outcomes cannot be determined. Using a quasi-experimental comparison group design, King, Vidourek, Davis, and McClellan (2002) found significant improvements in selfesteem, and positive connections to school, peers, and family, among at-risk 4th-grade students participating in the Healthy Kids Mentoring Program, compared to unmentored classmates. Students took part in relationship-building skills training, selfesteem enhancement, goal setting and pursuit, and academic assistance. The mentors were trained and encouraged to provide extra reading support to the participating youth. In addition, they also prompted the students to engage in positive interactions with their peers as part of their goal pursuit activities

(e.g., thanking one peer per day for their assistance or support in classroom activities). This strategy may have been a key factor contributing to the observed improvements in connections with peers. The authors concluded that having a project coordinator on site, utilizing community volunteers, getting parental and community support, keeping parents informed, promoting the skill components mentioned such as goal setting and pursuit, and providing mentors with ongoing training and support were critical for the successful implementation of the mentoring program. These research findings again implicate goal-related activities as potentially important in mentoring, but in combination with several other programmatic features. Similar results were reported by Garate-Serafini et al. (2001), who developed and implemented a goal-centered mentoring program. This school-based program was for low-income, minority youth with disabilities to help participants achieve personal goals and empowerment. The authors used a pre-post singlegroup design to assess effects. Staff noted the apparent importance of several goal-related procedures. These included providing training in setting relevant, challenging, achievable goals; recruiting mentors to help the youth achieve their goals; and having mentors and program staff support the youth in goal setting and attainment. Program staff also called attention to the need for goals to be consistent with the student’s values and culture (e.g., the goal of “independence” is often misunderstood in the Latino culture as pursuing separation from the family, which is not supported). Ultimately, 85% of participating youth were successful in achieving one or more academic, employment, and/or independent-living goals. Forneris, Danish, and Scott (2007) used a peer mentoring program called GOAL to teach 9th-grade minority students from a low-income community to set and pursue goals, solve problems, and seek social support to promote better adjustment to high school. Based on qualitative interviews with the participating students, the authors concluded that the adolescents who participated in the peer mentoring program appeared to have learned the steps for setting an achievable goal and to be better prepared to increase performance and reach their goals in any aspect of life. One of the objectives of the GOAL program was to teach adolescents a problem-solving strategy called STAR, which was used to learn how to make correct choices and reduce conflict that could interfere with reaching their goals (e.g., getting in trouble with the law). STAR stands for Stop and take a deep breath, Think of all the alternatives to the problem, Anticipate the consequences for each alternative, and Respond with the best choice. The “best” choice

Goals in Mentoring Relationships   89 is the one that gets the problem solver closest to his or her goal. The participants reported that the strategy was easy to learn and helpful in dealing with their day-to-day problems. Dennison (2000) reported on a peer mentoring/ tutoring program called the Big Buddies Project that used high school students to work with 3rd and 4th graders at risk of dropout. A single-group prepost design was used to evaluate outcomes for both the participating students and their high school mentors/tutors. Program staff worked closely with the peer mentors/tutors in a structured way during the entire year of engagement with their Little Buddies. The author argued that it is important that peer mentors/tutors be given enough training and supervision to effectively work with an identified at-risk student population. Although some improvements were apparent in several academic areas for the children who participated, the pre-post differences were not statistically significant. The author concluded that further research is needed to better understand if and how this program influences selfesteem, school attitudes, and school behaviors in children at risk for dropping out of school. In another peer mentoring program, in which students were instructed to set goals and ask mentors for help in pursuing their goals, Karcher, Nakkula, and Harris (2005) found evidence that the mentors’ perceptions of the quality of the relationship were a function of their mentees’ openness to seeking support from them and of the mentors’ initial feelings of self-efficacy. The authors concluded that programs should place less emphasis on preparing mentors to deal with their mentees’ risk status and more emphasis on promoting their sense of efficacy as mentors and role models and on helping mentees become more proactive in seeking support from their mentors to address their needs. This last strategy may be used to encourage mentees to seek active mentor support for their own goal setting and pursuit. Finally, Karcher et al. (2010) conducted two studies of school-based mentoring with a sample of 412 mentors participating in the Big Brothers Big Sisters School-Based Mentoring Impact Study. Their findings support the distinction between different styles of mentor interactions with youth— that is, relational (focusing conversations more on the development of a positive relationship between the mentor and mentee) versus instrumental (focusing conversations more on the goals and outcomes of the relationship). In particular, findings from the first study suggested that both instrumental and relational interactions made significant and unique contributions to relationship quality. An examination of the standardized factor loadings of the

observed variables on relationship quality revealed that both types of conversation topics were positively associated with relationship quality, but that the magnitude of the relational conversations pathway coefficient (.39) was more than three times as large as that of the instrumental (goal-directed) conversations coefficient (.11). However, both the content (indicators used) and the context (the absence of teenage mentees who may be most ready to benefit from instrumental interactions) may explain the relatively modest associations between goal-directed conversations and relationship quality. It is also possible that on other more task-related outcomes, such as behavior changes, grades, or attendance, goal-directed interactions may be better predictors (Hamilton & Hamilton, 2005). In summary, school-based mentoring programs have used goal setting with various degrees of success. One key consideration raised by the various research studies is the difficulty in ensuring the quality and consistency of the mentoring relationships in a given program (i.e., the right dose of good mentoring). Some mentors seem to be more dedicated and caring than others and more able to establish a positive relationship. When there is not sufficient training and supervision, the benefits of the relationship decrease. The consistency problem is important in the context of the present discussion because it is likely to have an impact on the effectiveness of goal setting and goal pursuit in mentoring relationships and programs. Evaluations of programs that have emphasized goal setting and goal pursuit in the relationships have pointed toward the benefits of goal setting for youth in the context of mentoring. However, because evaluations did not contrast the effects of mentoring youth with or without goal setting as a component of the program or examine changes in goal setting as a mediator of outcomes, available findings at present offer only indirect and inconclusive evidence of the value of incorporating a goal orientation into mentoring programs for youth. In addition, given the concerns about being overly directive in goal setting, it is noteworthy that findings from a couple of the studies suggest that youth mentees need to experience a sense of autonomy in the relationship. In the context of working on goals, this implies the importance of youth having some say in the process of setting the goals they will be pursuing and not simply accepting the ones preferred by the mentors or the program staff. Community-Based Mentoring Turning to community-based mentoring, Rhodes, Reddy, Roffman, and Grossman (2005)

90  MENTORING RELATIONSHIPS conducted an exploratory and confirmatory factor analysis of youth-reported mentoring relationship data from the national evaluation of the Big Brothers Big Sisters community-based mentoring program with a sample of 347 participating youth. Interestingly, they found that for one factor, perceptions of the mentor being helpful with solving problems loaded strongly (which in this program is likely to be related to goals set by and for the youth to address identified needs), whereas the remaining three measures had loadings on the same factor that were in a direction indicative of an absence of dissatisfaction and feelings of hurt or betrayal in the relationship. This aspect of their findings suggests the importance of the youth’s perceptions of the supporting role of the mentor with respect to goal pursuit over the duration of the relationship. This could also partly explain the apparent potential for a sense of disappointment when the mentor is perceived as not being supportive of the youth’s goal pursuit. Balcazar, Davies, Viggers, and Tranter (2006) reported on the pilot evaluation of a GAS approach used to assess the outcomes of five mentoring programs from a large city in the United Kingdom. The mentoring programs served youth offenders, youth with disabilities, at-risk college students, and underachieving youth. A total of 176 mentors were trained to use GAS to set short-term goals with their mentees and then support their mentees in goal pursuit. The authors used a pre-post design with multiple nonequivalent groups and assessed goal attainment after periods of 2 and 4 months. Results indicated significant and consistent increases in goal attainment among most of the participating youth compared with baseline assessments, except for youth offenders, for whom more modest changes were evident. Qualitative interviews with participating mentors suggested that GAS offered clear direction for mentors about how to focus their support and a useful framework for working with youth. However, some mentors complained about the need to receive additional training in using the goal attainment scale effectively. They also noted the difficulty of using GAS with youth who did not seem ready to set or pursue goals. This was especially the case for mentors working with youth with serious emotional problems and those working with youth offenders, many of whom were apparently not ready to open up to their mentors. These caveats are consistent with previous research findings (e.g., Blum & Jones, 1993; Larose, Tarabulsy, & Cyrenne, 2005; Rhodes et al., 2005) that underscore the importance of establishing trust and positive rapport between mentors and mentees in order to build effective relationships. In this study, the follow-up

evaluation was perhaps conducted prematurely. It may be that with youth who have more challenges, mentors require more time to build sufficient trust in order for goal-setting and goal-pursuit activities to be effective. In summary, although fewer research studies were identified in this category, community-based programs like Big Brothers Big Sisters are among the largest in the country. The research findings offer cautionary notes about the potential detrimental effect of forcing the youth to engage in goal setting and goal pursuit when they are not ready. On the other hand, the available findings point to goal setting as a strategy that can be useful for focusing the mentoring efforts and for assessing youth outcomes in mentoring relationships. It is also important to recognize that youth often perceive mentors as playing a critical role in supporting their goal pursuit, and they feel disappointed or abandoned when they perceive their mentors as not giving them enough support to pursue their goals. Meta-Analysis of Youth Mentoring Program Effectiveness In their recent meta-analysis of 73 independent evaluations of youth mentoring programs published between 1999 and 2010, DuBois et al. (2011) found that effect size (an estimate of benefits for youth stemming from program participation) did not change significantly as a function of whether or not programs were coded as having a goal-setting component. Furthermore, many of the articles reviewed lacked sufficient details to allow the raters to confidently identify the extent to which goal setting and pursuit was an integral part of the mentoring program (D. L. DuBois, personal communication, April 3, 2012). It is noteworthy, however, that program effects were stronger for programs that incorporated teaching/ guidance and advocacy roles for mentors, each of which is likely to be amenable to inclusion of goalsetting and goal-pursuit activities in their programs. Recommendations for Future Research The studies reviewed lead us to offer multiple recommendations for future research. First, they point to the need to attend to the characteristics of the mentoring relationship and ways to strengthen it to maximize the benefits and minimize potential harm to the mentees. Issues of selection, training, monitoring, and evaluation are likely to be central to a program’s ability to deliver a positive mentoring experience to participating youth. Future research focused on goal setting and goal pursuit in the

Goals in Mentoring Relationships   91 context of youth mentoring should attend carefully to these concerns and their implications for the success with which these activities can be integrated into programs and achieve desired effects. Second, further examination specifically of the interplay between relational and instrumental (goal-oriented) interactions as dimensions of mentoring relationship quality is necessary. The framework provided by Nakkula and Harris (2005 and this volume, Chapter 4) in which the significance of subjective, objective, and external factors for understanding and assessing both relational and instrumental facets of mentoring relationship quality would be useful for guiding the design of research along these lines. Third, also worthy of further research are both the quality of the match (compatibility) between the mentors and the youth and whether or not sharing common interests is more important in this regard than considerations such as ethnicity, race, or gender. Shared interests theoretically could be especially important in setting the stage for goal setting and pursuit in a relationship, further illustrating the potential fruitfulness of examining these activities in relation to other features of relationships and programs. Fourth, the issue of the appropriate dose of mentoring needs to be studied further. What minimal amount of engagement is typically necessary to build trust in the relationship? Because relationships are the heart of mentoring, successful mentoring, including goal-setting and goal-pursuit activities in particular, is not likely to be possible without strong relationships; therefore, mentors need to be trained in effective ways to develop rapport and communication with youth (Hayashi & O’Donnell, 2004; see Kupersmidt & Rhodes, this volume, Chapter 30). Similarly, mentees could also potentially benefit from training aimed at helping them clarify expectations about the relationship and at how to utilize mentors’ help in goal setting and goal pursuit. In addition, more research is needed to understand the timing of the goal-setting process in the relationship. In other words, how much mutual trust and understanding are necessary to effectively introduce goal setting into the mentoring relationship? Given the need for some youth to build trust with their mentors before they are able to engage in goal setting and goal pursuit, investigation of the effects of staggered introduction of the goalsetting process at various stages of relationship development with different groups of youth could help to clarify this issue. From a programmatic standpoint, future research is also needed to evaluate the challenges of goal-setting implementation. There is a need to identify the best strategies for teaching mentors and mentees to set high-quality goals, for assessing goal

progress, and for determining goal attainment. More research is necessary as well to determine when and how it is appropriate to involve other adults, such as teachers, siblings, or parents, in the process of setting goals and/or supporting goal pursuit. These individuals could provide valuable support for mentors and thereby increase the youth’s chances of success, but it is not clear at present how best to engage them or when. A final area in need of research is the cost of implementing goal-setting and goal-pursuit activities, such as GAS, in mentoring programs for youth. This type of analysis could provide schools and community organizations with valuable information about the feasibility of implementing goal-oriented mentoring programs. It could also address the interest of funders in the extent to which goal-setting and goal-pursuit activities may offer a cost-effective means of enhancing the impact of mentoring programs on youth (for an indepth discussion of issues involved with the economic evaluation of youth mentoring programs, see Foster, this volume, Chapter 34).

Practice Several sets of guidelines have been proposed for the development and implementation of effective mentoring programs for youth (see Big Brothers Big Sisters of America, 2003; Hayashi & O’Donnell, 2004; Roach & Elliott, 2005; Turner-Stokes, 2009). For the most part, however, goal setting and goal pursuit have not been a focus of recommended practice guidelines (e.g., MENTOR, 2009). There are, however, some noteworthy exceptions. Big Brothers Big Sisters of Canada (2011) proposed that setting goals can help meet the need of establishing a shared purpose in the mentoring relationship. To that end, the organization has addressed goal setting in its national standards for matching and match monitoring. These guidelines emphasize that (a) goals are required for all matches, (b) the word goal is simply an objective, an aim, an aspiration, or a desired result for the match toward which the match participants direct their efforts—the process of creating goals is far more significant to healthy outcomes than the actual goal itself, (c) goals must be youth centered (relevant to the youth’s life) and must be developed in collaboration, which means that the youth should be involved directly in setting and establishing goals, (d) goal setting can serve as an orientation to the relationship, particularly in the early phase of the match (however, the process does not have to be done at

92  MENTORING RELATIONSHIPS the match introduction meeting and, in fact, it is not advised at this point; allowing up to 4 months for the pairs to come up with their goals is recommended), and (e) goal setting allows matches to celebrate success and encourages achievements and accomplishments. These recommendations align closely with the theory and research we have reviewed in this chapter as well as our own recommendations for practice that are presented in this section. Based on available research, which we acknowledged has significant limitations, and on our experiences with youth and the development of goal-driven interventions designed to meet the needs of participating youth (e.g., Balcazar, GarateSerafini, & Keys, 2004; Balcazar, Keys, & GarateSerafini, 1995; Balcazar, Taylor-Ritzler, Dimpfl,

Portillo, Guzman, & Murvay, in press; McDonald, Keys, & Balcazar, 2007; Taylor-Ritzler et al., 2001), we recommend designing mentoring programs that include evaluation strategies that yield information regarding goal setting and pursuit, and its association with program outcomes. Turner-Stokes (2009) similarly suggests that GAS and standardized measures be used side by side in evaluating programs. In any case, GAS (or like methods) should be introduced with caution, making sure that the desires and wishes of the mentees are respected and that the youth are ready and motivated to set and pursue personal goals with the support of their mentors. Table 6.1 summarizes guidelines, derived as closely as possible from the research literature, that can inform the process of supporting goal setting and pursuit in mentoring relationships.

Table 6.1   Checklist for Practitioners Topic

Recommendations

Before goal setting

Evaluate the resources and support available onsite for goal setting and goal pursuit. Programs must have consistency in implementation and regular maintenance. Mentors should be familiar with the goal-setting process and learn how to use goal attainment scaling.

When to use goal setting and goal pursuit in mentoring programs for youth

Goal setting and goal pursuit should be considered (a) if the mentoring program is seeking to encourage a sense of future orientation and self-direction, (b) if goals will promote involvement by youth and by mentors, and/or (c) if the program aims to achieve particular objectives in the lives of the youth (i.e., behavior change, skill acquisition).

When to use goal setting and goal pursuit in mentoring relationships for youth

Goal setting and goal pursuit should be initiated in mentoring relationships with youth when (a) rapport and trust have been established between mentor and mentee, (b) the mentee is ready to begin talking openly and directly with the mentor, and (c) the goal-setting process is likely to give direction and/or clear purpose to the relationship.

Self-determination and collaboration in setting goals and goal pursuit

Mentors and mentees should collaborate to set goals and achieve positive outcomes; further, mentors should provide training in goal setting to their mentees, while allowing youth to take the lead in setting goals, taking into consideration their needs, desires, and interests. Attention should be paid to assessing the characteristics of the relationship between mentors and youth while considering the context. Youth should be encouraged to set and pursue their own goals, solve problems, and seek support as needed. Flexibility should be allowed in the goal-setting and goal-pursuit process, given that life situations and priorities could change over the course of the relationship. Goals should be consistent with youth’s values and culture.

Goals in Mentoring Relationships   93

Topic

Recommendations

Approach to goal setting and goal pursuit

Goal attainment scaling (or other comparable methods) should be used; this approach is individualized, with flexibility and accountability. Goal attainment scaling or another similar approach should be used to help youth pursue their goals by collaborating with their mentors to set realistic and specific tasks that break down the process into smaller, more attainable pieces. Goal pursuit should be monitored on a regular basis (once a week or at least twice a month). Mentors, with support and guidance from program staff, should offer help to overcome barriers that may appear to block progress. Mentors and others (e.g., program staff) should offer praise to reinforce progress and efforts of youth in pursuing their goals. Goal attainment should be celebrated.

We focus here on the process of implementing goal setting and GAS in order to assess outcomes achieved by participating youth. Because the scale is based on an individualized measurement system, program participants and their mentors are actively involved in specifying the goals and outcomes to be measured (see Figure 6.1). Examples of possibly stated goals are “to get a summer job,” “to open up a savings account,” “to get a B in math this semester,” “to improve my relationship with my sister,” and “to stay in school” (Balcazar et al., 1995). The process of GAS ideally will give participating youth an opportunity to increase self-awareness and develop a sense of themselves as individuals with the power to affect their lives and their futures in a positive way. The information contained on the GAS form can be collected by the mentor, who can summarize the various actions taken by the youth. Using the example of “getting a B in math” as a goal, the mentor and mentee start by rating the goal on its level of difficulty (1 = not difficult or easy [e.g., practice math problems 15 minutes per day]; 2 = moderately difficult [e.g., get a C in math this semester]; and 3 = difficult [e.g., get a B in math this semester]). Youth are encouraged to think about their goals in a realistic way. We recommend adding a measure of the level of functioning of the youth in order to account for individual differences and allow comparison among groups of participants. We have used simple ratings of levels of functioning with individuals with disabilities that could be modified to address levels of personal competence with values of 1 = high functioning (e.g.,

independent; interested in and able to pursue goals with minimal/moderate amount of help from mentor) and 2 = low functioning (e.g., dependent; unable to pursue goals on his or her own; unmotivated or requires extra help). The goal attainment scale can be determined according to the following criteria: (–1) deterioration (e.g., student dropped out of school), (0) goal dropped, (1) active pursuit of the goal, little progress (e.g., “I ended up with a D– in math class”), (2) active pursuit of the goal, some progress (e.g., “I ended up with a D+ in math class”), (3) accomplish less than desired goal, action may continue (e.g., “I got a C in math this semester”), (4) goal attained as expected, no further actions (e.g., “I got a B in math this semester”), and (5) accomplished more than expected success (e.g., “I got an A in math this semester”). We recommend that each pair identify the various possible outcomes from the beginning, so there will be no surprises when the pair comes together to evaluate the goal and determine the outcome. The result is a weighted goal score.2 Each mentee’s weighted goal attainment score—with a range from a low of –8 to a high of 40 points, depending on the weight values—could be compared across participants. Once the initial goals are written by the participating youth with the necessary support from their mentors, and they agree on the behaviors that would result in each of the values on the scale, the mentors and mentees can track progress in goal pursuit. Mentees are encouraged to set as many goals as they feel they can handle, drop or change their goals when desired, and propose new goals as their

Weighted goal attainment score = goal attainment score * ([degree of goal difficulty] + 2.5 [level of functioning]), where goal attainment score is –1 (deterioration) to 5 (more than expected success); degree of difficulty is 1 (easy), 2 (moderate), or 3 (difficult); and level of functioning is 1 (high functioning) or 2 (low functioning).

2

94  MENTORING RELATIONSHIPS circumstances may change. Mentors should receive training on goal setting and the application of GAS. This process involves practicing goal setting (see step 1 below) and identifying examples of the various levels of goal attainment for different types of goals (from –1 to 5), calculating several goal attainment scores using different types of hypothetical mentees, and examining different ways to encourage and support mentees in their goal pursuit. This training usually takes about 3 hours. The following are steps for implementing GAS in the context of youth mentoring programs: Step 1: Set goals. Mentor-mentee dyads are encouraged to set SMART goals (specific,

Figure 6.1

measurable, achievable, realistic, and time-related). This strategy is recommended to help mentors and mentees set goals that could be easily recognized and attained in a relatively short period of time (see Bovend’Eerdt et al., 2008, for a practical guide on writing SMART goals). Mentors are encouraged to teach mentees how to set SMART goals. Figure 6.1 presents an example of a goal-setting form used by the Big Brothers Big Sisters of Canada (2011). The form lists the goals, a brief description of the general strategies that the mentor and mentee identified to accomplish each goal, and the goal outcomes. The form becomes part of the official documentation of the mentoring relationship and is signed by a field supervisor.

Sample Form for Goal Setting From the National Standards of Big Brothers Big Sisters of Canada

Little Brother/Little Sister Name Goals 1 2 3 Strategies to Accomplish Goals Goal 1: Goal 2: Goal 3: Goal Outcomes Goal 1:

Outcome:

Description of Results

Goal 2:

Outcome:

Description of Results

Goal 3:

Outcome:

Description of Results

Caseworker Signature Source: Big Brothers Big Sisters of Canada.

Date:

Goals in Mentoring Relationships   95 Step 2: Assign a level of difficulty for each goal. Difficulty scores are decided by consensus between the mentor and the mentee. These scores are assigned for each goal on the basis of the perceived degree of complexity (i.e., number of steps required and/or difficulty of the activities involved). The levels are as follows: 1 = not difficult/easy, 2 = moderately difficult, 3 = difficult/ very difficult. The levels are determined using the participants’ input as to what is challenging and nonchallenging for them. Step 3: Determine the level of functioning. Level of functioning (1 = high functioning and 2 = low functioning) is assigned based on the abilities and capacity of the youth to pursue his or her goals independently. Youth who require significant extra help or who have limited experience or supports would be classified as low functioning. This information is obtained by engaging with the youth, by learning about his or her history (e.g., presence of a disability), and by talking to the parents and/or mentoring program staff. Because this value is used to weight the goal attainment score, it is not necessary to have it early in the mentoring relationship. It may be added once the mentor has sufficient information about the youth in order to make this judgment. The additional weight added to this factor was determined by a panel of experts (30 vocational rehabilitation counselors with extensive experience providing services to individuals with all types of disabilities) who

recommended adding the 2.5 additional weight because individuals with low levels of functioning require additional supports and assistance and are usually less likely to succeed in attaining their goals. Step 4: Determine the outcome for each goal. In this step, the mentees have the opportunity to assess the progress they have made toward achieving each goal by discussing their actions with their mentors. Mentors will start by reviewing the information they have gathered thus far from their interactions with the youth and by asking for clarification and updates along the way. The youth are asked to give details about their actions, attempted and completed. Sample probes include, “What steps have you taken to reach this goal since the last time we met?” “When did you take these steps?” and “Are you going to continue with this goal?” Using the information provided by the youth, the mentor and the mentee determine the status of each goal and assign a value (from –1 to 5) for the goal attainment score using the form shown in Figure 6.2. Remember that the various outcomes possibilities were identified after the pair wrote the goal. If a goal is not achieved, consider the following factors (from Big Brothers Big Sisters of Canada, 2011): (a) It may have been too difficult to achieve (see if it can be broken into smaller pieces), (b) it may have been developed without the active involvement and commitment of the youth (ask the youth about it and make appropriate corrections as needed), and

Figure 6.2  Goal Attainment Scale Participant:_________________________ Mentor:___________________________________________________________ Goal:__________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ Level of Difficulty: __________ Date goal set: ___________ Date goal reviewed: ___________________ GOAL ATTAINMENT SCORE (–1) Deterioration (0) Goal dropped, no intent to continue (1) Active pursuit of goal, little progress (2) Active pursuit of goal, some progress (3) Accomplished less than desired goal (4) Accomplished goal as expected (5) Accomplished more than expected success

ACTIONS/STEPS

96  MENTORING RELATIONSHIPS (c) the youth may be fearful of achieving a selfenhancing goal, or may have low self-esteem, or fear of being successful (talk to the youth to find out if these factors are present and, if so, work on addressing those issues before continuing with more goal setting). Step 5: Repeat the process with each goal proposed. The previous steps are repeated for any additional goals that may be set during the mentoring relationship. Goals that were dropped are also scored. Youth should be asked to review their lists of goals and progress reports to check for accuracy and get the input from their mentors. This information can also reinforce youth for their efforts at goal pursuit. Once each of the weighted goal attainment scores is computed, the mentor can aggregate the scores for all the goals the mentee pursued and report a total score, which we call the Total Goal Accomplishment Score per mentee.

Conclusion In this chapter, we have argued that mentoring relationships and programs that use goal setting and goal pursuit as part of training youth to plan and think positively about their future can be empowering and give purpose and direction to the relationships that form between mentors and mentees. Although many traditional mentoring programs do not actively incorporate goal setting, we encourage program developers to consider doing so in part because of the potential for goals to help youth develop both a better understanding of the potential benefits of the relationship and a greater appreciation for the feedback and support received from the mentor. One way to prepare youth to engage in goal setting and goal pursuit is to increase their awareness about their own strengths and to help them learn to recruit help from others. Despite a surprising lack of relevant research, we see great potential in this regard for natural mentoring. Learning help-recruiting skills may enable youth to recognize individuals in their social environment as potential mentors and learn how to approach them and initiate relationships that can be mutually beneficial for both the youth and the mentors (Balcazar et al., 1995). Within the context of such relationships, goal-pursuing youth may become more engaged in the mentoring process and proactive in goal attainment. A further potentially beneficial aspect of goal setting is its capacity to facilitate assessment of

program outcomes in terms of attainment of personal goals. That was the main necessity of the programs we collaborated with in the United Kingdom. Funders had received documentation about the process and the characteristics of the mentors and mentees, but they wanted to know more about the ways the programs were helping the participating youth. Satisfaction questionnaires and anecdotal reports were insufficient. Turner-Stokes (2009) also argued that GAS provides a useful reflection on outcomes that are of critical importance to the participants in the context of their own lives, which is something that traditional assessment measures do not provide. Although mentoring programs are often presented as being successful or beneficial without data to support these declarations, research evidence suggests that some programs can have detrimental effects on youth (DuBois et al., 2011; Rhodes et al., 2005). Therefore, high-quality evaluation of mentoring programs that involve goal setting, pursuit, and attainment is clearly indicated. We see it as important in this regard to recognize the limitations of goal setting and to take heed of related cautions. As we emphasized repeatedly in this chapter, the process of goal setting and goal pursuit needs to be implemented with care in order to promote respect for the youth’s desires and avoid mentor domination. In addition, the goal-setting process should not be rushed because many mentees can be expected to need time and experience to build trust in mentors and to develop an understanding of the purpose and personal value of a goal orientation as applied to their personal circumstances. We underscore in this regard our recommendation that both mentors and mentees receive appropriate training on goal setting and goal pursuit in order to minimize such risks and maximize the likelihood of goal attainment. In closing, we believe the weight of available evidence indicates that goal setting and goal pursuit, although likely to be controversial in some quarters, can be an important and value-added component in some mentoring relationships and programs for youth. Goal attainment scaling has played a constructive role in a number of programs that promote self-improvement and has demonstrated potential for benefiting mentoring programs and relationships as well. We anticipate that as the role of goal setting and goal pursuit in mentoring for youth becomes more widely explored through research, best practices will become more fully developed and thus the potential benefits of goal setting and goal pursuit will be more fully realized.

Goals in Mentoring Relationships   97

References Balcazar, F. E., Davies, G. L., Viggers, D., & Tranter, D. (2006). Goal attainments scaling as an effective strategy to assess the outcomes of mentoring programs for troubled youth. International Journal of School Disaffection, 4(1), 43–52. Balcazar, F. E., Garate-Serafini, T., & Keys, C. B. (2004). The need for action when conducting intervention research: The multiple roles of community psychologist. American Journal of Community Psychology, 33, 243–252. Balcazar, F. E., Keys, C. B., & Garate-Serafini, J. (1995). Learning to recruit assistance to attain transition goals: A program for adjudicated youth with disabilities. Remedial and Special Education, 16, 237–246. Balcazar, F. E., Taylor-Ritzler, T., Dimpfl, S., Portillo, N., Guzman, A., & Murvay, M., (in press). Improving the transition outcomes of low-income minority youth with disabilities. Exceptionality. Big Brothers Big Sisters of America. (2003, October). Standards of practice for one-to-one service for programs operating within affiliated sponsoring organizations. Philadelphia: Author. Retrieved from http://login.npwebsiteservices.com/Big_Brothers_ Big_Sisters_oNLFPLG/StandardsOfPractice BBBSA.pdf Big Brothers Big Sisters of Canada. (2011). National standard R1 matching and match monitoring. Burlington, Ontario: Author. Blum, D. J., & Jones, A. L. (1993). Academic growth group and mentoring program for potential dropouts. School Counselor, 40, 207–217. Bovend’Eerdt, T. J., Botell, R. E., & Wade, D. T. (2008). Writing SMART rehabilitation goals and achieving goal attainment scaling: A practical guide. Clinical Rehabilitation, 23, 352–361. Brunstein, J. C., Schultheiss, O. C., & Maier, G. W. (1999). The pursuit of personal goals: A motivational approach to well-being and life adjustment. In J. Brandstädter & R. M. Lerner (Eds.), Action & self-development: Theory and research through the life span (pp. 169–196). Thousand Oaks, CA: Sage. Dennison, S. (2000). A win-win peer mentoring and tutoring program: A collaborative model. Journal of Primary Prevention, 20, 161–174. Deutsch, N. L., & Spencer, R. (2009). Capturing the magic: Assessing the quality of youth mentoring relationships. New Directions for Youth Development, 121, 47–70. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Emmerson, G. J., & Neely, M. A. (1988). Two adaptable, valid, and reliable data-collection measures: Goal attainment scaling and the semantic differential. Counseling Psychologist, 16, 261–271.

Forneris, T., Danish, S., & Scott, D. (2007). Setting goals, solving problems, and seeking social support: Developing adolescents’ abilities through a life skills program. Adolescence, 42, 103–114. Garate-Serafini, T. J., Balcazar, F. E., Keys, C. B., & Weitlauf, J. (2001). School-based mentoring for minority youth: Program components and evaluation strategies. In R. Majors & J. Jollisfe (Eds.), The race education revolution: New directions and critical approaches to educating Black children (pp. 214–257). London: Falmer Press. Garwick, G. (1974). An introduction to reliability and the goal attainment scaling methodology. In T. J. Kiresuk & G. Garwick (Eds.), Program evaluation project report (pp. 65–67). Minneapolis, MN: Program Evaluation Project. Hamilton, M. A., & Hamilton, S. F. (2005). Work and service-learning. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 348–363). Thousand Oaks, CA: Sage. Hayashi, Y., & O’Donnell, C. R. (2004) A review of studies and websites: A report for the Melissa Institute for the Prevention and Treatment of Violence. Retrieved from http://www.melissainstitute.org/ documents/TMI_Mentoring_Report51-2.pdf Hurn, J., Kneebone, I., & Cropley, M. (2006). Goal setting as an outcome measure: A systematic review. Clinical Rehabilitation, 20, 756–772. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?” Testing a framework to guide mentor training and activity selection. New Directions for Youth Development, 2010 126, 51–69. Karcher, M. J., Nakkula, M. J., & Harris, J. (2005). Developmental mentoring match characteristics: Correspondence between mentors’ and mentees’ assessments of relationship quality. Journal of Primary Prevention, 26, 93–110. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois, & M. J. Karcher, (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. King, K. A., Vidourek, R. A., Davis, B., & McClellan, W. (2002). Increasing self-esteem and school connectedness through a multidimensional mentoring program. Journal of School Health, 72, 294–299. Kiresuk, T. J., & Choate, R. O. (1994). Applications of goal attainment scaling. In T. J. Kiresuk, A. Smith, & J. E. Cardillo, (Eds.), Goal attainment scaling: Applications, theory, and measurement (pp. 61–104). Hillsdale, NJ: Erlbaum. Kiresuk, T., & Sherman, R. (1968). Goal attainment scaling: A general method of evaluating comprehensive mental health programs. Journal of Community Mental Health, 4, 443–453. Kiresuk, T. J., Smith, A., & Cardillo, J. E. (Eds.). (1994). Goal attainment scaling: Applications, theory, and measurement. Hillsdale, NJ: Erlbaum. Langhout, R. D., Rhodes, J. E., & Osborne, L. (2004). An exploratory study of youth mentoring in an urban

98  MENTORING RELATIONSHIPS context: Adolescents’ perceptions of relationship styles. Journal of Youth and Adolescence, 33, 293–306. Larose, S., Tarabulsy, G., & Cyrenne, D. (2005). Perceived autonomy and relatedness as moderating the impact of teacher student mentoring relationships on student academic adjustment, Journal of Primary Prevention, 26, 111–128. LoSciuto, L., Rajala, A. K., Townsend, T. N., & Taylor, A. S. (1996). An outcome evaluation of Across Ages: An intergenerational mentoring approach to drug prevention. Journal of Adolescent Research, 11, 116–129. Malec, J. F. (1999). Goal attainment scaling in rehabilitation. Neuropsychological Rehabilitation, 9, 253–275. Marson, S. M., Wei, G., & Wasserman, D. (2009). A reliability analysis of goal attainment scale (GAS) weights. American Journal of Evaluation, 20, 203–216. McDonald, K. E., Keys, C. B., & Balcazar, F. E. (2007). Disability, race/ethnicity, and gender: Themes of cultural oppression, acts of individual resistance. American Journal of Community Psychology, 39, 145–161. McDougall, J., & Wright, V. (2009). The ICF-CY and attainment scaling: Benefits of their combined use for pediatric practice. Disability and Rehabilitation, 31, 1362–1372. MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. Nakkula, M., & Harris, J. (2005). Assessment of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 100–117). Thousand Oaks, CA: Sage. Portwood, S. G., Ayers, P. M., Kinnison, K. E., Waris, R. G., & Wise, D. L. (2005). Youth Friends: Outcomes from a school-based mentoring program. Journal of Primary Prevention, 26, 129–145. Rhodes, J. E. (2005). A theoretical model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43). Thousand Oaks, CA: Sage.

Rhodes, J. E., Reddy, R., Roffman, J., & Grossman, J. B. (2005). Promoting successful youth mentoring relationships: A preliminary screening questionnaire. Journal of Primary Prevention, 26, 146–167. Roach, A. T., & Elliot, S. N. (2005). Goal attainment scaling: An efficient and effective approach to monitoring student progress. Teaching Exceptional Children, 37(4), 8–17. Sipe, C. L. (2005). Toward a typology of mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 65–79). Thousand Oaks, CA: Sage. Steenbeek, D., Gorter, J. W., Ketelaar, M., Galama, K., & Lindeman, E. (2011). Responsiveness of goal attainment scaling in comparison to two standardized measures in outcome evaluation of children with cerebral palsy. Clinical Rehabilitation, 25, 1128–1139. Steenbeek, D., Ketelaar, M., Galama, K., & Gorter, J. W. (2007). Goal attainment scaling in pediatric rehabilitation: A critical review of the literature. Developmental Medicine & Child Neurology, 49, 550–556. Taylor-Ritzler, T., Balcazar, F., Keys, C., Hayes, E., & Garate-Serafini, T. (2001). Promoting attainment of transition related goals among low-income ethnic minority students with disabilities. Career Develop­ ment for Exceptional Individuals, 2, 147–167. Tennant, A. (2007). Goal attainment scaling: Current methodological challenges. Disability and Rehabilitation. 29, 1583–1588. Turner-Stokes, L. (2009). Goal attainment scaling (GAS) in rehabilitation: A practical guide. Clinical Rehabili­ tation, 23, 362–370. Turner-Stokes, L., & Williams, H. (2010). Goal attainment scaling: A direct comparison of alternative rating methods. Clinical Rehabilitation, 24, 66–73. Vansteenkiste, M., Simons, J., Lens, W., Soenens, B., & Matos L. (2005). Examining the motivational impact of intrinsic versus extrinsic goal framing and autonomysupportive versus internally controlling communication style on early adolescents’ academic achievement. Child Development, 76, 483–501.

7  MENTORING RELATIONSHIPS IN DEVELOPMENTAL PERSPECTIVE Gil G. Noam, Tina Malti, and Michael J. Karcher

Introduction Adaptive and positive development occurs in and with the support of significant relationships. Despite this fact, developmental research has guided mentoring practices only marginally. An important step is to provide a review of important theories and research about the mentoring of the developing child and adolescent. To reign in this broad topic, we focus on development and one-on-one mentoring relationships from middle childhood to middle adolescence, leaving young children, older adolescents, and young adults for other contributions. A critical question for us is the “fit” between children’s and adolescents’ developmental needs and characteristics and mentoring opportunities in order to gain insight into the gap between the promise and current research on mentoring. We will also introduce a new integrative framework built on our resilience research that can help in tailoring mentoring practices and programs to specific developmental needs and opportunities of young people. The time from middle childhood to middle adolescence is characterized by significant change and transition (E. Erikson, 1968). As young people move from middle childhood to adolescence, they typically experience an expansion in their social networks, and develop self-concepts that include for the first time a simultaneous awareness of their past, current, and future selves. Adolescents also experience a shift in relations with parents at the same time as their relations with peers take on new qualities of reciprocity and intimacy (Steinberg & Morris, 2001). As young people navigate the changes from childhood to adolescence, it is critical that they

receive adult guidance, support, and encouragement that foster adaptive and productive development. Mentoring practice is shaped by these developmental considerations, though often quite implicitly. MENTOR (2003) defines mentoring as occurring within relationships that bring “young people together with caring individuals who offer guidance, support, and encouragement aimed at developing the competence and character of the mentee” (p. 9). The existence of a committed and encouraging adult who believes in “me and my future” has proved to be an essential ingredient for many resilient youth who succeed despite great adversity (Noam & Hermann, 2002). While such relationships are strongly linked to the course of development of children and youth, too little direct connection has been made to the strong research base coming from the developmental sciences. Children and adolescents can be understood as active meaning-makers who construct and invent their world while adapting to it (e.g., Piaget, 1954). They can even overcome previous risks and trauma through their ability to give new meaning to old events and through exploration of alternative thoughts and behaviors. Explicitly or implicitly, mentors (themselves active meaning-makers) consider the developmental capacities, needs, and circumstances of their mentees. They must do so for the relationship to last and to deepen, because the way the mentoring pair make meaning of the world, each other, and the future can differ in fundamental ways. Bridges have to be built and common interests found so that the diversity of thoughts, feelings, and behaviors between them can enrich rather than derail their relationships and growth. 99

100  MENTORING RELATIONSHIPS Given the importance of understanding this active sense-making by mentors as well as mentees, it is surprising how little in-depth developmental knowledge and training has entered the mentoring world. It is not enough to know about the power of relationships. It is essential to view relationships as evolving, in no small part based on what motivates children and adolescents, depending on their developmental status. We share our perspective on “developmental mentoring” to aid program staff in creating environments and encouraging skills, among both mentors and mentees, that augment the natural processes of development-in-relationship. Our goal is to take developmental assumptions from being implicit and undefined in the field of mentoring generally, to explicit and thus more practical and useable.

Theory Because development always occurs in the context of significant relationships (Bowlby, 1969; Sroufe, 1997), the establishment of mentoring relationships with caring adults can be an effective strategy for fostering development in young people (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011; Noam & Hermann, 2002). But, specifically, how is a positive mentoring relationship established at different ages and how are positive developmental outcomes, such as prosocial values and connectedness to school, reached through interpersonal interventions such as youth mentoring? Developmental theories can help us understand how young people and mentors approach close relationships, which might provide insight into the processes that differentiate successful and unsuccessful matches (Scheithauer, Malti, & Noam, 2009). The processes of greatest interest for understanding mentoring and other significant relationships are human emotions, social-cognition, and the tandem processes of risk and resilience. We compare and contrast these processes within three major developmental paradigms. The first tradition regards attachment and how it shapes well-being, building on John Bowlby’s (1969) seminal work on the consequences of different styles of mother/caregiver-child relationships and the internal relational working models to which they give rise. The second theoretical tradition, introduced by Erik Erikson, is what we have called a “functionalist” approach (Noam, Kohlberg, & Snarey, 1983; Snarey, Kohlberg, & Noam, 1983). This approach emphasizes life tasks, such as school success, identity development, and the like, that need to be fulfilled at

key ages in the lifespan. The third perspective is that of cognitive development, building on Piaget’s (1954) work, but having been transformed by others into an integrative social-cognitive and socialemotional developmental perspective (Kohlberg, 1994; Loevinger, 1976; Noam, 1988). In the sections that follow we discuss these three essential developmental theories as they help explain or define the two transitions that are the focus of this chapter, and that have been the primary age range of most research on youth mentoring to date— that is, the move from middle childhood to early adolescence and from early adolescence to middle adolescence. In the final section of the chapter, we present a new developmental framework that arises out of the strengths and limitations of the existing paradigms. The clover model, also called the recursivetransformational model, relates developmental strengths and vulnerabilities and has implications for strengthening mentoring relationships and the programs that support them. Attachment Perspective on Development A developmental theory that is useful for understanding relationships between mentors and both children and adolescents is attachment theory (Bowlby, 1969). The attachment model has been used widely to describe individual differences in how children and adolescents interact with significant others, including parents, peers, and romantic partners. Attachment styles describe individual differences in representations of the availability and quality of support, beliefs about the extent to which the self is worthy of receiving love from others, and perceptions regarding appropriate responses to the distress of others, all of which contribute to the person’s internal working model of self-in-relationship. For those new to attachment theory, we describe its assessment and resulting classifications below. Attachment styles are inferred, during early childhood, through a procedure called the “strange situation,” in which adults observe the degree to which the child uses the parent as a safe base for exploring his or her environment. The child’s openness to initiating exploration of the surroundings, in both the presence and the absence of the primary caregiver, as well as the child’s seeking out of the parent (proximity seeking) and ability to be soothed by the return of an absent caregiver are the hallmarks of a secure attachment or internal working model of relationships. Attachment styles fall into three general categories reflecting the presence of anxiety feelings

Mentoring Relationships in Developmental Perspective    101 and avoidant behaviors in response to conflict, separations, and perceived loss in relationships. These styles reflect predictable behavioral responses that are formed in response to children’s perceptions of the security of their primary relationships. Some youth reflect a secure pattern. They demonstrate short-term anxiety in response to separations from the parent or caregiver, seek comfort and reassurance from the loved one when he or she returns, and are relieved by the loved one’s return, which is evidenced by their return to actively exploring their environment. Children who have experienced an inability to count on the caregiver’s return—or to expect that they will be comforted upon the loved one’s return—develop anxious expectations about their security in important relationships. Some become anxious-ambivalent, never really trusting that others will care for them over the long run. Such children frequently test their relationships by presenting demands for reassurance. The anxiousavoidant style reflects a lack of trust in others and a consequential distancing of the self from others. These children act as if they don’t need others, avoid closeness, resist comfort, and reject attempts at closeness as a protective mechanism. Obviously, we have written these descriptions quite generally, to make the transfer of these concepts to the mentoring relationship more direct. Attachment styles form in childhood and typically remain stable over time, yet they can be mollified by corrective emotional relationships that are positive and allow the child to overcome the anxiety they expect in relationships. Conversely, a secure attachment can be fractured by traumatic experience in later childhood, adolescence, or adulthood. In both childhood and adolescence, as well as in adulthood, attachment styles are revealed through the presence of anxiety, ambivalence, and avoidance in close relationships. For example, secure adolescents are low in anxiety and avoidance, whereas anxious-avoidant (insecure) adolescents are low in anxiety but high in avoidance. Anxious-ambivalent (insecure) adolescents are high in anxiety and low in avoidance. These qualities should not be confused with biologically based shyness or extraversion. Attachment styles reflect a set of beliefs, albeit unconscious, about the trust and security individuals expect from close relationships, including with parents, peers, or program-based mentors. Attachment has been linked increasingly to a “risk and resilience” perspective, which is also essential to understanding the risks and benefits of mentoring (Benard & Slade, 2009). For example, attachment style may influence the ability of adolescents to successfully attract mentors and maintain

relationships with them. Obviously anxiously attached adolescents who either test or dismiss and avoid relationships with adults are less likely to secure naturally forming mentoring relationships with adults; this makes them less resilient in times of stress without the available support of one or more natural mentors in their lives (Werner, 1986). Similarly, in program-based mentoring relationships, attachment styles may interfere with successful relationship development. Because attachment behaviors are most often elicited by high levels of stress and insecurity (Bowlby, 1969), conflict in mentor-mentee relationships are likely to evoke negative responses from individuals with insecure attachment patterns, whether the insecurity is the adolescent’s or the mentor’s. For example, fearful-avoidant youth who are at their core emotionally needy may be particularly likely to evoke the abandonment that they fear by rejecting their mentors or as a result of testing them to the point of exhaustion. Thus, attachment styles matter for the establishment and maintenance of mentoring relationships. Although attachment behaviors are first elicited in childhood, later attachment experiences in childhood also influence subsequent interpersonal behavior, relationship building, and the development of psychopathology. Mentoring is successful, from an attachment perspective, when a positive mentoring relationship becomes internalized or part of the internal working model (i.e., what is experienced outside the youth in the match becomes part of the inner life of the person) over time, altering the youth’s internal working model of relationships—their representational view of self and expectations of others. This is the essence of what successful mentoring provides many youth who need mentoring as a form of corrective emotional experience. As in infancy, though less intensely, this process happens again and again in close relationships throughout life and is what mentoring relationships attempt to do, as well. Functionalist Perspective on Development Erik Erikson’s epigenetic (biopsychosocial) model (1950) is a grand synthesis and extension of Freudian ego psychology that established the “eight ages of human being,” took psychoanalysis and developmental theory beyond childhood, provided an outline of adaptational mechanisms (“virtues”) at each age of the life cycle, as well as revealed the typical conflicts and vulnerabilities that can result from the failure to successfully achieve developmental competency (see J. M. Erikson, 1988). The theory describes the individual’s progressions through eight stages from

102  MENTORING RELATIONSHIPS early dependency and trust toward increasing individuation and later engagement with community, history, and culture. Each stage reflects new social demands— such as school success, intimate relationships, parenting, and retirement—that coax the emergence of skills that build on prior developmental accomplishments and vulnerabilities. Of particular importance for this chapter are the developmental phases of middle childhood and adolescence. In middle childhood, the child’s tasks are to acquire knowledge and skills, whereas in adolescence (especially middle adolescence) the tasks are to question identifications, take a perspective on the past, and develop a more cohesive sense of self, one that is consistent across time, contexts, and relationships. Erikson described these developmental tasks as the achievements of initiative, industry, and identity. These stages, notably, follow or build upon the developmental task of establishing (or experiencing) trust in close relationships. Trust (versus its absence, mistrust—what Bowlby called anxious-ambivalence or anxious-avoidance) provides security for the child to explore his or her environment. Thus, trust sets the stage for initiation of the self—what we refer to later, in the clover model, as the child’s assertion. Assertion is also apparent in Erikson’s next developmental task, that of industry. In middle childhood, the child moves beyond simply exploring the world; she begins to identify things she enjoys and is good at doing, and thus begins to develop mastery at some tasks relative to others. But recall that these stages are sequential, with each one building on the former. The child whose expectation of mistrust has inhibited her initiative into the larger world will have a much more constrained set of opportunities for industriousness or the mastery of skills and subsequently the identification of things uniquely of interest to him or her. In this way, industry is constrained by prior failures in development. On the other hand, each new stage provides the opportunity to overcome the significance of earlier vulnerabilities. Ego identity development, the third of Erikson’s tasks we consider, is the conscious sense of self that we acquire and develop in our daily interactions with others. It is shaped most dramatically in adolescence and it too is dependent on prior developmental achievements. The affirmation of a sense of selfworth in the elementary school years resulting from children’s successful initiation and assertion of themselves in their environment (initiative) allows the exercise of industrious engagement in activities and tasks that affords skills and mastery of interests (industry). It becomes the mid-adolescent’s task to integrate these interests, roles, and self-perceptions

into a cohesive self-concept that informs one’s identity development. Erikson’s model, however, also reveals the risks posed by development. On the one hand, the successful mastery of life tasks leads to growth of ego strength and fosters resilience or the capacity to overcome adversity or to persevere through challenging times. On the other hand, the shadow side (resulting from an unresolved crisis and the unsatisfactory achievement of functional competencies at earlier stages) causes disaffection, identity diffusion, and other maladaptive outcomes (E. Erikson, 1968). At each stage, people experience a typical conflict that serves as a turning point in development, either putting them at risk or increasing their resilience. These conflicts demand resolution, which can result in one’s developing a strengthening psychological quality or conversely revealing a potentially toxic shadow side of development. What is not always so evident in Erikson’s theory is the importance of belongingness at each stage after the first (“trust versus mistrust”) and the role it plays in supporting these developmental achievements. For example, a child may have the need to acquire skills, interests, or information, which can be achieved by specific structured activities (games, crafts, sports) that are largely cooperative and social in nature (Karcher & Benne, 2008). But the early adolescent has a need to explore and to see how he or she is viewed from the perspective of others, and thus has an equally strong but qualitatively different need for belonging and support (Noam, 1999). This sense of belonging is essential for development. Joan Erikson best illustrated the importance of belonging with her description of a “radius of significant others” that expands across development (J. M. Erikson, 1988; Karcher & Benne, 2008). In summary, in terms of Erikson’s functional approach, the focus of mentoring conversations and activities may be more fruitful if they address what is expected of the youth at each age (see Karcher & Hansen, this volume, Chapter 5). The mentor will want to help the child succeed in addressing the typical dilemmas, challenges, or tasks specific to the youth’s development, such that successfully mentoring a child will look very different than successfully mentoring an adolescent. The focus of successful interactions differs in terms of the distinct age-appropriate solutions for life tasks presented to youth. Social-Cognitive Perspective on Development Within each child and adolescent is an unfolding social-cognitive structure of meaning-making

Mentoring Relationships in Developmental Perspective    103 that the youth brings into relationships and uses to understand the self, relationships, and the world. Social-cognitive developmental theories describe this development in social cognition in terms of sequences of increasing levels of complexity and differentiation. Most notably is Piaget’s theory of cognitive development involving transformations that take place as an individual moves from one form of thinking to a qualitatively new form of thinking (Piaget, 1954). There are revolutions in how children and adolescents understand themselves (self) and relationships (persons, social and moral systems, and the like), and these distinct cognitions about the social world matter greatly for development and adaptation. Social-cognitive development always comes from the interaction between increasing cognitive maturation and interactions with others over time. Thus Piaget’s theory that gave rise to moral and social-cognitive psychology has a strong biological component but is simultaneously social. Of course, every individual’s social cognitions also will differ depending on that person’s prior relationship experiences, culture, and gender, but the stages have been viewed as general and consistent across cultures and historical moments. This part of Piaget’s theory has been criticized as much as the fact that Erikson tied his stages to the age of a person. Erikson (1950) noted that the social demands on children and adolescents differ in predictable ways and thus create specific crises tied to specific ages. Similarly, there are largely universal systematic socialcognitive styles of meaning-making that are typical for school-age children versus adolescents. When understood by mentors, these styles may facilitate stronger matches through enhanced empathy for the mentee’s perspective on the world. The typical adult who mentors a child must understand these ways of social meaning-making in order to understand what the child or adolescent does and does not mean to communicate and why. Perhaps most important for mentoring are the increasing ability from middle childhood to middle adolescence to take and coordinate the perspectives of self and other and the related changes in emotions one has about the self in relation to others. The mentor needs, to understand if (and how well) the mentee is capable of understanding the mentor’s perspective as well as recognizing the mentee’s own intentions, emotions, needs, and wishes, both in general and as they relate to their mentoring relationship. A mentee will have a general socialcognitive capacity to coordinate social perspectives (her own, her mentor’s, and her peer group’s), rendering her more or less capable of cooperating and taking the mentor’s perspective in negotiations.

The social-cognitive capacity is separate from, though influenced by, the mentee’s prior attachment history or experience of negotiating with adults (e.g., whether or not parents were open to negotiation and thus whether negotiation was modeled or encouraged). These prior experiences will shape how the individual’s social-cognitive capacity for perspective taking will be employed in the mentoring relationship. The young child who is unable (cognitively) or unwilling (emotionally) to coordinate multiple points of view may negotiate game play or match activities very differently than the adolescent who can easily coordinate multiple points of view (Selman & Schultz, 1990). Additionally, whether the adolescent’s full social-cognitive capacity is utilized will also depend on characteristics specific to that youth, such as a girl’s beliefs about what is culturally appropriate behavior for females when interacting with adults. Noam’s constructive-developmental model, a theory he introduced before the new model we present in the next section (Noam, 1988), reconstructs a picture of how developments in sense-making generally and perspective-coordination specifically manifest across childhood and adolescence. Noam is skeptical of the nature of post-Piagetian stages, as there is no evidence that the many aspects of skills, relationships, personality style, and the like should progress in a matter of firm, successive, and irreversible stages. Instead he views them as levels, way stations on a developmental pathway. The subjective-physical level of early childhood is one in which children evaluate actions in terms of physical consequences, in part, because they struggle to link external behavior to internal experiences. Children at this developmental level focus heavily on the world physically and actively, and in terms of most things (people and actions) being all good or all bad. These children make assessments based on external authorities. With the emergence of middle and late childhood comes the reciprocal-instrumental level of social-cognitive development. The intentions of both self and other can be assessed, understood, and used to engage in instrumental (e.g., cooperative) exchanges. The child understands an inner world exists for people apart from their physical appearance—one can smile yet be sad inside. This child has overcome impulsive, action-oriented responses and resolutions to problems. He or she can consider another’s motives, either to see the person eye-to-eye (with perspectives and needs equally valid to one’s own) or to manipulate the other (by knowing what the other person wants and that the other person can’t observe deceit unless

104  MENTORING RELATIONSHIPS one’s face or behaviors convey it directly). This level represents a significant advance over the action-oriented, egocentric thinking that precedes it; but it also evokes a self-consciousness and anxiety, such as among middle schoolers, who discover they can’t take anyone at face value—anyone could be two faced. No one can be trusted fully. But it is only at the next level, the mutualinclusive developmental position, that the youth, now typically approaching or entering high school, fully overcomes self-interest. The youth can see how actions affect others, not just internally (their feelings), but how actions now affect others’ responses in the future. One way to achieve this is to act consistently such that the youth appears, to himself or herself and to others, to be one and the same person—trustworthy and aware of his or her effect on others. Altruism emerges but so, too, does overly conformist behavior. This capacity to see oneself as part of a group, which endures over time and is bound by similarities in interest and mutual affection—can lead youth to be overly inclusive of others’ wishes for the sake of belonging. The mutual-inclusive developmental position also allows youth to have feelings about one’s feelings (such as anxiety, depression, and shame) that are the result of their own self-assessments. Other theoretical accounts in the social-cognitive theory tradition have moved away from a singular focus on social-cognitive development. One model is the social information processing (SIP) theory. According to SIP theory (e.g., Crick & Dodge, 1994), and consistent with Piaget’s theory, children’s and adolescents’ social-cognitive interpretation of a particular event influences their behavior. SIP theory differs from Piaget’s, however, in that the former describes a series of steps through which social information is processed and social behaviors are enacted. From this theoretical approach, the process of receiving and acting on information includes encoding, making attributions, selecting goals, generating responses, evaluating responses, and enacting responses. All of these steps are influenced by a social-cognitive “database” that includes beliefs about behaviors and interaction partners (see Selman, Beardslee, Schultz, Krupa, & Podorefsky, 1986). One style of social information processing is the maladaptive hostile attribution bias, which occurs when a child has the tendency to assume that others feel hostile toward the child. Typically, this style develops in reaction to real experiences of threatening and aggressive behavior of others (e.g., parents, siblings, peers), just as the avoidant attachment style results, in part, from feelings of rejection and neglect. Unfortunately, the typical behavioral

consequence of attributing aggressive feelings to those around oneself is to act in aggressive, selfprotective ways. Similarly, youth who have been subjected to others’ aggressive behaviors often lack sufficient modeling of alternative ways of resolving problems, which leaves them impaired not only in how they process social information, but also in the availability of behavioral responses. In essence, prior experiences and developmental maturation support or constrain effective interpersonal behaviors among youth in ways that mentors would benefit from understanding. From a social-cognitive developmental perspective, the individual uses a database of social information that is constrained by the individual’s social-cognitive capacities or developmental lens. Thus, how a youth responds to a mentor will reflect gendered, cultural, and other social expectations as well as the social demands placed on youth of their age (Erikson’s functionalist view), and also will be shaped by the manner in which prior relationship experiences have shaped their internal representations of self and others (attachment theory). But, according to social-cognitive theories, prior experiences with others have prepared the youth to encode information more or less accurately, to be aware of fewer or more alternative responses to a given problem, and thereby to shape the quality of their mentor-mentee interactions. From a reparative stance, the focus of mentoring can be to help a youth get unstuck from consistently poor, restricted, and defensive (i.e., egocentric) behavior patterns and to emerge from the mentoring relationship with a more adaptive way of thinking about the world and the people in it. Yet, the issue is more complex than that, because one’s defensive behavior may result from an avoidant attachment, the egocentric constraints of a subjective-physical developmental perspective, or culturally conditioned expectational biases— none of which has anything to do with the mentor. The mentor may provide the youth with the best opportunity to learn to behave more effectively. For example, a mentee may have revenge feelings that take the form of attributing hostility to the intention of others based on an avoidant interpersonal style that leads her to attack other people first (and ask questions later). She may be encouraged through the establishment of a more secure experience in the mentoring relationship. The mentor can help by modeling empathy and negotiating skills that use a more complex and adaptive set of skills that allow her to see and experience alternative ways of thinking about interpersonal conflicts and their resolution.

Mentoring Relationships in Developmental Perspective    105 A New Developmental Perspective— The Clover Model The three major theoretical traditions described in the previous section contribute greatly to our understanding of child and adolescent development and should be used increasingly to guide mentoring relationships. Their use also will be enhanced through theoretical advancement. The attachment paradigm is an extremely strong and well-researched method for understanding the vicissitudes of attachment problems well beyond the early parent-child relationship. Erikson has taught us to focus on the social and psychological demands of individuals who exist in a particular moment in history. To successfully rise to the challenges of each new era and each new developmental stage, Erikson illustrates the importance of success for young people in learning, identity development, and intimate relationships. This socio-historical perspective has also been an extremely important part of Vygotsky’s (1978) sociocultural theory, one that deserves its own chapter. Although social-cognitive theories are not as anchored in the essential life histories of individuals as these first two paradigms are, cognitive developmental theories reveal the continued progression in how meaning is made about the physical and social worlds, and this understanding is critical to mentors’ understanding of their mentees’ thoughts, intentions, and behaviors. The cognitive developmental view is crucial for mentors for two reasons. First, the individuals in a mentoring relationship will, by definition, differ in age and “wisdom.” Mentor and mentee will always have to work to bridge their sensemaking. Thus the mentor will need an understanding of these developmental differences to listen without judgment and to empathize. Second, the cognitive developmental understanding reveals a range of cognitive abilities that provides a starting place for contextualizing the youth’s experience within prior relationship (i.e., attachment) experiences and sociohistorical realities facing the youth. Despite providing major insights into human development and vulnerability, each of these theories has left major holes and contradictions. Coming out of psychoanalysis and ethology, attachment theory views the first few years of life as extremely formative. Although there is an updating mechanism of early attachment “working models,” the internalized ways in which interactions with the caregiver become part of the inner life, missing is a truly developmental theory. How do attachments develop later and normatively, not just how do the early attachments become revisited, replayed, or repaired?

Erikson’s work is the most comprehensive, having introduced a far-reaching approach to the whole life cycle from birth to death, embedding each generation into an historical context and translating Freud’s psychosexual stages to psycho-social ones. He also incorporated normal development and psychopathology into his crisis model. But, with the changing norms of society that followed the highly structured life-cycle expectations of the 1950s in Western society, these stages, tied as they are to ages, do not hold up any more. Girls and women can enter the generativity stage through procreation, which is biologically possible, and through medical advances, as late as age 50 or beyond. Puberty begins earlier and earlier and with it adolescence, now a stage that some researchers suggest begins at 10 and ends at 30. These changes are just part of a larger trend toward more individualized, subculture-specific life trajectories that reveal a need to reform or transform Erikson’s theory. Finally, the Piaget paradigm has faced a great deal of criticism about the firmness of the stage model, the lack of specific life history that the model allows, and the lack of ability to account for regression (that is, the normal return to earlier capacities under states of stress, confusion, and trauma). Piaget and also Kohlberg, tying their work so strongly to cognition, were able to keep their models of intelligence and morality tied to a tight cognitive model. Yet even this created problems; the system held together because cognition tends to be more formal and thus more stage appropriate. But researchers who developed these cognitive stage theories into ideas about faith, self, perspectivetaking, personality, and other strands of development ran into problems, because the stages break down under the weight of more and more content, domains, and styles. Why should they all be governed by one stage, rather than by many, or even by nondevelopmental dimensions interacting with developmental ones? Our response to these shortcomings, as well to the simultaneous power of these and the Vygotskyan paradigms, was to introduce a new developmental model that incorporates attachment, functionalist, and social-cognitive developmental theory applied to risk and resilience, and normal development (Noam & Malti, 2008; Noam, Young, & Jilnina, 2006). What began as a comparative endeavor of over two decades led to the introduction of a new theory. The model assumes that development from middle childhood to middle adolescence can be described as the leaves of a clover, each reflecting a particular kind of development (Noam & Malti, 2008). The four leaves of the clover—action, assertion,

106  MENTORING RELATIONSHIPS belonging, and reflection—are an attempt to create the most simple and efficient set of constructs, basic frames, and needs that are present from the beginning and continue throughout life. Our goal was to seek the minimum number of dimensions necessary to understand the needs and desires of children and adolescents in order to provide them with the right support and learning opportunities that engage and satisfy these needs. What we found through research as well as clinical and school observations is that action, assertion, belonging, and reflection are those four dimensions. We call these dimensions and their interactions the clover model because they do not follow each other sequentially, but are each present at all points of development. The leaves are not distinct entities; rather, they overlap like in a Venn diagram. Every individual exercises aspects of each of these four developmental processes and needs to balance them, but each leaf takes prominence for specific age groups. People move along a continuum, prioritizing the task of one leaf before another, but that priority does not mean that the other dimensions are not applicable. In early adolescence, for example, the belonging dimension becomes essential. But the physical needs of the action leaf, the issues of assertion (e.g., of will and trying to make

an impact), and reflection about self and the world also are active. These establish a new balance with belonging as the preeminent developmental tension. This model preserves a developmental point of view while broadening the scope from a stage-wise progression of sense-making (Piaget) or life tasks (Erikson) or a singular focus on relationships (Bowlby). Body, will, attachment, and cognition are in continuous exchange. They evolve together and apart, maintaining a tension between progression and regression. Understanding this concept can help mentors assess the strengths and risks posed by the particular developmental process facing youth their mentees’ age. The model’s utility lies in its use to understand the balancing act among these four essential elements throughout development. Table 7.1 defines what each of these terms represents. For the present purposes, we suggest mentors be trained to consider the four leaves of the clover—action, assertion, belonging, and reflection—when trying to understand a youth’s experience, plan a mentoring activity, or infer how the mentee may be experiencing a given event. We briefly reveal how each of these leaves reflects an integration of the three theories described previously.

Table 7.1   The Clover Model Chronology of Time

Clover

Development

Resilience

Risks

Early childhood to kindergarten age

Action

Thinking in impulsive terms; behavior defined in terms of consequences; self-focused emotions

Active, spontaneous, curious

Problems with behavior control: impulsivity and attention problems, hyperactivity

Middle childhood to early adolescence

Assertion

Proper behavior defined by what is best for the self; low empathy and trust; thinks and feels in self-centered terms

Leadership qualities, power oriented, boundaried

Externalizing problems: aggressive behavior, violence as revenge

Early to middle adolescence

Belonging

Ability to take others’ perspectives; seeking the approval of others; high empathy and orientation toward others

High interpersonal sensibility, prosocial orientation

Internalizing problems: feelings of depression and hopelessness

Middle to late adolescence

Reflection

Ability to coordinate perspectives of self and other; needing others to define who the self is

High sense of responsibility, reflective

Internalizing problems: feelings of loneliness and isolation

Mentoring Relationships in Developmental Perspective    107 Action. The child in early childhood is often all about action, though this tendency is slightly more applicable to boys than to girls. They need to be active, think better when active, and view the world in terms of concrete consequences of behavior. They are aware of and concerned with their own perspective, primarily, and with actively engaging in the world. Their goal is to learn about the world, experience mastery of different activities, and in doing so, satisfy immediate needs. But time spent in action also helps youth learn about who they are and whom and what they like. Assertion. Exploring their world is paramount for children, and to do this, they assert themselves, their wants and needs, and interests (now more verbally than via action). They can appear oblivious to the needs of others because they don’t yet have the capacity to fully coordinate multiple perspectives. Learning to do so will require asserting themselves and dealing with the reactions of others. Of course, what they anticipate the reactions of others may be will be shaped—for better or worse—by attachment history as well as by their expectations of hostility or support from others, and their cognitive abilities. Knowing a youth may be expecting hostility from others and understanding aggressiveness in terms of striving for assertion may help a mentor to normalize a mentee’s aggressiveness (Malti & Noam, 2009) and to be more optimistic that the mentor can provide a corrective emotional experience. Belonging. With a secure attachment, a youth will explore the world with ease, and if the youth has successfully achieved mastery during his or her industrious explorations of talents, interests, and skills, both of these experiences will dovetail with the emergence of the mutual-inclusive perspectivetaking skills that typically appear during the shift from middle to high school (sometimes earlier for girls). The result is a concern for belonging and a sense of allegiance with like-minded and affirming peers. A sense of belonging becomes all important, such that peers take on magnified importance, sometimes trumping family connectedness as the youth’s primary source of social affirmation. Of course, this can result in an overly conforming approach to peer relationships that limits further self-exploration and can even handicap a youth whose potential could catapult him or her academically or professionally beyond the reach of peers but who is too concerned with the need for belonging to risk exploring beyond the boundaries of peer or even cultural terrain.

Reflection. The teen who is able to take a perspective on his or her friendships, cultural group, or family can begin to reflect on ways in which he or she differs from others, in values, potential, interests, and needs. A deeper degree of identity exploration requires that the full force of mutual-inclusive perspective-taking abilities is applied to self-discovery. Reflection, which is hampered by anxious attachment but helped by the security of a secure attachment style, results in exploration beyond the comfortable boundaries of the collective views of peers, families, and other familiar people. Knowing this, a mentor can help the youth who wants (and now is able) to engage in deeper reflection to consider opportunities the youth has never considered. As should be clear from this brief overview, at each position on the clover, other leaves can take prominence for a brief or extended period or in specific relationships or contexts. And the process of one leaf may be needed to bridge two others. For example, the transition from the belonging to reflection positions may require an altogether new type of assertion to chart new educational territories or career terrain. Normative changes that occur from middle childhood to late adolescence mean that adolescents typically have different needs for belonging, assertion, action, and reflection than children (see Table 7.1 for the chronology of the developmental organizations). From a developmental perspective, it also means that significant relationships need to foster child and adolescent needs and thereby fill a unique function in the youth’s life. In the next section, we provide a brief review of research that illustrates some of the points made earlier, specifically about the importance of the three general theories described previously—attachment, functionalist, and social-cognitive developmental— and that speak to the utility of the clover model as a heuristic tool for supporting mentors. Following this section we consider some practices that seem particularly salient in terms of the developmental processes we have reviewed.

Research In this section, we review studies on child and adolescent mentoring with a focus on research that investigates developmental factors and is driven explicitly by a developmental perspective. For our review, we searched the literature using PsycInfo (CSA interface). Five searches were made, each included mentoring and youth as keywords located in abstracts. Then the following four terms were included as the third abstract keyword in separate

108  MENTORING RELATIONSHIPS searches: attachment, identity development, socialcognitive development or cognitive developmental, and belonging. These searches yielded 16, 17, 5, and 23 citations, respectively. Of these, 15 were dissertations (only 6 of which were focused on youth mentoring), 3 were overview chapters from the first edition of the Handbook of Youth Mentoring, and 2 were books. From this search emerged several school-based mentoring programs and communityand school-based mentoring studies. Attachment Processes in Mentoring We know from attachment research that youth who report abuse histories consequently bring anxiety, ambivalence, and avoidance into new close relationships, and that changes in working models take considerable time. This suggests, first, that youth with abuse histories may be most likely to have poor attachments or relational working models that interfere with the establishment of a trusting relationship with a mentor. Second, if mentoring relationships do leverage change by affecting mentees’ relational working models, such changes likely take considerable time to make. Consistent with these hypotheses, Grossman and Rhodes (2002), when analyzing data from the 1998 Big Brothers Big Sisters impact study, found the largest increases in academic, psychosocial, and behavioral outcomes of youth when mentoring lasted for a year or more. This finding suggests that long-term relationships may be necessary to affect mentees’ attachment-based working models of relationships. They also found that in relationships that ended prematurely, the mentees were significantly more likely to have had histories of abuse in relationships—suggesting those mentees’ working models may have undermined their mentoring experience. Cavell (see Cavell & Henrie, 2010) developed a mentoring program for aggressive children, based on an integrated model of risk and resilience and attachment theory. In the designed prevention program, called PrimeTime, mentors received extensive training in child-directed play skills and effective parenting. Children were randomly assigned to the PrimeTime mentoring program or to a standard mentoring condition intended to be inert because the mentors were trained minimally and multiple mentors sequentially mentored each youth for shorter periods of time (presumably lacking sufficient time to affect attachment-based working models). Children in both conditions showed significant improvement in ratings of aggression at posttreatment, but teacher-rated gains were lost at the 1-year follow-up. However, subsequent analyses revealed

the program effects were moderated by the youth’s interpersonal style, the mentor’s attachment style, and the nature of the interactions in the match: Alfonso and colleagues (2001, October) found that mentors with higher levels of ambivalence seemed less able to contain and manage conflict, reporting greater conflict in their mentoring relationships than did mentors with lower levels of ambivalence. Unexpectedly, the Alfonso et al. study (2001, October) found that the lowest ratings of relationship conflict come from high aggressive children with low ambivalent mentors. This finding suggests that the experience of having a low ambivalent mentor for high aggressive children may lead to a qualitatively different relationship for these children. (Alfonso, 2002, p. 7) Aggressive youth, whose prior experience in relationships may have led them to expect hostility and conflict in relationships, were most positively affected by the mentors whose approach was consistent with a secure attachment. Here we find that the mentors’ attachment style may interact with their mentees’ social information processing approach, suggesting that program staff should consider matching their most securely attached mentors with their more aggressive mentees. Some mentor behaviors may be particularly likely to evoke attachment anxiety, such as not showing up regularly. Empirical evidence indicates that inconsistent mentors make their mid-adolescent mentees feel less secure (Karcher, 2005). A study of 73 fourth and fifth graders in a weekly after-school peer mentoring program showed program participation to have overall positive effects on mentees’ connectedness to parents and school. This finding suggests positive peer mentoring may positively influence mentees’ relational working models (i.e., improving parent-child relationship quality), yet the same study found that mentors’ attendance was inversely related to mentee-reported changes in physical self-esteem. The repeated disappointment of mentees by absent mentors may negatively affect relational working models (e.g., affecting selfperceptions of their attractiveness to others and how much they can and should put their trust in others). These findings are not experimental, however, but rather are correlational, allowing for the possibility that youth with poor attachments (who already felt ugly) pushed away their mentors. The most important lesson from the attachment research is that past relationship histories are predictive of how youth enter mentoring relationships.

Mentoring Relationships in Developmental Perspective    109 Youth with troubled attachment histories sometimes test mentors as a function of their ambivalence toward others. If the mentor fails these tests (e.g., takes rebuffs personally or quits altogether), the failure confirms the youth’s initial lack of trust (Noam & Hermann, 2002). Erikson’s Functionalist Approach and Mentoring Developmentally Most research that draws on Erikson’s approach has focused on his later stages of generativity and wisdom, which are beyond the scope of the mentee age range to which we are applying developmental models. Such studies include experimental research on the Across Ages program (Taylor, 2000; 2006) and studies of mentor characteristics that predict psychological health (Westermeyer, 2004) or persistence as mentors (McAdams & St. Aubin, 1998; Parise & Forret, 2008). Perhaps the only youth mentoring study to look explicitly at developmental differences in the types of support to which youth are most open is by Liang, Spencer, Brogan, and Corral (2008). The authors interviewed 56 middle school, high school, and college students about the mentors in their lives and then analyzed the data qualitatively. Liang et al. (2008) found that trust, fun, and role modeling in mentoring were important across all phases of adolescence, but other mentoring processes varied developmentally. Liang et al. (2008) considered the form of mentor support called empowerment. They found differences between early and late adolescents and the types of support they sought. Consistent with research on adult mentoring, the younger mentees (here, early adolescents) described empowerment in terms of encouragement and verbal support, whereas older adolescents wanted more practical and functional assistance in dealing with key transitions or tasks in their lives. It is as if older adolescents, knowing better who they were and where they were going, wanted help achieving their goals, whereas younger adolescents simply wanted encouragement to explore new interests. Older youth were more sensitive to the issue of mutuality, wanting to be sure their mentoring relationships went two ways. For this reason, they were more interested in shared decision making and a mutual valuing of the relationship. That is, they were less open to a top-down relationship in which the mentor made decisions but the youth was the primary beneficiary of the relationship. This finding may be consistent with the self-focused nature of the developmental task during latency (initiative, industry, identity), and it directly parallels the progression

in cognitive development described earlier, in which mutuality and inclusiveness is how older adolescents make meaning in relationships. Cognitive Development Theories in Mentoring Two studies using randomized control designs illustrate two different ways to apply cognitive developmental theory in program planning and research. This first approach—what was initially called developmental mentoring but was later renamed the Cross-Age Mentoring Program (CAMP) because of the possible confusion with Morrow and Styles’s (1995) use of the term developmental style (see Karcher & Hansen, this volume, Chapter 5)—reflects the use of cognitive developmental theory to structure the mentoring relationship interactions. Structuring the Mentoring Meeting Time Develop­ mentally. Consistent with Ivey’s (1986) “developmental therapy,” in which the therapist tries to move through each level of Piaget’s cognitive development model by using first concrete language and then language requiring formal cognitive operations, CAMP has 4th and 5th graders paired with high-school-aged mentors (Karcher, Davis, & Powell, 2002) progress through the stages of social-cognitive development, specifically Selman’s (1980) stages of perspectivetaking development. Selman’s (1980) social developmental model of perspective-taking parallels Noam’s constructivist approach, in which youth first act out their thoughts (with minimal reflection), then they learn to speak their minds (sharing their subjective perspective) and then to coordinate their own and another’s perspective (by middle school), and finally they work from a group or shared perspective (by high school). Therefore, in every meeting, the mentor-mentee dyads first engage in icebreakers in which communication is physical rather than verbal. Then they share what happened since the last time they met—each sharing his or her own subjective perspective. Then they incorporate the “perspective” of the program staff member who shares with them the activity of the day. Finally, they work together in some way to collaborate—both accommodating their needs or wants to the other’s in service of their evolving relationships. Karcher (2008a) calls this approach walking the mentees “up the developmental ladder” in each meeting. To date, no experimental comparison of outcomes for programs that do and do not have dyads “walk up the developmental ladder” have been conducted. However, in two studies, Karcher and colleagues have reported positive effects of the program on achievement, connectedness to teachers and parents, and

110  MENTORING RELATIONSHIPS self-esteem (Karcher, 2005; Karcher et al., 2002). By contrast, large-scale randomized studies of other peer mentoring programs that have not explicitly used this format have yielded less impressive impacts (see Herrera, Kauh, Cooney, Grossman, & McMaken, 2008). We can only speculate at this point that the use of a social-cognitive developmental progression of activities is what differentiates these more and less effective programs. Social-Cognitive Developmental Abilities as Mode­ rators of Program Impacts. Little research has studied the developmental course of child and adolescent mentoring relationships (for an overview, see Keller, 2005). Although some studies have investigated developmental factors as mediators, moderators, or outcome variables in mentoring research, most of the extant studies used combined child and adolescent mentee samples with far more children than adolescents, which makes it difficult to compare teens and preteens. One school-based mentoring study (Karcher, 2008b) tested chronological age as a moderator of program impacts with Latino youth, and the findings suggest it may be particularly useful to think about developmental differences that take place during middle school and that have an effect on the receptivity and impact of school-based mentors for children and adolescents. To our knowledge, this is the only randomized study to include a sufficient number of elementary-, middle-, and high-school-aged youth to test the differential effects of nondirective youth mentoring in schools. In this study of 540 mostly Latino youth, mentors were trained in friendship development as the main approach with an emphasis on having fun and connecting rather than focusing explicitly on academics. Karcher (2008b) found positive effects of school-based mentoring for elementary-aged boys and negative effects for high-school-aged boys. A similar pattern was present for girls, but the positive effects for elementary-aged girls did not reach statistical significance. For both genders, there was a fourfold increase in the mentors’ focus on problems, grades, and attendance in high school compared to elementary school. This finding raises the question of whether school-based mentoring that is more relational and playful (than curricular, academic, or skills oriented) is better suited to the developmental needs of children than of adolescents or whether the mentees’ age elicits different behaviors from mentors that can undermine the relational, playful approach. The Clover Leaves Regarding developmentally informed research on mentoring in adolescence, some evidence indicates

the need to differentiate between different developmental needs in adolescence. For example, Hamilton and Hamilton (1993) found in a sample of young adolescents that assigned mentors often were discouraged by the failure of their mentees to respond positively to their support offers. Assuming that one of the processes through which mentors influence their mentees is by providing support, the determining factor could be the developmental readiness of the person mentored for specific forms of support. This finding is consistent with those from the previously cited study of Latino children and adolescents (Karcher, 2008b), which suggests mentoring that emphasizes interpersonal connection (belonging) may be better for children and early adolescents, whereas adolescents may desire something more akin to a goal-directed approach, taking a bigger-picture, future-oriented (reflection) orientation. But these goals may need to be collaboratively constructed to better fit the teenage mentees’ ability and desire to use mutual-reciprocal perspective-taking skills. Mentors who suppress the mentees’ exercise of emergent developmental skills in negotiation by unilaterally imposing goals in an effort to solve perceived problems may do more harm than good (Karcher & Nakkula, 2010). It seems likely that children and adolescents generally have different needs for support, but each youth is unique. So the mentor must ask, does my mentee need action, such as through sports, dance, and recreation? Or should I help him or her gain clarity of purpose and the ability to speak up (assertion)? Or could I help most by shoring up her sense of belonging and relationship? None of these approaches has to be mutually exclusive, but defining the mentee’s location on a developmental map may give the relationship a sense of direction. CAMP provides another example of the utility of the clover model (Karcher, 2008a). Viewed from the perspective of the model, when mentor-mentee dyads “walk up the developmental ladder,” they are also hopping across the clover leaves. In each 2-hour meeting, the first activity is an interactive icebreaker (action), followed by a touching-base time when each member of the dyad shares his or her perspective on the prior week (assertion of one’s own perspective), followed by an emphasis on deepening the relationship by taking another’s perspective through a collaborative connectedness activity (belonging). Finally, at the conclusion of the meeting, the mentor and the mentee each reflect on the meeting, the activity, and their relationship using the 3-2-1 activity (reflection; see Karcher, 2012). Karcher asserts (but has not tested) the hypothesis that, by starting with action and moving across the leaves to reflection, the dyad can achieve

Mentoring Relationships in Developmental Perspective    111 higher levels of perspective-coordination and interpersonal connectedness. Mentoring research reveals some of the developmental mediators and moderators of program success. Yet, research that relies explicitly on developmental theory and an integrative developmental approach, which considers key elements of individual development as stated in the clover model (i.e., social cognition, emotion, behavior), remains scarce. Studies that are developmentally informed and use rigorous research designs are even less developed in the field of child and adolescent mentoring. In the field of youth mentoring, research that systematically integrates a developmental perspective is still in its infancy. In addition, youth mentoring programs are often combined with other intervention components, and studies of them often lack methodological rigor, all of which hamper clear interpretations of research findings.

Practice The preceding section examined how developmental theory has or could be used to guide mentoring program design. Research on the Lunch Buddy program reveals that the mentor’s attachment status may interact with mentee’s attributional style, such that programs might be wise to try to use such information to match mentors and mentees. The structure of CAMP (Karcher, 2008a) uses a developmentally sequenced set of activities in each meeting, illustrating another way in which developmental theory might guide program development and practices. In this section, we suggest how other developmental processes described in the preceding section may be brought to program planning and mentor training. Attachment Theory How can attachment theory benefit the mentoring of children and adolescents? From a prevention literature view, mentoring can be a strategy for preventing children from a negative developmental course (risk prevention), a strategy for protecting children who are already on a risk trajectory (protection for at-risk populations), or a strategy for promoting all children’s and adolescents’ social and emotional competencies (universal prevention; see Cavell & Elledge, this volume, Chapter 3). Attachment theory and mentoring research suggest that programs serving youth in the first two groups, those at greater risk, may require special care in the selection, training, and support of mentors, as well as thoughtful matching in terms of the youth’s background.

Erikson’s Functionalist Theory Mentors are often viewed as older and wiser individuals, but also as individuals who expose their mentees to a variety of social and cultural experiences they might not otherwise have (Rhodes, 2002). Regardless of whether the match is confined to the school or is allowed free range to visit museums, parks, or cultural events, the same events can serve different developmental needs; mentors should assess how each interaction can foster development and support initiative, industry, and identity achievements when opportunities to do so arise. Younger children will benefit from more active explorations—such as through game play, tactile children’s museums, and outdoor activities that allow them to explore and direct themselves (initiative stage task). Mentors of children should align the child’s interests with new opportunities for engagement that allow them to explore freely. Youth just a few years older, those in the middle elementary years and beyond, benefit from and developmentally need to help make the decision, to assert and help codirect what the match does together. They will need help in identifying and relating their own interests to activities and opportunities for the match to engage in together, and they will benefit from repeated opportunities to develop mastery at new and interesting tasks (industry stage task). Although mentors do need to “get something” from mentoring, they don’t always need to have fun themselves. It is best if the mentor can try to get satisfaction and pride out of providing the mentee with meaningful opportunities for skill development and mastery. Belonging begets identity. Youth in middle and high school need a mentor to validate their needs for belonging and connection. Given their need for peer-based belongingness, peers’ perceptions can shape how the youth views the mentor. In late elementary school and early high school, how mentees believe their peers view the mentor-mentee dyad is critical to program success, but peer processes may operate differently at each age. A mentor’s presence may be a sign of specialness in elementary school (as seen in the Lunch Buddy program) or may signal a remedial intervention that is embarrassing, such as may be the case for high-risk youth whose mentors visit them at school during lunch (e.g., Karcher, 2008b). Mentors should try to recognize that, despite all of the material and emotional costs of “being a mentor,” mentees may value the perspectives of peers more frequently or highly than they do those of their mentors. It is this need for belonging that informs identity development. (This need is described more clearly by

112  MENTORING RELATIONSHIPS Joan than Erik Erikson [J. M. Erikson, 1988; also see Karcher & Benne, 2008]). It is out of the many experiences of adolescent belonging (with family, at school, but also among peers) that the adolescent mentee’s emergent sense of identity develops. Mentors who sympathize with (rather than feel rejected because of) the youth’s needs for belonging in peer groups are the mentors whose empathy and understanding make them invaluable to their mentees.

Figure 7.1   The Clover Model

Belonging Assertion

Social-Cognitive Theory One of the most important lessons that socialcognitive developmental theory presents to mentors is the awareness that between elementary and high school, cognitive developments occur that allow self-reflection and an awareness (or guesstimate) of what others are thinking about. This becomes a preoccupation during middle school, such that, as mentioned earlier, mentors who visit their mentees at school (in the class or in the lunchroom) may be more embarrassing than empowering. Peer processes that contribute positively to the impact of one program in elementary school may be undermining at later grades. Mentors should be taught to always try to view their match and their relationship from the youth’s perspective, knowing that the “group” or mutual perspective that emerges in middle school may work against them. Mentors can feel they are giving more than they are getting, because in truth they do give more than the child does. But some mentors may be overly frustrated by the seeming unwillingness of the mentee to take the mentor’s point of view—such as when a mentor feels taken advantage of when a youth asks for money, invites a friend to come along, or is absent when the mentor comes to meet the child. Mentors need to understand that, for elementaryaged youth, this apparent unwillingness may be a normal cognitive limitation that should not be taken personally, but rather expected and planned for. The inability of the elementary- or middle-schoolaged child to coordinate perspectives when solving problems—whether with a mentor, parents, peers, or siblings—also should be expected. In fact, the reason many youth need a mentor is to provide a safe context in which the youth can practice cognitive skills that are not his or hers today but with support and encouragement can be tomorrow— developmentally speaking (Vygotsky, 1978). Clover Model Mentors can use frameworks such as the clover model to better understand the developmental

Reflection Action

processes that may be at play in a given struggle facing their mentee, or that may interfere with establishment of the mentoring relationship, or that may explain the mentor’s own weaknesses and strengths. For example, is the mentor very reflective, but not very active in forging the relationship? This has implications for how the mentor will establish and maintain the relationship, and could indicate that the mentor may work better with older than younger mentees. As a second example, does the mentor model assertiveness and in other ways cull the mentee’s opinions and views of matters to foster communication? Third, does the mentor have a strong sense of belonging and connection in his or her own life? If so, this may help the mentor to coach the mentee in cultivating connections. If not, the mentor may not appreciate the mentee’s need for belonging or the mentor may inappropriately use the mentoring relationship as a way of getting his or her own belonging or social needs met. Table 7.2 provides suggestions for helping mentors to apply the clover model. A program can establish developmentally specific outcomes or program structures that make leaf-specific support material or activities available to mentors. Mentors can be trained to recognize and assess the developmental differences in each of the primary theories—attachment, functionalist, and social-cognitive development—so as to understand the clover-specific needs of mentees and thereby to

Mentoring Relationships in Developmental Perspective    113 Table 7.2   Checklist for Practitioners Based on Attachment Theory Consider using some assessment of adult attachment to determine who are the most securely attached mentors. Give aggressive youth; those with abuse, loss, abandonment histories; and youth with known problems trusting adults the more securely attached mentors. Help mentors size up the degree of trust and safety their mentees demonstrate (or report) in their relationships with adults and youth in their lives. Based on Erikson’s Functionalist Theory Train mentors to understand the processes involved in the developmental tasks of initiative, industry, and identity. Before and after mentors begin relationships with youth, have program staff help mentors make guesstimates about their mentees’ competencies and interest in exploration (initiative), industry (engaging in activities and moving toward skill mastery), and identity (understanding who they are, what they like and value, across contexts and relationships). Help mentors brainstorm topics of conversation, activities for field trips, and games to play together that reflect developmental tasks facing their mentees. Based on Social-Cognitive Developmental Theory Consider the self-consciousness that results from reciprocal and mutual perspective-taking skills, and determine whether there are risks posed by the mentoring context for the youth to feel stigmatized by (rather than a beneficiary of) a mentoring relationship. Send mentors into a context rich with the mentee’s peers (e.g., school class or lunchroom) for younger mentees to capitalize on relationship-enhancing peer processes, but not for older mentees to avoid triggering self-conscious feelings. Train mentors to understand the difficulty young children have taking others’ perspectives into account, and how this can look selfish, aggressive, immature, or inconsiderate (even when it is just normative). Based on the Clover Model Provide training to mentors in how to look for signs of the presence or absence of mentees’ drive for action, assertion, belonging, and reflection. In support calls and meetings, help mentors look for prosocial, relationship-enhancing, and fun opportunities to meet the mentees’ needs for action, assertion, belonging, and reflection (with emphasis on leaves related to the mentee’s age).

consider developmental interventions they could incorporate into mentoring (see Malti, Liu, & Noam, 2010). Such training helps mentors to understand their mentees’ developmental skills and limitations in self-control; clearly asserting needs; establishing belonging in familiar and unfamiliar contexts alike; and reflecting on their roles in problems or otherwise “taking a perspective” on themselves, their families, or their lives. Mentors may then be able to praise and affirm their child or adolescent mentees’ strengths and to focus their efforts

to help youth cultivate skills that may be lacking or inadequately used to achieve desired outcomes.

Conclusion To understand better how mentoring works, program developers and researchers need to investigate the influences of developmental characteristics (i.e., social cognitions, emotions, behavior) based on theoretical assumptions about the relations between

114  MENTORING RELATIONSHIPS these developmental characteristics, relationship quality, and outcomes of mentoring programs. Investigations of youth mentoring relationships and programs should be tailored to address concerns associated with developmental issues.

References Alfonso, L. (2002). Mentoring relationships with aggressive children: An investigation of putative correlates and putative outcomes. Unpublished dissertation, Texas A&M University, College Station, TX. Benard, B., & Slade, S. (2009). Listening to students: Moving from resilience research to youth development practice and school connectedness. In R. Gilman, E. Huebner, and M. Furlong, (Eds.), Handbook of positive psychology in schools (pp. 2353–2369). New York: Routledge. Bowlby, J. (1969). Attachment. New York: Basic Books. Cavell, T. A., & Henrie, J. L. (2010). Deconstructing serendipity: Focus, purpose, and authorship in lunch buddy mentoring. New Directions for Youth Development, 2010 126, 107–121. Crick, N., & Dodge, K. A. (1994). A review and reformulation of social information processing mechanisms in children’s social adjustment. Psychological Bulletin, 115, 74–101. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Erikson, E. H. (1950). Childhood and society. New York: Norton. Erikson, E. (1968). Identity, youth, and crisis. New York: Norton. Erikson, J. M. (1988). Wisdom and the senses: The way of creativity. New York: Norton. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30(2), 199–219. Hamilton, M. A., & Hamilton, S. R. (1993). Toward a youth apprenticeship system: A progress report from the Youth Apprenticeship Demonstration Project in Broome County, New York. Ithaca, NY: Cornell University Press. Herrera, C., Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, J. (2008). High school students as mentors: Findings from the Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Ivey, A. E. (1986). Developmental therapy. San Francisco: Jossey-Bass. Karcher, M. J. (2005). The effects of school-based developmental mentoring and mentors’ attendance on mentees’ self-esteem, behavior, and connectedness. Psychology in the Schools, 42, 65–78.

Karcher, M. J. (2008a). The Cross-Age Mentoring Program (CAMP): A developmental intervention for promoting students’ connectedness across grade levels. Professional School Counseling, 12(2), 137. Karcher, M. J. (2008b). The Study of Mentoring in the Learning Environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9(2), 99–113. Karcher, M. J. (2012). The Cross-Age Mentoring Program (CAMP) for Children with Adolescent Mentors: Program manual. San Antonio, TX: Developmental Press. Karcher, M. J., & Benne, K. (2008). Erik and Joan Erikson’s approach to human development in counseling. In K. Kraus (Ed.), Lifespan development in action: A case study approach for counseling professionals (pp. 199–238). Stamford, CT: Lahaska Press/ Cengage Learning. Karcher, M. J., Davis, C., & Powell, B. (2002). The effects of developmental mentoring on connectedness and academic achievement. School Community Journal, 12(2), 35–50. Karcher, M. J., & Spencer, R. (2010). The Study of Mentoring in the Learning Environment (SMILE): Multilevel modeling of setting-level program staff contributions to school-based mentoring program effectiveness. Unpublished manuscript. University of Texas at San Antonio. San Antonio. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher, (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Kohlberg, L. (1994). Stage and sequence: The cognitivedevelopmental approach to socialization. In B. Puka (Ed.), Defining perspectives in moral development. Moral development: A compendium (Vol. 1, pp. 1–134). New York: Garland Publishing. Liang, B., Spencer, R., Brogan, D., & Corral, M. (2008). Mentoring relationships from early adolescence through emerging adulthood: A qualitative analysis. Journal of Vocational Behavior, 72(2), 168–182. Loevinger, J. (1976). Ego development: Conceptions and theories. San Francisco: Jossey-Bass. Malti, T., Liu, C. H. J., & Noam, G. G. (2010). Holistic assessment in school-based, developmental prevention. Journal of Prevention and Intervention in the Community, 38, 244–259. Malti, T., & Noam, G. G. (2009). A developmental approach to the prevention of adolescent’s aggressive behavior and the promotion of resilience. European Journal of Developmental Science, 3(3), 235–246. McAdams, D. P., & de St. Aubin, E. (1998). Generativity and adult development: How and why we care for the next generation. Washington, DC: American Psychological Association. MENTOR. (2003). Elements of effective practice (2nd ed.). Alexandria, VA: Author. Morrow, K. V., & Styles, M. B. (1995). Building relationships with youth in program settings: A study of Big Brothers/Big Sisters. Philadelphia: Public/Private Ventures.

Mentoring Relationships in Developmental Perspective    115 Noam, G. G. (1988). A constructivist approach to developmental psychopathology. In E. Nannis & P. Cowan (Eds.), New directions for child development (pp. 91–122). San Francisco: Jossey-Bass. Noam, G. (1999). The psychology of belonging: Reformulating adolescent development. In A. H. Esman, L. T. Flaherty, & H. A. Horowitz (Eds.), Annals of the American Society of Adolescent Psychiatry (Adolescent Psychiatry: Developmental and Clinical Studies). Hillsdale, NJ: The Analytic Press. Noam, G. G., & Hermann, C. A. (2002). Where education and mental health meet: Developmental prevention and early intervention in schools. Development and Psychopathology, 14, 861–875. Noam, G., Kohlberg, L., & Snarey, J. (1983). Steps towards a model of the self. In B. Lee & G. Noam (Eds.), Developmental approaches to the self (pp. 59–142). New York: Plenum Press. Noam, G. G., & Malti, T. (2008). Responding to the crisis: RALLY’s developmental and relational approach. New Directions for Youth Development, 120, 31–55. Noam, G. G., Young, C. H., & Jilnina, J. (2006). Social cognition, psychological symptoms, and mental health: The model, evidence, and contribution to ego development. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathology, Vol. 1: Theory and method (pp. 750–794). Hoboken, NJ: Wiley. Parise, M. R., & Forret, M. L. (2008). Formal mentoring programs: The relationship of program design and support to mentors’ perceptions of benefits and costs. Journal of Vocational Behavior, 72(2), 225–240. Piaget, J. (1954). The construction of reality in the child. New York: Basic Books. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Scheithauer, H., Malti, T., & Noam, G. G. (Eds.). (2009). Developmentally appropriate preventions of aggression: Developmental science as an integrative framework. European Journal of Developmental Science, 3(3), 215–217.

Selman, R. L. (1980). The growth of interpersonal understanding: Developmental and clinical analyses. New York: Academic Press. Selman, R. L., Beardslee, W., Schultz, L. H., Krupa, M., & Podorefsky, D. (1986). Assessing adolescent interpersonal negotiation strategies: Toward the integration of structural and functional models. Developmental Psychology, 22(4), 450–459. Selman, R. L., & Schultz, L. H. (1990). Making a friend in youth: Developmental theory and pair therapy. Chicago: University of Chicago Press. Snarey, J., Kohlberg, L., & Noam, G. G. (1983). Ego development in perspective: Structural stage, functional phase, and cultural age-period models. Developmental Review, 3, 303–338. Sroufe, L. A. (1997). Psychopathology as an outcome of development. Development and Psychopathology, 9, 251–268. Steinberg, L., & Morris, A. S. (2001). Adolescent development. In S. T. Fiske, D. L. Schacter, & C. ZahnWaxler (Eds.), Annual Review of Psychology (Vol. 52, pp. 83–110). Palo Alto, CA: Annual Reviews. Taylor, A. S. (2000). Mentoring across generations: Partnerships for positive youth development. New York: Springer. Taylor, A. S. (2006). Generativity and adult development: Implications for mobilizing volunteers in support of youth. In E. G. Clary & J. E. Rhodes (Eds.), Mobilizing adults for positive youth development: Strategies for closing the gap between beliefs and behaviors. (pp. 83–100). New York: Springer. Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes (M. Cole, JohnSteiner, F., Scribner, S., Souberman, E., Trans.). Cambridge, MA: Harvard University Press. Werner, E. E. (1986). The concept of risk from a developmental perspective. Advances in Special Education, 5, 1–23. Westermeyer, J. F. (2004). Predictors and characteristics of Erikson’s life cycle model among men: A 32-year longitudinal study. International Journal of Aging & Human Development, 58(1), 29–48.

8  A BIOLOGICAL PERSPECTIVE ON POSITIVE ADULT-YOUTH RELATIONSHIPS Amber L. Allison and Elizabeth A. Shirtcliff

Introduction This chapter considers the topic of youth mentoring from a biological perspective. We describe changes that may occur within the body in the context of youth mentoring based on broader research literature that describes a reciprocal interaction between the environment, biology, and behavior. Although often overlooked, there is an implicit understanding that environmental factors can influence how we behave and how our bodies work. For example, when the environment does not provide food, we feel hungry. Biologists and physiologists describe a multitude of changes to the body that instantiate feelings of hunger as well as subsequent changes in feelings of fullness and pleasure when the environment later provides food. Alternatively, our behavior can influence how our bodies work; when children run fast during recess, they can feel their hearts race and blood pound. These physiological changes are visible signs that our bodies are responsive to the environment and our behavior. Within the field of biopsychology, there is growing appreciation and mounting empirical evidence that the connection between our environment, biology, and behavior extends deeper. Many environmental factors, some of which might be surprising, affect our bodies. For example, a multitude of changes occur in our bodies when we hug someone we like (Light, Grewen, & Amico, 2005). Even if we are not consciously aware of these biological changes, they can nevertheless influence how we interact with and engage our environment in dramatic ways. Rather than taking a reductionistic perspective, here we view biological mechanisms as another

way of appreciating the importance of environmental forces. One of the most powerful historical examples of the influence of our environment on psychobiological processes dates back to 1897 when Dr. Floyd Crandall described a terrible phenomenon: infants and children were dying after being hospitalized for an extended period of time. Surprisingly, they weren’t dying of any illness or physical problem. By the 1930s, physicians were familiar with this mysterious disease that caused delayed physical development, disruption of cognitive development, and eventually death. It wasn’t until the 1940s that Rene Spitz identified the medical cause: a lack of appropriate care and nurturing from a primary caregiver (Spitz, 1945). Spitz, together with other researchers such as Harry Harlow, revealed that we need support, social contact, nurturance, and love as much as we require food, shelter, and sleep. This provocative area of research pointed a finger at the importance of an attentive caregiver. Recently, the focus of this research has shifted to how a lack of love could be lethal, complementing the psychological approach by focusing on key components of biological systems that respond to social contact and affiliation. The vantage point emphasized in this chapter focuses on a bidirectional relationship between environmental factors and biology such that environments can “get under the skin.” This implies there is a pathway by which the things we do or the places we go are capable of altering who we are and how our bodies work. This perspective becomes especially important if we consider how this interplay between biology and environmental forces plays out across a young person’s development. 117

118  MENTORING RELATIONSHIPS This reciprocal interaction may be especially important to consider within the social context of youth mentoring. Important psychobiological changes may occur when mentors and youth interact, and these changes may serve as an underlying mechanism by which individuals experience and are influenced by social contact and affiliation. These changes are not visible nor are the individuals involved likely to be conscious of them. Nevertheless, they may reflect and influence the way the mentoring relationship is expressed. Investigation of biological mechanisms involved with youth mentoring should not be viewed as obviating the need for continued high-quality social or psychological research in this area; biological mechanisms are only interpretable within the psychosocial context in which that biology is expressed. This chapter explores the relationship between biology and mentoring ties. We begin by considering current broad-level theoretical models and empirical research relating environments, especially relationships, and biology. To aid in the interpretation of some of this information about biological processes, Table 8.1 defines important terms mentioned in the chapter. We also review the limited extant research that bears directly on biological processes in youth mentoring and highlight key directions for future research. In addition, we highlight potential practical implications of a biological perspective for youth mentoring practice and policy. Although the efficacy of mentoring

typically has been quantified by an individual’s status in domains such as education/work, mental health, and problem behavior (Rhodes & DuBois, 2006), the short- and long-term benefits attributable to biological processes associated with mentoring have been largely ignored. We intend this chapter to provide a compelling argument for the importance of considering how biology may influence and be reflected in mentoring interactions.

Theory How Might Youth Mentoring Get Under the Skin? The Psychobiological Approach The psychobiological approach is a broadlevel theory for understanding psychological phenomenon from a biological perspective (Gottlieb, Wahlsten, & Lickliter, 1998). This approach naturally extends from an evolutionary psychology perspective (i.e., the study of the functional and adaptive nature of traits based on the theory of natural selection) because it emphasizes that these biological processes have evolved to meet the demands of a constantly changing environment. A biological or an evolutionary foundation is thought to be expressed uniquely across individuals, and differently depending on a range of contextual factors. As such, a basic tenet of the psychobiological approach is that few psychobiological processes are expected

Table 8.1   Terms and Definitions Psychobiological approach

A theoretical framework relating environmental influences on individual development by considering how context interacts with biology

Homeostasis

The stability of physiological processes (i.e., optimal functioning)

Allostasis

An adaptive mechanism that allows an individual to adjust physiological processes in the face of a stressor

Cortisol

A hormone released whenever a stressor is perceived; social stressors result in the strongest cortisol response

Oxytocin

A hormone associated with social affiliation and bonding

Allostatic load

The wear and tear the body experiences as a result of chronic activation of allostatic mechanisms

Biological sensitivity to context (BSC)

The idea that individuals who are biologically reactive are most likely to be negatively affected by adverse environments and most likely to flourish in supportive environments

Attunement

The ability of two individuals’ physiological processes to be in sync

A Biological Perspective on Positive Adult-Youth Relationships    119 to be tied specifically to a single behavior or social interaction. For example, research regarding affiliative behavior of animals indicates that maternal behaviors (i.e., lactation) served as a foundation for developing social bonds that provide a secure environment for maternal behavior and, eventually, positive social interactions (Carter, Lederhendler, & Kirkpatrick, 1997). Carter and colleagues (1997) argued that the neural circuitry that originally developed to allow mothers to provide sustenance to their offspring was then co-opted to enable them to provide nurturance to and bonding with their children as well. Furthermore, this same neural circuitry allows children to bond with their parents (Uvnas-Moberg, 1999), sexual partners to feel bonded in romantic relationships (Bartels & Zeki, 2003), fathers to reciprocally bond with their children (Geary & Flinn, 2002), and friends to feel connected to each other (Taylor et al., 2000). It appears that we do not develop new biological processes every time we encounter a novel situation or environment because such changes are costly and time consuming, and they require many generations. Rather, psychobiological processes are similar across many types of social interactions, and a reasonable assumption is that the biological underpinnings associated with other social relationships are probably generally the same as those involved in mentoring relationships. The psychobiological approach has three main features: (1) an individual’s biological systems reciprocally interact with the individual’s environment, (2) this interaction is mediated by the individual’s cognitions and behavior, and (3) this interaction occurs across development. Based on these principles, the psychobiological approach emerges as an important overarching theoretical perspective when considering how environments get under the skin to influence developmental trajectories. As such, the psychobiological theory’s main features can be applied to youth mentoring. First, the bond that forms between the mentor and the youth is directly related to the biological processes influenced by the relationship. Second, the interaction is mediated by the individuals’ cognitions and behavior. The greatest biological changes are expected to occur whenever the youth perceives similarities with the mentor, and is given the opportunity to participate in a variety of mutually enjoyed activities. Third, biology and environments interact across development, especially during childhood and adolescence. Biological changes may appear small or minor, but they are expected to grow over time; potential benefits of youth mentoring may be most apparent as the mentoring relationship becomes an established component

of the child’s life. Childhood and adolescence have been identified empirically as sensitive periods, whereby the environment is more likely to affect biological processes (Shirtcliff, 2005). An important implication of this finding is that targeted changes or improvements in a child’s environment can be expected to change the child’s biology as well. What Are Some of the Basic Biological Processes Implicated During Youth Mentoring? Specific biological systems and processes are commonly considered environmentally responsive. The ability of the body to maintain functioning in the face of environmental changes has been appreciated since the French physiologist Claude Bernard (1813–1878) described the concept of the milieu intérieur, or internal body states constantly interacting with external forces (Bernard, 1957). Walter Canon (1871–1945) built on this notion by focusing on the consequential effects of context on biological systems required for homeostasis, such as pH, body temperature, and (especially relevant here) hormones (Canon, 1929). Homeostasis refers to physiological stability required for optimal functioning (Lupien et al., 2006). We feel our best whenever we preserve our body’s most favorable range of physiological arousal. Individuals interact behaviorally with the environment in attempts to maintain homeostasis, such as by taking off a sweater whenever body temperature rises or ingesting food when glucose levels are low. It is not realistic for an organism to remain at homeostatic equilibrium indefinitely due to the need to cope with various changes in internal and external environmental conditions. This is apparent if we think about children’s constant movement from one activity to another. Although the child is always himself or herself, it is not realistic for him or her to be in the same internal state while taking a test at school, relaxing at home, or running actively at soccer. Rather, it makes intuitive sense that the optimal internal state for that child may be different in the classroom, bedroom, or playground. The ability of physiological processes to adapt to different contexts is termed allostasis. Allostasis is a shortterm coping mechanism—activated in the face of salient events—that shifts the physiological systems outside of the normal homeostatic range. Allostasis demonstrates how physiological activity changes depending on environmental factors and enables the individual to remain stable across a wide variety of social contexts. Without allostasis, individuals would not be able to adequately manage environmental change.

120  MENTORING RELATIONSHIPS Hormonal processes have been identified as forerunners of allostasis and the ability of our body to reflect environmental forces. Hormones provide insight into the match of our biology with our social context (De Kloet, 2004). Hormones are remarkably responsive to the environment, constantly changing in response to our physical, social, and emotional worlds, and as such, they are frequently targeted in research on allostasis. The glands that secrete hormones adjust levels to match the demand of a particular situation. When we are in a socially salient environment or interaction, our body can increase hormones that promote affiliation, calmness, and reward. Similarly, when we are in a novel environment, our body can increase levels of hormones to facilitate alertness and memory. This adaptation allows us to explore the new environment for potential threats and remember those threats the next time we encounter that environment. The hormones most implicated in social situations include oxytocin and cortisol. Both hormones are socially responsive, but the individual’s past experience and perception of that interaction influences how much hormone is released. Oxytocin is generally released when bonding or affiliation is expected of the interaction (UvnasMoberg, 1998). Cortisol is generally released when a social threat is interpreted (Dickerson & Kemeny, 2004) or whenever the social information is particularly salient (Booth, Granger, & Shirtcliff, 2008). Oxytocin: How Does Affiliation Get Under the Skin? Oxytocin is tied to feeling connected to another person, as well as to the behavioral expressions of that connection, and is conceptualized as a “feelgood” hormone in social contexts (Uvnas-Moberg, 1998). Oxytocin is produced in the supraoptic and paraventricular nuclei of the hypothalamus and is released by the posterior pituitary, whereby it diffuses to several brain regions (Pfaff, Arnold, Etgen, Fahrbach, & Rubin, 2002). Oxytocin was first identified due to its release during breastfeeding, as it promotes lactation, maternal behavior, and other giving interactions in both animals and humans (Eriksson, Lundeberg, & Uvnas-Moberg, 1996). Breastfeeding women become more social and calm. Oxytocin is positively correlated with this increase in sociability and calmness (UvnasMoberg, Widstrom, Nissen, & Bjorvell, 1990). Because new biological processes are not established each time we encounter a new environment or relationship, previously existing processes are used during these new contexts. Oxytocin release has expanded, throughout the course of evolution,

beyond the mother-child bonding context and is now understood to be a physiological consequence of positive social interactions in general (Insel, 1992). Social bonds and support are associated with oxytocin release and long-term reductions in blood pressure and heart rate (Petersson, Alster, Lundeberg, & Uvnas-Moberg, 1996). A good marriage is associated with elevated oxytocin and reduced risk of several diseases (Grewen, Girdler, Amico, & Light, 2005). Oxytocin receptors are found in brain regions associated with reward, suggesting that oxytocin release during a positive social interaction is inherently reinforcing (Insel, 1992). Thus, oxytocin naturally released during a mentoring session has the potential to facilitate the creation of a bond between the individuals, while also instigating feelings of calmness; this bond and subsequent antistress state may contribute to the lasting effects of oxytocin, including reduced risk of diseases involving the cardiovascular system in both partners (Petersson et al., 1996). Cortisol: Why Is It Not Just a Stress Hormone? As opposed to oxytocin, which is associated with bonding and social affiliation, the role of cortisol in social interactions is complicated by the fact that cortisol can help us manage environmental input, particularly challenging social contexts, but it also can exert deleterious consequences if that social context overwhelms the individual’s resources or if the social feedback is threatening or fear-inducing. Cortisol has earned a negative reputation as a stress hormone, and early theories implied that high cortisol was maladaptive because it signified greater stress. The theory of allostasis, however, emphasizes that high cortisol may be adaptive depending on the context in which it is released. As explained below, the hypothalamic-pituitary-adrenal (HPA) axis, a component of an individual’s stress response system, factors significantly in our understanding of the role of cortisol in mediating an individual’s responses to different types of stressful and social situations. Basal Cortisol. Basal cortisol refers to the relatively stable levels of cortisol present naturally throughout the day. Cortisol is constantly released throughout the day, following a distinct pattern, with levels expected to be high in the morning, peaking within 30 minutes of awakening, and then declining steadily throughout the day. This rhythm provides the resources necessary to cope with moment-tomoment changes in the environment. High basal cortisol is associated with social competence, popularity, and social affiliation in school-aged children

A Biological Perspective on Positive Adult-Youth Relationships    121 (Tennes, Kreye, Avitable, & Wells, 1986), suggesting that higher basal cortisol levels may afford developing individuals the capacity to actively engage their social environments (Del Guidice, Ellis, & Shirtcliff, in press). Similarly, Shirtcliff et al. (2009) theorized that higher basal cortisol is associated with the ability to activate empathyrelated neurocircuitry during social interactions, whereas maintaining low cortisol during such interactions is associated with antisocial behavior. Reduced HPA axis activity contributes to an inability to respond to one’s own emotional or stressful experiences as well as those experienced by another. This evidence does not support the notion that high cortisol is maladaptive, but rather emphasizes that context-appropriate increases in cortisol allow the individual to appropriately interpret experiences salient to the self and others. Alternatively, deviation from context-appropriate cortisol levels or variability can indicate dysregulation (Siever & Davis, 1985) and can have implications for an individual’s ability to cope with stress as well as the individual’s overall health. Prolonged overactivity of the HPA axis as a consequence of chronic stress exposure may paradoxically result in chronically low cortisol throughout the day (Miller, Chen, & Zhou, 2007). For example, hypocortisolism is observed in adults with chronic fatigue syndrome or posttraumatic stress disorder and in some children who are victims of substantiated child abuse or neglect (Fries, Shirtcliff, & Pollak, 2008). On the other hand, elevated cortisol also can sometimes indicate dysregulation. Hypercortisolism has been associated with mental health symptom severity in adolescents, particularly during life transitions (Shirtcliff & Essex, 2008). Thus, in line with the theory of allostasis, successful and healthy adaptation to stressors requires context-appropriate cortisol levels, not just lower or higher values. Reactive Cortisol. As noted above, the HPA axis is a component of an individual’s stress response system. Stressors involving uncontrollability, unpredictability, and threats to the social self stimulate the strongest HPA reactivity in children (Gunnar, Talge, & Herrera, 2009) and in adults (Dickerson & Kemeny, 2004). Across a wide range of contexts spanning across laboratory and naturalistic settings, this research reveals that social forces are powerful (Gunnar & Vazquez, 2006). When a social situation is assessed to involve a threat to the individual, emotion-related neural circuitry within the brain activates the secretion of corticotropin-releasing hormone (CRH) from the hypothalamus. This hormone then promotes the anterior pituitary to secrete

adrenocorticotropin-releasing hormone (ACTH). ACTH travels throughout the entire blood stream and then acts on receptors in the adrenal cortex to stimulate the release of the hormone cortisol into the bloodstream (Dickerson & Kemeny, 2004). The physical effects of a surge of cortisol include adjusting heart rate, replenishing food reserves, trafficking immune cells, and promoting memory for emotional stimuli (McEwen & Wingfield, 2003). Psychological effects include increased memory for emotionally salient stimuli (Buchanan & Lovallo, 2001); moderately high cortisol in anticipation of a social encounter may facilitate learning and memory for that event and other preparatory activities (Booth et al., 2008). Release of cortisol is one mechanism the body uses as an individual attempts to cope with an environmental change and return to homeostasis. The stressor does not need to occur for a stress response to be initiated; cortisol levels may increase in anticipation of a social event or challenge (Klimes-Dougan, Hastings, Granger, Usher, & Zahn-Waxler, 2001). The anticipatory effect may be especially strong when a child anticipates or perceives negative judgment by another (Lewis & Ramsay, 2002). If these anticipations or perceptions are frequent, overactivation of the HPA axis and subsequent negative health outcomes are possible. HPA Axis Buffers. Supportiveness from adults can alter cortisol responses in positive ways (Gunnar et al., 2009). Specifically, research indicates that the presence of a supportive adult can either (a) alter the child’s cognitive appraisal of the environment as a stressor and give him or her strength to cope with the event without triggering a stress response (Taylor et al., 2008) or (b) alter a child’s peripheral biology such that his or her stress response system is less likely to be inappropriately activated (Liberzon et al., 2007). Social relationships can modify biology through cognitive-emotional interpretations of stimuli, which then influences the release of cortisol (Seeman, 2001). Research reveals that the perceived accessibility of a stable social support network helps to regulate the release of cortisol (Taylor et al., 2010) and can exert a positive influence on an individual’s acute thoughts, feelings, and behavior as well as long-term health status (Roy, Steptoe, & Kirschbaum, 1998). Additionally, effective early interventions can prevent HPA dysregulation associated with exposure to stress (Dozier et al., 2006). Brotman et al. (2007) found that, in children at risk for antisocial behavior (i.e., little or no reactivity to peer social challenge), a psychosocial intervention could normalize cortisol reactivity (i.e., produce an anticipatory increase in cortisol) and increase social

122  MENTORING RELATIONSHIPS competency. Similarly, foster children with a history of early life adversity typically show atypical cortisol levels throughout the day; after a familybased intervention, however, these children exhibited cortisol activity that was comparable to an agematched control group (Fisher, Stoolmiller, Gunnar, & Burraston, 2007). Unfortunately, not every social relationship is capable of buffering HPA responsivity. Research has shown that there are notable differences in the way children perceive and biologically respond to supportive relationships depending on the child’s previous life stress. Children who were raised initially in Russian and Romanian orphanages and later adopted continue to show emotion-regulation and attachment difficulties with their caregivers (Fries & Pollak, 2004). For these children, interacting with their caregivers stimulated a stress response similar to the response produced by interacting with a stranger (Johnson, 2000); they had lower oxytocin after interacting with their caregiver, and those with the lowest oxytocin had the highest stress response (Shirtcliff & Ruttle, in press). In contrast, children in the control group (i.e., those being raised in the families into which they were born) perceived interacting with a stranger as a social stressor whereas interacting with a parent buffered a stress response and increased oxytocin levels (Fries et al., 2008; Fries, Ziegler, Kurian, Jacoris, & Pollak, 2005). In fact, Shirtcliff and Ruttle (in press) report that those children who had the highest oxytocin levels also demonstrated the greatest HPA stress buffering, suggestive of a reciprocal connection between these biological processes. For biological processes to instantiate positive social relationships and support, the interaction must be perceived as stable and nurturing. Individual differences in reactivity may also contribute to the range of outcomes attributed to youth mentoring relationships. The theoretical notion of biological sensitivity to context (BSC; Boyce and Ellis, 2005) postulates that individuals who are highly reactive (“orchid children”) are more likely to be negatively affected by high-stress environments but also are more likely to flourish in low-stress or protective environments. Low reactivity individuals (“dandelion children”), on the other hand, are expected to develop relatively similarly across a range of environments, appearing buffered from deleterious consequences of stress but also appearing relatively immune to some positive aspects of supportive environments, as well. The relevance of these distinctions to a mentoring context is broadly consistent with research suggesting that there are notable individual differences in who benefits from the supportive and nurturing relationship that a mentor can

provide (Rhodes & DuBois, 2006). Following the BSC theory, we would predict that those children who benefit the most from youth mentoring are children who are biologically reactive and sensitive to environmental cues. These “orchid children” are more likely to be shaped by their environments, for better and for worse. Physiological Attunement: How Might Oxytocin and Cortisol Be Tied to Youth Mentoring? Physiological attunement occurs when a perceived connection exists between individuals and, under the skin, a reciprocal synchronization of physiological processes (Sethre-Hofstad, Stansbury, & Rice, 2002). Research involving mentoring indicates that, “at the most basic level, a necessary condition for an effective mentoring relationship is that the two people involved feel connected” (Rhodes & DuBois, 2006) and that the bond between the individuals is related directly to the effectiveness of the mentoring relationship in promoting desired outcomes. The subjective feeling of mutual closeness (i.e., feeling connected or bonded) between the youth and the mentor implies objective biological changes, namely hormone release, that occur as a result of the social contact. In the field of psychobiology, this process of feeling connected is evidence that environmental input influences processes under the skin. Attunement research provides evidence for biological cross-talk between individuals. SethreHofstad et al. (2002) found attunement in levels of cortisol in mother-infant pairs when the infant was exposed to a novel and challenging task. Mothers and infants matched in their fluctuations of cortisol throughout the task; however, the attunement finding was dependent on high maternal sensitivity. This testament fits into the psychobiological framework such that a cognitive and behavioral trait (maternal sensitivity) mediated the influence of an environmental condition (challenging task) on biological processes (stress responsiveness). Moreover, research also indicates that therapists (i.e., psychologists, psychiatrists, social workers) and patients can experience physiological attunement and that the patient reports an increase in perceived empathy as well as progress in those sessions where attunement occurs (Adler, 2007). These cognitive and biological changes occur only after a certain quality (not necessarily quantity) of bonding; this means that the parent, therapist, friend, or mentor must provide appropriately salient “doses” of support in order to alter biological processes. Both cortisol and oxytocin are biological reflections of the social environment and social

A Biological Perspective on Positive Adult-Youth Relationships    123 relationships, reciprocally inhibiting one another across the context of a salient social relationship. Heinrichs, Baumgartner, Kirschbaum, and Ehlert (2003) revealed that, during a laboratory social stressor, the combination of a significant other and intranasal oxytocin administration buffered the expected cortisol response, but this effect was found only if the significant other was perceived as supportive, and, surprisingly, social support was a greater buffer than the hormone alone. Rhodes and DuBois (2006) suggested that mentor characteristics can promote or impede the effectiveness of the relationship. If we expand the notion of physiological attunement to mentoring, we can predict that sensitivity, support, and stability conveyed by the mentor and perceived by the mentee would facilitate attunement of biological processes, perceived empathetic connection, and the most successful outcomes for the youth. Similar to the therapist-patient dyad, empathy has been identified as a factor crucial to the effectiveness of the mentor-mentee relationship (Rhodes & DuBois, 2006). Empathy has also been consistently related to cortisol levels (Shirtcliff et al., 2009) and oxytocin receptor subtypes (Rodrigues, Saslow, Garcia, John, & Keltner, 2009). A contextspecific balance of cortisol and oxytocin is necessary for an empathetic connection between individuals. For example, an increase in cortisol allows the individual to activate the appropriate empathy-related brain regions (Shirtcliff et al., 2009) and affords the individual the capability to interact with the salient social aspects of his or her environment. Simultaneously, an increase in oxytocin is required to facilitate a subjective feeling of connection and the desire to behaviorally express that feeling (Insel, 1997). Feeling the pain, emotion, and stress of another individual is critical to empathic processing and may be a necessary ingredient in successful youth mentoring in both members. Allostatic Load: How Can Mentoring Relationships Predict Health Outcomes? Our health depends on the functioning of our biological processes, either directly (because poor health is reflected in the functioning of biological systems) or indirectly (as environmental factors are reflected in the functioning of biological systems, which in turn influence health). Understanding how biological mechanisms are influenced by social relationships can lead to a better understanding of how mentoring relationships can protect against adverse health outcomes. For example, daily hassles (i.e., stress) are associated with an increased

occurrence of concurrent and subsequent health issues, such as flu, sore throat, headaches, and backaches (DeLongis, Folkman, & Lazarus, 1988). Consistent with the theory of allostasis, activating the HPA axis to cope with external changes is an adaptive response that enables survival in a range of environmental contexts. However, whenever the allostatic response becomes dysregulated, maladaptive outcomes are likely. It makes sense for our bodies to change acutely (by increasing body temperature, activating an immune response, and adjusting cortisol levels) whenever we are ill, but if these responses are prolonged, these systems can deteriorate and have long-term negative effects. McEwen and Stellar (1993) introduced the theory of allostatic load to describe the wear and tear the body experiences as a result of chronic exposure to stress. The biological changes resulting directly from frequent attempts to cope with environmental stress eventually constitute risk factors for deleterious health outcomes. In other words, the stress response system of an individual who repeatedly perceives benign stimuli as stressful will ultimately become less efficient, resulting in either too much or too little hormone release. These dysregulated hormone levels are considered a marker of allostatic load, which has been associated empirically with many health outcomes, including abdominal fat deposition, blood pressure elevation, high cholesterol, cardiovascular disease, and decreased immune function (Lupien et al., 2006). Though most research regarding allostatic load and health involves adult samples, there is reason to believe this theory is applicable throughout development (JohnstonBrooks, Lewis, Evans, & Whalen, 1998). In children, both physical (i.e., crowding and noise) and psychosocial (i.e., separation and violence) stressors can increase allostatic load and lead to negative mental health outcomes (Evans, 2003). DeLongis et al. (1988) and others (Gunnar et al., 2009) have postulated that psychosocial resources should influence the link between stress and health, and others have reported research findings consistent with the idea that supportive relationships provide a buffer against an adverse stress-health association (Seeman et al., 2002). By extension, because mentoring can be a form of psychosocial support (Barrera & Bonds, 2005), we would presume that this protective effect may expand to mentoring relationships. Figure 8.1 depicts how mentoring relationships may provide a buffer against the acute and long-term negative health consequences of stress exposure, in part, by increasing oxytocin levels, regulating context-appropriate cortisol levels, and facilitating attunement.

124  MENTORING RELATIONSHIPS Figure 8.1   The Theorized Psychobiological Benefits of Youth Mentoring

Environment Early/ Chronic Stress Daily Hassles Social Threats Uncontrollability Unpredictability Anticipation Health Outcomes Physical Health Mental Health Allostatic Load Immune Functioning Learning/Memory

Mentor Youth

Buffers Stable, Salient Social Relationships Social Support High BSC

Oxytocin Oxytocin

Cortisol Levels/ Reactivity

Research Research specifically relating psychobiology and youth mentoring is lacking. To date, the theoretical foundations described previously have not been investigated empirically within the context of mentoring. We have provided evidence of a strong empirical foundation from which to draw; however, a true understanding of how biological processes influence and are reflected by mentoring relationships necessitates research dedicated to this association. Some empirical research does exist relating the interplay between genes and supportive environments. Shanahan and colleagues (Shanahan, Erickson, Vaisey, & Smolen, 2007, 2008) have provided two experimental studies to address the association between environmental factors and the genetic risk inherent in individuals with a certain type of dopamine receptor (DRD2 Taq1A). In one study, the presence of a mentor counteracted a genetic propensity associated with worse school performance and less educational attainment (Shanahan et al., 2007). This study recruited 200 adolescents (grades 7–12) and found that the risk associated with the DRD2 TaqA1 gene was compensated by the presence of a teacher who acted as a mentor. The authors suggested that the mentors’ special skills and knowledge aid them in guiding at-risk students toward continuing education. This finding highlights the

Cortisol Levels/ Reactivity Attunement

necessity of equipping mentors with the appropriate competence in order to achieve effective outcomes. The other study found that supportive environments involving social capital (high parental SES, high parental involvement, and high-quality school) can similarly compensate for the genetic risk associated with the DRD2 TaqA1 gene (Shanahan et al., 2008). Albeit limited, these studies provide experimental evidence that external input in the form of mentoring relationships and supportive environments can get under the skin to alter genetic propensities. One such mechanism for this phenomenon is through the action of hormones. Hormones are the scaffolding for the individual’s genetic blueprint (McEwen & Wingfield, 2003); over seconds, minutes, hours, and days, hormones activate genes nearly everywhere in the human body, especially in emotion-related neural circuitry in the brain. Genes are relatively unchanging, but hormones allow their expression to vary across time, social contexts, physical environments, and developmental stages (Gottlieb et al., 1998).

Practice We expect that many of the biological factors described in the preceding sections are relevant to the immediate concerns of those involved in designing and implementing youth mentoring programs.

A Biological Perspective on Positive Adult-Youth Relationships    125 As such, it seems worthwhile to explore some of the possible policy-oriented reasons for exploring biological underpinnings of youth mentoring. Table 8.2 lists some important questions in the practical application of this information. First, biologically informed methods may indicate a mechanism for or provide a window into the etiology of developmental or health phenomena. The theory of allostasis attests to the observation that environments can influence biological functioning. As captured by the allostatic load theory, social environments can also get under the skin in deleterious ways, encouraging dysregulated biological functioning and subsequently exerting negative impacts on health. As a result, we recommend that mentor training include educating mentors on how to be a source of support, rather than a source of stress, for youth. This training would necessitate an emphasis on the importance of sensitivity and stability, while minimizing or avoiding negative evaluative interactions. Second, biological forces may indicate who is the most vulnerable to a particular disorder or, more relevant when considering mentoring, who is most likely to benefit from a supportive environment. Based on the previously mentioned theory regarding biological sensitivity to context, we believe it is important for mentors to understand that underlying individual differences in stress reactivity may influence the degree to which the mentoring relationship may alter biological processes and behavioral outcomes. Mentor training should incorporate ways to identify “orchid” versus “dandelion” children, as well as the appropriate

ways to approach these children. For example, if a mentee has high biological sensitivity to context, it becomes increasingly important for the mentor to provide a low-stress and protective environment. Finally, studies have begun to indicate that an initial biological vulnerability may be able to change as youth benefit from treatment and intervention. This developmental research is particularly provocative because individuals do not have conscious access to or control over these biological processes. Yet, these relatively objective biological outcomes still reflect the benefits of treatment and intervention. This perspective is important because it doesn’t engage a fatalistic perspective about children’s development. If mentoring is to be considered an effective intervention, mentors should be asked to spend quality time with youth in order to establish a consistent supportive relationship that would be salient enough to alter these underlying biological processes. This recommendation does not necessarily imply that the quantity of time spent together needs to be extreme, but rather that the mentor should use the time spent with the youth to provide a positive, supportive, and dependable environment. We caution against biological determinism or reductionism, which has been advanced before in hormone research with little knowledge gained (e.g., “raging” hormones; Buchanan, Eccles, & Becker, 1992). We do not suggest that individuals should be tested for certain biological markers before entering mentoring relationships, or that they should take hormones artificially to induce feelings of bonding (i.e., intranasal oxytocin administration; Kosfeld,

Table 8.2   Questions to Consider When Practically Applying Biology to Youth Mentoring Does mentor training educate mentors on the importance of being a source of support and a stress buffer? Throughout the mentoring experience, is the mentor accidentally becoming a source of stress for the youth, for example, by being an unstable or unpredictable mentor or by appearing socially threatening or overly evaluative? Are mentors trained in identifying biological sensitivity to context, and are they aware that the most sensitive child may be the most likely to benefit from mentoring? Throughout the mentoring experience, are mentors trained in how to adapt their behavior depending on the child’s reactivity and their developmental trajectories? Are mentors asked to provide sensitive and stable support in order to make the relationships salient enough to alter biological processes? Is the mentor aware that he or she may benefit biologically as well by the mentorship experience, through the process of attunement?

126  MENTORING RELATIONSHIPS Heinrichs, Zak, Fischbacher, & Fehr, 2005). Instead, we believe that biology can help explain some of the individual differences observed in mentoring outcomes and offer insight into some of the processes involved when mentors and mentees interact. A general lack of consistency exists in the findings regarding the effectiveness of mentoring (Rhodes & DuBois, 2006) perhaps in part because individual differences in biological mechanisms have been largely ignored, including related factors such as previous stress exposure, attunement, biological sensitivity, and developmental period. We have made a case for the importance of these factors when considering the biological changes (i.e., oxytocin and cortisol) and the subsequent trajectories (i.e., mental and physical health outcomes) that can occur as a result of successful mentoring relationships. Interdisciplinary collaborations of experienced practitioners and mentoring experts with biological researchers may provide fruitful new research directions, especially in the design and implementation of programs for training and educating mentors.

Conclusion Biological processes reflect our social interactions. These processes in turn influence how we interact within those interactions. This phenomenon captures the ability of environments to “get under the skin” and ultimately shape subjective experiences and objective health outcomes. Hormonal functioning is especially sensitive to social environmental contexts, and because hormones circulate throughout the entire body, these systems have a wide range of acute and long-term effects. Especially during development, hormones act as a biological negotiator in the relationship between the environment and health outcomes; hormone levels change according to the environmental demands and activate the biological systems necessary to cope with those demands. Contexts (including social interactions) help determine which level of which hormone is appropriate for adaptively managing the specific environmental input. Oxytocin and cortisol release are two major biological factors involved in the reciprocal relationship between environment and health outcomes. According to the theory of allostatic load, socially rewarding relationships lead to the release of oxytocin, which instantiates subjective feelings of closeness, thereby encouraging behavioral expression of this bond. Regulation of basal and reactive cortisol levels can prevent wear and tear on physiological

systems that would otherwise lead to poor mental and physical health outcomes. Across these systems, attunement may illustrate how the environment of the youth mentoring relationship may be translated into physiology in both members of the relationship. When this connection is reflected physiologically, a perceived connection reinforces the relationship as rewarding and individuals involved are more likely to invest additional time and energy in the relationship. The fact that this social information is capable of getting under the skin, for better or for worse, and influencing how we behave in social settings can be taken as indirect evidence of the importance of a wide range of social relationships for informing how we interpret and process our social worlds. Future interdisciplinary research is sorely needed to translate these psychobiological theories into practice in the youth mentoring context.

References Adler, H. (2007). Toward a biopsychosocial understanding of the patient-physician relationship: An emerging dialogue. Journal of General Internal Medicine, 22, 280–285. Barrera, M., Jr., & Bonds, D. D. (2005). Mentoring relationships and social support. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth men­ toring (pp. 133-142). Thousand Oaks, CA: Sage. Bartels, A., & Zeki, S. (2003). The neural correlates of maternal and romantic love. Neuroimage, 21, 1155–1166. Bernard, C. (1957). An introduction to the study of experi­ mental medicine. New York: Dover Publications. Booth, A., Granger, D., & Shirtcliff, E. (2008). Genderand age-related differences in the association between social relationship quality and trait levels of salivary cortisol. Journal of Research on Adolescence, 18, 239–260. Boyce, W., & Ellis, B. (2005). Biological sensitivity to context: I. An evolutionary-developmental theory of the origins and functions of stress reactivity. Developmental Psychopathology, 17, 271–301. Brotman, L., Gouley, K., Huang, K., Kamboukos, K., Fratto, C., & Pine, D. (2007). Effects of a psychosocial family-based preventive intervention on cortisol response to a social challenge in preschoolers at high risk for antisocial behavior. Archives of General Psychology, 64, 1172–1179. Buchanan, C., Eccles, J., & Becker, J. (1992). Are adolescents the victims of raging hormones? Evidence for activational effects of hormones on moods and behavior at adolescence. Psychological Bulletin, 111, 62–107. Buchanan, T., & Lovallo, W. (2001). Enhanced memory for emotional material following stress-level cortisol treatment in humans. Psychoneuroendocrinology, 26, 307–317.

A Biological Perspective on Positive Adult-Youth Relationships    127 Canon, W. (1929). Bodily changes in pain, hunger, fear, and rage. Boston: Branford. Carter, C., Lederhendler, I., & Kirkpatrick, B. (1997). The integrative neurobiology of affiliation. New York: New York Academy of Sciences. De Kloet, E. (2004). Hormones and the stressed brain. Annals of the New York Academy of Sciences, 1018, 1–15. Del Guidice, M., Ellis, B., & Shirtcliff, E. (in press). The adaptive calibration of stress responsivity: An integrative model. Neuroscience and Behavioral Reviews. DeLongis, A., Folkman, S., & Lazarus, R. (1988). The impact of daily stress on health and mood: Psychological and social resources as mediators. Journal of Personality and Social Psychology, 54, 486–495. Dickerson, S., & Kemeny, M. (2004). Acute stressors and cortisol responses: A theoretical integration and synthesis of laboratory research. Psychological Bulletin, 130, 355–391. Dozier, M., Peloso, E., Lindhiem, O., Gordon, K., Manni, M., Sepulveda, S., . . . Levine, S. (2006). Developing evidence-based interventions for foster children: An example of a randomized clinical trial with infants and toddlers. Journal of Social Issues, 62, 767–785. Eriksson, M., Lundeberg, T., & Uvnas-Moberg, K. (1996). Studies on cutaneous blood flow in the mammary gland of lactating rats. Acta Physiologica Scandinavica, 158, 1–6. Evans, G. (2003). A multimethodological analysis of cumulative risk and allostatic load among rural children. Developmental Psychology, 39, 924–933. Fisher, P., Stoolmiller, M., Gunnar, M., & Burraston, B. (2007). Effects of a therapeutic intervention for foster preschoolers on diurnal cortisol activity. Psycho­ neuroendocrinology, 32, 892–905. Fries, A., & Pollak, S. (2004). Emotion understanding in postinstitutionalized Eastern European children. Developmental Psychopathology, 16, 355–369. Fries, A., Shirtcliff, E., & Pollak, S. (2008). Neuroendocrine dysregulation following early social deprivation in children. Developmental Psychobiology, 50, 588–599. Fries, A., Ziegler, T., Kurian, J., Jacoris, S., & Pollak, S. (2005). Early experience in humans is associated with changes in neuropeptides critical for regulating social behavior. Proceedings of the National Academy of Sciences, 102, 17237–17240. Geary, D., & Flinn, M. (2002). Sex differences in behavioral and hormonal response to social threat: Commentary on Taylor et al. (2000). Psychological Review, 109, 745–750. Gottlieb, G., Wahlsten, D., & Lickliter, R. (1998). The significance of biology on human development: A developmental psychobiological systems view. In R. M. Lerner (Ed.), Handbook of child psychology: Theoretical models of human development (Vol. 1, pp. 233–273). New York: Wiley. Grewen, K., Girdler, S., Amico, J., & Light, K. (2005). Effects of partner support on resting oxytocin,

cortisol, norepinephrine, and blood pressure before and after warm partner contact. Psychosomatic Medicine, 67, 531–538. Gunnar, M., Talge, N., & Herrera, A. (2009). Stressor paradigms in developmental studies: What does and does not work to produce mean increases in salivary cortisol. Psychoneuroendocrinology, 34, 953–967. Gunnar, M., & Vazquez, D. (2006). Stress neurobiology and developmental psychopathology. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathol­ ogy (pp. 533–577). Hoboken, NJ: Wiley. Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003). Social support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress. Biological Psychiatry, 54, 1389–1398. Insel, T. (1992). Oxytocin—a neuropeptide for affiliation: Evidence from behavioral, receptor autoradiographic, and comparative studies. Psychoneuro­ endocrinology, 17, 3–35. Insel, T. (1997). A neurobiological basis of social attachment. American Journal of Psychiatry, 154, 726–735. Johnson, D. (2000). Medical and developmental sequelae of early childhood institutionalization in international adoptees from Romania and the Russian Federation. In C. Nelson (Ed.), The effects of early adversity on neurobehavioral development (pp. 113–162). Mahwah, NJ: Erlbaum. Johnston-Brooks, C., Lewis, M., Evans, G., & Whalen, C. (1998). Chronic stress and illness in children: The role of allostatic load. Psychosomatic Medicine, 60, 597–603. Klimes-Dougan, B., Hastings, P., Granger, D., Usher, B., & Zahn-Waxler, C. (2001). Adrenocortical activity in at-risk and normally developing adolescents: Individual differences in salivary cortisol basal levels, diurnal variation, and responses to social challenges. Developmental Psychopathology, 13, 695–719. Kosfeld, M., Heinrichs, M., Zak, P., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases trust in humans. Nature, 435, 673–676. Lewis, M., & Ramsay, D. (2002). Cortisol response to embarrassment and shame. Child Development, 73, 1034–1045. Liberzon, I., King, A. P., Britton, J. C., Phan, K. L., Abelson, J. L., & Taylor, S. F. (2007). Paralimbic and medial prefrontal cortical involvement in neuroendocrine responses to traumatic stimuli. American Journal of Psychiatry, 164, 1250–1258. Light, K., Grewen, K., & Amico, J. (2005). More frequent partner hugs and higher oxytocin levels are linked to lower blood pressure and heart rate in premenopausal women. Biological Psychology, 69, 5–21. Lupien, S., Ouellet-Morin, I., Hupbach, A., Tu, M., Buss, C., Walker, D., . . . McEwen, B. (2006). Beyond the stress concept: Allostatic load—A developmental biological and cognitive perspective. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathology. Vol. 2: Developmental neuroscience (pp. 578–628). Hoboken, NJ: Wiley.

128  MENTORING RELATIONSHIPS McEwen, B., & Stellar, E. (1993). Stress and the individual: Mechanisms leading to disease. Archives of Internal Medicine, 153, 2093–2101. McEwen, B., & Wingfield, J. (2003). The concept of allostasis in biology and biomedicine. Hormones and Behavior, 43, 2–15. Miller, G., Chen, E., & Zhou, E. (2007). If it goes up, must it come down? Chronic stress and the hypothalamicpituitary-adrenocortical axis in humans. Psycho­ logical Bulletin, 133, 25–45. Petersson, M., Alster, P., Lundeberg, T., & Uvnas-Moberg, K. (1996). Oxytocin causes a long-term decrease of blood-pressure in female and male rats. Physiology and Behavior, 60, 1311–1315. Pfaff, D., Arnold, A., Etgen, A., Fahrbach, S., & Rubin, R. (2002). Hormones, brain and behavior. San Diego, CA: Academic Press. Rhodes, J., & DuBois, D. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20(3). Rodrigues, S., Saslow, L., Garcia, N., John, O., & Keltner, D. (2009). Oxytocin receptor genetic variation relates to empathy and stress reactivity in humans. Psycho­ logical National Academy of Sciences, 106, 21437– 21441. Roy, M., Steptoe, A., & Kirschbaum, C. (1998). Life events and social support as moderators of individual differences in cardiovascular and cortisol reactivity. Personality and Social Psychology Bulletin, 75, 1273–1281. Seeman, T. (2001). How do others get under our skin? Social relationships and health. In C. D. Ryff & B. H. Singer (Eds.), Emotion, social relationships, and health (pp. 189–210). New York: Oxford University Press. Seeman, T., Singer, B., Ryff, C., Love, G., & Storms, L. (2002). Social relationships, gender, and allostatic load across two age cohorts. Psychosomatic Medicine, 64, 395–406. Sethre-Hofstad, L., Stansbury, K., & Rice, M. (2002). Attunement of maternal and child adrenocortical response to child challenge. Psychoneuroendo­ crinology, 27, 731–747. Shanahan, M., Erickson, L., Vaisey, S., & Smolen, A. (2007). Helping relationships and genetic propensities: A combinatoric study of DRD2, mentoring, and educational continuation. Twin Research and Human Genetics, 10, 285–298. Shanahan, M., Erickson, L., Vaisey, S., & Smolen, A. (2008). Environmental contingencies and genetic propensities: Social capital, educational continuation, and dopamine receptor gene DRD2. American Journal of Sociology, 114, S260–S286. Shirtcliff, E. (2005). Neuroendocrine and neural contribu­ tions to pubertal development, normative adolescent

development and affect-related behavior problems. Background paper for the Institute of Medicine Workshop on the Science of Adolescent Health and Development. Shirtcliff, E., & Essex, M. (2008). Concurrent and longitudinal associations of basal and diurnal cortisol with mental health symptoms in early adolescence. Developmental Psychobiology, 50, 691–703. Shirtcliff, E., & Ruttle, P. (in press). Immunological and neuroendocrine dysregulation following early deprivation and stress. In K. Heinz Brisch (Ed.), Attachment and early disorders of development. Munich, Germany: Klett-Cotta & Stuttgart. Shirtcliff, E., Vitacco, M., Graf, A., Gostisha, A., Merz, J., & Zahn-Waxler, C. (2009). Neurobiology of empathy and callousness: Implications for the development of antisocial behavior. Behavioral Sciences and the Law, 27, 137–171. Siever, L., & Davis, K. (1985). Overview: Toward a dysregulation hypothesis of depression. American Journal of Psychiatry, 142, 1017–1031. Spitz, R. (1945). Hospitalism—An inquiry into the genesis of psychiatric conditions in early childhood. Psychoanalytic Study of the Child, 1, 53–74. Taylor, S., Burklund, L., Eisenberger, N., Lehman, B., Hilmert, C., & Lieberman, M. (2008). Neural bases of moderation of cortisol stress responses by psychosocial resources. Journal of Personality and Social Psychology, 95, 197–211. Taylor, S., Klein, C., Lewis, B., Gruenewald, T., Gurung, R., & Updegraff, J. (2000). Biobehavioral response to stress in females: Tend-and-befriend, not fight-or-flight. Psychological Review, 107, 411–429. Taylor, S., Seeman, T., Eisenberger, N., Kozanian, T., Moore, A., & Moons, W. (2010). Effects of a supportive or an unsupportive audience on biological and psychological responses to stress. Journal of Personality and Social Psychology, 98, 47–56. Tennes, K., Kreye, M., Avitable, N., & Wells, R. (1986). Behavioral correlations of excreted catecholamines and cortisol in second-grade children. Journal of the American Academy of Child and Adolescent Psychiatry, 25, 764–770. Uvnas-Moberg, K. (1998). Oxytocin may mediate the benefits of positive social interaction and emotions. Psychoneuroendocrinology, 23, 819–835. Uvnas-Moberg, K. (1999). Neuroendocrinology of the mother-child interaction. Trends in Endocrinology and Metabolism, 7, 126–131. Uvnas-Moberg, K., Widstrom, A., Nissen, E., & Bjorvell, H. (1990). Personality traits in women 4 days post partum and their correlation with plasma levels of oxytocin and prolactin. Journal of Psychosomatic Obstetrics and Gynaecology, 11, 261–273.

9  SOCIAL NETWORKS AND MENTORING Thomas E. Keller and Jennifer E. Blakeslee

Introduction Youth mentoring relationships are embedded within webs of other relationships. The many individuals in the social networks of both mentee and mentor, such as the mentee’s parent or guardian, the mentor’s coworker, or the mentoring program coordinator, are bound to influence the nature and course of the mentor-mentee relationship. In turn, the mentoring relationship can influence the ways in which the mentor and mentee interact with other important people in their lives. Likewise, the mentoring relationship itself will alter the structure and quality of these social networks, if only by virtue of the new bond between mentee and mentor. Situating the mentoring relationship within its complex interpersonal environment provides a more comprehensive and accurate view of mentoring as a systemic intervention (Keller, 2005b), in which relationships are viewed as interdependent and “the quality of one relationship influences and is, in turn, influenced by other relationships” (Gjerde, 1986, p. 297). Attention to the interdependent connections between mentors and mentees and other individuals invites the use of networkoriented theories and research methods to better understand the development and influence of mentoring relationships. Such a perspective acknowledges the roles of program staff, family members, teachers, friends, and other community members in the life of a mentoring relationship. Thus, insights regarding social networks and mentoring have many practical implications for program development and implementation. This chapter begins with an overview of central concepts from social network theory, including the function of social capital and the evolution of

network structure. Next, the potential relevance of these concepts for mentoring is explored by applying them to specific mentoring scenarios. Although these concepts highlight the significance of interpersonal ties and interconnected networks that can be several degrees removed from a particular individual, this chapter’s scope is limited to interactions in which either a mentor or a mentee is directly involved. Accordingly, the research review addresses the relationships a mentor or mentee might have with someone outside of the dyad. The chapter concludes with a discussion of how understanding mentoring in a social network context could guide program practice.

Theory Social network concepts provide a way to explore the function of mentoring in the context of other relationships. Figure 9.1 presents an idealized representation of the mentoring system of mentee, mentor, parent, and program staff (Keller, 2005b). The mentoring system is situated within a wider network of possible mentor and mentee relationships. The figure illustrates how mentor and mentee have preexisting network ties with other individuals (solid lines), how mentoring creates new mentoring system ties (dashed lines), and how the mentoring relationship may lead to new network ties as the contacts of mentor and mentee begin to overlap (dotted arrows). Network concepts describe these network ties and also provide insights regarding the nature of support available to the mentee and the mentor, the factors affecting development of the mentoring relationship, and the potential effects of mentoring on participant networks. 129

130  MENTORING RELATIONSHIPS Figure 9.1   Ties in Hypothetical Mentor-Mentee Social Networks

School

Parent

Partner

Coach

Friends

Friends Mentee

Mentor

Relatives

Relatives

Community

Program staff

Community

Workplace

Existing New Potential For clarity, not all potential new ties are shown.

Social Network Theory A social network simply refers to the people in a social environment and all the relationship ties that link them. Each relationship tie may be strong or weak, one-sided or reciprocated, and transitory or lasting. Furthermore, the relationship tie between two individuals may serve a single role or multiple functions. These aspects of ties describe the content and flow of social processes—like the provision of social support—in networked relationships. The structure of a particular social network is the pattern of direct and indirect ties between a set of identified individuals or groups (Wellman, 1983). Network structure changes as members begin and end relationships or alter the way they interact with other members, by choice or circumstance. Basic structural properties include network size and network density, signifying the degree of interconnection of

network members (Table 9.1). Another distinguishing quality is network composition: do network members generally share attributes, or is membership diverse? The structural distribution of relationships reflects how social networks evolve. Clusters of strong ties are considered network cores, where members are embedded in a regular flow of communication and activity (Morgan, Neal, & Carder, 1997). Because these core relationships are interconnected, the network is structurally cohesive and resists disruption when individual ties disappear (Moody & White, 2003). A family is a common example of a network core, where relationships are strong, are interconnected, and may last lifetimes, although individual members may change over time. Each network core also has a periphery of less-connected weak ties— such as neighbors and coworkers—and these peripheral members are likely connected to other core

Social Networks and Mentoring   131 Table 9.1   Social Network Concepts and Principles Concepts Network size, a broad index of functional capacity, reflects the number of people directly or indirectly linked to each other. Density signifies the degree of interconnection among network members, based on how many ties among all those possible are actually present. Networks can be relatively dense or sparse. Tie strength indicates relational qualities like emotional closeness, reciprocity, frequency of communication, and the breadth of functional interaction and support between individuals. Network composition describes the aggregate nature of member attributes. Network composition can be relatively diverse in the range of member attributes, or very similar for a given attribute (homophily). Principles Strong ties develop slowly and tend to last. These form the core of a person’s network. Core networks are densely interconnected clusters of close relationships among people who share resources and monitor member needs. Strong ties and dense cores provide network closure and facilitate bonding social capital. Weak ties come and go more quickly (if they do not develop into strong ties). These form the periphery of a person’s network. Less dense (sparse) peripheral networks tend to be more diverse than the core network. Weak ties facilitate bridging social capital by connecting otherwise separate networks.

networks. Networks expand as new connections are made and as initially weak ties grow stronger. When the core expands with the development of relationships, the stability of the network increases, and more members are brought into an increasingly diverse periphery. One characteristic of networks illustrating their functional value is social capital. Social capital can be understood as the accessible information, resources, and support that individuals can draw on through their network relationships (Coleman, 1988). Social capital is often conceptualized as a bonding function of strong ties to family and community (Coleman, 1988), and a bridging function of weak ties to people outside of one’s close network (Burt, 1992). Bonding social capital is present when members of a network know each other and jointly influence attitudes and behavior, including a reciprocal obligation to provide support and enforce norms. This process is also called network closure (Coleman, 1988), and it embeds individuals in a dense core of predictably supportive relationships developed over time. On the other hand, bridging social capital refers to how individuals can use their network connections to obtain information and resources beyond the reach of their core relationships. This happens when weak ties—with a friend-of-a-friend or acquaintance—provide a bridge between separate

social groups (Granovetter, 1973). Thus, individuals benefit from a diverse periphery of network ties that bring within reach information and resources available in other networks. Both strong and weak ties influence the availability of social capital because each facilitates the flow of social resources through the network to the individual. Application of network concepts to youth mentoring illustrates how a network perspective is relevant to four organizing premises: (a) social networks determine the availability of mentors and the formation of mentoring relationships, (b) individuals other than mentors and mentees influence mentoring relationships, (c) mentoring has effects on relationships outside the match, and (c) mentoring can expand social network structures. Effects of Network Structure on Mentoring Network theory suggests that the presence of informal or natural mentors is more likely when a youth’s network is large, dense, diverse, and stable. The support and guidance provided by natural mentors depends not only on the frequency and quality of contacts but also on the type of social capital available through the relationships. Natural mentors who are core network members, such as relatives and family friends, may offer strong, durable bonds characterized by personal understanding, emotional

132  MENTORING RELATIONSHIPS support, and shared identity. These core network members would be connected to several other adults in the youth’s life, and this network closure would enable communication and coordination regarding youth activities and behaviors and thereby encourage accountability (Coleman, 1988). Natural mentors less connected to the core network may come from contexts that bring youth into contact with adults, such as schools, youth programs, jobs, and faithbased organizations (Hamilton et al., 2006). These mentors may be “institutional agents” linking youth to important social systems and bridging access to opportunities in other networks (Stanton-Salazar & Spina, 2008). However, some communities, due to limited resources or a history of oppression, may lack adults to serve as mentors and yield few network ties offering social mobility (Wilson, 1987). Though a single mentor may play multiple roles in the life of the mentee (Goldner & Mayseless, 2008), a large and diverse network may afford multiple relationships that each serve a primary function and combine to provide a full complement of support (Higgins & Kram, 2001). Young people whose networks lack size, diversity, and stability are less likely to experience these mentor-rich social environments and may benefit from additional adult support through formal mentoring programs (Freedman, 1993). Furthermore, the nature of the existing network may moderate the importance of a new mentoring relationship, such that mentoring may be relatively more significant when it plays a compensatory role for youth without strong networks or adequate support, as with youth in foster care (Gilligan, 1999). Finally, the networks of youth receiving services from systems such as child welfare, mental health, disabilities, or juvenile corrections, may necessitate mentor contact and coordination with a set of professionals involved in the mentee’s care (Britner, Balcazar, Blechman, Blinn-Pike, & Larose, 2006). Effects of Network Relationships on Mentoring The interdependent nature of social environments suggests that each person in the network of the mentor or mentee can influence the success of the mentoring relationship through second-order effects, defined as “the mediation of a dyadic process through the influence of a third party” (Bronfenbrenner & Crouter, 1983, p. 380). Interactions between the mentor and mentee may be affected without specific intent or awareness on the part of the third party, as issues related to someone outside of the match may be the focus of mentoring activity. For example, the mentor may work with the mentee on strategies for

refusing to engage in a peer’s negative behavior. Alternatively, competing demands on the mentor’s time and attention by a partner might interfere with the consistency and duration of meetings with the mentee. Likewise, the emotional tone of a prior interaction in another setting might spill over into the mentoring relationship (Westman, 2006). A mentor just berated by a coworker may exhibit irritation during an outing with the mentee, who then becomes upset. It is also possible that past experiences with a significant person may contribute to a representational model of relationships stored in memory that guides behavior in the mentoring relationship (Berlin & Cassidy, 1999). For example, a judgmental coach may have made the mentee wary of forming a relationship with another nonparental adult and predisposed to interpret the mentor’s behaviors for signs of rejection (Downey, Lebolt, Rincon, & Freitas, 1998; Larose & Tarabulsy, this volume, Chapter 20). Lastly, mentor-mentee interactions may be altered when the presence of a third person, such as the mentee’s sibling or a friend, creates a triad (Gjerde, 1986). In many cases, the influence of another person on the mentoring relationship may be purposeful and may serve to support or undermine the relationship. In this regard, the quality of communication and coordination within the network system composed of mentor, mentee, parent or guardian, and mentoring program worker may be particularly significant (Keller, 2005b). The parent or guardian may facilitate the development and maintenance of the mentoring relationship via interactions with the mentor or mentee. In establishing a relationship with the mentor, the parent can (a) provide valuable information about the history, interests, and needs of the mentee; (b) communicate family values and goals for the mentoring relationship; (c) help the mentor interpret the mentee’s behavior; and (d) sustain the mentor’s motivation through appreciation and encouragement. Likewise, the parent’s influence on the relationship can be mediated through the mentee by (a) reassuring the mentee about spending time with a new adult, (b) reinforcing expectations for appropriate behavior when with the mentor, (c) assisting with scheduling and keeping meetings with the mentor, and (d) monitoring the mentoring relationship and communicating or problem-solving any concerns that arise (Keller, 2005b). Conversely, the parent can hinder the development of the mentoring relationship by failing to communicate or cooperate with the mentor or mentee in the ways noted above. In addition, misunderstandings and conflicts between the parent and mentor could have negative repercussions for the

Social Networks and Mentoring   133 mentoring relationship. In more extreme cases, the parent or guardian has the power to block the relationship altogether by denying opportunities for contact between mentor and mentee (Keller, 2005b). The agency worker also plays a crucial role throughout the development of the mentoring relationship by providing training, offering advice, monitoring boundaries, making referrals for opportunities or services, and facilitating communication and problem-solving through initial and ongoing contacts with both mentor and mentee (Keller, 2005a). Similarly, individuals in the mentor’s network, such as a partner, coworker, or friend, may promote the mentoring relationship by accommodating or encouraging the mentor’s volunteer commitment even if they never meet the mentee. This is also true when individuals in the mentee’s network, such as a grandparent or a friend, support the mentee’s involvement with the mentor. By the same token, these other individuals may have reasons to downplay or disparage the mentoring relationship. Peers are particularly influential, and peer attitudes toward mentoring in general, or toward a particular mentor, may positively or negatively affect youth engagement in the relationship. Mentoring Effects on Network Relationships The reciprocal flow of effects through a network subsystem implies that what transpires in the mentoring relationship can positively or negatively influence how the mentor and mentee relate to other individuals in their respective networks (Keller, 2005b). For example, a mentor’s sense of accomplishment derived from a successful mentoring experience could be contagious to other personal relationships or could increase his or her support to others in the community. In contrast, facing difficulties or feeling overwhelmed in the mentoring relationship could strain another relationship if the mentor consistently complains, vents frustration, or seeks additional support from a partner or friend. Mentoring programs and individual mentors may have various goals, but most are likely to involve influencing the mentee’s behavior in social settings. Mentoring, as a relationship-based intervention, is expected to have effects in the social domain. Numerous mechanisms have been proposed to explain how a mentor might affect a mentee’s behavior toward others (Keller, 2007a; Rhodes, 2002; Rhodes, Spencer, Keller, Liang, & Noam, 2006). The mentor may teach social skills, promote self-regulation, facilitate communication, encourage empathy, and reinforce social expectations and norms. A mentor may role model appropriate

behavior and positive interactions in a variety of social settings. The relationship also may give the mentee valuable practice in seeking and receiving support. Thus, a sensitive and consistent relationship with a mentor may provide a “corrective experience” that changes the youth’s working model of relationships (Rhodes et al., 2006). A mentor also may have second-order effects on the mentee’s relationships with others. For example, the mentor may alter a stressful parent-child relationship by providing respite, helping each to understand the other’s perspective, providing each problem-solving advice, and offering each a more hopeful view of the youth’s future (Keller, 2005b). Mentoring Effects on Network Structure The introduction and growth of a mentoring relationship can shape youth network structure by increasing social capital (Hamilton & Hamilton, 2004; Rhodes, 2002). This may occur by strengthening existing bonds in the youth’s network, linking youth networks with mentor networks, or developing connections to entirely new networks. These structural changes facilitate the bonding and bridging functions of social capital, which are served jointly by stable network cores and increasing compositional diversity. Mentoring can enhance the stability of the youth’s core network by supporting network ties and facilitating interconnection between network members. Generally, the strength of a network tie reflects emotional affect, reciprocity, and communication frequency (Granovetter, 1973). A mentor can help the mentee to develop these aspects of relationships. A mentor also can tend to the mentee’s social network by actively engaging primary network members with regular check-ins—and encouraging members to check in with one another—to monitor youth needs and behavior by closing communication loops in the core network. When both youth and mentor have a connection with another network member, or when mentors help to strengthen the ties of any two members of the youth’s network, those ties are more likely to be a lasting part of the youth’s life (Degenne & Lebeaux, 2005). In these ways, the involvement of a mentor can contribute to a more stable network core that provides bonding social capital to the youth. On the other hand, bridging social capital depends on the social network’s compositional diversity and range. Core networks tend to be composed of members with similar attributes, but healthy networks also have a diverse periphery of weak ties with access to new information and resources (Burt, 1992;

134  MENTORING RELATIONSHIPS Granovetter, 1973). In most mentoring programs, the mentor and youth are embedded in different social spheres, and the mentoring relationship itself increases the diversity of both networks in terms of attitudes, beliefs, social norms, and sociodemographic position. This bridging link automatically increases youth access to resources and opportunities through potential ties to members of the mentor’s network. For example, a mentor may introduce the mentee to a coworker based on a shared interest or may contact a friend in pursuit of a youth-identified goal. These bridging ties expand the network periphery in terms of the diversity of members and the range of social groups from which members are drawn (Higgins & Kram, 2001). In youth mentoring, this kind of compositional expansion may increase youth access to educational and occupational mobility, which are critical functions of social capital (Furstenberg & Hughes, 1995). Social connections are generally how individuals, particularly adolescents, learn about and gain entry to higher education (e.g., scholarship information) and employment (e.g., informal referrals) (Granovetter, 1973). Similarly, site-based programs afford mentors opportunities to become familiar with the culture, context, and organizations in the youth’s community. By mentoring in a local school, for example, a mentor may become aware of issues facing schools and may advocate for the school system as well as the youth. Thus, the mentor’s network may expand with new relationships with teachers, staff, and other school stakeholders.

Research Studies employing methods established for social network analysis have not been conducted in the field of youth mentoring, and youth mentoring research rarely refers to the social network characteristics of mentors and mentees. Nevertheless, a growing literature provides evidence of relationships beyond the dyad shaping and being shaped by the mentoring relationship. The search for research literature was conducted with the Social Science Citation Index/ISI Web of Knowledge and PsycINFO databases and YouthMentoring listserv using the keywords mentor* or social* and network* or capital*, with youth*, adolescen*, or teen*. Also, bibliographies in located sources were used to identify additional references.

form of access to relationships with nonparental adults. Youth typically refer to relatives or kin when asked which nonparental adults have an important influence on them, make a difference in their lives, or otherwise act as mentors (Beam, Chen, & Greenberger, 2002; DuBois & Silverthorn, 2005). However, many youth also name natural mentoring relationships with someone acting in a professional capacity (e.g., teacher, counselor, minister) or with whom they have other forms of contact (e.g., coach, employer, neighbor) (DuBois & Silverthorn, 2005). In one study, the kinship status of these mentors corresponded with type of support for the mentee, with kin mentors providing long-term and diffuse support as members of the core network, and nonkin mentors providing more targeted and multidimensional support (Beam et al., 2002). Some research suggests how youth form natural mentoring relationships in networks with different compositional characteristics. Older youth in foster care are more likely to become acquainted with their nonkin natural mentors through formal contacts with the child welfare, school, or mental health systems than through informal channels involving family and friends (Munson & McMillen, 2006). Extended family and kin networks have traditionally been a source of supportive adult relationships for youth in African American communities (Taylor, 1996). Likewise, for a sample of Latino adolescents, most natural mentors represented familial relations, and almost all mentors were also of Latino background (Sánchez & Reyes, 1999). In a recent study, Latino youth who identified more natural mentors providing multidimensional support, especially from mentors in institutional networks, were more likely to receive subsequent mentoring (Sánchez, Esparza, Berardi, & Pryce, 2011). However, another study found the presence of nonfamily natural mentors among urban, low-income, immigrant Latino youth to be rare, concluding that despite the presence of eligible role models in the community, “institutional mechanisms for fostering these relations are weak or absent” (Stanton-Salazar & Spina, 2008, p. 251). Several studies document how youth development and after-school programs can serve this institutional connecting function when staff and volunteers establish natural mentoring relationships with youth (Hirsch, 2005; Jarrett, Sullivan, & Watkins, 2005). Effects of Network Relationships on Mentoring

Effects of Network Structure on Mentoring Natural mentoring arises in networks providing youth bonding or bridging social capital in the

Growing evidence illustrates how individuals beyond the mentor-mentee dyad contribute to mentoring experiences, reflecting the interdependence

Social Networks and Mentoring   135 of relationships of various strength and quality in networks. The benefits of school-based mentoring are greater for students who initially have relationship profiles featuring satisfactory, but not particularly strong, relationships compared to students with profiles characterized by either strongly positive or strongly negative relationships (Schwartz, Rhodes, Chan, & Herrera, 2011). Youth outcomes are also more robust for mentoring programs that address parent involvement (DuBois, Holloway, Valentine, & Cooper, 2002; Taylor & Porcellini, this volume, Chapter 31). This finding may reflect parent reinforcement of the mentoring relationship or greater program staff capacity. A recent study found specific program staff influences on school-based mentoring program effectiveness (Karcher & Spencer, in press). Two staff characteristics, program experience and extent of training, explained school-toschool variation in the degree to which mentored Latinas felt connected to the future. Reflecting network principles, a third factor that explained setting-level variation in mentee connectedness was the quality of relationships between program staff and school personnel. In a qualitative study that noted the contributions of program staff, Big Brothers Big Sisters (BBBS) volunteer mentors appreciated ideas for activities, guidance on building the mentoring relationship, and advice about how to deal with the mentee’s family (Morrow & Styles, 1995). A qualitative study of unsuccessful mentoring relationships cited cases that terminated prematurely when the program worker was unable to mediate mentor-mentee conflict or when the program worker became overly involved and created indirect communication patterns (Spencer, 2007). Recent qualitative research also provides insights regarding the roles that parents or guardians play in youth mentoring relationships, including as collaborators working with the mentor, coaches providing guidance to the mentor, or mediators trying to resolve issues in the mentoring relationship (Spencer, Basualdo-Delmonico, & Lewis, 2010). Positive and effective mentoring relationships are more likely when the mentor and parent are better acquainted and share understandings and expectations (Meissen & Lounsbury, 1981). However, constructive parent-mentor dialogue and appropriate role definitions may be difficult to achieve. Qualitative researchers observed that relations between parent and mentor frequently posed challenges, and their resolution was integral to the stability and success of the matches (Morrow & Styles, 1995; Styles & Morrow, 1992). For instance, some parents asked mentors to assume parent-like responsibilities for instilling values and

regulating child behavior; some mentees were reluctant to share confidences if mentors seemed too closely aligned with the parent (Styles & Morrow, 1992). However, in longer-lasting and more successful matches, mentors resisted efforts by parents to shape the direction of the match and instead focused on a youth-driven relationship. Mentorparent interactions also posed risks if the mentor’s involvement expanded to intervention in family problems, without prompting or consent from the family (Morrow & Styles, 1995). Mentoring Effects on Network Relationships Mentoring can improve mentee interactions with others, and a growing body of research documents promotion of prosocial behaviors and reduction of social isolation as mentoring operates in network context. Specifically, a meta-analysis of mentoring program evaluations showed evidence of improved social competence (DuBois et al., 2002), while a meta-analysis of three major randomized trials of school-based mentoring found significant effects for improving peer support and promoting relationships with nonfamilial adults (Wheeler, Keller, & DuBois, 2010). There also is evidence that improvements in mentored youth outcomes are due partially to the improved quality of the mentee’s other relationships. Investigating models of developmental change among mentees in a randomized trial of the BBBS community-based program, Rhodes and colleagues reported that improved parentchild relationship quality accounted in part for program effects on improved academic outcomes and reduced substance use (Rhodes, Grossman, & Reddy, 2005; Rhodes, Grossman, & Resch, 2000). Likewise, a study of a peer mentoring program in which high school students mentored younger students resulted in enhanced connectedness of mentees to their parents. Improvement in connectedness to parents was in turn related to gains in academic achievement. The parents corroborated these results by reporting that discussions of mentoring program activities provided them opportunities to engage positively with their children (Karcher, Davis, & Powell, 2002). In an experimental evaluation of a mentoring program for youth receiving mental health services, the parents of mentored youth reported less parenting stress and higher levels of parenting-related support than parents in the control group, while the mentees reported more satisfying parent-child relationships than did children on the waiting list (Jent & Niec, 2006). Whereas evaluation studies illustrate effects of program participation, other research has considered

136  MENTORING RELATIONSHIPS how aspects of the mentoring relationship correspond to social interactions in other relationships. For example, in a school-based mentoring program, students indicating higher levels of closeness to their mentors reported more improvement in trust and satisfaction in their relationships with their mothers and also had greater gains in teacher-rated social functioning (Goldner & Mayseless, 2009). Furthermore, students with closer mentoring relationships were more likely to credit mentoring with their improved social connections and support. In another prospective study with paid mentors, mentees rating higher mentoring relationship quality also reported higher levels of parent attachment, greater ability to form friendships with adults, more comfort with self-disclosure, and better school bonding after mentoring, even after controlling for baseline scores for these variables (Thomson & Zand, 2010; Zand et al., 2009). Lastly, some school-based program outcomes may be due partly to improved peer relationships, wherein the presence of the mentor in the school can enhance a child’s social reputation among peers and mentors can role model positive interaction in the peer group context (Cavell & Henrie, 2010). Mentoring Effects on Network Structure Although evidence that mentoring increases social capital is limited, some studies illustrate how this might happen through the expansion and exploration of resources in mentee networks. Investigating a program matching African American adolescent mothers to women in professional careers, Zippay (1995) described how the mentors introduced the mentees to their colleagues and provided educational and career information not typically available to the mentees. Research on the natural mentors of African American adolescent mothers showed that mentees with greater mentor support had more positive beliefs about their opportunity structure and participated in more career development activities, and these factors were associated with greater overall optimism in life (Klaw & Rhodes, 1995). Natural mentors helped African American adolescent mothers make better use of their networks and also buffered the negative effects of relationship problems in other network relationships (Rhodes, Ebert, & Fischer, 1992). Mentoring also may influence social capital by connecting youth to institutional networks. Linnehan (2001) found that a work-based mentoring program for urban African American students improved school attachment in terms of grades and attendance. Compared to nonmentored youth, those receiving mentoring believed school was relevant to their

future employment, and this belief also was associated with satisfaction with their mentor (Linnehan, 2003). No difference was observed between formal or informal mentoring, suggesting that the relationship itself increased connectedness through activities in the institutional network. Similarly, in an evaluation of a school-based program, urban minority youth who were “mentored as intended” by teacher mentors reported an increased sense of school belonging, while this measure declined in comparison groups (Holt, Bry, & Johnson, 2008). In another study, developmental mentoring with rural 4th and 5th graders facilitated school connectedness via structured program activities with older peer mentors from middle school (Karcher, 2005). Mentoring also connects youth to employment. Analyses of the national Add Health sample revealed that informal mentoring in adolescence and workplace mentoring in young adulthood were positively related to workforce attachment in terms of full-time employment (McDonald, Erickson, Johnson, & Elder, 2007). Young men with weak, bridging ties to friend and relative mentors during adolescence were nearly twice as likely to have full-time jobs in early adulthood, and young men who reported teachers as mentors were also more likely to be employed. For both men and women, receiving guidance and advice during adolescence was associated with later attachment to the workforce and relationships with workplace mentors. Conclusions/Future Directions A search of the literature for research on how social networks and relationships affect mentoring and how mentoring affects relationships and networks revealed few studies specifically examining these linkages. However, evidence gleaned from existing research lends preliminary, if tangential, support for the four organizing premises: (a) social networks determine the availability of mentors and the formation of mentoring relationships, (b) individuals other than mentors and mentees influence mentoring relationships, (c) mentoring has effects on relationships outside the match, and (d) mentoring can expand social network structures. The most compelling evidence is for the effects of mentoring on the mentee’s other relationships, particularly those with parents and peers. Because much of the mentoring literature focuses on evaluation of youth outcomes, these effects have been established through experimental and quasi-experimental studies. Other outcome research suggests that mentoring can enhance or extend social networks, but these studies typically rely on attachment to school

Social Networks and Mentoring   137 or workforce as proxies for network structure. Few studies investigate the development of mentoring relationships, and fewer still consider social factors external to the mentoring dyad. A network perspective suggests the need for greater attention to examining mentoring relationships with reference to their social context. For example, Britner et al. (2006) recommended evaluating changes in other youth support relationships as a result of mentoring. Similarly, DuBois, Doolittle, Yates, Silverthorn, and Tebes (2006) recommended that attention be paid to the size of youth networks and the interconnection between primary network members (e.g., mentor, parent, teacher), as these are rarely investigated. Future research assessing the networks that encompass the mentoring relationship should be conducted using established network analytic methods. As a burgeoning field of research, social network analysis is widely used to capture the interdependence of relationships in a complex network structure (Wasserman & Faust, 1994). Systematic network surveys can be used to measure basic network structure, including size, density, and composition (see Table 9.1). Further analysis can compare network ties, such as the relative strength of specific relationships (e.g., mentor-youth tie strength in terms of duration or contact frequency), or the distribution of strong and weak ties within the network. As the field of mentoring becomes more informed by network concepts, this type of network measurement would be as practical as any other survey assessment, although advanced network analysis requires specialized training.

Practice Consideration of youth mentoring from a network perspective yields many potential implications for program policy and practice. Although the research reviewed here helps to establish the relevance of applying network-oriented concepts to youth mentoring, the following observations and suggestions are presented tentatively due to the lack of mentoring research based on these conceptual frameworks. Table 9.2 offers recommendations for programs adopting a network orientation. Effects of Network Structure on Mentoring Supporting a youth’s development through mentoring might first involve a thorough assessment of the youth’s existing network for important relationships with adults and for the presence of other adults who could establish stronger relationships with the youth. Using methods developed for social network analysis (e.g., Wasserman & Faust, 1994), the pattern of relationships could be evaluated for both its structure and its ability to deliver functional support. The attributes of the individuals composing the network and the degree of interconnection between them could indicate, for example, that a youth has a dense core of close, supportive relationships within a tight-knit family or community. In this case, the youth may be in need of a mentor who can expand opportunities on the periphery by providing entry to new settings and networks. Alternatively, a youth may lack multiple

Table 9.2   Checklist for Practitioners Topic

Recommendations

Assessment

Assess the mentee’s network before matching with a mentor. Assess the mentor’s network before matching with a youth.

Matching

Match to encourage bonding or bridging in the youth’s social network.

Training

Have guidelines for mentor’s interactions with mentee’s parents, teachers, and peers. Have guidelines for mentee’s interactions with mentor’s partner, coworkers, and friends. Have training address how others in the youth’s social network, such as parents, peers, teachers, may affect the match (e.g., eliciting support, avoiding pitfalls). Have training address how the mentor’s existing network can support, or potentially hinder, participation in mentoring activities.

Support

Emphasize helping the mentee to build positive relationships with others. Emphasize introducing the mentee to new people to expand networks and increase opportunities.

138  MENTORING RELATIONSHIPS family ties and have contact primarily with adults in professional roles, which might suggest a mentor who could anchor the youth’s core personal network by establishing close relationships and focusing on bonding social capital. Beyond identifying existing needs or gaps in the network, assessment might suggest opportunities for the program to recruit and enlist particular adults for mentoring the youth. Such an approach might involve extra training and support for the prospective mentor and a formal commitment that promotes development of the mentee’s social network. An example is the FosterClub contract that enables youth in foster care to have conversations and form agreements with adults to clarify the types of support the adults are willing to provide. Alternatively, the role of a formal mentor might include helping the mentee forge stronger relationships with individuals in the youth’s existing network. The mentor might coach the mentee on reaching out to others or facilitate joint activities involving the mentor, mentee, and other adults. This approach begins with assessment of the youth’s social network across the ecologies of school, home, and peers. Even if the entire network cannot be assessed, the youth could be asked about the presence of other mentors and nonparental adults, their role in the youth’s life, and the support received from these network members (e.g., Sánchez et al., 2011). Assessing the youth’s network also may indicate how a new mentor might complement rather than compete with existing relationships. A goal may be to increase the diversity of the network by making sure the mentor is not duplicating types of instrumental support already available to the mentee through network ties. The assessment also might reveal issues or concerns of which the mentor should be aware, such as how an absent parent might feel about the youth having a mentor, how a sibling might also want to spend time with the mentor, or how the mentee might be involved with peers exhibiting negative behaviors. Likewise, a parallel assessment of the mentor’s existing social network could yield valuable information about his or her ability to make and keep a commitment to the youth. Such an analysis might indicate, for example, whether multiple ties to partner, family, friends, and supervisor demand the mentor’s time and attention and whether these individuals encourage and support participation in mentoring. The features of the mentor’s network also might suggest the capacity of the mentor to serve as a bridge to new contacts and opportunities for the mentee.

Effects of Network Relationships on Mentoring A social network perspective on youth mentoring highlights the importance of individuals beyond the mentor and the mentee in the success of the mentoring experience. In the case of formal program mentoring, agreement on goals and mutual cooperation among mentor, mentee, parent or guardian, and program staff may determine the overall success of the intervention (Keller, 2005b). In this regard, the program staff worker can play a pivotal role in encouraging network closure by monitoring the interactions among these individuals to actively prevent interpersonal conflicts that may inhibit positive outcomes. Program staff must be adept at assessing complex interpersonal situations, developing problem-solving strategies, and imparting advice and guidance. Indeed, the influence of program staff experience and training on mentoring outcomes seems particularly worthy of attention (Keller, 2007b). Furthermore, program design should incorporate monitoring and assessment of aspects of the social settings in which mentoring relationships are embedded (Deutsch & Spencer, 2009). Programs will vary in the degree to which parent, peers, and others are actively engaged in program activities along with the mentor and the mentee, but measurement of aspects of these relationships provides important monitoring and evaluation data. Further, it may also be valuable for each of these parties to receive preliminary training on the network of complementary roles and responsibilities involved in the overall mentoring intervention (Keller, 2005b). Mentoring Effects on Network Relationships Much of what happens naturally in a youth mentoring relationship is likely to contribute to the socialization of the mentee into ways of behaving and interacting in various settings. For programs that offer substantial latitude and rely on the mentor’s judgment to provide individualized attention to the mentee’s needs, perhaps the best recommendation is to encourage mentors to be more intentional about improving the mentee’s other social relationships. For programs that establish specific goals for the mentoring relationship, promotion of the mentee’s relationships with parents, teachers, peers, and others could be explicitly emphasized, and mentors could be trained on how to enhance the mentee’s social skills in these relationships. A structured curriculum could include lessons and activities on developing and maintaining strong relationships. Programs also could build the mentee’s relationships by inviting parents, friends, and others to participate

Social Networks and Mentoring   139 in activities and special events sponsored by the agency. For example, programs can combine curriculum designed explicitly to foster connectedness to peers, teachers, and family with special program events that bring these parties together with mentors and mentees (e.g., Karcher, 2005). By promoting close and constructive relationships in these ways, mentors can augment the bonding dimension of social capital by strengthening the network core. As an example, the Check & Connect program enhances primary social relationships to keep youth connected to school (Anderson, Christenson, Sinclair, & Lehr, 2004). In Check & Connect, mentoring is individualized based on multiple network factors, including school engagement, the quality of existing relationships, and the availability of community resources. Mentors, also called “monitors,” work to establish interconnections, increase communication, and improve collaboration among youth network members, including the family, school staff, and service providers. Mentors explicitly encourage problem solving among all network members, including problem solving with the youth to build social competence. Further, mentors maintain and expand youth connection to school through advocacy (e.g., negotiating with administrators to prevent an expulsion) and facilitation of youth participation in school activities. Mentoring Effects on Network Structure To foster expansion of the mentee’s social capital through new community connections, mentors could be encouraged to introduce mentees into various spheres of their lives. Mentors could receive training in how to strategically diversify the mentee’s network with bridging contacts into settings such as colleges and workplaces. Programs could maximize these opportunities by sponsoring group activities and visits, so that multiple matches can take advantage of each mentor’s connections. Diversification of the mentee’s network also can come through introductions to individuals who share similar interests and hobbies, can teach a skill, or may provide information and advice. Consequently, mentors can build mentees’ networks through venues such as clubs, service organizations, and family celebrations. Programs also could focus on the continuity of mentoring for youth participants over time. If formal mentoring is viewed as preparing youth for obtaining support in later mentoring relationships, then programs could be more explicit about strategies for establishing youth relationships with future formal and informal mentors. Mentors and program staff also could take an active role in both helping youth identify their next mentor and working with

the youth and the new mentor to make a smooth transition when the program match ends, thereby reducing the impact to the network when the youth’s tie to the program mentor becomes less involved. Other approaches include group and team mentoring, which provide opportunities for constructing a dense network of supportive relationships among multiple adults and youth (Kuperminc & Thomason, this volume, Chapter 18). Beyond offering youth instant access to an established network of mentors and peers, communication between mentors in these programs fosters mutual support and network closure. A model program explicitly focused on fostering network interconnection and expansion is the Blue Ribbon Mentor-Advocates (BRMA) program. To participate, youth first must “show a desire for additional interaction with adults, and their families must be willing to support the program through involvement with the mentor” (BRMA, n.d.). BRMA mentor-advocates establish a long-term, meaningful relationship with the mentee, while also promoting school success through advocacy with and on behalf of the youth and parents. The program aims explicitly to increase connections between families and schools and to add new network connections consistent with youth development goals, with the motivating premise that “these connections weave a tighter social fabric that improves all parts of our community” (BRMA, n.d.).

Conclusion Youth mentoring relationships operate within complex social environments. In reciprocal fashion, mentoring relationships are influenced by the social networks of the mentor and the mentee and in turn alter those networks. Network-oriented conceptual frameworks provide insights into these processes when applied to mentoring. Although these conceptual models can be used to generate hypotheses regarding potential mentoring scenarios, they rarely have served as the basis for empirical research. Growing awareness of social network influences and increasing access to network research methods offer great promise for understanding these expanded dimensions of the mentoring experience. Nevertheless, existing research findings buttress the claim that relationships beyond the mentoring dyad influence the development and impact of mentoring relationships. This fundamental notion is also consistent with experience rooted in practice. Many programs address

140  MENTORING RELATIONSHIPS issues involving individuals outside of the dyad, such as family members, teachers, and other professionals, though that aspect of practice may not be acknowledged with formalized guidelines or procedures. Although mentoring relationships and mentoring programs already function within social networks, explicit and strategic attention to network interactions might promote more productive relationships and greater program effectiveness. Programs that focus on enhancing the mentee’s social network have the potential not only to serve the needs of mentees but also to build strong and interconnected communities.

References Anderson, A. R., Christenson, S. L., Sinclair, M. F., & Lehr, C. A. (2004). Check & Connect: The importance of relationships for promoting engagement with school. Journal of School Psychology, 42, 95–113. Beam, M. R., Chen, C., & Greenberger, E. (2002). The nature of adolescents’ relationships with their “very important” nonparental adults. American Journal of Community Psychology, 30, 305–325. Berlin, L. J., & Cassidy, J. (1999). Relations among relationships: Contributions from attachment theory and research. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 688–712). New York: Guilford Press. Blue Ribbon Mentor-Advocate Program. (n.d.). Frequently asked questions. Retrieved from http://blueribbon mentors.org Britner, P. A., Balcazar, F. E., Blechman, E. A., Blinn-Pike, L., & Larose, S. (2006). Mentoring special youth populations. Journal of Community Psychology, 34, 747–763. Bronfenbrenner, U., & Crouter, A. C. (1983). The evolution of environmental models in developmental research. In W. Kessen (Ed.), Handbook of child psychology. Vol. 1: History, theory, and methods (4th ed., pp. 357–414). New York: Wiley. Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press. Cavell, T. A., & Henrie, J. L. (2010). Deconstructing serendipity: Focus, purpose, and authorship in lunch buddy mentoring. New Directions for Youth Development, 2010 126, 107–121. Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, S95–S120. Degenne, A., & Lebeaux, M. (2005). The dynamics of personal networks at the time of entry into adult life. Social Networks, 27, 337–358. Deutsch, N. L., & Spencer, R. (2009). Capturing the magic: Assessing the quality of youth mentoring

relationships. New Directions for Youth Development, 121, 47–70. Downey, G., Lebolt, A., Rincon, C., & Freitas, A. L. (1998). Rejection sensitivity and children’s interpersonal difficulties. Child Development, 69, 1074–1091. DuBois, D. L., Doolittle, F., Yates, B. T., Silverthorn, N., & Tebes, J. K. (2006). Research methodology and youth mentoring. Journal of Community Psychology, 34, 657–676. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Silverthorn, N. (2005). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92. Freedman, M. (1993). The kindness of strangers: Adult mentors, urban youth, and the new voluntarism. San Francisco: Jossey-Bass. Furstenberg, F. F., Jr., & Hughes, M. E. (1995). Social capital and successful development among at-risk youth. Journal of Marriage and the Family, 57, 580–592. Gilligan, R. (1999). Enhancing the resilience of children and young people in public care by mentoring their talents and interests. Child and Family Social Work, 4, 187–196. Gjerde, P. F. (1986). The interpersonal structure of family interaction settings: Parent-adolescent relations in dyads and triads. Developmental Psychology, 22, 297–304. Goldner, L., & Mayseless, O. (2008). Juggling the roles of parents, therapists, friends and teachers—A working model for an integrative conception of mentoring. Mentoring & Tutoring: Partnership in Learning, 16, 412–428. Goldner, L., & Mayseless, O. (2009). The quality of mentoring relationships and mentoring success. Journal of Youth and Adolescence, 38, 1339–1350. Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78, 1360–1380. Hamilton, S. F., & Hamilton, M. A. (2004). Contexts for mentoring: Adolescent-adult relationships in workplaces and communities. In R. M. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology (2nd ed., pp. 395–428). New York: Wiley. Hamilton, S. F., Hamilton, M. A., Hirsch, B. J., Hughes, J. N., King, J., & Maton, K. (2006). Community contexts for mentoring. Journal of Community Psychology, 34, 727–746. Higgins, M. C., & Kram, K. E., (2001). Reconceptualizing mentoring at work: A developmental network perspective. Academy of Management Review, 26, 264–288. Hirsch, B. J. (2005). A place to call home: After-school programs for urban youth. Washington, DC: American Psychological Association. Holt, L. J., Bry, B. H., & Johnson, V. L. (2008). Enhancing school engagement in at-risk, urban

Social Networks and Mentoring   141 minority adolescents through a school-based, adult mentoring intervention. Child & Family Behavior Therapy, 30, 297–318. Jarrett, R. L., Sullivan, P. J., & Watkins, N. D. (2005). Developing social capital through participation in organized youth programs: Qualitative insights from three programs. Journal of Community Psychology, 33, 41–55. Jent, J. F., & Niec, L. N. (2006). Mentoring youth with psychiatric disorders: The impact on child and parent functioning. Child & Family Behavior Therapy, 28, 43–58. Karcher, M. (2005). The effects of developmental mentoring on high school mentors’ attendance and on their younger mentees’ self-esteem, social skills, and connectedness. Psychology in the Schools, 42, 65–77. Karcher, M. J., Davis, C., & Powell, B. (2002). The effects of developmental mentoring on connectedness and academic achievement. School Community Journal, 12, 36–50. Karcher, M., & Spencer, R. (in press). The Study of Mentoring in the Learning Environment (SMILE): Multilevel modeling of setting-level program staff contributions to school-based mentoring program effectiveness. Keller, T. E. (2005a). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Keller, T. E. (2005b). A systemic model of the youth mentoring intervention. Journal of Primary Prevention, 26, 169–188. Keller, T. E. (2007a). Theoretical approaches and methodological issues involving youth mentoring relationships. In T. D. Allen & L. T. Eby (Eds.), Blackwell handbook of mentoring: A multiple perspectives approach (pp. 23–47). Malden, MA: Blackwell. Keller, T. E. (2007b). Program staff in youth mentoring programs: Qualifications, training, and retention. In J. E. Rhodes (Ed.), Research in Action Series (Issue 3, pp. 3–11). Alexandria, VA: MENTOR. Klaw, E. L., & Rhodes, J. E. (1995). Mentor relationships and the career development of pregnant and parenting African-American teenagers. Psychology of Women Quarterly, 19, 551–562. Linnehan, F. (2001). The relation of a work-based mentoring program to the academic performance and behavior of African American students. Journal of Vocational Behavior, 59, 310–325. Linnehan, F. (2003). A longitudinal study of work-based, adult-youth mentoring. Journal of Vocational Behavior, 63, 40–54. McDonald, S., Erickson, L. D., Johnson, M. K., & Elder, G. H. (2007). Informal mentoring and young adult employment. Social Science Research, 36, 1328–1347. Meissen, G. J., & Lounsbury, J. W. (1981). A comparison of expectations of volunteers, children, and parents in a Big Brother Big Sister program. Journal of Community Psychology, 9, 250–256.

Moody, J., & White, D. R. (2003). Structural cohesion and embeddedness: A hierarchical concept of social groups. American Sociological Review, 68, 103–127. Morgan, D., Neal, M., & Carder, P. (1997). The stability of core and peripheral networks over time. Social Networks, 9, 9–25. Morrow, K. V., & Styles, M. B. (1995). Building relationships with youth in program settings. Philadelphia: Public/Private Ventures. Munson, M. R., & McMillen, J. C. (2006). Nonkin natural mentors in the lives of older youth in foster care. Journal of Behavioral Health Services & Research, 35, 454–468. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E., Ebert, L., & Fischer, K. (1992). Natural mentors: An overlooked resource in the social networks of young, African American mothers. American Journal of Community Psychology, 20, 445–461. Rhodes, J. E., Grossman, J. B., & Reddy, R. (2005). The protective influence of mentoring on adolescents’ substance use: Direct and indirect pathways. Applied Developmental Science, 9, 31–47. Rhodes, J. E., Grossman, J. B., & Resch, N. L. (2000). Agents of change: Pathways through which mentoring relationships influence adolescents’ academic adjustment. Child Development, 71, 1662–1671. Rhodes, J. E., Spencer, R., Keller, T. E., Liang, B., & Noam, G. (2006). A model for the influence of mentoring relationships on youth development. Journal of Community Psychology, 34, 691–707. Sánchez, B., Esparza, P., Berardi, L., & Pryce, J. (2011). Mentoring in the context of Latino youth’s broader village during transition from high school. Youth & Society, 43, 225–252. Sánchez, B., & Reyes, O. (1999). Descriptive profile of the mentorship relationships of Latino adolescents. Journal of Community Psychology, 27, 299–302. Schwartz, S. E. O., Rhodes, J. E., Chan, C. S., & Herrera, C. (2011). The impact of school-based mentoring on youths with different relational profiles. Develop­ mental Psychology, 47, 450–462. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Spencer, R., Basualdo-Delmonico, A., & Lewis, T. O. (2010). Working to make it work: The role of parents in the youth mentoring process. Journal of Community Psychology, 39, 1, 51–59. Stanton-Salazar, R. D., & Spina, S. U. (2008). Informal mentors and role models in the lives of urban Mexicanorigin adolescents. Anthropology & Education Quarterly, 34, 231–254. Styles, M. B., & Morrow, K. V. (1992). Understanding how youth and elders form relationships: A study of four Linking Lifetimes programs. Philadelphia: Public/Private Ventures. Taylor, R. D. (1996). Adolescents’ perceptions of kinship support and family management practices: Association

142  MENTORING RELATIONSHIPS with adolescent adjustment in African American families. Developmental Psychology, 32, 687–695. Thomson, N. R., & Zand, D. H. (2010). Mentees’ perceptions of their interpersonal relationships: The role of the mentor-youth bond. Youth and Society, 41, 434–445. Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications. Cambridge, U.K: Cambridge University Press. Wellman, B. (1983). Network analysis: Some basic principles. Sociological Theory, 1, 155–200. Westman, M. (2006). Crossover of stress and strain in the work-family context. In F. Jones, R. J. Burke, & M. Westman (Eds.), Work-life balance: A psychological perspective (pp. 163–184). New York: Psychology Press.

Wheeler, M., Keller, T. E., & DuBois, D. L. (2010). Reviewing three recent randomized trials of schoolbased mentoring: Making sense of conflicting results. Social Policy Report, 24(3), 1–27. Wilson, W. J. (1987). The truly disadvantaged: The inner city, the underclass, and public policy. Chicago: University of Chicago Press. Zand, D. H., Thomson, N., Cervantes, R., Espiritu, R., Klagholz, D., LeBlanc, L., & Taylor, A. S. (2009). The mentor-youth alliance: The role of mentoring relationships in promoting youth competence. Journal of Adolescence, 32, 1–17. Zippay, A. (1995). Expanding employment skills and social networks among teen mothers: Case study of a mentor program. Child & Adolescent Social Work Journal, 12, 51–69.

PART IV Cultural Perspectives

10  RACE, ETHNICITY, AND CULTURE IN MENTORING RELATIONSHIPS Bernadette Sánchez, Yarí Colón-Torres, Rachel Feuer, Katrina E. Roundfield, and Luciano Berardi

Introduction I’m Mexican . . . but I actually feel that I would rather had a White mentor. Uh, I don’t know that’s just my internal racism [laughs]. Uh, but that’s, I mean that’s just how I feel. Like I feel like I would have better been served by a White mentor . . . when I think of professional, I just automatically picture someone who’s White. Uh, I guess that’s what I see when I think of professional, someone who’s White, which I probably shouldn’t cause I’m Mexican. This statement was made by a young highschool-aged teen in the United States who was matched with a mentor of Mexican descent in a mentoring program geared toward preparing lowincome youth for the college application process and the transition to college. His honesty about his own internalized racism shows the importance of considering the role of race, ethnicity, and culture in youth mentoring. This need is underscored by the increasingly diverse groups that are the focus of youth mentoring programs and initiatives both within the United States, such as youth from different immigrant groups (see Birman & Moreland, this volume, Chapter 24), and throughout the world (see Goldner & Scharf, this volume, Chapter 13). The first question typically asked by researchers and practitioners regarding the topic of race, ethnicity, and culture is, “Should youth and mentors be matched based on race?” We reviewed the literature related to this topic in our chapter for the first edition of this handbook (Sánchez & Colón, 2005)

and, like others (e.g., Liang & West, 2007), we concluded that rather than continue to focus on this question (for which simple or definitive answers have proved elusive) there is a need for a much deeper examination of issues of race, ethnicity, and culture within youth mentoring, such as those suggested by the quote that started this chapter. In pursuing this direction for the current chapter, we review theories (i.e., oppression, racial/ethnic identity, cultural competence) and related research that show promise for illuminating racial, ethnic, and cultural processes taking place within the youth, mentor, or program and that we did not cover in our original chapter. We also consider applied implications and provide a checklist for practitioners to consider in their programming. Before delving into the role of race, ethnicity, and culture in youth mentoring, it is important to understand differences between these terms, which often are used interchangeably. Race does not have a consensual theoretical or scientific definition (Helms, Jernigan, & Mascher, 2005). It is a social construction and it is not biologically determined, but individuals assign others to racial categories based on physical characteristics and make generalizations and stereotypes as a result (American Psychological Association [APA], 2003). Race has been used to maintain social hierarchies in societies (APA, 2003). Culture encompasses the beliefs, values, and patterned ways of thinking, acting, and feeling that are learned by individuals and transmitted to others (APA, 2003; Smedley & Smedley, 2005). Ethnicity refers to groups of people who share common cultural traits, such as language, history, 145

146  Cultural Perspectives religion, ancestry, traditions, and beliefs (Smedley & Smedley, 2005). In some contexts, ethnicity also may be about linking one’s ancestry to a particular tribal group. A limitation of studies of race is that even though researchers are aware that race is a social construction, they typically compare racial groups on a variety of behaviors and assume that any differences between the groups are due to cultural variables (e.g., values), ethnic membership, or biological characteristics (Helms et al., 2005). This ignores the complexity of race, ethnicity, and culture. We caution mentoring researchers not to make this same mistake, and, as a result, we do not review the differential effects of mentoring across racial/ ethnic groups. If we review this literature and find that mentoring is more effective for certain racial/ ethnic groups, then it would be difficult to explain this finding unless a racial, ethnic, or cultural process is measured. It is important to be attentive to the complexity of these constructs. For example, a mentor and a youth might be perceived as being of the same race, such as Black, but they might be of different ethnicities, such as African American, Kenyan, or Puerto Rican. The mentor, youth, and/or program staff might make assumptions about the similarities between them because of the perception that they share the same race, but their cultural values and traditions might be very different. These assumptions could impact the development of rapport and trust in the relationship and, in a program context, the effectiveness with which it is supported. As we move forward in this chapter, we cannot overstate the importance of keeping the preceding considerations in mind. We would like to underscore as well that the primary focus of our chapter is the U.S. context. We do so both because this is the context within which most of the studies reviewed here were conducted and because cultural issues in mentoring are considered from a more global perspective in Chapter 13 in this volume (see Goldner & Scharf). Integration of cultural models of mentors across these differing contexts represents an exciting frontier for the field, but not one that can be addressed in the context of this chapter.

Theory In this section, we consider four theories that are helpful in understanding how race, ethnicity, and culture may have implications for how mentoring relationships influence youth. The theories are the

similarity-attraction paradigm, oppression, ethnic and racial identity, and cultural competence. Our original chapter (Sánchez & Colón, 2005) covered three other theories, stereotype threat, cultural values, and Ogbu’s framework (1990) about ethnic minority youth, which we do not include here. We selected these theories because we view them as useful frameworks both for guiding understanding of existing research and charting directions for new research and for advancing consideration of issues of cultural competence in mentoring, an issue of particular interest to many practitioners. Racial, Ethnic, and Cultural Similarity or Dissimilarity The similarity-attraction paradigm proposes that individuals are attracted to those similar to themselves (Byrne, 1971). Although most research on the similarity-attraction theory focuses on similarity in attitudes and personality, the theory also suggests that, given the opportunity to choose their own mentors, youth would more often select mentors of the same race, ethnicity, or culture. Rhodes (2005) posits in her model of youth mentoring that mentors are more effective when they have a vested interest in the development of their mentees. Yet, the similarity-attraction paradigm suggests that some individuals may feel less of an interest in others who appear different based on race/ethnicity. From this perspective, relationships between mentors and mentees of similar racial, ethnic, or cultural backgrounds, who likely have more in common, could be expected to be more successful than those involving mentors and mentees of different backgrounds. With respect to therapeutic relationships, a meta-analysis of 10 studies that included African American and White clinicians and clients failed to find evidence that racial/ethnic similarity had an overall effect on retention in treatment, number of sessions attended, or psychological functioning at the end of treatment (Shin et al., 2005). Another meta-analysis (Maramba & Hall, 2002) of seven studies of racial/ethnic matching in therapy with more diverse samples, including African Americans, Latinos, Asian Americans, and Whites, found that clients matched with clinicians of the same race/ ethnicity were less likely to drop out after the first therapy session and that they attended more therapy sessions than those matched with a clinician of a different race/ethnicity. However, the effect sizes were small and there was no difference in psychological functioning between clients in racially similar relationships and those in racially different relationships. Considered with respect to youth

Race, Ethnicity, and Culture in Mentoring Relationships    147 mentoring, these findings suggest that the similarity or dissimilarity of the mentor and the youth along the dimensions of race or ethnicity should not be expected to be a robust predictor of relationship quality or youth outcomes. Looking beyond race/ ethnicity and at the cultural nuances and processes taking place in the relationships might be more important for understanding effective youth mentoring. The similarity-attraction paradigm itself, for example, calls attention to the potential importance of similarity of youth and mentor with regard to interests or personality. This would be consistent with research supporting the role of perceived similarity in both interpersonal attraction (Montoya, Horton, & Kirchner, 2008) and processes of social influence (Burger, Messian, Patel, del Prado, & Anderson, 2004) as well as the finding in a recent meta-analysis of youth mentoring program evaluations that matching youth and mentors on the basis of shared interests was associated with greater effectiveness in terms of youth outcomes (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). Oppression Oppression stems from an asymmetry in the distribution of resources between groups (Prilleltensky, 2003). It operates by systematically denying access to opportunities and resources to certain groups within a society. One source of oppression is discrimination (Prilleltensky, 2003). Research indicates that low-income, urban adolescents of color experience racial discrimination from adults and that the rate of this discrimination increases with age (Greene, Way, & Pahl, 2006). Such experiences may lead youth of color to feel a sense of cultural mistrust toward adults of a different race/ethnicity, which in turn could serve as a barrier to developing mentoring relationships with them either in a program context or in their own social networks. Cultural mistrust has been described as a pervasive attitude that racial minority individuals may have toward European Americans as a result of historical discrimination and oppression in the United States (Grant-Thompson & Atkinson, 1997). For example, Albertini (2004) found that 35%–50% of a sample of 359 West Indian and Haitian adolescents reported moderate to high levels of cultural mistrust toward Whites. Further, when participants were in the 9th grade, the more time they had spent in the United States, the more cultural mistrust they reported. Thus, it might be that over time many youth of color experience sufficient discrimination in the United States to develop a sense of mistrust toward Whites. It is also possible that youth of color

harbor cultural mistrust toward groups other than Whites (e.g., groups who are a different race/ethnicity) if those groups have been a source of discrimination in their lives. Racial and Ethnic Identity The terms racial identity and ethnic identity are often used interchangeably given the confusion about the meanings of race and ethnicity, but there are distinctions between the two. Because racial categories are sociopolitical constructs and individuals’ racial socialization experiences vary depending on whether one’s group is advantaged or disadvantaged, racial identity is described as individuals’ internalization of their racial socialization (e.g., discrimination, what it means to be Black; Helms, 2007; Sellers, Smith, Shelton, Rowley, & Chavous, 1998). Ethnic identity refers to a sense of belonging to a cultural group and participation in its cultural practices, regardless of one’s specific racial group (Helms, 2007). Research indicates that racial and ethnic identity play a significant role in minority youths’ development. In a comprehensive review of research on ethnic identity and adolescent well-being, Wakefield and Hudley (2007) found that a stronger ethnic identity was related to more positive academic, psychological, and social outcomes among ethnic minority youth. Similarly, a meta-analysis (HewattGrant, 2009) revealed that across childhood and adolescence, a stronger ethnic identity was associated with more positive psychological functioning as well as less personal dysfunction, such as depression and substance abuse. In Rhodes’s (2005) model, mentoring can promote positive youth outcomes through a range of processes, including identity development. She argues that youths’ internalizations of themselves may change as they identify with their mentors and see them as role models. Mentors’ views of mentees also may become incorporated into youths’ identities (Rhodes, 2005). Further, the social opportunities that mentors provide may facilitate identity development by affording experiences on which youth can draw to create their sense of self (Rhodes, 2005). With regard to racial/ethnic identity, mentors might be especially likely to affect youth of color in this area given the likely more salient role of racial/ethnic identity in their development (Wakefield & Hudley, 2007). The racial/ethnic identity of youth in same-race/ethnicity relationships could be especially likely to be affected because they identify with their mentors. Further, regardless of a mentor’s race, any racial attitudes (e.g., prejudices) that the mentor harbors could

148  Cultural Perspectives influence the youth’s racial/ethnic identity. These considerations suggest that mentoring relationships for youth can be important with respect to both mitigating the impact of and unwittingly reinforcing the negative effects of culturally involved themes of devaluation or oppression in the young person’s life. It is also possible that a youth’s racial/ethnic identity influences the youth’s view of his or her mentor(s). Internalized racism, furthermore, may lead to a weaker racial/ethnic identity and in turn influence the type of mentor that the youth prefers. Cultural Competence Cultural competence refers to the extent to which individuals have the capacity to effectively work with individuals of a cultural group (Sue, 2006). Sue (2006) described three general components of cultural competence: (a) cultural awareness and beliefs, (b) cultural knowledge, and (c) cultural skills. As applied to youth mentoring, the concept of cultural awareness would involve the extent to which mentors are cognizant of how their values and biases can play a role in their perceptions of mentees, including their problems and strengths, and in how they experience their relationships with mentees. Cultural knowledge would include learning about the mentee’s culture and worldviews as well as the mentee’s expectations for the relationship. Culturally specific skills involve the ability to apply cultural knowledge in relationships. Sue (2006) argued for three levels of analysis of cultural competence: provider, agency, and systems of care. The focus at the provider level is on those persons who work one-on-one with clients in a treatment or case management role, which in the case of mentoring programs would include both mentors and service delivery staff. Issues to consider would include interpersonal sensitivity and capacity for rapport building in interactions with youth and their families from varied cultural backgrounds (Sue, 2006). At the agency or organization level, the focus is on the extent to which the practices and other characteristics of the program or organization as a whole reflect a high level of cultural competence. Issues that might be examined include hiring practices, policies, and organizational climate and structures (Sue, 2006), all of which are relevant concerns for mentoring programs serving youth. Relevant considerations at the systems of care level would include, for example, the ways cultural competence is or is not reflected in the activities and practices of coalitions or intermediary organizations that seek to make youth mentoring available in a given community (Sue, 2006).

Given the commonalities between therapeutic and mentoring relationships (Spencer & Rhodes, 2005) and the posited need for mentoring relationships for youth to be based in feelings of mutuality, trust, and empathy (Rhodes, 2005), cultural competence may affect the very foundations of effective mentoring for young people. At the provider level, when a mentor possesses cultural awareness, specific cultural knowledge, and the skills to work sensitively with youth of a particular racial or ethnic group, the mentor may be more likely to develop a connection with the mentee that lays the groundwork for positive youth outcomes. The same would be expected for those staff charged with supporting the development of the relationship as well as the agency involved and the surrounding system of youth services and supports. Relevant to these possibilities, a meta-analysis of 76 evaluations found evidence that culturally competent mental health interventions (i.e., therapy) have a greater effect on treatment outcomes than traditional interventions (Griner & Smith, 2006). Further, therapeutic interventions that targeted a specific racial/ethnic group were substantially more effective than those that targeted a diverse group, and interventions in which the providers spoke the same language as clients were more effective than those interventions in which language matching was not discussed in the studies (Griner & Smith, 2006). The authors did not find differences in effectiveness related to whether interventions made other types of cultural adaptations (e.g., cultural sensitivity training, outreach to the community).

Research A variety of sources were used to identify research to be reviewed in this chapter. These included electronic databases such as PsycINFO, journals that publish youth mentoring research (e.g., Mentoring & Tutoring), reference sections of relevant articles, a listserv focused on youth mentoring and practice, and internet search engines. Our focus was on locating research that had direct relevance to the four theoretical perspectives discussed in the preceding section of this chapter. Although published studies conducted outside the United States were eligible for inclusion, our search yielded few such studies. Racial, Ethnic, and Cultural Similarity or Dissimilarity In accordance with the similarity-attraction paradigm, investigations of natural mentoring of African American, Asian American, Latino, and

Race, Ethnicity, and Culture in Mentoring Relationships    149 White youth have found youth most likely to identify mentors of the same race/ethnicity (Cavell, Meehan, Heffer, & Holladay, 2002; Haddad, Chen, & Greenberger, 2010; Klaw, Rhodes, & Fitzgerald, 2003; Sánchez, Esparza & Colón, 2008). Other explanations, however, are possible, such as youth simply having had greater levels of exposure to adults of the same race/ethnicity. At least half the youth in these studies, for example, identified kin as mentors. Research on role models similarly has found that African American and Latino youth tend to identify adults who are of the same race/ethnicity (Hurd, Zimmerman, & Xue, 2009; Stanton-Salazar & Spina, 2003). Research also has been conducted on youth and parent preferences regarding mentor race. A qualitative investigation of a mentoring program targeting mostly African American girls found evidence that participating girls preferred African American to White mentors when they were being paired (Schippers, 2008). In another qualitative study that gathered data from 13 racially/ethnically diverse parents whose children participated in a Big Brothers Big Sisters program (Spencer, BasualdoDelmonico, & Lewis, 2011), those parents of color who expressed a preference for their children to have a mentor of the same race thought that the similarity would facilitate relationship development because of shared cultural traditions and the mentor’s shared experience of being from an ethnic minority group. In support of this idea, based on findings of a qualitative study of a peer mentoring program for American Indian college students, the researchers concluded that having American Indian mentors facilitated relationship development because of their shared culture (Shotton, Oosahwe, & Cintrón, 2007). However, Spencer et al. (2011) reported additional themes of parents viewing the personal qualities (e.g., trustworthy, caring) of the mentor as more important than race and in some instances also feeling that cross-race mentoring would be beneficial because it would expose children to a different culture. In relevant experimental research (Hu, Thomas, & Lance, 2008), 126 White undergraduate students were randomly assigned to serve as a mentor or as a protégé in a hypothetical mentoring program. Participants were presented with 12 hypothetical profiles of either mentors or protégés who were White or Black, and they were asked the degree to which they would intend to begin a mentoring relationship with that person. Participants assigned the role of protégé were more likely to desire a samerace than a cross-race relationship, whereas there was no difference in the preferences of those acting as

mentors. These findings are similar to Schippers’s (2008) study of African American mentees. Hence, some evidence indicates that the race of the mentor may be more important to protégés than vice versa. As discussed earlier, the similarity-attraction paradigm (Byrne, 1971) suggests that mentors and youth of a similar race/ethnicity might have more successful relationships because of assumed commonalities between mentors and youth, which would lead to more positive youth outcomes. However, findings of research that compares whether being in a cross- versus same-race relationship are mixed and not suggestive of the importance of racial/ethnic similarity per se, as was noted in the previous edition of this chapter (Sánchez & Colón, 2005) and as continues to be reflected in recent research (Campbell & Campbell, 2007; Karcher, Roy-Carlson, & Benne, 2006; Taussig, Culhane, & Hettleman, 2007). In the meta-analysis referred to previously, programs that made it a practice to match youth with mentors of similar race or ethnicity did not demonstrate differential effectiveness overall (DuBois et al., 2011); furthermore, in the context of predictive models that also took into account other practices, same-race/ ethnicity matching was found be predictive of weaker program effects. In a noteworthy effort to examine the role of cross- versus same-race pairings, Ensher and Murphy (1997) used an experimental design to examine the effect of racial similarity on the quality of mentoring relationships. In this research, 104 youth aged 16–22 years in a summer internship program, 96% of whom were ethnic minority, were randomly assigned either to a mentor of the same race or to a mentor of a difference race. Most of the cross-race dyads were White mentors matched with ethnic minority youth (n = 33), but there were some crossrace relationships (n = 17) in which an ethnic minority mentor was matched with a different-race mentee. Mentors paired with same-race mentees reported liking their mentees more than mentors of different-race mentees. The investigators also found that male mentors with same-race youth reported a greater likelihood of continuing the relationship after the program ended than male mentors with cross-race youth, whereas this difference was not apparent for female mentors. When considering youth-report data, racial similarity was not predictive of youth satisfaction with the relationship or desire to continue the relationship after the program ended. Youth perceptions that their mentors were similar to them in outlook and values, however, were related to liking the mentor, relationship satisfaction, and the desire to continue the relationship. These findings are consistent with the possibility

150  Cultural Perspectives noted previously that similarity along other dimensions, such as interests and personality, might be more important than race for the quality of the relationship developed between a mentor and a youth mentee (see also Lederer, Basualdo-Delmonico, Spencer, & Wasserman, 2009). Oppression As mentioned earlier, youth of color experience racial discrimination from adults (Greene et al., 2006), which could have negative implications for whether or not they develop natural mentoring relationships, especially with adults who do not share their racial/ethnic background. Relevant to this possibility, Munson and McMillen’s (2006) study of foster care youth revealed that White adolescents were more likely than non-White youth to identify an unrelated adult mentor. A national study similarly found that African American and Latino youth were less likely than White youth to identify a grandparent, an uncle, or an aunt; a teacher or a coach; a neighbor; or an adult leader of a youth group as someone who cares about them (Girl Scout Research Institute, 2009). Although discrimination may well account in part for a relative lack of nonfamilial adults in African American and Latino youths’ lives, other explanatory factors are equally, if not more, plausible. It is possible, for example, that the familial adults of these youth are less available because of race-based stressors. Illustratively, another study using a national sample (Erickson, McDonald, & Elder, 2009) found that Black, Latino, and Native American youth from disadvantaged neighborhoods were less likely to report a natural mentor than were White and Asian youth from advantaged neighborhoods. Due to the confounding of youth race/ethnicity and neighborhood characteristics, it is unclear to what extent, if any, racial processes (e.g., racial discrimination) contributed to the differences found, independent of socioeconomic factors (even so, more macro-systemic issues relating to oppression are clearly germane when considering effects of socioeconomic disadvantage). As noted previously, the racial discrimination experienced by youth of color may cause some to feel a sense of cultural mistrust toward people outside their race/ethnicity. In a study designed to explore what role cultural mistrust might have within youth mentoring relationships, using a sample of 21 African American and Latina adolescent girls matched with White mentors, Sánchez, DuBois, Silverthorn, and Pryce (2012) found that greater youth-reported cultural mistrust of Whites was predictive of girls perceiving their mentors as

less culturally competent at 3 and 12 months as well as poorer quality relationships as rated by youth and mentors 1 year later. A study of 20 ethnically diverse, urban, low-income high school seniors and their mentors (Sánchez, Feuer, & Davis, 2012) similarly found that more reported cultural mistrust toward persons outside of one’s race/ethnicity was related to poorer quality mentoring relationships as rated by youth. Although preliminary, these studies suggest that cultural mistrust may serve as a barrier to developing quality relationships with mentors of a different race/ethnicity. Findings from a qualitative study of 14 African American middle school girls paired with a White or an African American female mentor point toward ways in which cultural mistrust might impede mentoring (Schippers, 2008). Mentors were students from a mostly middle- to upper-class college student body, whereas mentees came from poor neighborhoods; some mentors and mentees also differed on race. Some girls expressed disappointment when they found out that they were assigned to a White mentor rather than an African American, which the White mentors reported made them uncomfortable. Further themes included White mentors feeling that they were made fun of and/or ignored by the mentees and that the youth had less respect for them relative to the African American mentors. Schippers (2008) argued that the program was structured to make racial/class differences salient because it was designed to help African American girls who needed to be “fixed.” Specifically, the school counselor referred mentees to the program because of academic and/or behavioral problems. Perhaps these youth felt cultural mistrust toward the White mentors because they thought these mentors would try to change them or that they might not understand them. Racial/Ethnic Identity With regard to racial and ethnic identity, a study of 541 African American older adolescents transitioning into adulthood (Hurd, Sánchez, Zimmerman, & Caldwell, 2012) found that the report of having a natural mentor was related to more positive racial identity, including public regard (adolescents’ positive perceptions of how others view their race), private regard (how adolescents feel about their own race), and racial centrality (how central race is to adolescents’ identity). Another study (Yancey, Siegel, & McDaniel, 2002) similarly found that having a role model was related to a stronger ethnic identity among 744 ethnically diverse 12- to 17-year-old adolescents. Further, adolescents who knew their role models reported the most favorable ethnic identity.

Race, Ethnicity, and Culture in Mentoring Relationships    151 Consistent with theory reviewed previously that mentors’ behaviors also may be important in youths’ ethnic identities, Blash and Unger (1995) found that, among African American adolescent males, the number of adults in mentoring roles for the youth who were also active in the African American community was predictive of a stronger ethnic identity. It should be noted, however, that approximately half of the adults identified by youth were their parents; thus, the applicability of the study’s results to nonparental adult mentors is not entirely clear. In program evaluation research, an increase in ethnic identity was found among 34 Latina adolescents who were matched with Latina college student mentors by the end of the program (Kaplan, Turner, Piotrowski, & Silbert, 2009). However, because a comparison group was not included, it is hard to determine whether ethnic identity increased because of participants’ natural development or because of the program. In a more sophisticated research design, 40 adolescent girls who were mostly African American or Latina were randomly assigned to either a mentoring program with culturally tailored features or a traditional mentoring program (Pryce, DuBois, & Silverthorn, 2010). Although the findings were not statistically significant because of the small sample size, girls in the culturally tailored mentoring program had a stronger ethnic identity compared to those in the traditional program, and the effect was moderate (standardized mean difference effect size = .30, which generally would be considered substantial or noteworthy). Overall, these studies suggest that mentoring programs that have culturally tailored features may facilitate racial/ethnic identity development among adolescent youth of color. In an experimental study of 94 African American high school students (Linnehan, Weer, & Uhl, 2005), participants were provided a description of a hypothetical internship program at a chemical company where they were matched with a mentor. Participants were randomly assigned to receive either a White or a Black male mentor description. Analyses showed that participants with a weaker ethnic identity expressed a stronger belief that a White adult mentor would be competent and predictable than did students who had a stronger ethnic identity. Participants with a stronger ethnic identity also rated African American mentors as more competent than did those with a weaker ethnic identity. The researchers concluded that perhaps students with poor ethnic identities held negative stereotypes about African Americans and had more positive views toward Whites. In related research, Yancey et al. (2002) found that, compared to youth from other racial/ethnic groups, African American adolescents

both had the strongest reported ethnic identity and were more likely to identify ethnically similar role models. Because of the cross-sectional nature of these findings, they could be accounted for not only by youth with stronger ethnic identities preferring a same-race/ethnicity mentor, but also by a mentor of the same race/ethnicity contributing a stronger ethnic identity for the youth. Regardless, though, findings of the available studies do appear to implicate racial/ethnic identity as an important consideration within youth mentoring relationships. Cultural Competence In research pertinent to the role of cultural competence, Sánchez, DuBois, et al. (2012) found that among 40 young adolescent girls participating in an evaluation of a mentoring program, perceptions by youth that their mentors were culturally competent at 3 months were predictive of better quality relationships at 12 months as rated by both the youth and their mentors. Assessment of mentors’ cultural competence using the Scale of Ethnocultural Empathy (Wang et al., 2003), a self-report measure, at the beginning of the mentoring program was also related to better quality relationships at 12 months as reported by youth. In an evaluation study (Sánchez, Feuer, et al., 2012) of a mentoring program that targeted urban high school seniors applying to college, those who perceived their mentors to be more culturally competent reported better quality relationships. Because both measures were gathered at the same point in time, it is unclear whether cultural competence actually led to these better relationship outcomes. It is possible, for example, that mentors’ efforts in developing a closer relationship influenced the youth to perceive the mentor as culturally competent. In related qualitative research involving interviews of youth and adults in a mentoring program, findings revealed that one apparent reason why some relationships failed was that mentors were unable to bridge cultural differences (Spencer, 2007). Some mentors discussed perceived discrepancies between their own values and those of the youth and their families, but the mentors did not appear to be prepared to or know how to effectively deal with these differences. A further theme was a relative lack of awareness among mentors that misunderstandings taking place between them and their mentees seemingly stemmed from cultural differences or that they appeared to harbor prejudicial stereotypes that influenced their relationships. Conversely, based on qualitative research conducted with 12 Big Brothers Big Sisters matches, 10 of

152  Cultural Perspectives which were cross-race, it was concluded that mentors’ efforts in getting to know youths’ cultures may increase relationship quality (Lederer et al., 2009). In one subset of matches, categorized as Bridgers, mentors and youth reported that they grew closer because of cultural differences. Mentors in these matches made efforts to get to know their mentees and their families, and they were sensitive to and respectful of cultural differences. Deliberate sharing of cultural experiences and discussion of cultural differences within mentoring relationships also appeared to enhance the quality of these relationships. Roysircar, Gard, Hubbell, and Ortega (2005) observed changes in the cultural competence of 67 graduate students in a multicultural counseling course. Students provided one-on-one mentoring and tutoring to ethnically diverse adolescents. Based on Sue’s (2006) cultural competence framework, mentors were asked to write process notes about their meetings with mentees, including mentors’ biases and assumptions and how they relate to their interactions. The investigators found that instances of multicultural competence (e.g., being introspective about the learning challenges of mentees and about the role of their cultural values, racial attitudes, and various demographic characteristics in their dynamics with mentees) increased over time. This study provides evidence that the cultural competence of mentors should not be viewed as static or unchangeable. It is unclear, however, to what extent mentoring itself, the multicultural counseling course, self-reflection, or a combination of the three were responsible for the changes observed in mentors’ cultural competence. Hanlon, Simon, O’Grady, Carswell, and Callaman (2009) evaluated a group-mentoring program designed to be culturally tailored that served 237 6th-grade low-income, urban, African American students at a school that was randomly selected to be the intervention site. A demographically similar school was randomly selected to be the site in which comparison students were recruited. Youth who participated in the mentoring program had greater grade-point-average increases during the academic year compared to the 241 youth who did not participate. Further, teacher ratings of mentored youths’ cognitive problems decreased, compared to those of nonmentored youth. An evaluation of a culturally tailored program for Native Hawaiian students that included community mentors along with fieldwork with teachers and contextualized instruction similarly reported better educational outcomes for mentored youth than for nonparticipants (Yamauchi, 2003). The previously noted evaluation of a mentoring program for mostly African

American and Latina adolescent girls with culturally tailored components also found evidence of better mentoring relationship quality and youth outcomes (e.g., self-esteem, academic aspirations) for participants relative to those in the traditional mentoring program. It is unclear, however, to what extent the cultural components of these programs specifically contributed to observed findings. An important next step will be to conduct research that better isolates the potential contributions of cultural program components, such as through comparisons of mentoring relationships or youth outcomes across program conditions that do or do not include the cultural components of interest. With further relevance to the cultural competence of youth mentoring programs, a recent systemic review of studies of 22 mentoring programs in New Zealand concluded that most of the programs were highly culturally appropriate for the New Zealand context, but that they were less culturally appropriate when specifically targeting or including at least 15% Māori or Pasifika youth (Farruggia, Bullen, Dunphy, Solomon, & Collins, 2010). Half of the 17 programs with Māori youth ignored their culture, for example, and 21% of the programs were rated as highly culturally appropriate. Interestingly, the authors also found evidence of a trend for programs that were more culturally relevant to Māori culture to be less effective with regard to promoting desired youth outcomes. They speculated that perhaps in an effort to enhance the cultural relevance of mentoring programs for this population, fulfilling program goals and utilizing best practices might have been compromised or vice versa. Summary Much of the research on the role of race, ethnicity, and culture in mentoring is correlational and cross-sectional and based on small sample sizes. More studies with experimental and longitudinal designs are needed in order to clarify both the magnitude and the direction of potential causal associations involving racial and cultural processes (i.e., racial discrimination, racial/ethnic identity, cultural mistrust), mentoring relationship quality and youth outcomes, and, in particular, the extent to which such processes lead to closer mentor-youth bonds and more positive youth development. Further, researchers should examine how racial and cultural processes are related to youth mentoring (i.e., conduct studies testing meditational pathways of influence). For instance, does racial discrimination lead to more cultural mistrust, which then leads to poorer relationship quality? A further salient gap in

Race, Ethnicity, and Culture in Mentoring Relationships    153 the literature is that most of the research has focused on low-income, African American, Latino, and White youth in the United States. Relatively little attention has been given to other cultural groups, such as Asian Americans and Native Americans, or to youth from other geographical regions, such as rural and suburban areas or those outside the United States. Lastly, research is needed that more rigorously assesses the role of cultural competence in mentoring program effectiveness. For example, do cultural competence training programs enhance mentors’ and staff members’ cultural competence, and do the gains increase mentoring relationship quality?

field regarding how race/ethnicity should be considered. Some of the most important resources in the field, such as Elements of Effective Practice (MENTOR, 2009a), pay little attention to the role of race and ethnicity in mentoring programs. In the following sections, we offer recommendations to help address this gap, using the theoretical perspectives introduced earlier in this chapter as an organizing framework. Our recommendations are summarized in Table 10.1. We emphasize, however, that these recommendations are of necessity preliminary and largely untested, given the current state of the empirical literature in this area. Racial Similarity/Dissimilarity

Practice Despite evidence that race and ethnicity play an important role in mentoring relationships, there are limited research-based guidelines in the practice

Based on available research and consistent with current practices (e.g., Sawrikar, Griffiths, & Muir, 2008), we suggest that programs ask youth and their parents their preference with regard to mentors’ race/ethnicity when making matches. Regardless of mentor and youth race/ethnicity,

Table 10.1   Checklist for Practitioners Topic

Recommendations

Racial similarity/ dissimilarity

Consider the preferences of mentors, youth, and parents regarding matching based on race, ethnicity, and culture. Assist mentors and youth in finding similarities on other dimensions (e.g., shared interests). Consider providing same-race/ethnicity mentors to youth who have few same-race/ ethnicity role models or who might have internalized racism. Consider providing cross-race/ethnicity mentors to youth who have limited exposure to different cultures.

Oppression

Consider assessing the level of cultural mistrust in youth using instruments, such as the Cultural Mistrust Inventory for Adolescents (Sánchez & DuBois, 2006a), and using the data to tailor program support accordingly. Provide youth and mentors with safe opportunities to explore experiences such as prejudice and discrimination and feelings toward members of other cultural groups.

Ethnic identity

Assess youth ethnic or racial identity using scales such as the Multigroup Ethnic Identity Measure (Phinney, 1992) or Multidimensional Inventory of Black IdentityTeen (Scottham, Sellers, & Nguyen, 2008). Use this information to help inform whether your program is promoting healthy ethnic/racial identity among participating youth. Consider the potential benefits of matching youth with weaker ethnic identities with mentors with stronger ethnic identities to help catalyze the youth’s development of a positive identity.

(Continued)

154  Cultural Perspectives Table 10.1  (Continued) Topic

Recommendations Integrate into programs opportunities for youth to explore their racial or ethnic heritage with the support of their mentors and staff.

Cultural competence

Critically examine the cultural competence of mentors, the program/organization (e.g., staff, policies), and the network of institutions that support it and develop strategic plans for addressing areas of concern. Ensure that the program includes staff who share the same culture and language as the youth being served. Equally important, make a concerted effort to ensure that mentors recruited for the program include those whose cultural backgrounds are similar to those of youth. Consider providing training programs for mentors and staff that are organized around the components of Sue’s (2006) cultural competence framework (cultural awareness and beliefs, cultural knowledge, and cultural skills) and that include a discussion of the social construction of race. Consider arranging for consultants (e.g., tribal elders) to assist with enhancing the cultural competence of youth mentoring programs and organizations. Assess mentors’ cultural competence using the Ethnocultural Empathy Scale (Wang et al., 2003) and use the information gained to inform training of mentors and matching with youth. Assess youth perceptions of their mentors’ cultural sensitivity using the Cultural Sensitivity Scale–Mentee Report (Sánchez & DuBois, 2006b); use information gathered with this tool both to guide supervision and training of mentors and to inform evaluation as it relates to cultural competence of the program.

however, attention should also be given to helping them find similarities along other dimensions to facilitate the development of their relationships. Program staff, for example, can help mentors and youth find shared interests, and they can facilitate activities in which mentors and youth share information about their backgrounds and likes/dislikes. We also advise that program staff and mentors be supported in taking a strengths-based approach that focuses on the positive qualities in youth instead of the problems they may be experiencing, particularly among youth of color, who may have high levels of cultural mistrust. Although research fails to support a general emphasis on making either same- or cross-race matches, we believe that in some cases matching by race/ethnicity is likely to be helpful. Illustratively, youth who have few positive, same-race/ethnicity role models or who have internalized racism may benefit from a same-race match. Other youth, such as those who have not had much exposure to different cultures, may likewise be good candidates for a cross-race/ethnicity mentor. In these cases, culturally tailored support should be available to ensure

a productive match. For example, although youth with internalized racism may stand to potentially benefit from being paired with a same-race positive role model, they may initially have a negative reaction to being assigned a same-race mentor. Similarly, youth who have not had exposure to other cultures may need extra support in finding points of connection when paired with mentors with backgrounds different from their own. Oppression Because some youth may have experienced discrimination or oppression, we recommend that attention be given to assessing the youth’s familiarity and comfort with people of other races/ethnicities. The Cultural Mistrust Inventory for Adolescents (Sánchez & DuBois, 2006a) is one tool that can be used for this purpose. The information gathered then can be used to help create a successful match and to provide appropriately tailored support for the relationship. Youth with cultural mistrust should be given the opportunity to explore their cultural mistrust in a safe setting with program staff.

Race, Ethnicity, and Culture in Mentoring Relationships    155 Ethnic and Racial Identity The research reviewed previously suggests that mentoring programs can be used to promote a positive ethnic or racial identity among minority youth and that a positive ethnic or racial identity can facilitate other positive psychological, academic, and social outcomes. To this end, we recommend that opportunities for youth of color to explore their ethnic and racial identity with the support of their mentors be integrated into programs. Further, although the potential exists for a minority youth with a relatively weak racial or ethnic identity to express a preference during the matching process for an otherrace mentor, it may be beneficial to match the youth with a mentor who does share his or her ethnic or racial background, ideally one who exhibits a welldeveloped racial or ethnic identity himself or herself. Cultural Competence Beyond matching and providing support for matches, programs can take many steps to make their programs more culturally competent. Practitioner resources are available that suggest how to address issues relating to race, ethnicity, and culture at the program and organization levels. The Elements of Effective Practice Toolkit (MENTOR, 2005), for example, recommends providing mentors with training and opportunities that will help them to learn about their mentees’ cultures and to be sensitive to cultural differences. Also of value may be conducting a needs assessment to better understand the target community, hiring staff and volunteers who share the same culture and language as the target population (Liang & West, 2007), and including parents and family members in the mentoring relationship (Farruggia et al., 2010; Sawrikar et al., 2008). We also strongly recommend that programs assess the cultural competence of their staff and mentors and make active use of this information for purposes such as tailoring of training programs to staff or volunteer needs. It is also advised that organizational policies and written materials be culturally tailored, that culturally relevant topics (e.g., current events affecting participants) be discussed at staff meetings, that ongoing cultural competence training and support be provided to staff and mentors, that community agencies serving the same population collaborate, that members of the target population be involved in advisory structures, and that outreach activities be tailored (e.g., Liang & West, 2007; see MENTOR, 2009b). In pursuing these strategies, care should be taken to ensure that mentoring programs for youth are appropriately adapted to the context of the local culture. For some

cultural groups and contexts (e.g., those with more collective values), a group-based mentoring model could be more suitable than a traditional one-on-one approach. (For further discussion of the role of cultural values in youth mentoring, see Goldner & Scharf, this volume, Chapter 13.) From an evaluation standpoint, youth perceptions of their mentors’ cultural sensitivity should be assessed to help gauge whether the attention being given to cultural competence in the preceding areas (e.g., selection and training of mentors) is sufficient to meet the program’s needs. As noted in Table 10.1, a tool that can be used for this purpose is the Cultural Sensitivity Scale–Mentee Report (Sánchez & DuBois, 2006b). This scale consists of six items asking the youth to rate the mentor’s attitudes and behaviors in areas relevant to cultural competence. Some practice resources also provide specific training activities for program staff and volunteers (e.g., Liang & West, 2007). Sample resources on how to make a mentoring program more culturally competent include the Australian National Youth Affairs Research Scheme, which specifically addresses mentoring immigrant youth from the Horn of Africa (Sawrikar et al., 2008); and the Chapel Hill/Carrboro City Schools’ Blue Ribbon Mentor-Advocate program, which has created a training manual for mentors based on Sue’s (2006) cultural competence framework (G. Meyer, personal communication, January 2010). Cultural competence is an ongoing process; thus, staff and mentors should work on it continually.

Conclusion In the past 5 years, increased attention has been paid to the role of race, ethnicity, and culture in youth mentoring research and practice, but more systematic research and evidence-informed practice is needed. It is imperative that researchers and practitioners look beyond the potential for differential effects of simply the racial/ethnic backgrounds of mentors and youth. The complexity of the concepts of race, ethnicity, and culture presents a formidable challenge in giving more nuanced and in-depth attention to the role of these concepts in youth mentoring. Yet multiple theoretical frameworks offer promise to guide such efforts, as do preliminary findings of research that have utility for informing the practice of youth mentoring. In particular, whereas research on racial similarity continues to reveal mixed findings, investigations addressing the role of factors associated with similarity along dimensions other than race or ethnicity, experiences

156  Cultural Perspectives of oppression, racial/ethnic identity development, and cultural competence as considered at multiple levels of analysis have begun to provide greater insight into the roles of race, ethnicity, and culture in mentoring and how they may best be addressed in practice. The finding that youth of color are less likely than White youth to identify natural mentors demonstrates that perhaps institutional and societal racism has left their communities with fewer adults to fill this role. However, these very youth, who arguably are particularly in need of mentors and may be matched with ones through programs, may for reasons that are fully understandable harbor feelings of mistrust toward adults who are not of their racial/ethnic group. Research along these lines is in its infancy, but it appears that cultural mistrust may sometimes inhibit the quality of the relationships that youth of color develop with adults to whom they are matched in programs. Research needs to examine the prevalence and significance of such dynamics and how best to prevent and alleviate negative repercussions of cultural mistrust in the context of youth mentoring. The evidence that racial/ethnic identity plays a role in youth perceptions of mentors and that mentoring may facilitate racial/ethnic identity development is also noteworthy. Given these findings and the linkages of racial and ethnic identity to a range of desirable youth outcomes, further attention to practices to promote racial/ethnic identity in mentoring programs is clearly warranted. Research designs, however, will need to do a better job of delineating the specific contributions that such practices may make to program outcomes. Cultural competence is often mentioned in the youth mentoring research and practice literatures as an important component of mentoring, but the idea lacks systematic research. Given the diverse populations served by mentoring programs, considering cultural competence at the mentor, agency, and network-of-agencies levels may improve the effectiveness of youth mentoring programs overall, a so-far elusive goal (DuBois et al., 2011). We look forward to the next generation of development for the field, in which researchers and practitioners make considerations of race, ethnicity, and culture a top priority.

References Albertini, V. L. (2004). Racial mistrust among immigrant minority students. Child and Adolescent Social Work Journal, 21, 311–331. American Psychological Association. (2003). Guidelines on multicultural education, training, research, practice

and organizational change for psychologists. American Psychologist, 58, 377–402. Blash, R., & Unger, D. (1995). Self-concept of AfricanAmerican male youth: Self-esteem and ethnic identity. Journal of Child and Family Studies, 4, 359–373. Burger, J. M., Messian, N., Patel, S., del Prado, A., & Anderson, C. (2004). What a coincidence! The effects of incidental similarity on compliance. Personality and Social Psychology Bulletin, 30, 35–43. Byrne, D. (1971). The attraction paradigm. New York: Academic Press. Campbell, T. A., & Campbell, D. E. (2007). Outcomes of mentoring at-risk college students: Gender and ethnic matching effects. Mentoring & Tutoring, 15, 135–148. Cavell, T. A., Meehan, B. T., Heffer, R. W., & Holladay, J. J. (2002). The natural mentors of adolescent children of alcoholics (COAs): Implications for preventive practices. Journal of Primary Prevention, 23, 23–42. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Ensher, A. E., & Murphy, S. E. (1997). Effects of race, gender, perceived similarity, and contact on mentor relationships. Journal of Vocational Behavior, 50, 460–481. Erickson, L. D., McDonald, S., & Elder, G. H. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367. Farruggia, S., Bullen, P., Dunphy, A., Solomon, F., & Collins, E. (2010, July). The effectiveness of youth mentoring programmes in New Zealand. Youth Mentoring Network. Retrieved from http://www .myd.govt.nz/policy-and-research/what-works-inyouth-development.html#MentoringandYoungPeopl eASystematicReview2 Girl Scout Research Institute. (2009). Good intentions: The beliefs and values of teens and tweens today. New York: Girl Scouts of the USA. Retrieved from http://www.girlscouts.org/research/pdf/good_inten tions_full_report.pdf Grant-Thompson, S. K., & Atkinson, D. R. (1997). Crosscultural mentor effectiveness and African American male students. Journal of Black Psychology, 23, 120–134. Greene, M. L., Way, N., & Pahl, K. (2006). Trajectories of perceived adult and peer discrimination among Black, Latino, and Asian American adolescents: Patterns and psychological correlates. Developmental Psychology, 42, 218–238. Griner, D., & Smith, T. B. (2006). Culturally adapted mental health intervention: A meta-analytic review. Psychotherapy: Theory, Research, Practice, Training, 43, 531–548. Haddad, E., Chen, C., & Greenberger, E. (2010). The role of important non-parental adults (VIPs) in the lives of older adolescents: A comparison of three ethnic

Race, Ethnicity, and Culture in Mentoring Relationships    157 groups. Journal of Youth and Adolescence, 40, 310–319. Hanlon, T. E., Simon, B. D., O’Grady, K. E., Carswell, S. B., & Callaman, J. M. (2009). The effectiveness of an after-school program targeting urban African American youth. Education and Urban Society, 42, 96–118. Helms, J. E. (2007). Some better practices for measuring racial and ethnic identity constructs. Journal of Counseling Psychology, 54, 235–246. Helms, J. E., Jernigan, M., & Mascher, J. (2005). The meaning of race in psychology and how to change it: A methodological perspective. American Psycho­ logist, 60, 27–36. Hewatt-Grant, J. (2009). The relationship between ethnic identity and psychological health: A meta-analytic review. Dissertation Abstracts International, 69 (12-B), 7810. Hu, C., Thomas, K. M., & Lance, C. E. (2008). Intentions to initiate mentoring relationships: Understanding the impact of race, proactivity, feelings of deprivation, and relationship roles. Journal of Social Psychology, 148, 727–744. Hurd, N. M., Sánchez, B., Zimmerman, M. A., & Caldwell, C. H. (2012). Natural mentors, racial identity, and educational attainment among African American adolescents: Exploring pathways to success. Child Development, 83, 1196–1212. Hurd, N. M., Zimmerman, M. A., & Xue, Y. (2009). Negative adult influences and the protective effects of role models: A study with urban adolescents. Journal of Youth and Adolescence, 38, 777–789. Kaplan, C. P., Turner, S. G., Piotrowski, C., & Silbert, E. (2009). Club Amigas: A promising response to the needs of adolescent Latinas. Child and Family Social Work, 2, 213–221. Karcher, M. J., Roy-Carlson, L., & Benne, K. (2006, June). The role of mentor-mentee ethnicity similarity in mentees’ connectedness to other cultures. Poster symposium presented at the Annual Meeting of the Society for Prevention Research, San Antonio, TX. Klaw, E. L., Rhodes, J. E., & Fitzgerald, L. F. (2003). Natural mentors in the lives of African American adolescents mothers: Tracking relationships over time. Journal of Youth and Adolescence, 32, 223–232. Lederer, J., Basualdo-Delmonico, A., Spencer, R., & Wasserman, R. (2009, November). Barriers, bumps and bridges: How cultural differences shape the youth mentoring process over time. Paper presented at the 137th American Public Health Association Annual Meeting, Philadelphia. Liang, B., & West, J. (2007). Youth mentoring: Do race and ethnicity really matter? In J. E. Rhodes (Ed.), Research in Action Series (Issue 9, pp. 3–13). Alexandria, VA: MENTOR. Retrieved from http:// www.mentoring.org/downloads/mentoring_390.pdf Linnehan, F., Weer, C., & Uhl, J. (2005). AfricanAmerican students’ early trust beliefs in workbased mentors. Journal of Vocational Behavior, 66, 501–515.

Maramba, G. G., & Hall, G. C. N. (2002). Meta-analyses of ethnic match as a predictor of dropout, utilization, and level of functioning. Cultural Diversity & Ethnic Minority Psychology, 8, 290–297. MENTOR. (2005). How to build a successful mentoring program using the elements of effective practice: A step-by-step tool kit for program managers. Retrieved from http://www.mentoring.org/find_ resources/tool_kit/ MENTOR. (2009a). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. Retrieved from http://www.mentoring.org/program_ resources/elements_and_toolkits MENTOR. (2009b). Immigrant and refugee youth: A toolkit for program coordinators. Retrieved from http://www .mentoring.org/find_resources/immigrant_toolkit/ Montoya, R. M., Horton, R. S., & Kirchner, J. (2008). Is actual similarity necessary for attraction? A metaanalysis of actual and perceived similarity. Journal of Social and Personal Relationships, 25, 889–922. Munson, M. R., & McMillen, J. C. (2006). Nonkin natural mentors in the lives of older youth in foster care. Journal of Behavioral Health Services & Research, 35, 454–468. Ogbu, J. U. (1990). Mentoring minority youth: A framework. New York: Columbia University, Teachers College, Institute for Urban and Minority Education. Phinney, J. S. (1992). The multigroup ethnic identity measure: A new scale for use with diverse groups. Journal of Adolescent Research, 7, 156-176. Prilleltensky, I. (2003). Understanding, resisting, and overcoming oppression: Toward psychopolitical validity. American Journal of Community Psychology, 31, 195–201. Pryce, J. M., DuBois, D. L., & Silverthorn, N. (2010). Beyond the dyad: A mixed method analysis of group experiences and outcomes in a girls mentoring program. In N. L. Deutsch (Chair), Strength in numbers: The potential of group mentoring as an intervention for adolescents. Symposium presented at the Biennial Meeting of the Society for Research on Adolescence, Philadelphia. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43). Thousand Oaks, CA: Sage. Roysircar, G., Gard, G., Hubbell, R., & Ortega, M. (2005). Development of counseling trainees’ multicultural awareness through mentoring English as a Second Language students. Journal of Multicultural Coun­ seling and Development, 33, 17–36. Sánchez, B., & Colón, Y. (2005). Race, ethnicity, and culture in mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 191–204). Thousand Oaks, CA: Sage. Sánchez, B., & DuBois, D. L. (2006a). Cultural Mistrust Inventory for Adolescents. Unpublished document. Sánchez, B., & DuBois, D. L. (2006b). Cultural Sensitivity Mentee Measure. Unpublished document.

158  Cultural Perspectives Sánchez, B., DuBois, D. L., Silverthorn, N., & Pryce, J. (2012). The role of cultural mistrust and cultural sensitivity in girls’ mentoring relationships with adult women. Manuscript in preparation. Sánchez, B., Esparza P., & Colón, Y. (2008). Natural mentoring under the microscope: An investigation of mentoring relationship and Latino adolescents’ academic performance. Journal of Community Psychology, 36, 468–482. Sánchez, B., Feuer, R., & Davis, K. E. (2012). Cultural mistrust and cultural sensitivity in high school seniors’ mentoring relationships. Manuscript in preparation. Sawrikar, P., Griffiths, M., & Muir, K. (2008). Culturally and linguistically diverse (CALD) young people and mentoring: The case of Horn of African young people in Australia. Australia: National Youth Affairs Research Scheme. Retrieved from http://youthmen toringvic.org.au/publications-and-research/116/ Schippers, M. (2008). Doing difference/doing power: Negotiations of race and gender in a mentoring program. Symbolic Interaction, 31, 77–98. Scottham, K.M., Sellers, R.M., & Nguyen, H.X. (2008) A measure of racial identity in African American adolescents: The development of the Multidimensional Inventory of Black Identity—Teen. Cultural Diversity and Ethnic Minority Psychology, 14, 297 -306. Sellers, R. M., Smith, M. A., Shelton, J. N., Rowley, S. A. J., & Chavous, T. M. (1998). Multidimensional model of racial identity: A reconceptualization of African American racial identity. Personality and Social Psychology Review, 2, 18–39. Shin, S., Chow, C., Camacho-Gonsalves, T., Levy, R., Allen, I. E., & Leff, H. S. (2005). A meta-analytic review of racial-ethnic matching for African American and Caucasian American clients and clinicians. Journal of Counseling Psychology, 52, 45–56. Shotton, H. J., Oosahwe, E. S. L., & Cintrón, R. (2007). Stories of success: Experiences of American Indian students in a peer-mentoring retention program. Review of Higher Education, 31, 81–107.

Smedley, A., & Smedley, B. D. (2005). Race as biology is fiction, racism as a social problem is real: Anthropological and historical perspectives on the social construction of race. American Psychologist, 60, 16–26. Spencer, R. (2007). “It’s not what I expected:” A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Spencer, R., Basualdo-Delmonico, A., & Lewis, T. O. (2011). Working to make it work: The role of parents in the youth mentoring process. Journal of Community Psychology, 39, 51–59. Spencer, R., & Rhodes, J. E. (2005). A counseling and psychotherapy perspective on mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 118–132). Thousand Oaks: Sage. Stanton-Salazar, R. D., & Spina, S. U. (2003). Informal mentors and role models in the lives of urban Mexican-origin adolescents. Anthropology & Education Quarterly, 34, 231–254. Sue, S. (2006). Cultural competency: From philosophy to research and practice. Journal of Community Psychology, 34, 237–245. Taussig, H. N., Culhane, S. E., & Hettleman, D. (2007). Fostering healthy futures: An innovative preventive intervention for preadolescent youth in out-of-home care. Child Welfare, 86, 113–131. Wakefield, W. D., & Hudley, C. (2007). Ethnic and racial identity and adolescent well-being. Theory Into Practice, 46, 147–154. Wang, Y., Davidson, M. M., Yakushko, O. F., Savoy, H. B., Tan, J. A., & Bleier, J. K. (2003). The scale of ethnocultural empathy: Development, validation, and reliability. Journal of Counseling Psychology, 50, 221–234. Yamauchi, L. A. (2003). Making school relevant for atrisk students: The Wai’anae High School Hawaiian Studies Program. Journal of Education for Students Placed at Risk, 8, 379–390. Yancey, A., Siegel, J., & McDaniel, K. (2002). Role models, ethnic identity, and health-risk behaviors in urban adolescents. Pediatric Adolescent Medicine, 156, 55–61.

11  GENDER IN MENTORING RELATIONSHIPS Belle Liang, G. Anne Bogat, and Nicole Duffy

Introduction Mentoring practices are most effective when they are sensitive to individual differences and contextual factors, such as gender (Darling, Bogat, Cavell, Murphy, & Sánchez, 2006). Unfortunately, little research has focused explicitly on the influence of gender on the process and outcomes of youth mentoring relationships, and the results of existing work have been equivocal (Bogat & Liang, 2005). Thus, many questions remain regarding the differences in boys’ and girls’ needs in mentoring, the implications of same- versus cross-sex matches, and whether gender differences hold across other individual difference variables (e.g., race, ethnicity). This chapter integrates theory and research in and out of the mentoring field to shed light on the following questions: (1) Do boys and girls have different needs in mentoring relationships? (2) Do same-sex and cross-sex mentoring relationships vary in their impact on youth? and (3) Do gender differences in mentoring hold across various identity characteristics (e.g., race, ethnicity)? This current chapter builds on Bogat and Liang’s (2005) chapter in the first edition of the Handbook of Youth Mentoring by including new literature and findings pertaining to questions 1 and 2 as well as by introducing question 3 as a new area for exploration. Using a structure similar to the earlier chapter, we first briefly review developmental theories and research on gender differences in relationships that are relevant to youth mentoring. Next, we review research addressing the effects of gender on the characteristics and outcomes of mentoring relationships for youth, along with the limitations of such research and recommendations for future study. We then consider implications of the theory and

research reviewed for practices relating to youth mentoring and, in doing so, offer several recommendations for practice.

Theory Although there is no agreed upon overarching theory within which to study “gender of youth” as a main effect (i.e., direct influence) and “gender of youth–gender of mentor” interactions (i.e., how the combination of youth and mentor genders may be influential) in mentoring, theories from social psychology, developmental psychology, and related fields can elucidate the potential influence of gender in mentoring for young people. In this chapter, we apply theories of relationship development, friendship patterns, and help-seeking to understanding how boys and girls may relate in mentoring relationships. We consider similarity and attraction theory as well as attachment theory with respect to same- versus cross-matched mentoring. Gender identity theory is explored in considering the relevance of gender across other identity characteristics. Question 1: Do Boys and Girls Have Different Needs in Mentoring Relationships? Relationship Development. Gender socialization patterns begin shaping relational behavior from early childhood (Miller & Stiver, 1997). Indeed, feminist theorists contend that Western socialization emphasizes connection to others as the hallmark of healthy psychological development for girls and women (Jordan, Kaplan, Miller, Stiver, & Surrey, 1991). That is, even as sociocultural influences foster autonomy in boys and men, these same influences 159

160  CULTURAL PERSPECTIVES encourage relatedness in girls and women (Cross & Madson, 1997). Although traditional developmental theory has conceptualized adolescence as a time of increased autonomy from parents, contemporary theorists assert that girls’ relationships in and out of the family continue to be important throughout life (e.g., Gilligan, Lyons, & Hammer, 1990). Indeed, gender socialization is thought to affect girls’ and boys’ relational trajectories differently. The “gender-intensification hypothesis” (Hill & Lynch, 1983) suggests that the adoption of gender-specific beliefs and behaviors increases during adolescence. In addition, socialization pressures to conform to traditional sex roles have an increasing hold during this period of development, resulting in a widening gap in girls’ and boys’ relational attitudes and behavior over time (Galambos, Almeida, & Petersen, 1990). Thus, girls, compared to boys, may exhibit less striving for autonomy in adolescence and may be even willing to forgo certain relational qualities associated with traditional masculine norms, such as power and self-assertion. Indeed, it has been suggested that, during adolescence, many girls reach an impasse in which they feel they must silence their own thoughts and feelings in order to preserve relationships (a concept termed “loss of voice”; Gilligan et al., 1990). Relatedly, higher levels of femininity, as compared to androgyny or masculinity, are associated with lower levels of voice in girls (Harter, Waters, Whitesell, & Kastelic, 1998). These sacrifices in voice, and ultimately well-being, represent the personal cost involved in girls’ attempts to preserve certain relationships. Indeed, inauthentic relationships (i.e., relationships in which one is not able to fully voice or express true feelings, thoughts, opinions, and other aspects of one’s personhood) have been linked with low self-esteem, hopelessness, and depressed mood in adolescence (Harter, Marold, Whitesell, & Cobbs, 1996). Relational-Cultural Theory (RCT). Theorists have further elaborated the qualities of significant relationships and how they evolve over the lifespan (Jordan et al., 1991). In particular, relationalcultural theory (RCT) scholars have gleaned from their observations of female relationships four major growth-fostering characteristics: (a) mutual engagement, (b) authenticity, (c) empowerment, and (d) the ability to deal with difference or conflict (Miller & Stiver, 1997). Even those scholars who argue that these qualities are relevant for both males and females (Baumeister & Leary, 1995) concede that the qualities are expressed differently across gender. For example, although male-male relationships are essential to the development of young

men, men often behave according to masculine norms specifically in order to preserve their relationships and to avoid being rejected (Chu, Porche, & Tolman, 2005). In other words, boys may manifest intimacy differently than do girls: Self-disclosure is often considered a sign of deep intimacy for girls, whereas engaging in activities may be the way that boys develop intimacy (Pollack, 1999). These differences clearly have implications for the ways in which male and female mentors and mentees may spend their time together and develop close mentoring relationships. Friendship Patterns. Relatedly, theories on friendship development suggest that girls value intimacy in friendships more than boys do (Buhrmester, 1990). Research suggests that the quality of a friendship is more likely to influence girls’ adjustment (Berndt & Keefe, 1995) and that girls expect higher levels of intimacy, self-disclosure, and empathy than do boys (Clark & Ayers, 1993). Although girls may derive greater benefits from positive relationships, they may also be more negatively affected by their relationships. Indeed, psychological distress and crises in the lives of girls and women often stem from disconnections in relationships (Hagerty, Williams, Coyne, & Early, 1996). Because mentoring ties for youth share several features similar to friendships, the stakes may also be particularly high in girls’ mentoring relationships because of a pronounced potential for such ties to do not only good, but also harm in the lives of girls. Help-Seeking. Girls and boys also differ with regard to help-seeking behaviors. For example, boys report more negative views of help-seeking than do girls (Garland & Zigler, 1994), and girls are more likely than boys to problem-solve relationship issues with a friend (Sullivan, Marshall, & Schonert-Reichl, 2002). Gender differences related to coping with stress also have been reported; in times of need, girls often seek out emotional support (Greenberger & McLaughlin, 1998) and are more likely than boys to engage in “co-rumination” (excessive talk about problems in the absence of active coping; Rose, 2002). Theories explaining help-seeking behavior among boys is limited. Addis and Mahalik (2003), however, suggest that help-seeking in men is constrained by masculine gender role socialization (e.g., self-reliance), which ostensibly is communicated to boys in early development. Boys’ help-seeking may be moderated by individual characteristics (e.g., how closely they ascribe to gender norms) and by characteristics of the person from whom they are seeking help (e.g., male versus female peers) (Sears,

Gender in Mentoring Relationships   161 Graham, & Campbell, 2009). For example, high school boys are more likely to seek help from female peers than from male peers, and boys who ascribe to the traditional masculine norm of emotional control are unlikely to seek help from either male or female peers in times of distress (Sears et al., 2009). It appears that those who do may be more likely to seek help from informal (e.g., friends) rather than formal supports (e.g., adults, trained professionals) and are thus more susceptible to seeking help from peers that is ultimately ineffective (Timlin-Scalera, Ponterotto, Blumberg, & Jackson, 2003). Taken together, these differences in the relational development, friendship patterns, and helpseeking behaviors of boys and girls may have bearing on what they need and want out of mentoring relationships. First, because of the high value that adolescent girls place on intimacy and connection, compared to boys, they might be more open to mentoring (Weisz, Sandler, Durlak, & Anton, 2005) and consequently benefit more from it (Herrera et al., 2007), even though trusting bonds may take a longer time to form. Second, the literature on helpseeking described above regarding girls’ tendency to seek out emotional support suggests that, especially in times of stress, they may value psychosocial mentoring. Psychosocial approaches emphasize psychological and social needs, or the interpersonal relationship that forms between the mentor and mentee, whereas instrumental mentoring emphasizes problem solving and practical skill building (Flaxman, Ascher, & Harrington, 1988). In line with this possibility, a study on help-seeking behavior demonstrated that whereas girls relied more on emotional support than did boys during times of need, they did not differ in levels of seeking instrumental support (Greenberger & McLaughlin, 1998). Because the literature cited thus far tends to paint a picture of girls, compared to boys, as potentially more prone to seeking and receiving psychosocial help in times of need, the emphasis that many mentoring programs place on intervening in the lives of youth who are at individual and environmental risk may be particularly effective for girls and may contribute to the success of such programs for this population. Indeed, it would be interesting to test whether there are gender differences in the effects of mentoring in periods of high stress: Would girls do better than boys because of helpseeking preferences or styles? As we discuss later in this chapter, in our review of research on youth mentoring, the hypothesis that girls may benefit (or be harmed) more than boys as a function of whether the program is of high or low quality also has received some support.

Question 2: Do Same-Sex and Cross-Sex Mentoring Relationships Vary in Their Impact on Youth? Attachment Theory. Attachment theory addresses the process by which formative relationships (i.e., early parenting or other caregiving relationships) may shape the nature of future significant attachments—individuals develop lasting “working models” based on formative attachments that they apply to future relationships (Bartholomew & Shaver, 1998). A healthy orientation to relationships is classified as a secure attachment style and is characterized by trusting and favorable perceptions of self and others. When gender differences in attachment with parental figures have been found, girls have demonstrated higher quality of this attachment than have boys (e.g., Kenny & Donaldson, 1991). Youth also tend to have a higher quality of attachment with mothers than fathers (Paterson, Field, & Pryor, 1994). Such research suggests that female mentees and mentors might tend to have greater “attachment” or relationship quality than do males. Issues germane to the connection between the attachment construct and mentoring have been examined in a number of studies (e.g., Soucy & Larose, 2000). Rhodes (2005) contends that the success or failure of mentoring relationships is affected by the attachment orientation of both the mentor and the mentee. For example, Rhodes, Contreras, and Mangelsdorf (1994) found that youth who experienced accepting maternal relationships early in life were more likely to develop mentoring relationships. The authors suggest that these youths’ secure attachments allowed them to trust nonfamilial adults and to form close relationships with them. Thus, because girls, compared to boys, may have higher levels of secure attachment to parents, it stands to reason that they might be more likely to form secure attachments with adult mentors. On the other hand, one study found evidence that, compared to boys, girls with low levels of secure attachment may suffer the worst consequences (Georgiou, Demetriou, & Stavrinides, 2008). Specifically, girls with low scores in secure attachment style appeared to have an especially hard time forming natural mentoring relationships. Other research has suggested that in contrast to natural mentoring, in formal mentoring programs, girls are more likely than boys to have poor relationships with parents, yet longer-lasting matches (Rhodes, Lowe, Litchfield, & Walsh-Samp, 2008). Research on adolescent attachment to mother, father, and siblings suggests that there are same-sex and cross-sex differences in attachment relationships (Buist, Dekovic, Meeus, & van Aken, 2002).

162  CULTURAL PERSPECTIVES Moreover, in a study of adolescents that used a cohort-sequential design, the role that gender differences played in the nature of relationships with attachment figures seemed to shift during adolescence: change in the mean level of quality of attachment to the mother appeared to be nonlinear for boys, whereas girls’ mean level of attachment to their mothers showed a linear decline (Buist et al., 2002). Results for attachment to the father were opposite, with a linear decline in quality for boys, and a nonlinear development for girls (an overall decrease, with a period of increase from ages 13 to 15, and of decrease from 15 to 17). Quality of attachment to siblings showed differential development depending on gender composition of the sibling dyad. This research suggests that the outcomes of same-sex and cross-sex mentoring relationships may also be further complicated by differences across developmental stage in adolescence. Other evidence from the attachment literature suggests that same-sex mentoring relationships for girls may come with greater potential conflict. In one study, boys and girls differed in their styles of renegotiating parental relationships and managing the tension between connection and autonomy with same-sex and cross-sex parents (Flannery, 1991). At the same time, although mother-daughter relationships were characterized by greater attachment than mother-son relationships, mothers also struggled with greater intensity of conflict, which in turn was thought to increase negative feelings among adolescent girls and their mothers. To the extent that these findings are applicable to mentor-mentee relationships, they suggest that although matching gender for female mentors and mentees may provide certain benefits and safeguards, it may also mean as much or more conflict. Taken together, this literature suggests the importance of considering both mentors’ and mentees’ attachment styles and interpersonal histories in predicting processes and outcomes. Rhodes’s mentoring process model (2005) currently includes only the mentee’s interpersonal history as a moderator variable between mentoring and outcomes. Similarity and Attraction Theory. Considerations relating to possible gender effects in mentoring become further complicated when the interaction effects between mentor gender and mentee gender are taken into account. Several theorists have posited that individuals are attracted to (Byrne, 1971), and model themselves after (Bandura & Walters, 1963), those whom they perceive to be similar to themselves. It also has been argued that gender-matched role models are especially critical during adolescence when youth are developing their identities

and roles in society (Zirkel, 2002). Moreover, theorists have suggested that for females and other historically oppressed groups, seeing role models from their same group(s) can provide them with explicit examples of what they have the potential to achieve (Zirkel, 2002). Matching by gender may also be a safeguard against misunderstandings that ensue from differences in styles of relating. For example, the attachment literature has demonstrated that ways of approaching interpersonal conflict correspond with traditional gender roles, with secure males demonstrating less warmth and secure females showing more agreeableness (Pietromonaco, Greenwood, & Barrett, 2004). Thus, a securely attached female mentor might misconstrue a securely attached male mentee as uninterested or disconnected. Similarly, a girl with secure attachment may feel rebuffed when paired with a male mentor who demonstrates a lack of warmth in the context of relational conflicts. On the other hand, although little research has examined cross-sex matching in youth mentoring, some literature suggests that there could be benefits for such relationships among boys. Boys naturally seek help from female friends and their mothers (Sears et al., 2009; Timlin-Scalera et al., 2003), and the presence of female role models is associated with psychological well-being across gender (Bryant & Zimmerman, 2003). Workplace research on adult mentoring relationships suggests some downsides to cross-sex matching that may also be relevant in youth mentoring—both mentors and mentees reported fears that expressed interest in a mentoring relationship with an individual of a different gender would be misperceived by others as sexual interest (Gormley, 2008). In the same way, adult mentors’ interest in opposite-sex mentees may be interpreted as inappropriate. This perception may prevent adults from mentoring opposite-sex youth even when they might otherwise be a great match based on other criteria. In sum, the literature suggests that matching based on similarities beside gender—such as shared interests; personality characteristics; and relational styles, such as attachment orientation—may be even more relevant to the success of matches. Question 3: Do Gender Differences in Mentoring Hold Across Other Identity Characteristics (e.g., Race, Ethnicity)? Gender and Identity in Context. In the mentoring literature, gender has been used synonymously with sex and treated as a binary variable. However, gender theory suggests that the term gender is socially

Gender in Mentoring Relationships   163 constructed and therefore differs from the biologically driven term sex (Hare-Mustin & Marecek, 1988). Egan and Perry (2001) define “gender identity” as a “multidimensional construct encompassing an individual’s (a) knowledge of membership in a gender category, (b) felt compatibility with his or her gender group, (c) felt pressure for gender conformity, and (d) attitudes toward gender groups (p. 451). Research, furthermore, has shown gender identity to be a multidimensional variable that influences the psychosocial health of youth (Egan & Perry, 2001). Thus, mentoring programs tailored to be gender specific may benefit boys and girls differentially depending on their existing gender identity. For example, gender-tailored programs might be more beneficial for those who have strong, typical gender identities; alternatively, such programs might be especially beneficial for those with weaker gender identities because they have more to gain. Another question that remains is what is the effect of gender-specific programs on gender-atypical children? Research on this topic would be useful for facilitating understanding of the influence of gender identity development on mentoring success. Moreover, much of the aforementioned theory was derived with middle-class, European American samples and may apply differently to youth from diverse ethnic and socioeconomic backgrounds. For a youth whose ethnicity is of greater (or equal) salience to his or her identity than gender, the latter may play a lesser role in mentoring relationships. In this case, for example, matching pairs on ethnic similarity may supersede the importance of matching on gender (for a discussion of racial and ethnic issues in mentoring relationships, see Sánchez et al., this volume, Chapter 10). In their meta-analysis of 73 evaluations of mentoring programs, DuBois, Portillo, Rhodes, Silverthorn, and Valentine (2011) found that programs that matched participants on race and ethnicity did not exhibit greater effectiveness and concluded that more research in this area of matching is needed. Furthermore, mentors and mentees have multiple, intersecting identities that cannot be reduced or compartmentalized. It is thus likely that mentors and mentees will share some identity characteristics and differ on others. Karcher and Lee’s (2002) research on connectedness among boys and girls found that, in both the United States and Taiwan, girls reported more caring for and close contact with others. Other research, however, suggests that relationship patterns and development of girls and boys may differ across race, ethnicity, and social class. For example, Karcher and Sass (2010) found that girls reported higher connectedness to friends, siblings,

and teachers and that boys reported higher connectedness to their neighborhoods. However, these findings differed across race and ethnic groups. White students reported greater connections to neighborhood and friends than did students of color, and African American and Latino students reported stronger connections to siblings than did White students (Karcher & Sass, 2010). DuBois and Hirsch (1990) found that peer support differed by gender (girls reported more support) among European American but not among African American youth. In another study, White and Black girls from middleclass backgrounds were more likely to self-disclose to peers compared with those from low-income backgrounds (Way & Pahl, 2001). In a study of lowincome adolescents, Way and Chen (2000) found that Latino girls had more general friendship support than did boys, a finding that was not true of African Americans and Asian Americans. In addition, girls in all three ethnic groups had higher levels of close friendships than did the boys in these groups. Other research suggests that the relational development of girls differs across ethnic and socioeconomic groups. Not all girls, for example, seem to be equally susceptible to loss of voice, with some research indicating that this is less likely to occur in girls from low-socioeconomic-status backgrounds and among ethnic minority girls (Clark, 1999; Way, 1995). Taken together, this literature suggests considerable potential for gender to interact with other identity characteristics, to influence whether and how genderstereotypical relational behaviors play out in mentoring connections.

Research In this section, we review research on youth mentoring that addresses both the potentially unique ways in which mentoring manifests across gender and its potential differential effectiveness for boys and girls. Relevant articles were found through a rigorous search of PsycINFO and Google Scholar using terms such as mentoring, mentorship, youth, attachment, gender, and sex. Care was taken to include studies of different youth demographics (e.g., age, gender, race) and different types of measurement and research designs (e.g., quantitative, qualitative, cross-sectional). Additional searches included a focus on specific publications (i.e., Journal of Adolescence, Journal of Youth and Adolescence) that often report empirical findings on youth relationships, a review of the previous edition of the Handbook of Youth Mentoring, and consultation with researchers and theorists.

164  CULTURAL PERSPECTIVES We first review findings pertaining to sex differences in the impact of mentoring, specifically focusing on studies that have emerged since the first edition of the Handbook (i.e., after 2004), including several studies on school-based mentoring. Then we address research that is relevant to each of the three questions we used to organize our consideration of theory presented above. Recent evidence of the differential effectiveness of youth mentoring for boys and girls comes from several studies of school-based mentoring, each of which found significant differences in the impact of mentoring on outcome across sex. In a randomized study involving 516 Latino youth in 5th–12th grades (Karcher, 2008), findings indicated the strongest effects of mentoring program involvement for elementary-aged boys and high school girls, albeit on different outcomes. Among younger boys, those who were mentored had better academically related outcomes (e.g., connectedness to school and the future), whereas among older girls, those who were mentored reported more favorable relationships with friends and peers. Older, high-school-aged boys who were mentored, however, showed a consistent pattern of declines following mentoring relative to their nonmentored counterparts; particularly strong were declines in teacher connectedness following mentoring. This is the only study to date that reports on the influence of school-based mentoring for adolescent boys (in this case, Latino boys), and the results, although in need of replication, are troubling. In two other studies of school-based mentoring that did not include high-school-aged mentees, similar gender differences emerged. One large study of school-based mentoring, including over 2,300 elementary- and middle-school-aged youth, reported significantly greater improvements for mentored youth on scholastic efficacy, school bonding, parental relationships, and misconduct among females compared to males (Bernstein, Rappaport, Olsho, Hunt, & Levin, 2009, pp. D9–D11). A second study of 1,139 elementary and middle school youth participating in the Big Brothers Big Sisters of America school-based mentoring program similarly demonstrated that mentored girls had improved academic performance, class work, assignments completed, and misconduct, whereas boys did not improve (Herrera et al., 2007). On the other hand, the DuBois et al. (2011) meta-analysis of youth mentoring program evaluations conducted between 1999 and 2010 found that effect sizes were significantly stronger (more favorable) for samples that included larger percentages of male youth, a finding that persisted when controlling for other study characteristics. In

line with theoretical possibilities noted in this chapter, DuBois et al. speculated that girls and their mentors may be more likely to engage in emotionally focused interactions, some of which (e.g., co-rumination; Rose, 2002) may be counterproductive. They emphasized, however, that their finding of a difference in effect size associated with gender composition of study samples does not reflect a direct comparison of outcomes for male and female youth and, therefore, could be attributable to other characteristics of the studies or programs involved (DuBois et al., 2011). The hypothesis that girls may benefit (or be harmed) more than boys as a function of whether the program is of high or low quality also has received some support. A secondary analysis of Karcher’s (2008) school-based mentoring study, reported above, found that positive effects of mentoring on connectedness to the future were evident for elementary-aged girls only when their matches were supervised by the better-trained case managers who established more frequent communication with school office staff, whereas a similar difference was not apparent for boys (Karcher & Spencer, 2012). This study also revealed a negative (i.e., harmful) effect of mentoring on connectedness to the future for high school girls who had less experienced and less communicative case managers. These findings, although preliminary, suggest that girls’ outcomes associated with mentoring program participation may be influenced more heavily by program quality and support than are boys’ outcomes. Question 1: Do Boys and Girls Have Different Needs in Mentoring Relationships? Little research has examined explicitly whether boys’ and girls’ relational styles with mentors differ in the ways expected by the theoretical work cited thus far, and the findings are inconclusive. There is, however, some support for gender differences, as well as similarities. For example, in a study drawing on data from the landmark random assignment evaluation of the Big Brothers Big Sisters of America community-based mentoring program, Rhodes et al. (2008) found that girls’ mentoring relationships lasted significantly longer than those of boys, which confirmed qualitative findings from a previous study (Morrow & Styles, 1995). Moreover, girls were more satisfied than boys in long-term relationships, and less satisfied in short- and medium-term relationships. Girls in long-term relationships also rated mentoring as more helpful than did either the boys or the girls in shorter-term relationships. Thus, Rhodes et al. concluded that long-lasting relationships may be particularly well received for girls.

Gender in Mentoring Relationships   165 In addition, girls and boys may form relationships with their mentors that differ on two important characteristics: intimacy and autonomy. Although Rhodes (2002) suggested “meaningful conversation” as a critical mediator between mentoring and successful outcomes, boys may derive successful outcomes through engaging with mentors in “meaningful activity” (e.g., Rickwood & Braithwaite, 1994). Along these lines, research on adolescent girls suggests that they highly value and benefit from relationships with mentors that are characterized by psychosocial qualities, such as empathy and intimacy, and that these qualities help to foster strong relationships (Liang, Tracy, Taylor, & Williams, 2002; Spencer, Jordan, & Sazama, 2004). Furthermore, some natural mentoring studies have reported evidence that both younger (Liang, Tracy, Kenny, Brogan, & Gatha, 2010) and older (Liang et al., 2002) adolescent females tend to rate their relationships with a mentor higher in terms of such qualities than do their male counterparts. Studies on the efficacy of formal mentoring programs—even those that are focused on promoting school success—have indicated that feelings of closeness and emotional support are key ingredients of mentoring relationships associated with improvements in girls’ functioning (DuBois, Neville, Parra, & Pugh-Lilly, 2002). Other research suggests, however, that it may be overly simplistic or incorrect to assume that boys need only instrumental support, and girls need only psychosocial support. Qualitative interviews with boys and their male mentors revealed that it is important for boys to have close mentoring relationships with men who can model how one expresses emotional vulnerability without sacrificing manhood (Garraway & Pistrang, 2010; Spencer, 2007). Specifically, boys described mentoring relationships as characterized by strong, emotional connections through which they could share vulnerable emotions and self-disclose about confidential or difficult topics that they would not choose to speak about with other members of their social network. These descriptions challenge conventional stereotypes of masculinity and the assumption that boys are less likely to develop emotionally intimate mentoring relationships (Darling et al., 2006). In a qualitative study of relationship formation in sex-matched pairs of volunteers and youth in a Big Brothers Big Sisters program, male matches (mentor-youth pairs) were more likely than female matches to be characterized by developmental mentoring (i.e., mentoring that emphasizes psychosocial support and later incorporates goal-directed interactions) (Morrow & Styles, 1995). The girls’ matches

were more prescriptive (i.e., at first goal directed and then incorporating psychosocial support). Suggested explanations for this finding have centered on the nature of the participants (e.g., girl participants in the Big Brothers Big Sisters program had more behavioral problems than did boy participants) and program structure (e.g., the program tended to emphasize activities over talking, which may foster more satisfying friendships for boys than for girls) rather than a lack of interest in psychosocial-focused mentoring among girls (Bogat & Liang, 2005). Similarly, other studies suggest that merely emphasizing psychosocial mentoring for girls may not be enough. A small set of studies suggests that instrumental components of mentoring programs, such as role modeling and coaching, which serve to nurture the talents and abilities of the adolescent, are critical for girls as well as boys (Darling, Hamilton, & Niego, 1994). In a qualitative study of 12 female youth-adult pairs, Spencer and Liang (2009) provided evidence for the relevance of both psychosocial and instrumental mentoring for girls and that these two types of mentoring may be synergistic. In particular, close connections seemed to make it more likely that the mentee would ask for assistance or that the mentor would identify a need; at the same time, effectively offered instrumental support deepened the psychosocial bond among these pairs. Indeed, Darling (2005) argued that for girls and boys alike, instrumental mentoring including skill and knowledge development through teaching and challenging youth protégés may become even more important than psychosocial mentoring in later adolescence, when youth have more opportunities to receive emotional support from their peers and romantic partners. Question 2: Do Same-Sex and Cross-Sex Mentoring Relationships Vary in Their Impact on Youth? Most mentoring programs have a policy of matching by gender that makes it difficult to differentiate whether findings were the result of the mentor’s gender, the mentee’s gender, or some combination of the two. Existing research in and out of the mentoring field has been mixed. Some research on natural and program-based mentoring relationships suggests potential benefits of samesex matching. For example, research on natural mentors found that adolescent males with male role models engaged in less problem behavior and had better academic outcomes than did those without male role models (Bryant & Zimmerman, 2003),

166  CULTURAL PERSPECTIVES although the lack of comparative data on outcomes for males with and without female role models limits the interpretation of this finding. Gur and Miller (2004) hypothesized that male mentors serve as a role model for boys whose parents may be absent or otherwise incapable of fulfilling this role. However, two separate meta-analyses (DuBois, Holloway, Valentine, & Cooper, 2002; DuBois et al., 2011) have failed to provide evidence of any differential benefits for programs using same-sex matches. Similarly, a large-scale evaluation of the Big Brothers Big Sisters school-based mentoring program found that ratings of relationship closeness did not differ significantly between same- and cross-gender pairs (Herrera et al., 2007). A potential hypothesis that should be tested in future research is whether program format and goals can be moderators of same-sex matching benefits—for example, perhaps same-sex matching is more valuable in the context of psychosocial or community-based mentoring than in the context of instrumental or schoolbased mentoring. Furthermore, understanding the influence of cross- or same-gender matching can be advanced fully only by using a balanced design that compares groups of male mentor–male mentee, male mentor–female mentee, female mentor– female mentee, and female mentor–male mentee. Studies should also address mentee preference matching based on gender versus other characteristics. That is, we need to test whether there are individual differences in outcomes based on whether the mentee (and/or mentor) prefers to work with someone of the same gender. Question 3: Do Gender Differences in Mentoring Hold Across Other Identity Characteristics (e.g., Race, Ethnicity)? Much of the existing research on relationship development, friendship patterns, and help-seeking behavior has been conducted with middle-class European Americans. The findings of the small amount of research that examines gender differences in mentoring relationships across youth from various demographic and developmental backgrounds are mixed. There is, however, some evidence that the importance of relational bonds for females transcends other demographic differences. For example, in a study of 276 high-risk youth, Zand and colleagues (2009) found that girls reported higher quality relationships with their mentors than did boys; age and race were not significantly related to the mentor-youth bond. Some research also has suggested that individual mentee differences with regard to gender,

race, and age may influence the effectiveness of mentoring programs (Darling, Hamilton, Toyokawa, & Matsuda, 2002). Other research, furthermore, suggests that gender may interact with other demographic factors in complex ways. For example, in a study of natural mentoring ties among 122 adolescents who varied in race and gender, the strongest associations with relationship measures were for race (Hirsch, Mickus, & Boerger, 2002). Specifically, African American females (especially those from divorced families) reported stronger and more supportive ties to significant adults than did White females or Black or White males. The authors suggest that socioeconomic status, not just culture, may play a role in explaining these race differences. For further discussion of issues relating to mentoring and social class, see Deutsch et al., this volume, Chapter 12. In an evaluation of informal mentoring relationships among Mexican American high school students, no gender differences were found for mentor behaviors regarding the promotion of education and career (Flores & Obasi, 2005). In another study, informal teacher-student mentoring appeared to have a greater influence on females who selfidentified as sexual minority youth of color (Gastic & Johnson, 2009). Using semistructured interviews with college-aged mentors, Schippers (2008) reported case study findings from the mentors’ perspective on the intersection of race and gender in a mentoring program for African American middle school girls. The girls were matched with African American or White college women. The author describes her observations of how the girls expressed strong preferences for African American mentors and then voiced disappointment if they were subsequently assigned White mentors: “Pointing to the African American women, the girls say loud enough for everybody to hear, ‘I want her’ or ‘I hope I get her!’ . . . Later, when the mentoring pairs are announced, the girls express their preferences even more emphatically, squealing with delight when paired with an African American mentor and groaning in disappointment when paired with a White mentor” (Schippers, 2008, p. 78). This finding is interesting given that in their introductory remarks and icebreaker activity, program directors focused exclusively on gender as the most salient characteristic by verbalizing how the mentors and mentees would “bond as girls.” African American mentors’ notes and interview comments revealed that, compared with White mentors, they experienced greater emotional closeness, respect as authority figures, and bonding along gender lines with the girls. White mentors attempted to refer to

Gender in Mentoring Relationships   167 their shared gender status while with their mentees, but this shared gender status was not a consistently salient feature of their mentoring relationships, as it was for the African American pairs. Schippers (2008) suggested that racial differences were more salient than gender for the mixed-race pairs. Research findings taken together thus indicate that gender should be considered in the context of other individual differences, such as race, ethnicity, social class, and developmental stage. Summary and Directions for Future Research In sum, findings to date suggest that longterm mentoring relationships may be particularly beneficial to girls as compared to boys, at least as indexed by their self-reported satisfaction with such relationships, and that both girls and boys seem likely to benefit from a combination of goaldirected or instrumental support paired with psychosocial support characterized by emotional intimacy. Research on same-sex versus cross-sex pairs has yielded mixed findings, and it is possible that the benefits of same-sex matching of mentors differs in natural versus formal mentoring relationships. Finally, the intersection of gender with other identity characteristics has been found to be an important consideration for those designing mentoring programs. Future research should include mixed-methods approaches to help discern similarities and differences between boys and girls

with regard to mentoring relationships in both natural and formal settings as well as with regard to their responses to developmental (relationshipfocused) versus more instrumental (goal-focused) orientations of mentors in relationships. Longitudinal studies are also needed to aid researchers in discerning the potentially differential benefits gained from long-term mentoring relationships for boys and girls. Furthermore, understanding of the implications of matching on gender, potentially in combination with other characteristics, ideally should be advanced through designs in which groups of girls and boys at different stages of development are randomly assigned to same-sex or cross-sex mentors. Additionally, samples of youth and mentors that have shared and divergent identity characteristics (e.g., race, ethnicity, socioeconomic status, religion, family composition) may further enhance our understanding of similarities and differences across same-sex and crosssex mentoring pairs.

Practice We now turn to a consideration of implications of the theory and research reviewed for youth mentoring practice. Our discussion is organized around our three guiding questions, with key recommendations for practice summarized in Table 11.1.

Table 11.1   Checklist for Practitioners Topic

Recommendations

Gender match between mentor and mentee

Even though research on the topic of same- vs. cross-sex matching is lacking, programs should be open to the potential for being matched with a same-gender mentor to matter for some boys and girls. Programs should make sure that mentors pay attention to the gender match between mentor and mentee and address the issue with their mentors and mentees. Programs should be attentive to the possibility that some problems in cross-gender matches may be related to apprehension that the relationship will be perceived as sexualized. Training and in-services should address both the potential benefits and the potential limitations of same- vs. cross-sex matches (see text for particular topics that may be helpful to address). Because mentors in same- vs. cross-sex matches may have different needs, it may be helpful to provide separate training/in-services for both types of matches. (Continued)

168  CULTURAL PERSPECTIVES Table 11.1  (Continued) Relationship development and length

Programs should be designed with the understanding that there are gender-specific ways in which mentoring relationships may develop. For example, mentoring relationships may take longer to establish for girls; mentoring relationships may develop more quickly for boys. Programs should keep in mind that some evidence suggests girls may be particularly receptive to and valuing of longer-term mentoring relationships.

Type of mentoring

Programs should be designed to provide both psychosocial support and instrumental mentoring for both boys and girls because both types of support appear to be beneficial across gender. Programs should keep in mind that girls may be particularly likely to rely on mentors during times of stress. There is also some indication that boys like more activity-based mentoring, whereas girls may tend to prefer intimacy and connection. However, programs must pay attention to individual mentees’ preferences and differences.

Research has indicated that gender-specific programs that overgeneralize and stereotype girls’ needs and behaviors sometimes backfire, which suggests the need for mentor training to emphasize listening to girls’ interests without presumption, rather than imposing values. Gender and other identity characteristics

Currently there are no strong research findings that the interaction of gender and other identity variables influences feelings about the match or outcomes. However, programs should take into account the preferences of individual mentees and their caregivers regarding the gender and other identity characteristics of the mentors with whom they are matched.

Question 1: Do Boys and Girls Have Different Needs in Mentoring Relationships? Although the literature suggests that both boys and girls may benefit from instrumental and emotional support, differences in relationship development, friendship patterns, and help-seeking behaviors may still exist and have direct implications for mentoring youth. Thus, one useful focus for training and supervision may be to increase mentors’ attunement to these potential differences. For example, mentors may do well to attend to gender-specific tendencies in the pacing and development of mentoring relationships (e.g., for girls, mentoring relationships may take longer to establish, ultimately last longer, and be especially welcome in times of stress), interpersonal qualities (e.g., girls may emphasize “talking together,” and boys may emphasize “doing together”), and consequences (e.g., potentially different areas of impact). A number of gender-specific youth mentoring programs have emerged in recent years. Examples for boys include The Brothers Project and Boys to Men Mentoring Network; for girls, programs include Girls on the Run and GirlPOWER! However, with few exceptions (e.g., DuBois et al., 2008), these programs have not been examined using robust research methods. Therefore, even

though programs are tailoring interventions to fit the perceived gender-specific needs of boys and girls, little evidence suggests that such programs are successfully meeting these needs. For example, the work of Bay-Cheng, Lewis, Stewart, and Malley (2006) suggests that despite best intentions, genderspecific programs may backfire when adhering too closely to gender stereotypes and conventionality. Bay-Cheng et al. (2006) described how a program designed to foster “feminist” ideals, such as empowerment and giving voice, unwittingly stifled girls’ voices through stopping and shaping talk (especially talk about sex). This finding suggests the need for mentor training to emphasize listening to girls’ interests without presumption, rather than imposing values (Bay-Cheng & Lewis, 2006; Kaplan, Turner, Piotrkowski, & Silber, 2009). Question 2: Do Same-Sex and Cross-Sex Mentoring Relationships Vary in Their Impact on Youth? All in all, the literature is inconclusive on the relative advantages of same- versus cross-sex matching of mentors and mentees. Because crosssex matches are sometimes necessary to expedite the matching process and decrease time spent on

Gender in Mentoring Relationships   169 the wait list, it is important to know that there is no definitive evidence that same-sex pairing is more beneficial for youth. It may be more important to match youth and adults based on gender-associated styles that may be differentially applicable to each mentee. That is, to ensure a good fit between mentors and mentees, matching criteria may need to go beyond considerations of demographic variables to include relational styles. For example, the research reviewed in this chapter suggests that some youth may benefit from mentoring that emphasizes the slow development of a long-term relationship, “talking together” versus “doing together,” and added support in times of stress. With such possibilities in mind, it may be helpful to assess different relational styles of mentors and mentees, and match with attention to shared styles, rather than on gender alone. Moreover, initial mentor and staff training and ongoing in-service trainings should include an examination of both the benefits and the limitations to same- versus cross-sex matches. For example, mentors in cross-sex matches may be at greater risk of misperceiving the gender-specific needs of their mentees. Similarly, mentors in same-sex matches may be at risk of overidentifying with their mentees or assuming a particular mentoring style based on their notions of gender-based need, rather than their mentees’ actual needs. On the other hand, same-sex matches may provide for gender-role socialization in the form of a nonparental adult role model. This may be a particularly important consideration for young men and boys raised in female-headed single-parent families. Because mentors in sameversus cross-sex matches may have different needs, a case may be made for separate trainings for both types of matches, addressing as well specific needs among male and female subtypes. Question 3: Do Gender Differences in Mentoring Hold Across Other Identity Characteristics (e.g., Race, Ethnicity)? Those designing mentoring programs with issues of gender in mind should be aware that much of the literature to date has focused disproportionately on the experience of middle-class, White, female and male mentors, and that these experiences may not hold for mentors from other backgrounds. After finding modest effect sizes for gender and ethnic matching in a mentoring program aimed at enhancing academic outcomes (e.g., GPA, retention) for ethnically diverse college students, Campbell and Campbell (2007) suggested matching on gender and ethnicity when “feasible and convenient.”

Although there may be benefits to matching mentors on multiple areas of identity that are important to the mentee, research indicates that matching based on personality characteristics, interests, and relational styles may be just as effective in fostering a compatible relationship between mentors and mentees (see Pryce et al., this volume, Chapter 29). As studies suggest, same-sex matching and an emphasis on gender-specific programming do not in themselves ensure the efficacy of a mentoring relationship or program. Furthermore, training should raise awareness of the multiple and intersecting identities of mentors and mentees. As part of the matching process, information may be elicited from mentees on the salience of identity characteristics. For example, depending on the individual, it may be more important to match on identity characteristics other than gender. Thus, mentees and their families may be given a prematch survey assessing their preferences for matching based on a variety of identity characteristics, including gender, race, ethnicity, and socioeconomic status. Once matches are made, mentor training should include discussions of ways for mentors to understand and broach topics regarding gender and other intersecting individual difference variables with their mentees (Rhodes, Liang, & Spencer, 2009). Moreover, mentors should beware of assuming they share similarities with their mentees due to shared gender or other identity characteristics. At the same time, mentors in cross-sex pairs should be cautioned against underidentifying with mentees, based on this difference alone.

Conclusion This chapter has presented theory, research, and practice recommendations relevant to issues involving gender in mentoring relationships for youth. However, the reader should keep in mind that for virtually all of the topics considered here, the available evidence is limited in scope and quality and not conclusive. This includes, for example, whether boys and girls require different mentoring relationships, and if so in what specific ways, or whether same-gender matching is more beneficial for youth. The state of the research in such areas contrasts with theory and empirical literature outside of the mentoring field that suggests considerable differences in how boys and girls approach relationships and what they value in these relationships. One factor contributing to this discrepancy may be the methods used to examine outcomes when studying mentoring programs. There are two

170  CULTURAL PERSPECTIVES broad approaches to research: a variable-oriented approach and a person-oriented approach (Bergman & Magnusson, 1997). A variable-oriented approach focuses on the importance of variables (in the case of the research reviewed for this chapter, that variable would be gender). In this regard, research has by and large assumed that boys and girls represent two homogeneous groups, and, thus, that comparing these two groups tells us something about differences between the groups. However, the real world tells us that boys and girls are unlikely to be homogeneous groups; in fact, there may be as much within- as between-group heterogeneity. In our view, a more promising approach to understanding the influence of gender (as well as other identity attributes of youth) on mentoring relationships would be a person orientation. Such an orientation, with methods that reflect that orientation, would assume that specific variables cannot adequately explain human behavior (e.g., Bogat, Leahy, von Eye, Maxwell, & Levendosky, 2005; von Eye & Bogat, 2006). A person-oriented approach would, instead, examine profiles of individuals (both mentors and mentees) and programs and how these profiles intersect to potentially influence outcomes. For example, mentor and mentee backgrounds may matter. Girls or boys who live in a single-parent household with little exposure to role models from their own gender may well be better predisposed to work with a crossgender mentor. In summary, the extant research on gender and mentoring has provided an important foundation for practitioners interested in facilitating positive outcomes for boys and girls who participate in mentoring programs. However, a true understanding of gender and its interface with mentoring will require a deeper understanding of the broader array of characteristics that define the individuals who participate in programs as well as those defining the programs themselves.

References Addis, M. E., & Mahalik, J. R. (2003). Men, masculinity, and the contexts of help seeking. American Psychologist, 58, 5–14. Bandura, A., & Walters, R. H. (1963). Social learning and personality development. New York: Holt, Rinehart, and Winston. Bartholomew, K., & Shaver, P. R. (1998). Methods of assessing adult attachment: Do they converge? In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 25–45). New York: Guilford Press.

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 11, 497–529. Bay-Cheng, L. Y., & Lewis, A. E. (2006). Our “ideal girl”: Prescriptions of female adolescent sexuality in a feminist mentorship program. Affilia, 21, 71–83. Bay-Cheng, L. Y., Lewis, A. E., Stewart, A. J., & Malley, J. E. (2006). Disciplining “girl talk”: The paradox of empowerment in a feminist mentorship program. Journal of Human Behavior in the Social Environ­ ment, 13, 73–92. Bergman, L. R., & Magnusson, D. (1997). A person-oriented approach in research on developmental psychopathology, Development and Psycho­ pathology, 9, 291–319. Berndt, T. J., & Keefe, K. (1995). Friends’ influence on adolescents’ adjustment to school. Child Develop­ ment, 66, 1312–1329. Bernstein, L., Rappaport, C. D., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S. Department of Education’s Student Mentoring Program. Final Report. NCEE 2009-4047. Jessup, MD: National Center for Education Evaluation and Regional Assistance. Bogat, G. A., Leahy, K. L., von Eye, A., Maxwell, C., & Levendosky, A. A. (2005). The influence of community violence on the functioning of battered women. American Journal of Community Psycho­ logy, 36, 123–132. Bogat, G. A., & Liang, B. (2005). Gender in mentoring relationships. In D. L. DuBois, & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 205–217). Thousand Oaks, CA: Sage. Bryant, A. L., & Zimmerman, M. A. (2003). Role models and psychosocial outcomes among African American adolescents. Journal of Adolescent Research, 18, 36–67. Buhrmester, D. (1990). Intimacy of friendship, interpersonal competence, and adjustment during preadolescence and adolescence. Child Development, 61, 1101–1111. Buist, K. L., Dekovic, M., Meeus, W., & van Aken, M. A. (2002). Developmental patterns in adolescent attachment to mother, father and sibling. Journal of Youth and Adolescence, 31, 167–176. Byrne, D. E. (1971). The attraction paradigm. Ann Arbor, MI: Academic Press. Campbell, T., & Campbell, D. E. (2007). Outcomes of mentoring at-risk college students: Gender and ethnic matching effects. Mentoring & Tutoring, 15, 135–148. Chu, J. Y., Porche, M. V., & Tolman, D. L. (2005). The Adolescent Masculinity Ideology in Relationships Scale: Development and validation of a new measure for boys. Men and Masculinities, 8, 93–115. Clark, M. (1999). Adding the voices of Jewish girls to developmental theory. Women & Therapy, 22, 55–68. Clark, M. L., & Ayers, M. (1993). Friendship expectations and friendship evaluations: Reciprocity and gender effects. Youth & Society, 24, 299–313.

Gender in Mentoring Relationships   171 Cross, S., & Madson, L. (1997). Models of the self: Selfconstruals and gender. Psychological Bulletin, 122, 5–37. Darling, N. (2005). Mentoring adolescents. In D. L. DuBois and M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 177–190). Thousand Oaks, CA: Sage. Darling, N., Bogat, G. A., Cavell, T. A., Murphy, S. E., & Sánchez, B. (2006). Gender, ethnicity, development, and risk: Mentoring and the consideration of individual differences. Journal of Community Psychology, 34, 765–779. Darling, N., Hamilton, S. F., & Niego, S. (1994). Adolescents’ relations with adults outside the family. In R. Montemayor, G. R. Adams, & T. P. Gullotta (Eds.), Personal relationships during adolescence (pp. 216–235). Thousand Oaks, CA: Sage. Darling, N., Hamilton, S., Toyokawa, T., & Matsuda, S. (2002). Naturally occurring mentoring in Japan and the United States: Social roles and correlates. American Journal of Community Psychology, 30, 245–270. DuBois, D. L., & Hirsch, B. J. (1990). School and neighborhood friendship patterns of Blacks and Whites in early adolescence. Child Development, 61, 524–536. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. In J. E. Rhodes (Ed.), A critical view of youth mentoring (pp. 21–57). San Francisco: Jossey-Bass. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., Silverthorn, N., Pryce, J., Reeves, E., Sánchez, B., Silva, A., . . . Takehara, J. (2008). Mentorship: The GirlPOWER! program. In C. LeCroy & J. Mann (Eds.), Handbook of prevention and intervention programs for adolescent girls (pp. 325–365). Hoboken, NJ: Wiley. Egan, S. K., & Perry, D. G. (2001). Gender identity: A multidimensional analysis with implications for psychosocial adjustment. Developmental Psychology, 37, 451–463. Flannery, D. J. (1991). The impact of puberty on parentadolescent relations: An observational study of the relationship between affect and engagement in interactions, parent-adolescent conflict, and adolescent problem behavior. Dissertation Abstracts International, 52(5-A), 1688. Flaxman, E., Ascher, C., & Harrington, C. (1988). Youth mentoring: Programs and practices. New York: ERIC Clearinghouse on Urban Education, Institute for Urban and Minority Education, Teachers College, Columbia University.

Flores, L. Y., & Obasi, E. M. (2005). Mentors’ influence on Mexican American students’ career and educational development. Journal of Multicultural Counseling and Development, 33, 146–164. Galambos, N. L., Almeida, D. M., & Petersen, A. C. (1990). Masculinity, femininity, and sex role attitudes in early adolescence: Exploring gender intensification. Child Development, 61, 1905–1914. Garland, A. F., & Zigler, E. F. (1994). Psychological correlates of help-seeking attitudes among children and adolescents. American Journal of Orthopsychiatry, 64, 586–593. Garraway, H., & Pistrang, N. (2010). “Brother from another mother”: Mentoring for African-Caribbean adolescent boys. Journal of Adolescence, 33, 719–729. Gastic, B., & Johnson, D. (2009). Teacher-mentors and the educational resilience of sexual minority youth. Journal of Gay & Lesbian Social Services: Issues in Practice, Policy & Research, 21, 219–231. Georgiou, S. N., Demetriou, A. P., & Stavrinides, P. (2008). Attachment style and mentoring relationships in adolescence. Educational Psychology, 28, 603–614. Gilligan, C., Lyons, N. P., & Hammer, T. J. (1990). Making connections: The relational worlds of adolescent girls at Emma Willard School. Cambridge, MA: Harvard University Press. Gormley, B. (2008). An application of attachment theory: Mentoring relationship dynamics and ethical concerns. Mentoring & Tutoring: Partnership in Learning, 16, 45–62. Greenberger, E., & McLaughlin, C. S. (1998). Attachment, coping, and explanatory style in late adolescence. Journal of Youth and Adolescence, 27, 121–139. Gur, M., & Miller, L. (2004). Mentoring improves acceptance of a community intervention for court-referred male persons in need of supervision (PINS). Child & Adolescent Social Work Journal, 21, 573–591. Hagerty, B., Williams, R. A., Coyne, J. C., & Early, M. R. (1996). Sense of belonging and indicators of social and psychological functioning. Archives of Psychiatric Nursing, 10, 235–244. Hare-Mustin, R., & Marecek, J. (1988). The meaning of difference: Gender theory, postmodernism, and psychology. American Psychologist, 43, 455–464. Harter, S., Marold, D. B., Whitesell, N. R., & Cobbs, G. (1996). A model of the effects of perceived parent and peer support on adolescent false self behavior. Child Development, 67, 360–374. Harter, S., Waters, P. L., Whitesell, N. R., & Kastelic, D. (1998). Level of voice among female and male high school students: Relational context, support, and gender orientation. Developmental Psychology, 34, 892–901. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A., McMaken, J., & Jucovy, L. Z. (2007). Making a difference in schools: The Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia: Public/Private Ventures.

172  CULTURAL PERSPECTIVES Hill, J. P., & Lynch, M. E. (1983). The intensification of gender-related role expectations during early adolescence. In J. Brooks-Gunn & A. C. Petersen (Eds.), Girls at puberty: Biological and psychological perspectives (pp. 201–228). New York: Plenum Press. Hirsch, B., Mickus, M., & Boerger, R. (2002). Ties to influential adults among Black and White adolescents: Culture, social class, and family networks. American Journal of Community Psychology, 30, 289–303. Jordan, J. V., Kaplan, A. G., Miller, J. B., Stiver, I. P., & Surrey, J. L. (1991). Women’s growth in connection: Writings from the Stone Center. New York: Guilford Press. Kaplan, C. P., Turner, S. G., Piotrkowski, C., & Silber, E. (2009). Club Amigas: A promising response to the needs of adolescent Latinas. Child & Family Social Work, 14, 213–221. Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9, 99–113. Karcher, M. J., & Lee, Y. (2002). Connectedness among Taiwanese middle school students: A validation study of the Hemingway Measure of Adolescent Connectedness. Asia Pacific Education Review, 3(1), 95–114. Karcher, M. J., & Sass, D. (2010). A multicultural assessment of adolescent connectedness: Testing measurement invariance across gender and ethnicity. Journal of Counseling Psychology, 57, 274–289. Karcher, M. J., & Spencer, R. (2012). A mixed-methods, multi-site, multilevel study of program staff contributions to school-based youth mentoring program effectiveness. Unpublished manuscript, University of Texas at San Antonio. Kenny, G. A., & Donaldson, M. E. (1991). Contributions of parental attachment and family structure to the social and psychological functioning of first-year college students. Journal of Counseling Psychology, 38, 479–486. Liang, B., Tracy, A. J., Kenny, M., Brogan, D., & Gatha, R. (2010). The Relational Health Indices for Youth: An examination of reliability and validity aspects. Measurement and Evaluation in Counseling and Development, 42, 255–274. Liang, B., Tracy, A., Taylor, C., & Williams, L. (2002). Mentoring college-age women: A relational approach. American Journal of Community Psychology, 30, 271–288. Miller, J. B., & Stiver, I. P. (1997). The healing connection: How women form relationships in therapy and in life. Boston: Beacon Press. Morrow, K. V., & Styles, M. B. (1995). Building relationships with youth in program settings: A study of Big Brothers/ Big Sisters. Philadelphia: Public/Private Ventures. Paterson, J. E., Field, J., & Pryor, J. (1994). Adolescents’ perceptions of their attachment relationships with their mothers, fathers, and friends. Journal of Youth & Adolescence, 23, 579–600.

Pietromonaco, P. R., Greenwood, D., & Barrett, L. F. (2004). Conflict in adult close relationships: An attachment perspective. In W. S. Rholes & J. A. Simpson (Eds.), Adult attachment: New directions and emerging issues (pp. 267–299). New York: Guilford Press. Pollack, W. (1999). Real boys: Rescuing our sons from the myths of boyhood. New York: Owl. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43). Thousand Oaks, CA: Sage. Rhodes, J. E., Contreras, J. M., & Mangelsdorf, S. C. (1994). Natural mentor relationships among Latina adolescent mothers: Psychological adjustment, moderating processes, and the role of early parental acceptance. American Journal of Community Psycho­ ­­­­­logy, 22, 211–228. Rhodes, J., Liang, B., & Spencer, R. (2009). First do no harm: Ethical principles for youth mentoring relationships. Professional Psychology: Research and Practice, 40, 452–458. Rhodes, J., Lowe, S., Litchfield, L., & Walsh-Samp, K. (2008). The role of gender in youth mentoring relationship formation and duration. Journal of Vocational Behavior, 72, 183–192. Rickwood, D., & Braithwaite, V. A. (1994). Socialpsychological factors affecting help-seeking for emotional problems. Social Science and Medicine, 39, 563–572. Rose, A. (2002). Co-rumination in the friendships of girls and boys. Child Development, 73, 1830–1843. Schippers, M. (2008). Doing difference/doing power: Negotiations of race and gender in a mentoring program. Symbolic Interaction, 31, 77–98. Sears, H. A., Graham, J., & Campbell, A. (2009). Adolescent boys’ intentions of seeking help from male friends and female friends. Journal of Applied Developmental Psychology, 30, 738–748. Soucy, N., & Larose, S. (2000). Attachment and control in family and mentoring contexts as determinants of adolescent adjustment at college. Journal of Family Psychology, 14, 125–143. Spencer, R. (2007). “I just feel safe with him”: Close and enduring male youth mentoring relationships. Psychology of Men and Masculinity, 8, 185–198. Spencer, R., Jordan, J. V., & Sazama, J. (2004). Growthpromoting relationships between youth and adults: A focus group study. Families in Society, 85, 354–362. Spencer, R., & Liang, B. (2009). “She gives me a break from the world”: Formal youth mentoring relationships between adolescent girls and adult women. Journal of Primary Prevention, 30, 109–130. Sullivan, K., Marshall, S. K., & Schonert-Reichl, K. A. (2002). Do expectancies influence choice of helpgiver? Adolescents’ criteria for selecting an informal helper. Journal of Adolescent Research, 17, 509–531.

Gender in Mentoring Relationships   173 Timlin-Scalera, R. A., Ponterotto, J., Blumberg, F. C., & Jackson, M. A. (2003). A grounded theory study of help-seeking behaviors among White male high school students. Journal of Counseling Psychology, 50, 339–360. von Eye, A., & Bogat, G. A. (2006). Person-oriented and variable-oriented research. Concepts, results, and development. Merrill-Palmer Quarterly, 52, 390–420. Way, N. (1995). “Can’t you see the courage, the strength that I have?”: Listening to urban adolescent girls speak about their relationships. Psychology of Women Quarterly, 19, 107–128. Way, N., & Chen, L. (2000). Close and general friendships among African American, Latino, and Asian American adolescents from low-income families. Journal of Adolescent Research, 15, 274–301.

Way, N., & Pahl, K. (2001). Individual and contextual predictors of perceived friendship quality among ethnic minority, low-income adolescents. Journal of Research on Adolescence, 11, 325–349. Weisz, J. R., Sandler, I. N., Durlak, J. A., & Anton, B. S. (2005). Promoting and protecting youth mental health through evidence-based prevention and treatment. American Psychologist, 60, 628–648. Zand, D. H., Thomson, N., Cervantes, R., Espiritu, R., Klagholz, D., LaBlanc, L., & Taylor, A. (2009). The mentor-youth alliance: The role of mentoring relationships in promoting youth competence. Journal of Adolescence, 32, 1–17. Zirkel, S. (2002). Is there a place for me? Role models and academic identity among white students and students of color. Teachers College Record, 104, 357–376.

12  SOCIAL CLASS Nancy L. Deutsch, Edith C. Lawrence, and Angela K. Henneberger

Introduction The Horatio Alger story looms large in America’s national psyche. We are a society built on the idea of meritocracy. We believe in class mobility and resist the notion that social class operates as a structure that limits individual choice and ability (hooks, 2000; Huston & Bentley, 2010; Lareau, 2008). As a result, social class in America is an “untalked-about reality” (hooks, 2000, p. 31). When it is discussed, social class is often evoked either in combination or in competition with race for its influence on lives. Perhaps it is no surprise, then, that the literature on mentoring has largely ignored the role of social class in considering what makes for effective mentoring relationships. Although differential effects of mentoring programs on youth with varying socioeconomic backgrounds have received some empirical attention (e.g., DuBois, Holloway, Valentine, & Cooper, 2002; DuBois, Portillo, Rhodes, Silverthorn & Valentine, 2011), social class remains largely just under the surface of most mentoring research, despite its implicit presence in the national call for youth mentors. Mentors offer support and advice, serve as role models and advocates, and provide opportunities for youth to identify with positive adults (DuBois & Silverthorn, 2005; MENTOR, 2005; Rhodes, 2002). Embedded in this concept is the idea that mentors have access to resources unavailable to their mentees. Many programs appear to be based on the assumption that one benefit of mentoring is the provision of social capital to youth living in underresourced communities and/or families. Indeed, formal mentoring programs serve primarily youth from low socioeconomic status (SES) and racial/ethnic minority backgrounds, whereas most mentors come from White, upper-class backgrounds (Freedman, 1993;

Liang & West, 2007; R. Spencer, 2007a). Yet natural mentors, who are more likely to be from youths’ own communities and therefore to share social class status with youth, play an important role in youth development (DuBois & Silverthorn, 2005). In this chapter, we consider social class as one’s socioeconomic positioning within society, defined not only by family income but also by the education and occupations of adults in a child’s household. Such factors combine to create differences in financial resources and social power (Cohen, 2009). Because SES and class shape the environments in which youth live, we consider social class as a cultural context for youth development. Thus, we present a “culturally informed approach . . . [that considers] the rich, culturally textured beliefs, values, and practices” (Cohen, 2009, p. 197) of individuals from different social classes. We first review theories of social class and its influence on youth development, focusing on ways in which social class may be important for understanding the role of mentoring relationships in the lives of youth. We then propose a model of how cultural differences between mentor and mentee, particularly differences related to social class, might inform the focus and development of mentoring relationships. Finally, we review the research on the role of social class in mentoring and, building on our reviews of theory and research, discuss implications and recommendations for practice.

Theory Perspectives on Social Class What is social class? The answer depends on who you ask (Scott, 2008). Although the absence of 175

176  CULTURAL PERSPECTIVES social class in the mentoring literature may reflect Americans’ discomfort with social class, its complexity as a concept may also be a contributing factor. Definitions of social class revolve around money: how much you make, how you make it, and what you do with it (Hout, 2008). Yet social class also invokes capital (economic and cultural), workplace relations, and collective identity and interests (Lareau, 2008). Social class has been used to “designate differences in wealth, income, occupation, status, group identification, level of consumption, and family background” (Gordon, 1949, p. 262), but there is no consensus on whether class rests on income, employment, group identification, cultural differences, or some combination thereof (Gordon, 1949). In the field of youth development, social class is typically treated either as material/environmental or as cultural. In this section, we discuss how social class is defined and measured and why it is important to the field of mentoring. We believe that conceptualizing social class as a “social address” that affects the environments in which youth and their families interact (Huston & Bentley, 2010) allows for consideration of the mechanisms by which social class influences youth development and, therefore, is a useful approach for the field of mentoring. Social class affects youth through the provision of and access to material and social resources (e.g., quality child care and schools, social networks); cultural practices (e.g., beliefs and values); and processes such as discrimination, residential segregation, and stress that stem from material and cultural factors (Garcia-Coll et al., 1996; Huston & Bentley, 2010; Lareau, 2003). Mentoring may not be able to alleviate factors such as health, nutrition, and neighborhood conditions, but it may offer youth access to expanded social resources, human capital, and valued types of socialization and cultural practices. Some scholars claim that social class is not applicable to 21st-century America because there are no longer distinct, universal life experiences shared by individuals with common economic positioning (Kingston, 2000; Manza, Hout, & Brooks, 1995). Yet ample evidence indicates that social class still shapes lives in ways that are defined by both material and cultural aspects of class (Lareau, 2008). In the sections that follow, we discuss the material and cultural approaches to social class as they relate to youth mentoring and propose a model for considering the role of social class in mentoring programs and relationships. Material and Environmental. The material definition of social class includes family income and financial resources. These resources both influence and are

influenced by factors such as education, occupation, and social networks, which shape the environments in which youth live. For example, people with fewer economic resources have poorer health and well-being (Cohen, 2009), and studies have repeatedly demonstrated associations between children’s academic and social-emotional outcomes and SES (Duncan & Magnuson, 2003; Hattie, 2009; Huston & Bentley, 2010). Differing financial resources result in differing access to cultural and social capital such as academic tutors, after-school activities (Lareau, 2003), institutions, or people affiliated with different social networks. Family income may also affect youth through mediators such as material hardship and parental stress (Gershoff, Aber, Raver, & Lennon, 2007). Mentors should be made aware of the ways in which differential access to resources reflect important differences between them and their mentees and potential sources of misunderstanding. Cultural. The cultural definition of social class focuses on values, beliefs, and social norms that shape youths’ attitudes, behaviors, and experiences, sometimes providing a sense of communal identity with others who share one’s social class. The cultural approach posits that beliefs, values, and social practices become internalized predispositions that include, among other things, preferences for music, clothing, and entertainment and that differ between people of different social classes (Bordeau, 1987, cited in Kingston, 2000; Bordeau, 1998; Cohen, 2009). Different values and beliefs also lead to differing ways of interacting with institutions, such as schools or potential employers (Lareau, 2003). Furthermore, Lareau asserted that social class influences parenting practices, with parents in middleclass families engaging in more explicit cultivation of children and working-class and poor families adhering more to “natural growth” parenting practices. Snibbe and Markus (2005) reported cultural differences in models of agency, with individuals from lower-SES backgrounds tending to value flexibility, integrity, and resilience and individuals from higher-SES backgrounds tending to value control and individual influence. Whereas Kingston (2000) argues that social class does not correlate with cultural beliefs and values when education is removed from models, the scholars discussed in this section of the chapter consider education as part of one’s social class positioning. Another useful concept for considering social class as culture is Swidler’s (1986) notion of “culture as a ‘tool kit’ of symbols, stories, rituals, and worldviews, which people may use . . . to solve different kinds of problems” (p. 273). As such, social class

Social Class  177 influences the “strategies of action” that are available to individuals (Swidler, 1986). Taken together, this broader conceptualization of social class highlights how families, youth, and mentors from different class backgrounds may value different types of experiences or approaches to interacting with youth and social institutions. Within a mentoring relationship, it may be as essential for successful relationship development for a mentor to learn about and appreciate his or her mentee’s class “tool kit” as it is to understand his or her race or ethnic culture. This is an especially important point for mentoring programs to consider because our society tends to treat middle-class culture as the norm. Mentoring programs typically seek to empower youth from lower SES backgrounds by providing them with resources assumed to be missing from their communities. It has been suggested that successful mentoring relationships may serve to expand youths’ social capital (Philip, 2008; Putnam, 2000) by providing opportunities to interact with persons and institutions outside of youths’ immediate settings. In this model, youths’ access to both material and cultural aspects of social class is assumed to be enhanced by the presence of mentors who, presumably, come from a different class background than the youth. In fact, some mentors may infuse this approach into their mentoring (R. Spencer, 2007a). Mentoring from this perspective may engender a deficit view of youth. Yet researchers of youth programming increasingly argue for greater focus on the social capital and competencies of youth from lower SES backgrounds (M. B. Spencer, Dupree, & Hartmann, 1997). We hypothesize that if not addressed respectfully, differing class backgrounds and experiences could strain the abilities of mentors to make strong connections with their mentees; without a strong connection, mentoring is unlikely to be effective (Herrera, Sipe, & McClanahan, 2000; Rhodes, 2002; R. Spencer, 2006). Social Class and Youth Identity. The cultural view of social class implies a sense of group identity resulting from common experiences, interests, and socialization (Bordeau, 1998; Hout, 2008). This is similar to ethnic identity, emphasizing that people who share salient background characteristics may view themselves as having a common identity. That class identity may be more salient than previously thought is supported by evidence that social class can be a dividing line even within racial groups (Deutsch, 2008; Taylor et al., 2008). For example, middle-class Black families have been found to have more cultural practices in common with middle-class White families than with workingclass Black families (Lareau, 2003). At the same time,

just over one half of Americans report being middle class (Hout, 2008; Taylor et al., 2008), including 40% of those with incomes below $20,000 and onethird of those with incomes above $150,000 (Taylor et al., 2008), suggesting that many people do not identify with the labels we give to socioeconomic groups. There are not extensive studies of youths’ perceptions of social class, but ethnographic research (Deutsch, 2008; Deutsch & Theodorou, 2010; see also Huston & Bentley, 2010) and studies of social class stereotypes (Croizet & Claire, 1998) indicate that low-income youth are aware of their own social positioning in relation to others. Furthermore, because of the salience of the middle-class-as-normative view, the types of “microaggressions” (i.e., everyday verbal, behavioral, or environmental slights; Sue et al., 2007) that research has begun to identify in regards to race may occur around class as well. In mentoring, how youth view themselves in terms of social class identities may influence how they relate to mentors. The degree to which youth and their families are attuned to relative class positioning within mentoring relationships may make them more or less open to working with a mentor from a different background. Measurement of Social Class Measurement decisions are often based on available instruments, focused primarily on income, education, and occupational status (Grusky & Weeden, 2008; Lareau, 2008). Thus, most mentoring research has used measures that assess the material/ environmental aspects of social class. Typically, researchers focus on measuring the amount of money an individual or a family has (income) or on the way in which that money is earned (occupation; Hout, 2008; Manza et al., 1995). Some researchers use multiple SES measures to assess different components of social class, whereas others combine markers into one indicator or use only a single item (Duncan & Magnuson, 2003). Measuring youths’ social class is particularly complicated. Whereas adults “achieve” social class based on their occupations and incomes, children’s social classes are assigned to them based on their family, household, and neighborhood (Garcia-Coll et al., 1996). Researchers must decide whose status to measure, as children may or may not have access to the resources of their mother or father (Entwisle & Astone, 1994). Social class measures for youth rely typically on free/reduced lunch status, parental employment, and parental education (Liu et al., 2004; Sims, 1951; Wohlfarth, 1997). Free/reduced lunch status has been criticized, however, as an

178  CULTURAL PERSPECTIVES unreliable marker of class (Hauser, 1994). Longterm social class status is seldom captured but is important to consider, as there is surprising fluctuation in income, education, and occupation within families over time (Duncan & Magnuson, 2003). In studies of mentoring, social class has been measured in various ways, including through proxies for family income (e.g., free/reduced lunch status, family on public assistance, parental employment), household structure (e.g., single-parent family, number of people in household), parental education levels, and neighborhood characteristics (e.g., feelings of safety, levels of violence; Bernstein, Rappaport, Olsho, Hunt, & Levin, 2009; DuBois, Holloway et al., 2002; DuBois & Silverthorn, 2005; Herrera, Kauh, Cooney, Grossman, & McMaken, 2008). We recommend following Hauser’s (1994) suggestions to measure the head of household’s educational attainment and occupation as well as household income, family composition, and educational attainment of other adults in the household in order to assess components of social class that influence youth experiences and outcomes. Due to inconsistencies in measurement and concept definition, we may not yet know how a youth’s SES mediates the effects of mentoring or how the class match between mentor and mentee shapes the relationship. Because of the measures used in research, the information we have focuses on material aspects of social class, with very little knowledge of how cultural aspects of class influence mentoring. Yet based on theories of social class, we believe there may be important cultural factors that researchers and practitioners should take into account when considering social class and mentoring. Below, we propose a model for how mentoring programs can support mentors in recognizing and respecting youths’ social class and concomitant cultural resources, review the empirical literature on social class and mentoring, and provide suggestions for practice. A Model of Youth Mentoring and Social Class Most mentoring programs have as their charge to mentors the promotion of positive youth development through enhancement of youth agency and outcomes (Rhodes, 2002). However, whether a mentor is from the same or a different social class as a youth may have important implications for how a youth identifies with a mentor, a process recognized as critical to youth outcomes (Rhodes, 2005). Other important processes may include the mentor’s sense of efficacy, the experience of closeness within the relationship, and the mentor becoming an important

adult in the youth’s life (DuBois, Neville, Parra, & Pugh-Lilly, 2002; Parra, DuBois, Neville, PughLilly, & Povinelli, 2002). Social class could be an important consideration in accounting for the effects of mentoring relationships on youth in distinct ways, depending on the degree to which mentor and mentee share similar social class experiences and background. Drawing on intergroup contact and empowerment theories, we propose that it may be useful to conceptualize the role of social class in the development of effective mentoring relationships as a twostage process. A model of these pathways is shown in Figure 12.1. The social class backgrounds of the mentor and mentee directly influences their social capital and material and environmental resources. For mentors and mentees with dissimilar social class experiences, we suggest that youth outcomes may be influenced by the degree to which pairs focus on understanding, attending to, and navigating their cultural and material differences respectfully. The proposed focus during this initial phase of mentoring is based on intergroup contact theory (ICT) and is indicated as Stage 1 in the model presented in Figure 12.1 (Arrow A-1). We suggest that intergroup contact processes may help bridge potential differences in mentors’ and mentees’ cultural backgrounds related to class and facilitate relationship development (Arrow A-2), which in turn influences youth outcomes (Arrow C). On the other hand, for mentoring pairs that share similar social class experiences and do not need to navigate class and cultural differences, we hypothesize that their similarities may enhance relationship development (Arrow B) and allow them to focus more quickly on youth empowerment and outcomes, Stage 2 in the model (Arrow C). We expect that the strength and possibly also the direction of these pathways may be moderated by material and environmental factors that relate to social class, such as family income and access to resources like libraries and after-school activities, and by social capital factors, such as the youth’s or mentor’s access to people with links to jobs and higher education (Arrow D). For example, the impact of being paired with a mentor from an advantaged social class on the youth’s academic achievement may be greater when the youth lives in a community with few material or environmental resources to support educational success. In this way, factors relating to social class may be integral to the quality of mentoring relationships and the actual mechanisms and effectiveness of mentoring. That a focus on understanding of and respect for varying values, beliefs, and practices related to

Social Class  179 Figure 12.1   A Model of Mentoring and Social Class

Social Class Experiences and Background of Mentor and Mentee

A–1

Stage 1: Intergroup Contact Processes

A–2

B

C

Dyadic Relationship Processes

Stage 2: Youth Empowerment and Outcomes

D D Moderating Factors: Material resources Environmental resources Social capital

social class might be an important beginning frame for mentors and mentees with class differences builds on the work of Allport (1954), who suggested ICT as a means of reducing prejudice among groups of different races and ethnicities. ICT posits that four “essential” conditions must be present for intergroup contact to have positive results: (a) equal status between people in the situation, (b) common goals that are being worked toward, (c) cooperation and interdependence in working toward these goals, and (d) support for intergroup contact on an institutional level. An extrapolation to what conditions might be best for establishing a positive relationship when mentor and mentee differ in important ways (e.g., class, race, religion) might include that these mentoring pairs (a) engage in experiences that promote equal status between them, (b) mutually establish mentoring goals on which to work, (c) find ways to work collaboratively and cooperatively on these goals, and (d) receive program support for understanding and navigating their differences. When applied to mentoring, ICT and the proposed underlying processes suggest a possible path for reducing prejudice, including social class prejudice, and increasing mentoring effectiveness. We propose that attending to these processes is a critical first step in effective relationship development if cultural differences exist between mentor and mentee. The mentoring field already recognizes the necessity of fostering cultural sensitivity in mentors with regard to cross-race/ethnicity matches (Liang

& West, 2007; Sánchez & Colón, 2005). We hypothesize that due to the prevalence of the “middle-class culture as normative” approach to youth development in America, cross-class pairings are likely to have embedded class assumptions and biases that need to be understood as well. We propose that once prejudice reduction has occurred and biases are understood, mentoring pairs can successfully transition to a second stage (Stage 2 of the model) of addressing the more commonly understood goal of mentoring: “helping young people fulfill their potential” (MENTOR, 2012). This stage encompasses the role that mentoring relationships may have in fostering empowerment and other positive outcomes among youth (see Arrow C in Figure 12.1). When one considers mentoring as a means for enhancing youth agency, as this quote suggests, we believe that the processes delineated in empowerment theory and its application (Hur, 2006; Zimmerman, 2000) can be useful tools for practice. For example, collaboration and capacity-building are two key underlying processes that separate empowerment approaches to intervention from other approaches that are based in deficit views of problem solving (Zimmerman, 2000). When applied to mentoring, empowerment theory suggests that mentors develop a collaborative role with their mentees that involves valuing the culture of the youth and their community and building on that knowledge to increase their agency for positive development. Youth empowerment occurs when

180  CULTURAL PERSPECTIVES youth feel supported to develop their own capacity to overcome obstacles and “attain self-determination, self-sufficiency, and decision-making abilities” (Hur, 2006, p. 531). Based on our review of the literature on same-race mentoring relationships (Sánchez & Colón, 2005; Sánchez et al., this volume, Chapter 10) and natural mentoring relationships (R. Spencer, 2007b; Zimmerman, Bingenheimer, & Behrendt, 2005), which are more likely to be sameclass pairings, we hypothesize that mentor pairs from similar cultures, including class, may have already met the optimal conditions of intergroup contact and can more readily move to a focus on enhancing mentee outcomes through collaboration on goals and youth capacity-building.

Research In this section, we review the research that addresses the role of social class in mentoring relationships. Searches were conducted on databases (including PsycINFO, PsycArticles, and PubMed) using keywords mentoring and disadvantaged, social class, and/or at-risk. We also asked for unpublished studies on relevant mentoring listservs. Research findings are limited to those directly relevant to youth mentoring. This section first reviews the literature examining the effects of social class on outcomes in mentoring relationships. We then review the literature on mentor-mentee match with regard to social class. As noted above, social class is often defined within the context of race. However, in this chapter we present only those findings directly relevant to social class (see Sánchez et al., this volume, Chapter 10, for a discussion of race, ethnicity, and culture in mentoring relationships). Social Class and Outcomes One way in which social class may play a role in mentoring is by acting as a moderator of youth outcomes, as suggested in Figure 12.1. Formal mentoring programs tend to target youth from at-risk backgrounds, as ample evidence indicates that a positive relationship with an extrafamilial adult promotes resiliency in this population (for a review, see Werner, 1995). However, there is no authoritative definition of “at risk,” and the term is used to describe youth in many different conditions (e.g., living in poverty, experiencing abuse and neglect, displaying early symptoms of a behavioral disorder). In their meta-analysis of 55 studies of formal mentoring programs, DuBois, Holloway et al. (2002) found that mentoring had the greatest effects for youth

considered environmentally “at risk.” This was true whether youth experienced environmental risk in combination with individual risk or environmental risk alone, although no differential effects were apparent for youth living in single-parent versus twoparent households. It is not clear what measures were used to define environmental risk. Both single-parent household and low SES were also examined individually, with programs that focused on low-SES samples of youth demonstrating a marginally significant trend toward stronger effects in analyses conducted under assumptions of a fixed-effects model, but not when employing a generally more conservative random-effects model. The types of SES measures used are not known, although they were likely markers of material aspects of social class. Similar findings were reported in an updated meta-analysis conducted by DuBois and colleagues (2011), wherein environmental and individual-level risk were found to be interdependent in relation to estimates of program effect size, with the greatest program effects apparent for youth with either a relatively high level of environmental risk and a low level of individual risk, or vice versa. Family SES and single-parent household were again examined individually as potential moderators of effect size as were low neighborhood resources and presence of neighborhood risk factors (crime, drug use, and/or violence). None of these factors exhibited a significant association with program effect size (a marginally significant association was found for single-parent household in the direction of weaker effects for samples in which the majority of the youth came from such households, but this association was no longer evident when controlling for other moderators with which it was associated). The attention to a broader range of indicators relevant to social class was helpful in this more recent meta-analysis. Again, however, neither metaanalysis defined how SES (or related indicators such as neighborhood resources) was measured, likely because measurement of SES may have differed across studies within each meta-analysis. As a result, how material and cultural components of social class each influence mentoring effects is unclear. Another investigation, using data from the National Longitudinal Study of Adolescent Health and the Adolescent Health and Academic Achievement Study (Erickson, McDonald, & Elder, 2009), found that associations of natural mentoring relationships and academic outcomes were stronger among youth from low-income backgrounds compared to higherSES peers, particularly when these relationships were with teachers. Thus, there is some support for the pathway suggested in Figure 12.1, whereby

Social Class  181 environmental and material resources related to social class may moderate the effects of mentoring. Social Class and Mentor-Mentee Match A second way that social class may play a role in mentoring is via the mentor-mentee relationship and how the match between the participants’ social class backgrounds influences that relationship (see Figure 12.1). A qualitative study of parent involvement in mentoring conducted with 13 parents of youth who were participating in a community-based mentoring program found that parents hoped that mentors would provide their children with experiences that would “contribute to a broadening of their child’s sense of self and future possibilities” (R. Spencer, Basualdo-Delmonico, & Lewis 2011, p. 53), including educational opportunities. This motivation on the part of parents reflects the idea of mentoring as a source of social and human capital. A recent quasi-experimental study (Gaddis, in press) tested this hypothesis using a random sample of 355 youth from existing Big Brothers Big Sisters of America (BBBSA) applicants waiting to be assigned to a mentor. The study found no relationship between the type of match (same- versus crossclass) and youth outcomes. The author conjectured that, due to the age of the sample (mean age = 12) and the types of social capital that may be influential for low-income youth (education-related capital), the effect of social class on youth outcomes may not yet be apparent in his sample but could emerge as the youth age. He also suggested that providing children access to social capital does not guarantee they will be able to use that capital effectively. This finding is consistent with Swidler’s (1986) argument that people can “not take advantage of new structural opportunities” (p. 281) if they do not feel that they have the appropriate tools to do so within their own cultural tool kits. In a qualitative study involving in-depth interviews with 23 adult volunteer mentors and 21 youth mentees, R. Spencer, Lewis, and Basualdo-Delmonico (2007) found that three themes emerged regarding how mentors navigated social class differences between themselves and their mentees. Some mentors demonstrated sensitivity while others minimized or took a deficit view of social class differences. Overall, the mentors appeared to value upward mobility without demonstrating any understanding that there may be different models of success. The authors concluded that whereas actively navigating social class differences may increase processes such as identification and role modeling, minimizing differences may impede the development of close relationships and

taking a deficit view could even harm the youth’s own sense of self (R. Spencer et al., 2007). In fact, R. Spencer and Basualdo-Delmonico (this volume, Chapter 32) reviewed research suggesting that mentoring relationship failures may be due in part to difficulty in forming relationships across cultural barriers, with some mentors describing potential underlying biases rooted in cultural and social differences. Thus, for cross-class relationships, the first stage of our proposed model with a focus on cross-cultural understanding and prejudice reduction, as illustrated in Figure 12.1, may be particularly important. Support for using ICT’s four steps to address differences in social class within mentoring relationships comes from a meta-analysis of over 500 studies, about half of which involved nonracial and nonethnic samples (Pettigrew & Tropp, 2006). The meta-analysis concluded that while intergroup contact alone could facilitate prejudice reduction, establishing the four conditions proposed by ICT (equal status, common goals, cooperation, and institutional support) enhanced the positive effects of intergroup contact. In the same review article, Pettigrew and Tropp (2006) also summarized the research on what might be the underlying processes or mechanisms that promote optimal intergroup contact conditions leading to prejudice reduction. They posited that “reducing negative feelings such as anxiety and threat represents an important means by which intergroup contact diminishes prejudice” (p. 766). Intergroup anxiety, a likely reflection of feeling threatened or uncertain in the intergroup context, occurs when people are not sure “how they should act, how they might be perceived, and whether they will be accepted” (p. 767). Other mechanisms that have been suggested include being open to learning about the other, establishing affective ties, changing one’s own behavior, and reappraising one’s own position (Pettigrew, 2004). Little empirical work has been done with regard to same-class mentoring relationships. Mentors in natural mentoring relationships may be more likely than those in formal mentoring programs to have a similar social class status as the youth (e.g., aunt, neighbor) or extended involvement in the youth’s community (e.g., teacher). Such relationships have been found to be associated with positive youth outcomes (for reviews, see R. Spencer, 2007b; Zimmerman et al., 2005). Natural mentors may have particular strengths such as shared cultural ways of communicating (Hirsch, 2005). Yet potential limitations of such relationships have also been pointed out, particularly with regard to access to certain types of social capital, in this case, knowledge of college admissions procedures when a mentor and a

182  CULTURAL PERSPECTIVES mentee both come from communities with limited contact with institutions of higher education (Hirsch, Deutsch, & DuBois, 2011). One study of natural mentoring relationships with high school students in a STEM (science, technology, engineering, and math) summer program sheds some light on youths’ feelings about having mentors from a similar background (Syed, Goza, Chemers, & Zurbriggen, 2012). The authors focused primarily on ethnicity, although social class and ethnicity were difficult to untangle in their sample. Youth were asked a series of items that together assessed the amount of contact they had with mentors from a shared background as well as how important it was to them to have such mentors. Youth were instructed to think about ethnicity, gender, and social class as part of background in responding to the questions, but they were not asked about each dimension separately. Therefore, the relative importance of social class cannot truly be discerned. In an attempt to differentiate between these three dimensions of background, the authors tested for differences by participants’ gender, ethnicity, and social class (as measured by receipt of financial aid to attend the program). Whereas there were differences in levels of contact with similar-background mentors for youth from lower- versus higher-SES backgrounds (lower-SES youth reported less contact), there were no differences in importance. Non-White youth, however, did report greater importance of having a similar-background mentor than did White youth. The authors concluded that ethnicity is therefore more important in terms of mentor-mentee match. Yet there was significant overlap between ethnicity and SES in the sample, with non-White youth being more likely to come from lower-SES backgrounds, making it difficult to draw conclusions about the role of each. Furthermore, when the authors performed a cluster analysis to examine typologies of contact and importance over time, they found differences by both SES and ethnicity but did not have the sample size to test for the interaction between the two. They discussed the results primarily in terms of ethnicity, highlighting the continued downplaying of social class within mentoring research. Summary and Recommendations for Future Research Overall, few studies have specifically examined the social class of mentors and mentees. Viewed together, available findings suggest that mentoring may be more beneficial for youth from lower SES backgrounds. However, there are significant limitations to this evidence, including a primary focus on

material aspects of social class with limited and varied measures of youths’ SES and almost no measures of cultural aspects of class. When social class is reported, proxies such as household structure or risk status are often used. This approach inhibits our ability to understand social class as a broader, dynamic social positioning or culture that may influence mentoring in complex ways. Overall, we have some suggestions about how the material aspects of social class influence mentoring (i.e., youth in environments with fewer resources may benefit more), but we know very little about how the cultural aspects of social class influence mentoring relationships or programs. In studies that include an examination of cultural differences, race and ethnicity tend to be the focus, with limited, if any, consideration of social class. Additionally, the focus on the social class of either the mentee or the mentor alone, rather than on the match, limits our understanding of how social class may influence the dyadic relationship between mentor and mentee. The research to date that has yielded the richest information on the role of social class has used interviews as a primary method (R. Spencer, 2007a; R. Spencer et al., 2007). We recommend that researchers include a more nuanced and specific focus on social class within studies of mentoring. This should include measuring a variety of material and cultural aspects of class. Rather than relying on a single proxy, such as free/reduced lunch status, researchers should consider the variety of markers noted earlier and use measures in combination to examine social class both as a complex whole and as a concept with different components (e.g., income versus education versus occupational status). To examine social class as both a moderator of outcomes and an influence on the mentor-mentee relationship itself, as suggested in Figure 12.1, studies should include measures of the social class of both mentees and mentors. Furthermore, assessing not only the current social class status of mentors but their childhood social class as well could provide needed information about the role of mentor-mentee social class match and its influence on relationship processes. Due to measurement issues, particularly with capturing social class as culture and/or collective identity, qualitative methods may help researchers better understand the nuances of social class and how class as a culture may influence both the process and outcomes of mentoring.

Practice MENTOR: The National Mentoring Partnership’s most recent edition of Elements of Effective Practice for Mentoring recommends that after the mentoring

Social Class  183 relationship begins, programs provide mentors training on “increasing multicultural understanding, particularly regarding issues of race and class differences” (MENTOR, 2009). Indeed, although they did not specifically review training materials for social class, Kupersmidt and Rhodes (this volume, Chapter 30) recommended that programs pay attention to social class in their mentor training. Although many mentoring programs probably include social class issues when providing their mentors with training in cultural competence, in response to a solicitation to members of a listserv for youth mentoring research and practice with approximately 600 members (i.e., [email protected]), most of whom are practitioners (D. DuBois, personal communication, May 9, 2012), only three shared with us how they intentionally address these issues. Phoenix Youth At-Risk and BBBS of Northern Minnesota have adapted concepts from Payne’s (2005) research and teachings on a framework for understanding the culture of poverty and include them in mentor training. The goal of these activities is similar to the goal of ICT, that is, to reduce prejudice and increase understanding of differences (see also Kupersmidt & Rhodes, this volume, Chapter 30, for general information on mentor training not specific to social class issues). A popular speaker on poverty education, Payne focuses on providing middle-class educators and service providers with a better understanding of the challenges faced by families living in poverty. She addresses topics such as identifying resources to which people in poverty do and do not have access, uncovering hidden rule and value differences between social classes, and developing relationships with youth in poverty that are motivating. However, Payne’s approach is not without controversy. It has been criticized for blaming those in poverty for their situation, for promoting class stereotypes, and for not being empirically accurate (Gorski, 2005). Programs may also encourage mentoring activities that facilitate mentors’ recognizing and respecting youths’ cultural resources while simultaneously promoting equal status between mentor and mentee, thus addressing the first condition of ICT and aligning with the first stage of our proposed model (see Figure 12.1). The Young Women Leaders Program, a combined group and one-on-one after-school mentoring program at the University of Virginia that pairs college women and middle school girls, has incorporated several activities aimed at helping mentors appreciate the mentees’ world during the first month of the mentoring relationship (Lawrence, SovikJohnston, Roberts, & Thorndike, 2010). During their second group meeting, mentees are asked to identify

a woman in their family or community who they feel has been a leader in their lives and share why; the majority choose their mother or grandmother. These women are sent a certificate of appreciation from the group and later interviewed by the nominator’s mentor to learn what she sees as the mentee’s talents and abilities and in what areas the mentee might be helped to grow. Finally, the nominated women are invited to an appreciation dinner at the school, planned and prepared by the mentees and mentors, during which the honorees are introduced and celebrated. Combined, these activities provide a structured opportunity for the mentor, the mentee, and a significant adult in the mentee’s life to establish affective ties and learn about the other, two processes Pettigrew and Tropp (2006) suggest could reduce intergroup anxiety, tension, and prejudice. Given the lack of attention paid to social class in the mentoring field to date, the overlap between social class, race, and ethnicity in mentoring relationships, and the fact that the latter two have received much more attention, it would make sense for mentoring programs interested in focusing more on social class to review recommended best practices for increasing the racial and ethnic competency of mentors (see Sánchez et al., this volume, Chapter 10; Sánchez & Colón, 2005). Issues related to the culture of class could easily be included when training mentors to be culturally competent. For example, when introducing mentors to common racial “microaggressions” that mentees of color might experience, attention could also be paid to social class microaggressions (i.e., prejudices about housing, music, dress) and how to recognize and talk about them with their mentees. Educating mentors about the role of social structure and institutional prejudice regarding class may also be appropriate. Given the paucity of attention paid to social class issues in the mentoring field, we propose three directions for incorporating the role of social class in mentoring. These directions, and associated guiding questions that can be used to address social class issues in mentoring (see Table 12.1), are the focus of the remainder of this section of the chapter. First, programs should focus more intentionally on social class issues in prematch mentor training. Research suggests that an underlying mechanism in prejudice reduction is reducing anxiety and perception of threat (Pettigrew & Tropp, 2006). Mentors are more likely to feel anxious or threatened by social class differences if they are not sure how they will be perceived or accepted by their mentee. Prematch mentor training might focus on the interrelationships among SES, class, and culture as well as how to reduce the common biases and prejudices members

184  CULTURAL PERSPECTIVES Table 12.1   Guiding Questions and Recommendations for Practice Guiding Questions

Recommendations

Does the mentoring program see value and status of mentee’s social class as equal?

Social class differences should be viewed by mentoring programs and their staff as differences, not as deviancy or deficits. Attention should be paid as much to mentee’s strengths as to risks. Programs should promote mentoring characterized by curiosity and interest vs. judgment and reform. Programs should provide ongoing training and support to mentors for navigating class differences respectfully. Training may also be useful for mentees and parents to help identify and bridge class differences. Training and support should address both (a) material factors of social class (e.g., differential access to resources) and cultural factors (e.g., different values, practices, beliefs) and (b) class “microaggressions,” stereotyping, and prejudice. Programs should solicit feedback from members of the mentee’s community and social class (e.g., through use of an advisory council, focus groups, and/or having staff members who come from the same communities and social class backgrounds as the mentees). Programs should assess the degree to which mentors are culturally empathic as part of mentor and program evaluation (for possible measures, see Sánchez et al., this volume, Chapter 10).

Do mentors see value and status of mentee’s social class as equal to their own?

Through mentor training and suggested mentor-mentee activities, help mentors appreciate and value mentees’ social class culture (e.g., music, food, dress). Provide ongoing mentor training and support to ensure that mentors understand that differences are not an indication of deviance or deficit. Offer or provide suggested lists of activities, outings, and events that allow the mentor to engage in the mentee’s community as much as mentee engages in the mentor’s. Help the mentor to pay attention to the values and expectations of the mentee and the mentee’s family with regard to the mentoring relationship by communicating with parents on an ongoing basis and providing training and support for the mentor to do the same.

Do mentors and mentees develop common goals for the relationship that are respectful of the cultural aspects of social class?

Mentors and mentees should be supported in establishing and pursuing goals that lead to learning from each other and enhance cultural tool kits for both mentor and mentee (for further recommendations regarding goal setting and pursuit, see Balcazar and Keys, this volume, Chapter 6).

Do mentors prioritize collaboration and capacity-building with mentee?

Provide training and support to mentors on how to adopt a developmental (i.e., mentee-driven) vs. prescriptive (i.e., mentor-driven) approach in decision making. Ensure that mentor training emphasizes adopting a strengths-based vs. reform approach to mentee capacity-building. Provide mentors with suggestions for strategies and activities that allow opportunity and support for mentee’s autonomy development.

Social Class  185 of one social class may hold toward another. This could help provide a foundation for mentors to engage with intergroup contact processes to promote positive relationship processes (see Figure 12.1). Attention also should be paid to understanding both material factors (e.g., the reality of living with sparse financial resources) and cultural factors (e.g., how culturally textured practices, beliefs, and values of members of different social classes are different, not deviant) related to social class. Second, programs should provide mentors ongoing support for addressing issues related to social class throughout the course of their relationship. Since class issues related to social location are so rarely addressed directly in society at large, it is important for programs to provide mentors and mentees with prompts and support for reflecting on these issues. Mentors have cited feeling overwhelmed with the magnitude of issues faced by their mentees’ families as one reason for their mentoring dissatisfaction (R. Spencer, Basualdo-Delmonico, & Lewis, 2011). Such issues are likely often a consequence of mentees’ families’ financial circumstances. It would be important not only to include this focus in initial training but also to provide opportunities for mentors to return to a discussion of these issues as the mentoring relationship deepens. To ensure that the importance of addressing social class issues is conveyed, mentoring programs could formally assess the degree to which the mentor training and support they provide around social class issues increases mentors’ comfort with their mentees over the course of mentoring. Third, programs should provide mentors with a model for the development of mutual understanding and respect for issues related to SES and class. It is easy for high-SES mentors to assume that their primary goal should be to provide social capital to mentees living in underresourced families or communities. It is possible that a mentor’s motivation to “do good” or “give back” could lead to trying to “fix” the mentee, a deficit approach that will skew the power in the relationship. For cross-class pairing, approaching mentoring as a two-stage process could help mentors avoid such pitfalls (see Figure 12.1). Following ICT, the mentor would focus first on reducing social class prejudice and then, based on empowerment theory, next focus on the youth’s competencies rather than deficits. This is not to suggest that same-class pairs never have to consider both stages. As noted, social class encompasses both material and cultural aspects, and sharing aspects of one does not mean you necessarily share aspects of the other. For all mentors, understanding the capacities that youth already have in their “cultural tool kit” (Swidler, 1986) and focusing on valuing and adding

to them rather than replacing them offers a strengthsbased approach.

Conclusion Although there is little empirical evidence as to whether or how social class influences mentoring, a growing body of work suggests that it might. We propose that programs should consider the population of youth they serve, as well as the mentors they recruit, and ask themselves whether and how social class may be relevant to their activities. We suspect that many programs would benefit from considering both the material and the cultural aspects of social class more explicitly in how they approach staff, mentor, and mentee recruitment and training. In her classic book on mentoring today’s youth, Rhodes (2002) noted that “when the tool of change is a close relationship, everyone involved should proceed with caution” (p. 64). In this chapter, we have suggested that social class offers a cultural context for understanding youth development; it shapes the environment in which youth and their families live as well as their values, beliefs, and practices. Given that mentors are often from a different social class than the youth they mentor and could easily construe their role as providing a bridge to resources in the mentor’s world, we think caution is warranted. We suggest that mentoring programs should provide mentors with support and the opportunity to attend to any cultural differences they and their mentee may have, including class differences, in order to prevent potential misunderstandings. We have proposed a two-stage model of relationship development (see Figure 12.1) that suggests that the more cultural differences between mentors and mentees, the more mentors should first establish a sense of equal status and appreciation of cultural differences with their mentees before embarking on a change agenda. Social class remains an uncomfortable topic in the United States. Yet leaving it “untalked about” (hooks, 2000, p. 31) will not make it less real. We believe that for all mentors, whether they are in same- or crossclass matches, taking social class seriously and attending to its possible influence on youth, as well as on their own relationships with youth, should be an important part of the practice of mentoring.

References Allport, G. W. (1954). The nature of prejudice. Oxford, UK: Addison-Wesley. Bernstein, L., Rappaport, C. D., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S.

186  CULTURAL PERSPECTIVES Department of Education’s Student Mentoring Program. Jessup, MD: National Center for Education Evaluation and Regional Assistance. Bordeau, P. (1987). What makes a social class? On the theoretical and practical existence of groups. Berkeley Journal of Sociology, 32, 1–17. Bordeau, P. (1998). Practical reason: On the theory of action. Oxford, UK: Polity Press. Cohen, A. B. (2009). Many forms of culture. American Psychologist, 64, 194–204. Croizet, J. C., & Claire, T. (1998). Extending the concept of stereotype and threat to social class: The intellectual underperformance of students from low socioeconomic backgrounds. Personality and Social Psychology Bulletin, 24, 588–594. Deutsch, N. L. (2008). Pride in the projects: Teens building identities in urban contexts. New York: New York University Press. Deutsch, N. L., & Theodorou, E. (2010). Aspiring, consuming, becoming: Youth aspirations, gender roles, and identity. Youth & Society, 42, 229–254. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. In G. Noam & J. E. Rhodes (Eds.), A critical view of youth mentoring (new directions for youth development): Theory, research, and practice (Vol. 93, pp. 21–57). San Francisco: Jossey-Bass. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., & Silverthorn, N. (2005). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524. Duncan, G. J., & Magnuson, K. A. (2003). Off with Hollingshead: Socioeconomic resources, parenting, and child development. In M. H. Bornstein & R. H. Bradley (Eds.), Socioeconomic status, parenting, and child development (pp. 83–106). Mahwah, NJ: Erlbaum. Entwisle, D. R., & Astone, N. M. (1994). Some practical guidelines for measuring youth’s race/ethnicity and socioeconomic status. Child Development, 65, 1521–1540. Erickson, L., McDonald, S., & Elder, G. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367. Freedman, M. (1993). The kindness of strangers: Adult mentors, urban youth, and the new voluntarism. San Francisco: Jossey-Bass. Gaddis, M. (in press). What’s in a relationship? Examining race, class, and contact time as determinants of social capital in mentoring relationships. Social Forces.

Garcia-Coll, C., Lamberty, G., Jenkins, R., McAdoo, H. P., Crnic, K., Wasik, B. H., & Garcia, H. V. (1996). An integrative model for the study of developmental competencies in minority children. Child Development, 67, 1891–1914. Gershoff, E. T., Aber, J. L., Raver, C. C., & Lennon, M. C. (2007). Income is not enough: Incorporating material hardship into models of income associations with parenting and child development. Child Development, 78, 70–95. Gordon, M. M. (1949). Social class in American sociology. American Journal of Sociology, 55, 262–268. Gorski, P. C. (2005). Savage unrealities: Uncovering classism in Ruby Payne’s framework. Unpublished manuscript. Retrieved from http://www.edchange .org/publications/SavageUnrealities.pdf Grusky, D. B., & Weeden, K. A. (2008). Are there social classes? A framework for testing sociology’s favorite concept. In A. Lareau & D. Conley (Eds.), Social class: How does it work? (pp. 65–89). New York: Russell Sage Foundation. Hattie, J. (2009). Visible learning: A synthesis of over 800 meta-analyses relating to achievement. New York: Routledge. Hauser, R. M. (1994). Measuring socioeconomic status in studies of child development. Child Development, 65, 1541–1545. Herrera, C., Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, J. (2008). High school students as mentors: Findings from the Big Brothers Big Sisters school based mentoring impact study. Philadelphia: Public/Private Ventures. Herrera, C., Sipe, C. L., & McClanahan, W. S. (2000). Mentoring school-age children: Relationship development in community-based and school-based programs. Philadelphia: Public/Private Ventures. Hirsch, B. (2005). A place to call home: After-school programs for urban youth. Washington, DC: American Psychological Association. Hirsch, B., Deutsch, N. L., & DuBois, D. (2011). After-school centers and youth development: Case studies of success and failure. New York: Cambridge University Press. hooks, B. (2000). Where we stand: Class matters. New York: Routledge. Hout, M. (2008). How class works: Objective and subjective aspects of social class. In A. Lareau & D. Conley (Eds.), Social class: How does it work? (pp. 25–64). New York: Russell Sage Foundation. Hur, M. H. (2006). Empowerment in terms of theoretical perspectives: Exploring a typology of the process and components across disciplines. Journal of Community Psychology, 34, 523–540. Huston, A. C., & Bentley, A. C. (2010). Human development in societal context. Annual Review of Psychology, 61, 411–437. Kingston, P. W. (2000). The classless society. Stanford, CA: Stanford University Press. Lareau, A. (2003). Invisible inequality: Social class and child rearing in black and white families. American Sociological Review, 67, 747–776.

Social Class  187 Lareau, A. (2008). Introduction: Taking stock of class. In A. Lareau & D. Conley (Eds.), Social class: How does it work? (pp. 3–24). New York: Russell Sage Foundation. Lawrence, E., Sovik-Johnston, A., Roberts, K., & Thorndike, A. (2010). The Young Women Leaders Program mentor handbook (6th ed.). Charlottesville, VA: Young Women Leaders Program. Liang, B., & West, J. (2007). Youth mentoring: Do race and ethnicity really matter? In J. E. Rhodes (Ed.), Research in Action Series (Issue 9, pp. 3–13). Alexandria, VA: MENTOR. Liu, W. M., Ali, S. R., Soleck, G., Hopps, J., Dunston, K., & Pickett, T., Jr. (2004). Using social class in counseling psychology research. Journal of Counseling Psychology, 51, 3–18. Manza, J., Hout, M., & Brooks, C. (1995). Class voting in capitalist democracies since World War II: Dealign­ ment, realignment, or trendless fluctuation? Annual Review of Sociology, 21, 137–162. MENTOR. (2005). How to build a successful mentoring program using the elements of effective practice: A step-by-step tool kit for practitioners. Alexandria, VA: Author. MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. MENTOR. (2012). For mentors. Retrieved from http:// www.mentoring.org/get_involved/for_mentors Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. Payne, R. (2005). A framework for understanding poverty. Highlands, TX: aha!Process, Inc. Pettigrew, T. (2004). Intergroup relations and national and international relations. In M. B. Brewer & M. Hewstone (Eds.), Applied social psychology (pp. 225–242). Malden, MA: Blackwell Publishing. Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90, 751–783. Philip, K. (2008). She’s my second mum: Young people building relationships in uncertain circumstances. Child Care in Practice, 14, 19–33. Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community. New York: Touchstone Books/Simon & Schuster. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43). Thousand Oaks, CA: Sage. Sánchez, B., & Colón, Y. (2005). Race, ethnicity, and culture in mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 191–204). Thousand Oaks, CA: Sage. Scott, J. (2008). Reflection on “Class matters.” In A. Lareau & D. Conley (Eds.), Social class: How does it work? (pp. 354–358). New York: Russell Sage Foundation.

Sims, V. M. (1951). A technique for measuring social class identification. Educational and Psychological Measure­­ ment, 11, 541–548. Snibbe, A., & Markus, H. (2005). You can’t always get what you want: Educational attainment, agency, and choice. Journal of Personality and Social Psycho­ logy, 88, 703–720. Spencer, M. B., Dupree, D., & Hartmann, T. (1997). A phenomenological variant of ecological systems theory (PVEST): A self-organization perspective in context. Development and Psychopathology, 9, 817–833. Spencer, R. (2006). Understanding the mentoring process between adolescents and adults. Youth and Society, 37, 287–315. Spencer, R. (2007a). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Spencer, R. (2007b). Naturally occurring mentoring relationships involving youth. In T. D. Allen & L. T. Eby (Eds.), The Blackwell handbook of mentoring: A multiple perspectives approach (pp. 99–117). Malden, MA: Blackwell Publishing. Spencer, R., Basualdo-Delmonico, A., & Lewis, T. O. (2011). Working to make it work: The role of parents in the youth mentoring process. Journal of Community Psychology, 39, 51–59. Spencer, R., Lewis, T., & Basualdo-Delmonico, A. (2007, October). Mentoring in a multicultural society: A complex relationship. Presentation at ISPRC Diversity Challenge Conference, Boston, MA. Sue, D. W., Capodilupo, C. M., Torino, G. C., Bucerri, J. M., Holder, A. M., Nadal, K. L., Esquilin, M. (2007). Racial microaggressions in everyday life. American Psychologist, 62, 271–286. Swidler, A. (1986). Culture in action: Symbols and strategies. American Sociological Review, 51, 273–286. Syed, M., Goza, B. K., Chemers, M. M., & Zurbriggen, E. L. (2012). Individual differences in preferences for matched-ethnic mentors among high-achieving ethnically diverse adolescents in STEM. Child Development, 83, 896–910. Taylor, P., Morin, R., Cohn, D., Fry, R., Kochhar, R., & Clark, A. (2008). Inside the middle class: Bad times hit the good life. Washington, DC: Pew Research Center. Werner, E. (1995). Resilience in development. Current Directions in Psychological Science, 4, 81–85. Wohlfarth, T. (1997). Socioeconomic inequality and psychopathology: Are socioeconomic status and social class interchangeable? Social Science & Medicine, 45, 399–410. Zimmerman, M. A. (2000). Empowerment theory: Psychological, organizational, and community levels of analysis. In E. Seidman (Ed.), Handbook of community psychology (pp. 43–63). Dordrecht, the Netherlands: Kluwer Academic Press. Zimmerman, M. A., Bingenheimer, J. B., & Behrendt, D. E. (2005). Natural mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 143–157). Thousand Oaks, CA: Sage.

13  INTERNATIONAL AND CROSS CULTURAL ASPECTS IN YOUTH MENTORING Limor Goldner and Miri Scharf

Introduction Human development occurs within the context of cultural factors such as ideals, beliefs, values, norms, customs, and institutions. These social structures cognitively and emotionally construct the way people form their relationships (Bronfenbrenner & Ceci, 1994), as well as the different meanings that children and adolescents in different cultures ascribe to relationships, and their functions and roles in their lives (Bronfenbrenner & Ceci, 1994; Russell, 2007; Turnbull, Hernandez, & Reyes, 2009). Thus, cultures highlight different values such as self-reliance (Crystal, Kakinuma, DeBell, Azuma, & Miyashita, 2008) and the role of the immediate and extended family (Claes, Lacourse, Bouchard, & Luckow, 2001), which may also be reflected in the nature of mentoring relationships. Organized youth mentoring developed rapidly in the 1980s and 1990s in Canada, the United Kingdom, Australia, New Zealand, and Israel. Recently various states in South America and Western and Eastern Europe have started to develop one-on-one mentoring programs such as Big Brothers Big Sisters and others (Bodin & Leifman, 2011; Brady & O’Regan, 2009; Brumovská, 2007; Evans, Jory, & Dawson, 2005; Liabo, Lucas, & Roberts, 2005; Miller, 2002; Russell, 2007; Thurman et al., 2008). In addition, organized mentoring has begun to be adopted as a strategy to target social and health problems in developing countries (e.g., Kenya and Uganda) and in Far East nations, such as Korea, to deal with the risks of anomie, marginalization, and poverty (Shin & Rew, 2010; Thurman et al., 2008).

Several factors may account for the growing global popularity of mentoring. Nowadays, childhood and adolescence around the world are embedded in a relatively complex context that includes a shift in responsibility for welfare from the state to individuals, uncertainty regarding the future of the nuclear family, reliance on virtual relationships rather than face-to-face relationships, and confusion regarding identities and moral ethics (Goldner & Scharf, 2012; Philip & Hendry, 2000). Furthermore, the trend toward volunteerism among the middle class, and the desire of middle-aged adults to transfer their knowledge, traditions, and cultural values to youngsters also contributes to this direction (Freedman, 1992). Canada, reported to be the world leader in the percapita percentage of people involved in tutoring and mentoring, is an exceptional example of this tendency with more than 5,200 of these types of programs managed by educational and or nonprofit organizations (Miller, 2002). Finally, the desire for simple, flexible, and relatively inexpensive ways to help young individuals also encourages governments to adopt mentoring initiatives as a strategy for addressing social problems (Freedman, 1992). In fact, the increase in mentoring programs worldwide can be seen as part of the tendency toward privatization, where the state transfers some of its social responsibilities to nonprofit organizations and private volunteer and business organizations (Gidron, 2007). Two prominent examples of government support for expansion of youth mentoring can be seen in Britain (Colley, 2003; Hall, 2003; Meier, 2008; Miller, 2002; Philip, 2003; Philip, Shucksmith, & King, 2004) and in

189

190  CULTURAL PERSPECTIVES Israel (Fresko & Wertheim, 2001, 2006; Goldner & Mayseless, 2009). The prevalence of programs in various countries underscores the need to adopt theoretically guided culture-sensitive approaches in the field of youth mentoring. From a theoretical point of view it is important to identify both universal and culturespecific aspects of natural and organized mentoring relationships and to direct programs toward supporting these. Identifying similarities and differences in the mentoring experiences of young people across cultures could shed light on the main processes and the dynamics that contribute to the development of successful mentoring relationships. From a practical angle, a deeper understanding of the universal and culture-specific aspects of mentoring relationships can help practitioners to shape mentoring programs that suit a specific culture as well as to improve the training of mentors to achieve the best practice within the particular culture. This could promote better utilization of the resources invested in mentoring programs around the world. This is especially important given the small effect size typically found for the efficacy of organized youth mentoring (DuBois, Holloway, Valentine, & Cooper, 2002; DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011), and the understanding that even the best models of mentoring interventions are likely to be more effective in certain contexts and groups than in others (Rhodes, 2008; Rhodes & Lowe, 2009). Thus, our hypothesis is that mentoring relationships and programs will be more effective the more they are aligned with the values of the culture within which the youth is embedded (Ferrer-Wreder, Sundell, & Mansoory, 2012). Consequently, this chapter begins by addressing two main questions from a theoretical perspective. First, we explore the influence of cultural values in shaping the goals of mentoring relationships and programs. Second, we examine the potential influence of cultural values on the nature of the mentoring relationship. We then critically review research on youth mentoring from various cultures, and then end the chapter by discussing the implications for practice.

Theory In this section, we use three main theories as a theoretical framework within which to examine the influence of cultural values in shaping the goals and the nature of the mentoring. The first two theories, self-determination theory (Deci & Ryan, 2000a,

2000b) and regulatory focus theory (Higgins, 1997, 1998; Higgins et al., 2001), refer to the individual level, whereas the third theory, Schwartz’s three cultural dimensions value types (Schwartz, 1999), refers to the cultural and societal levels. In the mentoring realm, these theories may be useful for understanding cultural similarities and differences in the goals of mentoring relationships and programs, identities of those who become involved in mentoring, and characteristics of the mentoring relationship, including the type of mentoring activities. Theoretical Frameworks Self-determination theory (Deci & Ryan, 2000a, 2000b) posits that human motivation to initiate certain behaviors is linked to three basic universal human needs: autonomy, relatedness, and competence. According to the theory, these needs are essential for ongoing psychological growth, integrity, and well-being and serve to direct people in their behaviors. From the perspective of self-determination theory, an individual can develop in a healthy manner and achieve well-being only when all three needs are satisfied (Deci & Ryan, 2000a, 2000b). Autonomy refers to people’s drive to establish themselves as separate, independent individuals and to follow their inner wills, interests, values, and choices when acting, regardless of whether these actions are initiated independently or in response to the requests of others. Relatedness refers to individuals’ needs to achieve a sense of belonging and embeddedness by interacting with, connecting to, and caring for other people, whereas competence refers to people’s inherent desires to deal effectively with the environment and to achieve a sense of self-worth and validation (Deci & Ryan, 2000a, 2000b; Deci & Vansteenkiste, 2004). Although self-determination theory views the above needs as universal, the degree to which, and how, these needs are fulfilled or frustrated in different social contexts is assumed to vary (Deci & Ryan, 2000a, 2000b). According to regulatory focus theory (Higgins, 1997, 1998), people’s goal-directed behavior is regulated by two distinct motivational systems. These two systems, termed promotion and prevention, each serve a distinct survival function. The human promotion system is concerned with obtaining nurturance and underlies higher-level concerns with accomplishment and advancement. The system’s concerns relate to the pleasurable presence of positive gains and accomplishments. In contrast, the human prevention system is concerned with obtaining security, safety, and fulfillment of responsibilities. The prevention system’s concerns relate to the

International and Cross Cultural Aspects in Youth Mentoring    191 pleasurable absence of negative outcomes and the painful presence of losses. Regulatory focus theory proposes that the two systems use distinct means of regulating desired end-states and that individuals will vary in the degree to which one or the other system is their primary focus (Higgins, 1997, 1998). Whereas individuals with a promotion focus are inclined to use approach strategic means to attain their goals, individuals with a prevention focus tend to use avoidance strategic means (Higgins et al., 2001). Although Higgins and colleagues used this theory to understand individual achievement orientation at the psychological level, we believe that it could be useful for considering how different cultures may emphasize to differing degrees concerns relating to promotion or prevention in ways that, in turn, could be manifest in mentoring goals and practices. Schwartz (1999) referred to cultural dimensions value types and identified three dimensions of values, each comprising two poles that represent the implicitly or explicitly shared abstract ideas about what is desirable in a society. The first dimension describes the relationship between individuals and groups and is termed individualism versus collectivism. This contrast refers to two major questions: whose interests should take precedence and to what extent are individuals autonomous or embedded in their group. It is also described as contrasting independenceinterdependence and autonomy-relatedness. One pole of the dimension describes societies in which individuals are deeply rooted in their society and find their meaning largely through social relationships, identifying with their group, and participating in shared community activities. It captures qualities of conservatism, social order, security, and respect for tradition and family norms. People are socialized to meet the external demands of the society; to follow traditional customs; and to search for security, belonging, and embedding within the in-group. The opposite pole describes cultures in which individuals are viewed as autonomous. Individuals do not feel attached to any in-group; rather, there are numerous in-groups to which they can be attached and they find meaning in expressing their own uniqueness, preferences, traits, feelings, and motives (Schwartz, 1999). The second dimension refers to the relationship between individuals and their environment and encompasses qualities of mastery versus harmony. Cultures that value mastery emphasize qualities of competence and success, whereas cultures that value harmony highlight qualities of unity with nature, protection of the environment, and prevention of damage. The third dimension focuses on the power relationships between individuals and society and

refers to the ways in which societies ensure responsible behaviors that preserve their social fabric. Cultures that value power differences rely on hierarchical systems. These cultures emphasize an unequal distribution of power, roles, and resources. In contrast, cultures that focus on egalitarianism emphasize the transcendence of selfish interests in favor of the welfare of others and qualities of equality, social justice, freedom, and honesty. Implications for Mentoring in Different Cultural Contexts The preceding theories offer rich lenses through which to consider both similarities and differences in mentoring relationships and programs that may be evident across different cultural contexts. The cultural dimensions value types emphasized by Schwartz, for example, may be manifested in the goals of the mentoring and the various ways in which societies perceive protégés’ development and well-being (Al-Issa, 1995; Sagiv & Schwartz, 2000; Triandis, Bontempo, Villareal, Asai, & Lucca, 1988). For example, collectivistic societies may use the mentoring relationship as a mechanism for increasing protégés’ sense of belonging and relatedness, reducing their sense of alienation from the community, and strengthening their social skills and cooperation (satisfying relatedness needs according to self-determination theory). One of the main objectives will be promoting social cohesion and belongingness and preventing situations such as social disintegration (reflecting one of the goals emphasized in regulatory focus theory). In contrast, individualistic societies might perceive mentoring as a vehicle for achieving a sense of competence and independence. Mentoring will be used as a means to foster self-competence, self-worth, and personal success, and its major orientation will be toward promoting the autonomy of the individual. The relationship between individuals and society may be manifested in the mentoring practices and activities. For instance, in traditional cultures that emphasize hierarchical relationships, centralism, and relatedness and that underscore the superior role of adult authority figures, mentoring may be characterized by an asymmetrical relationship between mentors and protégés. In these cultures, mentors may play a more active and dominant role in the relationship than the protégés, especially in shaping the goals of the mentoring and making decisions concerning activities. Conversely, in cultures that underscore aspects of egalitarianism (Schwartz’s theory) and encourage autonomy and self-reliance and competencies (self-determination

192  CULTURAL PERSPECTIVES theory), the relationship between mentors and protégés may be more mutual and collaborative, with protégés taking major responsibility in shaping the goals of the mentoring, making decisions, and initiating activities. Moreover, based on the literature in the field of psychotherapy (Al-Issa, 1995; Triandis et al., 1988), there may be differences associated with cultural values in the ways of communication between mentors and protégés and the type of emotional provisions and support that mentors provide to protégés. For example, traditional and collectivistic cultures that favor the welfare of the community over that of the individuals and that value the restraint of strong feelings over aspects of intimacy (Al-Issa, 1995; Triandis et al., 1988) may discourage protégés from exposing their inner world and personal problems. In such relationships, protégés and mentors may focus more on aspects of cognitive and behavioral support than on emotional support, given that selfdisclosure of the individual’s (protégé’s) problems and feelings may be perceived as a threat to the cohesiveness of the family and the community. Furthermore, in such cultures mentoring meetings may be highly structured and include tangible and instrumental forms of support. In contrast, aspects of intimacy, companionship, and unity may be highly valued in more individualistic cultures that value aspects of general competence and development and assume that children’s promotion and development occurs in the context of warm relationships with adult caregivers. In these cultures, mentoring may focus on aspects of emotional support, as the self-disclosure of problems and feelings may be perceived as a key process in fostering relationships (Goldner & Mayseless, 2008; Miller, 2002). As a result, mentoring relationships may be relatively loose and experienced as a special kind of friendship (Goldner & Mayseless, 2008; Goldner & Scharf, 2012; Spencer, 2006, 2007).

Research In this section, we review studies of youth mentoring outside of the United States. Several studies directly address cross-cultural aspects of youth mentoring, whereas others address the success of a mentoring program or mentoring relationships within a particular culture and refer coincidentally to the social context. Studies were retrieved from PsycINFO, ERIC, and Google Scholar databases, and by using Google to search for the terms crosscultural or international and youth mentoring, or the combination of the phrase youth mentoring with

names of specific countries. Our literature review is not intended to be comprehensive, but rather emphasizes those studies that related most closely to our theoretical frameworks of interest. The review is organized in three sections: characteristics of the mentors and protégés, characteristics of the relationship, and goals and success of the mentoring. Characteristics of Mentors and Protégés Several studies of youth mentoring contexts outside of the United States refer to mentors’ characteristics. Some studies refer to these characteristics directly by comparing youth across different cultures, whereas others relate the information anecdotally. These studies examined a variety of variables. What are the backgrounds and characteristics of the mentors? Do their developmental stage (e.g., adulthood; married) and their maturity relate to mentoring effectiveness, and what is their formal role in the community? These dimensions may reflect the salient values or the importance of various roles in specific cultural and societal contexts. For instance, in a cross-cultural, cross-sectional study that explored the different perceptions of natural mentoring among 502 Chinese and 201 American 11th graders, the researchers found differences between the two cultures (Chen, Greenberger, Farruggia, Bush, & Dong, 2003). Chinese adolescents tended to identify their teachers as very important nonparental adult persons (VIPs) in their lives, whereas American adolescents were more likely to name church representatives and coaches as VIPs. Moreover, Chinese adolescents’ VIPs were older than the VIPs of the American adolescents. Another cross-sectional study conducted in Hong Kong among 124 adolescents aged 12–15 years reported findings suggesting that, unlike married U.S. mentors with children, who had been found to be at the greater risk of relationship termination (Grossman & Rhodes, 2002), those in Hong Kong who fit this profile were more likely to mentor effectively (Chan & Ho, 2008). In contrast and contrary to the expectations, no indications for a more central role of collectivistic values and the centrality of adults (Crystal et al., 2008; Rothbaum, Weisz, Pott, Miyake, & Morelli, 2000) in the Japanese culture were found in the cross-sectional study by Darling, Hamilton, Toyokawa, and Matsuda (2002) among 365 American and Japanese college students. Students in both cultures were most likely to choose adults (mainly parents) and relatives as their natural mentors. In some cases, the selection of the mentors may reflect a central place of companionship in views of

International and Cross Cultural Aspects in Youth Mentoring    193 mentoring within a culture. For example, the majority of mentors in a study of the Canadian Big Brother Big Sister organized mentoring program were young adults under the age of 30, single or never married (De Wit et al., 2007), which might offer a basis for cultivating more egalitarian and friendship-oriented relationships. Similarly, the nationally sponsored Perach program in Israel, which uses university and college students as mentors, underlines the resemblance between mentors and protégés, since both populations are close in age and dealing with academic challenges, one in elementary or high school and the other in the higher education system (Fresko & Wertheim, 2001, 2006; Goldner & Mayseless, 2009). In addition, an apparent trend toward the extensive use of business people as mentors in Canada and Britain may reflect the trend toward the use of paid professional mentors in Britain (Colley, 2003; Miller, 2002; Philip, 2003; Philip et al., 2004). Similarly, in Israel, the use of university and college students as mentors in return for scholarships is widespread across the country (Fresko & Wertheim, 2001, 2006). The rationale for drawing upon these groups as mentors hinges on the notion that protégés are lacking in some ways. To gain cultural capital and to extricate them from their adverse settings, paid mentors, or those with successful careers, can provide valuable assistance and role modeling for young people based on their own successful lives, thus advancing them a step into the competitive labor market (Philip, 2003). This notion aligns well with the focus on competence and individualism in these competitive societies, as highlighted in Schwartz’s typology and in self-determination theory. Protégés’ characteristics have usually been mentioned incidentally when researchers describe their study samples. In most cases, the protégé populations are described as children or youth at risk, and the mentoring programs emphasize aspects of prevention of future emotional distress, risk situations and delinquency, as well as promotion of functioning and increased competence. Following regulatory focus theory, the specific focus of concern often reflects a broader challenge characteristic of the society. For example, the protégés comprising the samples of numerous studies conducted in Africa are described as traumatized orphans, children, and youth who have lost their parents to AIDS or genocide (Onuoha, Munakata, Serumaga-Zake, Nyonyintono, & Bogere, 2009; Turnbull et al., 2009; USAID, 2007). For instance, Turnbull et al.’s cross-sectional study used quantitative and qualitative methods to explore the perceptions that mentors (N = 171, age

range = 25–44 years) had regarding their orphan protégés in Rwanda. Data indicated that some mentors felt that their protégés carried the bad deeds of their parents with them and caused problems in the community. In addition, most mentors felt that the community bore the responsibility for its orphan citizens. These findings highlight the community’s perceived responsibility to prevent future social disintegration and maintain cohesiveness (Turnbull et al., 2009). The attempt to prevent future negative outcomes associated with risk also emerges in westernized countries such as Israel and Britain. In Britain, mentoring interventions are usually used as a way to offer young people who are socially excluded, involved in the criminal justice system, or in risk situations a chance to build a better life (Meier, 2008). In Israel, the Perach project targets a relatively younger population of protégés, elementaryschool-aged children, to enable them to bridge gaps between disadvantaged and advantaged children in Israeli society as well as to help them realize their inner potential (Fresko & Wertheim, 2001, 2006). Characteristics of the Relationship Most studies that reference cross-cultural aspects in natural and organized mentoring have focused on the characteristics of the relationship and underscored the widespread notion of mentoring as a warm, caring relationship. For instance, Vazsonyi and Snider (2008) used a cross-sectional design to explore the differences among 2,735 Swiss apprentices and 368 U.S. part-time high-school employees (mean age = 16.6 years), and to examine the associations between the mentoring experience and protégés’ adjustment. They found no variation between the two cultures. In both settings, the high quality of the mentoring relationship, incorporating dimensions of support and role modeling, was associated with more favorable trends in the adolescents’ adjustment (Vazsonyi & Snider, 2008). The dominance of relationship quality was also found in a qualitative study (Brumovská, 2007) conducted in the context of three mentoring programs in the Czech Republic and in Sweden using semistructured interviews with mentors and coordinators (N = 14, age range = 26–31 years). The creation of a strong emotional bond was concluded to be a key feature in the development of successful mentoring relationships in both places. This type of relationship was characterized by feelings of closeness and trust, a relaxed atmosphere, and mutual enjoyment and, on the part of mentors, involvement, motivation, and satisfaction within the relationship. However, Swedish

194  CULTURAL PERSPECTIVES program staff tended to describe the mentor’s role as that of a professional caregiver and part of the child’s social welfare service, whereas Czech mentors were more apt to be perceived as friends (Brumovská, 2007). The Swedish emphasis on professionalism in the mentoring relationship may relate to the centrality of values of self-promotion, autonomy, and competence within this culture (as described in regulatory focus theory and self-determination theory), whereas the prominence of friendship within the Czech culture may underscore the value of, and need for, relatedness and belonging. In addition, and contrary to the researchers’ hypotheses, the earlierreferenced cross-cultural study of natural mentoring among Japanese and American college students (Darling et al., 2002) found no difference in the reported characteristics of the relationships between the two groups of adolescents. In both cultures, aspects of role modeling and respect were salient. Similar aspects of mentoring relationships were also indicated to be important in a study conducted in Hong Kong using quantitative and qualitative methods (Chan & Ho, 2008). Findings emphasized the significance of several qualities of mentoring relationships—such as sensitivity, availability, and trust—to the success of the mentoring. Finally, Philip and colleagues (Philip, 2008; Philip et al., 2004) conducted qualitative research using observation, individual and group interviews, and case studies of protégés, mentors, and managers in an attempt to learn about the main mechanisms in organized mentoring relationships in Scotland. Findings indicated that the close friendly nature of the relationship and the “ability to share a laugh” with the mentor distinguished the relationship from other kinds of relationships. The informal and relaxed interactions that occurred between the mentor and the protégé often constituted a “moment of meeting” between them and a turning point in strengthening the relationship. Furthermore, studying the various types of natural relationships in Britain, Philip and Hendry (1996, 2000) identified more heterogeneous types of mentoring than the classic one-on-one relationship (Keller & Pryce, 2010) that is more vertical in nature (Philip, 2003, 2008; Philip & Hendry, 1996, 2000). Examples of such relationships are friend-to-friend mentoring or peer-group mentoring between protégées of the same age. According to Philip and colleagues, at least within British culture, mentoring relationships are better understood as a set of processes among youth, adults, friends, and peers rather than as a discrete relationship between an older experienced adult and an inexperienced novice (Philip, 2003, 2008; Philip & Hendry, 1996, 2000).

The cross-cultural study by Chen et al. (2003), of natural mentoring among Chinese and American adolescents, emphasized the different types of support provided by VIPs, based on self-report questionnaires regarding the identity and function of the VIPs. Whereas the VIPs of Chinese youth provided qualities of emulation, role modeling, and academic support, American youths’ VIPs provided general support, fun, and companionship. In both cultures, only a few adolescents reported that their VIPs provided support for individuality. The centrality of education in the Chinese culture, and the emphasis on learning from older people, may be reflected in the focus on role modeling and academic change in natural youth mentoring within this culture (Chen et al., 2003). In terms of the values discussed by Schwartz (1999), it appears that natural mentoring relationships for young people in Chinese society may integrate qualities of individualism and selfpromotion (promoting academic achievements) with a traditional and centralistic point of view (preferring role modeling and instrumental support), perhaps reflecting the transition from communal to individual-industrial society taking place in the Chinese society (Chen et al., 2003). Mentoring Goals and Success Numerous studies have explored the goals and success of mentoring as an intervention for youth in numerous countries. The studies have focused particularly on promoting health habits, preventing future health problems, and decreasing distress. For instance, a study conducted in Uganda and South Africa (Onuoha et al., 2009) examined the contribution of natural mentors to the psychosocial mental health of three groups of 10- to 17-year-old adolescents: orphaned adolescents whose parents died of AIDS (n = 373), other orphaned adolescents (n = 287), and adolescents who were not orphaned (n = 290). Adolescents who reported having a natural mentor showed significantly lower levels of child abuse, social discrimination, anxiety, and negative mental health factors and higher levels of social support, self-esteem, and positive mental health factors than did adolescents without natural mentors (Onuoha et al., 2009). Differences on these outcomes between mentored and nonmentored youth were most prominent among AIDS-orphaned adolescents, most of whom were without both of their biological parents. In a longitudinal randomized assigned study conducted in Rwanda (N = 692) that explored the effectiveness of a mentoring intervention that focused mainly on mitigating adverse psychological outcomes among 19- to 27-year-old household protégés

International and Cross Cultural Aspects in Youth Mentoring    195 who experienced the death of a parent, the researchers found a decrease in feelings of marginalization, grief, depressive symptoms, and maltreatment and an increase in perceived community support, compared to a nonmentored control group (USAID, 2007). Finally, in a nonequivalent control group pretest– posttest design study in South Korea that explored the effectiveness of a mentoring intervention in promoting sexual health knowledge and behaviors among middle school children in group and individual mentoring sessions, improvements were found in sexual knowledge and attitudes among the treatment group (n = 17) compared to a randomized assigned control group (n = 16; Shin & Rew, 2010). In a longitudinal study of 141 German protégés and a stratified control group of 158 children (mean age = 8 years), Drexler, Borrmann, and MüllerKohlenberg (2011) examined the effectiveness of mentoring programs to promote resilience and children’s psychological competencies (reflecting the promotion focus in regulatory focus theory, as well as the values of competence and autonomy according to self-determination theory). The researchers found an increase in protégés’ motivation to learn, classroom participation, and competence to solve tasks individually. Their teachers reported that they showed greater confidence when dealing with new tasks, compared to their counterparts. Overall, from these findings it could be inferred tentatively that in societies in which youth cope with achieving basic needs, such as physiological and psychological health, mentoring interventions that focus primarily on promoting health habits and preventing psychological distress have potential to be effective. In such contexts, mentoring interventions may have the potential to address secondary needs such as promotion of self-competence and autonomy, as described by the self-determination and regulatory focus theories, only after more primary needs are addressed. In contrast, as illustrated by Drexler et al.’s (2011) research, in societies in which youths’ basic needs are more likely to be met, these latter needs, such as competence, may be able to be addressed successfully as a more primary focus of mentoring programs. Summary and Directions for Future Research In sum, research on cultural aspects of youth mentoring is fairly limited. Most studies have not used a cross-cultural research design, and cultural aspects for the most part have not been examined directly as part of the conceptualization or design of studies. However, a close look at the available findings suggests that certain aspects of mentoring, such

as a close and warm relationship, may be universally important for the development of the youth. Existing research also points toward the potential significance of specific cultural values in selecting and screening mentors, such as collectivism (e.g., selecting older mentors, relatives, or authoritarian figures such as teachers), the selection of prevention goals (e.g., decreasing health or psychological problems) or promotion goals (e.g., enhancing self-competence or autonomy) in the mentoring program, and the relative emphasis placed on companionship or friendship versus more authoritarian and instrumental dimensions in mentoring relationships. Evidence of whether these types of values, needs, or priorities of the surrounding culture facilitate more successful relationships and youth outcomes is for the most part lacking and calls for future exploration. Likewise, more research is needed to examine whether emulating specific facets or aspects of a culture is desirable for promoting mentoring effectiveness. Future research could address this question in the context of cross-cultural research geared toward identifying universal and unique aspects of mentoring relationships and exploring their role in the developmental trajectories of young people across various cultures. Along the same lines, it could be useful to examine the challenges and possibilities of cross-cultural matches (i.e., those involving a youth and an adult from different cultural subgroups within a given society) compared to samecultural matches in promoting the development of children and youth. (For a related discussion of the potential benefits of cross-race mentoring relationships under certain circumstances, see Sánchez and colleagues, this volume, Chapter 10). Moreover, we strongly recommend that future research incorporate cross-cultural concepts and theories (i.e. individualism, collectivism, hierarchy, egalitarianism, prevention, self-promotion, relatedness, competence) as a frame of reference to facilitate examination of the role of culture in shaping mentoring processes and outcomes. In doing so, future studies should consider how measures of cultural factors highlighted as potentially important by these (and other) theoretical perspectives may act as mediating or moderating influences in processes linking mentoring relationships and program participation to youth outcomes within different cultural contexts.

Practice Recommendations concerning effective practices for youth mentoring, for the most part, do not specify how cultural values and norms should be

196  CULTURAL PERSPECTIVES incorporated into mentoring programs. Instead, these recommendations generally offer universal suggestions regarding optimal ways to develop and support mentoring relationships within programs, such as fostering close relationships, length of the relationship and frequency of the meetings, parents’ involvement, and training and supervision for the mentors (e.g., MENTOR, 2009; Miller, 2002; Rhodes & DuBois, 2008). These recommendations also usually lean on the Western traditions in helping relationships, such as therapy (e.g. Spencer & Rhodes, 2005). Moreover, it is our impression that when designers of mentoring programs actively consider cultural aspects, the focus of their attention tends to be on matching mentors and protégés on the basis of cultural similarity. According to this perspective, cultural similarity—which involves shared values, background, and likelihood of personal compatibility—might enhance feelings of closeness and the opportunity for role modeling (see Sánchez and colleagues, this volume, Chapter 10). Additionally, most practical guidance that is available relating to culture addresses the issues of race and ethnicity among minorities in the United States and does not refer to cultural values within different cultures outside the United States. The increasing prevalence of mentoring programs around the world, and the challenges and joys of multiculturalism, underscore a need for more culturally informed models and guidelines for youth mentoring. The use of Schwartz’s typology, self-determination theory (Deci & Ryan, 2000a), or regulatory focus theory (Higgins, 1997, 1998), for example, as a guiding framework when designing mentoring programs or training mentors could potentially enhance the attunement of both programs and mentors to cultural values and their implications for effective mentoring of youth in different contexts. When defining the goals of the mentoring, practitioners and mentors alike should also be aware of the similarities and differences between their own needs (competence, relatedness, and autonomy), concerns (promoting selffunctioning and competence or preventing distress and social disintegration), and values (promoting individualism and competence or enhancing protégés’ belonging to the community) and those of their protégés. Mentors and practitioners should also be cognizant of the different positions that cultures attribute to individuals and community as well as the varying ways people in different cultures may view the power relationships between adults and youth, when communicating with, or relating to, protégés.

Furthermore, cultural knowledge may be valuable not only when dealing with mentoring across country contexts, but also when dealing with cultural differences within a given society, particularly when mentors and protégés have the potential to be embedded in different cultures. Acknowledging the variety of cultural values may help mentors to face their misattributions and stereotypes; to respond in a sensitive and flexible way to their protégés’ cues; and to avoid conflicts, misunderstandings, and mistrust between themselves, their protégés, and their families. Mentors’ misinterpretation of their protégés’ behaviors with their own cultural bias, assumptions, and assessments (e.g., a mentor interprets his or her protégé’s passivity as resistance when, in fact, the behavior reflects compliance, respect for adults, or restraint of feelings; Carpenter, Rothney, Mousseau, Halas, & Forsyth, 2008) has the potential to lead to unnecessary conflicts and premature termination of the mentoring relationship, thereby unintentionally inflicting harm. In this respect, mentors who act without appropriate cultural sensitivity and awareness and programs that fail to provide mentors with adequate training and support in this realm may be committing “cultural malpractice.” Thus, the need to promote and practice cultural sensitivity can, in fact, be regarded as an ethical imperative (see Rhodes, Liang, & Spencer, this volume, Chapter 35). Overall, as we noted at the outset of this discussion, little attention has been given to developing guidelines or recommendations that consider cultural aspects in youth mentoring relationships and programs. Most programs developed and implemented outside of the United States appear to have addressed this topic mainly during the phase of matching protégés and mentors by emphasizing pairing on the basis of cultural similarity. However, incorporating a wide perspective on cultural aspects in shaping mentoring programs, practices, and relationships could address a crucial need in promoting the success of the mentoring and reducing its potential for psychological harm in diverse international contexts. Given this importance, several practical recommendations are suggested (see Table 13.1). First, we suggest conducting prematch and ongoing training on cultural sensitivity, training mentors to be attentive to the role of cultural values in shaping their relationships with their protégés using relevant theories such as self-determination theory, regulatory focus theory, and Schwartz’s typology as guidelines. Second, we recommend that, in designing mentoring programs for youth, care be taken to incorporate a wide perspective on how cultural aspects may shape mentoring goals and activities. In doing so, programs should strive to acknowledge the similarities

International and Cross Cultural Aspects in Youth Mentoring    197 Table 13.1   Checklist for Practitioners Topic

Recommendations

Training

Conduct prematch and ongoing training on cultural sensitivity. Support mentors with the process of acknowledging the role of cultural values in shaping their relationships with their protégés, using training informed by theories such as self-determination theory, regulatory focus theory, and Schwartz’s typology. Design training with the goals of enabling mentors to face their misattributions and stereotypes relating to cultural values; to respond in a sensitive and flexible way to their protégés’ cues; and to be equipped to avoid conflicts, misunderstandings, and mistrust between themselves, their protégés, and their protégés’ families.

Shaping goals and activities

In designing programs, work to incorporate a wide perspective on how cultural aspects may shape mentoring goals and activities. Strive to acknowledge the similarities and differences between programs’ and mentors’ needs and those of protégés, so as to reduce the potential for conflicts, mistrust, and misunderstanding. Consider using the principles and constructs of self-determination theory, regulatory focus theory, and Schwartz’s typology of cultural values to guide these efforts.

Ethics

Add a multicultural perspective to ethical codes for mentoring relationships and programs to diminish possible damaging behaviors and malpractice when selecting and screening mentors and working with youth and mentors both within and across diverse cultural contexts.

and differences between programs’ and mentors’ needs, concerns, and values and those of their protégés as well as to reduce the potential for conflicts, mistrust, and misunderstanding. Finally, we suggest adding a multicultural perspective to ethical codes for mentoring relationships and programs so as to diminish possible damaging behaviors and malpractice when selecting and screening mentors and working with youth and mentors both within and across different cultural contexts.

Conclusion In the past 30 years, organized youth mentoring has developed rapidly around the world. The increasing global prevalence of mentoring programs, and the challenges and joys of multiculturalism within cultures, underscore the need to develop more culturally informed theoretical models and guidelines for youth mentoring in order to maximize its effectiveness. Moreover, when programs and mentors act without appropriate cultural sensitivity and awareness, they may address protégés’ needs wrongly and may commit “cultural malpractice”

that harms protégés’ well-being. Thus, the need to promote and practice cultural sensitivity can, in fact, be regarded as an ethical imperative. We have posited three theories that scholars and practitioners can use as a framework for understanding and responding effectively to cultural differences in values (Schwartz’s three cultural dimensions value types) as well as in basic needs and motivations (selfdetermination theory) and goals or areas of concern for well-being (regulatory focus theory). The theories describe various values, needs, and goals that characterize cultures and contexts and may be used as a guiding framework when designing mentoring programs or training mentors in the ways that different cultures shape mentoring goals, relationships, and practices. This approach could enhance the attunement of both programs and mentors and, hence, their effectiveness. For example, in cultures that emphasize aspects of individuation and egalitarianism (within Schwartz’s theory), autonomy and competence (within self-determination theory), and a promotion focus (within regulatory focus theory), mentoring may be perceived as a vehicle for enhancing protégés’ sense of psychological well-being, competence, and autonomy. The relationships between mentors and protégés may be more mutual, intimate, and collaborative, with

198  CULTURAL PERSPECTIVES protégés taking an active role in shaping the goals and practices of the relationship. In contrast, cultures that emphasize aspects of collectivism, power differences, relatedness, and prevention concern may be more likely to see value in using the mentoring relationship as a mechanism for increasing protégés’ sense of belonging, promoting social cohesion, and preventing social disintegration. Charting such differences and their implications for mentoring relationship quality and program effectiveness in different cultural contexts should constitute a high priority as mentoring initiatives for young people continue to extend their global reach.

References Al-Issa, I. (Ed). (1995). Handbook of culture and mental illness: An international perspective. Madison: Inter­ national Universities Press. Bodin, M., & Leifman, H. (2011). A randomized effectiveness trial of an adult-to-youth mentoring program in Sweden. Addiction Research and Theory, 19, 438–447. Brady, B., & O’Regan, C. (2009). Meeting the challenge of doing an RCT evaluation of youth mentoring in Ireland. A journey in mixed methods. Journal of Mixed Methods Research, 3, 265–280. Bronfenbrenner, U., & Ceci, S. J. (1994). Nature-nurture in developmental perspective: A bioecological theory. Psychological Review, 101, 568–586. Brumovská, T. (2007). Mentors as mediators and significant adults [International Master of Science in Social Work thesis]. Goteborg: University of Goteborg. Carpenter, A., Rothney, A., Mousseau, J., Halas, J., & Forsyth, J. (2008). Seeds of encouragement: Initiating an aboriginal youth mentoring program. Canadian Journal of Native Education, 31, 51-68. Chan, C. C., & Ho, W. C. (2008). An ecological framework for evaluating relationship-functional aspects of youth mentoring. Journal of Applied Social Psychology, 38, 837–867. Chen, C., Greenberger, E., Farruggia, S., Bush, K., & Dong, Q. (2003). Beyond parents and peers: The role of important non-parental adults (VIPs) in adolescent development in China and the United States. Psychology in the Schools, 40, 35–50. Claes, M., Lacourse, E., Bouchard, C., & Luckow, D. (2001). Adolescents’ relationships with members of the extended family and non-related adults in four countries: Canada, France, Belgium and Italy. International Journal of Adolescence and Youth, 9, 207–225. Colley, H. (2003). Engagement mentoring for “disaffected” youth: A new model of mentoring for social inclusion. British Educational Research Journal, 29, 521–542. Crystal, D. S., Kakinuma, M., DeBell, M., Azuma, H., & Miyashita, T. (2008). Who helps you? Self and other sources of support among youth in Japan and the USA. International Journal of Behavioral Develop­ ment, 32, 496–508.

Darling, N., Hamilton, S. F., Toyokawa, T., & Matsuda, S. (2002). Naturally occurring mentoring in Japan and the United States: Social roles and correlates. American Journal of Community Psychology, 30, 245–270. De Wit, D., Lipman, E., Manzano-Munguia, M., Bisanz, J., Graham, K., Offord, D. R., . . . , Shaver, K. (2007). Feasibility of a randomized controlled trial for evaluating the effectiveness of the Big Brothers Big Sisters community match program at the national level. Children and Youth Services Review, 29, 383–404. Deci, E. L., & Ryan, R. M. (2000a). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68–78. Deci, E. L., & Ryan, R. M. (2000b). The “what” and “why” of goal pursuits: Human needs and the self determination of behavior. Psychological Inquiry, 11, 227–268. Deci, E. L., & Vansteenkiste, M. (2004). Self-determination theory and basic need satisfaction: Understanding human development in positive psychology. Ricerche di Psichologia, 27, 17–34. Drexler, S., Borrmann, B., & Müller-Kohlenberg, H. (2011). Learning life skills strengthening basic competencies and health-related quality of life of socially disadvantaged elementary school children through the mentoring program “Balu und Du” (“Baloo and you”). Journal of Public Health, 20, 141–149. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Evans, I. M., Jory, A., & Dawson, N. (2005). International: Australia and New Zealand. In D. L. DuBois & M. J. Karcher, (Eds.), Handbook of youth mentoring (pp. 408–421). Thousand Oaks, CA: Sage. Ferrer-Wreder, L., Sundell, K., & Mansoory, S. (2012). Tinkering with perfection: Theory development in the intervention cultural adaptation field. Child and Youth Care Forum, 41, 149–171. Freedman, M. (1992). The kindness of others: Adult mentors, urban youth, and the new voluntarism. Cambridge, UK: Cambridge University Press. Fresko, B., & Wertheim, C. (2001). Mentoring by perspective teachers as preparation for working with children at risk. Journal of Education for Teaching, 27, 149–159. Fresko, B., & Wertheim, C. (2006). Learning by mentoring: Prospective teachers as mentors to children atrisk. Mentoring & Tutoring, 14, 149–161. Gidron, D. (2007). Researching the third sector and civil society in Israel: Its history and current status. Civil Society and the Third Sector in Israel, 1(1), 7–18. (Hebrew). Goldner, L., & Mayseless, O. (2008). Juggling the roles of parents, therapists, friends and teachers—A working model for an integrative conception of mentoring. Mentoring & Tutoring, 16, 412–427.

International and Cross Cultural Aspects in Youth Mentoring    199 Goldner, L., & Mayseless, O. (2009). Quality of mentoring relationships and mentoring success. Journal of Youth & Adolescence, 38, 1339–1350. Goldner, L., & Scharf, M. (2012). Mentoring adolescents. In R. J. R. Levesque (Ed.), The encyclopedia of adolescence (pp. 1722–1729). New York: Springer. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effect of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Hall, J. C. (2003). Mentoring and young people: A literature review. SCRE Research Reports 114. Retrieved from http://www.reports-express.co.uk Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 1280–1300. Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational principle. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp. 1–46). New York: Academic Press. Higgins, E. T., Friedman, R. S., Harlow, R. E., Idson, L. C., Ayduk, O. N., & Taylor, A. (2001). Achievement orientations from subjective histories of success: Promotion pride versus prevention pride. European Journal of Social Psychology, 31, 3–23. Keller, T. E., & Pryce, J. M. (2010). Mutual but unequal: Mentoring as a hybrid of familiar relationship roles. New Directions for Youth Development, 2010 126, 33–50. Liabo, K., Lucas, P., & Roberts, H. (2005). International: The UK and Europe. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 392–407). Thousand Oaks, CA: Sage. Meier, R. (2008). Youth mentoring: A good thing? Center for Policy Studies. Retrieved from http://www.cps .org.uk/files/reports/original/11102717000120080911SocialPolicyYouthMentoring.pdf MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. Retrieved from http://www.mentoring.org/program_resources/ elements_and_toolkits Miller, A. (2002). Mentoring students and young people: A handbook of effective practice. London: Kogan Page. Onuoha, F. N., Munakata, T., Serumaga-Zake, P. A., Nyonyintono, R. M., & Bogere, S. (2009). Negative mental health factors in children orphaned by AIDS: Natural mentoring as a palliative care. AIDS Behavior, 13, 980–988. Philip, K. (2003). Youth mentoring: The American dream comes to UK? British Journal of Guidance & Counseling, 31, 101–112. Philip, K. (2008). She’s my second mum: Young people building relationships in uncertain circumstances. Child Care in Practice, 14, 19–33. Philip, K., & Hendry, L. B. (1996). Young people and mentoring—toward a typology. Journal of Adolescence, 19, 189–201. Philip, K., & Hendry, L. B. (2000). Making sense of mentoring or mentoring making sense: Reflections on the mentoring process by adult mentors with young people. Journal of Community & Applied Social Psychology, 10, 211–223.

Philip, K., Shucksmith, J., & King, C. (2004). Sharing a laugh? A qualitative study of mentoring interventions with young people. Retrieved from http://www .jrf.org.uk Rhodes, J. E. (2008). Improving youth mentoring interventions through research-based practice. American Journal of Community Psychology, 12, 3–10. Rhodes, J. E., & DuBois, D. L. (2008). Mentoring relationships and programs for youth. Current Directions in Psychological Science, 17, 254–258. Rhodes, J. E., & Lowe, S. R. (2009). Mentoring in adolescence. In R. M. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology (3rd ed., Vol. 2, pp. 152–190). Hoboken, NJ: Wiley. Rothbaum, F., Weisz, J., Pott, M., Miyake, K., & Morelli, G. (2000). Attachment and culture: Security in the United States and Japan. American Psychologist, 55, 1093–1104. Russell, L. (2007). Mentoring is not for you!: Mentee voices on managing their mentoring experience. Improving Schools, 10, 41–52. Sagiv, L., & Schwartz, S. (2000). Value priorities and subjective well-being: Direct relations and congruity effects. European Journal of Social Psychology, 30, 177–198. Schwartz, S. (1999). A theory of cultural values and some implications for work. Applied Psychology: An International Review, 48, 23–47. Shin, Y., & Rew, L. (2010). A mentoring program for the promotion of sexual health among Korean adolescents. Journal of Pediatric Health Care, 24, 292–299. Spencer, R. (2006). Understanding the mentoring process between adolescents and adults. Youth and Society, 37, 287–315. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failure. Journal of Adolescent Research, 22, 331–354. Spencer, R., & Rhodes, J. E. (2005). A counseling and psychotherapy perspective on mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 118–132). Thousand Oaks, CA: Sage. Thurman, T. R., Snider, L., Boris, N., Kalisa, E., Nyirazinyoye, L., & Brown, L. (2008). Barriers to the community support of orphans and vulnerable youth in Rwanda. Social Science & Medicine, 66, 1557–1567. Triandis, H. C., Bontempo, R., Villareal, J., Asai, M., & Lucca, N. (1988). Individualism and collectivism: Cross cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology, 54, 323–338. Turnbull, B., Hernandez, R., & Reyes, M. (2009). Street children and their helpers: An actor-oriented approach. Children and Youth Services Review, 31, 1283–1288. USAID. (2007). Psychological benefits of a mentoring program for youth-headed households in Rwanda. Retrieved from http://www.popcouncil.org/horizons/ projects/Rwanda_PsychOVC.htm Vazsonyi, A. T., & Snider, J. B. (2008). Mentoring, competencies, and adjustments: American part-time employment and European apprenticeships. International Journal of Behavioral Development, 32, 46–55.

PART V Programs and Contexts

14  SCHOOL-BASED MENTORING Carla Herrera and Michael J. Karcher

Introduction Schools are a primary context for the development of mentoring relationships outside of the family. Many naturally forming mentoring relationships develop in schools between students and adults such as teachers and support staff (DuBois & Silverthorn, 2005a). Schools are also a common site for implementing formal mentoring programs for youth. Indeed, more volunteers serve as mentors in school-based mentoring (SBM) programs (28%) than in religion-based (24%), workplace (16%), or sports-related programs (9%) (AOL, 2002). SBM programs were designed, in part, to overcome many of the challenges facing communitybased mentoring (CBM) programs. SBM programs typically involve mentoring once a week in a supervised school setting. This creates a lower burden on volunteers than CBM programs (e.g., less time involved, less pressure to determine activities), making recruitment easier and attracting volunteers who are less able to manage the commitment required by CBM programs. In addition, by relying on teacher rather than parent referrals, programs have opportunities to proactively seek to engage youth who are not typically targeted by more traditional CBM programs (e.g., youth whose parents might not take the initiative to involve them in CBM programs, youth with particular difficulties in academic areas). The school context also can afford easier access to training, support, and supervision, all of which could decrease costs. These potential benefits have made SBM appealing both to schools and to mentoring agencies nationwide. Yet there are just as many factors that make implementing mentoring programs in schools more challenging, costly, and time intensive for program

staff than was initially posited (Herrera, 1999). First, although mentors may be attracted to the typically less intensive time commitment, the school’s 9-month academic calendar and segmented blocks of relatively brief time (e.g., 1 hour) for meetings may “water down” much of the relational intensity that is the cornerstone of effective matches in the community (Herrera, Sipe, McClanahan, Arbreton, & Pepper, 2000). Also, structural barriers including the summer break and youth mobility may decrease the likelihood that such long-term relationships will be achieved in SBM programs. Second, youth with more serious needs may not be well served by SBM. Although youth at greater academic, social, and behavioral risk are more easily accessible in schools, many SBM mentors may be unprepared for the challenges of working with these youth. Moreover, the inclusion of such youth may stigmatize the program, making it appear to be an effort to “address problems.” For these and other reasons, to the extent that SBM programs may have a relative advantage in terms of being able to “reach out to” youth in greater need, this may reflect a Pyrrhic victory of sorts in which much effort is expended for too little return. Finally, providing support to those mentors working in schools may be no less challenging than it is for those supporting matches in other types of mentoring programs. Volunteers taking off an hour each week from work to mentor, for example, may feel they do not have an extra hour for training and supervision. Additionally, mentors working in schools are likely to be much more constrained in the range of activities they can engage in and must successfully negotiate a complex system (the school) in their efforts to develop strong mentoring relationships. Thus, in some ways, staff support may be 203

204  Programs and Contexts needed even more in SBM than in other programs such as those that are community based. Considerable staff time is also likely to be required to foster relationships with the school and to ensure that school staff support mentors in ways that foster their success in developing strong mentoring relationships with youth as well as their sustained involvement as volunteers over time. In this chapter, we delve into recent research on mentoring in the school context. In doing so, our goal is for readers to take away a greater appreciation of both the opportunities and the challenges associated with fostering natural and program-supported mentoring relationships in schools. We hope this will help program staff avoid the challenges experienced by those who initially viewed SBM as either a mere extension of CBM or, even worse, as “CBM light.” SBM is a unique, complex program. Without clear efforts to make the most of its strengths and overcome its potential challenges, SBM can yield problems for schools, frustration for mentors, and disappointment for youth. Better understood, however, we believe SBM can help the mentoring field reach more youth and foster beneficial relationships for these youth. This chapter extends the coverage of the corresponding chapter from the first edition of the Handbook of Youth Mentoring (Portwood & Ayers, 2005) by reviewing findings from important new research on mentoring within the school context, by considering mentoring in schools from the perspective of new theoretical perspectives, and by offering updated and more detailed recommendations and guiding questions for the design and implementation of mentoring programs in schools. We begin with a consideration of theoretical perspectives, followed by a review of research and then a discussion of implications for practice.

Theory In this section, we describe the characteristics of school-based mentoring that may help the approach to achieve its intended goals. We then consider two theoretical perspectives to better understand the social factors that may foster or impede the cultivation and maintenance of effective mentoring relationships in school settings. The first is social expectancy theory (Rosenthal, 1993), and the second is social-cognitive developmental theory (Selman, 1980, 1990). Together, these theoretical perspectives highlight distinctive facets of the school setting that have the potential to shape the course and outcomes of mentoring relationships between adults and youth within this context.

The School as a Unique Social Context School-based mentoring relationships develop in a specific social context—the school. In theory, interactions in this context could help the mentoring relationship operate through pathways that may be especially helpful in combating some of the social and academic difficulties experienced by many school-aged youth. For example, interactions with an adult mentor can provide youth with valuable opportunities to practice and hone social skills that can be transferred to other social interactions (Rhodes, 2002). When matches meet during the school day, mentors are also given a unique glimpse into youth’s strengths and difficulties in peer interactions—a perspective that few adults outside of the school context have. This perspective can help mentors provide youth with valuable feedback on their peer interactions and how they can improve them, as well as engaging youth’s peers in positive interactions more directly. Positive ties with naturally established schoolbased mentors (i.e., those outside of the auspices of a formal program) may help youth see other schoolrelated adults (i.e., teachers) in a similarly positive light. Mentors who come from outside of the school environment as part of a formal program also may help improve youth’s relationships with teachers. This may be particularly true in programs involving regular interaction between mentors and teachers— mentors may help focus teachers’ attention on youth, share insights with them about their mentees, and give youth a new perspective on their teachers. Improved teacher relationships should then foster other improvements. For example, a teacher’s level of support can influence children’s motivation to achieve (Patrick, Anderman, & Ryan, 2002), their expectations for their own achievement (Goodenow, 1993), and their engagement in school (Klem & Connell, 2004). By engaging in academically focused activities with their mentees (see Herrera et al., 2000), school-based mentors can help improve academic performance directly (e.g., through homework completion) and also may focus discussions on school— including social and academic challenges in this context. Because many youth referred to schoolbased programs are struggling academically, this focus may be an important route through which formal SBM programs work. Specifically, academically at-risk students often see school and teachers in a negative light (Pianta, Stuhlman, & Hamre, 2002), have difficulty seeking help at school when needed (see Richman, Rosenfeld, & Bowen, 1998), and lack skills like time management that are

School-Based Mentoring  205 important for school success (Larose & Roy, 1995; see Larose and Tarabulsy, 2005). School-based mentors may offer suggestions for how to succeed at these critical tasks. By engaging youth in positive interactions at school, mentors may also give youth a more positive outlook on this context and an incentive to go to school. Studies suggest that participation in school-based activities increases youth’s school liking (e.g., Eccles & Barber, 1999)—which may in turn lead to improved attendance and academic performance. Although this general framework suggests some of the routes through which SBM might benefit youth, it ignores key developmental processes that may shape the effects of SBM in both positive and negative ways. The presence of peers, for example, is a powerful social force within the school context that can make—or break—the developing relationship. Next we discuss two theoretical perspectives that demonstrate the importance of peers and reveal possibilities for why peer influences appear to vary across grade levels. Social Expectancy Theory Somewhat unique to mentoring within the school context, peers are almost always present and their presence can shape the developing mentoring relationship, not just in the focus of match interactions but also in how the child perceives the mentoring relationship itself. We use as a starting point the assumption that often neither the mentor nor the mentee has a very clear sense of what mentoring is all about. This absence of clarity may plague mentoring relationships in schools more so than those based in the broader community because adults and students have a fairly clear understanding of how adults and youth are “supposed” to interact in schools, what their purpose for interacting is, and what the ultimate outcome of their interactions will be. Unfortunately, the prototypical adult-youth relationship in schools connotes the roles of teacher, tutor, principal, or coach for the adult, and for youth, only that of pupil (Karcher & Nakkula, 2010). None of these roles supports the type of interactions programs often tell participants they are trying to foster (e.g., “having fun,” “developing a friendship”) and the interactions that research suggests are characteristic of an effective mentoring relationship. Casual conversation, discussion of personal issues, and playful and fun interactions all appear to be essential elements of effective mentoring relationships, both formal and naturally forming ones (Karcher, Herrera, & Hansen, 2010; Parra, DuBois, Neville, Pugh-Lilly, & Povinelli, 2002; Spencer,

Jordan, & Sazama, 2004), but may be viewed as peripheral elements of what adult-youth relationships should look like in schools. This lack of clarity about what an SBM relationship should look like, what interactions it might constitute, and what its ultimate purpose is may undermine the effectiveness of these relationships, as a result of interpersonal expectancies. Rosenthal (1993) suggests that interpersonal expectancy effects (i.e., the social consequences of one’s expectations about others) often become manifest in self-fulfilling prophecies. Research has supported the existence of this phenomenon in a wide range of contexts and relationships, most notably, teacher-student relationships (de Boer, Bosker, & van der Werf, 2010; Rosenthal, 1993). Simply stated, what teachers are led to believe about their students’ capacity will ultimately contribute to the realization of that expectation through the teachers’ interactions with their students. Youth whose teachers expect them to underperform typically do, whereas those whose teachers expect superior performance will often live up to that expectation. In the school setting, the mentee has the opportunity (or burden) to consider his or her peers’ assessment of the purpose of mentoring. Thus, the presence of peers may alter the youth’s perception of the purpose of a mentor and the meaning of a mentoring relationship. In this way, the school-based context for the relationship could affect how the youth interacts with the mentor and whether the relationship succeeds or fails. The stigma associated with most educational enhancement interventions (e.g., tutoring, summer school, professional-delivered mental health interventions) likely helps to frame mentees’ initial beliefs about the meaning and purpose of their SBM relationship. In other social contexts, such as natural mentoring relationships with neighbors or family friends or those relationships that are program sponsored with interactions based in the broader community, the chemistry and perceived “connection” between the mentor and youth theoretically is less influenced by the presence of peers, allowing mentormentee interactions to serve as the main way in which youth assess the value, function, and meaning of the relationship. In schools, before sufficient time has passed to test out the relationship, youth may feel they need to contend with their peers’ assumptions about what mentoring means. Many youth (particularly older youth, as we will discuss) may assume that their peers see mentoring as a program for kids who are failing (academically or interpersonally). Such assumptions may contaminate the initial formation and ultimate direction of the relationship.

206  Programs and Contexts Attention to the ways in which perceptions of stigma may shape the effectiveness of a program is not new or unique to mentoring. Research suggests that any time a program or an intervention shifts from being delivered universally to all youth toward targeting individual youth more selectively, some degree of stigma emerges. These effects appear to be larger for boys and for acting-out or externalizing youth (Rapee et al., 2006), which may help explain findings we report later that suggest smaller or even negative effects of SBM for some boys and for some “higher-risk” youth. Stigma may also have very different effects depending on age (Hinshaw, 2006), so it may be crucial to take a developmental perspective in considering ways that stigma may be most problematic or threatening to the successful formation and continuation of a mentoring relationship in the school setting (see Noam et al., this volume, Chapter 7). We turn now to consideration of a theoretical perspective that is particularly well suited to attending to such developmental differences. Social-Cognitive Developmental Theory Social-cognitive developmental theory also can inform our understanding of potential effects of both stigma and peers on mentoring relationships in the school setting. This theory suggests that social perceptions are shaped by cognitive development, such that the processes of social expectations described previously will vary between childhood and adolescence (Selman, 1980, 1990). For example, not until middle school can children take their peers’ perspectives into account and coordinate them with their own. As a consequence, middle school students may become extraordinarily self-conscious and concerned with their peers’ assessments. By high school, many youth have attenuated some of the dread associated with negative peer evaluations by becoming more discriminating in whom they associate with and whose opinions matter. One implication of social-cognitive theory is that stigma may play an increasing role in SBM relationships for youth as they age. Elementary school children appear to profit universally from a teacher (and, arguably, by extension, a programbased mentor) who provides them with empathy, praise, and attention (Fredriksen & Rhodes, 2004; Van Ryzin, 2010). Psychodynamic developmental theories (Kohut & Elson, 1987) suggest that such relationships with teachers and formal mentors may mobilize or propel developmental processes that strengthen a sense of self, bolster confidence, and lead children to idealize adults. As a result, children may be motivated to demonstrate skills and talents

to the adults they idealize to elicit even more developmental nutrients (e.g., praise, attention) from them (Rhodes, Spencer, Keller, Liang, & Noam, 2006). Because younger elementary-aged children have not yet developed the capacity to weigh others’ assessments into their own perceptions of their relationships with others (and, hence, themselves), the benefits of such relationships with adults are generally not mitigated by perceptions of peers’ assessments of these relationships. As youth develop and progress toward middle school, a self-consciousness emerges that bridges the elementary-aged child’s unmitigated appreciation for the attention of a teacher or other adult (e.g., a mentor) with the more differentiated perception of adults that adolescents begin to establish. During this later stage of development, young adolescents become more discriminating about which adults’ perceptions of themselves they will incorporate into their own self-image. Middle-school-aged youth also become more aware of (i.e., differentiate for the first time) the roles that different adults play in larger society (e.g., teacher, counselor, mentor). This affords a new understanding of what a relationship with those adults may say to others about the youth. Whereas mentors might be viewed as just friendly adults by elementary school children, middle schoolers often start to view mentors as “adults who work with needy or struggling youth.” A more self-determined and consciously chosen identity emerges for adolescents that has the potential to make them more discriminating in their selection of who they will allow to mentor them and whose opinion will contribute to their developing identity (Erikson, 1968). The three theoretical perspectives we have shared suggest that SBM has the potential to positively and negatively affect child and adolescent development. Which effect it has on youth may be controlled, to some extent, by whether program staff apply these theories when crafting programs, selecting mentees, and training mentors. Using theory to train mentors in how to avert or deflect stigma from being associated with the match is important. But the first step should be to design programs (e.g., program goals, who gets served, what matches do together, where meetings occur) that leverage our theoretical understanding to ensure we are sending mentors into the right contexts to appropriately serve particular types of youth. For example, creating an SBM program solely for high-school-aged youth who are at high behavioral or academic risk could be a recipe for disaster without significant efforts to address the challenges inherent in serving this group at school. The potential problems of this specific

School-Based Mentoring  207 youth-program interaction not only are suggested by theory, but also have been revealed in research on youth mentoring in schools, which we review in the next section. Here we find lessons drawn from research about the advantages and disadvantages of using SBM to serve youth who differ in age, racial or socioeconomic background, or social or academic competency.

Research In this section, we review existing research on mentoring in the school context (SBM), considering first research on natural or informal mentors and then turning to research on the effectiveness of formal SBM programs. In our review, we link our discussion of findings to the theoretical perspectives discussed previously. We conclude by offering a synthesis of current findings and promising future directions for research. Natural Mentors in Schools For many youth, their most meaningful and longest lasting mentoring relationships are with members of their extended family (Zimmerman, Bingenheimer, & Behrendt, 2005). However, the school context appears to run a close second as a place in which mentoring ties with adults are cultivated (Darling, Hamilton, Toyokawa, & Matsuda, 2002; Greenberger, Chen, & Beam, 1998). DuBois and Silverthorn (2005b), for example, in an analysis of a nationally representative sample of youth from the Add Health study, found that nearly a third of the mentors that young adults reported having were associated with their schools (i.e., teachers, counselors, or coaches). And these naturally occurring school-based relationships may serve important protective functions. For example, Black, Grenard, Sussman, and Rohrbach (2010) found that having a school-based mentor was positively associated with school attachment for high school students, which was then associated with lower reports of engagement in a number of risk behaviors including substance use and violence. Gastic and Johnson (2009) found that “informal teacher-mentors” were particularly important for sexual minority youth (especially females) in terms of predicting subsequent college attendance. Yet, it appears that middle-class youth more easily form a connection with middle-class teachers (Delpit, 1995; Erickson, McDonald, & Elder, 2009), and that associations between having one’s teacher as a natural mentor and subsequent educational

attainment may be confounded with social class— that is, higher social class may lead both to having or reporting a teacher as a mentor and to higher educational attainment. Indeed, some research suggests that ethnic minority youth, who are more likely to be from poor and working-class backgrounds, are more likely than White youth to view extended family members as their primary natural mentors (Sánchez & Reyes, 1999) and report less connectedness to teachers (Karcher & Sass, 2010). The potentially important and often hidden role of social class in mentoring is addressed more fully by Deutsch et al. (this volume, Chapter 12). Although several of the studies described above found that having natural mentors in schools is associated with an increased likelihood of graduation and postsecondary education, few studies have explicitly considered the role of socioeconomic status as a moderator. One exception is a study by Erickson et al. (2009) using the Add Health dataset, which includes several thousand youth and provides a longitudinal view of the role of natural mentors in the lives of youth from families with more versus fewer resources. In their study, claiming an adult at school (teacher, school counselor, or coach) filled a mentoring role was associated with both higher 12th-grade GPA and achievement in higher education than claiming to have had a friend, relative, or community mentor (e.g., religious leader, employer, neighbor, friend’s parent, or helping professional). They found positive associations between having a natural mentor in one’s family and educational achievement for all youth, but the positive association between having a teacher or other adult at school as a mentor and academic outcomes was greatest for poor youth. What these studies make clear is that adults at school can be among youth’s primary adult support figures and that these relationships appear to benefit youth. These benefits may, in part, be attributable to the fact that school-based relationships can endure over time (assuming youth continue to attend a given school). Most also allow frequent contact, which may enable these mentors to provide a protective function through regular monitoring of youth’s behavior, such as has been reported in the Achievement Mentoring program that enlists school staff as mentors to youth (Johnson, Holt, Bry, & Powell, 2008). Nevertheless, not all youth avail themselves of or are as successful at entering into relationships with adult mentoring figures at school. Boys, in particular, appear to be at a disadvantage when it comes to cultivating ties with natural mentors in schools. Whereas youth most often report natural mentors in their lives

208  Programs and Contexts who are the same sex as they are, most adults who are available to fill this role in the school setting (i.e., teachers), especially during elementary school, are likely to be female (Zimmerman et al., 2005). Such considerations highlight how issues relating to natural and programmatic mentoring in the school setting may be interrelated and inform practice considerations. In particular, there may be a greater need for programmatic mentors for boys in elementary and middle school than for boys in high school, given the paucity of male teachers at these lower grade levels. This is good, given findings we present later in which the high-school-aged boys seem to be more harmed than helped by nondirective program-based mentoring (Karcher, 2008). Recent research highlights another example of how natural and program-based mentoring may complement each other in the school setting. Schwartz, Rhodes, Chan, and Herrera (2011) concluded, based on analyses of data from the Public/ Private Ventures impact study of the Big Brothers Big Sisters (BBBS) SBM program (discussed in more detail later in this section), that it was those youth who had neither strong nor particularly weak social ties with peers and adults who benefited the most from psychosocial (i.e., “Big Brothers Big Sisters style”) SBM. The authors argued that youth may need some basic relationship skills to form effective mentoring relationships; however, if a child already has very strong relationships within his or her social network, or is incapable of forging strong relationships, mentoring may not be as effective. At the same time, natural mentoring research suggests that fewer of such socially middling youth typically recruit natural mentors than do those youth who are most and least lacking the forms of support mentors provide (Zimmerman et al., 2005). Thus, program-based mentors may be very useful in helping to fill this gap for those youth “in the middle.” Overall Effectiveness of Formal Mentoring Programs in Schools Despite the relative newness of SBM as a program model, its effectiveness has been evaluated rigorously in three recent large-scale randomized controlled trial (RCT) evaluations (something that took almost 100 years for CBM programs). In combination with the findings of a handful of earlier, notably smaller RCTs of SBM (e.g., Cavell & Hughes, 2000; LoSciuto, Rajala, Townsend, & Taylor, 1996), those reported in the recent RCTs enable us to answer questions about impacts more definitively than was possible in the prior edition

of the Handbook. Yet, as we highlight below, a number of questions remain regarding this model of mentoring. Each of the large-scale RCTs focused on slightly different questions and yielded slightly different findings. One study examined the “absolute” impact of providing federal funding to establish or support SBM programs targeting at-risk youth (Bernstein, Rappaport, Olsho, Hunt, & Levin, 2009). Another focused specifically on the effectiveness of the BBBS SBM program model (i.e., the “potential” impact of SBM when implemented well; Herrera et al., 2007). Finally, the third study focused on one program within one city and tested the additive, or relative, effects of providing SBM as a supplement to other services provided to youth at risk for academic underachievement (Karcher, 2008). In short, one assessed absolute impact, another potential impact, and a third the additive effects of SBM. The Abt Study of the Impact of the U.S. Department of Education’s Student Mentoring Program Funding Initiative. A national study conducted by Abt Associates followed 2,573 students for one school year in 32 SBM programs funded under the U.S. Department of Education’s Student Mentoring Program (Bernstein et al., 2009). The study was designed to test the extent to which the funded SBM programs were effective. Thus, efforts were made to involve programs that were representative of the 255 mentoring programs funded in 2004 or 2005. Programs ranged widely in their experience with SBM and in the type of agency overseeing services. Some were established, whereas others were just starting their programs. The results of this study provided the most rigorous assessment to date of the value of SBM, at least as it is presently implemented in its myriad manifestations across the United States. Once the researchers applied fairly conservative standards for assessing statistical significance (i.e., the Benjamini-Hochberg correction for false discovery), they were unable to detect any statistically significant impacts of SBM on average (across all programs collectively). But these null findings tell only part of the story. Despite the laudable rigor of the Abt study, important issues undermine the unequivocal acceptance of their conclusion of no effects. First, in addition to the findings presented in the main text of the report, the authors also presented (in an appendix) separate impact estimates for each participating program for all of the outcomes. Considering the outcomes that Portwood and Ayers (2005) suggested SBM is most likely to affect (i.e., school

School-Based Mentoring  209 connectedness, here “school efficacy and bonding”), effect sizes ranged from d = .38 to d = –.35, suggesting that some programs yielded moderately large positive effects whereas others yielded similarly large negative effects. Thus, to simply state that mentoring had “no effect” masks both the risks and rewards of SBM (to borrow the subtitle from Rhodes’s [2002] book). Some programs showed evidence of effectiveness, while others did not, which begs the question of what might account for such differences. Second, the authors also presented impact analyses using the original outcome measures (as opposed to the new combined scales they created to improve reliability) and applying the conventional statistical significance level of .05 (see Bernstein et al., 2009, Appendix D). These analyses revealed statistically significant favorable impacts of SBM on measures of perceived scholastic efficacy, supportiveness of relationships with other adults, future orientation, absences, and truancy. All of these findings reflect small effects (none larger than a standardized mean difference of .15). And assessing statistical significance without correcting for chance findings due to conducting multiple tests of significance (i.e., false discovery) inflates the likelihood of reporting an effect by chance. These “raw” findings, using the original scales and without correction for false discovery, contradicted the “no effects” conclusion of the main report. Yet, the raw findings (and the size of these effects) were in line with findings from another RCT—that of the BBBS SBM program. In fact, we believe the “no effects” conclusion of the Abt study surprised policymakers, program staff, and the public alike (see Viadero, 2009), in part, because Herrera et al. (2007) and Karcher (2008) had reported fairly positive results in their RCTs examining SBM. Of course, these other studies were of more homogeneous and well-established mentoring programs and used less conservative tests of significance. Public/Private Venture’s (P/PV’s) Study of the BBBS SBM Program. In 2007, P/PV released a report of findings from its RCT evaluating the effectiveness of the BBBS SBM program. The study was conducted in collaboration with Big Brothers Big Sisters of America and focused on 10 BBBS programs nationwide. A total of 1,139 youth were followed for one and a half school years. Programs were diverse in geography and representative of both small and larger BBBS agencies, but they also were required to have strong leadership, wellestablished relationships with partner schools, and at least 4 years of experience implementing SBM.

Any of these factors might have increased their likelihood of yielding impacts. The results of this study suggested that BBBS SBM yielded statistically significant impacts on school-related outcomes (average standardized mean difference effect size of .08) consistent in size with those commonly reported in the field for youth mentoring generally and for SBM programs more specifically (DuBois, Holloway, Valentine, & Cooper, 2002). However, unlike P/PV’s 1995 impact study of BBBS CBM (Tierney, Grossman, & Resch, 1995), the 2007 study of SBM did not yield statistically significant impacts on non-schoolrelated outcomes (e.g., parent-child relationship quality, misbehavior outside of school). Rather, apart from program participants being more likely than controls to report having a special adult in their lives, all of the statistically significant impacts in this study were on school-related attitudes, performance, and behavior (e.g., attendance). Also, most of these impacts were no longer evident at the second follow-up (when only half of the mentees remained matched), thus potentially indicating a lack of durability for the effects of one school year of SBM. The study’s method of assessing statistical significance (using a significance level of p < .10, without limiting reporting of impacts to those that were evident with correction for false discovery) was also less conservative than that used in the Abt study. Although this appears to be a somewhat liberal approach to assessing statistical significance, it is the same level of significance used in the original P/PV study of BBBS CBM from 1995. This is important to note because most programs use the 1995 P/PV study as the “gold standard” evidence that CBM works. Thus, if critics deem the approach taken in the 2007 SBM study as too liberal, then they must also acknowledge that they are challenging the veracity and rigor of the earlier BBBS CBM study. It is also important to note that several key outcomes were significant even after correcting for false discovery (noted in the report’s appendices). The SMILE Study of the Communities in Schools of San Antonio Mentoring Program. Initiated in 2003, one year prior to Herrera et al.’s study, the Study of Mentoring in the Learning Environment (SMILE; Karcher, 2008) addressed the question of whether SBM has any additive effects on youth outcomes beyond the effects of school-based case management services provided by a local affiliate of the national Communities in Schools (CIS) organization in San Antonio, Texas. The study followed 516

210  Programs and Contexts predominately Latino youth in 5th through 12th grades over one school year (involving proportionally more high-school-aged youth than did the Abt and P/PV studies) and involved a large number of college-aged and Latino mentors. As noted, the SMILE study addressed a different question than the other two studies. Namely, it asked whether there is an additional benefit of providing mentors to youth who are already receiving other support services provided by CIS, such as group counseling, tutoring, and enrichment activities. Therefore, the findings we report for this study are about relative efficacy—that is, the benefits of participating in an SBM program for youth already receiving support services relative to youth who receive only the nonmentoring support services. It is not a test of absolute efficacy (when treatment is compared to no treatment), but rather an estimate of the additive effects of mentoring beyond the effects of an existing program. Overall, the SMILE study found statistically significant effects (at p < .05, without adjustments for false discovery) favoring youth who were assigned to also participate in the SBM program on just 4 of 21 outcome measures: youth reports of self-esteem (connectedness to self in the present and global self-esteem) and peer relationships (connectedness to peers and social support from friends). However, as we explore in greater detail below, impacts varied significantly depending on the gender and age of youth. Providing a mentor in addition to other support services alone yielded considerably larger impacts for young boys and teenage girls. But there were negative outcomes for high school boys. That is, younger boys benefited from “SBM plus” whereas older boys appear to have been harmed. While the direction of the effects for mentored high school boys were generally negative, only connectedness to teachers was significantly lower for boys who received a mentor in high school. The SMILE study is the only one of the three SBM studies that included a large enough sample of high-school-aged participants to examine the effects of SBM on older students, and, as noted, the results are not promising, at least for boys. These findings are in no way definitive, but they do raise some red flags regarding the use of nondirective SBM with high-school-aged boys and perhaps in particular those from ethnic minority groups who may be especially vulnerable to stereotype threat related to the stigma of school-based interventions when seen as mental health treatment. See Noam et al. (this volume, Chapter 7) for an extended analysis of how developmental competencies may have influenced these outcomes.

Integrative and Meta-analytic Views of Findings From the Recent RCTs of SBM. Despite the differences across these three recent RCTs, a general review of their findings reveals impacts that align with earlier findings from smaller-scale and, in some instances, quasi-experimental evaluations of SBM programs. This includes support for the conclusion of Portwood and Ayers (2005) in the earlier edition of the Handbook that the effects of SBM appear to be concentrated on outcomes related to the school domain. Despite evidence that SBM influences schoolrelated outcomes, only one of the three studies found impacts on grades specifically, and these were teacher reported rather than based on report cards. The size of the effect of BBBS SBM on academic outcomes was also virtually equivalent to estimates 10 years prior of the effects of the BBBS CBM program on academic outcomes, suggesting that if SBM affects grades it likely does not have a bigger impact on them than other mentoring approaches. Additionally, hypotheses presented earlier suggesting that SBM might be particularly powerful in improving student-teacher relationships have not been borne out. None of the three studies found direct improvements in student-teacher relationships, and in fact the SMILE study found SBM could negatively affect teacher connectedness among older boys. Wheeler, Keller, and DuBois (2010) used metaanalytic techniques to statistically synthesize findings that were reported across the three studies. They concluded that SBM can be modestly effective for improving selected school-related, but not academic, outcomes (i.e., support from non-familial adults, peer support, perceptions of scholastic efficacy, school-related misconduct, absenteeism, and truancy). Supporting findings from an earlier metaanalysis of 55 mentoring programs (DuBois et al., 2002), the actual size of these effects (i.e., the difference between the progress made by youth in the treatment group versus that in the control group) was fairly small (ranging from .07 to .18)—comparable to the size of effects yielded from CBM, though on a narrower set of outcomes. Other school-based programs (e.g., tutoring) have been found to yield similar, though in some instances, somewhat larger, impacts on academic outcomes. Tutoring yields an effect size of .26 on reading (that is, a quarter of a standard deviation difference between tutored and nontutored youth on reading achievement; Ritter, Barnett, Denny, & Albin, 2009). The effect sizes of after-school programs range from .10 to .30, depending on the outcome (Durlak, Weissberg, & Pachan, 2010). Effect

School-Based Mentoring  211 sizes for school-based drug-use prevention programs range from .17 to .19 (Tobler et al., 2000). By comparison, class size reduction effects range from .11 to .20 for math and reading achievement, respectively (Nye, Hedges, & Konstantopoulos, 1999). Moderators of SBM Program Impacts Recent research also has begun to point toward some of the circumstances under which SBM may be most (and least) effective. In this section, we consider results pertaining to how the effects of participation in an SBM program may be influenced by program “moderators” (i.e., factors that moderate a program’s effectiveness) such as youth and mentor characteristics. We also discuss how other factors, such as characteristics of the mentoring relationship and program practices, are associated with outcomes. Reflecting the developmental and interpersonal expectancy theories described above, we find that youth characteristics (e.g., grade level, risk, gender) are associated with differences in impact. These theories also provide insights into when, for whom, and under what conditions program practices, such as training, are essential. Youth Characteristics. As we have noted, both theory and recent studies suggest that age may play an important role in SBM’s effectiveness because of the presence of peers. The SMILE study findings of positive effects for young boys and negative effects for older boys is a good example of how development may moderate the effects of SBM. The effects of the program differ as a function of the characteristics of the youth, in this case, the youth’s age. These findings are consistent with the developmental and social expectancy theories presented earlier and point to the important role of peers in the presence of the match. For younger youth, their peers’ presence may propel the mentor’s impact; whereas for older youth, it may have the opposite effect. Younger, elementary-aged youth may feel proud to have a mentor and gain status with their peers through the focused attention of an adult (Herrera, 1999). Conversely, high school mentees who are better able to see the presence of a mentor from multiple perspectives might be embarrassed by this attention as it could indicate to others that the student has some kind of problem. This may help explain why high-school-aged boys in the SMILE study showed decrements in connectedness to teachers after meeting with a mentor. Although most other work studying both inand out-of-school mentors has not found significant differences between the benefits that girls and boys

receive from mentoring, Herrera et al. (2007) found that girls received selected benefits from SBM that boys did not (i.e., improvements in academic performance and decreases in school misbehavior). This could, in part, reflect the fact that girls and boys may be referred to mentoring programs for different reasons. Rhodes, Lowe, Litchfield, and Walsh-Samp (2008), for example, found that in CBM programs, referred girls had more problems at baseline in relationships with their parents. Because BBBS CBM targets youth from single-parent (often female-led) households, boys may be referred simply because they could benefit from a missing role model, whereas girls may be referred because they are already having difficulties with their (maternal) caregiver (see Rhodes et al., 2008). Findings from the SMILE study further suggest that gender might interact with age in determining what youth get out of the program. Specifically, older boys may be the least likely to benefit from SBM and younger boys the most likely for two reasons. First, younger boys may have fewer options for natural male mentors in the school setting (as noted earlier). Second, young boys do not have the capacity to, or burden of, trying to make sense of social expectations by peers. Several studies suggest that although SBM may help programs recruit the most “at-risk” youth, these youth may not benefit the most from mentoring. Namely, studies hint that those youth who are most at-risk academically or socially may receive fewer benefits from SBM participation (Herrera et al., 2007; Schwartz et al., 2011). These findings are at odds with the common program mission— often based on pressure from schools to prioritize these youth—of serving youth “who need the program the most.” The evidence to date suggests that it may not be those youth who can get the most out of the program. Youth with very serious needs may benefit from more intensive services than a 1-houra-week program can provide. Mentor Characteristics. SBM has attracted new groups of volunteers who, for various reasons, could not (or would not) have participated in CBM programs. In particular, older volunteers, high school and college students, and mentors recruited through businesses participate in SBM more often than in CBM (Herrera et al., 2000). This influx of new volunteers has made even more important the question of which groups of mentors might be most successful with youth in the context of SBM programs. We do not discuss high school student mentors here (but see Karcher, this volume, Chapter 16) except to make one theoretically relevant point. In the P/PV study of the BBBS SBM program, the only favorable program

212  Programs and Contexts impact that appeared to be unique to teen-mentored youth was an improvement in feeling accepted by their peers. As suggested by the theoretical perspectives discussed earlier, this may be due, in part, to the added social value of having an older student to spend time with in the presence of one’s peers (Herrera, Kauh, Cooney, Grossman, & McMaken, 2008), or perhaps teen mentors are less likely to evoke the stigma associated with adult “helpers.” Yet, in these BBBS programs, high school students appear overall to have been much less effective than adult volunteers, with positive program impacts concentrated nearly exclusively on youth who had the latter as volunteers (Herrera et al., 2008). These findings could be due in part to the fact that the high school student mentors were less consistent in their program attendance and appeared to have more difficulty focusing on their own mentees when their peers were present (Herrera et al., 2008). Consistent with Rosenthal’s (1993) interpersonal expectancy theory, described earlier, some evidence also indicates that mentors’ expectations may play a key role in shaping the quality of their mentoring relationships with youth in SBM programs. In analyses of data from the P/PV study, Karcher, Davidson, Rhodes, and Herrera (2010) found that high school student mentors who viewed children as fun and interesting prior to being matched, developed more emotionally engaging relationships with their mentees (from the youth’s perspective), than mentors who did not. These expectations also interacted with youth characteristics such that mentors with positive attitudes were most effective with those youth who were the least academically connected and successful—these youth were more engaged in the relationship than disconnected mentees with more negative mentors, and they reported stronger relationships with their teachers at followup than did disconnected control youth—whereas less positive mentors actually had negative effects on the least academically at-risk youth. Match Characteristics. In a longitudinal mixedmethods investigation of 27 matches in BBBS SBM programs, Keller and Pryce (2010, 2012) identified a variety of match styles and reported evidence of their being associated with differential outcomes for youth. Notably, matches with mentors who took on a hybrid role that included components of both “horizontal” friendship and the “vertical” provision of adult wisdom and guidance were the most successful. Relative to youth with mentors characterized by other styles (i.e., tutor, friend, or acquaintance), youth rated their relationships more positively and experienced greater

improvements in aggression and depressive symptoms. What also becomes apparent from their research is that it is difficult to identify a match’s potential by assessing match characteristics only at one point in time. Some matches struggle initially but later make significant breakthroughs; others start off well but then lose their steam. Only over time did the different roles of mentors they observed become apparent. It is the interaction between what matches do together and how this varies over time that determines the nature of a match (Karcher & Nakkula, 2010). When focusing on outcomes assessed at the final time point of the P/PV study in the second school year, Herrera et al. (2007) found that the SBM relationships associated with the most favorable youth outcomes were those that were higher quality (as indicated by youth reports of “closeness” to their mentor) and lasted past the first school year. Youth with matches that were higher quality but that ended early had better outcomes than nonmentored youth (i.e., the control group) albeit on only a few scattered measures. Those whose matches were lower quality but lasted into the second school year experienced no clear pattern of differential outcomes relative to nonmentored youth. In contrast, those with lower-quality relationships that ended after the first school year had significantly poorer outcomes relative to controls in several areas. This pattern of results suggests that ending a relationship on a bad note may be more disruptive or harmful than ending a strong relationship or working on one that is struggling. However, the issue of match length needs to be addressed more systematically because those youth who experience (and likely help to cultivate) long-term and high-quality relationships may be inherently different from those who do not. Moreover, other work suggests that prioritizing multiple-year matches may not always be the best strategy. There is some evidence that rematching (e.g., as a strategy to create longer matches for youth) is associated with setbacks, at least when relationship endings are unplanned (Grossman, Chan, Schwartz, & Rhodes, 2012). And studies of the Lunch Buddy program (e.g., Cavell & Hughes, 2000; see Cavell & Henrie, 2010) suggest that shorter matches that are well structured and that end at a predetermined time (midpoint of the school year) can be effective. Future research will need to test rigorously whether multiple matches with planned endings are as effective as having a single mentor for the same amount of time. Understanding more about the importance of match length vis-à-vis the school characteristics

School-Based Mentoring  213 that are associated with it is also crucial. Most SBM models constrain interactions between mentors and youth not only to the school building but also to the school calendar. Under these circumstances, youth and mentors communicate infrequently, if at all, over the summer months. However, in the P/PV study of the BBBS SBM program, some agencies did encourage matches to keep in touch. Those matches that communicated at least twice a month (21%) were more likely to continue into the fall. The quality of these relationships also improved more from the spring to the following fall than those of matches that had not communicated in the summer (Herrera et al., 2007). Program Practices. Mentor training and support appear to be two of the most important practices in SBM. Both variables are linked positively with relationship quality and match continuation (Herrera et al., 2007). Staff communication with mentors (an indicator of support) and program staff’s communication with school office staff (Karcher & Spencer, 2012) also have been linked with more benefits for youth mentored in schools—particularly in programs using young volunteers (Herrera et al., 2008). In the P/PV study, schools in which case managers reported better relationships with the school liaison, those in which the case manager spent more time at the school, and schools that were more invested in the program (e.g., had program-dedicated space and staff) yielded slightly longer matches than schools without these characteristics (McMaken, Herrera, & Kauh, 2008). Mentors’ ratings of their access to school resources and space are also linked positively with their reports of relationship quality and continuation of the match into a second school year (Herrera et al., 2007). Note that the direction of effects in several of these findings is unclear. For example, those mentors taking advantage of opportunities for training and support may be those who are already most committed to the relationship (i.e., strong relationships may encourage more use and positive perceptions of training and support rather than vice versa), highlighting the need to test the role of these practices more rigorously. Summary and Recommendations for Future Research Our understanding of mentoring in the school context has benefited from several recent high-quality studies. The findings of these studies underscore the apparent benefits that can stem from natural mentoring relationships that emerge between youth and

various adults in schools as a natural outgrowth of the education process. Likewise, studies of formal mentoring programs indicate their potential to strengthen, albeit modestly, a range of noteworthy school-related outcomes. Results, furthermore, point toward a number of factors relating to characteristics of youth and mentors, their relationships, and program practices as important in shaping levels and areas of impact of SBM programs on youth. Yet, we still have a long way to go in understanding how both natural and program-driven mentoring works in the school setting, under what circumstances it works best, and (particularly in school-based programs) whether and for whom it may lead to stigmatization. Although the recent RCTs provide rigorous tests of the size of SBM program impacts, the evidence these studies have yielded regarding the importance of specific program practices, mentor types, and youth characteristics is much more preliminary. Future studies that dig deeper into these issues will be of vital importance in improving SBM programs. The research to date suggests that not everyone can be an effective school-based mentor and not all children benefit equally from the program. Careful consideration should be given to whether youth with different needs and at different ages would benefit more from a program-sponsored mentor or from efforts to help foster and cultivate natural mentoring relationships between teachers and students. Understanding more about which youth benefit most, which mentors yield the most benefits for specific youth, and how to make effective matches would help programs know where to target their resources. Until researchers can outline what makes a strong SBM program and what types of participants work best at different grade levels, agencies will be left to fend for themselves in designing and implementing a huge range of practices that have yet to gel into one proven model.

Practice Recent studies provide some guidance as to how school-based mentoring programs and initiatives to support natural mentoring ties in schools might shape themselves in their efforts to help youth the most. Drawing on this evidence, in Table 14.1 we outline our recommendations for practice and guiding questions that should be prioritized as the field moves forward. In the remainder of this section of the chapter, we elaborate on these recommendations.

214  Programs and Contexts Table 14.1   Recommendations and Guiding Questions for Practice Recommendations Carefully develop key school partnerships. Develop a formal memorandum of understanding (MOU) with each host school that outlines mutual roles and responsibilities. In the MOU, a. ensure that at least one staff member at each host school is assigned to support and coordinate the program and is given authority and resources for carrying out this role; b. identify appropriate places in the school for mentoring to take place; and c. agree how to provide teachers and staff with appropriate orientation regarding the practices and benefits of school-based mentoring (SBM) and to enlist their support in recruitment, monitoring, and evaluation procedures. Structure programs to allow enough time for mentoring relationships to grow and develop, keeping in mind that at least one school year may be needed for strong, effective relationships to be formed. Mentors should be able to commit to this time frame, unless the program commits to providing a coordinated series of mentors to each youth over shorter periods of time (e.g., each semester; see Cavell & Henrie, 2010). Consider how the age or needs of the youth targeted may be best served by the program. Program-made matches may be particularly useful for boys in elementary and middle schools that have few male teachers (and fewer opportunities for same-sex natural mentoring relationships to form). Consider whether youth needs are best met by programmatically created mentoring relationships or naturally occurring school-based relationships. For example, sexual minority youth may benefit most from natural mentoring relationships, whereas youth with neither very strong nor very weak social ties may benefit greatly from programmatically created relationships. A school-based program could provide mentors for youth directly or help youth recruit natural mentors and foster their development. Consider ways to match mentors with mentees that capitalize on the shared meeting location of the school as well as their chemistry and shared interests (such as the “meet-n-greet” activity in which a group of youth meet a group of volunteers, and individuals from both groups note partners that they would like to get to know better). Guiding Questions Considering the evidence that many SBM programs may be best equipped to serve “medium” risk youth, what risk backgrounds are likely to be optimal for the youth served by your program? How might this focus change the way your program recruits and supports mentors, youth, and the schools in which you operate? And how is this information conveyed to key stakeholders (e.g., schools, mentors)? Does your program integrate any developmentally specific program elements that are likely to yield the strongest matches, such as having matches in elementary school meet in the presence of peers (e.g., in the cafeteria) or having older high-school-aged mentees provide input into the nature and purpose of their mentoring relationships? How can your program be designed and implemented to avoid evoking perceptions of stigma regarding mentoring as a program for misbehaving youth, or youth with mental health or academic needs? What changes could your program implement to lessen such stigma, such as involving more lower-risk youth or framing the program differently to the public and schools in which you work? What strategies can your program use to avoid potential negative effects of the academic calendar (e.g., summer match contact) and take advantage of opportunities that are presented by this calendar (e.g., end-of-year match closure activities and celebrations)? Given the challenges and costs involved in coordinating summer activities, is your program better off focusing on supporting matches that are limited to the school year, but are “renewable?” Do the likely benefits of operating a summer program outweigh the costs for your program?

School-Based Mentoring  215

How can your program capitalize on the school context (e.g., using shared location for innovative matching or introduction events, being present to implement closure activities, engaging teachers as partners in learning more about youth participants and ensuring they benefit from the program) to foster more effective match beginnings and endings? How can your program take advantage of its location in the school to involve and support parents in ways that foster desired program outcomes?

Selecting and Supporting Schools Research suggests that SBM programs simply run better in supportive schools. Many agencies assume that they must give every school exactly what it asks for to gain access to its students. Yet some schools may not be ready for a partnership with the program. School buy-in should be assessed prior to collaborating and should be secured at all levels. Programs should clearly state their goals. Schools should sign a memorandum of understanding outlining their role and provide a staff person with time allocated to the program and dedicated space if possible. Programs must weigh the benefits of being able to serve youth (the neediest of whom are often in underresourced schools) with being able to provide youth with top-notch services. The latter may be difficult in overwhelmed schools. When schools do come on board, programs need to foster these relationships through frequent communication with teachers and staff, feedback on how youth are benefiting, and continuous efforts to ensure that the needs of both the school and its students are being met. Once schools are selected, programs should work with them in a way that takes advantage of potential strengths. For example, program staff can enlist school staff to help with program recruitment, management, and supervision and to provide materials and activities to make the program run more smoothly. School staff can be enlisted to provide insights into individual children’s needs and to serve as partners in ensuring that these needs are met. School staff can also shape the mentor’s experience in ways that encourage—or discourage—further participation. Programs need to ensure that schools welcome and support their volunteers both by stressing this expectation up front and by checking in with volunteers to ensure that they are continuing to get the support and encouragement they need from the school. Training Staff Program staff also need training on how to develop and sustain positive relationships with schools and how to ensure that matches are well

supported. Staff need to understand how the school context may shape match interactions and how developmental changes in peers’ awareness of the match may foster success in elementary school matches but undermine high school matches. The potential merits of recruitment strategies that avoid stigmatizing the most at-risk youth and that consider both youth needs and mentor attitudes toward youth should also be highlighted in staff training. Selecting Youth and Mentors Both theory and research to date suggest that the youth who are most likely to be referred by teachers (i.e., those with serious academic, social, and behavioral challenges) may not be best served by schoolbased programs without significant additional efforts to recruit, train, and support mentors who can meet these specific needs. Instead the average schoolbased program may be most likely to benefit youth “in the middle”—for example, youth who have some challenges with their relationships, but not so many that they are incapable of forming a strong relationship with a mentor (Schwartz et al., 2011). Serving older adolescents in school-based programs can also raise significant challenges— programs need to present and structure themselves in ways that reflect youth’s growing need for autonomy. One way of doing this is to solicit specific goals that youth may have for their own program involvement and to build the program around these goals. Programs should also consider and adjust to potential mentee resistance to meeting with a mentor in the presence of their peers. Less progress has been made in determining who are the most effective school-based volunteers, except to note that volunteer age is a crucial factor, with high-school-aged volunteers, on average, being much less effective than adult volunteers when provided similar degrees and forms of support and training (Herrera et al., 2008). These younger volunteers appear to be more effective with additional supports than without such supports (see Karcher, this volume, Chapter 16). But all programs should approach using younger, high-school-aged

216  Programs and Contexts volunteers with care. For example, programs should give careful consideration to research suggesting that teen mentors’ attitudes matter—those with positive attitudes toward youth tend to be most effective particularly with academically at-risk youth, while those with less positive attitudes may negatively influence youth with few risk factors. In short, programs that cannot implement the necessary structure to support peer mentoring programs should probably rely instead on adults alone. Training Mentors Good volunteer training should be a part of all mentoring programs (see Kupersmidt & Rhodes, this volume, Chapter 30). In SBM, training may require some components beyond what is standard in CBM. For example, training should help mentors understand the school context, school rules, and how to manage (and benefit from) the input of peers; how the youth’s age could affect his or her expectations (as well as those of peers) about the mentor’s role; and what kind of support they should and shouldn’t expect from school staff. Even more important, schools may put subtle (or not so subtle) pressure on the mentor to focus on academics during meeting time. Training should help mentors understand how to engage in academic activities while at the same time allowing youth to shape their meetings and ensure that a central goal is to foster a strong, supportive bond with youth through collaborative interactions (Karcher et al., 2010).

Youth Involvement One of the lessons that research to date suggests is the need for older youth to have constructive, meaningful input into the program. Giving youth (and mentors) a voice in creating their match is a start (Karcher, 2007), but alone it is not likely to overcome the possible stigmatization resulting from peers seeing SBM as a program for youth who need help. Older youth may need additional opportunities to shape the form and function of the program in which they participate. This is not unique to mentoring but is a practice that seems to enhance the effectiveness of school-based mental health interventions more generally (Feldman & Silverman, 2004). One example of this practice was implemented in a high school in Norfolk, Virginia (R. Royster-Davis, personal communication). The first attempt to bring in a group of mentors to work with student mentees was not received well by the youth, who conveyed their disinterest in the program. Following this failed effort, program staff subsequently divided the identified youth into separate groups of boys and girls, and talked with them to see how a mentor might best serve their specific goals. By allowing the youth to set the parameters and goals for the program, staff felt they had created a degree of commitment they had not previously seen in the program. Programs could benefit from finding other ways to help older youth become “producers” of their own developmental experiences (Lerner, 1982) and foster true youth-adult partnerships (Wong, Zimmerman, & Parker, 2010).

Matching

Activities

In SBM programs, agencies have a limited number of mentors to match with a limited number of referred children, and they have to make these matches fairly quickly to increase the number of months a child can be matched within a given school year. Thus, it is our impression that, in most programs, significant efforts are not devoted to creating “perfect” matches (as is done in CBM). School location, gender, and scheduling are often the most salient matching factors. For SBM matches to yield strong bonds and life-changing experiences, we think more attention will need to be paid to this aspect of the program. Programs should capitalize on the fact that all of their mentees are present in the school at the same time, which provides the option of conducting a “meet-n-greet” between mentors and mentees to allow participant voice in the matching process (Karcher, 2007; for further discussion of mentor-youth matching, see Pryce et al., this volume, Chapter 29).

Another way that age and gender may moderate program impact is through the fit or alignment of the program’s activities with the inclinations or interests of the youth (which likely vary across grade levels and genders). For example, boys may value social and playful interactions with their mentors more than do girls (Bogat & Liang, 2005), and the school system does not easily afford such interactions as well or as often as it does the more communication-based and instrumental interactions that are favored by girls. Thus, to serve boys well, program staff may need to consider ways to foster interactive, fun, and companionship-based mentoring interactions in the school setting. Some mentoring programs in school and community contexts have strategically attempted to incorporate activities that pull youth outside their comfort zone. The GirlPOWER! site-based program in which matches meet in the community on the weekend is one example that uses activities to help

School-Based Mentoring  217 girls engage in career planning and exploration in ways that might be more comfortable for boys and differ from girls’ more typical approach to relationship building in mentoring. Conversely, programs that can help boys practice relationship building and communicating about personal matters could help them to develop the kinds of relating skills and connection forming that may be less common for them. Both approaches, however, require careful planning and mentor support. These approaches have usually been reported in CBM programs, but given the constraints that schools present for match interactions, particularly at the high school level, thoughtfully developing such approaches for school-based matches could fill a void. Match Support The school context provides the perfect opportunity for giving matches high-quality support— particularly in programs in which matches meet in one space at a set time—because staff can be present during match meetings. Given this opportunity, we recommend that program staff think about creative ways to provide their matches with support beyond the minimum requirements, depending on the needs of the school and the specific matches within that school. Staff can meet with small groups of volunteers to discuss specific topics. They can observe match interactions to troubleshoot. Staff can identify mentors with strengths in specific areas and match them with volunteers who are struggling with that issue, or meet with all mentors every week for a short debriefing. Mentors can complete logs after their meetings to note concerns to be discussed during support calls. But the setting can also work against in-depth support because staff who see matches every week may think that additional one-on-one support is not needed, when it may still be crucial. Thus, we strongly urge programs to ensure that staff provide match support beyond “just being there” because those matches that need the most help may be the least likely to take the initiative to approach staff even if they see them every week. Match Closures How matches end can be just as important as how they begin (see Spencer & Basualdo-Delmonico, this volume, Chapter 32). The quality of this ending has the potential to determine whether youth interpret it as the natural closing of an important relationship or as the child’s failure. In school-based matches, closures may be particularly important because, at least at present, relatively

few SBM programs are able to keep matches active across multiple years (see Herrera et al., 2007). Moreover, because the SBM program often will end every year for summer vacation, program staff know when most matches will end (if only for the summer). In this way, the school provides a unique opportunity to capitalize on the power of effectively closed matches. It is our impression that most SBM (and CBM) programs do not have clear guidelines for how to help matches navigate the closure process. Although the expectation of most SBM programs may be that mentors will sustain the relationship for at least a school year, programs are likely to also aspire to have relationships last well beyond this time frame. For this reason, SBM programs often may find it difficult to deal with relationship closures and thus avoid investing resources into strengthening this process. Perhaps programs and staff (like some mentors) feel that closing a relationship is admitting a failure (Spencer, 2007). Such feelings notwithstanding, we recommend that every mentor training explain how to close SBM matches well and that staff should have clear guidelines on how to support youth and mentors through relationship endings (Spencer & Basualdo-Delmonico, this volume, Chapter 32).

Conclusion Much has been learned about mentoring in the school context since the first edition of the Handbook. School-based mentoring programs have been the topic of some of the most rigorous research on youth mentoring to date. Yet there is considerable room (and now guidance) for making these programs more effective. Indeed, Big Brothers Big Sisters of America (2009) has used information from recent studies to develop and implement an “enhanced” model of SBM. Other programs can similarly use available findings to bolster program practices and potentially foster greater impacts. At the same time, research needs to continue to investigate “nextgeneration” questions regarding the effectiveness of SBM programs. Thanks to recent groundbreaking studies, we can say with good confidence that SBM programs can work. What is needed now is more indepth study of the how, for whom, when, and why of SBM program effectiveness. Likewise, naturally occurring mentoring relationships with adults in the school setting, though likely to have the potential for significant impacts, have been relatively neglected among researchers and, thus, deserve much greater

218  Programs and Contexts consideration. Forging more integrated strategies that harness the likely complementary strengths of natural and program-supported mentoring relationships in schools seems to us a particularly promising future direction for research and practice in this area.

References AOL. (2002). Mentoring in America. New York: AOL Time Warner Foundation. Bernstein, L., Rappaport, C. D., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S. Department of Education’s Student Mentoring Program: Final report. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education, U.S. Department of Education. Big Brothers Big Sisters of America. (2009). Enhanced school-based mentoring preliminary findings and high school Bigs. Unpublished document. Black, D. S., Grenard, J. L., Sussman, S., & Rohrbach, L. A. (2010). The influence of school-based natural mentoring relationships on school attachment and subsequent adolescent risk behaviors. Health Education Research, 25, 892–902. Bogat, G. A., & Liang, B. (2005). Gender in mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 205–217). Thousand Oaks, CA: Sage. Cavell, T. A., & Henrie, J. L. (2010). Deconstructing serendipity: Focus, purpose, and authorship in Lunch Buddy mentoring. In G. G. Noam, M. J. Karcher, & M. J. Nakkula (Eds.), New directions for youth development: Theory, practice, and research. The structure of mentoring interactions (pp. 107–121). Hoboken, NJ: Wiley. Cavell, T. A., & Hughes, J. N. (2000). Secondary prevention as context for assessing change processes in aggressive children. Journal of School Psychology, 38, 199–235. Darling, N., Hamilton, S., Toyokawa, T., & Matsuda, S. (2002). Naturally occurring mentoring in Japan and the United States: Social roles and correlates. American Journal of Community Psychology, 30, 245–270. de Boer, H., Bosker, R. J., & van der Werf, M. P. C. (2010). Sustainability of teacher expectation bias effects on long-term student performance. Journal of Educational Psychology, 102(1), 168–179. Delpit, L. (1995). Other people’s children: Cultural conflict in the classroom. New York: New Press. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Silverthorn, N. (2005a). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92.

DuBois, D. L., & Silverthorn, N. (2005b). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524. Durlak, J. A., Weissberg, R. P., & Pachan, M. (2010). A meta-analysis of after-school programs that seek to promote personal and social skills in children and adolescents. American Journal of Community Psychology, 45(3), 294–309. Eccles, J. S., & Barber, B. L. (1999). Student council, volunteering, basketball, or marching band: What kind of extracurricular involvement matters? Journal of Adolescent Research, 14, 10–43. Erickson, L. D., McDonald, S., & Elder, G. H., Jr. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367. Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton. Feldman, N., & Silverman, B. (2004). The “let’s talk about it” model—Engaging young people as partners in creating their own mental health program. In K. E. Robinson (Ed.), Advances in school-based mental health interventions: Best practices and program models (pp. 1–24). Kingston, NJ: Civic Research Institute. Fredriksen, K., & Rhodes, J. (2004). The role of teacher relationships in the lives of students. New Directions for Youth Development, 103, 45–54. Gastic, B., & Johnson, D. (2009). Teacher-mentors and the educational resilience of sexual minority youth. Journal of Gay & Lesbian Social Services: Issues in Practice, Policy & Research, 21, 219–231. Goodenow, C. (1993). The psychological sense of school membership among adolescents: Scale development and educational correlates. Psychology in the Schools, 30, 79–90. Greenberger, E., Chen, C., & Beam, M. R. (1998). The role of “very important” nonparental adults in adolescent development. Journal of Youth and Adolescence, 27, 321–343. Grossman, J. B., Chan, C., Schwartz, S., & Rhodes, J. E. (2012). The test of time in school-based mentoring: The role of relationship duration and re-matching on academic outcomes. American Journal of Community Psychology, 49, 43–53. Herrera, C. (1999). School-based mentoring: A first look at its potential. Philadelphia: Public/Private Ventures. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Herrera, C., Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, J. (2008). High school students as mentors: Findings from the Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia: Public/Private Ventures. Herrera, C., Sipe, C. L., McClanahan, W. S., Arbreton, A., & Pepper, S. K. (2000). Mentoring school-age children: Relationship development in community-based and

School-Based Mentoring  219 school-based programs. Philadelphia: Public/Private Ventures. Hinshaw, S. P. (2006). Stigma and mental illness: Developmental issues and future prospects. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathology, Vol 3: Risk, disorder, and adaptation (2nd ed., pp. 841–881). Hoboken, NJ: Wiley. Johnson, V. L., Holt, L. J., Bry, B. H., & Powell, S. R. (2008). Effects of an integrated prevention program on urban youth transitioning into high school. Journal of Applied School Psychology, 24(2), 225-246. Karcher, M. J. (2007). Meet-n-greet: A mentor-mentee matching approach for increasing the prevalence of naturally self-selected mentoring partners in program-based matches. Unpublished manuscript, University of Texas at San Antonio. Retrieved from http://www.utsasmile.org Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9, 99–113. Karcher, M. J., Davidson, A., Rhodes, J. E., & Herrera, C. (2010). Pygmalion in the program: The role of teenage peer mentors’ attitudes in shaping their mentees’ outcomes. Applied Developmental Science, 14, 212–227. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. New Directions in Youth Development, 126(Summer), 51–70. Karcher, M. J., & Nakkula, M. J. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. New Directions for Youth Development, 2010 126, 13–32. Karcher, M. J., & Sass, D. (2010). A multicultural assessment of adolescent connectedness: Testing measurement invariance across gender and ethnicity. Journal of Counseling Psychology, 57(3), 274–289. Karcher, M. J., & Spencer, R. (2012). The Study of Mentoring in the Learning Environment (SMILE): Multilevel modeling of setting-level program staff contributions to school-based mentoring program effectiveness. Unpublished manuscript, University of Texas at San Antonio. Keller, T. E., & Pryce, J. M. (2010). Mutual but unequal: Mentoring as a hybrid of familiar relationship roles. New Directions in Youth Development, 126, 33–50. Keller, T. E., & Pryce, J. M. (2012). Different roles and different results: How activity orientations correspond to relationship quality and student outcomes in school-based mentoring. Journal of Primary Prevention, 33(1), 47–64. Klem, A. M., & Connell, J. P. (2004). Relationships matter: Linking teacher support to student engagement and achievement. Journal of School Health, 74, 262–273. Kohut, H., & Elson, M. (1987). The Kohut seminars on self psychology and psychotherapy with adolescents and young adults. New York: Norton.

Larose, S., & Roy, R. (1995). Test of reactions and adaptation in college (TRAC): A new measure of learning propensity for college students. Journal of Edu­ cational Psychology, 87, 293–306. Larose, S., & Tarabulsy, G. M. (2005). Academically atrisk students. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 440–453). Thousand Oaks, CA: Sage. Lerner, R. M. (1982). Children and adolescents as producers of their own development. Developmental Review, 2, 342–370. LoSciuto, L., Rajala, A. K., Townsend, T. N., & Taylor, A. S. (1996). An outcome evaluation of Across Ages: An intergenerational mentoring approach to drug prevention. Journal of Adolescent Research, 11, 116–129. McMaken, J., Herrera, C., & Kauh, T. J. (2008). School and program factors associated with longer match length in school-based mentoring. Presented at the annual meeting of the Society for Prevention Research, San Francisco, CA. Nye, B., Hedges, L.V., & Konstantopoulos, S. (1999). The long-term effects of small classes: A five-year follow-up of the Tennessee class size experiment. Educational Evaluation and Policy Analysis, 21, 127–142. Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. Patrick, H., Anderman, L. H., & Ryan, A. M. (2002). Social motivation and the classroom social environment. In C. Midgley (Ed.), Goals, goal structures, and patterns of adaptive learning (pp. 85–108). Mahwah, NJ: Erlbaum. Pianta, R. C., Stuhlman, M. W., & Hamre, B. K. (2002). How schools can do better: Fostering stronger connections between teachers and students. New Directions for Youth Development: Theory, Practice, Research, 93, 91–107. Portwood, S. G., & Ayers, P. M. (2005). Schools. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 336–347). Thousand Oaks, CA: Sage. Rapee, R. M., Wignall, A., Sheffield, J., Kowalenko, N., Davis, A., McLoone, J., & Spence, S. H. (2006). Adolescents’ reactions to universal and indicated prevention programs for depression: Perceived stigma and consumer satisfaction. Prevention Science, 7, 167–177. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J., Lowe, S. R., Litchfield, L., & Walsh-Samp, K. (2008). The role of gender in youth mentoring relationship formation and duration. Journal of Vocational Behavior, 72, 183–192. Rhodes, J. E., Spencer, R., Keller, T. E., Liang, B., & Noam, G. (2006). A model for the influence of mentoring relationships on youth development. Journal of Community Psychology, 34, 691–707.

220  Programs and Contexts Richman, J. M., Rosenfeld, L. B., & Bowen, G. L. (1998). Social support for adolescents at risk of school failure. Social Work, 43, 309–323. Ritter, G. W., Barnett, J. H., Denny, G. S., & Albin, G. R. (2009). The effectiveness of volunteer tutoring programs for elementary and middle school students: A meta-analysis. Review of Educational Research, 79, 3–38. Rosenthal, R. (1993). Interpersonal expectations: Some antecedents and some consequences. In P. D. Blank (Ed.), Interpersonal expectations: Theory, research, and applications (pp. 3–24). New York: Cambridge University Press. Sánchez, B., & Reyes, O. (1999). Descriptive profile of the mentorship relationships of Latino adolescents. Journal of Community Psychology, 27, 299–302. Schwartz, S. E. O., Rhodes, J. E., Chan, C. S., & Herrera, C. (2011). The impact of school-based mentoring on youth with different relational profiles. Developmental Psychology, 47, 450–462. Selman, R. L. (1980). The growth of interpersonal understanding: Developmental and clinical analyses. New York: Academic Press. Selman, R. L. (1990). Toward the application of socialcognitive developmental theory. Clinical Psychology Review, 10, 567–588. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354.

Spencer, R., Jordan, J. V., & Sazama, J. (2004). Growthpromoting relationships between youth and adults: A focus group study. Families in Society, 85, 354–362. Tierney, J. P., Grossman, J. B., & Resch, N. L. (1995). Making a difference: An impact study of Big Brothers Big Sisters. Philadelphia: Public/Private Ventures. Tobler, N. S., Roona, M. R., Ochshorn, P., Marshall, D. G., Streke, A. V., & Stackpole, K. M. (2000). Schoolbased adolescent drug prevention programs: 1998 meta-analysis. Journal of Primary Prevention, 20, 275. Van Ryzin, M. (2010). Secondary school advisors as mentors and secondary attachment figures. Journal of Community Psychology, 38, 131–154. Viadero, D. (2009). “No effects” studies raising eyebrows. Education Week, 28(27), 1, 14–15. Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials of schoolbased mentoring: Making sense of mixed findings. Social Policy Report, 24(3), 1–21. Wong, N., Zimmerman, M., & Parker, E. (2010). A typology of youth participation and empowerment for child and adolescent health promotion. American Journal of Community Psychology, 46, 100–114. Zimmerman, M. A., Bingenheimer, J. B., & Behrendt, D. E. (2005). Natural mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 143–157). Thousand Oaks, CA: Sage.

15  AFTER-SCHOOL PROGRAMS Megan A. Mekinda and Barton J. Hirsch1

Introduction Since publication of the first edition of the Handbook of Youth Mentoring, after-school programs have remained in the spotlight as a key component of the youth services infrastructure. Such programs continue to serve a wide range of youth in terms of age, racial and socioeconomic background, and geographic location. However, findings are mixed with regard to the overall effectiveness of such programs to foster targeted outcomes including academic achievement, skill development, and general psychosocial well-being. Considerable variation in program quality helps to account for differences in effectiveness, and recent work in the field has focused increasingly on improving the quality of both program design and implementation to best meet the needs of participating youth. Significantly, almost without exception this work highlights staff as among the most important features of high-quality, effective programs, if not “the critical ingredient” (Grossman et al., 2002; Hirsch, Deutsch, & DuBois, 2011; Rhodes, 2004). In short, positive youth-staff relationships—particularly those that reflect elements of care and support characteristic of mentoring— appear to be a primary reason that youth attend afterschool programs and why they choose to stay (Arbreton, Sheldon, & Herrera, 2005; Deschenes et al., 2010; Walker & Arbreton, 2004). This chapter examines the nature, determinants, and consequences of mentoring within after-school programs. We expand on theory and research presented in the corresponding chapter from the first edition of the Handbook (Hirsch & Wong, 2005). Our

conclusion remains the same: After-school programs are unique and potentially rich contexts for mentoring. New to this chapter is an elaborated theoretical model for understanding mentoring in after-school programs. Also new are findings from a series of recent studies that provide greater insight into the nature of adult-youth relationships within the programs, their effect on program quality, and associations with outcomes for both youth and staff. The final section outlines recommendations for practice to foster and support strong mentoring ties within the after-school program context. Our discussion focuses on school- and communitybased comprehensive after-school programs such as the Boys & Girls Clubs of America and Beacon Centers. These programs are often multiaged, and many youth attend for multiple years. They typically offer a wide range of activities including homework help, organized recreation, and free time to relax and socialize. However, the overall structure and specific offerings vary considerably from one program to the next. Although more focused after-school activities such as organized sports or fine and performing arts groups certainly have potential as contexts for mentoring, they have been less explored in the literature with regard to relationships between adult leaders and participating youth.

Theory The theoretical model presented in Figure 15.1 is an elaborated version of that published in the original edition of the Handbook (Hirsch & Wong, 2005).

1

Preparation of this chapter was facilitated in part by an IES pre-doctoral fellowship awarded to the first author and awards to the second author from the William T. Grant Foundation and the Wallace Foundation.

221

222  PROGRAMS AND CONTEXTS We retain the main features of the model, including its focus on potential mentoring behaviors of program staff, environmental and individual characteristics that establish the context for mentoring, and resulting outcomes for both youth and adults. Additions to the model reflect recent work that provides valuable insight into the nature of mentoring relationships within after-school programs and that stresses the centrality of positive adult-youth relationships to overall program quality. The model also now incorporates the qualities and contributions of nonstaff mentors, such as volunteers from the community or partnering organizations. Characteristics of Mentoring Relationships We begin by discussing important characteristics of mentoring relationships because the heart of our concern is what goes on between youth and mentors within the after-school program context. Mentoring Arrangement. To capture the varied nature of mentoring ties within comprehensive after-school programs, we expanded our original framework

Figure 15.1

to recognize different mentoring arrangements in terms of which and how many persons are involved. Qualities shared by many programs encourage considerable variety in relationship structure. For example, programs are typically staffed by multiple adults, which allows for both group mentoring between one staff member and multiple youth and also collective mentoring between a single youth and multiple staff. Intensive, one-on-one mentoring relationships are also possible as youth may develop a particularly strong bond with a particular staff member. Programs that partner with formal mentoring organizations create even greater opportunity for one-on-one mentoring. Multiage programs, particularly if youth attend for several years, allow for trilevel mentoring, in which youth mentored by one or more adults in turn serve as mentors for younger participants (Deutsch, 2008). In a similar vein, there is also the possibility of reciprocal mentoring, in which youth, by virtue of a particular skill set or knowledge base, are able to provide guidance and instruction to program staff (Kafai, Desai, Peppler, Chiu, & Moya, 2008). The variety of mentoring arrangements observed in after-school programs supports a broader conceptu-

A Revised Framework for Investigating Mentoring Relationships in After-School Programs

Characteristics of Mentoring Relationships Environmental Characteristics • Program/ Organization • Community • Family Individual Characteristics • Youth • Staff/Volunteer

Mentoring Arrangement • • • • •

Individual Group Collective Trilevel Reciprocal

Program Experience • Attendance • Engagement • Retention

Outcomes

Mentoring Behaviors • • • •

Emotional Support Guidance/Teaching Sponsorship/Advocacy Learning/Collaboration

• Youth • Staff/Volunteer

After-School Programs  223 alization of youth mentoring as the field continues to refine the construct (DuBois & Karcher, 2005). For example, trilevel and reciprocal mentoring challenge the notion that mentors must be adults or at least older than the mentee, as well as the idea that mentoring is a hierarchical or unidirectional relationship (Kafai et al., 2008). Mentoring Behaviors. We maintain a focus on three broad facets of mentoring behavior observed within after-school programs. Emotional support reflects empathy for youth and responsiveness to their personal needs and perspectives, which can foster deep mutual respect and caring. Guidance or teaching reflects support for cognitive and skill development through instruction and role modeling. Sponsorship or advocacy reflects effort to provide access to connections or experiences that promote development. To this list we have added learning/collaboration, a form of reciprocal mentoring in which youth teach adults a specific skill they have mastered or the pair works together to solve a problem or complete a project (Kafai, Desai, Peppler, Chiu, & Moya, 2009; Kafai et al., 2008). Mentors who act as learners or collaborators can model valuable learning behaviors, including strategies to deal with information one does not understand. Mentoring in this way also provides opportunities for youth to take an active leadership role, which can be particularly important among older participants to retain their interest in the program (Deschenes et al., 2010). Determinants of Mentoring Behavior Relationships between adults and youth in afterschool programs are not synonymous with mentoring relationships. Some adult-youth interactions are characterized by conflict and tension with few redeeming positive qualities (Dworkin & Larson, 2006; Hirsch et al., 2011). Even positive relationships can fail to incorporate the mentoring behaviors described above (Rhodes, 2004). Therefore, it is of great importance to identify factors that influence the incidence and quality of mentoring within afterschool settings. In this section, we consider individual and environmental characteristics that shape the overall context for mentoring. Individual Characteristics. Individual characteristics of both program staff and participants are important determinants of mentoring behavior. Their relative age, gender, and ethnic and socioeconomic background can serve as the basis for rapport building, expose each party to valuable new perspectives and insights, or perhaps inspire conflict, depending on

the degree of similarity between mentor and mentee. Youth’s age, gender, and background also influence their developmental needs and consequently the specific mentoring behaviors from which they are most likely to benefit. Personality and social attributes of youth and adults help shape the nature and quality of their interactions, including attachment history, mental health, social skills, and attitudes (Eccles & Gootman, 2002; Rhodes, 2002; Zimmerman, Bingenheimer, & Behrendt, 2005). Finally, shared interests (e.g., with regard to activities, tastes, or preferences) can serve as the basis for in-depth relationships, particularly within after-school programs where many youth are initially drawn by the types of activities offered (Grossman & Bulle, 2006), thereby potentially amplifying the benefits of mentoring ties (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). Environmental Characteristics. With regard to environmental determinants of mentoring behavior, potentially most important are the program and organizational characteristics that shape the nature of adult-youth interactions. Basic characteristics like enrollment, attendance, and youth-to-staff ratio determine the number of youth and staff available for such relationships and also possible mentoring arrangements, in other words, opportunities for oneon-one, collective, or group mentoring. Similarly, the age range of youth and degree of participant retention from year to year affects the extent to which older youth and veterans of the program can engage in trilevel mentoring. Activity offerings are also important to consider. Programs typically sponsor a range of activities from structured homework help to semiorganized sports and informal socializing. Activities provide the framework for youth-staff interactions and may elicit different mentoring behaviors. For example, structured activities create opportunities for youth and staff to bond over shared interests and for staff to provide instruction and feedback for skill development. Opportunities for informal socializing may provide greater opportunity for youth to discuss personal issues and solicit emotional support (Walker & Arbreton, 2004). Organizational practices that affect the recruitment, training, and retention of staff may have a direct impact on the formation and quality of mentoring relationships (Grossman & Bulle, 2006). Recruitment practices determine the qualities that staff bring to the program, including their interest in and ability to relate to young people. Programs can differ in the extent to which they prioritize relationship building, which is reflected in the level of support and training

224  PROGRAMS AND CONTEXTS for staff members to fulfill mentoring roles. High staff turnover can undermine the formation of longterm, stable relationships characteristic of effective mentoring (Arbreton et al., 2005; Rhodes, 2002; Rhodes & Lowe, 2009; Walker & Arbreton, 2004). One final but important organizational characteristic is the level of collaboration between the after-school program and outside organizations. Mentoring within after-school programs need not be the exclusive responsibility of program staff, and partnerships with formal mentoring organizations are a promising way to increase participants’ access to caring adults (Higginbotham, Harris, Lee, & Marshall, 2006; Forum for Youth Investment, 2006). Such partnerships offer logistical and economic benefits to both organizations. Volunteer mentors are provided reliable access to youth as well as a dependable meeting place. The presence of volunteer mentors lowers the youth-adult ratio in the program and thus offers young participants a degree of individual attention that might not be possible without the additional manpower. Staff from both organizations may have the opportunity to learn from each other, drawing from their respective backgrounds, training, and experiences with youth. Overall, such partnerships have the potential to create a network of caring adults in the lives of young people rather than a system of independent organizations that compete for the time and attention of youth as well as for limited sources of funding. Characteristics of youth’s family and community may affect mentoring relationships within after-school programs in important ways. Parental involvement (or lack thereof) can be a source of support or frustration for the mentor (DuBois, Holloway, Valentine, & Cooper, 2002; for an in-depth discussion of the role of the family in the context of youth mentoring, see Taylor & Porcellini, this volume, Chapter 31). Community characteristics such as incidence of violence and gang activity may dictate youth’s needs and be frequent topics of discussion and guidance with their mentors. They may also directly affect youth’s ability to travel to and from the program safely and therefore their frequency of contact with program staff. Program Experience and Outcomes It is reasonable to expect that youth engaged in mentoring relationships within after-school programs could experience many of the same outcomes associated with mentoring in other contexts. These include gains in social-emotional and identity development, cognitive and intellectual growth, and greater access to social and cultural capital (Rhodes & Lowe,

2009). However, after-school programs are not simply contexts for mentoring. They typically offer a wide range of resources and activities, which provide youth with access to peer networks, opportunities to develop valuable skills and competencies, and on the most basic level, a safe space to avoid high-risk activity on the streets. Within such programs, positive adult-youth relationships may be critical not only for their direct benefits, but also for the indirect benefits they provide by facilitating more meaningful involvement of youth in the program itself (Arbreton et al., 2005; Grossman & Bulle, 2006; Kuperminc et al., 2005; Rhodes, 2004; see also Cavell & Elledge, this volume, Chapter 3). In short, mentoring relationships may enhance positive outcomes associated with participation in after-school programs, which include gains in mental health and academic achievement as well as reductions in high-risk behavior (Mahoney, Larson, Eccles, & Lord, 2005). The revised framework highlights how mentoring in this way may serve an important mediating function with regard to the positive impact of after-school programs on participants. Adults are likely to experience their own set of outcomes as a result of their relationships with youth. For example, the act of mentoring may affect program staff’s sense of self-efficacy and job satisfaction. Benefits associated with the act of mentoring might also be experienced by program youth involved in trilevel or reciprocal mentoring arrangements.

Research We searched for studies directly related to mentoring in after-school programs through three main sources: (a) recommendations from colleagues with expertise in after-school programs and/or youth mentoring, (b) online databases and search engines, specifically PsycINFO and Google Scholar (search terms included derivations of the words mentor, relationship, after-school, and program), and (c) references cited in relevant publications. As with the review in the first edition of the Handbook, we found the most compelling and detailed evidence of mentoring in after-school programs to come from a series of recent in-depth qualitative investigations. The work provides essential insight into the complexities and nuances of mentoring at particular sites, promoting a greater understanding of the unique contributions of mentoring to youth’s program experience. Accordingly, the present review draws heavily from two main bodies of qualitative work. First, Hirsch and colleagues provided a series of in-depth studies of several Boys & Girls Club

After-School Programs  225 sites that serve predominantly minority and lowincome youth. Findings from Hirsch (2005) were reviewed in the first edition of the Handbook (Hirsch & Wong, 2005). Here we discuss two subsequent investigations. The first examined three club sites of high, moderate, and low quality in an effort to identify program characteristics that lead to success and failure (Hirsch et al., 2011). Hirsch and colleagues drew from ethnographic observations, survey data, and interviews with youth and staff. Their analysis focused extensively on youth-staff relationships, particularly the synergies they create with other key site characteristics to shape youth’s experience in the club and the resulting outcomes over the course of a year. In the other study, Deutsch (2008) presented qualitative data from 4 years of participant observation at a single Boys & Girls Club site. She used field notes, interviews, and case studies of adolescent participants to examine the confluence of gender, race, and class in identity construction of marginalized youth, highlighting the central role of relationships in this process. In the second body of work, Kafai and colleagues (Kafai et al., 2008; Kafai et al., 2009) presented qualitative data on mentoring in Computer Clubhouses, programs typically housed within larger community centers in underserved areas. The Clubhouses are not as comprehensive as the Boys & Girls Clubs studied by Hirsch and colleagues; rather, they engage youth in technology-related design and discovery projects with the ultimate goal of helping them develop generalizable skills and self-confidence. Data consist of volunteer mentors’ field notes from their interactions with youth and their responses during a group interview. In the following sections, we review important findings from each of these studies and incorporate results from additional mixed-method and quantitative investigations when available. In particular, Public/Private Ventures offers a rich body of mixedmethod evaluations of a range of after-school programs nationwide including Boys & Girls Clubs (Arbreton, Bradshaw, Sheldon, & Pepper, 2009), Beacon Centers (Grossman, Campbell, & Raley, 2007; Walker & Arbreton, 2004), and others (e.g., Deschenes et al., 2010; Grossman et al., 2002). Although these studies do not focus on mentoring per se, they attend to adult-youth relationships as a key component of after-school programs overall. Their findings reveal how aspects of these relationships appear to shape youth’s experience and contribute to program quality and effectiveness. All

2

findings are organized according to their relevance to the theoretical model in Figure 15.1. We focus especially on evidence that supports new additions to our framework, such as the range of mentoring arrangements and the role of mentoring as a mediator of program effectiveness. Characteristics of Mentoring Relationships Collectively, the studies confirm a wide range of mentoring arrangements and behaviors within after-school programs. For Hirsch et al. (2011), of particular note was the importance of collective mentoring, a phenomenon that has received little attention in the mentoring literature to date. The researchers described collective mentoring as a “powerful synergistic process” in which the combined efforts of multiple staff were observed to enhance greatly the level and quality of support provided to a particular youth. Case studies of individual youth also documented instances of intensive mentoring from a single staff member, which at times reflected a level of care and intimacy to suggest the mentor’s status as a surrogate parent. Hirsch and colleagues argued that, just as the Clubs were distinctively suited for collective mentoring, they could also provide support for meaningful oneon-one relationships in ways superior to stand-alone mentoring programs. The case of one Club member, who chose the pseudonym Pocahontas,2 illustrates how the Club environment facilitated her strong bond with Manuel, a staff member she described as “like a father” (p. 68). In contrast to stand-alone programs in which mentors and mentees are deliberately matched, in the Club Pocahontas sought out the staff member to whom she was naturally drawn, affording her a sense of agency and ownership in the relationship. The pair engaged in a range of activities on a near-daily basis, promoting a degree of consistency, breadth, and depth in the relationship more difficult to achieve in stand-alone programs. Also important, the researchers noted that the relationship was cross-gender, a rare occurrence in stand-alone programs but extremely valuable to Pocahontas who, at that time in her life, was mourning the loss of her father and eager for paternal-like companionship. Deutsch (2008) highlighted Boys & Girls Clubs as unique contexts for trilevel mentoring as they represent one of the few places in which youth can interact with adults and same-age or younger peers simultaneously. She concluded that opportunities to

All staff and youth names included in this chapter are pseudonyms used by the original researchers.

226  PROGRAMS AND CONTEXTS receive support and guidance from adult staff while also serving as role models for younger participants established an appropriate balance of relatedness and autonomy for youth. Trilevel mentoring also helped youth to develop and be recognized for positive qualities such as leadership and responsibility. For example, Deutsch shared the story of Lorenzo, a longtime participant at one of the clubs, whose close relationship with Charles (a staff member) and selfdescribed status as a role model for other youth allowed him to establish a meaningful place for himself within the club and contributed positively to his self-concept. In analyzing an instance of trilevel mentoring observed on the basketball court, Deutsch explained, “During this game Lorenzo has the opportunity to interact with Charles as a mentee, showing off a bit when he makes a good play and receiving validation from Charles. Yet he is also able to act as a mentor, mimicking Charles’s style of friendly and fun interaction with the younger kids, both encouraging them and making them laugh” (p. 74). Also important, Deutsch asserted that the trilevel mentoring observed here benefits not only Lorenzo, but also the younger members of the club: While they [younger participants] have Charles as an adult model of who they may be down the road, they also have Lorenzo as a model of who they may be in a couple of years. Furthermore, Lorenzo provides a model of self that includes individual traits and achievements, as well as a position within the community. Youth observe Lorenzo’s relationship with Charles and the role he himself plays within the club. Thus, younger members can see a possible future role for themselves in the club. (p. 74) Deutsch’s findings concerning the potential benefits of trilevel mentoring are consistent with a recent meta-analysis of formal mentoring programs, which indicated positive effects of cross-age peer mentoring comparable in magnitude to those of adult-youth mentoring (DuBois et al., 2011). Furthermore, in their study of a range of after-school programs nationwide, Deschenes and colleagues (2010) found that providing leadership opportunities was the single strongest predictor of high retention among older participants. Interviews with program providers revealed that, for high school youth, such opportunities included mentoring of younger participants. Finally, Grossman et al. (2007) concluded from program observation that peer teaching and mentoring was a strategy used by some Beacon instructors to enhance youth engagement. However, the researchers’ quantitative analyses did not reveal

a significant link between cooperative peer learning and engagement or learning outcomes. With regard to mentoring behaviors, Kafai and colleagues described a spectrum within Computer Clubhouses ranging from mentor directed (i.e., mentors as teachers and facilitators) to youth directed (i.e., mentors as passive observers or learners; Kafai et al., 2008; Kafai et al., 2009). Mentors most often reported co-constructing or collaborating with students, which, the researchers argued, is evidence of mentoring as a reciprocal and equitable relationship. Kafai et al. (2008, 2009) attributed the prevalence of collaboration to the nature of the design activities, in which demands for creativity and experimentation placed mentor and mentee on a level playing field. Notably, the content of youth-mentor interactions within the Clubhouses appeared to be focused largely on issues related directly to the project itself. Grossman et al. (2007) observed a similar pattern in the Philadelphia Beacon Centers: For program activities that targeted the development of a specific skill (e.g., karate), instructor support was almost exclusively instrumental (i.e., skill-specific guidance and teaching). Yet, even in the absence of emotional guidance, youth rated instructors as supportive, seemingly because they interpreted direction and feedback as evidence of care and responsiveness. It is important to note, however, that some Beacons instructors did in fact provide emotional support, investing time to learn about and address the needs and interests of individual youth and to hang out and have fun. The findings of Hirsch and colleagues (2011) illustrate the balanced nature of mentoring possible within after-school programs. Staff of the Boys & Girls Clubs demonstrated a full range of mentoring behaviors and took advantage of the comprehensive nature of the clubs to support youth in a variety of domains. Again, the case of Pocahontas and Manuel is illustrative. Data reveal instances in which Manuel assisted Pocahontas with homework, coached her through feelings of sadness and loneliness in the wake of her parents’ death, and promoted her physical health and safety by encouraging her to eat well and escorting her to her home in a dangerous neighborhood. In addition, by inviting her to attend staff dinners, he facilitated positive relationships with other Club leaders, enhancing the potential for collective mentoring. Thus, Manuel served as tutor, confidant, nurturer, protector, and advocate, representing three of the four facets of mentoring behavior identified in our model (there were no reported instances of Manuel in an obvious collaborative role; Hirsch et al., 2011). The different facets of mentoring do not always complement each other within the after-school

After-School Programs  227 program setting. Indeed, a qualitative study of three youth activism programs revealed the potential for tension among them. Kirshner (2008) explained how adult participants had to balance their desire to empower youth with the complex demands of a successful social activism campaign, the substantive focus of each program. The adults responded to this tension in distinct ways, revealing a series of tradeoffs among deliberate mentoring strategies. In one program, they prioritized youth leadership by acting predominantly as facilitators, a role that corresponds roughly to our concept of advocacy. Although this strategy promoted a sense of ownership and belonging among youth, it created fewer opportunities for them to observe and model the behavior of seasoned activists. In another program, adults adopted an apprenticeship model by providing considerably more direction for youth, reflecting the teaching behavior included in our framework. In this case, youth had less autonomy but appeared to develop a more sophisticated skill set. Finally, in the third program, the adults promoted joint work, a strategy that corresponds to our conception of collaboration. Here, adult-youth interactions were collegial but lacking in support for new skill development, which limited less experienced youth to peripheral involvement in the campaign. The findings highlight the need for program providers to be mindful of the effect of specific mentoring behaviors on youth’s experience and to use these behaviors strategically to support program goals. Determinants of Mentoring Behavior Findings are mixed regarding the effects of individual characteristics on mentoring relationships. In the Boys & Girls Clubs and San Francisco Beacon Centers, background similarities between staff and youth (e.g., growing up in the same neighborhood) were observed as a basis for bonding and allowed staff to draw from shared experience to provide targeted support and guidance (Deutsch, 2008; Hirsch et al., 2011; Walker & Arbreton, 2004). In contrast, most mentors in the Computer Clubhouses did not share background characteristics with their mentees. Dissimilarities did not seem to prevent the formation of mentoring ties; however, as noted the interactions did appear more limited in terms of content. Grossman et al. (2007) reported that individual staff characteristics (specifically age, gender, racial similarity to youth, and levels of education and training) were statistically unrelated to youth’s ratings of adult support in the Philadelphia Beacon Centers. The researchers cautioned that these findings are extremely preliminary, given the

small (N = 30) and likely unrepresentative sample of adult staff. However, they are consistent with the finding from the recent meta-analysis of youth mentoring programs that effects were strongest when mentors and youth were matched based not on race or ethnicity, but rather on similarity of interests (DuBois et al., 2011). Attitudes and personality characteristics of youth and staff influenced the nature of mentoring relationships within the programs studied. For the adolescents in Deutsch’s study (2008), of particular importance was staff members’ respect for youth, which distinguished them from other adults with whom youth interacted, particularly teachers. Hirsch et al. (2011) focused less on general personality traits and more on the compatibility of specific youth-staff dyads. The case studies included a portrait of a young person who shared a strong mentoring tie with one staff member but an adversarial relationship with another, which suggests the value of natural mentoring ties that capitalize on mutual attraction. However, the case studies also revealed instances in which natural ties failed to develop. This aspect of Hirsch and colleagues’ findings suggests the need for targeted training for staff to better serve as mentors and also speaks to the potential benefits of a partnership with a formal mentoring program. With regard to program and organizational characteristics, the most notable insights from our present review concern the relationship between program activities and mentoring behavior. Activities are a defining feature of after-school programs and often the major draw for participants; determining their influence on the nature of adult-youth relationships is key to understanding these programs as contexts for mentoring. Some studies suggest that certain types of activities lend themselves more naturally to specific mentoring behaviors than others. As noted earlier, highly task- or skill-oriented activities in the Computer Clubhouses and Philadelphia Beacon Centers appeared to elicit mostly instrumental support from adults (Grossman et al., 2007; Kafai et al., 2009). Drawing from observations of a wider range of programs, Grossman and colleagues (2002) reported that enrichment activities such as sports or fine arts were most effective in fostering adult-youth relationships characterized by warmth, intimacy, and the provision of emotional support. In contrast, academic activities (e.g., tutoring and homework help) tended to be more formal and narrowly focused. The occasional absence of activity may also be conducive to strong adult-youth relationships. For example, in the Boys & Girls Clubs studied by Hirsch et al. (2011), opportunities for

228  PROGRAMS AND CONTEXTS downtime and informal socializing allowed youth to seek out adults with whom they related and avoid those they disliked. Informal socializing also allowed for moments of one-on-one interaction, which is important for individualized, targeted mentoring. The studies also demonstrate that mentoring behavior can be independent of activity structure or substantive focus. For example, Walker and Arbreton (2004) reported similar levels of adult warmth and support across arts and recreation, academic, career development, and leadership activities in the San Francisco Beacon Centers. Conversely, Kirshner (2008) revealed disparate mentoring behaviors across three programs despite their shared activity focus (a social activism campaign). He attributed these differences in part to varying philosophies about working with youth, a reflection of the priorities and values of the programs themselves. We conclude from our review that program goals—and the skill and intention of program staff to carry them out—are a stronger determinant of mentoring behavior than activity type. However, program providers can be strategic in selecting activities most likely to support specific mentoring objectives. Outcomes The research reviewed here provides some fairly compelling evidence that adult-youth relationships help to mediate the positive impact of after-school programs on participants. At the most basic level, youth who build strong relationships with program staff or volunteers attend more regularly and for longer periods of time, which means they have greater exposure to other valuable aspects of the program environment (Deschenes et al., 2010; Grossman et al., 2007; Walker & Arbreton, 2004). Findings also suggest a more direct impact of mentoring on youth outcomes. In their evaluation of Boys & Girls Clubs, Arbreton and colleagues (2009) determined that more frequent participation in the Clubs predicted positive gains in character and citizenship, academic success, and healthy lifestyle, the three priority outcome domains of the organization. The researchers drew from youth and adult interview data to determine that adult-youth relationships within the Clubs, characterized as stable and supportive, were consistently one of the key program features to be linked with promotion of positive outcomes across these domains. Furthermore, Grossman and colleagues (2007) reported a positive statistical relationship between youth’s ratings of adult support and self-reported enjoyment, engagement in program activities, and, especially, level of learning.

Qualitative data detail how program staff engaged in “online” mentoring in which they provided direct guidance as youth navigated diverse situations during program hours, such as conflict with peers or difficulty mastering a particular skill or activity (Hirsch et al., 2011; Kafai et al., 2008). For example, Grossman et al. (2002) reported that adult leaders “were able to explain the relevance and importance of the activity or the skills [youth] were developing, challenge youth to push beyond their comfort level, encourage them to persevere and praise their accomplishments” (p. 24). These findings suggest that mentors’ provision of in-the-moment support and their role in enhancing youth’s level of involvement can lead to a more enriching program experience overall. In a rare purely quantitative evaluation of adult-youth relationships in after-school programs, Serido, Borden, and Perkins (2011) found that youth who perceived more positive relationships with adults reported a stronger sense of voice. In turn, youth with a stronger sense of voice perceived greater benefits to program participation, such as the development of new skills and a sense of personal fulfillment. These findings suggest that adult-youth relationships may have had both direct and indirect impacts on youth’s perceived benefits. Findings were cross-sectional and correlational, which means one must be cautious in making causal interpretations. Nevertheless, the study represents a step in the right direction in terms of empirically testing specific mechanisms through which mentoring in afterschool programs affects youth outcomes. Of the studies reviewed, the research of Kafai and colleagues (2008) is unique in its focus on the experiences and outcomes of mentors. They argued that working with youth from less privileged backgrounds allowed mentors to reflect on complex social issues including negative stereotypes of the youth, disparities in access to knowledge and resources, and the implications of mentors’ own status within society. Mentors’ experiences as learners and collaborators also appeared to challenge their preconceptions of mentoring as a top-down process and to foster greater appreciation for the contributions of the youth. It is important to note that these findings are limited to mentors’ self-reported change in perceptions and attitudes (via field notes and group interviews) without any measure of related behavioral or psychosocial outcomes (e.g., self-efficacy). Summary and Recommended Future Directions Ultimately, the studies reviewed here represent a valuable addition to the research base examined in the original Handbook chapter and further our

After-School Programs  229 understanding of mentoring within after-school programs. However, there remains considerable room for growth. We call for future research that employs a wider range of methods, particularly quantitative investigations that track longitudinally the relationship between qualities of mentoring ties and aspects of youth experience and growth. Qualitative data from the studies reviewed here indicate valuable topics of investigation. Of particular note, Deutsch (2008) argued that mentoring relationships in the Boys & Girls Clubs resulted in positive gains in youth identity formation. Staff’s ability to serve as positive role models and a constant source of support and encouragement for youth’s own potential to succeed appeared to expand youth’s understanding of possible and future selves in ways that challenged negative stereotypes of minority, economically disadvantaged youth. We hope future work will also include attention to the impact of organizational practices, particularly the addition of a formal mentoring component (e.g., through partnerships with outside mentoring organizations).

Practice In this final section, we offer recommendations for practice based on the preceding considerations and evidence of potentially successful strategies from

the programs reviewed (see Table 15.1). In short, for high-quality mentoring to become a reliable component of after-school programs, our view is that it should be an explicit goal, reflected in each aspect of the organization. Here we focus on practices related to staff, adult-youth interactions, and finally, community resources. Staff Mentoring within after-school programs depends on the sustained presence of dedicated staff who are willing and able to assume a mentoring role. We believe staff recruitment, training, and incentives are perhaps the smartest investments to be made with often very limited resources. To the extent possible, programs must be selective in their hiring, targeting individuals who respect youth, care about their well-being beyond the interests of the program, and set high expectations for their behavior and development (Deschenes et al., 2010; Deutsch, 2008; Grossman & Bulle, 2006). Staff who share background characteristics with youth (e.g., young age, home community) may be particularly well-equipped to build strong mentoring ties, given their personal understanding of youths’ perspectives and daily struggles (Grossman et al., 2002; Hirsch et al., 2011; Walker & Arbreton,

Table 15.1   Checklist for Practitioners Topic

Recommendations

Staff

Recruit staff with the ability and motivation to serve as mentors. Provide targeted training, feedback, and encouragement to support staff in their mentoring efforts. Arrange for adequate incentives (e.g., competitive salaries) to retain staff for periods long enough to form strong mentoring ties.

Adult-youth interactions

Make mentoring an explicit priority during program activities. Select activities to reflect specific mentoring goals. Allow youth to move among activities and/or incorporate periods of informal socializing to encourage more in-depth, personal interaction and natural bonding.

Community resources

Provide support and encouragement for older youth to act as mentors for younger participants. Involve individuals from the community in program activities to expose youth to a broader range of potential mentors and increase opportunities for one-on-one mentoring. Consider well-planned and clearly articulated partnerships with formal mentoring organizations such as Big Brothers Big Sisters.

230  PROGRAMS AND CONTEXTS 2004); however, charisma and passion for youth work are likely more essential qualities. Once hired, staff are likely to benefit from ongoing support and instruction. Conversations during staff meetings or case conferences about strategies for effective mentoring (including those focused on the collective mentoring of a particular youth; Arbreton et al., 2005), mentoring workshops, and other forms of training (e.g., that relate to interpersonal skills and youth development) should be routine (Walker & Arbreton, 2004). Expectations for high-quality mentoring might also be endorsed through administrators’ formal evaluations of staff and recognition of successful efforts. Perhaps the biggest challenge for after-school programs regarding staff is retention. High turnover is a pervasive issue that directly affects the potential for long-term, personal connections with youth (Grossman et al., 2002). Competitive salaries, benefits, and additional incentives (e.g., opportunities for promotion and full-time status) are likely crucial to attract high-quality staff members and retain them for periods long enough for meaningful mentoring ties to form (Arbreton et al., 2005; Grossman et al., 2002; Walker & Arbreton, 2004). Adult-Youth Interactions In our view, second in importance only to the presence of high-quality staff is the existence of opportunities for staff to interact with youth in ways that foster mentoring ties. Our discussion of the relationship between activity type and mentoring behaviors indicates the potential for intentional design of activities to support overall program objectives, including specific goals for mentoring. For example, programs that value a broad range of mentoring behaviors may be best served by a broad range of activity offerings. Those that value natural mentoring ties may be best served by creating opportunities for informal socializing or by allowing youth to move freely among activities so that they can seek out staff with whom they most relate. In any case, we strongly recommend that all decisions concerning activity structure and focus reflect a clear understanding among staff of the nature and degree of mentoring expected to occur across program components. Community Resources Although existing research has focused largely on paid staff, after-school programs would be well served to consider how others from the program or outside community might also contribute valuable

mentoring services. For example, they might engage older youth as mentors for younger participants (i.e., deliberately encourage trilevel mentoring; Deutsch, 2008). Programs might also involve individuals from the community in planned projects or activities, adults who can serve as valuable sources of information and other forms of social capital (Jarrett, Sullivan, & Watkins, 2005). Finally, partnerships with formal mentoring programs are a promising source of both manpower and expertise, which ideally translates to more frequent and higher-quality attention to individual youth. For example, in the past decade Big Brothers Big Sisters groups have partnered with hundreds of Boys & Girls Club sites nationwide to improve their mentoring services (Quotah, 2009). At the Twin Cities partnership site, youth and their mentors are able to meet at the Clubs, facilitating reliable contact, and engage in activities there together, allowing mentors to interact with the youths’ peers and program leaders and thus potentially strengthening the youths’ social networks (Forum for Youth Investment, 2006). To be effective, such partnerships likely must be well designed with details regarding shared resources, responsibilities, and practice clearly communicated, perhaps through joint training and regular meetings, or even outlined in a formal contract (Quotah, 2009). Specifically, it may be key for both organizations to be clear about sources of mutual funding and how it will be allocated, specific goals for mentoring, how formal matches will be arranged, and expectations for volunteer mentors’ roles in the after-school program (e.g., when they will meet with mentees and their level of involvement in regular activities). Such steps may encourage a synergistic and truly collaborative relationship between organizations, rather than a mere cohabiting of physical space (Forum for Youth Investment, 2006; Quotah, 2009). Readers interested in learning more about how to develop a formal mentoring program within the after-school setting should review the guidelines for mentoring program development presented by Weinberger (2005) in the first edition of the Handbook.

Conclusion Our review highlights after-school programs as especially promising contexts for youth mentoring. The presence of multiple staff, widespread opportunity for meaningful adult-youth interaction, and the breadth of life concerns addressed present rich opportunities for enhanced mentoring. However,

After-School Programs  231 recognizing the potential of mentoring within such programs is less than half the battle. To realize the benefits of after-school programs, practitioners must commit to supporting high-quality mentoring— in its diverse forms—within their program settings. It also requires researchers to make concerted efforts to tackle persistent gaps in our knowledge of mentoring and after-school programs to best inform practice and policy. We look forward to continued progress on both fronts.

References Arbreton, A., Bradshaw, M., Sheldon, J., & Pepper, S. (2009). Making every day count: Boys & Girls Clubs’ role in promoting positive outcomes for teens. Philadelphia: Public/Private Ventures. Arbreton, A., Sheldon, J., & Herrera, C. (2005). Beyond safe havens: A synthesis of 20 years of research on the Boys & Girls Clubs. Philadelphia: Public/Private Ventures. Deschenes, S. N., Arbreton, A., Little, P. M., Herrera, C., Grossman, J., Weiss, H. B., with Diana Lee. (2010). Engaging older youth: Program and city-level strategies to support sustained participation in out-ofschool time. Philadelphia: Public/Private Ventures. Deutsch, N. L. (2008). Pride in the projects: Teens building identities in urban contexts. New York: New York University Press. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Karcher, M. J. (2005). Youth mentoring: Theory, research, and practice. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 2–11). Thousand Oaks, CA: Sage. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Dworkin, J., & Larson, R. (2006). Adolescents’ negative experiences in organized youth activities. Journal of Youth Development [online], 1(3). Retrieved from http://www.nae4ha.org/directory/jyd/ Eccles, J. S., & Gootman, J. A. (Eds.). (2002). Community programs to promote youth development. Washington, DC: National Academies Press. Forum for Youth Investment. (2006). Out-of-school time policy commentary #11: People, places, and possibilities: Integrating mentoring with after-school. Washington, DC: The Forum for Youth Investment, Impact Strategies, Inc. Grossman, J., & Bulle, M. J. (2006). Review of what youth programs do to increase the connectedness of youth and adults. Journal of Adolescent Health, 39, 788–799.

Grossman, J., Campbell, M., & Raley, B. (2007). Quality time after school: What instructors can do to enhance learning. Philadelphia: Public/Private Ventures. Grossman, J., Price, M. L., Fallerath, V., Jucovy, L. Z., Kotloff, L. J., Raley, R., & Walker, K. E. (2002). Multiple choices after school: Findings from the Extended-Service Schools Initiative. Philadelphia: Public/Private Ventures. Higginbotham, B. J., Harris, V. W., Lee, T. R., & Marshall, J. P. (2006). Youth and families with promise: A multi-component youth development program. Journal of Youth Development, 1(3), 1–9. Hirsch, B. J. (2005). A place to call home: After-school programs for urban youth. Washington, DC: American Psychological Association. Hirsch, B. J., Deutsch, N. L., & DuBois, D. L. (2011). After-school centers and youth development: Case studies of success and failure. New York: Cambridge University Press. Hirsch, B. J., & Wong, V. (2005). After-school programs. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 364–375). Thousand Oaks, CA: Sage. Jarrett, R. L., Sullivan, P. J., & Watkins, N. D. (2005). Developing social capital through participation in organized youth programs: Qualitative insights from three programs. Journal of Community Psychology, 33, 41–55. Kafai, Y. B., Desai, S., Peppler, K. A., Chiu, G. M., & Moya, J. (2008). Mentoring partnerships in a community technology centre: A constructionist approach for fostering equitable service learning. Mentoring & Tutoring: Partnership in Learning, 16, 191–205. Kafai, Y. B., Desai, S., Peppler, K. A., Chiu, G., & Moya, J. (2009). The multiple roles of mentors. In Y. B. Kafai, K. A. Peppler, & R. N. Chapman (Eds.), The Computer Clubhouse: Constructionism and creativity in youth communities (pp. 90–99). New York: Teachers College Press. Kirshner, B. (2008). Guided participation in three youth activism organizations: Facilitation, apprenticeship, and joint work. Journal of the Learning Sciences, 17, 60–101. Kuperminc, G. P., Emshoff, J. G., Reiner, M. M., Secrest, L. A., Niolon, P. H., & Foster, J. D. (2005). Integration of mentoring with other programs and services. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 314–333). Thousand Oaks, CA: Sage. Mahoney, J. L., Larson, R. W., Eccles, J. S., & Lord, H. (2005). Organized activities as developmental contexts for children and adolescents. In J. L. Mahoney, R. W. Larson, & J. S. Eccles (Eds.), Organized activites as contexts of development: Extracurricular activities, after-school and community programs (pp. 3–22). Mahwah, NJ: Erlbaum. Quotah, E. (2009, October 15). Two national youth charities strike partnerships to share resources. Chronicle

232  PROGRAMS AND CONTEXTS of Philanthropy. Retrieved April 25, 2010, from http://philanthropy.com/article/Two-National-YouthCharities/57780/ Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E. (2004). The critical ingredient: Caring youth-staff relationships in after-school settings. In G. G. Noam (Ed.), After-school worlds: Creating a new social space for development and learning (pp. 145–161). San Francisco: Jossey-Bass. Rhodes, J. E., & Lowe, S. R. (2009). Mentoring in adolescence. In R. M. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology (3rd ed., Vol. 2, pp. 152–190). Hoboken, NJ: Wiley.

Serido, J., Borden, L. M., & Perkins, D. F. (2011). Moving beyond youth voice. Youth & Society, 43, 44–63. Walker, K. E., & Arbreton, A. (2004). After-school pursuits: An examination of outcomes in the San Francisco Beacon Initiative. Philadelphia: Public/ Private Ventures. Weinberger, S. G. (2005). Developing a mentoring program. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 220–233). Thousand Oaks, CA: Sage. Zimmerman, M. A., Bingenheimer, J. B., & Behrendt, D. E. (2005). Natural mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 143–157). Thousand Oaks, CA: Sage.

16  CROSS-AGE PEER MENTORING Michael J. Karcher

Introduction Since the previous chapter on this topic was published in 2005, much has changed on the landscape of peer mentoring. The 2008 economic downturn, resulting in withdrawal of public and philanthropic investments in youth development programs, was followed by some surprising studies on both youth mentoring in schools and peer mentoring in particular. Given this state of affairs, one could imagine this chapter would begin on a dour note—fewer peer mentoring programs are active now than a half a dozen years ago, support for peer programs has declined, and the residual effects of the economic downturn on public coffers and private foundations has made the resources necessary for operating a fully functional peer mentoring program few and far between. These are all true. However, during this same period we have learned more about how to run effective peer mentoring programs, and many agencies and organizations learned from the hard lessons in recent years and have fortified their programs in many ways. We have also learned from research about the positive effects of peer mentors for the mentors, in terms of gains in academic connectedness, self-esteem, and cultural competencies, as well as in core qualities necessary for successful citizenship such as responsibility, leadership, and hopefulness. In short, having found the glass half empty (from recent studies that I describe in this chapter), and seeing the diminishment of available resources further deplete the glass, there have been reasons for concern; conversely, we have seen a cup half full, as well. The fact that we can have more faith in the benefits of these programs when run well should lead us to increase our efforts to improve peer mentoring programs.

As a starting place, I put forth a definition of cross-age peer mentoring, the importance and value of which should become apparent to readers in later sections, even if readers question the necessity of some of the overly prescriptive details of this definition. In cross-age peer mentoring, a middle- or highschool-aged mentor (a youth at least 2 years older than the child being mentored) and mentee meet regularly, usually weekly, for a sustained, consistent period of time (minimally 10 times; ideally 20 or more) to engage in conversations, play, or curricula/ structured activities (ones that do not directly or solely teach information or skills in which the mentee has been found lacking) that help to forge a close relationship in which the mentee experiences empathy, praise, and attention from the mentor. Furthermore, staff or other program stakeholders should view the developing relationship (and not the attainment of specific skills or knowledge gains) as the primary mechanism of change. This level of detail may seem arbitrary, and certainly is subject to critique and modification, but it is necessary to differentiate cross-age peer mentoring from the host of other peer programs available to youth in schools. Origins of Peer Mentoring One of the important points to keep in mind about peer mentoring is that it is not really new. Although it usually takes place in schools, all the evidence I have found suggests the practice of peer mentoring well predates the formal introduction of adult mentors in schools, what we call “schoolbased mentoring” in Chapter 14. Variants of peer mentoring appear to have been in place in schools across the United States as early as the 1960s even though most reports about mentoring in schools date 233

234  PROGRAMS AND CONTEXTS the emergence of school-based mentoring as starting in the late 1990s following a swell in the number of adult volunteers wanting to mentor. Colin Powell’s America’s Promise campaign, propelled by the first President Bush’s 1990 “Thousand Points of Light” initiative, relied on the Big Brothers Big Sisters (BBBS) program evaluation led by Grossman and Tierney (1998) as evidence of the power of volunteerism in general and of mentoring in particular (Benson, 1999).1 School-based mentoring began, at least for the BBBS organization, after Grossman and Tierney’s (1998) evaluation of the BBBS communitybased mentoring program resulted in more mentors volunteering and the concomitant need for more contexts in which to place them. In a 2007 report, Carla Herrera and colleagues suggested that BBBS started to move into schools in earnest only 10 years prior to the evaluation of school-based mentoring detailed in that report. Cavell (2012) and Herrera, Grossman, Kauh, Feldman, McMaken, and Jucovy (2007) reported that, between 1999 and 2006, the number of schoolbased matches increased fivefold from 27,000 to 126,000; during this same period, teen mentors were first trained and matched in the BBBS organization. In 2000, fewer than a third of mentors were under the age of 21, but by 2006 nearly half of the school-based mentors were teenagers. Consistent with the emergence of teen mentors in the BBBS program, it seems that most people think adult school-based mentors were on the scene before peer mentors were, yet it appears peer mentoring preceded adult-youth mentoring in schools by about 20 years. In Mentoring a Movement, Susan Weinberger (2005) reported she and others in the Norwalk, Connecticut, school district implemented the first school-based mentoring program in 1983. Bill Milliken didn’t even found Communities in Schools until 1977. So school-based mentoring programs appear to have emerged at the earliest in the 1980s, while peer mentoring programs were in place by the 1960s (see Hebeisen, 1973; Varenhorst, 1983). It may be that the concept of “peer mentoring” was not commonly known until the term mentoring had become a common household word. This occurred in the late 1990s (also as a result of attention directed to Grossman and Tierney’s evaluation of BBBS). Some observers would say that the BBBS

1

program has used the term mentoring for 100 years. This may be true, but they use the term Bigs, not mentors. Milliken (2007) reflected, “I’m not even sure the term was even used back in 1960, but if you had to call us anything (other than crazy), I suppose we were mentors” (p. 14). In the 1970s and 1980s, there were ample reports of mentoring programs calling mentors “Buddies” (O’Donnell, Lydgate, & Fo, 1979) and “Companions” (Goodman, 1972). Similarly, cross-age peer mentoring has suffered from inconsistent naming over the past 50 years. The earliest reports of peer mentoring programs refer to the mentors as “Pals” or “Friends” and were more inclined to call mentoring “counseling” or “helping” (see Heneisen, 1973; Varenhorst, 1983). Indeed, there are important differences between cross-age peer mentoring and the various other peer programs, such as peer helping, peer counseling, peer tutoring, or peer support (see Karcher, 2007). Most peer helping, tutoring, counseling, and support programs (both in high schools and in colleges) have youth helping same-age peers. This defies the definition of a mentor. Peer mentors must, by definition, be “older and wiser” (Rhodes, 1994). Therefore, programs in which youth serve, assist, or support same-age peers should not be considered mentoring. I consider cross-age peer mentoring to be occurring only when there are two or more years’ difference between the mentor and the mentee (Karcher, 2007). Although the terms high school mentor, teen mentor, or peer mentor all convey the age of the mentor, these terms alone do not confirm that the mentor is indeed older (and presumably wiser and more mature) than the mentee. Because the terms peer mentor, teen mentor, and high school mentor are sometimes used by others to refer to interactions between same-age peers (such as in college or high school peer mentoring programs), for our purpose I reconfirm the definition of the mentor as being someone significantly older than the mentee by adding the adjective “cross-age” to peer mentoring that occurs between teen mentors and their younger mentees. Even using just this definition, we can find several cross-age peer mentoring programs as early as the 1960s. Hamburg and Varhenhorst (1972) described the Palo Alto School District

Grossman, currently head of the Department of Labor’s evaluation efforts in the Obama administration, was also a lead investigator of the Big Brothers Big Sisters High School Bigs peer mentoring program (Herrera, Kauh, Cooney, Grossman, & McMaken, 2008) described in this chapter. It is ironic that one of her evaluation efforts can be credited with helping to introduce mentors into schools, while another report, some 10 years later, is sometimes blamed for the diminution of support for school-based peer mentoring programs (at least in the BBBS network).

Cross-Age Peer Mentoring   235 Peer Counseling Program. Peer counseling, depending on how it was structured, was sometimes much like cross-age peer mentoring today. The use of students as providers of psychosocial support to other youth, and the dual benefits of such nonprofessional helping programs, were thoughtfully described 50 years ago by Reinherz (1964). Hebeisen initiated the first study of peer mentoring in 1970 (funded by the National Institutes of Health), suggesting their PEER program had been around in the 1960s. The PEER program was done in a group context and was modeled after a parent effectiveness training program. The book Peer Program for Youth, which is based on the PEER program, is really a curriculum for older adolescents to use when working with younger children in a supportive, befriending capacity. An accompanying book, Extend: Youth Reaching Youth, emphasized the importance of training youth to work with youth (Fletcher, Norem-Hebeisen, Johnson, & Underwager, 1974). This program and research study emerged from the Youth Research Center, which later became the Search Institute. So there is considerable evidence that peer mentoring (in the form of peer “helping” or “counseling”) was in play in the late 1960s. In the previous edition of the Handbook of Youth Mentoring (and more recently elsewhere; Karcher, 2007), I went to considerable lengths to explain how mentoring differs from peer tutoring, counseling, and helping. In the past 30 years, there has been considerable differentiation of peer mentoring, counseling, tutoring, and helping. Each of the latter three have become more task and problem focused during this time (see Goodlad, 1998; Topping, 1996), whereas peer mentoring, at least as implemented and practiced in the past decade by the BBBS program, has been quite the opposite. It is now more relationship focused and also has been, until recently, much less structured. Cross-age peer mentoring (at least in BBBS) had become, up until Herrera et al.’s 2008 report, a group-based program sometimes lacking structure, focus, or goals. Many programs employ match agreements and goal setting at the inception of the match, but programs often fail to keep these goals a priority for matches. Without a clear structure, many such programs reflect little of the original wisdom of the well-organized peer programs (described above) that preceded them. Although peer mentoring has been in existence for 50 years, its prevalence has increased dramatically in the past 15 years and multiple peer mentoring models have emerged. This is illustrated best and most concretely by the number of youth served through peer mentoring in the BBBS programs in

the United States and Canada. In the United States, Big Brothers Big Sisters of America (BBBSA) started using teen mentors in 2001, but within 5 years, nearly 45% of youth served in BBBSA were mentored by teenage Bigs (Herrera et al., 2008). In Canada, more than 60% of Big Brothers Big Sisters of Canada (BBBSC) agencies offer a teen mentoring program, and of these more than 90% are less than 10 years old. In both countries, the large majority of matches are between one teenage mentor and one child. Group or team approaches exist, and are more popular in Canada (Cavell, 2012), but they occur much less frequently in both countries. Also of relevance for this chapter’s focus is that in both the United States and Canada, the use of a curriculum to structure teen mentoring interactions in BBBS has been rare, but there appears to be a growing recognition even in BBBS of the need to provide activities, structure, and guidance (Cavell, 2012; Karcher, Hansen, & Herrera, 2010). A growing body of research, virtually all appearing since the first edition of the Handbook, points to the problems that result from unstructured, unfocused, group-based peer mentoring. It is noteworthy that the first programs, noted above, were structured using a curriculum or planned activities. This practice, I suggest later, may be an essential one for safe and effective cross-age peer mentoring programs. Those early pioneers recognized that such structural components helped to minimize what we now call “deviancy training effects,” when peers reinforce one another’s unconventional, antiauthority attitudes and behaviors, as can happen in unstructured peer-group contexts, both among the mentees and among the mentors. This term was not named as such until the 1990s, but the effects of these processes were evident in some of the earliest mentoring program studies (see McCord, 2003). To expand on these points, this chapter begins with an overview of the theories and concepts that I think are key to establishing and maintaining safe, effective peer mentoring programs, at least based on extant literature. The research supporting these practices and program structures are described in the second section. In addition, the benefits to mentors (the older teens) as well as the unique benefits of peer mentoring to support youth development of mentees in the areas of cultural identity, healthy behaviors, and social connections also are described. Finally, in the section on practices, I go further into detail about the practices that the research literature and unpublished reports have reported to be useful in structuring programs. I apply a list of empirically based best practices (originating from DuBois, Holloway, Valentine, & Cooper’s [2002] meta-analysis) to several

236  PROGRAMS AND CONTEXTS existing peer mentoring programs to detail the landscape of peer mentoring as it is practiced currently. Descriptions of the specific program structures that may be necessary to operate a peer mentoring program effectively—like matching procedures, youth development curriculum, and closure practices—also are highlighted in that last section.

Theory Helper Therapy Principle One of the selling points for cross-age peer mentoring programs since their first reported use has been that the intervention serves mentors and mentees alike, regardless of whether it was framed in terms of the program fostering positive youth development or preventing risk-taking behaviors. Much has been written recently about the effects of involving teens in roles of responsibility and the effect of doing so on their attitudes toward adults, selfperceptions, and engagement in school (Wong, Zimmerman, & Parker, 2011). Less in vogue right now among program proponents and researchers, but perhaps more important in the eyes of those ultimately responsible for deciding whether a peer program will be supported in a school (e.g., principals), are the dual benefits of peer mentoring simultaneously helping older and younger youth at risk at the same time. In fact, one of the earliest references to peer support programs was in a special issue of the journal Social Work that focused on the indigenous helpers and the helper therapy principle (Perlmutter & Duram, 1965). The helper therapy principle, described by Riessman (1965), suggests that there are identifiable (and thus testable) processes that may lead those who provide support services to benefit as much as those who receive them. Some of these mechanisms of change in attitudes and behavior are the consequence of being placed in a role in which one is “doing something worthwhile in helping someone in need” (p. 30), engaging in “self-persuasion through persuading others” (p. 31), and experiencing the sense of importance and status associated with the role of helper. In addition, Riessman suggests “some children develop intellectually not by being challenged by someone ahead of them, but by helping somebody behind them, by being put in the tutor-helper role” (p. 29). One additional benefit Riessman proposes is that often the helper “becomes more efficient, better motivated, and reaches a new stage in helping skill” (p. 28) and becomes more open to “the possibility of embarking on a teaching

career” (p. 29). Therefore, selecting mentors who may already be inclined toward careers in the helping professions may facilitate such benefits and career-congruent skill developments. Deviancy Training One of the earliest reports of using teens in a way “similar to Big Brother programs” made two important points (Perlmutter & Durham, 1965). The first point was regarding the benefits to the mentors, noted earlier. The second point emphasized the need for programmatic efforts to do no harm to the mentees involved in the program. Perlmutter and Durham’s program involved a formal mentor application process, including a personal interview; mandatory monthly training; monthly written feedback for each mentor provided by a caseworker; and the closing of all matches at the end of each academic year. Not all of these practices are found in even the most structured peer mentoring programs in operation today. But perhaps the most important programmatic guideline for this pioneering program was to make sure “the teenage volunteer operat[ed] with minimal autonomy with children of low vulnerability whose needs can be met through a nonprofessional helping relationship” (p. 46). “Children with behavior problems were not chosen, since the teenager was not prepared or qualified to handle ‘acting out’ or aggressive behavior” (p. 43). The exclusion of behaviorally at-risk youth is critically important, but it has been, in my experience, the hardest program component to sell to principals. The possibility of iatrogenic or adverse consequences of mentoring have been documented in the literature for almost 50 years, although it is not a point many program staff or researchers have wanted to discuss until lately (see Spencer, 2007), and it is even more critical to consider in peer programs. These iatrogenic effects often result from the presence of deviance training that occurs in the context of group interventions with youth, regardless of whether the helper is professional or nonprofessional (Dodge, Dishion, & Lansford, 2006a). These negative consequences are especially likely to occur “under conditions of poor supervision and lack of structure” (Dodge et al., 2006a, p. 3). “Deviancy training occurs when a peer displays antisocial behavior or talks about it and other peers positively reinforce that behavior by smiling or giving verbal approval and high status to the first peer” (p. 5). Peer program coordinators should consider several moderators of deviancy training effects. First, both younger children and youth who are moderately involved in deviant behavior—neither delinquent nor

Cross-Age Peer Mentoring   237 nondelinquent, but on the fence—are more susceptible. Helper experience matters, such that more experienced helpers (trained, seasoned professionals) could curb these effects, but novice helpers (teen mentors being perhaps the most novice helpers imaginable) have the worst outcomes. Finally, structured programs yield fewer negative outcomes than unstructured programs. Group time with peers and especially unstructured interactions should be minimized; high degrees of structure supervised by adult staff should be in place to support the dyads of teen mentors and younger mentees. Professionals operating peer programs must learn about these processes and what they can do to prevent them by reading academic and applied materials on the topic (e.g., Dodge et al., 2006a, 2006b). Youth Development and Self-Psychology Two things that differentiate peer mentoring from other peer interventions are the role of the relationship and the breadth of the outcomes targeted by programs. In peer mentoring, youth development is targeted broadly and it is believed to be facilitated by the establishment of a close relationship that affords empathy, trust, and mutuality (see Rhodes, 2005). The mentoring relationship—specifically the mentee’s experience of its developmental properties—is what is believed to mediate or leverage the change in the mentee. One theory of human development that may be particularly helpful in understanding this developmental process is Kohut’s self-psychology theory. Although Kohut’s writings (1977) about his psychoanalytic theory are dense, the theory is very simple. Kohut believes all humans seek from others opportunities to receive mirroring, to idealize other individuals, and to experience a sense of twinship or oneness with select others. The mirroring is often received from others in relationships in the forms of empathy, praise, and attention. The opportunities for idealization follow from these mirroring experiences and occur in relationships that provide clear, consistent structure, such that the individuals in these relationships know what they can expect from the other person because they feel confident in their knowledge of the other’s interests, values, and goals. This confidence develops when the other acts in a consistent manner. In short, we need others in our lives who provide empathy, praise, and attention in the context of clear, consistent structure. We need this emotional support and this structuring from parents and peers alike. When this support and structure is in place, they provide or represent the two “poles of development” (mirroring experiences

and idealizing experiences) that set the stage for the person to form a connection with others whom they view as similar and sympathetic to themselves. Mentors can provide just this opportunity for “twinship,” sameness, oneness, or “We-ness” (Karcher, 2012a). But, as with parents, this we-ness cannot be achieved in the absence of empathy, praise, and attention or in a relationship in which the partner is inconsistent, unpredictable, and untrustworthy. Therefore, peer mentoring programs should foster both these emotional experiences and a clear, consistent structure. Adolescent Connectedness Theory Program curricula can be used to help foster development, but not always in the ways program staff and mentors expect. Cross-age peer mentoring programs can use a curriculum in part to provide a consistent way to structure the mentoring relationship, so that these developmental processes can occur. One must be sure, however, that the curriculum itself does not supersede the relationship in importance. The curriculum can also be used to focus the match on a broad array of topics relevant to youth development. To target a diffuse array of skills, behaviors, and attitudes critical for successful growth and development, the curriculum could take an ecological approach. An ecological approach considers not just the mentoring dyad but also the other important relationships and contexts in the youth’s life. The adolescent’s ecology includes relationships with peers, parents, friends, and teachers. It involves the contexts of home, school, neighborhood, and cultural or religious places and practices. The adolescent’s world broadens beyond the present to include the future, whereas most children experience themselves only in the present (Karcher, 2012a; Karcher, Holcomb, & Zambrano, 2008). Youth engage in their social ecology through behaviors and feelings, not only through thinking (which is the common method of engagement upon which most prevention curricula rely), such that youth connect through action and feelings to each of these worlds. Programs that use curricular activities to help mentors engage in discussion about these relationships and contexts, or that allow the practice of new, more adaptive and functional behaviors in these relationships, not only help the mentor get to know the youth better, more holistically, but also allow multiple opportunities for growth and change rather than narrowly targeting a few behavioral skills or prosocial attitudes. In peer programs, it may be more appropriate to use models of intervention that target the broader social ecology of adolescent

238  PROGRAMS AND CONTEXTS connectedness as program goals than models targeting specific knowledge or skills that are more appropriately served by peer tutoring, peer counseling, peer education, and peer helping. This approach also may provide a helpful heuristic tool for differentiating between such programs.

Research This section summarizes recent research on crossage peer mentoring. It focuses on both the theoretical concepts presented above and the key practices used to buttress programs that have been the subject of research. Many more programs are described in the Practice section than appear here, because this section reviews only empirical studies of the effectiveness of programs or specific processes. A PsycINFO search was conducted using the term peer mentoring and either youth or children, yielding 25 citations. The majority of articles (88%) were published after 2005. Most were excluded because they focused on peer mentoring for parents/teachers (4 articles), included mentors who were the same age as the mentees or who were over age 18 (4), described peer education (3), or simply argued that it was a good idea (5). Using the term peer mentor yielded only one additional study. The term teen mentor yielded two additional dissertations. Two other studies were identified by contacting staff at BBBSA and BBBSC. A web search of the terms peer mentoring yielded many program descriptions and evaluation reports. Most of these programs, however, were not really cross-age peer mentoring programs or the evaluations had flawed or weak designs. Peer Mentoring Research: To Be or Not to Be Considered Evidence Few of the studies resulting from the abovedescribed search inform readers about peer mentoring. Many researchers inaccurately defined interventions as “mentoring” that appear to be tutoring or peer education or that embed peer activities in programs providing mentoring relationships with adults, and many used research designs that prohibit causal attribution. Undermining the utility of many studies of peer mentoring is a lack of clarity about how peer mentoring differs from problem-focused tutoring, helping, or counseling. Often this results from the absence of information on the nature of the interactions that took place between older and younger youth. The web-based search for peer mentoring described above, for example, yielded several

promising programs called “peer mentoring.” Upon closer inspection, most of these programs placed little emphasis on the mentoring relationship (e.g., in program descriptions or evaluations), thus indicating that they were peer helping, skills-training, tutoring, or educational programs. The Teen Trendsetters program, for example, which involved 3,500 students in Florida in 2010, is a great program, but there is little evidence that it provides mentoring or that changes demonstrated by participating youth in a recent evaluation can be attributed to the program. Bessell and Kloosterman (2011) reported, “Teen Trendsetters™ is a program that engages high school students in tutoring and mentoring while simultaneously helping underperforming elementary school second- and thirdgraders improve their reading skills” (p. 5). Their report provided no evidence that this intervention involved more than teens reading with elementaryaged students, nor did it provide other evidence that this was a mentoring program. Equally problematic, their program evaluation also omitted a comparison group and included considerable missing data, so that little could be inferred from the report about program impacts. Another common problem in “peer mentoring” research reports is confusion between peer education and mentoring. One program evaluation (Johnson, Holt, Bry, & Powell, 2008) included 8 weeks of “selective” structured mentoring by adults (“Achievement Mentoring”) in addition to 16 weekly “universal” group activities led by older peer leaders (Powell, 1993). (Here, “selective” means that only “at-risk” youth received mentoring and it was from adult mentors [who were teachers at the school], but all youth [“universal”] participated in youth-led group activities weekly; see Cavell & Elledge, this volume, Chapter 3, for further explanation of terminology used in the prevention field.) Included in the “manualized universal program,” called Peer Group Connection by Powell (1988), were 97 freshman who met in groups of 12–15 with pairs of upperclassmen student peer leaders for 16 weeks to discuss 16 modules covering peer pressure, goal setting, school connectedness, and other processes related to program outcomes. Twenty program youth deemed at greater risk of dropping out were also assigned an adult mentor. Findings suggested the program had larger effects on the higher-risk youth on two of four outcomes. But almost half of these “higher-risk” youth had an adult mentor, whereas presumably none of the “lower-risk” program youth or comparison group youth (a cohort of 60 students from a gym class) had an adult mentor. The authors reported the “integrated program”

Cross-Age Peer Mentoring   239 had a larger effect (d = 1.11) on the more at-risk program youth; however, it was integrated (i.e., including both a mentor and older-peer-led group prevention activities) only for half of the highest-risk youth. That is, the low-risk youth got a different program (no mentors), leaving it unclear whether the adult mentoring was the important active ingredient for the highrisk youth, whether the peer activities had any effect on their own, or whether changes observed among high-risk youth reflected the processes of a regression toward the mean among those most at risk. Fortunately, a stronger test of this peer program on longitudinal outcomes (Johnson, Simon & Mun, in press) is described in the next section. A dissertation by Mathews (2007) created a similar confound in two ways. First, the peer mentoring was really peer tutoring in social skills over 16 sessions by an older youth. More specifically, “Resilient Peer Training” was provided in which “mentors” were trained that “a peer mentor is a teaching friend” (p. 108). This peer program was designed by Fantuzzo, a leader in the field of peer tutoring for two decades. Such programs—in which program or school staff identify youths’ deficits and the program targets these deficits for remediation by older peers who are instructed to teach the skills in which the mentees are deficient—are more appropriately labeled peer helping, education, or tutoring. Further, by coupling this “peer mentoring” with two other interventions (i.e., social skills training and token economy reinforcement), no unique effect of this peer support program could be identified. Another evaluation of a promising multicomponent intervention program appears to report impacts from peer mentoring but provides little evidence of this. Peer mentoring was only one of the programs in the Chicago Life Directions: Peers Inspiring Peers program (Life Directions, 2006). The sample included 58 students who participated in the Peer Mentor program and 481 students who were part of the other program components, which included “Peer Motivation” and “Neighborhood Enrichment.” Teachers and students completed surveys regarding program impacts on student attendance, school participation, and behavior, but only descriptive statistics were provided. The lack of inferential statistics, an absent comparison group, aggregation of data across programs, and small sample size make it difficult to link any impacts to peer mentoring, per se. In one multicomponent study of a peer mentoring program, 10th to 12th graders mentored 137 9th graders who failed one or more classes in their first term in high school. Although these youth received after-school credit-recovery activities in addition to mentoring, and the comparison group for the study

was the prior year’s freshman class, reports of gradewide declines in failure rates were attributed to the presence of the peer mentoring program (Chew & Wallace, 2008). The percentage of 9th-grade students who failed major subjects (e.g., science, social studies, English, math) did decline the year the peer mentoring program was implemented (from 15.6% to 11.7%), but serious methodologic shortcomings limit the viability of these attributions. For example, it seems unrealistic to infer that changes across the entire class (N = 632) were due to the program, when only 137 9th-grade students participated in it. That would require the peer mentoring program to have had a massive contagion effect on the rest of the students in the school. There also was no evidence that the two cohorts were sufficiently comparable to allow a comparison. But perhaps most important, the school-based program was implemented solely to assist students who were at risk of academic failure with little explicit reference to the role that the mentoring relationship was expected to play in mediating these programmatic effects. The point of highlighting these studies that do not speak to the effects of peer mentoring is simply to suggest that perhaps most of the research and evaluation reports of peer mentoring programs that one finds either are not studies of peer mentoring or their designs are such that no program effects can be linked to the peer mentoring uniquely. This is troublesome. Evaluations of Cross-Age Peer Mentoring Programs To remove other types of peer programs from consideration, I consider in this section only programs that meet the strict definition of cross-age peer mentoring provided earlier: Cross-age peer mentoring entails a school-aged youth at least 2 years older than the child being mentored (e.g., “older and wiser”) meeting regularly, usually weekly, for a sustained, consistent period of time lasting at least 10 times (ideally 20 or more). They engage in conversations, play, or curricular activities (that do not directly or solely teach skills in which the mentee has been found to be deficient) that serve to forge a close relationship in which the mentee experiences empathy, praise, and attention. Furthermore, program staff (or researchers) should clearly articulate that it is this relationship, not specific skills or knowledge gains, that is believed to foster the youth’s development. Note that programs in which college students mentor school-aged children or teens are excluded from this definition. The overlap between programs such as Lunch Buddy

240  PROGRAMS AND CONTEXTS (Cavell & Henrie, 2010), peer mentoring in college (e.g., Larose et al., 2011; Smith-Jentsch, Scielzo, Yarbrough, & Rosopa, 2008), and peer mentoring in schools deserves attention, but it will not receive it here. This chapter focuses primarily on high-schoolaged mentors of youth. High School Bigs Program. Herrera’s evaluation of the BBBS High School Bigs program in 2008 provided perhaps the most rigorous test of peer mentoring to date—perhaps too rigorous given the degree of program development that preceded it and the absence of prior studies of programmatic efficacy, moderators of effects, and mediating mechanisms of change (Cavell, personal communication). Up until the time of this evaluation, the program consisted of relatively unstructured weekly, school-based meetings between a teenage mentor and an elementary- or middle-school-aged mentee. I underscore “until the time of this evaluation” because after this evaluation, the program was thoroughly redesigned by Keoki Hansen and members of the High School Bigs Demonstration Program at BBBS, and this new model was being field tested between 2009 and 2011 (see Karcher, Hansen, & Herrera, 2010). The original program was dyadic in nature, but it often occurred in a larger group context and it was generally unstructured. In Herrera et al.’s (2008) study, half (49%) of the mentors were high school juniors, another quarter were seniors, and the rest were sophomores or freshmen. Two-fifths received course credit for mentoring. Bigs and Littles usually met in groups during or after school, but sometimes they met individually during the day or over lunch. The focus of the interactions was on relationship development, and any use of curricular activities was not consistent across sites. When present, curricula were used extemporaneously (e.g., as a result of an individual coordinator’s or mentor’s planning at a given site). The mentors took a more developmental, relationship-oriented (at first) approach with their mentees/Littles. Herrera et al. reported that “relationships with high school Bigs were similar in length and quality of those with adults” (p. iii). But this may be where the similarities ended. The difference between effects for adult mentors and teen mentors was stark. For all mentors combined, Herrera et al. (2007) found impacts on 9 of 23 academic outcomes (and none on the 8 nonacademic outcomes), but when program effects were examined separately for adult and teen mentors, adult mentors’ mentees demonstrated statistically significant improvement on 12 of 31 outcomes. Comparing youth who received a teenage mentor to youth on the waitlist yielded only one program

effect. “Littles matched with high school Bigs improved relative to their non-mentored peers in only one measure, teacher-reported social acceptance” (Herrera et al., 2008, p. 2). (Note: These are all at a p-value less than .10; all representing small effects from d = .09 to d = .24). None of the effects of the teenage mentors on these outcomes approached an even “small” effect size (all d < .15). Several program practices were associated with statistically significant outcomes, however, and these are worth noting. Although these reveal the important role that programmatic structure plays in achieving outcomes from peer mentoring, not all are easy to understand. Matches with teen mentors that took place in the context of other matches in one location lasted longer, which is good, but their mentees felt they got less attention than mentees meeting alone with their mentors. Therefore, group meetings may be appealing for mentors (allowing them to meet their social needs) and thereby help to retain mentors across years, but efforts to redirect mentors’ attention back to their mentees may be critical. This may be done through training or through the use of planned activities. Indeed, teen mentors who viewed their training as higher in quality and mentors who received more than 2 hours of training had higher quality relationships. Similarly, more frequent communication with program staff was associated with statistically significant improvements by mentees on five social and academic outcomes (Herrera et al., 2007, p. iv). The mentor’s attitude also seems critically important. Karcher, Davidson, Rhodes, and Herrera (2010) found that BBBS teen mentors who valued youth more highly in general had mentees who were more emotionally engaged in the match. Additionally, the interaction of mentor’s attitude and mentee’s risk level moderated program outcomes. Mentees who were more disconnected (i.e., at greater social, academic, and behavioral risk) but were matched with mentors who valued youth in their community more highly (e.g., saw them as resources) reported significantly higher teacher relationship quality after mentoring than did similarly disconnected control group youth, whereas teachers reported more disruptive behavior among those mentees least at risk who were matched with mentors who held less positive attitudes toward youth in their community (compared to low-risk control-group youth). Therefore, mentors’ attitudes toward youth may serve as one tool for identifying mentors. Karcher, Davidson et al. (2010) reported the items in the attitude toward youth scale so that agencies could use those items for recruitment purposes.

Cross-Age Peer Mentoring   241 Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors (Karcher, 2008, 2012a). CAMP is a year-long, after-school dyadic peer mentoring program that uses developmental theory to structure mentor and mentee interactions. Child developmental theory is used to structure the 2-hour sequence of interactions that takes place during each weekly meeting, and also provides one of the two core frameworks used to organize the curricular activities. The other framework is a theory of adolescent connectedness across the youth’s social ecology, which provides a definition for what it means to promote connectedness and what domains of connectedness are critical to focus on in the program. For example, the first-year curriculum focuses on promoting connectedness to teachers, peers, friends, family, self, reading, and culture. CAMP also includes specific matching and match-closure procedures as well as structured family involvement and teacher involvement to ensure these elements are present. One factor that may distinguish the original High School Bigs program (not the revised model described in the next section) from CAMP, and thus what might account for the differences in their impacts, may be the presence of a curriculum in CAMP to provide a clear, consistent structure around which to organize relationship development. Peer tutoring and education also may use curricula, but peer mentoring is differentiated from them in its focus on relationship development and the careful use of curricular activities for the sake of the relationship, not to effect specific outcomes. The curriculum content, however, is designed to be changed to suit local needs of specific populations of youth; testing CAMP effects using another curriculum would be a way to test the hypothesized role of the curriculum in program impacts. However, another explanation may be that in studies of CAMP, the program may have been implemented with greater fidelity (implemented as intended), and the High School Bigs programs studied by Herrera et al. (2008) may have varied widely in program fidelity or not had clear implementation guidelines at all. One study using peer mentoring to prevent obesity provides a particularly interesting example of the fine line between peer mentoring and peer education and provides another test of the role of a program curriculum in peer mentoring. Based on the definition provided above, peer mentoring should not be overly problem focused, and program planners should not view as the mechanism of change specific gains in information or skills learned through the program, even if the program teaches skills and those skills are used as outcome measures of program impact. In many cases, the

main way to differentiate between peer mentoring and peer education or tutoring is in terms of the hypothesized role that the mentoring relationship plays in changing behavior. Cross-age peer mentoring also has been used to influence eating habits. Smith (2011a) hypothesized that it was the mentoring relationship that would boost mentees’ self-efficacy and, in turn, change behavioral intentions regarding eating and exercise. To specifically test the mediating effect of the peer mentoring relationship, Smith (2011b) randomly assigned half of the youth to a comparison group in which teens taught the same health education curriculum in a group format. Therefore, any differences in outcomes, such as behavioral intention and health (body mass index), should be the effect of the dyadic mentoring relationship in one group but not the other. In Smith’s study, 72 students in 3rd and 4th grades were divided randomly and assigned to the two conditions: (a) curriculum delivered by teens in a group format (n = 27), and (b) curriculum delivered in the context of a dyadic peer mentoring relationship (n = 25). In addition to the small sample size (yielding low statistical power sufficient to detect only mediumsized effects), the duration of the mentoring was only 8 weeks. The mentors received $100 for participating in the 2-day training used in CAMP (Karcher, 2012b), additional training in delivering the curricular materials (which were adapted for dyadic rather than group delivery), and weekly supervision. Using pairedsamples t-tests, Smith measured pre-post changes within each intervention group separately (the effect sizes reported here, d, reflect by how many standard deviations the youth differed after the intervention, compared to their pre-intervention scores). Statistically significant changes were found after 8 weeks for body mass index for mentored youth (d = –.41), but not for curriculum-only youth (d = –.26). Mentored youth also had significant improvements in behavioral intentions (d = .35), but the curriculum-only youth did not (d = .06). Comparing pre-post change between the mentoring and peer education groups yielded a statistically greater gain in intentions to eat healthfully among mentored youth (d = .35) but no difference in body mass index. Smith hypothesized that this change in intentions would result from improved self-efficacy that would occur mostly among mentees. Counter to expectations, however, the mentees gained in nutritional knowledge and attitudes toward eating healthfully, while the curriculum group youth reported greater self-efficacy at posttest. Therefore, although the data did not support the hypothesis about how teen mentoring may work, the probability that the teen mentoring program yielded larger healthy behavior changes was supported.

242  PROGRAMS AND CONTEXTS Karcher (2005b) reported another attempt to assess the relative effects of the CAMP curriculum versus the mentoring relationship. In the context of a randomized impact study of CAMP, a quasiexperimental comparison was made to estimate whether changes in enabling outcomes were associated more strongly with exposure to the curriculum (mentee’s attendance) or exposure to the mentor (mentor’s attendance). These enabling outcomes (change in social skills and self-esteem) were expected to be “caused” by time spent with the mentor and would mediate (or explain) the effects of program participation on distal outcomes (i.e., connectedness). Karcher found significant between-group differences in changes in connectedness to school and to parents (consistent with prior research, see Karcher, Davis, & Powell, 2002) favoring mentees. Among mentees, mentor attendance was, as hypothesized, significantly related to improvements in five of eight measures of social skills, self-esteem, and connectedness. None were associated with mentee attendance. However, no evidence was found that program effects on connectedness to school and to parents were mediated by these enabling outcomes. Future research should test some of the helper therapy principles and other theories. Smith’s study supports the argument that it is the relationship between mentor and mentee that is associated with peer mentoring program success, but how the relationship brings about these outcomes remains unclear. For example, although Cavell’s Lunch Buddy program (Cavell & Henrie, 2010) uses collegeand not high-school-aged mentors with elementaryaged mentees, research on the program may inform our understanding of peer mentoring processes. It may be that in peer mentoring, as in the Lunch Buddy program and as proposed by Riessman (1965), improved peer interaction quality and social status for mentees is improved by a mentor who is publically empathic, praising, and attentive in the presence of the mentee’s peers. This may explain how the relationship effects positive outcomes in group-based peer mentoring programs like CAMP. Other Cross-Age Peer Mentoring Programs. One group-based peer mentoring program that provides considerable structure to support mentors is the Peer Group Connection (Powell, 1993), which was evaluated longitudinally to assess its influence on high school graduation. In a study by Johnson, Simon, and Mun (in press), 16 seniors were trained as peer leaders and assigned in pairs to work with small groups of 12 first-year students. The mentors provided an average of 18 weekly “outreach sessions” that coached relationship, emotional, and academic

skills to small groups of freshmen. Program participants were provided two additional booster sessions their sophomore year. The incoming 9th graders in 2005 were randomly assigned to receive the Peer Group Connection program (n = 94) or become part of the control group (n = 174). Researchers then used logistic regression to predict graduation likelihood 4 years later in 2009. After including a group of 35 baseline characteristics as predictors, including gender, researchers found virtually no differences in the likelihood of graduating between program and control youth. The propensity scores for the program youth (.72) and controls (.70) suggested the randomization was successful in that groups were very similar before participation in the intervention. The main effect analyses of program impact on graduation reached only marginal significance (to be expected given the sample size). Program youth graduated at a rate of 77% and controls at a rate of 68%. Planned moderator analyses were then conducted to estimate program impact across genders and for students at high and low risk for graduation. Risk was determined by a cutpoint of .60 on the propensity score. The school’s graduation rate was typically 70%, so the .60 cutoff put 70% of the sample above this point. Thus the 30% below .60 were those not expected to graduate based on characteristics measured in 2005. The effect was moderated by gender, with a statistically significant effect of the program for boys but not for girls; the effect for boys was moderated by risk level. Boys at high risk for dropping out were twice as likely to graduate if they participated in the program (60%) than if they did not (30% graduation rate). Note, however, that these finding may be spurious given the small sample sizes. The proportion of graduated to not-graduated high-risk males in the program (n = 14, 8:6 = 1.33 odds of graduating) and the control group (n = 28, 8:20 = .4 odds of graduating) yields an impressive odds ratio of 3.325. High-risk boys who participated in the program were more than three times more likely to graduate from high school if they participated in the program. Although this program requires a lot of infrastructure—with the course for the mentors taught weekly as perhaps the most costly—and there is no evidence that a one-on-one relationship was established between mentors and mentees, this appears to be a very promising program. Finally, a study with older peer mentors in the workplace revealed the importance of studying key developmental processes, like those described by Kohut (1977), as mediators of program effects. Westerlund, Granucci, Gamache, and Clark (2006)

Cross-Age Peer Mentoring   243 used workplace mentors to facilitate the skills of teenage cosmetology students in a vocational training program. The students/mentees ranged in age from 16 to 18 years, while the mentors ranged in age from 18 to 61 years. Single-subject multiplebaseline designs were used in which baseline levels were estimated for such skills as setting rollers and combing out hair. Then the mentors provided task and emotionally supportive training interventions. This could be viewed as a tutoring study, of course, in which researchers looked to see how demonstrations, corrective feedback, and descriptive praise helped students learn vocational tasks. But there was an enduring relationship of more than ten meetings, and the authors hypothesized that the relationships that developed, specifically the degree of social support that was experienced, more than the instructional interventions alone, would be what facilitated the mentees’ confidence and skills. The researchers examined the co-occurrence of mentee comfort levels as a function of mentor praise and encouragement, and how skill development was related to experiences of increased social support. Two of the four mentees nominated their mentors as individuals in their “circle of support,” and researchers identified the importance of praise and clear, consistent structure in fostering that social support. This research was innovative in its planned assessment of the role that relational support and relationship building played in the development of vocational skills. Mentors, Moderators, and the Helper Therapy Principle at Work The earlier study by Karcher, Davidson et al. (2010) that found mentors with more positive attitudes toward youth had a bigger effect on those mentees at greater behavioral risk is consistent with another study of mentor attitudes as a moderator of program impact. In a study with a small sample (33 mentors, 27 mentees) that correlated mentor and mentee characteristics, Karcher and Lindwall (2003) found that mentors who reported higher levels of social interest, as measured by Crandall’s (1991) social interest scale, were more likely to choose the more challenging youth to mentor and were more likely to persist into a second year as a mentor than those with lower social interest. Brewer (2009) also found social interest to be higher among peer mentors. A second study reported by Karcher and Lindwall (2003) compared 120 youth, half of whom volunteered as teen mentors. Peer mentors reported more conventional connectedness to school, family,

reading, and their future than did youth who did not volunteer to mentor. Taken together, these findings indicate that peer mentors in general and more successful peer mentors in particular tend to differ from their peers in their connectedness to school, their attitudes toward youth in general, and their social interest. Recruiting those mentors who value youth more and who report greater social interest may increase the chances that strong and enduring relationships form with their mentees. For this reason, Karcher included the social interest scale as the first part of the online training at www.highschoolbigs.org. After mentors complete the survey online they can direct their scores to program staff for use in mentor selection or matching efforts. There also is some quasi-experimental evidence that peer mentors benefit from participating as mentors. Consistent with the helper therapy principle, in a comparative study of 111 rural high school students (Karcher, 2009a), those who volunteered to serve as peer mentors reported greater changes on multiple measures of academic connectedness and self-esteem across one academic year than did youth in their same classes who did not serve as mentors. In this way, serving as a mentor may indeed allow mentors an opportunity to practice helping roles that may serve to inform career choices and strengthen their empathy and relationship skills. It is worth noting, however, that Karcher and Lindwall (2003) found that mentors with the greatest social interest and who chose more challenging mentees reported declines in connectedness to school following the experience (although they persisted as mentors the following year anyway), and some mentors felt so challenged that they reconsidered whether they should continue to pursue their goal of being a teacher (Karcher, 2006). However, it may be that becoming more realistic about the challenges of helping (and perhaps of being a teacher), as well as the experience of struggling to persist through a difficult mentoring relationship and ultimately succeeding, are likely important life lessons that may serve to bolster resilience and steel youth for later interpersonal challenges. Cross-Age Peer Mentoring and Inclusion of Diverse Populations The use of peer mentoring to foster mentees’ conventional connectedness to school has been reported by other researchers as well. Peer mentoring has been used to facilitate connectedness to school and a sense of belonging at school for youth who may otherwise feel marginalized or “different” in some way. Like Carter and Hughes (2005), who

244  PROGRAMS AND CONTEXTS provided within-grade peer support to youth with special needs by pairing them with a regular education student (but which does not qualify as crossage peer mentoring because of age similarity between participants), others have reported using peer mentoring to help nonmainstream youth feel more welcome in their new schools. One study used a very short-term peer mentoring approach to foster peer attachment among immigrant youth (Yeh, Ching, Okubo, & Luthar, 2007). Twenty-three immigrant high school students self-selected to participate in a school-based peer mentoring program. They were each matched with another high school student with whom they met once a week individually, once weekly in small groups, and during lunch once a week. The program was relatively short, lasting only 3 months, but the weekly dosage (number of contacts) was relatively high for school-based mentoring. Yeh et al. (2007) reported significant outcomes on only one of three outcomes: peer attachment. Using a pre-posttest research design, the researchers tested for changes among participating youth in college and career self-efficacy, academic selfefficacy, and peer attachment (trust and need for closeness). Significantly higher peer attachment scores for both the trust scale (d = .35) and the need for closeness scale (d = 1.33) were reported at posttest. No differences were found on college, career, or academic self-efficacy. However, the absence of a comparison group, the self-selected nature of the sample, and the small sample size strongly challenge the assertion that these changes were due to program participation. In a qualitative dissertation by Steiman (2006), examining processes and perceptions among an even smaller group of nine elementary students who participate in peer mentoring, the mentored 3rd-, 4th-, and 5th-grade immigrant students reported greater self-confidence and a stronger sense of attachment to the school community after participation in the peer mentoring program. Newcomer students who had been in the United States for less than a year were matched with middle- and high-school-student mentors. Again, the pairs met three times a week (high dosage) at the elementary school for an 8-week period (low duration). The main purpose of the program was to help alleviate some of the social and emotional stress associated with being in a new country. Structured and unstructured interviews were conducted with teachers, mentors, and newcomer students, and a parent focus group was also conducted at the end of the program. Document analysis of some writing samples of mentors and mentees

were conducted to identify themes, trends, and patterns. Interpretations of the qualitative data collected from mentors, mentees, teachers, and parents in this study support the finding that peer mentoring can assist immigrant children in gaining a stronger connection to and belongingness within the school environment. The use of different types of data (interviews, observations, focus group) increased credibility of the researcher’s interpretations and provided a means of triangulation, but the lack of a comparison group restricts any causal interpretation of data. Several studies of programs that met the proposed definition for cross-age peer mentoring were problematic due to validity threats posed by the research design; therefore, these studies are not described here. O’Hara’s study (2011) had two comparison groups, one of which was an attention placebo (“reading mentor”), but the samples were too small to provide reliable outcome estimates. Like O’Hara, another study (Rosenblum et al., 2005) reported declines in conduct or behavioral problems after peer mentoring, but this study had a treatment group that was compromised by self-selection. Only half of the mentees attended the program, which meant the control group no longer served as the counterfactual comparison for the treatment group (e.g., most likely the more behaviorally at-risk in the treatment group self-selected out). Attendance was used instead as the measure of dosage, increasing the self-selection bias; only 70% of participants were available at the 1-year posttest. An otherwise good mixed-methods dissertation on peer mentoring also was hampered by attrition as well as a poor response rate, imbalanced membership, and survey reporting across genders that undermined the utility of the study’s quasi-experimental design (Finckler, 2003). Finally, another welldone study (Eddy, 2011) included a detailed literature review but unfortunately selected an effect size two or three times what should be expected based on meta-analyses (e.g., DuBois et al., 2002). This is surprising given that the intervention was only 10 weeks long, suggesting the need to attempt to detect even smaller effects. Finally, this study used 5th graders as reading mentors (like O’Hara’s [2011] control group) to 1st and 2nd graders. It gave little indication that the mentors were trained or encouraged to develop a personal relationship with their mentees, leading one to question whether this was a mentoring or tutoring intervention. All of these studies have merits and strengths but reveal important barriers to effectively studying cross-age peer mentoring.

Cross-Age Peer Mentoring   245 Conclusion and Future Research Recommendations It is probably fair to say the research on peer mentoring is fairly weak at this point with respect to breadth and depth, and much future work is needed. In terms of breadth, there are very few studies of cross-age peer mentoring. Few programs have more than one or two studies of effectiveness, and randomized controlled comparisons are rare. Several studies of peer mentoring have been of little utility because the research design incorporated other interventions into the treatment package or the studies provided insufficient detail to differentiate peer mentoring from peer helping, tutoring, counseling, and education. In terms of depth, despite there being multiple rich program descriptions (e.g., Crooks, Chiodo, Thomas, & Hughes, 2010; Willis, Bland, Manka, & Craft, 2012), few studies of program moderators or active ingredients have been done to empirically link program practices with outcomes. What evidence I did find, such as of the importance of curricula and other support practices, I piecemealed together inductively by comparing studies rather than basing it on formal comparisons of such practices explicitly or within a given program (e.g., dismantling design) more deductively or affirmatively. Therefore, we clearly need more research on different types of mentoring program designs, but the field also needs more detailed and specific tests of what program practices yield the largest outcomes. Theory-based tests would be useful as well, because those few studies that have tried to test theoretical mediators of program effects have not been successful to date.

Practice In the first edition of the Handbook of Youth Mentoring, Karcher (2005a) described several recommendations for practice, which included important implications for structuring programs, screening mentors, supervising matches, and training mentors. Fortunately, since the Handbook’s first edition, various programs have developed novel approaches that respond to some of these recommendations. Several of these are described in this section. In particular, the BBBSA High School Bigs has undergone a thorough overhaul and now includes a host of best practices, CAMP curriculum and training have been manualized, and the BBBSC High School Teen Mentoring program materials have been developed and disseminated. This section describes each of these resources, as well as other peer mentoring

training and support materials. Like the excellent summary of key practices written by Garringer and MacRae (2008), I focus specifically on program practices that have been identified in the peer mentoring literature, and when possible I have used the program descriptions (e.g., manuals) instead of research studies alone. To frame this presentation of peer mentoring practices, this section uses DuBois et al.’s (2002) empirically and theoretically identified best practices in youth mentoring. Program Best Practices: An Overview of Practices Used in Several Peer Programs DuBois et al. (2002) conducted a meta-analysis of 55 studies of youth mentoring program impact. This review included no peer mentoring programs, but a more recent review of youth mentoring program impacts did (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). That review found no statistically significant difference in the program impacts for adult versus peer/teenage mentors, nor was any evidence reported that mentor type moderated the association between program practices and program outcomes. Therefore, the best practices identified by DuBois et al. (2002) likely apply to peer mentoring as well. DuBois et al. (2002) measured effects associated with both theory-based and empirically based best practices. Theory-based practices are those practices recommended in mentoring literature, whereas empirically based practices are those that are not necessarily discussed in mentoring literature but that were found to be associated with greater effect sizes in the meta-analysis. DuBois et al. (2002) identified three theory-based practices—parent involvement, use of structured activities, and ongoing training— that explained over 25 percent of the between-program variability in impacts. Other theory-based practices include mentor screening and recruitment, matching based on specific criteria, training from pre-match throughout the life of the match, and monitoring of program implementation. Important empirically based best practices identified in the meta-analysis include using a non-school setting and selecting mentors from the helping professions. Because there were relatively few peer mentoring studies to consider, and because significant strides have been made in the past 50 years in other peer helping methods, I briefly describe noteworthy peer helping programs—namely, Peer Power (Tindall, 2009a, 2009b, 2009c, 2009d) and Peer Buddy Program (Hughes & Carter, 2008)—and stand-alone materials for training and support for peer programs in general (e.g., Expanding the Spirit of Mentoring; Cox,

246  PROGRAMS AND CONTEXTS 2006). Then I focus specifically on peer mentoring programs, as defined earlier: BBBSA’s revised High School Bigs program, BBBSC’s High School Teen Mentoring program (Government of Alberta, 2010a), the Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors (Karcher, 2008, 2012a), Mentoring Works (Avani, 1998), and the California Friday Night Live Mentoring Program (Scott-Nakai, 2008). In the sections that follow, I review training materials, comprehensive peer support programs, comprehensive cross-age peer mentoring programs, and specific best practices. Unfortunately, to my knowledge none of these training activities or curricula has been studied in a way that would allow any estimation of its efficacy relative to another’s.

and Varenhorst (1983) is a pioneer in the peer helping field. This 159-page resource includes a host of skills-based training activities that are similar to the training materials for the CAMP intervention described earlier (Karcher, 2012c) and that focus on communication and assertiveness skills. It also provides activities to help youth understand their own assets in order to help them better cultivate assets among the peers they serve in the program. Training in peer mediation also is provided. For those whose peer mentoring program takes place in a school that utilizes the Search Institute’s developmental assets model, this set of training materials will complement those curricular goals well. Additional assetbased mentor training and support materials also are available (Probst, 2006).

Mentor Training Expanding the Spirit of Mentoring (Cox, 2006). Cox’s 56-page training material includes activities that can be modified to fit various program needs and, although targeted to adolescents, can be used with various age groups. Training activities focus on developing team-building, communication, and strength-building skills among mentors. Peer Connection Program (Powell, 1988). Available through the Princeton Center for Leadership Training, this set of leadership training materials has been used by multiple researchers and doctoral students to train mentors in the peer mentoring programs they studied. The focus of the training is on leadership, group skills, and problem solving. The Peer Group Connection is a peer leadership training program intended to be used by educators to prepare upperclassmen to work with freshmen and sophomores in a mentoring or other supportive role. In their recent impact evaluation, Johnson et al. (in press) explained that, using the materials, “[h]igh school juniors and/or seniors become trained peer leaders who meet once a week with [small groups of] freshmen in outreach sessions designed to develop skills, promote a respectful school culture, nourish meaningful connections, and strengthen relationships among students across grades. Booster sessions are provided during students’ sophomore year to reinforce learning from the previous year” (p. 5). The program provides considerable mentor training by coupling the mentoring program with an academic service-learning class that provides opportunities to prepare for the mentoring meetings and to debrief about meetings afterward. An Asset Builder’s Guide to Training Peer Helpers (Varenhorst, 2003). This resource is targeted at training teenage peer helpers rather than mentors,

Meaningful Mentoring (Bowman & Bowman, 2005). For program coordinators who meet weekly with their mentors and need additional training activities that are unique from those described above, as well as for training mentors who will be working with mentees on their own without a curriculum (which I do not advise), this set of activities may be helpful. Using the metaphor of the mentor as a copilot, this 155-page book includes activities focusing on training mentors in the skills they need to communicate, listen, and address problems as they arise. Peer Helping Programs With Useful Program Design and Training Materials Peer Buddy Program (Hughes & Carter, 2008). The Peer Buddy Program was designed to promote supportive interactions between students with disabilities and their general education peers. It is a structured program, and peers commit to volunteering for a full academic year. Hughes and Carter suggest their buddies “may describe themselves as a friend, tutor, advocate, or instructor in relation to their peers with disabilities” (p. 90). These relationships are between same-age peers, and thus may provide peer support but don’t qualify as “older and wiser” peer mentors. The program guide includes a host of useful procedures for programs and training teens, and supplemental training materials are available (Carter, Cushing, & Kennedy, 2009). Peer Programs: An In-depth Look at Peer Programs—Planning, Implementation and Admini­ stration (Tindall & Black, 2009) and the Peer Power Set (Tindall, 2009a, 2009b, 2009c, 2009d). The Peer Power program was developed for middle, high school, and higher education students and may be

Cross-Age Peer Mentoring   247 most useful to teachers coordinating service-learning courses. Generally, youth should be at least 12 years of age to participate in the training. The training materials were developed to assist peer helpers in successfully communicating with their peers and mediating conflict (like Varenhort’s, above). Although Peer Power is not a cross-age peer mentoring program, the two volumes of training and reflection activities could be useful for training peer mentors as well. One has to be impressed with the materials the author has assembled. Any professional whose work is to facilitate peer programs should have a set of these materials. Peer Programs is the only book I’ve found that discusses the development of the peer program professional (you, perhaps), the history and future of peer helping efforts, different models (not just activities) for training, and program details like budgeting and staff team development. The Peer Power series, including Strategies for the Professional Leader: Becoming an Effective Peer Helper and Conflict Mediator (Volumes 1 and 2) and Workbook: Applying Peer Helper Skills (Volumes 1 and 2), covers training strategies and includes the accompanying worksheets for youth on topics from substance use, coping, and mental health, to the use of peer helping in various forms (tutoring, peer education, peer helping, and peer mentoring—though only 16 of the 401 pages in Volume 2 of the workbook discuss mentoring). The experience, vision, comprehensiveness, coverage, and field-tested wisdom in these materials is unsurpassed. Youth Helping Youth: A Handbook for Training Peer Facilitators (Myrick & Erney, 2004). This 266 page spiral-bound guide is nearly as comprehensive as the Peer Power set described above, but it is contained in one book. Like Tindall’s work, Youth Helping Youth provides not just training activities but the bigger picture on the history of peer helping, tips for peer helping professionals, evaluation suggestions, and ideas on how to organize and manage a peer helping program. It is a more succinct set of materials covering the basics of teaching listening skills, communication and peer support skills, and both decision-making and problem-solving skills. It has less information on how to set up and manage a program than the Peer Buddy Program but much more information on training peer helpers. It covers a narrower range of training activities than Tindall, but is not redundant with Peer Power materials. It does duplicate some material from another text by these authors called Caring and Sharing. But as with Tindall’s work, these materials have stood the test of time and are useful and insightful.

Three Comprehensive Peer Mentoring Programs In this section, I overview three programs currently in existence that include all or most of the best practices identified by DuBois et al. (2002). Two of these are BBBS programs and the third is CAMP. However, I begin this section by describing a program that no longer exists because it informed or could have informed all three of the other programs. The Teen Mentoring Initiative (not the one associated with 4-H in the United States) was a collaboration between YouthLaunch and Big Brothers Big Sisters of Central Texas that was intended to support High School Bigs in Texas. In 2004, YouthLaunch published a resource manual: The ABCs of Teen Mentoring, which was, as it was entitled, “an implementation guide for BBBS coaches.” This short manual covered not only enrollment, matching, and relationship development specifics, but also guidance on how to structure the Big-Little RAP (“reflection and planning”) conversations. It also emphasized the importance of district and campus partnerships, mentor training, and closure procedures. For mentors, a 46-page handbook (The Next Big Thing) was provided that overviewed mentor roles, how the program worked, mentoring tips, communication skills information, and 20 pages of activities to use. YouthLaunch closed its offices in 2011, and these materials are presently unavailable. They are, however, still in use by some of the largest BBBS agencies (M. O’Teeter, personal communication) and may have informed Canada’s impressive materials for training and supporting teen mentors. High School Teen Mentoring Program—Big Brothers Big Sisters of Canada. BBBSC is Canada’s main provider of youth mentoring programs. A wider scope of mentoring program formats are conducted through BBBS of Canada than BBBS of America, and BBBSC devotes considerable resources to supporting their various youth mentoring efforts. In this program, mentor-mentee matches meet for a decided period of time, once a week, to participate in relationship-building activities. The High School Teen Mentoring Activity Book and Handbook (Government of Alberta, 2010a, 2010b) are resources developed by the BBBSC: Edmonton and Area agency and Alberta Education. The 56-page mentor handbook is similar to that developed by YouthLaunch. In fact, on the last page it notes the “original manual was received from Big Brothers Big Sisters of North Texas with permission to use the content” (p. 56), suggesting a link may exist between the BBBSC and YouthLaunch materials. The 120-page High School Teen Mentoring Activity Book provides a host of activities “to assist mentees in discovering their interest in talents, how

248  PROGRAMS AND CONTEXTS they best learn, possible career pathways, learning after high school, and how to make decisions and start planning” (p. 1). Although the mentees targeted by the activities (those in grades 3–5) may seem too young to benefit from career guidance provided by their mentor-as-career-coach, most of the activities are quite age appropriate self-exploration activities. Perhaps the real career coaching happens for the mentors/Bigs who, consistent with the helper therapy principle, may benefit as much or more from these activities as do the mentees/Littles. In addition, there are additional resources (High School Teen Mentoring Bin Resources) available for mentors to use with their mentees. These fantastic materials are available at alis.alberta.ca/publications (key phrase: mentoring). What they lack that the YouthLaunch materials provided more explicitly in their Resource Manual are specifics about how to recruit participants, match them, support the matches, and close the matches. Apparently the specifics needed to get the program off the ground and securely established in a school are left to the agencies. Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors. CAMP is a highly structured developmental mentoring program that pairs high school students with elementary and middle school students from grades 4–6 (Karcher, 2008). Together, the mentor-mentee pairs play games, participate in structured activities, and join others in events that link the youth with their families, teachers, and communities. CAMP’s main purpose is to increase connectedness of youth to their teachers, families, communities, and futures (Karcher & Santos, 2011b). This program is designed to be tailored to meet local school and youth needs. As such, it has been used with different curricular foci and with older mentees (Boy With a Ball, 2012; Sar & Bledsoe, 2011; Smith, 2011b). Learn more at www .crossagepeermentoring.com. Implementation of CAMP is guided by the Program Manual (Karcher, 2012a), which provides the sort of programmatic specifics not found in the BBBSC Teen Mentoring materials described earlier. CAMP and its program, manual, training materials, and mentor handbook were developed between 1995 and 1998 and evaluated as used in the St. Stephen’s Episcopal School in Austin, Texas (Karcher et al., 2002). Like the BBBSC Teen Mentoring program, it is accompanied by both a Mentor Handbook and a theory-based curriculum of structured activities. The Mentor Handbook (Karcher, 2012b) is linked to the Training Guide (Karcher, 2012c), which includes over 30 training activities. The curriculum, which is intended to promote

social skills and connectedness across the adolescent social ecology (e.g., connectedness to teachers, school, peers, reading) was developed and evaluated in Columbus, Wisconsin (Karcher, 2005b, 2008, 2009a; Karcher & Lindwall, 2003). The Program Manual is a guide for evaluating CAMP using validated measures of connectedness and social skills, so that the program’s impact can be measured in terms of these outcomes. The Program Manual also addresses the specifics of negotiating with school districts to establish the program; the details on what research says about who should mentor and be mentored in peer programs; and the specific steps for matching youth, involving parents, structuring meetings, evaluating impact, and closing matches. CAMP is currently under review for inclusion in the National Registry of Evidencebased Programs and Practices. Mentoring Works!: A Peer Helping Program (Avani, 1998). Avani’s Mentoring Works program is structured to match an older student with a younger student. Avani suggests the program can be applied across building levels (e.g., high-school-aged mentor with elementary-aged mentee), within a building level (upperclassmen mentoring freshman), and with mentees and mentors in the same grade. Avani’s training materials are for use with both mentors and mentees. The 70-page spiral-bound Facilitator’s Guide provides specifics on setting up a peer program, mentor and mentee recruitment, and what to do on the first meeting. The Facilitator’s Guide also includes a host of activities that mentors and mentees can use to interact together. The program emphasizes short- and long-term goal setting, critical-thinking and problem-solving skills, and communication skills. It is not clear that these activities would be sufficiently engaging to maintain the attention of elementary-aged children, as many worksheet activities in the accompanying Student Workbook use language not likely to appeal to or be developmentally appropriate for children. The materials include three posters and pamphlets for marketing purposes. Friday Night Live Mentoring Program. Although this program does not have a set of materials that are available for use by the public, it does present a skeleton of its program components online, and it has been described as an exemplary peer mentoring program (Scott-Nakai, 2008). The peer mentoring is part of a larger set of programs, but the specifics of the peer mentoring program, and specifically the mentor-support strategies it employs, are summarized fully online at http://www.yli.org. The program includes two lead adult staff, mentor and mentee

Cross-Age Peer Mentoring   249 training, parent orientation specific matching procedures, a 16-week curriculum, longer interactive group activities, and recommended summer booster activities. It also includes ongoing relationship monitoring procedures, ongoing parental feedback and engagement procedures, closure activities, and mission- and model-driven program evaluation. Assuming these procedures are implemented consistently and competently, this program is a good model to emulate. The Revised High School Bigs Model. Recently developed and currently being field-tested by BBBSA, this program does not currently have materials that are available to the public, and it has not been implemented federation-wide in all BBBSA agencies, but it is a much more comprehensive program than the one Herrera et al. (2008) evaluated. Working in collaboration with Herrera, Karcher, and staff from the Rochester and Kentuckianna agencies, Keoki Hansen led the development of this comprehensive model (Karcher, Hansen, & Herrera, 2010; Hansen, Swinton, & Christensen, 2009). Briefly reviewed here are the key elements: Big/teen training and mentor orientation; Little training; staff training; recruitment,

screening, and enrollment guidelines; matching; and match meetings (e.g., location, duration, focus, structure). Specific training on how mentors and mentees should choose their activities is based on theoretical and empirical evidence of what works best in this program (see Karcher, Herrera, & Hansen, 2010; Karcher & Hansen, this volume, Chapter 5). Perhaps most important, this model specifies the minimum amount of ongoing support that program staff should provide, ways to involve parents and the importance of doing so, how to secure administrator buy-in and establish school partnerships, and the need for effective closure practices. Essential Youth Mentoring Best Practices in Peer Mentoring: From Match to Closure Key practices in the order in which they occur during an academic school year, as most peer mentoring programs operate in schools, frame the next section. Given the importance of mentor training, I address elements of program training in a separate section later in the chapter. Most of these practices are summarized as questions for practitioners in Table 16.1.

Table 16.1   Checklist for Practitioners Creating the Right Context for the Program Is it clear to all involved—school administration, teachers, parents, mentors, and children—that the program is indeed to be a cross-age peer mentoring program, and not a tutoring, peer education, peer counseling, or peer helping program? Has a budget been developed such that sources of funds have been secured to cover any curriculum, food, transportation, or other after-school program essential materials? Have program coordinators been relieved of other duties in order to adequately plan, prepare, implement, monitor, and evaluate the program (whether the coordinator is a school employee or external program staff member)? Have you secured buy-in from school administration and negotiated and documented in writing what the school will provide (e.g., transportation, location, copies) and for how long? Have mentee identification procedures been approved by the school that will lessen the likelihood that stigma or deviancy training processes will be associated with the program by ensuring that only a fraction of program youth are at high behavioral or emotional risk? Key Startup Procedures Do you have forms and procedures for securing consent from mentees’ and mentors’ parents? Have recruitment and screening procedures for prospective mentors been selected? Will the process of selecting mentors use recommendations from teachers and school staff, and will an interview of youth be conducted by a program staff member? Program Fidelity and Evaluation Have you established a plan for monitoring program implementation and assigned to specific individuals responsibilities for monitoring each element of program fidelity? (Continued)

250  PROGRAMS AND CONTEXTS Table 16.1  (Continued) Have you considered how you might keep an implementation log or track attendance? Will you have mentors use a reflective journal, complete activity reflections following the day’s activities, or in other ways keep track of program fidelity and mentor satisfaction? Have measures of relationship quality and program satisfaction been identified? Have you aligned any curriculum with program outcomes, identified adequate measures of constructs to use in program evaluation, and made sure these measures reflect outcomes of interest to the community, school administration, parents, and youth? Have measures of primary outcomes been identified (a) that reveal the program has as its primary goal youth development generally, not skill or content mastery, and (b) that underscore this is not a tutoring, education, or counseling program? Has any evidence of the scales’ validity and reliability, when used with similar populations (age, race, risk level), been found to support the use of these (versus some other) outcome and process measures? Involving Stakeholders and Participants When determining how mentees are assigned to their mentors, do you have plans to utilize choice or other input from youth? Do you have plans to ensure parent and teacher involvement? Do program structures encourage parents to communicate with mentors and the program coordinator? Will parents be encouraged to serve as volunteers or otherwise be involved in program activities that may promote connectedness between children and their parents, between parents and the mentor, and between parents and the school staff? Have you considered whether teachers may assist in identifying participants, in implementing (and developing) the summer program, or in other ways to foster teacher buy-in? Program Structure and Activity Focus How will you provide structure for the weekly meetings? How can you prepare from the start for the end of the match? How will you structure, supervise, and reinforce match closure practices? What policies, practices, and consequences are in place to foster mentor commitment, enthusiastic use of planned activities (e.g., curriculum use), and consistent attendance? Training How will you train mentors continuously throughout the program? What will you do to make these requirements crystal clear to mentors before they commit to the program? When and how will you provide mentors with needed training in the role of mentors, the ways they should and should not interact with each other or with mentees, and how to identify times when they need to seek additional support? When during the year (and during the school day, specifically) will you provide ongoing training both to teach mentors the goals and specifics of any curriculum that is used in the program, and to help them deal with developmental issues related to the match (e.g., about gift-giving policies in winter, honest closure in spring) that often arise after the initial training? Will you provide training on key issues such as the relationship life cycle, active listening and effective communication, peer and social negotiation strategies, conflict resolution and mediation, the need to model enthusiastic use of the curriculum, how to reflect on the match, ways to monitor the quality of the match, and how to discourage deviancy training? How will you monitor who has participated in required trainings? Do you have a plan to identify who needs and who has received additional training?

Cross-Age Peer Mentoring   251 Having suggested fairly consistently so far that some sort of planned activities or themed curriculum of activities may be useful to orient mentors and mentees, I do not restate that point here. And, because so few programs have proposed the best practices for mentoring activities, and no research has examined what types are best, I refrain from identifying such practices as well. Research and time may tell us this, but right now we know too little to speak with confidence. The state of the research to date also cannot speak to the basics of program format and meeting location, time, and duration. Seemingly contradictory findings, like those mentioned earlier (e.g., teens like meeting in groups, but their mentees don’t), make it hard to be confident in any specific recommendations. But other promising practices seem less equivocal, and so I describe these in greater detail. Recruitment and Screening of Prospective Mentors. One of DuBois et al.’s (2002) theory-based best practices is related to the recruitment and screening of prospective mentors. In particular, recruiting mentors from the helping professions was an empirical best practice. Several programs reviewed in this chapter include teacher referral and recommendations and suggest conducting interviews to determine whether the mentor is mature enough to serve in the role of a peer mentor. For example, the Mentoring Works program bases its selection of mentors on recommendations from teachers and school staff, followed by an interview with a program facilitator (Avani, 1998). The Teen Mentoring Initiative (YouthLaunch, 2004) used a similar approach; if there were too many applicants, a selection committee used a scoring matrix to rank the most promising applicants. In CAMP, mentors who have higher scores on social interest and attitudes toward youth are given priority because research suggests they are the most helpful (Karcher & Lindwall, 2003; Karcher, Davidson et al., 2010). Matching: How Mentees Are Assigned to Their Peer Mentors. The ways in which matches are made shapes the outcome of the mentoring experience for both the mentor and the mentee. A key element of effective mentoring programs is a systematic matching method (DuBois et al., 2011), and using interest surveys to match by similar interests is the predominant method (e.g., O’Hara, 2011). However, virtually no research has examined how often this practice takes place or the outcomes of varying types of matching procedures (see Pryce et al., this volume, Chapter 29). Equally common—especially in school settings with matches not meeting at a set time as

part of a larger group—is for matches to be based on scheduling or availability (see Herrera & Karcher, this volume, Chapter 14). For example, in the Peer Buddy Program, matching is done by selecting volunteers in the same classes, pairing students with shared interests/experiences, or pairing students who know each other and/or live near each other (Hughes & Carter, 2008). Self-selection is another viable approach. In his Mentoring Works program, Avani (1998) describes matching mentors and mentees with whom they feel “connected” and suggests one day of the program be set aside especially for determining matches. Avani suggests this matching meeting take place after two or three “getting to know you” meetings are held. This allows mentors and mentees to intuitively decide with whom they feel most “connected.” Avani, however, does not give clear guidelines as to how this matching ought to occur. In contrast, in the “meet-n-greet” approach (see Karcher, 2012a), mentors and mentees interact using icebreakers and afterward list those people they liked meeting to provide program staff with additional information for matching (Pryce et al., this volume, Chapter 29). In the meet-n-greet procedure, small groups of five to six mentees and mentors come together to meet, interact, and get to know each other. Before the interaction time, both mentors and mentees are instructed to try to remember the names of one to three individuals whom they enjoyed meeting. Afterward, mentors and mentees go off into separate groups and are asked to simply list the names of individuals they remembered meeting and with whom they enjoyed interacting. They are explicitly asked not asked to rank their choices in order to avoid the disappointment of a first choice being unavailable. Following this process, program staff take the names and create matches, using all other available information (interests, schedules, etc.), attempting to make matches based on the mutual preferences of mentors and mentees when possible. A quasi-experimental pilot study of the meet-n-greet process (Karcher & Santos, 2011a) found higher relationship quality in matches in which one or both of the participants were matched with someone they listed than when neither partner received someone they listed. Monitoring Program Implementation. DuBois et al. (2002) described the importance of monitoring program implementation. Several mentoring programs describe methods for monitoring their programs, ranging from keeping log sheets and taking attendance (YouthLaunch, 2004) to maintaining a reflective journal (Hughes & Carter, 2008). CAMP

252  PROGRAMS AND CONTEXTS (Karcher, 2012a) encourages participants to complete activity reflections following the day’s activities to ensure that the mentees’ needs have been met, and also to have pairs reflect on what can be changed and improved in the next meeting. Earnest efforts are encouraged by linking the information written down to a competitive between-match game called the Newlymatched Game in which mentors use these notes to answer questions about their mentees. Parent and Teacher Involvement. Parent involvement is key. The revised High School Bigs model (Karcher, Hansen, & Herrera, 2010) explicitly requires program staff to communicate with the parents of both Bigs and Littles multiple times each year, to report on their child’s absence from the program, and to invite parents to in-person and end-ofyear activities. Information for parents is provided throughout the child’s involvement and a parent guide is provided. CAMP hosts quarterly “Super Saturday” events in which parents spend time with their children’s mentors and see the work their children have done. Saturday events have included trips to the zoo, a picnic at a public park, or a mini-carnival at the school. The intention is to provide time for parents to understand what their children have been doing, as some evidence indicates that the academic achievement gains from CAMP are mediated by improvements in youth reports of parental connectedness after program participation (Karcher et al., 2002). This day serves to promote connectedness between children and their parents, and between the parents and both the mentor and the school staff. Another component of CAMP is strategic teacher involvement through a teacher connectedness activity. During this 1- to 2-month series of transactions, the mentee and the mentor learn interview skills, practice these skills, conduct an interview with a teacher, and relate the teacher’s experience to their own experiences. The goal is to increase the children’s caring about teachers, to prompt improved behaviors in and attitudes toward school as a function of this caring, and engage the teachers in the mentoring program to secure their support. Meeting and Match Closure Practices. Both BBBSA and BBBSC describe the importance of “wrapping up” the matches effectively (Government of Alberta, 2010; YouthLaunch, 2004), but only recently has BBBSA detailed the specifics of this process (e.g., how many absences a mentor can have before the match is closed). In the revised High School Bigs model, the same closure ritual used in CAMP is employed, when it is deemed necessary to close a match. By contrast, all CAMP matches are closed at

the end of the year, so CAMP takes additional steps to prepare mentees and mentors for the ending by using regular relationship reflections and practice closures. The daily 3-2-1 activities and quarterly reflections prepare youth to be successful at saying goodbye at year’s end (Karcher, 2012a). The 3-2-1 activity is a helpful approach to structuring the opening and closing of each mentoring session. At the start, the mentor and the mentee take turns describing three good things that occurred the prior week, two bad things that happened, and one thing they hope to be different in the next week. At the end of the meeting, the same 3-2-1 process occurs, but this time the reflection is on the day’s activities or the relationship itself (with mentors usually having to model this relationship reflection for quite a while before mentees feel comfortable trying it). This makes sure mentors and mentees know what’s going on in each other’s lives, that they get immediate feedback on their interactions and the curricular activity for the day, and that they air problems quickly. In addition, quarterly “relationship reflections” take place, which involve the mentor and the mentee reflecting on what worked well in the match, what challenges they faced in their match, and what they could change to avoid that problem in this and future relationships. This procedure helps youth experience a successful goodbye by coaching them in how to reflect on their experience. Successes in goodbyes are critical for youth who have had little control over the losses of important people in their lives. Training Mentors Both intuitively as well as empirically, training is related to the formation of a strong cross-age mentoring relationship (Herrera et al., 2008). DuBois et al. (2002) described training, both prematch and throughout the duration of the match, as a theory-based and empirically based best practice for all youth mentors. Training teens to mentor may be a bit more challenging than training adults simply because adults are generally more cognitively mature, but also because, as described earlier, teen mentors are more likely than adults to be distracted by their peers. In the programs described earlier, there was considerable overlap in terms of the training that mentors received. Common themes and training content across the training programs include a focus on training mentors about the relationship life cycle, active listening, negotiation strategies, conflict resolution, and mentor roles. Relationship Life Cycle. Most programs describe the cycle of mentoring relationships from “hello

Cross-Age Peer Mentoring   253 to goodbye.” Manuals describe specific tips for interacting with mentees at different stages in the process, as well as goals that they might set at specific stages. The training prepares mentors for dealing with the awkwardness of the beginning of the relationship and the importance of celebrating their progress and successes in the relationship as it ends. Active Listening and Effective Communication. Communication and active listening are key components of all the training programs. Some programs provide training on effective communication, including the use of body language, eye contact, clarifying and reflecting, and attending. The YouthLaunch (2004) materials train mentors to use conversation builders to show their mentees that they are important by being empathic, paying attention, praising effort, and caring about what their mentees are saying. Mentors are also given a list of conversation blockers, such as criticism and sarcasm, which might prevent mentees from engaging in relationships. They are trained in the basics of nonverbal communication (e.g., to “SOFTEN” up: smile, open, lean forward, touch, eye contact, and nod; YouthLaunch, 2004, p. 24). Interactive rather than didactic training seems especially developmentally appropriate. Although interactive training could allow for deviancy training, it also may meet youths’ valid needs for socializations in ways that inhibit deviancy training (or at least allow its expression away from the mentees). For example, to introduce the importance of communication in relationships, facilitators in the Mentoring Works program (Avani, 1998) introduce several role plays in the training exercise to help mentors identify some of their everyday mannerisms that might be ineffective communication strategies. Negotiation Strategies. Included under the umbrella of negotiation strategies are the ways in which mentors and mentees decide what they are going to do during a meeting, how they will collaborate, and what the main goals of the mentoring session will be. Training in the importance and means of collaborating when making decisions about activities and conversation topics seems critical given the empirical support for such collaboration (Karcher, Herrera et al., 2010) and the developmental sophistication required for mentors to scaffold the collaboration process for mentees who otherwise could not perform it (Karcher, 2008; Karcher, 2012a). The YouthLaunch training program emphasizes a process titled “Discovering and Negotiating.” In this

stage of training, mentors are taught how to negotiate goals for the relationship. Conflict Resolution/Mediation. Multiple programs addressed problem solving and conflict resolution in their training (Tindall & Black, 2009; Varenhorst, 2003). In the Mentoring Works program, trainees use a problem-solving checklist to resolve hypothetical conflicts in case studies. Karcher’s (2012c) training teaches cognitive problem-solving techniques at different developmental levels and provides exercises to illustrate them. Mentor Roles. Mentors, especially high school mentors, can experience confusion in their new role. Several of the programs use training time to explore the various roles mentors might have, as well as to address any confusion they might experience due to their new role as a mentor. CAMP uses video presentations and discussions to describe mentor roles. Using video clips, such as from Disney’s Jungle Book and the Kung Fu Panda movies, the trainings explore ways in which the films’ characters illustrate different mentoring styles (Karcher, 2009b). A CAMP training activity developed by Ze’v Korn uses hats to help mentors consider the different hats they will wear as a cross-age peer mentor. Involving High School Mentors in Creation of the Curriculum. Given the evidence emerging in the field of positive youth development, and in particular on the importance of involving youth in creating the intervention programs in which they serve (Wong et al., 2011), finding ways to strategically involve mentors in developing the curricular activities makes sense. The YouthLaunch (2004) training recommends letting the teen Bigs help with as much planning as possible. The CAMP curriculum guide and Training Guide (Karcher, 2012c; Karcher & Judson, 2012) provide information on how program staff should develop theoretically consistent curricula and how to involve mentors in curricula development. Mentors are explicitly taught the theory and coached to apply it to their own lives in order to develop activities for their mentees that are meaningful and programmatically consistent. Training Mentors to Monitor the Status of the Relationship. Mentors will be most effective in seeking out the support they need and thereby less likely to experience negative effects of being overwhelmed by challenging relationships (e.g., Karcher & Lindwall, 2003) when they are trained in how to monitor their own relationship. The YouthLaunch (2004) materials train mentors in conducting RAP sessions for this

254  PROGRAMS AND CONTEXTS purpose. These sessions are conducted to monitor and support the matches in an informal way without a formal agenda. The questions that guide the RAP sessions are, “what?,” “so what?,” and “now what?” Mentors describe what happened in the last week with their mentees, how these events affected their relationships, and what can be done to further develop the relationships. The CAMP curriculum also encourages this in 3-2-1 activities and relationship reflections. While these activities are intended to prepare the mentors and the mentees to more effectively perform the match termination ritual at the end of the year, the written information that is collected also helps them identify and record how the match is going so that the mentor or program staff (who read it afterward) can identify red flags as they emerge. These RAP sessions and end-of-meeting 3-2-1 activities help mentors plan to improve their relationship in the future but also help mentors provide mentees empathy, praise, and attention in a clear, consistent format.

Summary DuBois et al.’s (2002) list of best practices provides a lens for understanding how cross-age peer mentoring programs may be organized to achieve maximal impact. It is vital that practitioners implementing cross-age peer mentoring programs incorporate the necessary techniques and key practices. It is also important when drafting program manuals to include specific details regarding how to tailor program practices developed for adults or community mentoring settings to schools and to teen mentors. Finally, and perhaps unique to peer mentoring, the use of an orienting activity (a curricular or planned activity) may be a useful practice for organizing mentors and mentees in dyads that have prosocial engagement and minimal deviancy training. But there are likely other ways besides using curricular activities to do what is needed—that is, to foster the developmentally critical experience of receiving empathy, praise, and attention from a mentor within a clear, consistent structure (the program). However connection and sustained attention between mentor and mentee is cultivated, it is important to keep the focus of mentors and mentees on the matches; the program processes or practices that achieve this focus may be what differentiates crossage peer mentoring from other peer interventions.

Conclusion I suggest that, to foster thriving matches, cross-age peer mentoring programs need to create a holding

environment that allows the power of peer relationships to take shape and operate over a sustained period of time. Good peer mentoring programs should provide the opportunity for a child to idealize an older peer from whom the mentee receives empathy, praise, and attention. Successful peer mentoring, in my opinion, requires program structures that minimize mentor-to-mentor interactions (particularly their spontaneous, youth-focused interactions) and fosters supportive teen-child interactions. I described as a “promising practice” in crossage peer mentoring the careful use of a general curriculum, like those described in the pioneering peer mentoring work of the 1960s, 1970s, and 1980s described earlier. But the use of any curriculum must not be intended to fix kids, solve problems, or teach academic skills; rather, its purpose should be to promote positive youth development generally, and primarily to help orient the mentor and the mentee toward one another. This is key. The curriculum’s purpose is to contain and direct the participants’ attention to one another, not necessarily to impart information or teach specific skills. Despite evidence of the utility of planned activities; the importance of matching, training, supervision, and closure practices; and the apparent benefits of cross-age peer mentoring for mentors and mentees, and in particular for different combinations of types of youth, we have not really scratched the surface of what there is to know about what makes cross-age peer mentoring work. We must acknowledge that we have not exercised due diligence to know what practices are most helpful (Cavell, 2012). Between now and the next edition of the Handbook of Youth Mentoring, I hope more studies of cross-age peer mentoring are conducted that exemplify high fidelity of implementation, strong elements of programmatic support, and research designs that can tease apart the unique benefits of different programmatic structures. Only through such research can cross-age peer mentoring move from a good idea with lots of training activities available to support it to a systematic approach to intervention and the promotion of positive youth development that can become part of a school counselor’s arsenal of empirically supported intervention options.

References Avani, N. (1998). Mentoring works! A peer helping program: Facilitator’s guide. Plainview, NY: The Bureau for At-Risk Youth.

Cross-Age Peer Mentoring   255 Benson, P. L. (1999). An initial look at America’s Promise: Successes, challenges, and opportunities. Philadelphia: Public/Private Ventures. Bessell, A. G., & Kloosterman, V. I. (2011). Teen Trendsetters™ Reading Mentoring: 2009–2010 Florida Program Evaluation Report. Unpublished report. Miami, FL: University of Miami. Bowman, R. P., & Bowman, S. C. (2005). Meaningful mentoring. Chapin, SC: Youthlight. Boy With a Ball. (2012). Fixing academic failure: The Boy With a Ball Harlandale Cross-Age Mentoring Program. Retrieved from http://www.boywithaball .com/news-velocity-1.html Brewer, C. (2009). Social interest and self-efficacy levels among high school volunteer mentors and their nonmentor peers: A comparison study. Unpublished thesis, Walden University, Minneapolis, MN. Carter, E. W., Cushing, L. S., & Kennedy, C. H. (2009). Peer support strategies for improving all students’ social lives and learning. Baltimore, MD: Paul H. Brookes. Carter, E. W., & Hughes, C. (2005). Increasing social interaction among adolescents with intellectual disabilities and their general education peers: Effective interventions. Research and Practice for Persons with Severe Disabilities, 30(4), 179–193. Cavell, T. A. (2012). Teen mentoring: Findings from three studies. Big Brothers Big Sisters Canada national convention, BMO Institute for Learning, Toronto, Ontario, Canada. Cavell, T. A., & Henrie, J. L. (2010). Deconstructing serendipity: Focus, purpose, and authorship in lunch buddy mentoring. New Directions for Youth Development, 126, 107–112. Chew, A., & Wallace, B. C. (2008). So no child is left behind: A peer mentoring/tutoring program for atrisk urban youth attending a college preparatory high school. In B. C. Wallace (Ed.), Toward equity in health: A new global approach to health disparities (pp. 507–528). New York: Springer. Cox, R. (2006). Expanding the spirit of mentoring: Simple steps and fun activities for a flourishing peer mentor or peer support programme. Invercargill, New Zealand: Essential Resources Educational. Crandall, J. E. (1991). A scale for social interest. Special Issue: Social interest. Individual Psychology: Journal of Adlerian Theory, Research & Practice, 47(1), 106–114. Crooks, C. V., Chiodo, D., Thomas, D., & Hughes, R. (2010). Strengths-based programming for First Nations youth in schools: Building engagement through healthy relationships and leadership skills. International Journal of Mental Health Addiction, 8, 160–173. Dodge, K. A., Dishion, T. J., & Lansford, J. E. (2006a). Deviant peer influences in intervention and public policy for youth. Social Policy Report, XX(I), 3–18. Dodge, K. A., Dishion, T. J., & Lansford, J. E. (Eds.). (2006b). Deviant peer influences in programs for youth: Problems and solutions. New York: Guilford.

DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Eddy, C. L. (2011). The influence of cross-age mentoring on primary-aged children’s reading scores. Un­­ published dissertation, Walden University, Minneapolis, MN. Finckler, J. (2003). Evaluation of a teen mentor program. Dissertation Abstracts International Section A: Humanities and Social Sciences, 1178. Fletcher, K. R., Norem-Hebeisen, A., Johnson, D. W., & Underwager, R. C. (1974). Extend: Youth reaching youth (A group interaction plan for training youth). Minneapolis, MN: Augsburg Publishing House. Garringer, M., & MacRae, P. (2008). Building effective peer mentoring programs in schools: An introductory guide. Folsom, CA: Mentoring Resource Center. Goodlad, S. (1998). Mentoring and tutoring by students. London: Kogan Page. Goodman, G. (1972). Companionship therapy: Studies in structured intimacy. San Francisco: JosseyBass. Government of Alberta. (2010a). High school teen mentoring activity book. Alberta, Canada: Advanced Education and Technology. Government of Alberta. (2010b). High school teen mentoring handbook. Alberta, Canada: Advanced Education and Technology. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work? An impact study of the Big Brothers Big Sisters Program. Evaluation Review, 22(3), 403–426. Hamburg, B. A., & Varenhorst, B. (1972). Peer counseling in the secondary schools: A community mental health project for youth. American Journal of Orthopsychiatry, 42(4), 566–581. Hansen, K., Swinton, J., & Christensen, T. (2009). Rethinking how we work with high school Bigs. Big Brothers Big Sisters national conference, Miami, Florida. Hebeisen, A. (1973). Peer program for youth: A group interaction plan to develop self-esteem, self-understanding, and communication skills. Minneapolis, MN: Augsburg Publishing House. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Herrera, C., Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, J. (2008). High school students as mentors: Findings from the Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures.

256  PROGRAMS AND CONTEXTS Hughes, C., & Carter, E. W. (2008). Peer buddy programs for successful secondary school inclusion. Baltimore, MD: Paul H. Brookes. Johnson, V. L., Holt, L. J., Bry, B. H., & Powell, S. R. (2008). Effects of an integrated prevention pro­­ gram on urban youth transitioning into high school. Journal of Applied School Psychology, 24(2), 225–246. Johnson, V., Simon, P., & Mun, E. Y. (in press). A peer-led high school transition program increases graduation rates among Latino males. The Journal of Edu­ cational Research. Karcher, M. J. (2005a). Cross-age peer mentoring. In D. L. DuBois, & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 266–285). Thousand Oaks, CA: Sage. Karcher, M. J. (2005b). The effects of school-based developmental mentoring and mentors’ attendance on mentees’ self-esteem, behavior, and connectedness. Psychology in the Schools, 42, 65–77. Karcher, M. J. (2006). What happens when high school mentors don’t show up? In L. Golden & P. Henderson (Eds.), Case studies in school counseling (pp. 44–53). Alexandria, VA: ACA Press. Karcher, M. J. (2007). Cross-age peer mentoring. In J. E. Rhodes (Ed.), Research in Action Series (Issue 7). Alexandria, VA: MENTOR. Karcher, M. J. (2008). The Cross-Age Mentoring Program: A developmental intervention for promoting students’ connectedness across grade levels. Professional School Counseling, 12(2), 137–143. Karcher, M. J. (2009a). Increases in academic connectedness and self-esteem among high school students who serve as cross-age peer mentors. Professional School Counseling, 12(4), 292–299. Karcher, M. J. (2009b). Neither Baloo nor Bagheera alone could get Mowgli safely to the man village: Helping mentors decide what to do with their mentees. Presentation at the 4th National School-Based Mentoring Conference, Kansas City, Missouri, June 17. Karcher, M. J. (2012a). The Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors: Program manual. San Antonio, TX: Developmental Press. Karcher, M. J. (2012b). The Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors: Mentor handbook. San Antonio, TX: Developmental Press. Karcher, M. J. (2012c). The Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors: Training guide. San Antonio, TX: Developmental Press. Karcher, M. J., Davidson, A., Rhodes, J. E., Herrera, C. (2010). Pygmalion in the program: The role of teenage mentors’ attitudes in shaping their mentees’ outcomes. Applied Developmental Science, 14(4), 212–227. Karcher, M. J., Davis, C., & Powell, B. (2002). Developmental mentoring in the schools: Testing connectedness as a mediating variable in the promotion

of academic achievement. School Community Journal, 12, 36–52. Karcher, M. J., Hansen, K., & Herrera, C. (2010). High School Bigs model: Draft first year (with research evidence and essential elements). Philadelphia: Big Brothers Big Sisters of America. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. New Directions for Youth Development, 2010 126, 51–69. Karcher, M. J., Holcomb, M., & Zambrano, E. (2008). Measuring adolescent connectedness: A guide for school-based assessment and program evaluation. In H. L. K. Coleman & C. Yeh (Eds.), Handbook of School Counseling (pp. 649–669). Mahwah, NJ: Erlbaum. Karcher, M. J., & Judson, B. J. (2012). The Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors: Connectedness curriculum. San Antonio, TX: Developmental Press. Karcher, M. J., & Lindwall, J. (2003). Social interest, connectedness, and challenging experiences. What makes high school mentors persist? Journal of Individual Psychology, 59, 293–315. Karcher, M. J., & Santos, K. (2011a, March 30). Natural selection: Exploratory test of the meet-n-greet approach to mentor-mentee matching in youth mentoring programs. Paper presented at the 2011 biennial meeting of the Society for Research in Child Development, Montreal, Quebec, Canada. Karcher, M. J., & Santos, K. (2011b). Promoting connectedness through developmental interventions: Adapting the Cross-Age Mentoring Program (CAMP) for youth in Asia. Asian Journal of Counseling, 18(1/2), 125–147. Kohut, H. (1977). Restoration of the self. New York: International Universities Press. Larose, S., Cyrenne, D., Garceau, O., Harvey, M., Guay, F., Godin, F., . . . Deschênes, C. (2011). Academic mentoring and dropout prevention for students in math, science and technology. Mentoring & Tutoring: Partnership in Learning, 19, 419–439. Life Directions. (2006). Peer motivation and peer mentor program evaluation (2005–2006): Executive summary. Chicago: Life Directions: Peers Inspiring Peers. Mathews, T. L. (2007). Effects of a comprehensive peer engagement program on children with socially withdrawn behaviors and histories of maltreatment. Unpublished dissertation, University of Kansas, Lawrence. McCord, J. (2003). Cures that harm: Unanticipated outcomes of crime prevention programs. Annals of the American Academy of Political and Social Science, 587, 16–30. Milliken, B. (2007). The last dropout: Stop the epidemic! Carlsbad, CA: Hay House. Myrick, R. D., & Erney, T. (2004). Youth helping youth: A handbook for training peer facilitators. Minneapolis, MN: Educational Media Corporation.

Cross-Age Peer Mentoring   257 O’Donnell, C. R., Lydgate, T., & Fo, W. S. (1979). The Buddy System: Review and follow-up. Child Behavior Therapy, 1(2), 161–169. O’Hara, D. (2011). The impact of peer mentoring on pupils’ emotional literacy competencies. Educational Psychology in Practice, 27(3), 271–291. Perlmutter, F., & Durham, D. (1965). Using teen-agers to supplement casework service. Social Work, 10(2), 41–46. Powell, S. R. (1988). The peer group handbook. Lawrenceville, NJ: Princeton Center for Leadership Training. Powell, S. R. (1993). The power of positive influence: Leadership training for today’s teens. Special Services in the Schools, 18, 119–136. Probst, K. (2006). Mentoring for meaningful results: Assetbuilding tips, tools, and activities for youth and adults. Minneapolis, MN: SEARCH Institute Press. Reinherz, H. (1964). The therapeutic use of student volunteers. Children, 11, 137–142. Rhodes, J. E. (1994). Older and wiser: Mentoring relationships in childhood and adolescence. Journal of Primary Prevention, 14(3), 187–196. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. (pp. 30–43). Thousand Oaks, CA: Sage. Riessman, F. (1965). The “helper” therapy principle. Social Work, 10(2), 27–32. Rosenblum, A., Magura, S., Fong, C., Curry, P., Norwood, C., & Casella, D. (2005). Effects of peer mentoring on HIV-affected youths’ substance use risk and association with substance using friends. Journal of Social Service Research, 32(2), 45–60. Sar, B. K., & Bledsoe, L. K. (2011). Investigation of the effectiveness of a developmental mentoring model as an intervention/prevention strategy for juveniles of varying levels of risk among middle school youth in Metro Louisville. Two-year grant, funded by the Office
of Juvenile Justice and Delinquency Prevention. Scott-Nakai, K. (2008). Effective peer mentoring: The California Friday Night Live Mentoring Program. U.S. Department of Education Funded Mentoring Resources Center Case Study Series. Folsom, CA: Mentoring Resource Center. Smith, L. H. (2011a). Cross-age peer mentoring approach to impact the health outcomes of children and families. Journal for Specialists in Pediatric Nursing, 16(3), 220–225. Smith, L. H. (2011b). Piloting the use of teen mentors to promote a healthy diet and physical activity among children in Appalachia. Journal for Specialists in Pediatric Nursing, 16(1), 16–26.

Smith-Jentsch, K. A., Scielzo, S. A., Yarbrough, C. S., & Rosopa, P. J. (2008). A comparison of face-to-face and electronic peer-mentoring: Interactions with mentor gender. Journal of Vocational Behavior, 72(2), 193–206. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22(4), 331–354. Steiman, A. H. K. (2006). Peer-mentoring for newcomer immigrant students. Unpublished doctoral dissertation, University of California, Los Angeles. Tindall, J. A. (2009a). Peer power, book one—strategies for the professional leader: Becoming an effective peer helper and conflict mediator (4th ed.). New York: Routledge. Tindall, J. A. (2009b). Peer power, book two—strategies for the professional leader: Applying peer helper skills (3rd ed.). New York: Routledge. Tindall, J. A., & Black, D. R. (2009). Peer programs: An in-depth look at peer programs—planning, implantation, and administration (2nd ed.). New York: Routledge. Topping, K. (1996). Reaching where adults cannot. Peer education and peer counselling. Educational Psycho­ logy in Practice, 11(4), 23–29. Varenhorst, B. (1983). Real friends: Becoming the friend you’d like to have. San Francisco: Harper & Row. Varenhorst, B. (2003). An assets builder’s guide to training peer helpers. Minneapolis, MN: Search Institute Publications. Weinberger, S. G. (2005). Mentoring a movement: My personal journey. Norwalk, CT: Author. Westerlund, D., Granucci, E. A., Gamache, P., & Clark, H. B. (2006). Effects of peer mentors on work-related performance of adolescents with behavioral and/or learning disabilities. Journal of Positive Behavior Interventions, 8(4), 244–251. Willis, P., Bland, R., Manka, L., & Craft, C. (2012). The ABC of peer mentoring—what secondary students have to say about cross-age peer mentoring in a regional Australian school. Educational Research and Evaluation: An International Journal on Theory and Practice, 18(2), 173–185. Wong, N., Zimmerman, M., & Parker, E. (2011). A typology of youth participation and empowerment for child and adolescent health promotion. American Journal of Community Psychology, 46, 100–114. Yeh, C. J., Ching, A. M., Okubo, Y., & Luthar, S. S. (2007). Development of a mentoring program for Chinese immigrant adolescents’ cultural adjustment. Adolescence, 42(168), 733–747. YouthLaunch. (2004). Teen mentoring initiative resource manual (1st ed). Austin, TX: Author.

17  ELECTRONIC MENTORING AND MEDIA Carmit-Noa Shpigelman1

Introduction The Roots of E-Mentoring Over the past several decades, face-to-face (FtF) mentoring programs for youth have become increasingly widespread. The presence of a caring mentor, often an adult, has long been recognized as an important resource in the lives of at-risk youth (DuBois, Holloway, Valentine, & Cooper, 2002). Changes in the physical and social environments of youth, such as new patterns beyond the traditional family, parents who work around the clock, overcrowded schools, and less cohesive communities, have dramatically reduced youth’s ready access to caring adults (Rhodes, Spencer, Saito, & Sipe, 2006). At the same time, electronic media, which has become an accepted and common means of communication, has opened up a new mentoring opportunity for youth, termed electronic mentoring (e-mentoring). E-mentoring programs arose for many of the same reasons that FtF mentoring programs were developed, that is, as a means to compensate when naturally occurring mentoring relationships were not equitably available (Hamilton & Scandura, 2003). Similar to FtF mentoring programs, e-mentoring programs also focus on creating educational and vocational opportunities for underprivileged or underrepresented populations (Bierema & Merriam, 2002). However, as opposed to FtF mentoring, less is known about the dynamics and the implications of e-mentoring (Bierema & Merriam, 2002; Miller & Griffiths, 2005).

In the first edition of the Handbook of Youth Mentoring, Miller and Griffiths (2005) presented the unique features of e-mentoring versus FtF mentoring. To complement their chapter, this chapter focuses on typologies of e-mentoring programs, the challenges of e-mentoring for youth, and models and strategies for coping with these challenges, while also reviewing additional empirical studies of e-mentoring published since the first edition. Definition of E-Mentoring E-mentoring refers to a relationship in which a mentor, usually a more experienced or an older person, provides guidance and support to a less experienced or younger person (the mentee) via distance communication technologies. E-mentoring is also referred to as telementoring, online mentoring, or virtual mentoring. The term telementoring is routinely used when the mentoring relationships are conducted via telephone or media that transmit voice communication electronically, whereas the term e-mentoring refers primarily to text-based electronic communication (Single & Single, 2005). The Development of E-Mentoring Over the years, computer innovations have led to several new forms of electronic communication, each with unique implications and applicability for e-mentoring (Mallen, Day, & Green, 2003). Computer-mediated communication (CMC) was originally defined as a form of electronic written communication. In accordance with this definition,

1 This work was supported by the U.S. Department of Education, National Institute on Disability and Rehabilitation Research (H133P060003). The views expressed here do not necessarily represent the policy of the Department of Education.

259

260  PROGRAMS AND CONTEXTS the first communication tool to be developed and used was electronic mail (e-mail) by which electronic text messages were exchanged between two or more individuals via the internet. Later, another type of asynchronous communication was developed—online discussions, also referred to as bulletin boards or web forums. In online discussions, people can post messages, usually questions and answers on a related topic, which are displayed in chronological order, or organized into thematic groups for others to log in and view (Benford, 2008). Most e-mentoring programs have been based on asynchronous communication, which is communication that is not dependent on the physical presence of the users. Each user can send messages or reply to others at different times (Harris & Figg, 2000). As

computer technologies have progressed, CMC and e-mentoring, respectively, have expanded to include multimedia and synchronous communication (i.e., the users are present and respond in real time) conducted with the aid of audio tools (microphone) and/ or video tools (webcam). There has been increasing use of telephone and video conferencing via the internet in e-mentoring programs as well as the use of webcams in conjunction with instant messaging services. Initiated in 1993, the first large-scale K–12 education e-mentoring program, the Electronic Emissary Project, matched K–12 public school students and teachers with subject-matter experts in various disciplines and was delivered primarily by asynchronous communication such as e-mail and web forums (see Table 17.1). Over the years, the online interaction

Table 17.1   Development of Online Mentoring Environments Program

Description

Public Relations Society of America (PRSA), College of Fellows

1989: Online mentoring program for novices in the area of public relations

http://www.prsa.org/leadership/cof DO-IT: Disabilities, Opportunities, Internetworking, and Technology

1992: E-mentoring community providing academic and career support for youth with disabilities

http://www.washington.edu/doit http://emissary.wm.edu

1993: The first large-scale K–12 education e-mentoring program

International Telementoring Project

1994: E-mentoring program for K–12 public school students

Electronic Emissary Project

http://www.telementor.org MentorNet http://www.mentornet.net iMentor http://www.imentor.org I Could Be http://www.icouldbe.org DHC: Digital Heroes Campaign http://www.mentoring.org Second Life http://secondlife.com SimTeach http://www.simteach.com GED Exit Option http://www.broward.k12.fl.us/

1997: E-mentoring network for diversity in engineering and science 1999: A New York City–based youth mentoring organization that uses e-mail communication to enhance in-person youth mentoring 2000: E-mentoring program for building self-esteem, knowledge of basic financial principles, and career exploration for lower-income middle school and high school students 2000: E-mentoring program that promotes youth’s development and prevents problems in behavioral, academic, and psychological functioning 2003: A 3D virtual world in which users can socialize, customize an avatar, and interact with each other via text or voice chat 2006: An immersive learning environment for university instructional designers, faculty, and administrators 2006: An online version of a traditional mentoring program for high school students at risk of leaving school without completing graduation requirements

Electronic Mentoring and Media   261 conducted in the program has expanded to include synchronous communication such as chats and audio and video conferencing (Harris, Rotenberg, & O’Brysan, 1997). However, most online mentoring programs are still delivered by asynchronous communication, which allows time for reflection before responding (Ensher, Heun, & Blanchard, 2003). By comparison, most online tutoring, coaching, and learning programs are delivered by synchronous communication (Eröz-Tuğa & Sadler, 2009). The second generation of the internet, also known as web 2.0, has created new environments for e-mentoring such as social network websites and virtual worlds. In these online environments, users can create the content, share, and collaborate with other users. O’Neill (2011) argued that, in the context of e-mentoring, these environments might allow local communities to build social capital (networks, norms, and trust that enable participants to act together more effectively to pursue shared objectives) and to promote educational achievements (Putnam, 1995). E-mentoring conducted through virtual worlds, threedimensional (3D) graphical environments such as Second Life, may also build social capital. The users of such virtual worlds “inhabit” a remote environment and interact with other users via their own graphical self-representations known as avatars; they can socialize, take part in group or individual activities, and create and exchange virtual items and services in a way more akin to the nature of FtF mentoring programs (Barnes, 2003; Benford, 2008; Kamel Boulos, Hetherington, & Wheeler, 2007). However, a few e-mentoring programs have been conducted in 3D environments and most of these targeted e-learning (especially for adult education), such as the SimTeach program. To summarize, e-mentoring may be conducted textually, by audio only, or by video (webcam). It also can be delivered synchronously through chat rooms or instant messaging, or asynchronously through e-mail or discussion lists (Leh, 2001). Similar to traditional mentoring programs, e-mentoring may be conducted in different modes (individual versus group) and as different types, as described below.

Theory Typologies of E-Mentoring Ensher et al. (2003) have proposed a typology of e-mentoring. In the first type, CMC-only, mentoring is provided online by e-mails, chats, instant messages, and the like. For example, MentorNet is a yearlong program that brings students and professionals together entirely by e-mail. The purpose of

the program is to partner undergraduate and graduate females with industry professionals in science, technology, engineering, and math (STEM) as a way to increase female representation in science professions (MentorNet, 2008). In the second type, CMC-primary, the majority of mentoring interactions (more than 50%) are conducted online, but they may also be supplemented by other media tools such as telephone calls and FtF interactions. For example, the Public Relations Society of America (PRSA) College of Fellows matches senior professionals in the area of public relations and mentees who are novices, who are transitioning to the field, or who just need specific advice. Mentors and mentees typically are matched via a computerized process and then meet face to face for a 20-minute period at the annual conference. Most of the pairs communicate only via the internet, whereas others may also communicate occasionally via telephone or in person (Ensher et al., 2003). In the third and last type, CMC-supplemental, the majority of mentoring is done in person, yet the relationship is supplemented via synchronous and asynchronous online support. Mentors and mentees may be in the same location or have a strong possibility of physically seeing each other, so CMC provides a supplemental communication option between them. For example, in the Student-Teacher On-line Mentoring Project (S.T.O.M.P.), participating at-risk students from low-income middle schools are paired with education majors from local colleges to read and discuss books face to face, but they also supplement their discussions with e-mail communication (Ensher et al., 2003). The Focus of E-Mentoring Programs A useful complement to Ensher et al.’s (2003) typology would be consideration of the program’s goal. Like traditional (FtF) mentoring programs, e-mentoring programs may aim to provide instrumental and/or relational support. In the field of youth mentoring, instrumental support refers to primacy being placed early in the relationship on teaching information and learning skills required for youth’s vocational development, while the relational support refers to placing primary emphasis early in the match on youth’s emotional well-being (see Karcher & Hansen, this volume, Chapter 5). Some e-mentoring programs prioritize one form of support, while other programs provide instrumental support as well as relational support (Burgstahler & Cronheim, 2001; Miller & Griffiths, 2005; Rhodes & DuBois, 2008). There are instrumental e-mentoring programs, also known as goal-directed programs (see Karcher

262  PROGRAMS AND CONTEXTS & Hansen, this volume, Chapter 5), which typically aim to provide vocational or career support, particularly online business coaching or training such as the University of Washington’s DO-IT (Disabilities, Opportunities, Interworking, & Technology). The DO-IT program supports the academics and careers of youth with disabilities through the use of computer and networking technologies (Darling, 2005). Other instrumental e-mentoring programs are educational or curriculum-based programs. Children are trained by an expert mentor such as an educator or a highly skilled professional who provides intellectual resources, support, and guidance on school topics (Miller & Griffiths, 2005). An example of such programs is the International Telementoring Project, which was established in 1994 and has become the largest e-mentoring program in the world. The program pairs K–12 public school students with professionals in technology companies to assist students with their class projects (Bennett, Hupert, Tsikalas, Meade, & Honey, 1998). While the instrumental programs focus primarily or initially on educational or vocational issues, the developmental e-mentoring programs focus primarily and initially on personal development and empowerment of the individual. Since the internet has become a social medium, online relationships conducted through e-mentoring programs can be similar to those developed in-person in terms of their breath, depth, and quality (Bargh & McKenna, 2004). In this sense, the online skills required from e-mentors are similar to ones required from e-counselors, such as demonstrating effective listening, providing guidance and feedback, and reinforcing mentees’ self-esteem (Barak, Hen, Boniel-Nissim, & Shapira, 2008). An example of developmental e-mentoring programs is the Virtual Mentoring Program, an Israeli e-mentoring program designed to provide socioemotional support for youth with a wide range of disabilities (physical, sensory, emotional, behavioral, or intellectual disabilities). The program connects young adult mentors with disabilities and teen-aged mentees with disabilities via e-mail. The mentors serve as adult role models for the mentees, sharing their life experiences and focusing on positive aspects of the disability (Shpigelman, Reiter, & Weiss, 2008; Shpigelman, Weiss, & Reiter, 2009). The e-mentoring programs discussed previously, and especially the developmental programs, may contribute more significantly to adolescents’ than to children’s lives. Cognitive abilities and expressive written communication skills—both vital in CMC environments—are much more limited among children than adolescents. In addition, children have had fewer opportunities to experience

serious risks that may provide fodder for the mentoring dyad to discuss, whereas adolescence is marked by coping with changes in the family and peer contexts and coping with moving from elementary to middle and high school. Throughout discussions with mentors about these changes, adolescents may be able to crystallize their personal identity and shift behavioral tendencies including beliefs, expectancies, goals, perceptions, and interpretations about themselves and others with whom they interact more than may be the case for children (Darling, 2005). A Proposed Conceptual Framework Based on a review of the literature, we have developed a conceptual framework that characterizes a successful e-mentoring process for youth aimed at providing primarily relational support (Shpigelman, Reiter, & Weiss, 2009). The Electronic Socio-Emotional Support (ESES) conceptual framework presents a model of essential elements in operating an e-mentoring program with youth (see Figure 17.1). Although it was developed for youth with disabilities, it has implications for providing e-mentoring to youth without disabilities (Shpigelman, Reiter, & Weiss, 2009). The ESES framework lists the interpersonal and programmatic preconditions required for effective e-mentoring, the forms of support that serve as mediators through which e-mentoring relationships develop, and the personalization processes that allow the e-mentoring relationship to foster youth’s self-development. The mentee’s drive for entering into an e-mentoring program may emanate from intrinsic and/or extrinsic sources (see Figure 17.1). Intrinsic and Extrinsic Motivation to Participate. Intrinsic sources of motivation are internal to the mentee, whereas an extrinsic drive to participate in a program is propelled by forces or factors outside the mentee. A mentee may seek electronic support on his own if he is sufficiently motivated and aware of its possibilities. This may occur when the mentee is self-motivated to explore career opportunities or seek guidance on academic projects (Asgari & O’Neill, 2004; DiRenzo, Linnehan, Shao, & Rosenberg, 2010). However, many at-risk youth may not be exposed to (and thus are unaware of) the potential benefits of e-mentoring for their personal development (Darling, 2005). In such cases, they may best be motivated by extrinsic sources, such as peers, family members, caregivers, or teachers. These sources may be necessary to encourage potential mentees and to expose them to the opportunities offered by e-mentoring (Shpigelman, Weiss,

Behaviors

Note: CMC = computer-mediated communication.

Source: Adapted from Shpigelman, Reiter, & Weiss (2009).

Emotions

Social Support

Tangible Assistance

Emotional Support

H

Personal Processes: Assimilation and Generalization

T

Health

Self-Empowerment

PA

Informational Support

Task-Based CMC ES ES P

S

H

AT

Online Communication

CMC and Mentoring Support

Self-Disclosure and Empathy

Building Trust

Coordination of Expectations

Preconditions: Hardware and Software Computer Literacy Mentoring Experience and Training Type and Quality of CMC Personality

Perceptions

Extrinsic

The Mentee's Motivation

Ongoing Activities

Demography

ES E

Intrinsic

Figure 17.1   The Electronic Socio-Emotional Support (ESES) Process

Moderators

Environment

263

264  PROGRAMS AND CONTEXTS & Reiter, 2009). Even when mentees are engaged in the e-mentoring program by extrinsic sources, they need to have a basic level of intrinsic drive to commence and especially to sustain involvement in the program. For the relationship to be maximally effective, intrinsic motivation must be cultivated if it is not already present (Spencer, 2007). Of course, mentoring relationships can be nurtured but should not be forced (Bierema & Merriam, 2002), and the relative contribution of the internal and external factors is dynamic, depending on the individual’s personality and supports in his or her environment. Preconditions. The next part of the ESES framework consists of the material preconditions needed to enable and facilitate the process of developing an e-mentoring relationship (see Figure 17.1). These preconditions include the availability of the hardware and software needed to make computer access possible for youth. Adapted hardware may be particularly important for youth with disabilities. Another precondition is ensuring that technical support is available. Mentors and mentees also must be computer literate (i.e., being adept at computer and internet usage), have sufficient communication skills to interact via the internet (such as reading comprehension and expressive written communication skills), and have the ability to find, evaluate, and use information technology (Bennett el al., 1998; Hoffman & Blake, 2003). According to the ESES framework, the development of successful e-mentoring relationships also depends on appropriate training for both mentees and mentors and the type and the quality of CMC (sufficiently frequent, two way, and task based) (Kasprisin, Single, Single, & Muller, 2003; Knight, Zheng, & Pipkin, 2009; Single & Single, 2005). Pretraining may enhance individuals’ confidence in their ability to participate in an e-mentoring program. Pretraining has considerable importance in electronic environments because it should focus on support and coping techniques unique to CMC, such as promoting mentors’ ability to handle actingout behavior and intense emotions expressed in the e-mail messages, and encouraging both mentors and mentees to communicate and express themselves (Burgstahler & Crawford, 2007; Shpigelman et al., 2008; Shpigelman, Weiss, & Reiter, 2009). When sufficient preconditions are in place, movement in the development of the mentor-mentee relationship proceeds down the ESES path (i.e., the e-mentoring process), as illustrated in Figure 17.1. Support and Activities. Several task-based CMC supportive behaviors contribute to strong e-mentoring relationships. The ESES framework is based on

Cutrona and Suhr’s (1992) Social Support Behavior Code that was originally developed for supportive behaviors occurring during FtF interactions. In the ESES framework, the mentoring relationship needs to provide (a) informational support (e.g., providing information about a stressful event or how to deal with it), (b) tangible assistance (e.g., providing or offering to provide goods or services needed in a stressful event), (c) social support (e.g., communicating belonging to a group of persons with similar interests and concerns), and (d) emotional support (e.g., communicating caring, respect, and confidence in abilities) (Shpigelman, Weiss, & Reiter, 2009). Each of these behaviors is influenced by factors such as the demographics, personalities, and health of those involved as well as the nature of the environment in which the interactions occur; these factors may moderate the effectiveness of each type of support on the desired programmatic outcomes of changes in selfperceptions, emotions, and behaviors. According to the ESES framework, an e-mentoring program should focus on ongoing interactions (i.e., task-based CMC) that reflect a defined task or area of learning that will make the electronic communication more authentic and attractive for youth (Miller & Griffiths, 2005). By engaging in cognitive tasks conducted via interactions with mentors, mentees may acquire and refine new thinking and skills, becoming more receptive to adult values, advice, and perspectives (Rhodes & DuBois, 2008). In addition, several ongoing activities are critical. First, mentors and mentees have to coordinate their ongoing expectations about their e-mentoring relationships. Second, providing ongoing relational support by building trust, self-disclosure, and empathy should be viewed as an essential component of online communication (Shpigelman, Weiss, & Reiter, 2009). In accordance with other FtF mentoring models (Buche, 2008; Keller, 2005), building trust, self-disclosure, and empathy should occur continuously. This may be obtained by online communication that includes visual and vocal cues. An online communication conducted via text chat or audio or video telephone (Skype or video conferencing) may enhance the social presence (i.e., the feeling of being together) of social interaction with a virtual or remotely located communication partner (Lombard & Ditton, 1997; Postmes, Spears, & Lea, 2002; Rao, Angelov, & Nov, 2006; Short, Ederyn, & Bruce, 1976; Tanis & Postmes, 2003). Personalization Processes. Successful movement down the ESES path is hypothesized to lead the mentoring relationship from ongoing, taskbased CMC activities to two intertwined personal

Electronic Mentoring and Media   265 processes—assimilation and generalization—which are what make the mentoring relationship valuable beyond the computer-mediated environment itself. During the e-mentoring relationship, the mentee is encouraged to synthesize information, thoughts, and feelings experienced via the electronic process and to develop safe and healthy behaviors (Shpigelman, Reiter, & Weiss, 2009; Shpigelman, Weiss, & Reiter, 2009). This may occur through the previously described ongoing activities and electronic communication that focus directly or indirectly on issues that concern youth, such as their social life, relationships in the family, studies, career, travel, and other issues raised by the mentee. It may also occur through the feeling of social presence that can be facilitated by adding visual and auditory graphic icons, such as emoticons (J or L), to the electronic correspondence. These emoticons are designed to supplement a text message with emotional content (Barnes, 2003; Leh, 2001). Ideally, the ESES process will result in the mentee’s perceptual, emotional, and behavioral self-empowerment; in other words, the mentee acquires or strengthens the necessary psychological resources, such as self-efficacy, self-esteem, identity, and interpersonal skills that will enable goal achievement (Amichai-Hamburger, McKenna, & Tal, 2008). To summarize, the process described in the ESES framework commences when a mentee seeks electronic support. Certain material preconditions, such as hardware and software, computer literacy skills, and mentoring training, enable and facilitate the process. The mentor may supply various types of support that, in the ESES framework, are hypothesized to result in positive self-perception, emotional, and behavioral changes in the mentee. The next section reviews the findings of studies that have explored the feasibility and utility of the framework’s components, or that either support or challenge this framework’s viability.

Research My research search strategy included narrow keywords such as e-mentoring, electronic mentoring, online mentoring, virtual mentoring and telementoring, as well as general keywords related to the e-mentoring field (i.e., computer-mediated communication and online support). The search was conducted on three major databases: ERIC, PsycINFO, and ProQuest. The studies included in this review present the outcomes of youth e-mentoring programs, both instrumental and relational. Few peer-reviewed studies have been published on e-mentoring for youth

(Miller & Griffiths, 2005). In accordance with my extension of adding instrumental and developmental program types to Ensher et al.’s (2003) typology of e-mentoring programs—as CMC-only, CMC-primary, and CMC-supplemental—I present studies that explored instrumental and developmental e-mentoring programs separately. Instrumental E-mentoring Programs O’Neill and Gomez (1998) examined the ability of an instrumental e-mentoring program, the Computer Supported Cooperative Work (CSCW) system, to sustain long-term online relationships. The researchers analyzed qualitative data on 26 lengthy e-mentoring relationships between students in grades 7–12 and volunteer scientists who advised the youth on science projects. The findings indicated that utility is a major sustaining force for both the mentees and the mentors. On the one hand, the mentees need to feel that they are benefiting from the e-mentoring relationships (i.e., that the relationships have productive utility for them). On the other hand, the mentors need to perceive the relationships as easily manageable (e.g., that the e-mentoring relationship requires only a few minutes of time invested each week) and beneficial (e.g., that it gives volunteer mentors opportunities to break from their routine work and contribute to the future of their chosen field). The researchers emphasized the importance of increasing the visibility of e-mentoring relationships for the volunteer mentors through the use of web forums that present the problems of the mentees, the advice provided by the mentors, and the success of the mentees afterward. Although this research focused primarily on sustaining and promoting satisfaction of volunteer mentors, it may be useful for strengthening programs for mentees. The visibility of e-mentoring relationships through web forums may motivate mentors and mentees to engage in and sustain e-mentoring relationships, by bolstering extrinsic motivation, which is the starting point in the proposed ESES framework described earlier. DiRenzo et al. (2010) surveyed 1,381 lowerincome middle school and high school students who participated in I Could Be (ICB), an instrumental e-mentoring program that is based on a commitment of one academic year and that focuses on professional development of youth. The researchers examined the relationship between mentoring outcomes (such as program satisfaction, and general, career, and task self-efficacy) and the mentees’ preprogram attitudes and experiences. The results indicated a positive relationship between a mentee’s prior internet experience and motivation to achieve

266  PROGRAMS AND CONTEXTS the program outcomes of general self-efficacy and task-efficacy. That is, motivated mentees with greater internet experience prior to entering an e-mentoring program were more apt to seek their mentors’ guidance, to perceive the relationship as being easy to maintain, and to feel the technology is easy to utilize. This finding underscores the importance of mentees’ intrinsic motivation and their need to be computer literate, both preconditions proposed in the ESES framework. O’Neill, Asgari, and Dong (2011) explored the meaning of success in online mentoring (i.e., how to evaluate and measure the success). The researchers analyzed survey and interview data from an online mentoring program in which 17 history experts supported 108 high school students in historical research. They found that an online mentoring process requires trade-offs between mentees’ satisfaction or perceived success and the intended outcomes of the program. Online mentoring demands a working understanding of how the various parts of the program interact. This trade-off perspective affirms the importance of the ESES framework component related to the ongoing coordination of expectations. Thomson (2009) conducted another study that supports the potential of instrumental e-mentoring for youth at-risk academically. She interviewed five former students who participated in an e-mentoring program for at-risk high school students who met with local business mentors. The interviewed mentees were 18 years of age or older, participated in the program at least 1 year, and were perceived by the school to have made a successful transition to adulthood in terms of enrollment in postsecondary education or full-time employment, with no involvement with the judicial system. Qualitative content analysis revealed that the interviewed youth felt their participation in the e-mentoring program helped them to achieve their goals of successfully graduating from high school. Several mentees reported that their mentors served as a motivator for them when their own parents’ support was absent. The findings indicated that e-mentoring programs may promote not only the professional development of at-risk youth but also their emotional well-being. Consistent with the ESES framework, the components of e-mentoring that were found to be most effective were building a relationship of trust and conducting FtF meetings during the program. Another study, conducted in the United Kingdom (Liu, Macintyre, & Ferguson, 2012), found that allowing a range of communication channels was an important factor in the success of e-mentoring. The researchers used web 2.0 technology—a platform

called SocialLearn—to develop an e-mentoring service for strengthening the connections between students and staff members (mentors) in a distance learning setting through the Open University. They conducted a 6-month pilot study with 12 staff members who used this new online social learning platform. Thematic analysis was applied to screen captures, audio scripts, and interviews. The researchers found that the new platform had the potential to support more diffuse relationships that would support a sense of community and “generalized reciprocity.” During the pilot, the users completed their profiles and quickly began to establish connections; they then used those connections to locate and make other connections. The users also wanted to be able to differentiate between varied types of connection (e.g., instant messages and social network websites). The researchers noted that users needed to be able to see others (through their activities) and to be able to understand and interpret what others were doing. Murphy (2011) also found that blended mentoring (e-mail plus talking on the phone or meeting face to face) increased positive outcomes for both students and mentors. The findings of both studies provide further support for the need for the social presence obtained through FtF meetings and online communication in e-mentoring programs, as indicated in the ESES framework. Developmental E-Mentoring With regard to developmental e-mentoring, few programs have been designed especially for the socio-emotional support of youth. Accordingly, only a few peer-reviewed studies have been published on developmental e-mentoring programs. An example of such a program may be the Digital Heroes Campaign (DHC), as shown in Table 17.1. In 2000–2002, 242 youth aged 13–17 years were paired with mentors with whom they were to correspond exclusively via e-mail for a minimum of 6 months. Rhodes et al. (2006) assessed the nature, types, and quality of the online relationships developed during DHC using a qualitative methodology in which they conducted a narrative analysis of the content of the messages. In addition, two independent coders rated the quality of the messages based on a numeric scale set up by the researchers. Furthermore, interviews were conducted with a subsample of the mentors and with the supervisors. The online relationships tended to be of relatively high quality, in that they were youth centered and both the mentors and the mentees seemed emotionally engaged and satisfied with the relationships. This finding supports the

Electronic Mentoring and Media   267 need to provide ongoing relational support by building trust, self-disclosure, and empathy, as indicated in the ESES framework. About half of the online relationships broke up before the end of their 6-month commitment period. The optimal duration for conducting a successful e-mentoring program for youth has not been included in the ESES framework and should be explored further. The finding that e-mentoring relationships can reflect depth, be satisfying, and contain some of the same ingredients of FtF relationships is supported, in part, by studies that explored the use of the ESES framework in the Israeli e-mentoring program for youth with disabilities (Shpigelman et al., 2008; Shpigelman, Reiter, & Weiss, 2009; Shpigelman, Weiss, & Reiter, 2009). The studies included qualitative content analysis of 860 electronic correspondences of 13 mentor-mentee pairs and 18 responses to post-program evaluation questionnaires. Three evaluators conducted a quantitative analysis by rating a sample of the e-mail messages. The results indicated that e-mentoring relationships can become positive and supportive relationships. The majority of the mentors and the mentees evaluated the e-mentoring program to be an enjoyable experience. They were satisfied with the experience and the opportunity to assist others. Questionnaire responses suggest that some participants felt that communicating via e-mail appeared to reduce the visibility of their disability, which enabled them to speak about life experiences (e.g., family, friends, hobbies, studies) and to help them feel that they share some characteristics with typical youth. However, some of the participants, especially the mentees, had a harder time establishing a feeling of presence in the computer-mediated environment and felt that they needed more FtF communication to develop a deeper relationship. This finding supports the inclusion of FtF meetings as well as online communication in e-mentoring programs, as found in previous studies and indicated in the ESES framework (Buche, 2008; Lenear, 2006; Liu et al., 2012; Thomson, 2009). Conducting FtF meetings in addition to the online communication may be vital to enhance a feeling of presence in online environments. Other studies support the inclusion in the ESES framework of preconditions and ongoing training (Kasprisin et al., 2003; Single & Single, 2005). Due to the unique features of e-mentoring relationships, such as corresponding in a more spontaneous fashion and removing barriers created by first impressions (Barnes, 2003; Miller & Griffiths, 2005; Rhodes, 2003; Rhodes et al., 2006), training for both mentors and mentees appears critical to promote successful outcomes of e-mentoring programs.

Kasprisin et al. (2003) explored the impact of a series of interactive, web-based case studies that were developed as training modules for mentors and mentees in MentorNet. Using a control group experimental design, they randomly assigned 200 participants to a condition in which interactive online training was required. The other 200 were assigned to a condition in which the training was optional. The mentees in the mandatory group exhibited improved outcomes in terms of their involvement, which means that the trained mentees exchanged e-mail messages more frequently with their e-mentors. Engaging in one type of training versus the other was not significantly related to the other two variables examined in the study (i.e., overall satisfaction and perceived value of participation in the e-mentoring program). However, the study results, especially those related to the involvement rate, were limited by the constraints of selfreported questionnaires, as also constrained the study by DiRenzo et al. (2010). By contrast, a study conducted by Culpepper (2008) indicated that the implementation of a structured e-mentoring model did not have a statistically significant positive effect on at-risk students’ psychological (self-esteem, career indecision), behavior (attendance), or academic success (achievement). The participants in Culpepper’s study were students enrolled in the GED Exit Option program, a dropout prevention program, as described in Table 17.1. The participants were 32 mentors and 91 youth randomly selected to the control group (nonmentored, n = 45) or the mentored group (n = 26) for a period of about 6 months. The researchers analyzed quantitative and qualitative online surveys, including the GED test scores. They found no significant differences between the groups on psychological and behavior success or on academic success (GED test scores). The study included the three major phases for conducting a successful e-mentoring program described in Single and Muller’s (1999) Model of Structured E-Mentoring (Single & Muller, 1999): planning, program structure, and assessment. Although one program structure, the mentor-mentee matching, was done randomly, this is an uncommon practice. More important, the insufficiently small sample size for the study resulted in underpowered analyses that greatly increased the risk of not finding a program effect had one been present. In conclusion, most of the studies described here focused on specific, formal e-mentoring programs for at-risk adolescents, conducted in educational or professional settings over a period of 6–12 months, and the programs have been instrumental rather than developmental (DiRenzo et al., 2010;

268  PROGRAMS AND CONTEXTS O’Neill & Gomez, 1998; Thomson, 2009). Most reports included relatively small samples, with theoretical reviews or anecdotal reports being more common (Miller & Griffiths, 2005). Few of the studies included comparison groups (Culpepper, 2008; Kasprisin et al., 2003) or used both quantitative and qualitative research methods (Culpepper, 2008; Shpigelman et al., 2008; Shpigelman, Weiss, & Reiter, 2009). Furthermore, most of the research methods involved self-report measures, which could be problematic because the participants might have responded in a socially desirable manner (Culpepper, 2008; DiRenzo et al., 2010; Kasprisin et al., 2003). Given the state of research on e-mentoring to date, needed are explorations of the effectiveness of e-mentoring programs using structured frameworks and models to test hypotheses about the essential components of e-mentoring. The importance of key moderators in the ESES framework (i.e., demographic, personality, health, and environmental factors); specific mediators of program outcomes, such as the level of instrumental and relational support; and the effect of dyad and group e-mentoring on program outcomes should be explored further (DiRenzo et al., 2010; Lenear, 2006). In addition, researchers should include larger samples of youth (given the 50% dropout rate that has been reported during some e-mentoring program) and explore not only the outcomes of e-mentoring but also the dropout rates and reasons for dropping out of e-mentoring programs. With regard to research methods, both qualitative and quantitative methods should be used in order to yield insights that go beyond primarily theoretical reviews and to overcome the socially desirable response problem associated with self-report questionnaires. A holistic approach is recommended, which means including questionnaires that will filled out—before and after program participation— by mentees, mentors, teachers, coordinators, and other related individuals; conducting observations on the process itself; and collecting and analyzing the electronic correspondences. Finally, future research should explore the optimal duration of an e-mentoring process; the importance and elements of training; the need for FtF contact; and the moderating effects of youth characteristics such as different computer skills, gender, personalities, and needs.

Practice Key Challenges of E-Mentoring Beyond the potential benefits, e-mentoring faces a number of key challenges. When using text-only

communication, the recipient remains unaware of the sender’s emotions unless the sender expresses them explicitly within the text. The attainment of social presence is important for the development of successful e-mentoring relationships, as indicated in multiple studies (O’Neill & Gomez, 1998; Shpigelman et al., 2008; Shpigelman, Weiss, & Reiter, 2009; Thomson, 2009). Another key challenge described in the literature is skill deficiency, including language limitations such as reading comprehension and expressive written communication, and also lack of computer and internet usage skills and typing skills. These limitations may lead to misreading or misunderstanding of the electronic correspondence and yield an ineffective e-mentoring process. This problem has been reported in mentoring of adults (Cravens, 2003; Ensher et al., 2003; Mallen et al., 2003) as well as in youth mentoring (DiRenzo et al., 2010; Shpigelman et al., 2008; Shpigelman, Weiss, & Reiter, 2009). The literature indicates that conducting CMConly e-mentoring programs may be difficult due to the possibility of technology failures (computer, software, or internet) and time delays that may lead to feelings of abandonment, disillusion, panic, and frustration for both mentors and mentees (Ensher et al., 2003; Mallen et al., 2003). In this sense, the literature indicates also that a development of meaningful e-mentoring relationships requires a level of maturity. Emotional difficulty with making personal disclosures, sharing emotions online, and writing about problems and failures, as well as successes, may preclude the development of close online relationships and may result in superficial or purely informational communication (Cravens, 2003; Darling, 2005; Ensher et al., 2003). Coping With the Challenges To overcome these challenges, selecting the type of e-mentoring program to be conducted is critical. Ensher et al. (2003) suggest deciding whether the program will be CMC-only, CMCprimary, or CMC-supplemental. Then it is important to establish an on-site coordinator at each school or organization that will conduct the e-mentoring program, including administration, technical support, and online facilitation required to sustain online relationships (O’Neill & Gomez, 1998). Mentor and mentee recruitment, pretraining, and matching; ongoing support; and online facilitation of the mentor-mentee relationships are key components that require time and attention (Collier, 2009). The coordinator has to guide the mentors and inform them of techniques they can use to overcome the

Electronic Mentoring and Media   269 challenges of e-mentoring, such as using emoticons, synchronous text, audio or video online communication, as well as conducting FtF meetings. However, the literature suggests that online relationships are different from traditional relationships and that sustaining online relationships over time is much more difficult. Therefore, it is critical to engage the participants in online and youth-centered communication. This should start by matching mentor-mentee pairs on the basis of mutual interests, and perhaps consider giving the participants the opportunity to choose their own partners (Collier, 2009; Kendall, 2007). After the matching, the participants should be engaged in online task-based activities, such as those proposed in the ESES framework or revealed in research on promising programs (DiRenzo et al., 2010; O’Neill & Gomez, 1998; Rhodes et al., 2006). To date, there are no definitive data on what activities are most effective. According to Lenear (2006), integrating youth e-mentoring programs into an existing structured program, such as a school-related program, may promote more frequent communication by mentees, because teachers or other assistants are available to motivate and help the mentees.

Time is another critical factor. Rhodes et al. (2006) found that about half of the online relationships in their study broke up before the end of the 6-month commitment period. Research on another e-mentoring program that ran for one academic year found that one year was too long a period to sustain frequent communication (Shpigelman et al., 2008; Shpigelman, Weiss, & Reiter, 2009). On the other hand, research on short-term e-mentoring programs (i.e., 3 months) has not produced meaningful or consistent results (Collier, 2009). These researchers recommend that e-mentoring programs be conducted over a period of 6–12 months. However, the question of the optimal time should be explored further. To summarize, based on a review of the literature and research, to implement a successful e-mentoring program, it is recommended that each phase of the program be considered. Consistent with traditional FtF mentoring, the early phase includes establishing program goals, recruiting participants, and matching pairs or groups, whereas later phases include providing ongoing support and, finally, evaluating the program’s implementation and outcomes. Other factors appear to be uniquely critical for e-mentoring, as shown in Table 17.2. These include selecting

Table 17.2   Checklist for Practitioners Topic

Recommendations

Establish goals

Define a clear purpose for the program and specific tasks for the electronic exchange.

Select technology

Select appropriate technology to use to support the e-mentoring process, such as email, electronic distribution lists, web forums, real-time chat (with or without webcam).

Develop the communication structure

Develop other communication channels, such as separate online groups for the mentors and the mentees so that they will be able to share experiences with their peers.

Develop guidelines and tutorials

Develop procedural and behavioral guidelines for mentors, mentees, teachers, and parents. The guidelines should be simple and straightforward and help participants understand their roles and responsibilities in the program. The materials should be distributed to the participants when they first enter the program and also periodically throughout the program.

Guide participants

Provide orientation and pretraining for mentors and mentees. Pretraining must include coping techniques via CMC—such as the ability to handle acting-out behavior and intensity of emotions as expressed in the email messages—and encouraging participants to communicate, ask appropriate questions, and express themselves.

Provide ongoing support

Communicate with each participant on a regular basis in order to achieve a realistic understanding of expectations, to answer questions and to identify solutions to any difficulties that the participant may be facing, and to ensure sufficient two-way electronic communication.

Manage online activities

Monitor and promote active involvement of the participants, such as by sending discussion questions to the group or privately encouraging individual mentors or mentees to participate in the e-mentoring process.

270  PROGRAMS AND CONTEXTS appropriate and accessible technology, including the development of various communication channels; supplying tutorials and conducting pretraining for both mentors and mentees; and, finally, engaging the participants by online task-based activities.

Conclusion The field of e-mentoring for at-risk youth is relatively nascent. However, the studies that have emerged, especially in the past 5 years, have supported the considerable potential of e-mentoring for personal growth and empowerment of children and especially adolescents. The e-mentoring typologies, framework, and studies highlighted in this chapter reveal the unique features of e-mentoring, such as how it provides freedom from the restrictions of location and time and allows greater access to diverse populations. In addition, the literature and existing preliminary research suggest that negative feelings related to one’s physical appearance (e.g., overweight, disability) and vocal characteristics (e.g., stutter) as well as social skill deficits may be lessened in e-mentoring programs (Collier, 2009; Scealy, Phillips, & Stevenson, 2002; Shpigleman et al., 2008; Shpigelman, Weiss, & Reiter, 2009). However, not every youth is equally able to participate successfully in an e-mentoring program. The ESES framework, which is based on research, suggests that e-mentoring is suited for youth who are computer literate (i.e., have good computer and internet usage skills), who have expressive written communication skills, including the creative independence it takes to hold up their end of the written dialogue, and who have a basic level of intrinsic motivation to join an e-mentoring program (DiRenzo et al., 2010; Hoffman & Blake, 2003). Therefore, it may not work for all youth, although it may provide a unique adjunct for at-risk youth whose difficulties may interfere with the processes of FtF mentoring. Future technological developments, such as multidimensional environments, also may enhance the expansion of the e-mentoring phenomenon. It is hoped that this growth in technology will be coupled by growth in the knowledge base about best practices in e-mentoring so that its full potential may be reached.

References Amichai-Hamburger, Y., McKenna, K. Y. A., & Tal, S. A. (2008). E-empowerment: Empowerment by the internet. Computers in Human Behavior, 24, 1776–1789.

Asgari, M., & O’Neill, D. K. (2004). What do they mean by “success”? Contributions to perceived success in a telementoring program for adolescents. Paper presented at the American Educational Research Association, San Diego, CA. Barak, A., Hen, L., Boniel-Nissim, M., & Shapira, N. (2008). A comprehensive review and a meta-analysis of the effectiveness of Internet-based psychotherapeutic interventions. Journal of Technology in Human Services, 26, 109–160. Bargh, J. A., & McKenna, K. Y. A. (2004). The internet and social life. Annual Review of Psychology, 55, 573–590. Barnes, S. B. (2003). Computer-mediated communication: Human-to-human communication across the internet. Boston: Allyn & Bacon. Benford, P. (2008). The use of internet-based communication by people with autism. Unpublished PhD. thesis, University of Nottingham, UK. Bennett, D., Hupert, N., Tsikalas, K., Meade, T., & Honey, M. (1998). Critical issues in the design and implementation of telementoring environments. Center for Children and Technology. Retrieved from http://cct .edc.org/admin/publications/report/09_1998.pdf Bierema, L. L., & Merriam, S. B. (2002). E-mentoring: Using computer mediated communication to enhance the mentoring process. Innovative Higher Education, 26, 211–227. Buche, M. W. (2008). Development of trust in electronic mentoring relationships. International Journal of Networking and Virtual Organizations, 5, 35–50. Burgstahler, S., & Crawford, L. (2007). Managing an e-mentoring community to support students with disabilities: A case study. Association for the Advancement of Computing in Education Journal, 15(2), 97–114. Burgstahler, S., & Cronheim, D. (2001). Supporting peerpeer and mentor-mentee relationships on the internet. Journal of Research on Technology in Education, 34(1), 59–74. Collier, M. L. (2009). An investigative study of the effects of e-mentoring on transition planning for secondary students with learning disabilities. Unpublished PhD thesis, Department of Special Education, University of Utah. Cravens, J. (2003). Online mentoring: Programs and suggested practices as of February 2001. Journal of Technology in Human Services, 21, 85–109. Culpepper, D. W. (2008). Determining the quality and impact of an e-mentoring model on at-risk youth. Unpublished dissertation, University of South Florida. Retrieved from http://scholarcommons.usf.edu/etd/198 Cutrona, C. A., & Suhr, J. A. (1992). Controllability of stressful events and satisfaction with spouse support behaviors. Communication Research, 19, 154–174. Darling, N. (2005). Mentoring adolescents. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 177–190). Thousand Oaks, CA: Sage. DiRenzo, M. S., Linnehan, F., Shao, P., & Rosenberg, W. L. (2010). A moderated mediation model of e-mentoring. Journal of Vocational Behavior, 76, 292–305.

Electronic Mentoring and Media   271 DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. Ensher, E. A., Heun, C., & Blanchard, A. (2003). Online mentoring and computer-mediated communication: New directions in research. Journal of Vocational Behavior, 63, 264–288. Eröz-Tuğa, B., & Sadler, R. (2009). Comparing six video chat tools: A critical evaluation by language teachers. Computers & Education, 53, 787–798. Hamilton, B. A., & Scandura, T. A. (2003). E-mentoring: Implications for organizational learning and development in a wired world. Organizational Dynamics, 31, 388–402. Harris, J. B., & Figg, C. (2000). Participating from the sidelines, online: Facilitating telementoring projects. ACM Journal of Computer Documentation, 24, 227–236. Harris, J., Rotenberg, L., & O’Brysan, E. (1997). Results from the Electronic Emissary Project: Telementoring lessons and examples. Denton, TX: Texas Center for Educational Technology. Retrieved from http:// www.tcet.unt.edu/pubs/em/em01.pdf Hoffman, M., & Blake, J. (2003). Computer literacy: Today and tomorrow. Journal of Computing Sciences in Colleges, 18(5), 221–233. Kamel Boulos, M. N., Hetherington, L., & Wheeler, S. (2007). Second Life: An overview of the potential of 3-D virtual worlds in medical and health education. Health Information & Libraries Journal, 24(4), 233–245. Kasprisin, C. A., Single, P. B., Single, R. M., & Muller, C. B. (2003). Building a better bridge: Testing e-training to improve e-mentoring programmes in higher education. Mentoring & Tutoring, 11, 67–78. Keller, T. E. (2005). The stage and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook for youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Kendall, D. L. (2007). Does choice matter? The impact of allowing mentees to select their own mentors. Unpublished PhD thesis, Department of Psychology, University of Central Florida. Knight, P., Zheng, W., & Pipkin, C. (2009). An investigation of electronic mentoring practice (Mentor Kenosha & Racine Report). School of Business and Technology, University of Wisconsin-Parkside. Retrieved from http://www.uwp.edu/departments/community.part nerships/documents/E-Mentoring_ReportZhengKnightmg.pdf Leh, A. S. C. (2001). Computer-mediated communication and social presence in a distance learning environment. International Journal of Educational Tele­ communications, 7(2), 109–128. Lenear, P. E. (2006). The effect of an internet-based mentoring program on the transnational distance and interaction between mentors and mentees. Unpublished doctoral thesis, Department of Education, University of Illinois at Urbana-Champaign.

Liu, H., Macintyre, R., & Ferguson, R. (2012). Exploring qualitative analytics for e-mentoring relationships building in an online social learning environment. Proceedings of the 2nd International Conference on Learning Analytics & Knowledge, Vancouver, BC, Canada. Lombard, M., & Ditton, T. (1997). At the heart of it all: The concept of presence. Journal of Computer Mediated Communication, 3(2). Retrieved from http://jcmc.indiana.edu/vol3/issue2/lombard.html Mallen, M. J., Day, S. X., & Green, M. A. (2003). Online versus face-to-face conversation: An examination of relational and discourse variables. Psychotherapy: Theory, Research, Practice, Training, 40, 155–163. MentorNet. (2008). Students’ perceptions of the value and need for mentors as they progress through academic studies in engineering and science (A Report to the National Science Foundation Concerning a Small Grant for Exploratory Research (SGER)). San Jose, CA: Author. Retrieved from http://www.mentornet .net/documents/files/evaluation/studentperceptions .completereport.pdf Miller, H., & Griffiths, M. (2005). E-Mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 300–313). Thousand Oaks, CA: Sage. Murphy, W. M. (2011). From e-mentoring to blended mentoring: Increasing students’ developmental initiation and mentors’ satisfaction. Academy of Management Learning & Education, 10, 606–622. O’Neill, D. K. (2011). Local social capital through e-mentoring: An agenda for new research. Learning, Media and Technology, 36, 315–321. O’Neill, D. K., Asgari, M., & Dong, Y. R. (2011). Tradeoffs between perceptions of success and planned outcomes in an online mentoring program. Men­toring & Tutoring: Partnership in Learning, 19, 45–63. O’Neill, D. K., & Gomez, L. M. (1998). Sustaining mentoring relationships on-line. In S. Greenberg & C. Neuwirth (Eds.), Proceedings of CSCW 98: ACM conference on computer-supported cooperative work (pp. 325–334). New York: Association for Computing Machinery. Postmes, T., Spears, R., & Lea, M. (2002). Intergroup differentiation in computer-mediated communication: Effects of depersonalization. Group Dynamics, 6, 3–16. Putnam, R. (1995). Tuning in, tuning out: The strange disappearance of social capital in America. Political Science and Politics, 28, 664–683. Rao, B., Angelov, B., & Nov, O. (2006). Fusion of disruptive technologies: Lessons from the Skype case. European Management Journal, 24, 174–188. Rhodes, J. E. (2003). Online mentoring: The promise and pitfalls of an emerging approach. Retrieved from http://www.mentoring.org/research_corner/11_03_ online.adp Rhodes, J. E., & DuBois, D. L. (2008). Mentoring relationships and programs for youth. Current Directions in Psychological Science, 17, 254–258.

272  PROGRAMS AND CONTEXTS Rhodes, J. E., Spencer, R., Saito, R. N., & Sipe, C. L. (2006). Online mentoring: The promise and challenges of an emerging approach to youth development. Journal of Primary Prevention, 27, 497–513. Scealy, M., Phillips, J., & Stevenson, R. (2002). Shyness and anxiety as predicators of patterns of internet usage. CyberPsychology and Behavior, 5, 507–515. Short, J. A., Ederyn, W., & Bruce, C. (1976). The social psychology of telecommunication. New York: Wiley. Shpigelman, C. N., Reiter, S., & Weiss, P. L. (2008). E-mentoring for youth with special needs: Prelimi­nary results. CyberPsychology & Behavior, 11, 196–200. Shpigelman, C. N., Reiter, S., & Weiss, P. L. (2009). A conceptual framework for electronic socio-emotional social support for people with special needs. International Journal of Rehabilitation Research, 32, 301–308. Shpigelman, C. N., Weiss, P. L., & Reiter, S. (2009). E-mentoring for all. Computers in Human Behavior, 25, 919–928.

Single, P. B., & Muller, C. B. (1999, April). Electronic mentoring: Issues to advance research and practice. Paper presented at the 12th annual meeting of the International Mentoring Association, Atlanta, GA. Retrieved from http://www.mentornet.net/Docu ments/Files/EmentoringIssues.doc Single, P. B., & Single, R. M. (2005). E-mentoring for social equity: Review of research to inform program development. Mentoring & Tutoring, 13, 301–320. Spencer, R. (2007). “It’s not what I expected” - A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331-354. Tanis, M., & Postmes, T. (2003). Social cues and impression formation in CMC. Journal of Communication, 53, 676–693. Thomson, E. (2009). The perceived impact of an e-mentoring program on at-risk students’ successful transition to adulthood: A multiple case study. Doctoral dissertation, School of Education, Drake University.

18  GROUP MENTORING Gabriel P. Kuperminc and Jessica D. Thomason

Introduction According to a recent national poll, nearly 3 million adults are involved in formal and sustained one-onone relationships with young people (MENTOR, 2006). Notably, more than half (51%) of the adult respondents to that poll reported that they worked with more than one youth. Moreover, Sipe and Roder (1999) estimated that 20% of youth mentoring organizations employ group mentoring for at least part of their program offerings. Multiple young people engaged in sustained relationships with one or more mentors represents one of the most common formats in which youth mentoring occurs. Research in the area, however, has focused on one-on-one mentoring; little is known about how group mentoring compares to one-on-one mentoring, and there is little understanding of its effectiveness (DuBois & Karcher, 2005; Karcher, Kuperminc, Portwood, Sipe, & Taylor, 2006). In this chapter, we outline a practical and theoretical framework for understanding group and team mentoring. We define group and team mentoring and consider theoretical perspectives from group psychotherapy and mutual help interventions, as well as perspectives on the developmental significance of peer and adult relationships. We next review research on both informal and formal group mentoring, paying particular attention to evidence of overall effectiveness and to research that addresses our theoretical framework. We then consider implications for practice. The chapter concludes with some broader observations about future prospects and priorities for group mentoring research and practice. Despite the popularity of one-on-one mentoring, researchers and practitioners have questioned its ability to reach large numbers of youth. Specifically,

Herrera, Vang, and Gale (2002) noted that a group format is preferred by many adults and youth who may feel put off by the one-on-one format. The group format may be more culturally appropriate for many ethnic minority groups (Lindsay-Dennis, Cummings, & McClendon, 2011; Utsey, Howard, & Williams, 2003). Consistent with this notion, Herrera et al. (2002) found higher proportions of ethnic minority mentors choosing to participate in group mentoring as compared to the one-on-one format. Group mentoring may also offer a more cost-effective alternative to traditional one-on-one dyadic mentoring relationship programs (Herrera et al., 2002). Beyond these practical considerations, it is reasonable to ask whether we can honestly expect group mentoring to have positive impacts in the lives of young participants comparable to those observed in the more traditional one-on-one format. Critics have questioned whether group members experience the same intensity and quality of relationship with a mentor that is thought to be at the heart of mentoring (e.g., Rhodes, 2002). Others caution that group interventions may cause more harm than good as a result of peer pressure toward increasing negative or antisocial behavior (e.g., Dishion, McCord, & Poulin, 1999). In response, we feel a conceptual and theoretical case for effective group mentoring should be made to address these concerns.

Theory Group mentoring has not been defined consistently, in part reflecting the wide variation in how it is implemented (Karcher et al., 2006). The most recent edition of the Elements of Effective Practice for Mentoring (MENTOR, 2009, p. 25) differentiates 273

274  PROGRAMS AND CONTEXTS group mentoring (one mentor working with up to four young people) from team mentoring (several adults working with small groups of young people, in which the adult-to-youth ratio is not greater than 1:4). In practice, the number of youth in a group can vary from as few as 2 to as many as 32, and groups may include a single mentor, or two or more mentors working in teams (Herrera et al., 2002). Groups may also be composed of multiple mentor-mentee pairs who come together for some activities (e.g., Deutsch, Henneberger, Wiggins, & Lawrence, 2010). Philip and Hendry (2000) identified “individual-team mentoring” as one of five types of informal mentoring and did not differentiate group and team mentoring. Beyond that study, empirical research has not yet examined how mentoring processes or outcomes differ as a function of the number of mentors present or the ratio of mentors to youth. Thus, although we appreciate the importance many people working in the field place on this distinction, we regard team mentoring as a special case of the more general category of group mentoring. Throughout the chapter we use the term group mentoring to refer to either “natural” or programmatic mentoring contexts in which one or more mentors work with at least two youth. With regard to the mentor-to-youth ratio, we would agree that ratios much larger than 1:4 risk becoming counterproductive, but argue that research is needed to provide better guidance on this issue (see also Karcher et al., 2006). Regardless of the specific format, we argue that group mentoring must involve an intentional focus on interpersonal relationships and incorporate the core elements of effective youth mentoring relationships identified by DuBois and Karcher (2005): mentor(s) with greater experience, offering guidance intended to facilitate growth and development of mentees, and the development of an emotional bond between mentor and mentees. Group mentoring is differentiated from other group interventions (e.g., group psychotherapy) based on its emphasis on relationship building and group processes as a primary means of targeting developmental achievements among group participants. Our conceptualization begins with the idea that the group structure may offer some unique advantages (Karcher et al., 2006). Consider the interactions that are likely to occur in group mentoring. Similar to one-on-one mentoring, youth in group mentoring will experience direct interaction with a mentor (or mentors), although at times that interaction may be shared with other group members (see Figure 18.1, regions A and B). Indeed, Herrera et al. (2002) found that mentees reported less relationship intensity with mentors in group as

compared to one-on-one mentoring. However, the effect of this reduced intensity may be offset by experiences in other types of interactions. Specifically, youth are likely to interact directly with group members (region C) and to observe interactions between mentor(s) and other group members in which they are not directly involved (region D). In the case of team mentoring, youth may also have the opportunity to observe interactions among multiple mentors. A comprehensive theoretical model for group mentoring must take these types of interactions into account, addressing their developmental significance, as well as their potential for contributing to or detracting from positive developmental outcomes. Our theoretical perspective on group mentoring draws from conceptual and empirical work in group psychotherapy and mutual help interventions (Barrera & Prelow, 2000; Cox, Vinogradov, & Yalom, 2005). We also incorporate perspectives on the developmental significance of relationships with peers and adults (Hartup, 1989). Insights From Group Psychotherapy Current theories of group psychotherapy assume that interpersonal relationships are a major driver of human psychological development, and argue that the group setting offers a specific therapeutic context in which individuals “are provided with a varied array of interpersonal relationships, that with proper guidance, will permit them to identify, explore, and alter maladaptive interpersonal behavior” (Cox et al., 2005, p. 1329). Group therapy emphasizes interpersonal learning: The group is seen as a social microcosm that provides individuals with opportunities to try out new behaviors, to observe and learn from others’ (effective and ineffective) behavior, and to receive feedback from peers and the therapist. Patients gain a sense of hope through watching others set goals and improve, and learn that others share many of their own concerns and experiences. Dies (2000) and Richmond (2000) argued that group therapeutic processes are particularly pertinent to working with children and adolescents as they navigate the developmental challenges of managing peer relationships and establishing a sense of independence. A common theme throughout the theoretical and empirical work on group psychotherapy is an emphasis on the role of the therapist as a visionary leader who is skilled in group dynamics and conflict resolution, and who is able to manage multiple roles as authority figure and friend, strict enforcer of rules, and trusted coach. A rich empirical literature supports the efficacy of group treatments for children and adolescents,

Group Mentoring  275 Figure 18.1   Opportunities for Interaction Within Group Mentoring

A

Target youth

Mentor C D

B

Other group members

particularly those that use cognitive-behavioral therapy (CBT) techniques. For example, meta-analyses of group CBT for anger problems, substance abuse, and general symptomatology in children and adolescents have found large effect sizes (Battjes et al., 2004; Petrocelli, 2002; Sukhodolsky, Kassinove, & Gorman, 2004). These strong findings suggest that the structured teaching-based and skill-oriented nature of CBT interventions may be key to their effectiveness. Thus, one consideration in evaluating the effectiveness of group mentoring is to consider the extent to which mentors structure the group experience for youth. Insights From Mutual Help and Support Group Interventions Watkins (2005) discussed the potential for small group interventions to contribute to personal, relational, and collective well-being. He noted that the small group format affords opportunities for individuals to share their concerns with attentive peers in a context where they not only receive help but also may provide help to others. Led by skilled facilitators with knowledge of group dynamics and a vision of the potential for change, group members can explore common issues and work toward solutions in ways

that enhance their ability to cope with difficulties in their lives and forge a sense of belonging and cohesion within the group. Group members may also identify areas of common concern that can lead to collective action toward change. Mutual help and support group interventions share with group therapy an emphasis on the importance of group leaders in structuring the groups’ activities, but perhaps place greater emphasis on the group processes themselves (i.e., cohesion, identification with peers and the group leader, and the provision and receipt of help) as the key mediators of the links between group participation and positive developmental outcomes. Unfortunately, little research on such interventions for youth has examined the extent to which social support, deriving from either group leadership or group processes, mediates the link between participation and the achievement of positive outcomes (Barrera & Prelow, 2000). Insights From Developmental Theory Understanding the role of peer influences in child and adolescent development is pivotal for building a theory supporting the potential for group mentoring. A particular concern involves the potential for “contagion” of negative behaviors when

276  PROGRAMS AND CONTEXTS youth who are at risk for aggression and other behavioral problems are grouped together in interventions that are intended to be preventive or therapeutic (Dishion et al., 1999). While raising important concerns, these findings of negative effects have been contradicted by several studies (e.g., Burleson, Kaminer, & Dennis, 2006; Huefner, Handwerk, Ringle, & Field, 2009; Weiss et al., 2005). Ang and Hughes’s (2001) meta-analysis of group-based social skills training interventions for antisocial youth found little evidence of iatrogenic effects, but did find that effect sizes for positive outcomes were stronger when intervention groups involved a mix of prosocial and antisocial youth rather than only antisocial youth. Thus, it appears that contagion effects are strongest when groups consist predominantly of youth already exhibiting antisocial behaviors, and don’t seem to be present in well-run group counseling for children and adolescents. It is also important to consider the possibility that the relationships formed in a group mentoring context can lead to positive development. Developmental researchers have long held that interpersonal relationships both with peers and with adults are fundamental to positive development. For example, Hartup (1989) differentiated “vertical” relationships that children form with individuals who hold greater knowledge and social power from “horizontal” relationships that children form with individuals who hold the same amount of social power. Vertical relationships offer protection, security, and opportunities for the development of basic social skills. Horizontal relationships form the contexts in which social skills are practiced and elaborated. Though the nature of these relationships evolves with development, both remain important across developmental transitions from early childhood through adolescence. Keller (2005) discussed such processes within dyadic youth mentoring relationships, but their manifestation in the larger group mentoring format has not been examined. For example, it has been observed that school-based mentoring capitalizes on the opportunity for peers’ observations of mentor-mentee relationships to elevate the status of younger mentees (Cavell & Smith, 2005; Herrera & Karcher, this volume, Chapter 14). The group mentoring format may increase the involvement of peers in the mentoring process dramatically while limiting opportunities for deviancy effects that may be present in peer mentoring in which no adult serves as a moderator. Proposed Theory of Group Mentoring Our theory of group mentoring draws on the notion of groups as social microcosms involving

youth in both vertical and horizontal relationships. Exposure to the varied relationships and interactions depicted in Figure 18.1 has the potential to engender experiences of group cohesion, feelings of closeness with the mentor, and opportunities for mutual peer support not readily available in traditional adult-youth dyadic mentoring relationships (see Figure 18.2). These processes also occur at the group level, such that groups as a whole can be described by the extent to which they are characterized by close mentor-mentee relationships, cohesion, and mutual help. These processes (connection, cohesion, and mutual support at the group level), in turn, mediate the link between participation in group mentoring and its desired outcomes. In sum, we argue from a theoretical perspective that group mentoring has the potential to contribute to positive developmental outcomes in youth. The potential mechanisms of change operate via the (vertical) relationship between mentors and mentees (as in traditional one-on-one mentoring) as well as through (horizontal) processes of group cohesion and mutual help. Future refinements to this model will need to specify the extent to which individual experiences (e.g., feelings of closeness with a mentor) versus group-level experiences (e.g., aggregate measures of group cohesion) operate as mediators of the links between participation and intended outcomes.

Research Group interventions for youth are ubiquitous, but not all approaches incorporate mentoring as an intentional and central feature. Our literature review revealed studies focused on a range of target populations, from group treatments of youth with identified disorders (e.g., depression) to youth considered to be at risk for developmental difficulties. We considered informal group mentoring that often occurs in settings designed for other purposes (Felner et al., 2001; Hirsch, 2005; Van Ryzin, 2010) as well as formal and programmatic group interventions. In our review of formal group mentoring programs, we emphasized initiatives that stress prevention or the promotion of youth development rather than treatment or psychotherapy. To identify research specifically on group-based youth mentoring, we consulted databases of published works in a variety of fields (e.g., Academic Search Complete, Family & Society Studies Worldwide, PsycARTICLES, PsycEXTRA, Psychology & Behavioral Sciences Collection, PsycINFO, Social Work Abstracts, Sociological Collection, ProQuest

Group Mentoring  277 Figure 18.2   Theoretical Model of Group Mentoring

Group member

Closeness with mentor

Desired outcomes

Group cohesion

Mutual support

Education Journals, Educational Abstracts). We also drew on our knowledge of existing programs and included evaluations of programs described in unpublished reports and evaluations presented at major professional conferences. We identified examples of “natural group mentoring” in formalized settings (e.g., schools or after-school programs) in which group mentoring occurred systematically (e.g., advisory programs) or spontaneously (e.g., facilitator meeting informally with groups of youth) and examples of formal group mentoring programs. For inclusion, formal programs had to include an explicit and intentional focus on the relationship between leader(s) and the group members as a key factor in the program’s theory of change. Informal Group Mentoring in Formalized Settings Youth are commonly placed in groups for a variety of educational and recreational activities; thus, it is typical to find everyday settings that have the basic ingredients for group mentoring to occur (Hamilton et al., 2006). For example, school reforms often include efforts to create small, personalized learning environments and to restructure the school day to create spaces in which teachers can meet regularly with small groups of students. The School Transition Environment Project (STEP; Felner et al., 2001) restructured middle and high school teachers’ roles to include administrative duties, counseling, guidance/advisement, direct communication with parents, and liaison to student support services. Evaluations of STEP and its successor, Project HiPlaces, have consistently found decreased rates of

school dropout, improved academic performance, and reduced behavioral problems among participants relative to comparisons. Unfortunately, it is impossible to tease apart the specific contribution of advisory groups (which closely resemble group mentoring) to promoting these positive outcomes. Though the organization of classrooms affords the opportunity for teachers to serve as mentors for groups of students they see for extended periods of time each week, fewer than 5% of children and youth identify their teachers as mentors (Hamilton et al., 2006). However, Van Ryzin (2010) found that a larger proportion of youth (about 40%) identified their teachers as attachment figures in schools implementing formal teacher-homeroom advisory programs. Moreover, students who nominated their teachers reported greater school engagement and had greater gains in academic achievement and emotional adjustment than those who did not. Hirsch (2005) described natural group mentoring within the context of Boys & Girls Clubs in a large urban area. His mixed-method account described how certain adult workers at the clubs were able to build trust and gain the respect of children and adolescents. As the youth developed a strong sense of attachment toward and identification with these adults, the adults in turn served as role models and were able to use their authority as directors of club activities to involve youth in extended group discussions and activities focused on a broad range of youth issues. Similar findings were reported in subsequent mixed-methods research with Boys & Girls Clubs that reported indepth case studies of youth and clubs (Hirsch, Deutsch, & DuBois, 2011).

278  PROGRAMS AND CONTEXTS The studies by Hirsch and colleagues (Hirsch, 2005; Hirsch et al., 2011) and Van Ryzin (2010) suggest that informal group mentoring in formalized youth contexts holds promise for promoting youth development, but it appears that such mentoring is unlikely to occur unless systematic efforts are made to increase their likelihood (e.g., establishing teacher-student advisory groups in schools). Thus, we turn in the following section to an examination of formal group mentoring programs. Group Mentoring Programs We identified evaluations of 10 formal mentoring programs that met our criteria. Table 18.1 summarizes program descriptions and evaluation findings from these programs. We note program goals, settings, target populations, and structural characteristics. We briefly describe group activities and the extent to which implementing activities reflects a task focus. We viewed a high level of task focus as reflecting activities driven by formalized programmatic goals (e.g., strict adherence to a curriculum), whereas a low task focus reflected activities driven by group members’ interests. Programs are listed according to the research design with which they have been evaluated, from the most rigorous (i.e., randomized experiment) to the least (i.e., nonexperimental). Before turning to our review of the studies in Table 18.1, it is worth noting that a recent meta-analysis of youth mentoring program evaluations reported that the effectiveness of programs using a group format did not differ significantly from that observed for those utilizing the more traditional one-on-one format (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). Details regarding the number or types of group mentoring programs included were not provided. Nonetheless, the finding is noteworthy, especially given that, as in a prior meta-analysis (e.g., DuBois, Holloway, Valentine, & Cooper, 2002), this review found evidence of significant overall effects of mentoring programs on emotional, behavioral, social, and academic outcomes. Overview of Findings. All of the programs were site based and most targeted youth identified as being at risk for school difficulties or behavioral problems. Six programs specifically targeted ethnic minority youth or culturally diverse populations and incorporated an understanding of culture and ethnicity as a central programmatic feature. Targeted youth ranged in age from 8 to 18, and mentor(s)-to-mentee ratios ranged from 1:3 to 1:10. Only three programs used two or more than two mentors working as a team (thus limiting the extent to which we were

able to explore the distinction between group and team mentoring). The evaluations assessed impacts on a broad range of outcomes, including academic performance, interpersonal skills, cultural/spiritual development, and behavioral/emotional problems. Variation in the extent of task focus ranged from programs that followed prescribed curricula (e.g., Jent & Niec, 2009) to ones where activities were organized primarily around participants’ interests (e.g., Kuperminc & Cummings, 2010). Only five evaluations included experimental or quasiexperimental designs, only about half of the evaluations reported significant effects on targeted outcomes in quantitative analyses, and small sample sizes limited the ability of studies to detect significant effects. Few evaluations examined processes suggested by our theoretical model as mediators of program effects, yielding few insights about the utility of our model. In the remainder of this section, we describe three representative programs to illustrate the diversity of approaches to group mentoring. We conclude the section by summarizing current knowledge about the effectiveness of group mentoring, given the strengths and limitations of extant studies, and offer directions for future research. Cognitive Behavioral Principles Within Group Mentoring (Jent & Niec, 2009). This program took place at a rural community mental health center and targeted children, 8–12 years of age, identified as having emotional or behavioral problems. Mentors were undergraduate or graduate students in social work or counseling. Each group consisted of two mentors working with four to eight children. Sessions included group discussions regarding social problem solving and interaction skills. Each session featured a specific task set with clear goals and activities as determined by the program. Additionally, the groups engaged in child-determined activities based on their interests. These activities were considered “practice” for the youth to engage in prosocial behavior, and children were rewarded for positive behavior with praise and a token economy. The evaluation, a randomized trial, found significant increases among intervention children (n = 42) in their use of social problem-solving skills and significant reductions in externalizing and internalizing behavior problems relative to controls (n = 38). Unfortunately, there was no way to tease apart the effectiveness of the group mentoring component from the program’s therapeutically oriented skills training component. Youth Development Program (House, AlvarezJimenez, McCoy, Lapidus, & Kuperminc, 2006; Kuperminc & Cummings, 2010). This program

279

Village Model of Care (Carswell et al., 2009)

(Jent & Niec, 2009)

Cognitive Behavioral Principles Within Group Mentoring

Goal: African American cultural heritage, school bonding, social skills development, and academic achievement Setting: On-site after-school program Target: 6th-grade African American youth in high-risk urban areas Group Composition: Approximately 20 students; ages 11–15; 45.5% male Mentor(s): African American college students, recent graduates (1:10 ratio)

Goal: Reduce behavior problems Setting: Rural community mental health center Target: Children with emotional and behavioral problems Group Composition: 4–8 children per group, ages 8–12 Mentor(s): Two mentors per group; minimum of 24 hours of training; half-hour supervision per week (1:4 ratio)

Goal: Dropout prevention, academic success Setting: After-school program Target: Youth at academic risk in middle and high school Group Composition: Six pairs of 8th-grade students at low, medium, or high academic risk; predominantly Hispanic and Asian Mentor(s): Two trained adult facilitators, mostly college students (1:6 ratio)

Twelve Together

(Dynarski et al., 1998; U.S. Department of Education, 2007)

Description

Program

Table 18.1   Group Mentoring Program Descriptions and Evaluation Findings

Activities: Moderate task focus; study-skills exercises, tutoring/ homework help, discussion of self-control topics, career opportunities, African American cultural heritage, recreational/ social activities; activities for family: parental support services; community outreach (scheduled gatherings, outings with families, field trips)

Activities: High task focus; group discussion; didactics related to social problemsolving and social interaction skills; child-determined activities based on group interests

Design: Randomized controlled trial

Activities: Low task focus; weekly group discussion based on student interest (personal, family, and social issues); homework assistance, trips to college campuses, annual weekend retreat to create group cohesion

(Continued)

Key Findings: Participants had significantly higher grade point averages and higher teacher ratings of behavior at posttest

Sample Size: 478 students; 237 in intervention group, 241 in comparison group

Design: Quasi-experimental

Key Findings: Increases in use of social problem solving skills; decreases in parentreported externalizing and internalizing problems

Sample Size: 80

Design: Randomized controlled trial

Key Findings: 8% of students in intervention dropped out; 13% of control group dropped out (not significant); no impact on progressing in school (as measured by highest grade completed)

Sample Size: 246; 130 intervention, 116 controls

Evaluation

Group Processes

280

Setting: In school during regular hours

(House et al., 2006; Kuperminc et al., 2010)

The Kuumba Group (Washington et al., 2007)

(Moon, Cristenson, & Redford, 1996; Murphy et al., 1997)

TeamWorks

Goal: Increase peer relationship quality and school engagement

Youth Development Program

Setting: Churches, community settings

Goal: Increase spirituality and adaptive functioning at home, school, and in relationships; cultural strengths

Activities: Moderate task focus; videos, recreational activities, discussions, values clarification

Sample Size: 6 youth participants; 6 relative caregivers

Design: Nonexperimental

Key Findings: Perceptions of mentor instrumental support related to improved attitudes toward adults and the future; perceptions of mentor as role model related to improved attitudes toward adults, community, the future, and cultural pride

Sample Size: 120

Mentor(s): Three mentors (teacher, college student, community member) (1:4 ratio)

Key Findings: Improved attitudes toward community, relationships with adults

Sample Size: 60; 30 intervention, 30 controls

Study 1 Design: Quasi-experimental, posttest only

Key Findings: Decrease in school problem behavior; decrease in school belonging; increase in school engagement for immigrants

Sample Size: 152 (102 participants, 50 controls)

Design: Quasi-experimental

Evaluation

Study 2 Design: Nonexperimental, pre- and posttest, mixed quantitative/qualitative

Activities: High task focus; biweekly meetings; curriculum focused on team-building, leadership, and community service

Activities: Low-to-moderate task focus and high relational focus; combination of planned and group-initiated discussions and activities

Group Processes

Group Composition: 10–12 students in 6th–8th grades in urban, low-SES schools; primarily Black and Hispanic

Target: Youth at academic risk in middle and high school

Setting: After-school program

Goal: Improve attitudes toward self, family, school, community, adults, other ethnic groups, and the future; improve participation in school

Mentor(s): Undergraduate students; training, ongoing supervision (1:7 ratio)

Group Composition: 4–7 students matched by gender

Target: Ethnically diverse male and female high school students

Description

Program

Table 18.1  (Continued)

281

(Khan, 2006)

TALKS Mentoring Movement

Setting: After-school program

(Deutsch et al., 2010)

Mentor(s): Adult volunteers from the community (1:3 ratio)

Group Composition: Three students, usually racially mixed with one excelling student, one average student, and one “at-risk” student

Target: Children with disciplinary problems, grades 3–12

Setting: In school during regular hours

Goal: Improve behavior, academics

Mentor(s): College-aged women

Group Composition: Group activities involving 1:1 mentoring pairs

Target: 7th-grade girls

Goal: Relational development for adolescent girls and their mentors

Mentor(s): African American male adults; clinicians, graduate students, and business professionals (1:6 ratio)

Activities: High task focus; activities and discussion focused on respect, peer pressure, relationships with siblings, anger management, work ethic, understanding different races and cultures, and having a positive attitude

Activities: High task and high relational focus; weekly group discussion and activities

exercises, art media, business plan development, breathing and relaxation to reduce stress and promote well-being; incorporation of Afro-centric themes such as Nguzo Saba, ritual adapted from Kwanzaa holiday, and Let the Circle Be Broken

Target: African American males, ages 9–17

Group Composition: Six per group

Group Processes

Description

Young Women Leaders Program

Program

(Continued)

Key Findings: Positive relationship quality with mentor and others, vision for the future, ability to manage conflict, selfesteem, perceptions of support in social environments, few self-reported aggressive behaviors

Sample Size: 124

Design: Nonexperimental; mixed methods

Key Findings: Mentor-mentee relationship quality related to satisfaction; self-reported improvements in academic and interpersonal skills

Sample Size: 134 mentors and mentees (quantitative); 98 post-program interviews (qualitative)

Design: Nonexperimental; mixed methods

Key Findings: No significant difference in spirituality; focus groups indicated better home and school behavior

Evaluation

282

Goal: Empower marginalized youth through critical consciousness about their exclusion from the mainstream

Setting: School

Group mentoring program

(Garcia, 2007)

Setting: Foster care agency

(Utsey et al., 2003)

Mentor(s): African American, male volunteers; quarterly training; ratio unknown

Group Composition: males aged 12–16 years; number per group unknown

Target: African American male adolescents in foster care

Goal: Reduce self-destructive behavior, increase adaptive behavior

Therapeutic mentoring group

Mentor(s): Three adult leaders (facilitator, co-facilitator, assistant); graduate and undergraduate students (3:10 ratio)

Group Composition: Multiethnic; maximum of 10 per group

Target: Alternative school students; history of attendance, behavior, and/or motivational problems

Description

Program

Table 18.1  (Continued)

Activities: Low task and high relational focus; discussion on issues important to members such as sexuality, foster care, education, and careers; bimonthly recreational activities as determined entirely by mentors

Activities: High task focus, moderate relational focus; phases: engagement, co-constructive learning, and transformation activities; use of multicultural counseling theory

Group Processes

Key Findings: Members seemed to experience change in attitude and behavior; all seemed to experience positive relationship with Black male role model; consistency of mentors’ behaviors and presence thought to be a major factor in group’s effectiveness

Sample Size: 6

Design: Case study

Key Findings: Positive group climate unrelated to outcomes

Sample Size: 121

Design: Cross-sectional, no control group

Evaluation

Group Mentoring  283 was designed to increase peer relationship quality and school engagement among students in a culturally diverse and predominantly immigrant high school student body. Undergraduate psychology majors were matched with groups of five to seven high school students. Mentors were recruited based on faculty recommendations and were required to have completed a semester-long experiential course on interpersonal communication. They received a full day of training and weekly ongoing supervision by graduate psychology students. Mentors earned course credit throughout a full academic year. Activities were a combination of planned (by the mentor) and group-initiated discussions and activities surrounding a variety of issues, many of which focused on racial and ethnic issues pertaining to their school. House et al.’s (2006) quasiexperimental evaluation (n = 102 participants and 50 demographically matched controls), found reductions in behavior problems and increases in school engagement (but only for immigrant students) relative to controls. Unexpectedly, participants reported declines in perceived school belonging. Subsequent qualitative analysis of mentors’ weekly process notes (Cummings, 2010) suggested that group discussions about racial conflict and perceived discrimination among some groups may have contributed to negative feelings about the school. On the other hand, in partial support of the cultural congruence hypothesis, Kuperminc & Cummings (2010) found that youth with a positive ethnic identity were more likely than others to experience high levels of mutual support and connectedness with mentors. Thus, the decline in school belonging might reflect important ethnic identity developments that were not captured fully during the window of the study. Of relevance to our theoretical model, Kuperminc & Cummings (2010) examined effects of program participation on the quality of peer and school relationships, and moderating effects of perceived mutual help and connectedness with the mentor (a measure of group cohesion was omitted because of high correlations with the other process variables). Youth who perceived a high level of mutual help increased their sense of school belonging and peer relationship quality. Youth who reported high connectedness with their mentor increased their interpersonal skills. Thus, the study provides preliminary support for the theoretical model. The Young Women’s Leadership Project (Deutsch et al., 2010; Sovik-Johnston, Lawrence, Deutsch, & Lee, 2009). Seventh-grade girls identified as being at risk for academic or social-emotional problems were paired with a college-aged woman

and participated in both one-on-one and paired activities over the course of one academic year. One-on-one mentoring occurred for at least 1 hour per week, and eight mentor-mentee pairs met as a group after school for 2 hours per week. A highly structured curriculum for group activities included issues related to self-concept, academic achievement, body image, and healthy decision making. Participant survey responses revealed perceived improvement in academics and interpersonal skills with more pronounced gains in these areas for girls from lower socioeconomic backgrounds. Deutsch et al.’s (2010) qualitative observations found the group setting could be protective for girls who were dissatisfied with their paired mentor, and that mentees played an active role in shaping their experiences in the group through strategies of “reaching out” (i.e., mutual support) to others and working out negative emotions (e.g., conflict, rejection). Summary and Recommendations for Future Research Results from the set of four studies presented in Table 18.1 with rigorous experimental or quasiexperimental designs have tended to show gains in academic and behavioral functioning among program participants relative to controls. Despite promising findings, the research is limited in at least four ways. First, because most of the programs were multifaceted, it is difficult to tease apart effects of group mentoring from those of other programmatic activities. Future studies might employ factorial designs in which multiple versions of an intervention (e.g., group mentoring alone, group mentoring plus social skills training) could be compared to a nonintervention control (see Kuperminc et al.’s [2005] chapter in the first edition of the Handbook of Youth Mentoring for further reading on the challenges facing multicomponent program evaluations). For example, Randall, Eggert, and Pike (2001) studied the effectiveness of two suicide prevention protocols for high school students at risk of school dropout and suicide potential. One group received assessment, brief counseling, and facilitation of social support from school personnel and a parent. After receiving the same intervention, a second group also participated in a 12-session skills training and social support group. Participants in the combined intervention showed increased personal control, problem-solving coping, and perceived family support. Both experimental conditions showed decreased depression and enhanced self-esteem relative to untreated controls. This study illustrates

284  PROGRAMS AND CONTEXTS how a factorial design can provide insight into the added value of a group process on an intervention’s effectiveness. Second, the studies offer little insight into the mechanisms through which group mentoring can be effective. Future studies, employing both qualitative and quantitative methodologies are needed. The nonexperimental studies by Garcia (2007), Deutsch et al. (2010), and Murphy, Soto, and Gopez (1997) measured some of the program processes thought to mediate the association between participation and expected outcomes. The processes they investigated are similar to our proposed model that considers closeness with mentor and group cohesion as potentially mediating the relationship between the individual’s participation and desired outcomes. Deutsch et al. inferred links between perceived relationship quality with mentors and participants’ satisfaction in the program. Garcia measured perceptions of group climate as mediating the relationship between program participation and psychosocial development. Murphy et al. found that changes in youths’ attitudes toward adults, community, the future, and cultural pride were linked to perceived instrumental support from mentors and mentors as role models. Our work with the Youth Development Program has begun to quantify program-related experiences that predict participating youths’ social development (Kuperminc & Cummings, 2010). As these efforts move forward, it is critical that investigators take into account the interplay of group- and individual-level processes. For quantitative studies in particular, this can be accomplished through multilevel modeling that accounts for the fact that individuals are nested within groups; such analysis requires large samples and careful specification of variables at both individual and group levels. Third, the idea that a group mentoring format may be particularly congruent with the interdependent relational style of youth from cultural or ethnic minority groups (Lindsay-Dennis et al., 2011; Utsey et al., 2003) has not yet been tested formally. Future research will need to investigate this possibility. Fourth, research must move beyond descriptive studies that document instances in which informal group mentoring structures emerge in the context of formalized youth settings. The fact that such structures seem to emerge “naturally” in contexts where groups of young people engage in sustained relationships with one or more adults attests to their potential for engaging youth and adults. Studies are needed that can elucidate the conditions under which informal group mentoring is most likely to

occur. Also needed are studies that can clarify the relative strengths and limitations of groups varying in level of formality (e.g., naturally occurring vs. intentionally formed groups).

Practice Group mentoring has practical appeal as a lower cost option, compared to one-on-one mentoring. Group mentoring may also increase the reach and offer a more culturally congruent alternative for ethnic/cultural minority youth and provide other unique benefits over traditional dyadic approaches. To date, however, few resources for designing and implementing group mentoring programs exist. In this section, we offer recommendations for practice in each of several areas of group mentoring that are informed by theory and research reviewed in this chapter (see Table 18.2). Our recommendations also draw from available resources specific to group mentoring (Sherk, 2006), from resources for mentoring in general (MENTOR, 2009), and from our own and our colleagues’ (e.g., N. Deutsch, personal communication, November 1, 2011) experience. Structuring the Program Among the key early decisions in designing and implementing a group mentoring program are to determine the size of groups and the number of mentors per group. Proponents of group psychotherapy suggest that the use of co-therapists affords opportunities for modeling behavior (e.g., male and female therapists can model positive gender roles), sharing responsibility, and working off one another’s complementary skills (e.g., differences in experience). Group size and mentee-to-mentor ratios also must be considered. Recommendations of four youth per mentor have been offered (MENTOR, 2009; Sherk, 2006), but there is little evidence to support this ratio. However, groups exceeding 8–10 people (including mentors) may make group exercises unwieldy (Sherk, 2006). Age differences between mentors and mentees must also be considered. The only instance of negative effects in our literature review occurred in our own study (House et al., 2006), in which youth reported declines in sense of school belonging. Subsequent investigation suggested that many groups had difficult discussions about racial tension and perceived discrimination at the school, issues that many of our mentors were unprepared to deal with in satisfying ways. Although older mentors might

Group Mentoring  285 Table 18.2   Checklist for Practitioners Topic

Recommendations

Structuring the program

Mentor to mentee ratios of 4–6 youth for each mentor; large groups may make group activities unwieldy. Consider having at least two mentors per group to enhance opportunities for role modeling, sharing responsibility. Consider age difference between mentor(s) and mentees.

Recruitment and training of mentors

Screen potential mentors using best practices for one-on-one mentoring. Consider consulting a child therapist with expertise in youth groups. Adapt existing training materials developed for one-on-one mentoring (see Kupersmidt & Rhodes, this volume, Chapter 30). Include instruction on child/youth development topics (risk and resilience, peer and adult relationships), tailored to the population to be mentored. Devote substantial time to developing skills in group facilitation (managing multiple roles, stages of group process, handling disruption).

Recruitment and training of mentees

Establish clear criteria for eligibility, emphasizing shared interests and goals among mentors and group members. Introduce mentees to each other; help them understand concept of mentoring. Engage group mentors in establishing ground rules and understanding how to get the most out of the experience.

Matching

Peer-to-peer: Avoid homogeneous groups of youth with behavioral problems to limit likelihood of harmful effects; emphasize shared interests. Mentor-to-mentor: If using multiple mentors, consider how they may complement one another’s experience, gender roles, energy level, etc. Mentor(s)-to-group: Match experienced mentors with groups with most potential for difficulties; emphasize shared interests with group members.

Cultural competence and ethnic identity

Assess cultural competence and ethnic identity of mentors and mentees (see Sánchez et al., this volume, Chapter 10). Identify preferences and concerns regarding matching, comfort with individuals of other groups. Provide training and ongoing consultation as necessary.

Ongoing supervision, reflection, and planning

Monitor group process: Periodic observation by program managers and consistent logging of group activities and processes by mentors. Document stages of group process. Vary types of activities in order to keep it fresh. Plan for termination.

have been better equipped to manage such difficult themes, it is also possible that mentees felt more comfortable raising difficult issues with mentors who were close to their own age. Program designers are encouraged to weigh the potential advantages and disadvantages of each approach.

Recruitment and Training of Mentors Initial screening of potential mentors should follow the best practices developed for one-on-one mentoring programs (see Sherk, 2006, and Kupersmidt & Rhodes, this volume, Chapter 30),

286  PROGRAMS AND CONTEXTS and training materials can be adapted from existing sources (e.g., MENTOR, 2009). It is important to pay particular attention to training in facilitation of group dynamics and peer relationships. It may be helpful to engage the assistance of a child therapist with group expertise. Such an expert can help mentors to identify the progress of the group through the early stages of initial group formation, testing limits and resolving authority issues that may be necessary before the group is able to fully engage in its work (Dies, 2000). Such consultation can also be important at a program’s conclusion, providing mentors with strategies for helping group members to recap what they have learned, deal with the emotions inherent in saying goodbye, and learning to move on. Recruitment and Training of Mentees Establishment of eligibility criteria will depend largely on program goals. Because the goals of the Youth Development Program were broadly articulated to create welcoming settings that would encourage connections to school and to peers among a predominantly cultural minority and immigrant student body, we established groups based on shared interests among and between mentors and mentees. Whereas this approach may not be applicable to other programs, we believe that programs should make intentional efforts to encourage group formation and ownership. This includes initial discussions with mentees to “train” them in the concepts of mentoring and how to get the most out of the experience. Early discussions should also engage youth in establishing ground rules and procedures for the group, including those related to confidentiality and respect (Sherk, 2006). Matching A critical task for one-on-one mentoring is to consider the chemistry in a mentor-mentee pair (see Pryce et al., this volume, Chapter 29). For group mentoring, this task is complicated by the need to consider peer-to-peer, mentor-to-mentor (if using multiple mentors), and mentor-to-group matches (Sherk, 2006). One of the primary concerns regarding peer-to-peer matches is the likelihood of harmful effects when youth with high levels of behavioral problems are grouped together. It may be wise to limit the number of youth already exhibiting behavior problems to no more than one per group (Dodge, Dishion, & Lansford, 2006). If using multiple mentors, consider how they might complement one another with regard to skills, experience, and roles

(gender, personality, etc.). For example, teaming an experienced older mentor with a younger, less experienced mentor might have addressed the concerns raised in the Youth Development Program example. Cultural Competence and Ethnic Identity Herrera et al. (2002) found that group mentoring programs tend to attract higher proportions of youth and adults from ethnic minority groups than do one-on-one mentoring programs. Although research is only beginning to address this issue, preliminary evidence points to the importance of careful attention to issues of cultural competence and ethnic identity. As suggested by Sánchez et al. (this volume, Chapter 10), we recommend that programs incorporate an assessment of mentors’ and mentees’ ethnic identity and cultural competence to assist in identifying matching preferences and levels of comfort in interacting with individuals of different ethnic or cultural groups. These issues should be integrated into initial and ongoing training; it may also be advisable to engage external consultants with expertise in this area. Ongoing Supervision, Reflection, and Planning The fact that most group mentoring programs are site based and (at least somewhat) curriculum driven offers opportunities for supervision, reflection, and planning that might not be readily available in other mentoring contexts. We recommend a twopronged approach including periodic observation (e.g., by program staff or mentors serving as guest observers to other groups), and consistent logging of group activities and impressions of the group process by mentors. These processes are useful for documenting basic program information (e.g., attendance) and charting progress of individuals and the group as a whole. Such information can be used to identify problems and strategize solutions (e.g., working with a disruptive mentee), monitor the stages of group development, reflect on what is working and what needs to be changed, gauge youths’ levels of energy and engagement, and plan for the ending of the group.

Conclusion In this chapter, we have offered a preliminary theoretical model for understanding the potential effectiveness of group mentoring. Our model draws from theoretical frameworks of group psychotherapy, mutual support group interventions, and theories regarding the significance of peer and adult

Group Mentoring  287 relationships for psychosocial development. We argue that intended outcomes are likely to emerge when group members experience a high level of connectedness with the mentor or mentors, and when a high level of group cohesiveness takes hold in which group members experience themselves in the roles of both helper and recipient of help (i.e., mutual help). Our review of the literature provided limited support for the effectiveness of group mentoring across a broad range of outcomes, including academic performance, interpersonal skills, cultural/ spiritual development, and behavioral and emotional problems. Limited support was also found for the theoretical model insofar as one study found associations linking perceived mutual support to increases in youths’ sense of belonging to school and the quality of their peer relations, whereas connectedness to mentors was linked to increases in youths’ perceived interpersonal skills. It is notable that group mentoring has continued to grow as a prominent alternative despite concerns that it represents only a “watered-down” version of traditional one-on-one mentoring and amid cautions about potential negative effects of group interventions for youth. Moving research and practice on group mentoring forward will require an appreciation of the unique ways that a group setting can contribute enriching experiences that promote youth development. One intriguing aspect is the potential synergy of group mentoring for youth of color, many of whom have interpersonal orientations that favor interdependence with others. Given the challenges often observed of effectively mentoring youth of color (Sánchez et al., this volume, Chapter 10), group mentoring may offer a particularly valuable alternative for many ethnic minority youth. Whereas past work on group psychotherapy and mutual help interventions provides insight into the requisite skills for effective management of groups, future research will need to clarify how the mentor role is both similar to and different from the roles of group therapist or facilitator and how it is both similar to and different from the traditional one-on-one mentoring role. As research and practice on group mentoring begin to address these issues, we can continue to expand the repertoire of effective approaches to youth mentoring.

References Ang, R., & Hughes, J. (2001). Differential benefits of skills training with antisocial youth based on group composition: A meta-analytic investigation. School Psychology Review, 31, 164–185.

Barrera, M., & Prelow, H. (2000). Interventions to promote social support in children and adolescents. In D. Cicchetti, J. Rappaport, I. Sandler, & R. Weissberg (Eds.), The promotion of wellness in children and adolescents (pp. 309–340). Washington, DC: CWLA Press. Battjes, R., Gordon, M. S., O’Grady, K. E., Kinlock, T. W., Katz, E. A., & Sears, E. C. (2004). Evaluation of a group-based substance abuse treatment program for adolescents. Journal of Substance Abuse Treatment, 27, 123–134. Burleson, J., Kaminer, Y., & Dennis, M. (2006). Absence of iatrogenic or contagion effects in adolescent group therapy: Findings from the Cannabis Youth Treatment (CYT) Study. American Journal on Addictions, 15, 4–15. Carswell, S. B., Hanlon, T. E., O’Grady, K. E., Watts, A. M., & Pothong, P. (2009). A preventive intervention program for urban African American youth attending an alternative education program: Background, implementation, and feasibility. Education & Treatment of Children, 32(3), 445–469. Cavell, T. A., & Smith, A. (2005). Mentoring children. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 160–176). Thousand Oaks, CA: Sage. Cox, P. D., Vinogradov, S., & Yalom, I. (2005). Group therapy. In R. Hales, S. Yudofsky, & G. Gabbard (Eds.), The American Psychiatric Publishing textbook of psychiatry (5th ed., pp. 1329–1373). Arlington, VA: American Psychiatric Publishing. Cummings, L. (2010). Evaluating the influence of participation in a diverse high school-based group mentoring program. Unpublished doctoral dissertation, Georgia State University. Deutsch, N., Henneberger, A., Wiggins, A., & Lawrence, E. (2010, April). Fostering connection: Mentoring groups as a context for relational development for adolescent girls and their mentors. Presented at the biennial meeting of the Society for Research on Adolescence, Philadelphia, PA. Dies, K. (2000). Adolescent development and a model of group psychotherapy: Effective leadership in the new millennium. Journal of Child and Adolescent Group Therapy, 10, 97–111. Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: Peer groups and problem behavior. American Psychologist, 54, 755–764. Dodge, K. A., Dishion, T. J., & Lansford, J. E. (2006). Deviant peer influences in programs for youth: Problems and solutions. New York: Guilford Press. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Karcher, M. J. (Eds.). (2005). The handbook of youth mentoring. Thousand Oaks, CA: Sage. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of

288  PROGRAMS AND CONTEXTS the evidence. Psychological Science in the Public Interest, 12, 57–91. Dynarski, M., Gleason, P., Rangarajan, A., & Wood, R. (1998). Impacts of dropout prevention programs: Final report. A research report from the School Dropout Demonstration Assistance Program evaluation. Princeton, NJ: Mathematica Policy Research. Felner, R. D., Favazza, A., Shim, M., Brand, S., Gu, K., & Noonan, N. (2001). Whole school improvement and restructuring as prevention and promotion: Lessons from STEP and the Project on High Performance Learning Communities. Journal of School Psycho­ logy, 39, 177–202. Garcia, A. J. (2007). Investigating processes in a positive youth development program for multi-problem youth. Unpublished doctoral dissertation, Florida International University. Hamilton, S. F., Hamilton, M. A., Hirsch, B. J., Hughes, J., King, J., & Maton, K. (2006). Community contexts for mentoring. Journal of Community Psychology, 34, 727–746. Hartup, W. (1989). Social relationships and their developmental significance. American Psychologist, 44, 120–126. Herrera, C., Vang, Z., & Gale, L. (2002). Group mentoring: A study of mentoring groups in three programs. Philadelphia: Public/Private Ventures. Retrieved from http://www.ppv.org/ppv/publications.asp?section _id=22 Hirsch, B. J. (2005). A place to call home: After school programs for urban youth. New York: Teachers College Press. Hirsch, B. J., Deutsch, N. L., & DuBois, D. L. (2011). After-school centers and youth development: Case studies of success and failure. New York: Cambridge University Press House, L. D., Alvarez-Jimenez, A., McCoy, N., Lapidus, R. B., & Kuperminc, G. P. (2006, May). Mentoring culturally diverse youth: Group mentoring processes and effects. Presentation at the 14th annual meeting of the Society for Prevention Research, San Antonio, TX. Huefner, J., Handwerk, M., Ringle, J., & Field, C. (2009). Conduct disordered youth in group care: An examination of negative peer influence. Journal of Child and Family Studies, 18, 719–730. Jent, J. F., & Niec, L. N. (2009). Cognitive behavioral principles within group mentoring: A randomized pilot study. Child and Family Behavior Therapy, 31, 203–219. Karcher, M. J., Kuperminc, G., Portwood, S., Sipe, S., & Taylor, A. (2006). Mentoring programs: A framework to inform program development, research, and evaluation. Journal of Community Psychology, 34, 709–725. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Khan, C. (2006). Promoting youth development through school-based group mentoring. Dissertation Abstracts International, 67.

Kuperminc, G., & Cummings, L. (2010, April). Group mentoring for culturally diverse youth: The role of group process in promoting positive peer relations. Presentation at the biennial meeting of the Society for Research on Adolescence, Philadelphia, PA. Lindsay-Dennis, L. A., Cummings, L., & McClendon, S. C. (2011). Mentors’ reflections on developing a culturally responsive mentoring initiative for urban African American girls. Black Women, Gender & Families, 5(2), 66–92. MENTOR. (2006). Mentoring in America 2005: A snapshot of the current state of mentoring. Washington, DC: Author. Retrieved from http://www.mentoring. org/start_a_program/evaluation/2005_national_poll/ MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Washington, DC: Author. Retrieved from http://www.mentoring.org/find_resources/ele ments_of_effective_practice/ Moon, A., Christenson, C., & Redfern, J. (1996). TeamWorks evaluation: An initial evaluation. Los Angeles: University of California–Los Angeles School of Public Policy. Murphy, S., Soto, D., & Gopez, A. (1997). TeamWorks evaluation report: An investigation of program processes and outcomes. Claremont, CA: Claremont McKenna College Kravis Leadership Institute. Petrocelli, J. (2002). Effectiveness of group cognitivebehavioral therapy for general symptomatology: A meta-analysis. Journal for Specialists in Group Work, 27, 92–115. Philip, K., & Hendry, L. B. (2000). Making sense of mentoring or mentoring making sense? Reflections on the mentoring process by adult mentors with young people. Journal of Community Psychology, 10, 211–223. Randall, B., Eggert, L., & Pike, K. (2001). Immediate post intervention effects of two brief youth suicide prevention interventions. Suicide and Life-Threatening Behavior, 31, 41–61. Rhodes, J. (2002). Group mentoring. Research Corner on the website of MENTOR. Retrieved from http:// www.mentoring.org/access_research/group/ Richmond, L. (2000). Reflections on a thirty five year experience with adolescent group psychotherapy. Journal of Child and Adolescent Group Therapy, 10, 110–118. Sherk, J. (2006). Designing and implementing a group mentoring program. Santa Rosa, CA: Center for Applied Research Solutions. Retrieved from http:// www.carsmentoring.org/publications/index. php?category_id=34 Sipe, C. L., & Roder, L. E. (1999). Mentoring in schoolage children: A classification of programs. Philadelphia: Public/Private Ventures. Sovik-Johnston, A., Lawrence, E. C., Deutsch, N. L., & Lee, J. L. (2009). Designed for girls: Outcomes from a school-based mentoring program for middle school girls. Manuscript submitted for publication. Sukhodolsky, D., Kassinove, H., & Gorman, B. (2004). Cognitive-behavioral therapy for children and adolescents: A meta-analysis. Aggression and Violent Behavior, 9, 247–269.

Group Mentoring  289 U.S. Department of Education. (2007). What Works Clearinghouse: Twelve together. WWC Intervention Report. Retrieved from http://ies.ed.gov/ncee/wwc/ interventionreport.aspx?sid=527 Utsey, S. O., Howard, A., & Williams, O., III (2003). Therapeutic group mentoring with African American male adolescents. Journal of Mental Health Counseling, 25, 126–139. Van Ryzin, M. (2010). Secondary school advisors as mentors and secondary attachment figures. Journal of Community Psychology, 38(2), 131–154.

Washington, G., Johnson, T., Jones, J., & Langs, S. (2007). African-American boys in relative care and a culturally centered group mentoring approach. Social Work With Groups, 30, 45–69. Watkins, M. (2005). Small group and individual interventions. In G. Nelson & I. Prilleltensky (Eds.), Community Psychology (pp. 211–231). New York: Palgrave Macmillan. Weiss, B., Caron, A., Ball, S., Tapp, J., Johnson, M., & Weisz, J. R. (2005). Iatrogenic effects of group treatment for antisocial youth. Journal of Consulting and Clinical Psychology, 73, 1036–1044.

19  WORK AND SERVICE-LEARNING Stephen F. Hamilton and Mary Agnes Hamilton

Introduction Paid employment, unpaid internships, and servicelearning are common experiences for youth that also can be contexts for mentoring. Mentoring relationships in these contexts may be established formally but might also occur naturally as youth and adults engage in work and service activities together. In this chapter, we describe an abbreviated theoretical framework for understanding effective mentoring in such relationships that draws from our previous chapter (M. A. Hamilton & Hamilton, 2005). The review of the empirical literature on mentoring of youth in work and service emphasizes the research published since that chapter. Mentoring, Work, and Service-Learning This chapter is primarily, though not exclusively, about natural or informal mentoring—that is, mentoring that occurs without a formal matching process and outside of a mentoring program. Our focus instead is on the potential of work experience and service-learning programs to be what Freedman (1993) called “mentor-rich environments” (p. 111). In Karcher and Nakkula’s (2010) terms (see Karcher & Hansen, this volume, Chapter 5), we are most concerned with goal-directed interactions between youth and mentors and in instrumental relationship styles. Work and service-learning share several characteristics that can make them optimal contexts for mentoring. Both give youth adult-like roles in which they are responsible for producing something of value to others. In both, young people have the opportunity gain new knowledge and skills. Compared with youths’ roles as student and family

member, the roles of worker, apprentice, intern, and volunteer are more active and entail more obligations to others, obligations that have a visible impact on other people in a way that ordinary schoolwork does not. Work and service potentially provide contexts in which young people and adults naturally develop relationships around common interests and activities; some of these relationships become mentoring relationships. By examining work and service-learning more closely, we hope to learn more about how to realize their potential as places in which mentoring can occur. Ultimately we would like to know how to create optimal contexts for mentoring and how to orient youth and adults to make use of these contexts. We think of this approach as fostering natural mentoring. This emphasis does not exclude the designation of specific adults as mentors in the contexts of work and service, which can increase the likelihood that mentoring will occur in these contexts, but our focus is primarily on naturally forming relationships. Considerable evidence supports the utility of promoting natural mentoring relationships in workplaces and schools. Erickson, McDonald, and Elder (2009) found that the impact of natural mentors may be quite substantial, in contrast to the rather modest effects found in most evaluations of mentoring programs (DuBois, Holloway, Valentine, & Cooper, 2002; DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). Using data from the National Longitudinal Study of Adolescent Health (Add Health), Erickson and his colleagues found that youth from families with more resources (defined by parents’ education, family income, and living with two parents) were more likely than others to have a person who fit their definition of an informal mentor. But having a mentor, especially one who was a 291

292  PROGRAMS AND CONTEXTS teacher, was associated more strongly with higher educational achievement among youth with limited resources. Indeed, after grades and other contributing factors were controlled, having a teacher mentor was predictive of an increased proportion of those youth with limited resources who subsequently enrolled in college from 35% to 65%, almost matching the enrollment of youth with more resources (e.g., highly educated parents) who did not have teacher mentors (67%). Having a teacher-as-mentor also was associated with a much smaller increase in the likelihood of college enrollment among youth with highly educated parents (67% to 75%) (p. 359). In a subsequent analysis of Add Health data using mentoring data from Wave 3 and employment data from Wave 4 of the study when participants ranged in age from 25 to 31, McDonald and Lambert (2011) found that having a natural mentor at school or work was associated with higher status and skill level of subsequent employment, using propensity scores to control for likelihood of having a mentor. Although earnings were not higher for young adults who were mentored as youth, this could have resulted from those in lower-status and lower-skill jobs working longer hours. Moreover, the authors argued that higher-quality employment during young adulthood is likely to lead to higher earnings later in life as well as greater satisfaction, wellbeing, and stability (p. 27). Scope and Significance Work and service are normative experiences for youth. The 2009 Current Population Survey found the annual labor force participation rate of 16–19 year olds to be 37.5%, down from 52% in 2000 (not seasonally adjusted; U.S. Department of Labor, 2010). Youth frequently report finding mentors at work. Mortimer (2003) found that 71% of employed youth (in Wave 3, when most were high school juniors) reported that their supervisors were almost always or often “willing to listen to problems and help find solutions” and 38% reported feeling “extremely or quite close to their supervisors.” This confirms the expectation that workplaces can be contexts for mentoring. Call and Mortimer (2001) found that some work environments provide an “arena of comfort” for youth experiencing stress in their families. “Support from supervisors at work is found to moderate the effect of family discomfort on employed adolescents’ well-being, self-esteem, and mastery” (p. 127). In the Add Health survey, young adults were asked to name one mentor since age 14; 10% named an employer or a coworker (DuBois & Silverthorn, 2005). Vazsonyi and Snider (2008)

found that having a workplace mentor predicted improvements in young people’s psychosocial competencies and adjustment and that high school students with part-time jobs in the United States were as likely to have a workplace mentor as Swiss apprentices in a much more formal system. According to the most recent annual survey of volunteer activity conducted as a supplement to the Current Population Survey, 26% of 15–19 year olds served as volunteers in 2009. This rate generally tracks the rate for all age groups and is down from the high of 30.4% reported in 2005 (Corporation for National and Community Service, 2010). When young people aged 12–18 were questioned themselves about volunteer activity, they reported even higher rates (Corporation for National and Community Service, 2005, p. 2). We have found no data on the prevalence of mentoring in the context of service-learning, but the presence of a teacher or other adult whose role is necessarily broad and who is in regular contact with youth in a range of settings suggests the possibility of mentoring relationships in some cases. As youth engage in important tasks with adults in the community, it is also possible that some of those adults become mentors. Definitions We use the term mentor broadly. The standard definition of a mentor, posited as the ideal by most mentoring programs, gives the mentor a quasiparental role: a person who cares about a youth, has a long-term commitment, and engages in multiple activities with the youth during which he or she teaches and advises the youth about a wide range of topics (see Bronfenbrenner’s definition, quoted by S. F. Hamilton & Hamilton, 2004, p. 396, and the “functional roles” of a mentor—what a mentor does—cited by S. F. Hamilton & Darling, 1996, pp. 201–208). Note that simply being designated a mentor in the context of a program does not make an adult a mentor by this definition; that is accomplished only by the performance of mentoring roles. Mentoring’s scope and duration may vary depending on the relationship, and a youth may rely on a series of mentors over time (S. F. Hamilton & Hamilton, 2004, pp. 397–398). Work and service-learning are goal-directed activities. With reference to “purpose” in Karcher and Nakkula’s (2010) framework, work is controlled mostly by adults and serves conventional goals of adults or society; service-learning usually offers more scope for youth control and for collaboration. Mentoring in these contexts will usually have a strong instrumental focus. What distinguishes a mentor in these contexts from an adult

Work and Service-Learning   293 who is a supervisor or a teacher but not a mentor is the formation, over time, of social and emotional bonds and the extension of the relationship beyond the formal requirements of the adult’s position. In this chapter, we define work as paid employment including both formal jobs (e.g., regular hours as a retail clerk) and informal jobs (e.g., episodic babysitting or lawn mowing), as well as unpaid internships and similar arrangements in workplaces. Service-learning refers to activities that combine specific educational purposes with the aim of benefiting others. According to the National Service-Learning Clearinghouse (2011), “Service-Learning is a teaching and learning strategy that integrates meaningful community service with instruction and reflection to enrich the learning experience, teach civic responsibility, and strengthen communities” (p. 1). Teaching and Mentoring Mentors in the context of work and servicelearning are likely to emphasize teaching as their principal “functional role,” instructing youth in specific knowledge and skills required to accomplish their shared goal. School-sponsored servicelearning is usually led by a classroom teacher, but almost always includes out-of-school activities that entail less formal interactions. Service-learning conducted in youth organizations (e.g., scouts, religious organizations, 4-H) is usually led by an adult who is not a schoolteacher. Adults who are not formally charged with leadership may also become mentors, such as nonsupervisory workers and adults encountered in the course of a service-learning project. Nonetheless, the centrality of teaching is a distinctive feature of mentoring in the context of work and service. It raises questions about where the boundaries lie between teaching and mentoring and about how informal, out-of-classroom teaching occurs. These are questions we pursue below.

Theory What makes work and service promising contexts for mentoring? How may mentors enhance the developmental impact of these experiences? The ecology of human development and social learning theory provide the basis of our theoretical framework for answering these questions. We also consider some theoretical perspectives on learning outside of school, especially through apprenticeship. The ecological theory of human development (Bronfenbrenner, 1979; Bronfenbrenner & Morris, 2006) suggests that engaging in task-oriented activities

with someone who is not a parent can promote development, especially when the activity becomes more complex over time and when the balance of power shifts progressively in favor of the youth. These changes in the relationship take an extended time period. Forming deep connections between the mentor and others and with individuals across multiple settings is important from this perspective. Bandura’s (1986) social learning theory tells us that learning occurs when the learner pays attention to the model, remembers, acts, and is motivated. We refer to these subfunctions as learner engagement. The learner’s cognitive transformation of observations is also critical; designated “reflection” in the definition above, this is what differentiates service from a service-learning activity. People must make personal sense of observations for them to become memorable and to generalize specific observations as rules, principles, and mental models. Teachers not only demonstrate what to do but also explain the processes by thinking out loud. Teachers also assess and provide feedback on the learner’s performance. This sequence of teaching behaviors is incorporated in our framework for understanding workplace teaching and learning. These perspectives, our research on apprenticeship in Germany (S. F. Hamilton, 1990), and extended dialog with educators and employers in our youth apprenticeship demonstration project (M. A. Hamilton & Hamilton, 1997) have led us to distinguish three kinds of competencies that apprentices in contemporary workplaces need: technical, personal, and social. This framework offers a way to comprehend what young people can learn from work and service. Technical competence includes the knowledge and skill required to do a specific job. With the exception of computer use, which is ubiquitous, technical competencies varied widely among employers offering apprenticeships (e.g., manage accounts receivable, draw blood, or test the quality of photographic paper). Personal and social competencies refer to what it takes to enact the role of worker successfully, in addition to technical competence. Personal competencies are grouped under the headings of self-confidence, initiative, motivation, continuous improvement, and career planning. Social competencies are categorized in terms of systems, rules, teamwork, and communication. In our research, all participating employers, in such diverse fields as health care, manufacturing, and insurance, agreed on the personal and social competencies they wanted their apprentices to learn. Not only were these competencies “generic,” applicable in diverse workplaces, but they can also be acquired

294  PROGRAMS AND CONTEXTS and refined in nonwork settings, notably but not exclusively in service-learning. This framework for understanding how adults teach youth at work is also applicable to servicelearning. We see teaching in those contexts as comprising the following behaviors: demonstrate proper job performance, explain how to perform properly, explain why certain performance is needed and why it is achieved in a certain way, monitor performance, question reflectively, and engage in problem solving. Finally, Karcher and Nakkula’s (2010) distinction between goal-directed and relational foci for mentoring encounters and between developmental and instrumental styles is consistent with our position, which is that a goal-directed focus and instrumental style are starting points for mentoring relationships in work and service-learning contexts. A more relational focus and developmental style may then emerge.

Research Workplace mentoring has been studied primarily among adults, less among youth at work. (See Eby & Allen, 2008, for a treatment of both lines of research in an introduction to a journal issue devoted to linking them.) Mentoring in the context of servicelearning has not been a topic of research, however. Our discussion of the research is organized first according to principles derived from the ecology of human development and social learning theory, followed by research on teaching and learning in work contexts. Our search of the literature included use of reference lists in published works, and keyword searches of the Cornell library database and Google Scholar using combinations of mentor(ing) with youth employment, jobs, work, and service-learning. However, our review is selective, including studies that bear on the theoretically derived topics. Goal-Directed Activity The importance of goal-directed activity in mentoring within workplace and service-learning contexts gains support from the meta-analysis conducted by DuBois et al. (2002, Table II, pp. 173– 176), which found relatively stronger effects (albeit not statistically significant at p < .05) both for mentoring programs that were located in workplaces and for those that provided structured activities for youth and mentors. However, a subsequent meta-analysis by Eby, Allen, Evans, Ng, and DuBois (2008) found somewhat larger effects for academic and workplace mentoring, albeit with college students and adult

samples, respectively, compared to “youth” mentoring more generally. Halpern (2005) drew the same conclusion from his research on an after-school program for Chicago youth, and it is a major theme in his book (Halpern, 2009). In the most rigorously designed study of workplace mentoring we found, Rollin, Kaiser-Urlery, Potts, and Creason, (2003) examined the impact of workplace internships for 8th-grade students identified as being at risk for engaging in school violence. Specific results varied among the three middle schools involved, but compared to matched control groups, interns had fewer days of suspensions and other sanctions and committed fewer infractions in school. Although the authors attributed these beneficial effects to mentoring, the theoretical basis for the intervention, notably social cognitive career theory, was that work and service would “encourage realistic outcome expectations, goal-setting, and enhancement of effective work-place behaviors, such as communication, social skills building, anger management, and coping skills” (p. 405). This study supports the idea of embedding mentoring in activities that entail learning skills. Complex Activities Successful skill development in the workplace yields multiple positive outcomes for youth. Mortimer (2003) found that young workers who learned new skills on the job reported increased efficacy and reduced depression (pp. 157–158, 160). Those who reported they had more chances to learn at work while they were high school students were more likely to say the jobs they held 7 years after graduation were related to their career goals (pp. 192–193). Apparently, working at challenging jobs is an effective form of career exploration. Halpern (2009) provided many examples of youth mastering difficult work skills with support from adults. We would expect the novelty and challenge of service-learning also to provide both opportunity and motivation to learn new skills. However, most studies have evaluated academic learning, not work-related skills. Balance of Power Larson, Walker, and Pearce’s (2005) comparative study of youth-driven and adult-driven programs found advantages in both. The authors recommended a balanced approach in which adults help keep youth on track while also building their sense of ownership and experience with real responsibility. They did not trace a progressive shift over time, but their recommendation sounds like what

Work and Service-Learning   295 Bronfenbrenner (1979) called a “balance of challenge and support” (p. 288). Many of the cases described by Halpern (2009) demonstrated comparable balance. Examining “modern apprenticeships” in the United Kingdom, Fuller and Unwin (2004) found considerable reciprocity, with apprentices frequently teaching other apprentices and sometimes adult workers as well. They did not test whether reciprocity was beneficial, but their findings challenge the assumption that apprentices always play the role of novice and are only the recipients of instruction. Extended Time Mortimer (2003) concluded that the nature and quality of work determine its impact on youth and that critical aspects of work for youth include its intensity and regularity. She defined high-intensity work as 20 or more hours per week. Regular workers were employed more than 18 months during high school, excluding summers. She found developmentally optimal work experience to be low intensity but regular. Linnehan (2001) found that a work-based mentoring program was beneficial to urban high school students’ attendance and grades but only for those who were in the program for a school year. Those who were “irregular” and left the program after a semester were no different from the comparison group. Consistent with the finding that longer time periods are associated with greater impacts from mentoring (Grossman & Rhodes, 2002), one of the three quality elements identified by the Corporation for National and Community Service (2005) is service-learning activity that lasts at least one semester, as it was associated with higher self-reported impact in their study (see also Billig & Weah, 2008). Connections Across Settings Work experience is more valuable when it is connected to school learning (Stern, Finkelstein, Urquiola, & Cagampang, 1997). According to Kahne and Sporte (2008), civic participation is enhanced by service-learning but also by discussing civic and political affairs with parents and by residence in a neighborhood in which civic activity is common. Although neither of these studies addressed mentoring, they suggest that a mentor’s connections with people in other settings will enhance effectiveness, for example, family and school (see also Keller and Blakeslee’s discussion of mentoring’s effects across mentees’ social networks, this volume, Chapter 9).

Learner Engagement Both service-learning and work-based learning epitomize engaged learning (National Research Council & Institute of Medicine, 2004; Newmann, 1996). Both tend to generate enthusiasm; students describe them as more real and more consequential than classroom work. Youth report high levels of satisfaction with even routine jobs (Newman, 1999). However, little is known about what specifically teachers or mentors can do to engage youth (Blumenfeld, Kempler, & Krajcik, 2006). Cognitive Transformation The importance of reflection is a constant theme in research and theory on service-learning. Its presence distinguishes service from servicelearning (McLellan & Youniss, 2003; Stukas, Clary, & Snyder, 1999). Reflection is a means of achieving what Bandura (1986) referred to as cognitive transformation, not only observing, but thinking about what has been observed and integrating it into a conceptual whole (“personalizing”). We view reflection as something a mentor can elicit to make work and service better opportunities for learning (M. A. Hamilton & Hamilton, 2002). Teaching in the Context of Work and Service To test and refine our conceptual framework on teaching and mentoring at work, we conducted face-to-face interviews with 66 workplace mentors who were recommended as good or outstanding and whose experience was at least a full-time summer or part time for a school year. Their accounts of what and how they taught were confirmed by telephone interviews with 61 youth who had worked with these mentors. First, we asked the mentors to list the main things they tried to teach their high school interns or apprentices. This revealed their goals, which fit well into the categories of technical, personal, and social competence. Next, we asked them how they taught, beginning with an open invitation to describe their teaching, followed by probes designed to elicit stories of specific situations in which they attempted to achieve each of their goals. This gave us a picture of their teaching behavior. We found that the teaching behaviors we had identified (demonstrating, explaining how, explaining why, monitoring, questioning, and problem solving) fit their descriptions of what they did with youth. All of the mentors spoke about teaching using the first four types of teaching behaviors, which we then called “universal.” Questioning and problem solving were described less frequently (by 76% and

296  PROGRAMS AND CONTEXTS 47% of respondents, respectively), and consequently we defined these two teaching behaviors as “challenging” for mentors to perform and for protégés to respond to because they require youth to think about and apply previous learning (M. A. Hamilton & Hamilton, 2002, p. 68). The following quotation from an interview with an outstanding mentor illustrates problem solving. This mentor was describing how he worked with an intern to create an exhibit: We had to create a beach environment. We had a space within the exhibit that was 12 feet this way and 14 feet that way and went from so many inches deep at one end to zero at the other end as it ramped up the beach. “How much water is in there? How did you figure that out?” You’ve got to use approximations, you’ve got to use triangulations, trigonometry, geometry, you know, straight measuring skills, scaling skills. “How much water is in there when we lower the water level six inches when the tide goes out? Now draw that up in a picture, label all of the parts and put the dimensions in. The tide’s going to get down in five minutes, so that this exhibit doesn’t take half a day for you to see the tide go down. How many gallons of water do you have to pump? How fast does the pump have to pump that water?” This is gallons per minute, and we have all those cute conversions with 60s involved in it. So that kid sat down and figured all of that stuff out and totally tickled the hell out of himself when he realized he knows how to do this. “This is what geometry is all about. I always wondered about that.” Notice that the mentor did not know all the answers, but engaged the intern in an extended give-and-take to work them out together. We used this and other vignettes in our training of workplace mentors with the goal of increasing their use of questioning and problem-solving (see Table 19.1). Project-Based Learning The beach environment case is simultaneously an exemplar of mentoring using problem solving and an illustration of the way categories overlap. It is work-based learning, but in the context of a nonprofit organization it could also be considered service-learning. A science teacher might view it as an excellent example of project-based learning. This is a mode of instruction that is consistent with the scientific consensus on how powerful learning occurs (Bransford, Brown, & Cocking, 2000; Lave

& Wenger, 1991; Resnick, 1987; Rogoff, 1990; Rogoff & Lave, 1984). Crawford (2000) provided a detailed case study of one science teacher’s projects that illustrates the opportunities for teachers acting as mentors when they take students out of the classroom and involve them in activities with real-world implications on which they work collaboratively with students having some choice in their roles (“personalization”). Indeed, mentor is one of the ten roles that Crawford observed the teacher performing over the course of the year (pp. 931–932). Teaching and instructional design in work and service contexts can be enriched by theory, research, and good practice in project-based learning. Research Summary and Recommendations The extant research is best described as encouraging of the idea advanced in this chapter of utilizing goal-directed activities in work and service as contexts for mentoring. We know that youth find natural mentors at work. Although research has not been done on mentoring in service-learning, it appears to offer comparable opportunities. There is ample evidence that both can be beneficial to youth, provided they are of high quality. The presence of an adult mentor is surely critical to quality. Even in the workplace, however, research on mentoring is largely retrospective and based on self-reports and, thus, is subject to considerable selection and recall biases. Lacking are prospective studies that have examined programs that have provided instruction and support to mentors (e.g., in the use of the teaching behaviors in our framework), assessed whether mentoring defined as such occurred, and compared outcomes for those mentored and those not. The research encourages the hypothesis that having mentors in workplaces and service-learning programs would yield long-term educational and developmental outcomes, but does not demonstrate such effects. The size of the effects of “informal” mentors in schools, the mentorfriendly features of workplaces and service-learning programs, and the number of young people in those contexts are promising and bode well for further research and continued practice.

Practice The practice of mentoring youth in the context of work and service is underdeveloped. It usually happens informally, without explicit planning or support. The implications of our chapter apply to three

Work and Service-Learning   297 quite separate groups of practitioners: those who support mentoring; those who provide and facilitate work experiences, especially work-based learning; and those who organize and instruct service-learning experiences. We address each of these in order. Supporting Mentoring For those who advocate for and create mentoring opportunities for youth, the message of this chapter was stated at the outset: an alternative to mentoring programs that is especially promising for high-school-aged youth is creating work and servicelearning opportunities in which mentors are readily available. Mentors may or may not be assigned in these settings, especially in workplaces. Without creating matches, program staff can still actively promote mentoring through orientation, training, and support. Another approach might be orienting and supporting youth in identifying and recruiting adults to be their mentors (Hobbs, 1982). The

National Guard Youth Challenge program uses this approach, which they call Youth Initiated Mentoring (Schwartz, 2012). Work Experience Intentionally fostering mentoring in youth jobs could enhance their developmental benefits, as is done in the Hillside Work-Scholarship Connection program created by Wegmans supermarkets in Rochester, New York (see Table 19.1). Service-Learning Service-learning program organizers might take proactive steps with both youth and adults to encourage them to seek mentoring relationships and support those relationships as they develop. Resources for conducting high-quality service-learning programs are plentiful, but, like the research literature, they have little to say about relationships between youth and adults.

Table 19.1   Practice Questions, Recommendations, and Resources Key Questions

Recommendations

Resources

What work and servicelearning program structures enhance teaching and mentoring?

Review resources on project-based learning.

http://www.hightechhigh.org/ projects

Review resources on service-learning.

http://www.learnandserve.gov

Review key components of school-towork systems.

http://www.human.cornell.edu/ fldc/yis/research/publications .cfm

What orientation, training, and support can improve teaching in workplaces and service-learning?

Address specific technical competencies for the job/task.

http://www.human.cornell.edu/ fldc/yis/research/cywp.cfm

Address generic personal and social competencies. Train on •• How to use questioning and problem solving. •• How to elicit reflection. •• How to collaborate with youth to set and achieve goals.

How can adults be prepared and supported to be effective mentors to youth in these contexts?

Orient adults to roles and boundaries. Orient adults to consider their capacity and comfort level in forming social and emotional bonds with youth.

http://cywparchive.human .cornell.edu/mentoring/ training.html

Support mentors to solve problems and maintain commitment. (Continued)

298  PROGRAMS AND CONTEXTS Table 19.1  (Continued) Key Questions

Recommendations

Resources

Encourage mentors to make connections beyond their own setting, especially to the young person’s family and school. Consider the costs and benefits of matching youth with mentors in work and service-learning.

Youth may want to choose a mentor they admire, see as a role model, can talk to, etc. The person who is assigned to supervise and instruct a young person may not be the best mentor. Young people surrounded by adults may need a designated adult to turn to for encouragement and advice at the outset.

How can young people be aided in identifying potential adult mentors in work and service-learning?

Review materials to aid youth and adults in establishing mentoring relationships without being externally matched.

http://www.ngycp.org/site/ mentor

Encourage youth leadership in Are youth aware of and able to accept responsibility community projects. to take initiative for learning?

http://www .theinnovationcenter.org/

How can the value of youth Recruit and train mentors. jobs unrelated to school be enhanced with a mentoring component?

http://www.hillside.com/ ServicesDetail.aspx?id=830

Conclusion Mentoring in the context of work and servicelearning entails a higher proportion of teaching and advising than mentoring in other contexts. In workplaces, the presumed function of an adult mentor is to teach a young person how to perform a job. As a relationship develops over time, teaching moves easily into advising as the lessons turn to how to be a good worker anywhere—in our terms, adults not only teach technical competence but also advise youth to help them develop greater personal and social competence. Compelling evidence has been found that valuable mentoring occurs in workplaces. We have made the case that it can, should, and probably does occur in service-learning programs, as well. In both contexts, more explicit attention is needed, in both research and practice, to understand and strengthen mentoring as a key component of the experience for youth. Adults in both contexts should be supported in performing their mentoring

http://cywparchive.human.cornell .edu/mentoring/training.html

role and in drawing boundaries around it as necessary. Youth should be aided and encouraged to identify and recruit mentors and to attend to their teaching and advice. Broad community support is needed for mentoring in these contexts. Employers and service organizations should be informed of the critical contribution of mentor-rich environments to the development of youth, as well as the benefits to mentors and their institutions. Parents also should be involved so they understand and approve of the work and service-learning activities in which their adolescents engage, and so they can help foster the relationships their children develop with adults beyond the family.

References Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

Work and Service-Learning   299 Billig, S. H., & Weah, W. (2008). K-12 service-learning standards for quality practice. Minneapolis, MN: National Youth Leadership Council. Retrieved from http://www.nylc.org Blumenfeld, P. C., Kempler, T. M., & Krajcik, J. S. (2006). Motivation and cognitive engagement in learning environments. In R. K. Sawyer (Ed.), The Cambridge handbook of the learning sciences (pp. 475–488). New York: Cambridge University Press. Bransford, J. D., Brown, A. L., & Cocking, R. R. (Eds.). (2000). How people learn: Brain, mind, experience, and school. Washington, DC: National Academies Press. Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press. Bronfenbrenner, U., & Morris, P. (2006). The ecology of developmental processes. In W. Damon & R. M. Lerner (Vol. ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (6th ed., pp. 793–828). New York: Wiley. Call, K. T., & Mortimer, J. T. (2001). Arenas of comfort in adolescence: A study of adjustment in context. Mahwah, NJ: Erlbaum. Corporation for National and Community Service. (2005). Building active citizens: The role of social institutions in teen volunteering (Brief 1). Washington, DC: Author. Corporation for National and Community Service. (2010). Volunteering in America: Volunteering of teenagers (age 16–19). Retrieved from http://www.volunteering inamerica.gov Crawford, B. A. (2000). Embracing the essence of inquiry: New roles for science teachers. Journal of Research in Science Teaching, 37, 916–937. DuBois, D., Holloway, B., Valentine, J., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., & Silverthorn, N. (2005). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524. Eby, L. T., & Allen, T. D. (2008). Moving toward interdisciplinary dialogue in mentoring scholarship: An introduction to the special issue. Journal of Vocational Behavior, 72, 159–167. Eby, L. T., Allen, T. D., Evans, S. C., Ng, T., & DuBois, D. L. (2008). Does mentoring matter? A multidisciplinary meta-analysis comparing mentored and nonmentored individuals. Journal of Vocational Behavior, 72, 254–267. Erickson, L. D., McDonald, S., & Elder, G. H., Jr. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367.

Freedman, M. (1993). The kindness of strangers: Reflections on the mentoring movement. San Francisco: Jossey-Bass. Fuller, A., & Unwin, L. (2004). Young people as teachers and learners in the workplace: Challenging the noviceexpert dichotomy. International Journal of Training and Development, 8, 32–42. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30(2), 199–219. Halpern, R. (2005). Instrumental relationships: A potential relational model for inner-city youth programs. Journal of Community Psychology, 33, 11–20. Halpern, R. (2009). The means to grow up: Reinventing apprenticeship as a developmental support in adolescence. New York: Routledge. Hamilton, M. A., & Hamilton, S. F. (1997). Learning well at work: Choices for quality. Washington, DC: U.S. Government Printing Office. Retrieved from http:// www.human.cornell.edu/fldc/yis/research/publica tions.cfm Hamilton, M. A., & Hamilton, S. F. (2002). Why mentoring in the workplace works. In J. Rhodes (Ed.), A critical view of youth mentoring (pp. 59–89). New Directions for Youth Development: Theory, Research, and Practice, No. 93. San Francisco: Jossey-Bass. Hamilton, M. A., & Hamilton, S. F. (2005). Work and service. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 348–363). Thousand Oaks, CA: Sage. Hamilton, S. F. (1990). Apprenticeship for adulthood: Preparing youth for the future. New York: Free Press. Hamilton, S. F., & Darling, N. (1996). Mentors in adolescents’ lives. In K. Hurrelmann & S. F. Hamilton (Eds.), Social problems and social contexts in adolescence: Perspectives across boundaries (pp. 199–218). New York: Aldine de Gruyter. Hamilton, S. F., & Hamilton, M. A. (2004). Contexts for mentoring: Adolescent-adult relationships in workplaces and communities. In R. M. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology (pp. 395–428). New York: Wiley. Hobbs, N. (1982). The troubled and troubling child. San Francisco: Jossey-Bass. Kahne, J. E., & Sporte, S. E. (2008). Developing citizens: The impact of civic learning opportunities on students’ commitment to civic participation. American Educational Research Journal, 45, 738–766. Karcher, M. J., & Nakkula, M. J. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. In M. J. Karcher & M. J. Nakkula (Eds.), Play, talk, learn: Promising practices in youth mentoring (pp. 13–32). New Directions for Youth Development: Theory, Research, and Practice, No. 126. San Francisco: Jossey-Bass. Larson, R., Walker, K., & Pearce, N. (2005). A comparison of youth-driven and adult-driven youth programs: Balancing inputs from youth and adults. Journal of Community Psychology, 33, 57–74.

300  PROGRAMS AND CONTEXTS Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, UK: Cambridge University Press. Linnehan, F. (2001). The relation of a work-based mentoring program to the academic performance and behavior of African American students. Journal of Vocational Behavior, 59, 310–325. McDonald, S., & Lambert, J. (2011). The long arm of mentoring: Informal adolescent mentoring and employment outcomes in young adulthood. Unpublished paper prepared for the U.S. Department of Labor. McLellan, J. A., & Youniss, J. (2003). Two systems of youth service: Determinants of voluntary and required youth community service. Journal of Youth and Adolescence, 32, 47–58. Mortimer, J. T. (2003). Working and growing up in America. Cambridge, MA: Harvard University Press. National Research Council & Institute of Medicine. (2004). Engaging schools: Fostering high school students’ motivation to learn. Committee on Increasing High School Students’ Motivation to Learn. Board on Children, Youth, and Families, Division of Behavioral and Social Sciences and Education. Washington, DC: National Academies Press. National Service-Learning Clearinghouse. (2011). What is service learning? Retrieved from http://www.service learning.org/what-service-learning Newman, K. S. (1999). No shame in my game: The working poor in the inner city. New York: Knopf and the Russell Sage Foundation. Newmann, F. M. (Ed.). (1996). Authentic achievement: Restructuring schools for intellectual quality. San Francisco: Jossey-Bass.

Resnick, L. B. (1987). Learning in school and out. Educational Researcher, 16(9), 13–20. Rogoff, B. (1990). Apprenticeship in thinking: Cognitive development in social context. New York: Oxford University Press. Rogoff, B., & Lave, J. (Eds.). (1984). Everyday cognition: Its development in social context. Cambridge, MA: Harvard University Press. Rollin, S. A., Kaiser-Urlery, C., Potts, I., & Creason, A. H. (2003). A school-based violence prevention model for at-risk eighth grade youth. Psychology in the Schools, 40, 403–416. Schwartz, S. E. O. (2012). Youth initiated mentoring: The influence of a new model of mentoring on program outcomes in the National Guard Youth Challenge Program. Unpublished manuscript, University of Massachusetts, Boston. Stern, D., Finkelstein, N., Urquiola, M., & Cagampang, H. (1997). What difference does it make if school and work are connected? Economics of Education Review, 16, 213–229. Stukas, A. A., Jr., Clary, E. G., & Snyder, M. (1999). Service learning: Who benefits and why. Social Policy Report, 13(4), 1–19. U.S. Department of Labor. (2010). Bureau of Labor Statistics, Current Population Survey, 2009. Retrieved from http://data.bls.gov Vazsonyi, A. T., & Snider, J. B. (2008). Mentoring, competencies, and adjustments in adolescents: American part-time employment and European apprenticeships. International Journal of Behavioral Development, 32, 46–55.

PART VI Special Populations

20  ACADEMICALLY AT-RISK STUDENTS Simon Larose and George M. Tarabulsy1

Introduction Many educators consider formal and natural mentoring to be unique experiences that may prevent school dropout. It is widely held that pairing an academically at-risk student (AARS) with a wellintentioned adult, “competent” peer, or experienced teacher typically may improve the student’s adjustment. The proliferation of mentoring programs implemented in schools and communities around the world without prior careful evaluation illustrates the predominance of this view (Randolph & Johnson, 2008). Yet recent studies and meta-analyses suggest that the impact of mentoring on the development of AARSs is modest at best and can even be detrimental under certain conditions (Blinn-Pike, 2007; Eby, Allen, Evans, Ng, & DuBois, 2008; Wood & Mayo-Wilson, 2012). A comprehensive evaluation of the benefits of mentoring for AARSs first requires a clear understanding of the prevalence, determinants, and impact of the risk of academic failure. Without such insight, conclusions regarding the effectiveness of mentoring among this population will remain unclear. This chapter constitutes an important update of the chapter first published in 2005. It includes several novel elements, notably an analysis of recent theoretical perspectives on mentoring AARSs, a review of empirical studies published since 2000, and a proposition of key reflective actions for mentoring practitioners. Who are academically at-risk students? AARSs represent a heterogeneous population. There are many ways for students to be considered “at-risk”

for lower academic achievement and school dropout. AARSs may present negative internal forces acquired over the course of their development, and/ or they may have experienced negative external influences that undermine their school adjustment. For instance, negative internal forces may include behavior and/or emotional problems in early childhood (e.g., aggressiveness and hyperactivity), poor study skills in early adolescence, and career indecision in late adolescence. Indicators of negative external influences may include family or communitybased poverty, unstable family circumstances (e.g., divorce), peer rejection, and ethnic and sexual discrimination. Such negative forces and influences have the potential to alter the quantity and quality of academic support available to students by challenging parental school involvement and academic relationships with peers and teachers (Tyler & Lofstrom, 2009). Within this context, it is not surprising that AARSs are more likely to drop out of school. High school dropout continues to be an important social and economic problem in western nations. In 2005, 10% of Canadians between the ages of 20 and 24 had not graduated from high school or were not enrolled in high school (G. Bowlby, 2005). In 2007, this same statistic stood at 16% in the United Sates among those aged 16–24, for a total of 3.3 million young adults (Cataldi, Laird, KewalRamani, & Chapman, 2009). In countries of the European Union, 15% of young adults aged 18–24 had failed to obtain a high school diploma in 2007, although rates vary significantly between countries (e.g., 5% in Poland and 36% in Portugal;

1

This chapter was supported by grants from le Fonds Québecois de la Recherche sur la Société et la Culture and from the Social Humanities Research Council to the first author.

303

304  Special Populations Pour la Solidarité, 2009). In most of these countries, dropout rates are higher among boys, ethnic minorities—especially African Americans and Hispanics in the United States—and youth of lower socioeconomic status. School dropout has been associated with a number of individual and social consequences, such as receiving social aid, having physical or mental health problems, being involved in illegal activities, and becoming the parents of children who are more likely to drop out (Tyler & Lofstrom, 2009). The next section focuses on some of the theoretical perspectives on which recent AARS mentoring research has been based. These perspectives help to document research that has addressed the effectiveness of mentoring, and they elaborate on potential underlying mechanisms at the heart of the mentoring process.

Theory Approaches to mentoring research that involve AARSs may be grouped into three categories. The first category comprises atheoretical research. Such research justifies the relevance of mentoring through empirical evidence (e.g., that mentoring relationships should be sustained over a period of at least 6 months), but it does not rely on any theories or perspectives to describe mentoring processes. The second category includes research based on established theory or general models used in fields outside mentoring (e.g., attachment theory) without addressing issues specifically related to the mentoring experience (e.g., Larose, Bernier, & Soucy, 2005). The third category includes recent models inspired by mentoring research that sheds light on the mechanisms involved in the experience of being mentored (specific models; e.g., Keller, 2005). In the next section, we describe the general and specific models. General Models Attachment theory (J. Bowlby, 1982) is often cited in research on mentoring AARS (Gormley, 2008). According to this research, attachment patterns that developed over time (secure, ambivalent, or avoidant) influence a student’s likelihood of seeking a mentor or agreeing to be mentored, and thereby moderate the outcomes related to the mentoring process. Youth raised in fragile environmental contexts characterized by low parental sensitivity may have developed negative perceptions of self and others that are believed to undermine their faith in others as sources of support and in the usefulness

of using such support in crisis situations. Certain youth may actively seek help from others (i.e., ambivalent attachment) only to feel unsupported as a result of their intense focus on their emotional turmoil or inability to detach from the conflict of their family of origin. Other youth do not actively seek out support since they have come to expect dysfunctional interpersonal reactions from others and prefer to rely on themselves to overcome their difficulties (i.e., avoidant attachment). In a theoretical paper, Gormley (2008) suggested that insecure attachment styles of both mentees and mentors can significantly limit the capacity of mentoring to improve youth development (moderating effect). Prevention theories are often mentioned when attempting to identify the degree of mentoring required to promote school completion (Randolph & Johnson, 2008). Mentoring can be based on a selective prevention approach targeting students who display moderate risk. In such cases, students are identified using a series of risk and contextual factors and mentoring is provided before they encounter academic failure. Mentoring may also be offered to those who have experienced repeated academic failure (indicated prevention). The mentoring activities are then more targeted and prescriptive and often are combined with other approaches, such as tutoring, social skills development, and remedial courses. Finally, mentoring may be used to promote academic success of all students regardless of their risk level (universal prevention). Intervention in this case is often less intense and aims mainly at encouraging students’ vocational development and feelings of connectedness to their school. Given the highly heterogeneous profiles of the AARS population, we propose that the application of selective prevention strategies constitutes the most promising avenue for mentoring AARSs. Such an approach helps target negative internal and external influences before high-risk situations become overly complex and more difficult for mentors and nonprofessionals to address. Further, selective prevention strategies may provide mentors guidance in targeting the kind of support and the activities they might engage in with mentees. This hypothesis is consistent with the concept of resilience, often used in prevention research (see Cavell & Elledge, this volume, Chapter 3), which posits that youth raised in fragile environmental contexts characterized by economic and social adversity can nevertheless fare well when exposed to protective factors such as natural and formal mentoring (Randolph & Johnson, 2008). Another perspective that has shed light on AARSs’ mentoring processes is derived from

Academically At-Risk Students  305 mentoring functions theory. This theory, which was adapted for mentoring college students (Crisp & Cruz, 2009), posits four interconnected functions: (1) psychological and emotional support, (2) support for setting goals and choosing a career path, (3) academic subject knowledge support, and (4) role modeling. Proponents of this view believe that mentoring serves as a psychosocial resource that enables AARSs to fulfill their security and affiliation needs (function 1). It entails providing emotional and social support, such as listening and taking part in shared social activities. Mentoring is also thought to provide AARSs with a context that allows them to evaluate their strengths and aspirations (function 2). Coaching, challenging current plans and progress, and setting goals are examples of actions that fall under this function. For others, mentoring enables AARSs to enhance their knowledge and acquire strategies useful to academic success (function 3). Teaching study strategies and tutoring are examples of actions stemming from this function. Finally, mentoring can serve as a locus of identification (function 4). By promoting institutional values and sharing their experiences about the school, mentors provide mentees with opportunities to bond and develop attitudes and behaviors that will help them progress. Specific Models Three specific models offer interesting research avenues for understanding the mechanisms involved in mentoring AARSs. First, the youth mentoring model (Rhodes, 2005) suggests that mentoring relationships built on reciprocity, trust, and empathy promote the development of AARSs by increasing their social skills and emotional well-being (e.g., asking teachers for help), fostering their cognitive skills (e.g., critical thinking and self-awareness), and exposing them to a positive role model with whom they can identify. This theoretical approach also suggests that these “developmental gains” gradually change the dynamics between AARSs and the important people in their lives, such as parents, peers, and teachers, and that these changes account for the effects of mentoring on the subsequent academic and social adjustment of AARS. Second, the systemic model of youth mentoring (Keller, 2005) posits that mentor-mentee relationships must be understood within the larger context of all possible relationships, including those among mentors, mentees, parents, and teachers. It stipulates that these relationships are interconnected and that the activities, discussions, and emotions experienced during mentoring both influence and

are influenced by the mentee’s other relationships. The model also suggests that the actions of mentors can indirectly influence parent-adolescent and teacher-adolescent relationships. When interacting with their mentees, mentors can validate and reinforce the views of parents and teachers; provide them with autonomy support, which, in turn, can help them better manage their relationships with authority figures; or simply brighten their lives, generating positive emotions that can be transposed to other relationships. Third, unlike the two previous models, the mentoring sociomotivational model proposed by Larose and Tarabulsy (2005), which was inspired by self-determination theory (Deci & Ryan, 1985), places additional importance on mentor behaviors. According to this model, four sets of mentor behaviors are critical for improving AARS development: structure, engagement, autonomy support, and competence support. A mentor who establishes clear guidelines in terms of mentoring objectives, activities, and functioning (structure); who openly and respectfully discusses personal, academic, and career issues with the AARS (engagement); who accepts and validates the AARS’s personal choices without exercising any control or pressure (autonomy support); and who is able to increase the AARS’s feelings of competence following negative experiences (competence support) should develop a more productive relationship with the student. The resulting positive relationship is believed to foster AARSs’ feelings of competence, relatedness, autonomy, and support, which might improve their social and academic adjustment. Both general and specific models provide promising avenues for research and intervention with AARSs. First, they suggest the existence of several interactive psychological processes whose effects may influence relationships other than those between AARSs and mentors. These processes include identification, attachment, critical reasoning, self-reflection, perceived social and instrumental support, resilience, and satisfaction of motivational needs. Such processes should be part of targeted experimental and longitudinal investigations in future mentoring research. Second, several models suggest the importance of multilevel interventions (with at-risk students, but also with their parents, teachers, and peers) and organized actions that reflect the rules and values of the significant individuals in the lives of AARSs. They propose that mentors’ actions within the mentoring relationship could generate positive collateral effects, which could result in a less fragile environmental context for AARSs. For

306  Special Populations example, promoting the expectations of parents and teachers within the mentoring context may improve relationships between AARSs and their parents and teachers and, possibly, encourage parents to become more involved in their youth’s education, and teachers to build stronger interpersonal bonds with the mentored youth.

open to formal mentoring program participation. Such findings give us a better sense of the youth who are most likely to form natural mentoring relationships and helps identify those who will need assistance to establish a formal mentoring relationship through a program. The challenge, then, is to identify strategies to attract to mentoring those with limited personal resources who might also benefit from mentoring.

Research Here, we review findings of empirical studies that (a) focus on the determinants of student involvement in mentoring, (b) address theoretical predictors of mentoring relationship quality for AARSs and other student populations, and (c) explore several hypothesized effects and processes driving the mentoring experience of AARSs. The studies reviewed have, for the most part, been identified through the usual databases, PsycLit and ERIC, with the following keywords: youth, mentoring, student, at-risk student, high school, college, academic adjustment, and performance. All empirical studies published since 2000 that dealt with formal or informal mentoring and included samples of elementary, high school, and/or college students were considered. Determinants of Participation in Mentoring In a prospective study, Larose and colleagues (2009) showed that low-risk students with the most personal resources (i.e., positive help-seeking attitudes, perceived support from friends and school motivation), but fewer environmental resources (low maternal income and education, leaving home to attend college, belonging to a family of recent immigrants, not having any siblings who previously studied in college), were more likely to accept the support of a formal mentor upon entering college. These observations are in part similar to those reported in a longitudinal study on natural mentoring involving over 12,000 American youth (Erickson, McDonald, & Elder, 2009), which found that students with the most personal resources (e.g., personality, physical appearance, and college aspirations) and environmental resources (e.g., parental income and education, friends, peer network centrality) were more likely than other students to have had a mentor in their lives. These findings suggest that youth who are the least socially and academically equipped have a lower chance of crossing paths with a natural mentor or of accepting formal mentoring opportunities. They also propose that youth with modest personal resources living in difficult circumstances may be

Determinants of Mentoring Relationship Quality for AARSs We know that mentor backgrounds in helping professions (DuBois, Holloway, Valentine, & Cooper, 2002), mentor perceptions of their own competence as mentors (i.e., mentor efficacy; Karcher, Nakkula, & Harris, 2005), and the previous attachment patterns of AARSs (Larose, Bernier, & Soucy, 2005) are personal factors that affect mentor relationship quality (MRQ). The quality of training and supervision offered to mentors (Rhodes & DuBois, 2006), the possibility of choosing one’s mentor (Kendall, 2007), and the duration and frequency of relationships (Grossman & Rhodes, 2002) are significant organizational parameters that positively influence MRQ. The content and manner of negotiating mentoring activities also contribute to MRQ, but the latter process seems more critical. A recent study on mentoring of low-risk college students (Larose, Cyrenne, Garceau, Brodeur, & Tarabulsy, 2010) found that MRQ was positively influenced by a set of mentor behaviors that combined emotional engagement and reciprocity with some directivity (i.e., a more authoritative style). In this study, MRQ was not found to be affected by specific mentoring content (i.e., discussing academic issues, doing specific activities, or trying to solve specific problems). Similarly, in a study of Big Brothers Big Sisters (BBBS) school-based mentoring students, Karcher, Herrera, and Hansen (2010) reported unique contributions made to MRQ by both relational and goaldirected activities. They reported that collaboratively negotiated activities of any kind were best for student outcomes. Taken together, these studies suggest that what may matter most for establishing positive bonding with AARSs is not so much the specifics of the mentoring activities per se, but rather the manner in which mentors and mentees choose the experiences in which they engage. The similarity of interests between mentors and AARSs also appears to be a determinant of MRQ. A research synthesis of mentoring programs intended for different categories of high-risk adolescents

Academically At-Risk Students  307 showed that AARSs who perceived high levels of similarity with their mentors in terms of vision, perspective, and values, but not in terms of demographics, reported greater liking and satisfaction with their mentors (Sipe, 2002). Alternatively, one study has shown that students considered at risk because of low high school grades displayed more adaptive behaviors and perceptions in mentoring and earned higher grades when their attachment orientation to their parents was in contrast to their mentor’s relational style (Bernier, Larose, & Soucy, 2005). AARSs presenting dismissing attachment tendencies benefited more from working with mentors who valued dependency, relationships, and closeness, whereas AARSs presenting preoccupied/anxious attachment tendencies were more adapted when working with mentors who valued self-reliance, achievement, and autonomy. Providing the AARS with a challenging relational stance that is not in line with the student’s own seems to reinforce a process of exploration and change that may positively affect the course of the relationship. Effects and Explanatory Processes of Mentoring Formal Mentoring. In the past 15 years, many independent—sometimes small-scale—experimental and quasi-experimental studies have suggested that formal mentoring of AARSs can lead to changes in a number of cognitive, emotional, and behavioral outcomes. For example, formal mentoring has been found to improve attitudes toward school and helping, academic confidence, school connectedness, perceptions of parental and teacher relationships, vocational and reading skills, participation in college preparatory activities, and persistence in college (see Blinn-Pike, 2007; Eby et al., 2008, for reviews). Yet, conclusions from meta-analyses suggest that these findings must be interpreted with caution. In fact, the reported effects of formal mentoring were modest, with Cohen’s d coefficient rarely exceeding .20 (Blinn-Pike, 2007; DuBois, Holloway et al., 2002). Two such analyses even suggested that the outcomes of formal mentoring for youth are much more limited than those associated with workplace mentoring and volunteer tutoring (Eby et al., 2008; Ritter, Denny, Albin, Barnett, & Blankenship, 2006). Most studies of the impact of formal mentoring involve short-term evaluations. Only three studies followed up samples after the end of the mentoring program. The first evaluated the Quantum Opportunity Program (QOP), an intense case management and mentoring after-school program for lowachieving students (Rodriguez-Planas, 2009). The

mentoring component involved assigning 15–25 youth to case managers with whom they were expected to develop trusting relationships. Case managers were also required to establish links with schools, families, and friends. In addition to mentoring, QOP offered students developmental activities, community and educational services, and financial incentives. Using a randomized controlled trial, the authors showed modest short-term gains on high school graduation and on the pursuit of postsecondary studies, especially among younger adolescents, girls, and lower achieving students. However, these effects decreased over time, and 5 years after the program, the QOP had no detectible impact on educational or employment outcomes for its participants. The second study evaluated the long-term effects (15 months after the match) of the BBBS school-based mentoring program using a randomized controlled group design and involving the participation of 10 agencies in the United States (Herrera et al., 2007). Researchers followed AARSs in primary and high school, where 52% of the experimental group students continued to receive mentoring after 1 year. The study showed no longterm impact of the program on the majority of the academic (e.g., general academic performance, GPA, quality of class work) and nonacademic (e.g., substance use, social acceptance, and relationship with parents) outcomes examined. The challenge of proving long-term effects was also documented for AARSs at early elementary levels. In a prevention program for aggressive elementary school students, Hughes, Cavell, Meehan, Zhang, and Collie (2005) compared the impact of two mentoring programs that lasted three semesters and involved college students as mentors. The first program (PrimeTime) combined community-based mentoring (weekly visits) with a focus on child skills training and consultation for parents and teachers. The second program (Lunch Buddy) was a stand-alone, school-based mentoring program that involved two visits a week during lunch times in the presence of nonmentored peers, with a different mentor being assigned each semester. Using a randomized controlled trial, the authors found that in the short term, both programs reduced children’s externalizing problems and increased their levels of academic and behavioral skills. Surprisingly, at the 1- and 2-year follow-ups, the effects observed among participants of the PrimeTime program had faded completely, whereas Lunch Buddy participants were evaluated by their teachers as displaying a higher skill level and fewer externalizing problems. This was found to be the case even though the

308  Special Populations mentoring relationship of the Lunch Buddy program was of lower intensity than that of the PrimeTime program (i.e., multiple shorter matches rather than one longer match). Factors related to the quality of interactions with peers were used to explain these findings. To summarize, the studies that have examined the long-term effects of formal mentoring suggest that acquired progress is difficult to maintain. Various factors, including the initial characteristics of students and mentors, the intensity and degree of contact, or the nature of the relationship forged between the mentor and the mentee, may explain this situation. In this context, it is pertinent to examine the factors that potentially buffer or emphasize the effects of formal mentoring on AARSs (i.e., moderating factors). Moderating Processes. Although risk level has often been considered an important moderator of mentoring effects, research findings in this area are mixed. Stronger effects for high-risk students on social and behavioral adjustment indicators were reported when student risk was estimated using academic indicators (e.g., low GPA, high absence rate from school; Rodriguez-Planas, 2009; Whiting & Mallory, 2007). Some studies also suggested that boys with low academic scores are especially responsive to formal mentoring (Whiting & Mallory, 2007). Other studies have found that mentoring was helpful for all students, but that risk moderated mentoring outcomes. For example, in a study evaluating a program that used both peer leaders and adult mentoring, high-risk students (grades below 70% and/or more than eight school absences) improved their ability to resist peer pressure during the transition to high school, whereas low-risk students enhanced their capacity for making friends (Holt, Bry, & Johnson, 2008). Conversely, MorrowHowell, Jonson-Reid, McCrary, Lee, and Spitznagel (2009) showed that the effect of formal mentoring on reading comprehension was less pronounced for a higher risk, special education group of students than for non–special education students, suggesting that the impact of mentoring may be less evident when a student’s risk level presents greater psychological or learning challenges. Finally, one study found that when the risk was based on students’ feelings of being disconnected from school, having a mentor who held positive views regarding youth behaviors led AARSs to be more emotionally engaged in the mentoring relationship and, subsequently, to report stronger relationships with their teachers (Karcher, Davidson, Rhodes, & Herrera, 2010). The hypothesis that greater psychological or learning difficulties moderate youth mentoring

outcomes is consistent with studies that included emotional indicators in their definition of risk. Among students with high school grades that were low, but sufficient for college admission, those who reported more insecure attachment relationships to their parents were less likely to hold positive perceptions of mentoring and, subsequently, more likely to report high levels of conflict with their teachers (Larose, Bernier, & Soucy, 2005). This latter finding underlines the challenge of mentoring AARSs when emotional factors are considered. Certain characteristics of mentors themselves also seem to play an important role in moderating the effects of mentoring. The effects of mentoring AARSs appear to be greater when mentors feel they have the competence to help (Parra, DuBois, Neville, Pugh-Lily, & Povinelli, 2002), hold highly positive attitudes toward youth (Karcher, Davidson, et al., 2010), and express high motivation for selfenhancement (Karcher et al., 2005), as well as when they have a background in a helping profession (DuBois, Holloway et al., 2002), are not poor, and are not married (Grossman & Rhodes, 2002). Studies also suggest that high school and college student mentors have more difficulty generating positive results among at-risk mentees than adult mentors or teachers (Converse & Lignugaris/ Kraft, 2009; Hughes et al., 2005; Whiting & Mallory, 2007). Some contextual factors also play a role in program effectiveness. These factors include initial and ongoing structured training for mentors, monitoring of program implementation, and parental involvement (DuBois, Holloway et al., 2002). In addition, programs viewed as effective allow for more extended contact between mentors and AARSs, engage AARSs in social and academic activities, and structure interactions in such a way as to favor the development of AARS autonomy in decision making (Blinn-Pike, 2007; Morrow-Howell et al., 2009). Mediating Processes. Although several models suggest the presence of different mediating processes to explain mentoring outcomes (see the “Theory” section, above), such processes have rarely been studied. Rhodes and colleagues (Rhodes, Grossman, & Resch, 2000; Rhodes, Reddy, & Grossman, 2005) found that improvements in the quality of relationships with parents mediated the impact of the BBBS community match programs on youth self-worth, school value, and grades, as well as on peer relationships, but only for youth involved in relationships that lasted more than 12 months. This study included 959 adolescents and used a randomized controlled trial design. The authors claimed that

Academically At-Risk Students  309 this mediating process may have been the result of a gradual change in mentee representation models of relationships with parents or a decrease in the tension experienced between youth and their parents. They also suggested that short-term mentoring does not provide sufficient opportunity for the development of secure relationships. Similarly, other reports revealed that perceived connectedness to parents and perceptions of support from significant adults outside mentoring act as mediators of the link between formal mentoring and mentee academic adjustment ((DuBois, Neville, Parra, & Pugh-Lilly, 2002; Karcher, Davis, & Powell, 2002). Two quasi-experimental studies explored the role of interpersonal processes in mentoring as mediators of AARS outcomes. The first study was conducted with students who had failed more than half of their first semester courses. Results showed that mentees who had a positive working alliance with their mentors (i.e., agreement on goals, positive bonding) were more likely to improve their academic competence, participation in class, tendency to seek help from teachers, and academic perseverance than were mentees in less collaborative mentoring relationships or students in a control group (Larose, Monaghan, Chaloux, & Tarabulsy, 2010). The second study, conducted with low-achieving students, indicated that youth-perceived autonomy and relatedness support in a teacher-student mentoring relationship led to better academic adjustment (Larose, Tarabulsy, & Cyrenne, 2005). Such studies support the argument laid out in several models, suggesting that the impact of mentoring on dropout prevention may be explained in part by the satisfaction of motivational needs, the quality of the mentoring relationship, and an improvement in youth relationships. Natural Mentoring. Studies on the impact of natural mentoring of AARSs are much less common than those focused on formal mentoring. Here, we present the conclusions of one of the largest longitudinal studies conducted in the United States on natural mentoring and education (Erickson et al., 2009). Beyond the effect of personal and environmental resources (e.g., parental income and education, number of friends, school size, physical appearance, and personality), natural mentoring was positively associated with academic performance in high school and with youths’ subsequent educational status (i.e., highest degree achieved). For youth with existing resources, natural mentoring with relatives (e.g., brother, sister, grandparent) had a positive effect on educational status. When these resources were limited, mentoring by a teacher was identified as having the most positive

impact. This last finding is consistent with observations made by DuBois and Silverthorn (2005), indicating that at-risk youth with nonfamilial mentors were more likely to complete high school than those who identified a familial mentor. In addition to having an impact on AARSs in the classroom, teachers appear to be meaningful models for preventing student dropout among this population. The number of natural mentors, their characteristics, and the type of relationships they build with their mentees also play a significant role in preventing academic difficulties. In a correlational study among 140 Latinos from an urban public high school serving a predominantly low-income student population (Sánchez, Esparza, & Colon, 2008), the presence of a natural mentor was associated with fewer absences in class, greater educational expectations, and a strong sense of school belonging. Further, the number of reported mentors was positively related to academic outcomes. Finally, mentors’ education, frequency of contact, relationship length, and total form of support provided by mentors were related positively to student academic outcomes. In sum, research on mentoring AARSs over the past 15 years has mainly addressed the shortterm impact of mentoring and its moderating processes. Research on natural mentoring is clearly less abundant than that on formal mentoring. As such, many issues will require further careful analysis in future works: What factors attract youth to formal mentoring? How can mentoring reach the most emotionally at-risk youth? Who is best positioned and skilled to support AARSs? What are the long-term effects of formal and natural mentoring? What developmental mechanisms account for these long-term effects? Future research must account for several important methodologic parameters such as longitudinal follow-up; the systematic quantification of effect sizes; the use of randomized designs, including various “mentoring” conditions; and controlling for confounding factors before and during the mentoring experience.

Practice This review draws out the need to implement mentoring programs with great care and by considering an important number of parameters. This section describes a series of key reflective actions that we perceive to be important in designing, implementing, and managing mentoring programs for AARSs (see Table 20.1). One such action specifically concerns natural mentoring.

310  Special Populations Table 20.1   Checklist for Practitioners Recommendations Promote formal mentoring among academically at-risk students (AARSs) with fewer personal resources (e.g., negative help-seeking attitudes, low school motivation) by targeting them in recruitment efforts (both seeking them out and promoting their participation through general invitation messages). Quickly identify the factors involved in a youth’s lack of academic success (e.g., personal, contextual) and determine the nature and level of risk, by using diagnostic profiles. Attract mentors who are teachers or who have experience in the helping professions, such as is currently practiced in the Achievement Mentoring Program (Holt et al., 2008). Offer mentors comprehensive initial training that aims to achieve the following goals: (a) identify the protective and risk factors related to school dropout; (b) enhance mentors’ abilities to empathize, to be authentic, and to collaborate; (c) learn how to practice efficient academic problem solving; and (d) understand the potential effects of students’ academic and interpersonal profiles on their expectations, emotions, and behaviors throughout the mentoring process. Encourage mentors to intervene with parents and teachers (e.g., serving as an advocate) by informing them about program objectives, rules and policies, mentor profile and role, and program expectations with regard to parent and teacher collaboration. Allow AARSs to select their mentors or match them based on shared academic or vocational interests. Offer mentors support and supervision during the intervention and foster their perception of competence. Providing the mentor with early feedback on the mentee’s view of the match may serve to bolster mentor efficacy. Adjust the quantity of mentoring based on the student’s level of risk, and be clear about the expected duration of the relationships (e.g., one year, one semester, longer). Establish mechanisms to support mentoring relationships over multiple years, ideally across key normative school transitions (e.g., middle school to high school). Provide mentors and AARSs with support to fulfill this goal. Create academic cultures that value natural mentoring between adults and AARSs through strategies such as social marketing, focus groups, and social network connections.

Program Promotion, AARS Screening, and Mentor Selection As shown in the “Research” section (e.g., Larose et al., 2009), students who lack personal resources are not as likely to have the opportunity to be mentored or to agree to take part in a formal program. Thus, program managers may wish to implement promotional approaches that will help AARSs understand that mentoring is not a threat, but rather a useful tool that may fulfill personal and academic needs. Furthermore, advertising the instrumental functions (e.g., help with school work) and vocational functions (e.g., enhanced knowledge

of the job market) of mentoring may attract more AARSs, particularly some boys who may feel uncomfortable in close relationships. Few programs establish diagnostic profiles of the youth they seek to help. Yet doing so might be helpful to identify the degree and origin of an AARS’s difficulties and, therefore, better prepare the mentor to adapt the quantity of mentoring provided. Even major national mentoring programs, such as the QOP or the BBBS school-based mentoring program, were not able to generate meaningful long-term outcomes on the academic success and educational status of AARSs, indicating the need to

Academically At-Risk Students  311 better understand and be responsive to the needs of participating students, potentially improving the intervention’s long-term effect. The reviewed studies also underline the importance of choosing teachers as mentors for AARSs (e.g., Converse & Lignugaris/Kraft, 2009) and questions the effectiveness of programs that call on high school and college mentors (e.g., Whiting & Mallory, 2007), except those who are especially well suited to serve as mentors. Although younger mentors should not be excluded from mentoring programs devoted to AARSs, they should be selected carefully based on maturity, social skills, and interpersonal abilities. In our view, young mentors should also be provided with substantial training and supervision. It may be helpful to have them mentored themselves by more experienced adult mentors. Mentor Training and Supervision Ten years ago, some studies estimated that nearly half of the mentoring programs implemented in schools contained no more than 2 hours of mentor training (see Larose & Tarabulsy, 2005). A recent national survey on practices of BBBS programs in the United States suggested that the situation has not improved significantly since that time. In fact, only 1% of the agencies surveyed reported providing their mentors more than 4 hours of initial training (Big Brothers Big Sisters of America, 2009). Initial training and mentor supervision are key practices that should not be neglected. Fortunately, an increasing number of programs devoted to AARSs are investing equal amounts of resources in training and supervising mentors. For example, the MIRES program (Larose et al., 2011), which aims to help students complete science programs at the college level, offers 2 days of initial training, as well as several individual and group meetings throughout the mentoring program. It also features a hands-on component, whereby mentors practice interacting with a mentee through simulated videotaped scenarios under the supervision of instructors. The program and mentoring meetings are based on the sociomotivational model of mentoring. Mentors learn how to structure mentoring relationships on instrumental and emotional levels, and to support mentee autonomy. Although the content of training sessions might vary significantly from one mentoring program to another, research suggests that certain goals should be prioritized when intervening with AARSs (Crisp & Cruz, 2009; Keller, 2005; Larose, Cyrenne et al., 2010; Tyler & Lofstrom, 2009). These goals

are as follows: to identify the protective and risk factors related to school dropout; to enhance mentors’ abilities to empathize, to be authentic, and to collaborate; to learn how to practice efficient academic problem solving; to understand the potential effects of students’ academic and interpersonal profiles on their expectations, emotions, and behaviors throughout the mentoring process; and to intervene with parents, peers, and teachers. We also believe that supervision should both equip mentors to respond to the timely needs of mentees and aim at consolidating their perceived competence in their ability to be helpful. Supervisors may use modeling and constructive feedback to achieve these goals in their meetings with mentors. Matching Mentors and Mentees Research suggests that matching AARSs and mentors based on similarity of views helps to strengthen the quality of the bond (Sipe, 2002). Allowing AARSs to select their mentors appears to be a promising strategy to achieve this goal (Kendall, 2007). However, freely choosing a mentor implies complex logistics and a level of resources that most organizations do not have. Matching AARSs and mentors must therefore be based on criteria that respect both the mentee and the resources available to the mentoring organization, in a way that will maximize the quality of the future relationship. Such criteria may include personal, social, and vocational interests; mentoring expectations; and views about the involvement of parents and teachers in the mentoring process. The Mentoring Relationship Ensuring frequent, high-quality contact is a recommendation that applies to all mentoring programs, regardless of the specific clientele targeted. In light of the findings on the long-term effects of mentoring AARSs (e.g., Herrera et al., 2007), the issues that must be raised concern the intensity and responsiveness of the specific mentoring experience. Should a youth experiencing a difficult single-parent context and one who has had behavioral problems since kindergarten be offered the same type of mentoring? At the elementary level, does a student with reading and writing difficulties not require a more specialized mentor than another experiencing anxiety about the transition to high school? Is 1 year of mentoring sufficient to lessen the impact of a highrisk situation that, in some cases, may have begun in early childhood? In our opinion, the quantity of mentoring must be adjusted to the level of risk for

312  Special Populations academic failure, and in certain cases, an offer of mentoring over several years might be considered. Reinforcing the link between the mentor and the AARS’s teacher is an aspect of mentoring practice that merits further attention. This type of action is at the core of several intervention and mentoring programs, such as the Check and Connect Program (Sinclair, Christenson, & Thurlow, 2005) and the Achievement Mentoring Program (Holt et al., 2008), intended for primary and high school AARSs. Within such programs, mentors are expected not only to conduct a systematic followup of mentees’ academic behavior and achievement, but also to be in constant contact with the mentees’ teachers. In this context, the mentor becomes a kind of mediator, informing teachers of mentee progress, so that teachers may be sensitive to these aspects and reinforce this progress within the classroom setting. Natural Mentoring Natural mentoring is a unique experience that benefits the youth involved. Unfortunately, those who need it most have fewer opportunities to access it. This knowledge alone justifies the importance of implementing mentoring programs. It also calls upon schools and communities to consider the predominant culture in their environments and adopt measures that will foster the creation of natural mentoring relationships between adults and youth.

Conclusion In conclusion, many scientists and practitioners view mentoring as a preventive action having the potential to support academic functioning in AARSs and reduce the risk of dropping out of school (Tyler & Lofstrom, 2009; Wheeler, Keller, & DuBois, 2010). However, the causal relationship between taking part in mentoring programs and the reduction in high school dropout is not as convincing as it might be. The body of knowledge reviewed in this chapter is vast, though highly focused on the shortterm effects of programs implemented in schools and communities. Future research would do well to address the long-term outcomes of mentoring relationships and examine the dynamics within these relationships (e.g., mentor strategies, mentee behaviors, conflicts, activities). In pursuing both of these objectives, the highly heterogeneous profiles of atrisk students should also be considered. This will no

doubt lead to more comprehensive and nuanced intervention models that will better meet the needs of the AARS population.

References Bernier, A., Larose, S., & Soucy, N. (2005). Academic mentoring in college: The interactive role of student’s and mentor’s interpersonal dispositions. Research in Higher Education, 46, 29–51. Big Brothers Big Sisters of America, compiled by Wheeler, M., & DuBois, D. L. (2009). Analysis of responses to agency practices survey for Big Brothers Big Sisters of America’s community-based mentoring program. Unpublished manuscript. Blinn-Pike, L. (2007). The benefits associated with youth mentoring relationships. In T. D. Allen & L. T. Eby (Eds.), The Blackwell handbook of mentoring: A multiple perspective approach (pp. 165–187). Malden, MA: Blackwell Publishing. Bowlby, G. (2005). Provincial drop-out rates: Trends and consequences. Education Matters: Insights on Education, Learning and Training in Canada, 2(4). Bowlby, J. (1982). Attachment and loss: Attachment (Vol. 1., 2nd ed.). New York: Basic Books. Cataldi, E. F., Laird, J., KewalRamani, A., & Chapman, C. (2009). High school dropout and completion rates in the United States: 2007 compendium report (No. NCES 2009-064). Washington, DC: U.S. Department of Education, National Center for Education Statistics. Converse, N., & Lignugaris/Kraft, B. (2009). Evaluation of a school-based mentoring program for at-risk middle school youth. Remedial and Special Education, 30, 33–46. Crisp, G., & Cruz, I. (2009). Mentoring college students: A critical review of the literature between 1990 and 2007. Research in Higher Education, 50, 525–545. Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. New York: Plenum Press. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth, a meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. In G. G. Noam (Ed. in chief) & J. E. Rhodes (Ed.), A critical view of youth mentoring (New Directions for Youth Development: Theory, Research, and Practice, No. 93, pp. 21–57). San Francisco: Jossey-Bass. DuBois, D. L., & Silverthorn, N. (2005). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92. Eby, L. T., Allen, T. D., Evans, S. C., Ng, T., & DuBois, D. L. (2008). Does mentoring matter? A multidisciplinary meta-analysis comparing mentored and non-mentored

Academically At-Risk Students  313 individuals. Journal of Vocational Behavior, 72, 254–267. Erickson, L. D., McDonald, S., & Elder, G. H., Jr. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367. Gormley, B. (2008). An application of attachment theory: Mentoring relationship dynamics and ethical concerns. Mentoring & Tutoring: Partnership in Learning, 16, 45–62. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Herrera, C., Baldwin Grossman, J., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a difference in school: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Holt, L. J., Bry, B. H., & Johnson, V. L. (2008). Enhancing school engagement in at-risk, urban minority adolescents through a school-based, adult mentoring intervention. Child & Family Behavior Therapy, 30, 297–318. Hughes, J. N., Cavell, T. A., Meehan, B., Zhang, D., & Collie, C. (2005). Adverse school context moderates the outcomes of two selective intervention programs for aggressive children. Journal of Consulting and Clinical Psychology, 73, 731–736. Karcher, M. J., Davidson, A. J., Rhodes, J. E., & Herrera, C. (2010). Pygmalion in the program: The role of teenage peer mentors’ attitudes in shaping their mentees’ outcomes. Applied Developmental Science, 14, 212–227. Karcher, M. J., Davis, C., & Powell, B. (2002). The effects of developmental mentoring on connectedness and academic achievement. School Community Journal, 12, 35–50. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. In M.J. Karcher & M. J. Nakkula (Eds.), Play, Talk, Learn: Promising Practices in Youth Mentoring (pp. 51-70). Belmont, MA: Jossey-Bass. Karcher, M. J., Nakkula, M. J., & Harris, J. (2005). Developmental mentoring match characteristics: Correspondence between mentors’ and mentees’ assessments of relationship quality. Journal of Primary Prevention, 26, 93–110. Keller, T. E. (2005). A systemic model of the youth mentoring intervention. Journal of Primary Prevention, 26, 169–188. Kendall, D. L. (2007). Does choice matter? The impact of allowing proteges to select their own mentors. Doctoral dissertation, University of Central Florida. Larose, S., Bernier, S., & Soucy, N. (2005). Attachment as a moderator of the effects of security in mentoring on subsequent perceptions of mentoring and relationship quality with college teachers. Journal of Social and Personal Relationships, 22, 399–415.

Larose, S., Cyrenne, D., Garceau, O., Brodeur, P., & Tarabulsy, G. (2010). The structure of effective academic mentoring in late adolescence. New Directions for Youth Development, 2010 126, 123–140. Larose, S., Cyrenne, D., Garceau, O., Harvey, M., Guay, F., & Deschênes, C. (2009). Personal and social support factors involved in students’ decision to participate in formal academic mentoring. Journal of Vocational Behavior, 74, 108–116. Larose, S., Cyrenne, D., Garceau, O., Harvey, M., Guay, F., Godin, F., . . . & Deschênes, C. (2011). Academic mentoring and dropout prevention for students in math, science and technology. Mentoring & Tutoring: Partnership in Learning, 19, 419–439. Larose, S., Monaghan, D., Chaloux, N., & Tarabulsy, G. (2010). Working alliance as a moderator of the impact of mentoring relationships among academically at-risk students. Journal of Applied Social Psychology, 40, 2656–2686. Larose, S., & Tarabulsy, G. (2005). Academically at-risk students. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. (pp. 440–453). Thousand Oaks, CA: Sage. Larose, S., Tarabulsy, G., & Cyrenne, D. (2005). Perceived autonomy and relatedness as moderating the impact of teacher-student mentoring relationships on student academic adjustment. Journal of Primary Prevention, 26, 111–128. Morrow-Howell, N., Jonson-Reid, M., McCrary, S., Lee, Y., & Spitznagel, E. (2009). Evaluation of Experience Corps, student reading outcomes (CSD Research Report 09-01). St. Louis, MO: Washington University, Center for Social Development. Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. Pour la Solidarité. (2009). L’Union européenne s’intéresset-elle au décrochage scolaire? État des lieux et perspectives en Europe. Série Affaires sociales. Randolph, K. A., & Johnson, J. L. (2008). School-based mentoring programs: A review of the research. Children and Schools, 30, 177–185. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois and M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43). Thousand Oaks, CA: Sage. Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20, 3–20. Rhodes, J. E., Grossman, J. B., & Resch, N. L. (2000). Agents of change: Pathways through which mentoring relationships influence adolescents’ academic adjustment. Child Development, 71, 1662–1671. Rhodes, J. E., Reddy, R., & Grossman, J. B. (2005). The protective influence of mentoring on adolescents’

314  Special Populations substance use: Direct and indirect pathways. Applied Developmental Science, 9, 31–47. Ritter, G., Denny, G., Albin, G., Barnett, J., & Blankenship, V. (2006). The effectiveness of volunteer tutoring programs: A systematic review. Campbell Collaboration Reviews of Intervention and Policy Evaluations (C2-RIPE). Philadelphia: Campbell Collaboration. Rodríguez-Planas, N. (2009). Longer-term impacts of mentoring, educational services, and incentives to learn. Retrieved from http://www.cemfi.es/ research/ Sánchez, B., Esparza, P., & Colon, Y. (2008). Natural mentoring under the microscope: An investigation of mentoring relationships and Latino adolescents’ academic performance. Journal of Community Psychology, 36, 468–482. Sinclair, M. F., Christenson, S. L., & Thurlow, M. L. (2005). Promoting school completion of urban secondary youth with emotional or behavioural disabilities, Exceptional Children, 71, 465–482.

Sipe, C. L. (2002). Mentoring programs for adolescents: A research summary. Journal of Adolescent Health, 31, 251–260. Tyler, J. H., & Lofstrom, M. (2009). Finishing high school: Alternative pathways and dropout recovery. The Future of Children, 19, 77–103. Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials of schoolbased mentoring: Making sense of mixed findings. Social Policy Report from SRCD, 24(3), 1–26. Whiting, M. A., & Mallory, J. E. (2007). A longitudinal study to determine the effects of mentoring on middle school youngsters by nursing and other college students. Journal of Child and Adolescent Psychiatric Nursing, 20, 197–208. Wood, S., & Mayo-Wilson, E. (2012). School-based mentoring for adolescents: A systematic review and meta-analysis. Research on Social Work Practice. Retrieved from http://rsw.sagepub.com/content/early/ 2012/01/03/1049731511430836

21  JUVENILE OFFENDERS Wing Yi Chan and David B. Henry

Introduction Recent data have indicated decreases in juvenile arrest rates, but juvenile delinquency continues to be a serious concern in the United States (Puzzanchera, 2009). In 2008, the juvenile murder arrest rate was 3.8 arrests per 100,000 juveniles. In the same year, juveniles accounted for 16% of all violent crime arrests and 26% of all property crime arrests. The mission of the Office of Juvenile Justice and Delinquency Prevention (OJJDP) is to “develop and implement effective and coordinated prevention and intervention programs . . . and to provide treatment and rehabilitative services tailored to the needs of juveniles and their families” (OJJDP, 2010a). OJJDP provides a list of model programs, one of which is mentoring (OJJDP, 2010b). However, research on the effectiveness of mentoring for juvenile offenders is inconclusive. The role of natural mentors needs further exploration, the long-term effects of mentoring are inconclusive, the mechanism of change has yet to be defined, and the effects of mentoring for female and minority juvenile offenders have not been investigated fully. We explore these topics in this chapter. In this chapter, we review relevant theory and the latest empirical literature on mentoring of juvenile offenders and offer suggestions for future research and practice. We expand on the same chapter in the previous edition (Blechman & Bopp, 2005) in three ways. First, we explore the role of natural mentors and how naturally occurring mentoring poses both opportunities and challenges in furthering our knowledge about mentoring of juvenile offenders. Second, we focus on the promise of a systemoriented approach for delivering effective mentoring programs for juvenile offenders. Finally, we offer

recommendations for practitioners who work to deliver mentoring programs to juvenile offers.

Theory Does mentoring have potential to reduce recidivism and promote positive transition after release among juvenile offenders? If mentoring of juvenile offenders is an effective strategy, how does it work? In this section, we apply the resilience framework to suggest why and how mentoring can be a theoretically effective intervention for juvenile offenders. Resilience is defined as “achieving good outcomes in spite of serious threats to adaption or development” (Masten, 2001; see also Zimmerman, Bingenheimer, & Behrendt, 2005, for a discussion of resilience theory as it applies specifically to natural mentoring relationships for youth). As the definition suggests, the phenomenon of resilience requires two conditions: the presence of risk and positive adaptation in the presence of risk. One aim of resilience research is to identify factors that protect individuals in face of risk and enable individuals to achieve positive adaptation. Masten and Coatsworth (1998) identified three broad categories of protective factors: individual factors (e.g., intelligence), family factors (e.g., supportive parents), and community factors (e.g., connections to adults in the community). An intervention that uses a resilience framework would attempt to bolster the strength of these protective factors or to create opportunities for these factors to develop if they are absent. Mentors develop a personal relationship with youth to provide “companionship, support, and guidance” (Keller, 2005). Thus, mentors are one example of community protective factors suggested by Masten 315

316  SPECIAL POPULATIONS and Coatsworth. Furthermore, mentors can help to strengthen other protective factors in the lives of youth. For example, mentors can assist youth in negotiating the school system, help resolve conflicts between youth and their parents by communicating youths’ perspectives to their parents, and model positive behaviors (e.g., encourage youth to participate in their communities). Theoretically, mentors who work within multiple ecological contexts (e.g., family, peers, school) are likely to be most successful in promoting resilience because resilience is a process that takes place across multiple levels and involves multiple ecological contexts (Masten, 2007). When mentoring happens beyond the individual level, mentors have the potential to connect youth to different social networks and to gain access to different resources that can promote positive adaptation. These considerations may be especially important for mentoring of juvenile offenders, a population for which interventions geared to multiple ecological contexts (e.g., multisystemic therapy) are among the most effective available (Henggeler & Schoenwald, 2011). Consistent with this perspective, Blechman and Bopp (2005) concluded that for young offenders, mentoring appears to be an intervention that holds greater potential when it involves the families of the youth involved. Within this broad understanding, mentoring theoretically has the potential to be effective in promoting positive adaptation after release for juvenile offenders for several reasons. Mentoring has the potential to provide alternative adult support and guidance, which is often missing or inadequate in the lives of juvenile offenders (Belshaw & Kritsonis, 2006). Mentors often may be able to help juveniles gain access to resources in the community that are necessary for successful transition (e.g., counseling, employment, medical services). This may be particularly likely to occur when adult juvenile system staff members also serve as informal mentors. Because of the nature of their formal employment, these adults are familiar with the juvenile justice system and with the resources available to juvenile offenders. Thus, a comprehensive approach to mentoring that engages multiple social contexts (e.g., family, school) may be critical for successful mentoring of juvenile offenders. In the remainder of this chapter, we explore how mentoring can become a part of this comprehensive approach to addressing the impact of multiple social contexts of youth. One possible strategy is to examine the influence of naturally occurring mentoring because natural mentors are members of youths’ existing social contexts. Therefore, our review of the research emphasizes

understanding the possible effects of naturally occurring mentoring of juvenile offenders.

Research Review of the Literature Similar to the conclusion reached by Blechman and Bopp (2005), based on their review of available empirical research in the previous edition of the Handbook of Youth Mentoring, our analysis suggests that mentoring holds promise for reducing recidivism of juvenile offenders under certain conditions. Additionally, our assessment of available evidence suggests that naturally occurring mentoring deserves greater attention from researchers and practitioners. Furthermore, developing relationships with youth prior to release appears to be a critical component of a successful intervention. More generally, although some studies have demonstrated the potential of mentoring for juvenile offenders, critical gaps in understanding persist, including the long-term effects of such interventions and the mechanisms through which mentoring may effect change for youth who become involved with the juvenile justice system. In conducting our literature review, we used two social science search engines: PsycINFO and ERIC. We limited our search from 2004 to the present so as to focus on research that has appeared since the prior edition of the Handbook. We found three peer-reviewed articles on the topic of mentoring of juvenile offenders using a combination of the following keywords: mentoring, juvenile offenders, recidivism, and juvenile system. We also included three evaluation reports in the present review. Two were conducted by the Washington State Institute for Public Policy on a mentoring program for juvenile offenders in Seattle. The goal of the Seattle program is to “reduce youth violence and delinquency.” The third evaluation report is an annual report of Indiana’s juvenile reentry program, Aftercare for Indiana Through Mentoring (AIM). Also included in this review are two recent meta-analyses (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011; Jolliffe & Farrington, 2007). Peer-Reviewed Studies. Since the review by Blechman and Bopp (2005), three additional peerreviewed studies have examined the impact of mentoring on juvenile offenders, producing conflicting findings. Bouffard and Bergseth (2008) evaluated a reentry program that included a mentoring component

Juvenile Offenders  317 and evaluated the program’s short-term effects on a group of juvenile offenders. The authors provided a description of the program, which is unusual in this literature. Transitional coordinators (TCs), who are full-time staff members, develop informal mentoring relationships with youth. Job responsibilities of TCs include arranging community services, developing individualized transition plans, and providing case management. TCs are not formally trained as mentors, so mentoring occurs naturally as part of their provision of comprehensive services to youth. The Bouffard and Bergseth (2008) study used a quasi-experimental design to compare the shortterm outcomes of 49 youth (control group) matched on demographic characteristics to 63 youth who participated in the reentry program (experimental group). The control group received regular supervised probation. Outcomes included positive drug tests, risk and needs assessment (i.e., risk factors include substance use and deposition to offenses; and needs domains include peer relations and employment needs), total new official contacts, and new official contacts specifically for criminal offenses. These outcomes were assessed 6 months after release. Findings supported the reentry program’s effectiveness in reducing drug use, risk, and recidivism. Youth in the reentry program had a significantly lower rate of positive drug tests and lower risk in comparison to their counterparts in the control group. Also, youth in the reentry program had fewer new official contacts within 6 months after release than participants in the control group. Thus, this study provides evidence suggesting that a reentry program that includes informal mentoring can be effective in reducing short-term drug use and recidivism among juvenile offenders. A second goal of the Bouffard and Bergseth (2008) study was to examine the implementation of the reentry program. It was hypothesized that youth in the program would receive more services than those in the control group because of the assignment of TCs. Implementation was assessed by the number of staff contacts with youth and number of referrals and services received by each youth. Participants in the reentry program had significantly more total contacts than participants in the control group, supporting the hypothesis. Youth in the reentry program were referred to a variety of services, but the program’s effects on referrals could not be assessed because data on referrals were not collected for the control group. In another study of juvenile reentry, Marsh and Evans (2009) explored outcomes associated with the quality of relationships with staff in juvenile correction settings and whether the type of relationship

with staff was associated with perceived success on release. Although no mentoring program was officially implemented, the existence of some form of informal mentoring relationships with staff was assumed. The study included 543 juveniles from Alaska, Idaho, Nevada, and Oregon. The majority of participants were White and male. A self-report instrument measured the perceived quality of the mentoring relationship (Rhodes, Reddy, Roffman, & Grossman, 2005). When completing the questionnaire on perceived quality of relationship, respondents were asked to think of a staff member they respected. Perceived likelihood of success was measured by the Post-Detention Likelihood to Succeed (Evans, Brown, & Killian, 2002). Cluster analysis found three different types of relationship with staff: (a) balanced relationships, characterized by high levels of effective problem solving and closeness to staff, (b) practical relationships, characterized by effective problem solving but low levels of closeness, and (c) engaged relationships, characterized by high levels of closeness but low levels of effective problem solving. All three types of relationship included trust and positive affect. Youth who had balanced relationships with staff reported the highest level of perceived success on release, followed by the practical and engaged groups. Although the extent to which youth received mentoring from staff is unclear, the balanced relationship resembles an effective mentoring relationship in that the staff member engages the youth regularly and provides guidance and support (Todis, Bullis, Waintrup, Schultz, & D’Ambrosio, 2001). Baltodano, Mathur, and Rutherford (2005) reviewed 10 studies focused on promoting successful community transition among juvenile offenders with disabilities, and they concluded that mentoring by adults is a promising intervention. Their review analyzed two studies of the impact of a social skills training intervention on female juvenile offenders, three descriptive studies of individual variables (e.g., attribution style, temperament, goal cognition) as risk or protective factors for recidivism, two descriptive studies of juvenile offenders’ perceptions of their transition process, and three descriptive studies of juvenile offenders’ engagement upon release (e.g., attending school, working, sense of belonging to community). The findings from these 10 studies identified a wide range of factors that may influence recidivism and promote successful transition for juvenile offenders with disabilities. These include perception of internal control, positive peer influence, and quality of programming (Baltodano et al., 2005). Here

318  SPECIAL POPULATIONS we focus on the results related to adult mentoring and support. Two of the 10 studies found an association between having a supportive adult who serves as an informal mentor and positive outcomes among juvenile offenders with disabilities. One study (Yellin, 1996) found that among juvenile offenders who have learning disabilities, those who had an adult advocate were less likely to be rearrested a year after their initial release. A second study (Johnston, 2003) found a similar relationship. Juvenile offenders with disabilities who reported having caring and supportive adults were more likely to be successfully engaged (i.e., working or in school, not detained or committed since release, and not institutionalized for substance or emotional problems since release) than those who didn’t have access to caring adults. Although these two studies were not designed to examine the impact of mentoring, interviews with participants revealed the potential value of supportive adults in promoting successful transition for juvenile offenders with disabilities. Evaluation Reports. Unlike the programs described in the peer-reviewed studies, the programs included in the two evaluation reports discussed here involve a formal mentoring component whose goal is to reduce recidivism and assist with reentry into the community. In 1995, the city of Seattle received funding from the OJJDP to design and implement a comprehensive and community-based program to reduce recidivism. The Washington State Institute for Public Policy was asked to evaluate the impact of the mentoring program. Mentors for this program were volunteers at first recruited from churches. Later, mentors were recruited through distributing posters, engaging local services, and advertising on the internet. The program attempted to recruit mentors from diverse cultural backgrounds. The mentors were expected to make a 1-year commitment, complete a 1-day training, meet with the youth monthly for the 6 months prior to release and weekly after release, and attend monthly meetings on mentoring skills. Most of the mentors were female. Youth committed to a facility served by the Seattle Office of the Juvenile Rehabilitation Administration (JRA) were invited to participate in the program. If a youth was interested, the JRA mentoring program manager would meet with the youth to determine eligibility and to facilitate a match based on ethnicity, gender, personality, and interest. In 2002, the Washington State Institute for Public Policy published preliminary findings on the effectiveness of the mentoring program in reducing recidivism (Barnoski, 2002). Youth who were

included in this analysis were released to the community between February 1997 and September 2000. A control group was selected and was matched to the mentoring group by ethnicity, risk assessment, and gender. The same number of juvenile offenders (N = 78) were in both groups. Within each group, 31 were male, 31 were African American, and 7 were Native American. Recidivism was assessed by administrative records. In comparison to the control group, juvenile offenders who were in the mentoring program had lower rates of recidivism after 12 months of release (54% in the control group, compared to 45% in the mentoring group). However, the difference was not statistically significant, which the authors attributed to small sample size. A follow-up study was conducted, and the findings were presented in a final report (Drake & Barnoski, 2006). The follow-up analysis examined the differences in recidivism after 24 and 36 months of release in addition to the initial 12-month followup. At the 12-month follow-up, the mentoring group had lower rates of overall recidivism, felony recidivism, and violent felony recidivism, but the differences only approached statistical significance. The differences between the two groups diminished over time. The authors suggested that the differences diminished past 12 months because the youth were not receiving mentoring after 1 year of release. The program was implemented statewide in 2002. The other evaluation report we examined is the annual report of the Aftercare for Indiana Through Mentoring (AIM) program (AIM, 2004). The goal of the program is to “support Indiana’s incarcerated youth in making the transition from corrections to community through healthy relationships with adult mentors.” In 1996, AIM initially followed the structure of the Michigan State University program in which undergraduate students provided assistance to first-time offenders. College students were recruited to participate as mentors to help the juvenile offenders make the transition back to the community in Indianapolis. One unique characteristic of AIM is that the program included an advisory board to elicit feedback from both college mentors and juvenile offenders to ensure that the program would be relevant and effective for both the mentors and mentees. Over the years, AIM has evolved and expanded. By 2005, the program had expanded to provide mentoring to juvenile offenders released from nine correctional facilities in Indiana (e.g., Richmond, Bloomington, and Lafayette) and had secured funding from AmeriCorps to hire staff members to direct the capacity-building efforts. The report (AIM, 2004) presented findings from the evaluation conducted in 1997. Participants

Juvenile Offenders  319 included all youth released from the Plainfield Juvenile Correctional Facility and returned to the Indianapolis metropolitan area. Youth were randomly assigned to three groups: (a) in the AIM program with prerelease preparation but not assigned to a mentor, (b) in the AIM program with prerelease preparation with an assigned mentor, and (c) not in the AIM program. The report presented findings on recidivism at 12-month and 48-month follow-ups. At the 12-month follow-up, the AIM with mentor group had the lowest rate of recidivism (25%), followed by the AIM without mentor group (29%) and the not in AIM group (39%). A similar pattern was found at the 48-month follow-up: The AIM with mentor group experienced 44% recidivism, the AIM without mentor group experienced 50% recidivism, and the not in AIM group experienced 62% recidivism. An additional analysis took a closer look at the recidivism rate of a subgroup of the AIM with mentor group. The subgroup consisted of mentors and youth who worked closely together (mentors who fulfilled their commitment and youth who actively participated in the program). The recidivism rates of this subgroup were the lowest at the 12-month and 48-month follow-ups, at 13% and 28%, respectively. Further, the evaluation compared the effect of AIM on recidivism between high-risk and medium/ low-risk youth. Risk assessment, however, was not clearly operationalized in the report. The differences in arrests and felony arrests between the AIM group and the not in AIM group were larger among high-risk individuals; the AIM group had lower rates of arrests. Meta-Analyses. A meta-analysis of youth mentoring program evaluations published between 1999 and 2010 (DuBois et al., 2011) generally found that too few studies were available on which to derive reliable estimates of the relationship between mentoring and outcomes such as reoffending or successful transition to community among juvenile offenders. A second meta-analysis (Jolliffe & Farrington, 2007) assessed the impact of mentoring on reducing recidivism in the United States and in the United Kingdom. The analysis included 16 studies published between 1981 and 2005. Half of the studies investigated the impact of structured mentoring as a component of a comprehensive treatment for juvenile offenders. The rest of them examined a stand-alone mentoring program. The structure and quality of the mentoring programs varied. On average, the findings suggested that mentoring reduced reoffending by 4%–11%. However, results varied according to the studies’ methodologic quality. Whereas evidence of a significant impact of mentoring on reduced reoffending was found for

seven studies that were less methodologically rigorous, a significant impact was not apparent among only those studies that were of the highest methodologic quality. The analysis also found evidence that the frequency and length of meetings between mentor and mentee were important components of effective programs. Youth were less likely to reoffend when they participated in programs in which they spent more time with their mentors (i.e., at least once a week). Another important finding was that mentoring programs that lasted longer were not necessarily more effective. One possible explanation for this finding is that quality control is more difficult in longer-term mentoring programs. Finally, Jolliffe and Farrington (2007) found that the effectiveness of mentoring in reducing recidivism was strongest when it was part of a comprehensive approach to helping juvenile offenders make a successful transition. This finding is consistent with results from other studies reviewed thus far, and with the general notion of a system-oriented approach to mentoring, in which mentoring is part of a comprehensive intervention. Summary and Recommendations for Future Research Literature Review Summary. Together, the three peer-reviewed studies, two meta-analyses, and three evaluation reports suggest that both naturally occurring mentoring and structured mentoring programs can be effective in reducing recidivism and promoting successful transition among juvenile offenders. Naturally occurring mentoring of juvenile offenders appears to have some potential to promote positive outcomes; however, none of the three reviewed studies of naturally occurring mentoring confirmed that mentoring actually took place between the adults and juveniles. Two of the three studies assumed that any close relationships that developed between adult staff members and juvenile offenders would naturally include mentoring. A closer examination of the potential mentor behaviors (e.g., monitoring, guidance, positive regard; see Todis et al., 2001) is needed to determine how naturally occurring mentoring takes place and whether it is helpful to the transition of juvenile offenders. The findings on structured mentoring programs are promising but not strong. The Jolliffe and Farrington (2007) meta-analysis found evidence of a positive effect on recidivism only among a subset of studies that were the least methodologically rigorous and the average effect size was small. In addition, they found no evidence to suggest that the positive effects of mentoring would last beyond the

320  SPECIAL POPULATIONS duration of the mentoring program. Similarly, the two evaluations conducted by the Washington State Institute for Public Policy found that the difference in recidivism rates between the experimental group and the control group diminished at 24- and 36-month follow-up periods. Thus, it appears that, as is true for many interventions, the effects of mentoring for juvenile offenders fade over time. At the very least, the long-term effects of mentoring for this population remain to be determined. The AIM evaluation is the only study that found significant differences between juvenile offenders assigned to a mentoring program and juvenile offenders who did not receive mentoring. The findings suggest that the mentoring component reduced recidivism above and beyond prerelease preparation at both 12 and 48 months. Also important, the AIM study was a methodologically rigorous evaluation that used random assignment and tested whether mentoring added value to traditional prerelease preparation. Guidelines for Future Research. Although some evidence suggests that mentoring is a potentially effective intervention to reduce recidivism and ease juvenile offenders’ transition back to the community, there is a clear need for additional research to inform practice or policy decisions. This research should prioritize further investigations to identify the conditions under which mentoring is effective for juvenile offenders. When compiling the following list of guidelines for future research, we considered the reviewed studies described here as well as the existing literature on youth mentoring. Three of the reviewed studies found evidence for a positive impact of naturally occurring mentoring and structured mentoring on juvenile offenders (AIM, 2004; Baltodano et al., 2005; Bouffard & Bergseth, 2008). However, the processes through which mentoring may have effected changes in drug use and recidivism remain unclear. To further our understanding of how mentoring brings about change, research needs to evaluate various potential mediating mechanisms, such as providing emotional support, modeling positive behaviors, and facilitating access to community resources (Cullen, 1994). Assistance in accessing community resources was a prominent feature of one reentry program reviewed in the previous section (Bouffard & Bergseth, 2008). In that program, mentoring was assumed to occur naturally as the adult case managers developed relationships with their youth clients. Thus, important research questions include the following: (a) What are some possible mediators (i.e., mechanisms of change) that lead to change associated with mentoring? (b) What are the critical components of

a successful mentoring intervention? (c) Do different combinations of components lead to different results? (d) Are there components or activities that should not be part of mentoring? Mentoring for juvenile offenders remains today largely a “black box” whose internal processes are poorly understood, despite general optimism about its positive effects. A question of equal importance is, For whom among the juvenile offender population does mentoring work? In the AIM program, 60% of participants were either attending school or working. Would AIM have had a similar impact on juvenile offenders who had withdrawn from school and were unemployed? Of further note is that studies available for consideration in this and other reviews of mentoring for juvenile offenders have included mostly male and White participants. It remains unclear whether mentoring would have the same effects for female and racial/ ethnic minority youth. Would the effective components be similar, or would different components be required to address the unique cultural and sociopolitical experiences of female and racial/ethnic minority juvenile offenders? Some research on the impact of mentoring of female adolescents, for example, suggests that gender match can be particularly important to fostering positive relationships between mentors and female mentees (Sosik & Godshalk, 2000) and that female adolescents are in greater need of longer mentoring relationships than male adolescents (Rhodes, Lowe, Litchfield, & Walsh-Samp, 2008). Likewise, when mentoring youth from minority backgrounds, it is important to consider the youth’s racial and ethnic identity development and the experiences of oppression (Kaplan, Turner, Piotrkowski, & Silber, 2009; see also Sánchez and colleagues, this volume, Chapter 10). It is possible that these factors would interact with the experiences of institutionalization to influence the outcomes of mentoring of female and minority juvenile offenders. A disproportionally large number of minority youth are arrested for crimes, and juvenile arrest rates for females have increased while rates for males have decreased (Puzzanchera, 2009). Effective mentoring can help to address the overrepresentation of minorities by helping youth to make a successful transition to community life and by reducing the likelihood of future crime. Thus, it is critical for research on mentoring of juvenile offenders to address the effects of mentoring interventions with female and minority juvenile offenders. Three of the reviewed studies suggest that naturally occurring mentoring can be a promising intervention for juvenile offenders. In two of these studies, mentors were adult staff members at correctional

Juvenile Offenders  321 facilities or social service agencies providing transitional services. In addition to examining the impact of mentoring on the juveniles, research needs to address the effects of mentoring on the informal mentors. To what extent do informal mentors receive encouragement and support for mentoring activities? Do staff serving as informal mentors experience additional job stress as a result of their additional responsibilities as mentors? An exploration of the needs, expectations, and challenges of formal and informal mentors would provide valuable information for developing successful mentoring programs. Such exploration would also shed light on the conditions that are most conducive to the development and success of naturally occurring mentoring relationships for juvenile offenders both inside and outside of the juvenile justice care system. Another area that deserves research attention is the selection of mentors, particularly natural mentors. Naturally occurring mentoring happens organically, and it is possible for youth to connect with peers or older adults who model poor behaviors. Research on peer influence suggests that mentoring-like relationships among delinquent youth can result in a process of “deviancy training” in which youth school one another in delinquent behaviors (Dishion, Spracklen, Andrews, & Patterson, 1996). Future research must consider the potential unintended negative consequences of poor natural mentoring and how to discourage the formation of such relationships. Finally, although research suggests that mentoring can effect change across multiple developmental domains (DuBois et al., 2011), only two of the six

reviewed studies (i.e., the three peer-reviewed studies and the three evaluation reports) included positive developmental outcomes in addition to recidivism to evaluate successful transition. The process of transition is multifaceted, and recidivism is the most undesirable of many important potential outcomes of successful transition (Griller-Clark, 2004). Future studies of mentoring should assess other markers of successful transition, such as school engagement, employment, and community participation (Baltodano et al., 2005). Mentoring can have a potentially significant impact on promoting engagement with school and community, increasing access to social services, and seeking employment among juvenile offenders. In one of the reviewed studies, informal mentors who were also case managers were in charge of helping the participants to access social services and assisting the participants in their educational and vocational pursuits (Bouffard & Bergseth, 2008). It would be fruitful to investigate whether mentoring can have an impact on these positive developmental indicators of successful transition.

Practice In this section, we present some suggestions to practitioners who deliver mentoring services to juvenile offenders (see Table 21.1). We have three recommendations: (a) develop a partnership with researchers, (b) apply a comprehensive approach to mentoring, and (c) implement naturally occurring mentoring.

Table 21.1   Checklist for Practitioners Topic

Recommendations

Partnerships with researchers

Develop partnerships with researchers to facilitate evidence-based approaches to mentoring program development, program evaluation, and ongoing program improvement for services focused on mentoring for juvenile offenders.

A comprehensive approach to mentoring

Develop partnerships with other institutions that serve juvenile offenders (e.g., school, mental health services). Facilitate opportunities for mentors to meet with family members, teachers, and case managers to coordinate services for juvenile offenders. Encourage and train mentors to serve as mediators between school and family, and between school and the juvenile justice system.

Naturally occurring mentoring

Recruit adults who are already working with juvenile offenders to serve as informal mentors (e.g., teachers, case managers). Support adults who are in positions to provide informal mentoring to juvenile offenders.

322  SPECIAL POPULATIONS First, the success of mentoring programs depends on the translation of research into practice (DuBois et al., 2011). It is critical that researchers make research findings easily accessible for practitioners who implement mentoring programs in the community. One efficient way is to publish detailed manuals of structured mentoring programs online for practitioner use (Rhodes & Lowe, 2008). Manuals of research-based programs could be modified to describe how to implement and adapt the program to fit the needs of the local agency and its clients. Another way for researchers to partner with practitioners is through conducting evaluation. Researchers can assist agencies to develop capacity to conduct evaluation and an organizational culture that encourages it. Evaluation, both of process and outcome, is critical to the continued development of mentoring interventions for juvenile offenders. Evaluation results not only benefit agencies that are providing services; they also help researchers test and refine theories regarding mentoring of juvenile offenders. With these considerations and possibilities in mind, we view a collaborative relationship between researchers and practitioners as essential to developing, delivering, and improving effective mentoring practices for juvenile offenders. Second, we propose a comprehensive systemoriented approach to mentoring that involves schools and other institutions that serve juvenile offenders. This approach is similar to the caregiver team approach suggested by Blechman and Bopp (2005). Such an approach can include meetings of parents, teachers, case managers, and mentors to assess needs and coordinate services. Such an approach can be particularly effective for involving informal mentors in the transition process. Interventions that address risk factors across multiple ecological contexts (e.g., family, school, neighborhood), such as multisystemic therapy (Henggeler, Schoenwald, Borduin, Rowland, & Cunningham, 2009), are likely to be particularly successful in promoting positive behaviors and preventing future delinquency (Henggeler & Schoenwald, 2011). Belshaw and Kritsonis (2006) offer some strategies on how to better coordinate services across multiple ecological systems to facilitate partnership among mentors, teachers, and administrators. Mentors can serve as mediators between family and school and between school and the juvenile justice system. Similarly, teachers and school administrators can work with mentors to learn about the youth’s needs and problems outside of school that affect educational performance. A coordinated effort that engages multiple ecological contexts of the youth may enhance the effectiveness of mentoring for juvenile offenders.

Third, the studies reviewed here suggest that naturally occurring mentoring can be somewhat effective in reducing recidivism and assisting in the transition back to the community for juvenile offenders. If so, how can policymakers and program planners encourage the development of natural mentoring relationships in contexts already involving youth offenders? Policy shifts that recognize and encourage informal mentoring can be cost-effective ways to improve outcomes for youth offenders.

Conclusion The literature reviewed here provides a glimpse of the promise of both structured and informal mentoring for facilitating the transition back to the community and generally improving the life outcomes of youth involved in the juvenile justice system. It also points out significant gaps in our knowledge. Although the evidence is generally promising, effects of mentoring for juvenile offenders suggested thus far are not strong, leaving room for doubt about the effectiveness of mentoring for this population. Despite this uncertainty, the literature points us in some promising directions. Naturally occurring mentoring emerged from the review as a variation of mentoring that can be potentially beneficial to juvenile offenders as they make their transition back to the community. Another finding from this review is that a system approach to mentoring may be particularly effective as a strategy that offers the potential to effect change across multiple ecological contexts of the youth. When mentors engage the larger social networks of youth in addition to providing support and guidance at the individual level, they have a potentially greater impact. Because successful transitions involve multiple ecological contexts, mentoring that has the ability to create connections across contexts may be especially promising. To further our understanding of how mentoring can promote positive transition among juvenile offenders, we see much potential value in research and practice taking a systems perspective to investigate and capitalize on the mechanisms through which mentoring may effect change. Whenever an intervention attempts to create changes within a system, either at one level or across multiple levels, unintended consequences are possible (Trickett, Kelly, & Vincent, 1985). Naturally occurring mentoring is itself an unintended, though potentially desirable consequence of service provision for youth offenders and it

Juvenile Offenders  323 deserves examination. We look forward to a next generation of studies that can shed greater light on these promising avenues for improving the mentoring that is made available to this population.

References Aftercare for Indiana Through Mentoring (AIM). (2004). Annual report: Indiana’s juvenile reentry program. Retrieved from http://aim.spea.iupui.edu/AIM_ annual_report.doc Baltodano, H. M., Mathur, S. R., & Rutherford, R. B. (2005). Transition of incarcerated youth with disabilities across systems and into adulthood. Exceptionality, 13, 103–124. Barnoski, R. (2002). Preliminary findings for the Juvenile Rehabilitation Administration’s mentoring program. Seattle: Washington State Institute for Public Policy. Belshaw, S. H., & Kritsonis, W. A. (2006). National implications in education and juvenile justice: Bridging the gap between court order juvenile mentoring programs and secondary educators. Doctoral Forum: National Journal for Publishing and Mentoring Doctoral Student Research, 3(1). Blechman, E., & Bopp, J. (2005). Juvenile offenders. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 454–466). Thousand Oaks, CA: Sage. Bouffard, J. A., & Bergseth, K. J. (2008). The impact of reentry services on juvenile offenders’ recidivism. Youth Violence and Juvenile Justice, 6, 295–318. Cullen, F. (1994). Social support as an organizing concept for criminology: Presidential address to the Academy of Criminal Justice Sciences. Justice Quarterly, 11, 527–559. Dishion, T. J., Spracklen, K. M., Andrews, D. W., & Patterson, G. R. (1996). Deviancy training in male adolescent friendships. Behavior Therapy, 27, 373–390. Drake, E. K., & Barnoski, R. (2006). Recidivism findings for the Juvenile Rehabilitation Administration’s mentoring program: Final report. Seattle: Washington State Institute for Public Policy. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Evans, W. P., Brown, R., & Killian, E. (2002). Decision making and perceived postdetention success among incarcerated youth. Crime & Delinquency, 48, 553–567. Griller-Clark, H. (2004). Transition services for youth with disabilities in the juvenile justice system. (EDJJ Training Module #8). Retrieved from http://www.edjj .org/focus/TransitionAfterCare/docs/TrainModule .pdf

Henggeler, S. W., & Schoenwald, S. K. (2011). Evidencebased interventions for juvenile offenders and juvenile justice policies that support them. Social Policy Report, 25(1). Retrieved from http://www.srcd.org/ Henggeler, S. W., Schoenwald, S. K., Borduin, C. M., Rowland, M. D., & Cunningham, P. B. (2009). Multisystemic therapy for antisocial behavior in children and adolescents (2nd ed.). New York: Guilford Press. Johnston, J. (2003). In the arena: Youth in transition from juvenile corrections to the community. Unpublished doctoral dissertation, Arizona State University, Tempe. Jolliffe, D., & Farrington, D. P. (2007). A rapid evidence assessment of the impact of mentoring on re-offending: A summary. (Home Office Online Report). Retrieved from http://www.youthmentoring.org.nz/research .cfm#faq132035 Kaplan, C. P., Turner, S. G., Piotrkowski, C., & Silber, E. (2009). Club Amigas: A promising response to the needs of adolescent Latinas. Child & Family Social Work, 14, 213–221. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Marsh, S. C., & Evans, W. P. (2009). Youth perspectives on their relationships with staff in juvenile correction settings and perceived likelihood of success on release. Youth Violence and Juvenile Justice, 7, 46–67. Masten, A. S. (2001). Ordinary magic: Resilience processes in development. American Psychologist, 56, 227–238. Masten, A. S. (2007). Resilience in developing systems: Progress and promise as the fourth wave rises. Development and Psychopathology, 19, 921–930. Masten, A. S., & Coatsworth, J. D. (1998). The development of competence in favorable and unfavorable environments: Lessons from research on successful children. American Psychologist, 53, 205–220. Office of Juvenile Justice and Delinquency Prevention. (2010a). Mission statement. Retrieved from http:// www.ojjdp.gov/about/missionstatement.html Office of Juvenile Justice and Delinquency Prevention. (2010b). OJJDP model program guide: Mentoring. Retrieved from http://www.ojjdp.gov/mpg/prog TypesMentoring.aspx Puzzanchera, C. (2009). Juvenile arrests 2008. Retrieved from http://ojjdp.ncjrs.gov/ojstatbb/publications/ StatBBAbstract.asp?BibID=250498 Rhodes, J. E., & Lowe, S. R. (2008). Youth mentoring and resilience: Implications for practice. Child Care in Practice, 14, 9–17. Rhodes, J. E., Lowe, S. R., Litchfield, L., & Walsh-Samp, K. (2008). The role of gender in youth mentoring relationship formation and duration. Journal of Vocational Behavior, 72, 183–192. Rhodes, J. E., Reddy, R., Roffman, J., & Grossman, J. B. (2005). Promoting successful youth mentoring relationships: A preliminary screening questionnaire. Journal of Primary Prevention, 26, 147–167.

324  SPECIAL POPULATIONS Sosik, J. J., & Godshalk, V. M. (2000). The role of gender in mentoring: Implications for diversified and homogenous mentoring relationships. Journal of Vocational Behavior 57, 102–122. Todis, B., Bullis, M., Waintrup, M., Schultz, R., & D’Ambrosio, R. (2001). Overcoming the odds: Qualitative examination of resilience among formerly incarcerated adolescents. Exceptional Children, 68, 119–139. Trickett, E. J., Kelly, J. G., & Vincent, T. A. (1985). The spirit of ecological inquiry in community research.

In E. C. Susskind & D. C. Klein (Eds.), Community research: Methods, paradigms, and applications (pp. 283–333). New York: Praeger. Yellin, E. M. (1996). Resilience: A model of success for adjudicated youth. Unpublished doctoral dissertation, Arizona State University, Tempe. Zimmerman, M. A., Bingenheimer, J. B., & Behrendt, D. E. (2005). Natural mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 143–157). Thousand Oaks, CA: Sage.

22  YOUTH WITH MENTAL HEALTH NEEDS David C. R. Kerr and Cheryl A. King1

Introduction Significant numbers of youth have mental health needs, defined here as clinically significant disruptions in behaviors, thoughts, or emotions that cause distress and impaired social or academic functioning (American Psychiatric Association, 2000). Based on a nationally representative sample, the 3-month prevalence of a psychiatric diagnosis was 13.3% among American children aged 9–16, and lifetime prevalence (i.e., having ever had a disorder) by age 16 years was conservatively estimated at 36.7% (Costello, Mustillo, Erkanli, Keeler, & Angold, 2003). Given these rates, it is clear that youth with mental health needs often participate in community-oriented mentorship programs. Yet few studies exist on programs specifically designed to address their needs. Existing Treatments for Youth With Mental Health Needs Effective Mental Health Treatments. The promise of mentorship for helping youth with mental health needs must be understood in the context of the evidence base for existing treatments for such youth. Research now supports the efficacy of multiple psychosocial and medication treatments for children when delivered via structured protocols in carefully controlled settings by well-trained clinicians (Silverman & Hinshaw, 2008). Effect sizes (standardized estimates of the strength of treatment effects relative to a comparison condition) for these treatments, however, depend on the problem focus. Specifically, cognitive-behavioral 1

therapy (CBT) and interpersonal psychotherapy are well-established treatments for major depressive disorder (MDD), with effect sizes in the small to medium range (Brown et al., 2008; David-Ferdon & Kaslow, 2008). The combination of antidepressant medication (fluoxetine) and CBT has demonstrated a medium to large effect on primary MDD symptoms (Treatment for Adolescents With Depression Study Team, 2004). CBT has been deemed probably efficacious for youth with obsessive-compulsive disorder with medium to strong effect sizes (Barrett, Farrell, Pina, Peris, & Piacentini, 2008), and for generalized anxiety and phobias with medium effect sizes (Silverman, Pina, & Viswesvaran, 2008). Parental behavioral management training is a well-established and highly efficacious treatment for attention-deficit/hyperactivity disorder (ADHD; Pelham & Fabiano, 2008), oppositional defiant disorder (ODD), and conduct disorder (Eyberg, Nelson, & Boggs, 2008), with medium to large effect sizes. Pharmacologic treatments alone have small to medium effects on symptoms of anxiety disorder and MDD, and moderate effect sizes for ADHD (Brown et al., 2008; Hidalgo, Tupler, & Davidson, 2007). Taken together, this range of effect sizes indicates that many but not all youth benefit from available evidence-based treatments for mental health difficulties and that, among those who do benefit, many do not experience a full resolution of their problems. Thus, even for youth who are receiving a high standard of clinical care, clearly there is room for other forms of intervention, such as mentoring, to yield additional benefit. Unfortunately, many youth do not receive this standard of care, and there

The authors thank Adam Horwitz, Katherine Lewis, and Miranda White for their assistance.

325

326  SPECIAL POPULATIONS are wide gaps between findings from laboratorybased efficacy trials and those based on usual community-based services. Specifically, a metaanalysis of 32 randomized clinical trials found that evidence-based treatments were associated with more favorable outcomes than were usual care (effect sizes were small to medium; Weisz, JensenDoss, & Hawley, 2006), and in a meta-analysis of 14 trials, psychotherapy as usually practiced in communities was found to be no more effective on average than no intervention (i.e., average effect size was essentially zero; Weisz & Jensen, 2001). These striking limitations suggest windows of opportunity for mentoring to benefit youth directly and perhaps also indirectly by facilitating interventions so that they can exert maximum benefit. Reaching and Maintaining Youth in Treatment. Even though effective treatments are available, several obstacles exist to engaging and maintaining youth mental health treatment. For instance, many youth with mental health problems simply do not present for treatment. A recent retrospective study carefully examined 5 years of systematic mental health screenings in six high schools (Husky, Sheridan, McGuire, & Olfson, 2011). The majority of the 19.6% of students identified as at risk were not receiving treatment. Moreover, only 41.9% of the students referred for community services actually accessed treatment. Youth also often show poor treatment adherence because they depend on adults for transportation and to support, participate in, or administer the intervention. Indeed, an estimated 50%–75% of children referred for treatment do not initiate or complete recommended treatment (Kazdin, Siegel, & Bass, 1990). Mentoring: A Promising Approach Given the modest state of the science with respect to reducing youth mental health problems, there is a need for and openness to innovative and adjunctive intervention approaches. Mentoring is one such approach. However, despite the evidence base in nonclinical populations, systematic research on mentoring for youth with identified mental health needs is lacking. Though mentorship cannot replace traditional treatment, there are theoretical reasons to expect it may reinforce and enhance mechanisms of change targeted by mental health treatments. Furthermore, a promising feature of mentoring is that it may influence youth through different pathways. Specifically, treatments often focus on internal changes, such as in how youth perceive and react to events. Yet such an approach

in isolation may overestimate youth agency and capacity to manage the often chronic and severe contextual stresses that influence their problems. In contrast, mentoring by its very nature introduces change in a youth’s external circumstances. Chapter Overview In this chapter, we first consider theoretical models of mentoring with special attention to the implications for youth with mental health needs. We then review research on mentoring of youth mental health problems, and we suggest areas in need of further study. Finally, we offer practice recommendations. We use the general term externalizing problems to refer to behaviors of undercontrol that are most apparently problematic to others (e.g., aggression, defiance) and the term internalizing problems to refer to significant internal distress (e.g., anxiety, depression). We use DuBois and Karcher’s (2005) definition of a mentoring relationship as one in which a mentor with greater experience or wisdom provides guidance to support a mentee’s development, within the context of an emotional bond of mutual trust and caring.

Theory General Mentoring Framework Most youth with mental health problems grapple with the same developmental tasks as other youth. Thus, Rhodes’s (2005) widely known developmental model of youth mentoring also can be expected to apply to youth with mental health needs. Rhodes posits that mentoring relationships characterized by mutuality, trust, and empathy indirectly lead to positive youth outcomes by supporting social-emotional, cognitive, and identity development. In particular, youths’ experiences with supportive mentors may alter their working models of themselves and of their relations with adults such that alternative ways of being can be imagined and pursued. Of great relevance to this chapter, the model assumes that individual and contextual factors affect the formation of such a relationship and moderate the associations between mentorship and developmental course, which would include mental health outcomes. Compared to others, youth with mental health problems are more likely to have temperamental characteristics and limitations in social competence that may interfere with the formation of a strong bond with a mentor. These youths’ relationship histories also more often may

Youth With Mental Health Needs   327 have involved inconsistency, disappointment, and power struggles, leading them to expect or inadvertently re-create these patterns. Finally, transitions in caregiving and residence are associated with youth mental health problems (e.g., Wickrama, Conger, Lorenz, & Jung, 2008), and such experiences may interfere with mentoring relationships. The social-emotional development facilitated by mentoring may improve broader social functioning by improving youth relationships with parents and peers (Rhodes, 2005). Given that family conflict and negative adult and peer models can affect the onset, exacerbation, or maintenance of mental health disorders, mentors’ influences on social functioning and the provision of alternative working models of self and other may be especially salient for youth with mental health needs. It also is useful to consider the potential promise of mentoring using the theoretical models and key themes that guide interventions for youth mental health problems. These models suggest several ways in which mentors could help youth with mental health needs. Mentoring and Theory in Mental Health Treatment General Issues: Treatment Initiation, Adherence, and Destigmatization. Mentors may make a significant impact on youth with mental health needs by identifying these needs and facilitating entry into treatment. In many cases, mentors, by virtue of their professional roles, social networks, and histories in their communities, can be valuable sources of local information. Mentors also may be particularly effective at advocating for youths’ needs, either directly with parents or indirectly by communicating with mentoring program staff who can advise parents. Given the importance of referring parents and youth to effective treatment (as opposed to just any treatment; see review above), mentoring programs could have considerable impact on youth mental health outcomes by directing families to providers who offer high standards of care. Mentors also may enhance youth outcomes by supporting engagement and ongoing adherence with recommended treatment. They may do so by explicitly encouraging it, instilling hope, and offering pragmatic support (e.g., transportation). The stigma associated with mental health problems and treatment can undermine treatment adherence (Sirey et al., 2001). Though research is lacking, mentors may fight stigma, for example, by talking comfortably and nonjudgmentally about mental health problems and treatment and reframing treatment-seeking as a sign of strength and maturity.

Thus, we propose that Rhodes’s (2005) model may be elaborated for youth with mental health needs: Effects of mentoring on positive outcomes could be mediated partially by facilitative effects on treatment linkage and adherence. Internalizing Problems: Stress, Support, and Coping. Diathesis-stress models often are used to explain the onset of youth internalizing disorders as well as recurrences of episodic disorders, such as MDD. In these models, underlying biological, interpersonal, or cognitive vulnerabilities (e.g., the tendency to make negative internal attributions regarding the causes of stressful events) are particularly likely to have negative effects when stressors reach sufficient magnitude or when important behavioral routines and patterns of reinforcement become disrupted. Because social support is believed to buffer youth against the negative effects of stress, treatment goals often include developing and utilizing support networks. General CBT strategies for internalizing disorders include cognitive restructuring as well as fostering problem-solving and emotion regulation skills (Clarke, DeBar, & Lewinsohn, 2009). Though less systematic and intensive than CBT, mentoring has been conceptualized similarly to involve scaffolding of social-emotional and cognitive development (Rhodes, 2005). Mentors provide social support and may naturally model and reinforce realistic appraisals, adaptive coping, and problem solving, or they could be trained to do so effectively. Whereas youth with positive characteristics attract and retain supportive adult mentors (Rhodes, 2005), those with internalizing problems may have difficulty doing so. For example, Murray and Greenberg (2001) found that youth receiving school services for emotional disturbance had poorer affiliation with teachers than did nondisabled youth or youth receiving other services (e.g., learning disability), as evidenced by self-reports on items such as “there is a teacher . . . I can count on when I have a problem.” Depressive symptoms have been found to weaken youths’ supportive relationships (Stice, Ragan, & Randall, 2004) and to elicit interpersonal stressors that perpetuate depression (Kercher & Rapee, 2009). To the extent that internalizing problems such as irritability, social withdrawal, and low energy are manifested interpersonally, it may be challenging for mentors to become engaged and remain motivated with these youth. Several other therapeutic interventions used to treat specific internalizing disorders may occur naturally during mentoring or, with training and encouragement, may be comfortable extensions of the relationship. First, given that disruptions in

328  SPECIAL POPULATIONS important routines and behaviors influence the onset and maintenance of depression, an important component of CBT, known as behavioral activation, is aimed at increasing youth engagement in pleasurable activities and counteracting tendencies to withdraw from reinforcement opportunities (e.g., Clarke et al., 2009). Shared activity with a mentor can create new routines and engagement in pleasurable activities for youth who would not do so otherwise. Second, a key efficacious ingredient of behavioral treatments for anxiety disorders is gradual exposure to feared situations (which often are social in nature), the avoidance of which drives impairment in daily living (Silverman, Pina et al., 2008). With some briefing on treatment rationale, mentors seemingly could support youth-driven efforts to confront and cope with such situations. Finally, interpersonal psychotherapy focuses on shoring up social support, resolving acute interpersonal crises, and enhancing communication skills and interpersonal effectiveness to treat depressive and eating disorders (e.g., Murphy, Cooper, Hollon, & Fairburn, 2009). Mentoring naturally serves the social support function that helps buffer the impact of crises, and mentors could be trained and encouraged to provide constructive feedback regarding youths’ positive and maladaptive communication patterns. To summarize, mentoring relationships in their natural form can provide youth with opportunities for recreation, social stimulation, and interpersonal feedback, all of which may complement or overlap with clinical approaches used to help youth with internalizing problems. Specialized training of mentors in these areas may be of further benefit. Thus, mentoring is a theoretically promising approach for influencing the course and formal treatment of internalizing disorders. Externalizing Problems: Modeling and Monitoring. Externalizing problems often have origins in the intersection between challenging temperamental characteristics and ineffective parenting. These origins are relevant to mentoring because by the time youth with externalizing problems reach later childhood and adolescence, they may have long histories of coercive or disengaged relationships with adults. Youth affiliation with deviant peers, neighborhood factors such as the absence of prosocial adult models and the presence of antisocial ones, and poor parental monitoring all appear to influence problem behavior trajectories (Woo & Keatinge, 2008). Thus, interventions for youth with externalizing problems emphasize the constructive engagement of adults. Specifically, treatments teach parents to

reinforce appropriate behavior, provide consistent consequences for misbehavior, monitor youth whereabouts, limit unsupervised time and access to deviant peers, and coordinate efforts with school staff. Mentors are a promising resource for furthering these aims for several reasons. They can both reinforce the positive behaviors that youth demonstrate (e.g., school attendance) and discourage the antisocial behaviors that they disclose or exhibit. Similarly, during shared activity with youth, mentors can reinforce adaptive communication skills (e.g., respectfully disagreeing) and prosocial behaviors (e.g., sharing). Mentors can support socialemotional and cognitive development by modeling problem solving, conflict management, and communication. Also important, they may stimulate identity development for youth with externalizing problems by serving as role models for developmental achievements related to career, family, and community engagement; by exposing youth to alternative realities and imagined futures; and by sharing knowledge regarding community resources and career opportunities (Rhodes, 2005). Finally, mentors can contribute to the protective effects of adult monitoring on youth behavior. Unlike parents, mentors do not use knowledge of a youth’s behaviors and whereabouts to set limits in their environment (e.g., stopping contact with a problematic friend). Yet mentors may foster the kind of relational climate that promotes youth disclosure and perception of adult oversight that is protective. For example, in a study that asked youth about the nonparental adults who they viewed as most important in their lives, peer risk factors were found to be less associated with youth problem behaviors when youth believed these supportive adults would disapprove of problem behavior (Beam, Gil-Rivas, Greenberger, & Chen, 2002). This finding highlights the influence a mentor may have as a supportive nonparental authority figure. In sum, there are good reasons to expect that mentors have the potential to supplement and support treatment efforts for youth with externalizing problems and to assist with important socialization goals. Suicidal Thoughts and Behaviors: Belonging, Mattering, and Monitoring. The vast majority of youth who die by suicide had at least one psychiatric disorder (Shaffer, Gould, Fisher, & Trautman, 1996). Thus, effective mental health treatment is an important focus of suicide prevention (U.S. Public Health Service, 1999), and mentors’ efforts that complement or overlap with the treatment models described above are relevant here. Additionally,

Youth With Mental Health Needs   329 according to the interpersonal-psychological theory of suicide (Joiner, 2005), three conditions are necessary and sufficient for an individual to die by suicide: perceptions of a lack of belonging, the sense that one is a burden to others, and the acquired capacity to enact lethal self-harm. The theory’s corollary— that disrupting any of these conditions will prevent suicides—has become influential in the field of suicide prevention and has clear application to the role that mentoring may play in reducing youth suicide risk. Specifically, mentors offer shared activity and emotional closeness (i.e., belonging) and may instill in youth a sense of mattering (vs. burdensomeness). A mentoring relationship may be a distal protective factor that gradually impacts youth with respect to suicide risk as well as a proximal one that offers a safety line of support and informal monitoring. In this regard, there could be benefits associated with training mentors to recognize signs of potentially elevated risk for suicidal behavior and to respond appropriately as a concerned and caring adult. Critically, however, the importance of a mentoring relationship in a vulnerable youth’s life cuts both ways: Mentors also can communicate (e.g., through inconsistent contact or abrupt termination of the relationship) that his or her connection with a youth is not valued or is burdensome.

Research Our review of the research first summarizes key studies of natural mentoring and mentoring programs with nonclinical samples that considered mental-health-related problems (e.g., depressive symptoms) as outcomes or as moderators of mentoring effects. Second, we consider studies of mentoring for youth with identified mental health problems. Third, we review findings from multicomponent interventions for youth with mental health needs in which one or more components involve mentoring. Finally, we offer recommendations for future research given the state of the evidence. Our literature search strategy included Google Scholar and PsycINFO searches using the term mentor in conjunction with (a) general terms and stems such as mental health, psychiatr-, and disorder, (b) specific diagnoses and symptoms such as depress-, or (c) specific interventions such as wraparound. Studies also were identified from the reference sections of relevant articles, and using Google Scholar to identify articles that had cited these works. Finally, the editors recommended several articles for inclusion in this review. Given space limitations,

we focus primarily on research that has met relatively high standards of rigor. Studies of Mentoring in Nonclinical Samples Having a natural mentor has been linked to a decreased likelihood of problem behavior outcomes (e.g., substance use, gang membership, serious fighting, and risk taking) but has been weakly related or unrelated to depressive symptoms, anxiety, or suicidal ideation (DuBois & Silverthorn, 2005a; Eby, Allen, Evans, Ng, & DuBois, 2008). Conversely, among those who had a natural mentor, youth closeness with their mentor was not significantly linked with problem behavior outcomes, other than less drug use, but was negatively associated with depressive symptoms and suicidal ideation (DuBois & Silverthorn, 2005b). Summarizing these patterns, DuBois and Silverthorn (2005b) suggested that, without a close relationship, mentoring may not be sufficient to influence psychological symptoms and well-being, whereas the mere presence of a concerned adult mentor may protect against problem behaviors. Even when natural mentoring has been linked to positive outcomes, this has only partially mitigated the observed associations of individual and environmental risks with mental health outcomes (DuBois & Silverthorn, 2005a). Several large studies of youth participating in formal mentoring programs also have considered mental health outcomes and whether effects are moderated by mental-health-related youth characteristics. Using data from the landmark randomized controlled trial (RCT) of Big Brothers Big Sisters (BBBS), Grossman and Rhodes (2002) found that youth who had mentoring relationships of longer duration (more than 12 months) showed positive outcomes relative to those in the control group, including decreases in substance use. However, evidence was not found for benefits of shorter relationships, and in some cases, these relationships appeared to be detrimental. Specifically, relative to the control group, alcohol use increased among youth in relationships lasting less than 6 months, and self-worth and perceived scholastic competence declined for those in relationships lasting less than 3 months. Critically, youth who had been abused or who were referred for psychological problems had shorter mentoring relationships. These findings suggest that mentoring can be beneficial, inert, or harmful and that youth with mental health needs appear to be at increased risk of being worse off for having been mentored. DuBois and colleagues conducted two related meta-analyses of studies of mentoring programs and

330  SPECIAL POPULATIONS paid special attention to effects for youth deemed at risk due to individual characteristics, environment, or both (DuBois, Holloway, Valentine, & Cooper, 2002; DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). First, DuBois et al. (2002) analyzed data from evaluations of youth mentoring reported from 1970–1998. To be included, studies had to have compared youth participants’ outcomes before and after mentoring, and/or outcomes for youth who did and did not receive mentoring. Effect sizes indicated benefits of mentoring, with particularly strong effect sizes for samples of youth who were at both individual and environmental risk. However, those at individual risk only did not benefit from mentoring, on average. The authors speculated that, in the absence of environmental risk, mentors may have attributed youth problems to internal causes, a potential barrier to empathy and effective role fulfillment. It also is possible that mentors were able to mitigate the effects of environmental risk on individual risk (e.g., a child whose conduct problems appear related to family chaos), but they were not able to mitigate individual risk (e.g., a strong biological vulnerability to mental disorder) that had no clear contextual causes. DuBois et al. (2002) did not examine different domains of individual risk, thus limiting conclusions they could make regarding mentoring of youth with mental health as opposed to other problems (e.g., school failure). To better understand the variability in the overall null effect of mentoring programs for youth at individual risk, DuBois and colleagues (2002) further analyzed the effects of program characteristics. Results indicated that mentoring programs following “best practices” (e.g., expectations about frequency of contact) had more positive effects on youth outcomes; this finding held as well when considering specifically studies in which samples consisted of youth experiencing only individual risk factors. In contrast, programs that did not follow best practices were not simply ineffective, but rather, on average, had effects in the direction of causing harm. The DuBois et al. (2011) meta-analysis examined studies published between 1999 and 2010 that compared mentored with nonmentored youth. Outcomes tended to improve on average for mentored youth and remain the same or decline for nonmentored youth. These positive effects were observed in domains relevant to mental health, including psychological/emotional and conduct problems, although program effects on substance use specifically (assessed in relatively few studies) were weaker and nonsignificant. Additional findings of this review suggested more beneficial effects for some youth with high levels of individual risk, including those

serving youth with behavior problems specifically. In this respect, DuBois and colleagues (2011) speculated that more recent mentoring programs may have evolved to be better equipped to address the needs of youth with higher individual risk characteristics. Because this group likely included many youth with mental health needs, this is a promising possibility though one that requires further evaluation. In sum, studies support the promise of natural and programmatic mentoring for youth with mental health needs, perhaps especially for youth with needs based on problem behavior and other individual risk characteristics. Although some research suggests that mentoring has the potential to be harmful for youth with mental health needs, this is not a general finding. It seems likely that harm can be mitigated (a) by following best practices (DuBois et al., 2002; Grossman & Rhodes, 2002), and (b) by providing mentors with basic training about anticipated relationship challenges and ways to respond to concerns about mental health problems (see “Practice” section). Studies of Mentoring for Youth With Mental Health Needs The Youth-Nominated Support Team (YST), Version I (C. A. King et al., 2006), is a psychoeducational social network intervention for adolescents who are psychiatrically hospitalized with serious suicidal ideation or a recent suicide attempt. It was designed to assist these adolescents in the transition from psychiatric hospitalization to home and community setting. Youth nominate up to five caring persons (up to one peer) from family, school, and community domains who are then asked to act as support persons by participating in a psychoeducational session and maintaining weekly supportive contact with the youth for 6 months. In turn, YST staff members have regular contact with the support persons to address questions and concerns. YST is conceptualized as providing psychoeducation to support persons and providing positive emotional support to adolescents, enhancing their sense of connectedness. Also, support persons are informed of youth diagnoses and treatment plans, and are asked to inquire about and encourage adherence with recommended treatment. Thus, YST is consistent with recommendations that mentoring be used to enhance the implementation of interventions that target established risk and protective factors (see Cavell & Elledge, this volume, Chapter 3). In a randomized trial, 289 adolescents were assigned to YST (plus treatment-as-usual) or treatment-as-usual groups. At the end of the 6-month intervention,

Youth With Mental Health Needs   331 there were no significant group differences in suicide attempt rates, but there was a significantly greater reduction in suicidal thoughts for girls in YST compared to the treatment-as-usual group. A second randomized trial was conducted for YST, Version II (YST-II; C. A. King et al., 2009). Given YST findings that peers were less likely to remain involved and delayed the intervention due to an extended informed consent process, and concerns about burdening youth with a perceived responsibility for potentially suicidal peers, YST-II enlisted only adult support persons. Additionally, YST-II was implemented for only 3 months, given that in YST, (a) most support persons were not maintained for 6 months, (b) length of support person involvement (3 vs. 6 months) was not associated with outcome, and (c) many support persons reported that 6 months of ongoing check-ins with staff became burdensome. The YST-II study was conducted with 448 suicidal and psychiatrically hospitalized adolescents. Youth in the YST-II condition showed a significantly greater decrease in suicidal ideation after 6 weeks, particularly if they had a history of multiple suicide attempts (small to moderate effect size), but effects did not persist at a 1-year follow-up. Among youth with histories of no suicide attempt or a single attempt, YST-II was associated with greater decreases in functional impairment at 3- and 12-month followups (small effect sizes). Taken together, YST and YST-II had no identifiable negative effects (and no suicides in intervention groups), but had only small to moderate, time-limited positive effects. Given the vulnerable population targeted and the effects observed during the high-risk posthospitalization period, even modest effects are important. Still, findings indicate that more intensive, longer-term, and multifaceted interventions are needed to alter negative trajectories of suicidal adolescents. The trials had several strengths that are instructive to future studies of mentoring for youth with mental health needs, including manualized training and supervision of support persons and interventionists, and the effort to demonstrate effectiveness and generalizability by limiting adolescent exclusion criteria. Other Reviewed Programs. We identified two other studies of mentoring for youth with current mental health needs. Jent and Niec (2006) studied a supplemental mentoring program in which youth (aged 8–12 years) with a psychiatric disorder received mentoring from an undergraduate for 8 weeks or more. Second, K. A. King, Vidourek, Davis, and McClellan (2002) followed a sample of 4th graders who met one or more high-risk criteria (e.g., thresholds of sadness,

substance use, or low self-esteem) and who participated for 5 months in the Healthy Kids Mentoring Program. Both programs require rigorous evaluation. Studies of Multicomponent Interventions That Include Mentoring Several interventions have integrated mentoring as one of several components of an intervention for youth with mental health problems. A limitation of all of the studies reviewed for the present purposes is that the unique effects of the mentoring component have not been established. We review programs implemented in school contexts to help at-risk or high-risk youth, as well as other programs directed toward very seriously disturbed youth. Cognitive-Behavioral Group Mentoring. Jent and Niec (2009) aimed to integrate features of mentorship and group CBT. In an RCT, 86 youth aged 8–12 years old who were referred to community mental health for mentoring services were randomized to the CBT-mentoring group or to a 3-month waitlist. Two paid mentors met with three to seven youth in groups for 12 weekly, 4-hour sessions of discussion, psychoeducation, and activities. Many activities took place outside of the mental health center (e.g., swimming), and mentors provided transportation to reduce treatment barriers. Mentors received a minimum of 24 hours of initial training and weekly supervision by an experienced clinician. Behavioral principles including modeling, praise, and token economies were used to reinforce appropriate social behavior among participants. Mentors also engaged in supportive conversations with youth. Youth in the CBT-mentoring group condition showed more improved social problemsolving and greater decreases in externalizing and internalizing symptoms after 3 months, compared to the waitlisted controls. This study provides an innovative model for how training and supervision of mentors and delivery of a program may be coordinated through a mental health services center. Such a platform suggests opportunities for coordinating and integrating mentoring and traditional services that might be more challenging to arrange across sites, such as between a mentor from a church-based program and a therapist in private practice. Another important strength of the model is the effort to reduce barriers to treatment by transporting youth to the group when necessary. The authors’ use of a group mentoring format is efficient, but contrasts with the traditional conceptualization of mentoring as a oneon-one relationship. However, DuBois et al. (2011)

332  SPECIAL POPULATIONS found evidence of effectiveness for programs in which mentoring occurred in group contexts. We recommend that CBT-mentoring be compared with another active intervention (such as usual services) to demonstrate effectiveness. Additionally, it is not clear whether this fusion of widely used clinical intervention strategies with supportive relational processes should be considered mentoring any more than traditional group CBT for youth with mental health needs or the relatively widespread use of trained paraprofessionals to work with children who have behavioral problems. In the mental health field, paraprofessionals include parents, teachers, and others who are not mental health professionals but who have been trained to deliver a particular intervention. There is evidence to suggest that trained and supervised paraprofessionals may be as effective as mental health professionals, particularly when treating undercontrolled behavior problems (Weisz, Weiss, Han, Granger, & Morton, 1995). Furthermore, traditional CBT group leaders could be considered adult role models, and they typically attend to relational processes in order to promote respect, trust, and inclusion. Thus, we recommend explicit testing of whether the mentoring and CBT components of Jent and Niec’s (2009) intervention contribute uniquely to positive outcomes (i.e., group mentoring vs. group CBT-mentoring vs. group CBT). Additionally, we expect CBT-mentoring groups that are tailored further according to youth mental health problem (e.g., anxiety disorders) will be more effective than general CBT interventions for youth with a range of problems. This expectation suggests directions for future studies. PrimeTime. In an RCT, PrimeTime, a multicomponent school- and community-based program was compared to a single-component school-based mentoring program (Lunch Buddy) for 145 2ndand 3rd-grade children (from 13 schools) who showed significant aggression based on teacher reports and/or peer nominations. Children assigned to PrimeTime received weekly mentorship from an undergraduate student for three semesters. They also participated in weekly skills training groups focused on conflict resolution and affect regulation, which were provided by the undergraduate mentors’ doctoral student supervisors. Finally, these same doctoral students provided home and school consultation to the children’s parents and teachers up to twice each month. Children assigned to the Lunch Buddy program met twice weekly during school lunch periods with a new undergraduate mentor each semester; mentors attended an orientation session but had no formal training or supervision. Surprisingly, children

in the less-intensive Lunch Buddy condition showed superior behavioral outcomes at 1- and 2-year follow-ups, especially in higher adversity schools, though PrimeTime was more effective in low adversity schools (Hughes, Cavell, Meehan, Zhang, & Collie, 2005). These findings are even more surprising given that the Lunch Buddy relationship was deliberately “diluted,” and, indeed, youth rated these relationships as less emotionally supportive than PrimeTime mentoring. Results of an earlier, smaller RCT (Cavell & Hughes, 2000) involving 2nd- and 3rd-grade youth identified as aggressive by peers and teachers showed comparable outcomes for youth in PrimeTime compared to those receiving community-based mentoring with limited training and support. Design features of these trials and interventions make it impossible to distinguish the positive and negative effects of specific program components. The lack of a no-treatment control group also means the potentially positive effects of Lunch Buddy and standard mentoring programs cannot be discerned. Additionally, it is not clear how the effects of the mentoring components of these trials would compare to those of behavior management programs, which are the best supported interventions for decreasing child aggression (Eyberg et al., 2008). Other components of these programs are of interest and merit further study. Illustratively, PrimeTime trained mentors to allow rather than avoid conflict. This approach contrasts with other programs that focus on positive attention (e.g., Jackson’s [2002] intervention for children at risk for delinquency teaches mentors to anticipate and avoid conflict). Which of these approaches is more effective at strengthening the mentoring relationship and contributing to social-emotional development of aggressive children deserves formal scrutiny. The Check and Connect Program. The Check and Connect Program uses a school-based paraprofessional described as mentor, advocate, and academic service coordinator (Sinclair, Christenson, Lehr, & Anderson, 2003). The role of this consistent adult figure is to promote school connection across multiple years. He or she maintains an emotionally supportive relationship with a student, tracks the student’s academic functioning, and adjusts contact with the student and family accordingly. This includes increasing contact when appropriate, modeling and coaching cognitive-behavioral problemsolving approaches, encouraging participation in school activities, and coordinating school services (e.g., behavioral plan, tutoring). In one trial, 9th graders with behavioral or emotional disabilities

Youth With Mental Health Needs   333 who were randomly assigned to Check and Connect showed lower rates of school dropout and higher rates of attendance and engagement in individualized education plan (IEP) transition plans (i.e., working toward career and community goals) over a period of 4–5 years, compared to control youth (Sinclair, Christenson, & Thurlow, 2005). However, youth referred for mental health problems were not examined separately, and mental health outcomes were not reported. Early Risers. Check and Connect was integrated into Early Risers, a multicomponent prevention program aimed at kindergarteners identified as at-risk for aggression and other serious conduct problems (those with severe emotional-behavioral disorders were excluded from the trial). Early Risers also provided a summer school intervention that included a peer mentoring system, in which at-risk youth were paired with low-risk children to encourage modeling of social skills and prosocial peer affiliation. In an RCT, August, Hektner, Egan, Realmuto, and Bloomquist (2002) reported positive effects of Early Risers for social and academic outcomes, but not for aggression, hyperactivity, or impulsivity, after 3 years of continuous intervention in program participants, relative to controls. Six-year outcomes supported fewer ODD symptoms among Early Risers participants, compared to controls, but no differences in rates of ODD diagnosis or in drug involvement or conduct disorder symptoms (Bernat, August, Hektner, & Bloomquist, 2007). In support of Rhodes’s (2005) model of mentorship, effects of Early Risers were partially mediated by improved social skills (Bernat et al., 2007). Wraparound Services. Wraparound is a case management approach for youth with serious mental health problems (e.g., Burns, Schoenwald, Burchard, Faw, & Santos, 2000; Epstein et al., 2003). Wraparound teams typically are comprised of the youth, family members, formal and informal service providers, and natural supports in the family and community, which can include mentors. Controlled research on the effectiveness of this widely implemented approach is limited to three RCTs and four studies using nonequivalent matched controls; a meta-analysis based on these studies supported only small mean effects on mental health outcomes (Suter & Bruns, 2009). The independent effects of mentor involvement in wraparound teams have not been evaluated. One study of fidelity to the wraparound model found that 25% of the 112 team meetings examined included a mentor (Epstein et al., 2003). Bidirectional influences

may exist between mentors and the broader system of care for impaired youth. That is, mentors may provide the team with observations from a nonfamily, nonprofessional perspective, and the team may help the mentor better support the youth’s goals. Although promising, we cannot assume this approach is optimal or even beneficial without controlled research. For example, youth may value and benefit from a mentoring relationship that is perceived as untainted by their histories and functioning in other domains (family, school, therapy). Multisystemic Therapy (MST). A primary goal of MST for youth with serious behavioral disturbances is to treat youth in their natural environments using family- and community-based services, rather than place them in expensive and potentially ineffective or even deleterious residential placements. MST teams identify and intensively intervene on the multiple family, community, and service system influences on problem behaviors. A component of this complex intervention emphasizes cultivation of indigenous family supports, which can include informal youth mentors and formal community mentorship programs (Henggeler & Lee, 2003). The evidence base for MST is extensive. Numerous RCTs support its effectiveness compared with usual services (e.g., individual therapy, incarceration, or inpatient hospitalization) for youth with diverse problems, including those in psychiatric crisis, youth from maltreating families, and juvenile violent and sexual offenders (see Henggeler & Lee, 2003). To our knowledge, the unique contributions of mentoring to MST effectiveness have not been assessed. Multidimensional Treatment Foster Care (MTFC). MTFC is a family-based behavioral program for youth in out-of-home care placements, including youth with severe behavioral disturbances. Like MST, the MTFC model emphasizes that treatment gains made in natural settings are more likely to persist than those made in standard-care conditions in which troubled youth reside together. Thus, youth live with trained and supervised foster parents, attend community schools, and receive individual and family therapy; all services are centrally supervised and coordinated. The effectiveness of MTFC relative to usual services is supported for youth referred from mental health, child welfare, and juvenile justice systems (see Chamberlain & Smith, 2003). Two MTFC components are highly relevant to mentorship. First, trained foster caregivers not only implement a behavioral program, which would be akin to functioning as trained mental

334  SPECIAL POPULATIONS health paraprofessionals, but they also may develop a genuine caring relationship with youth. Indeed, one mediator of beneficial effects of MTFC on boys’ reduced delinquency and arrests in the year following care placement was a more positive adult caretaker-youth relationship, as reported by youth and caretakers (Eddy & Chamberlain, 2000). Second, youth in MTFC have weekly individual meetings with a gender-matched behavioral support specialist (BSS), often in community settings such as restaurants. BSSs provide emotional support and use behavioral principles to help youth interact in prosocial ways with individuals in the community. Consistent with Rhodes’s (2005) mentoring model, foster care parents and BSSs may positively influence youth social-emotional development through modeling and behavioral intervention. They may contribute to identity development by helping youth imagine alternative futures for themselves and by offering social and cultural capital that may be lacking in their present networks. Summary and Research Recommendations The evidence base for mentoring of youth with mental health needs is limited, likely because the literature on targeted mental health treatments for a number of child and adolescent mental health problems is still emerging, and because there has been limited research on stand-alone or adjunctive mentorship approaches. The strongest evidence for the unique effects of mentoring comes from four RCTs conducted within relatively narrowly defined population groups, compared to the broad range of mental health problems with which youth present (Cavell & Hughes, 2000; Cavell, Elledge, Malcolm, Faith, & Hughes, 2009; Hughes et al., 2005; C. A. King et al., 2006, 2009). A logical next step is to broaden the evidence base by experimentally testing the unique effects of the other stand-alone mentoring interventions for youth with mental health needs such as those described above. Studies of programs that integrate mentoring and interventions for mental health problems (e.g., Jent & Niec, 2006, 2009; K. A. King et al., 2002) should follow the lead of clinical trials by testing problem-tailored interventions in less diagnostically heterogeneous samples as this approach should make it more likely for them to demonstrate efficacy and be adopted in clinical practice. The lack of appropriate controls in studies of mentoring for youth with mental health problems limits conclusions not only about benefits, but also about harm or reduced benefit compared to other services. Given the evidence reviewed on the potentially detrimental effects of some mentoring

relationships (DuBois et al. 2002; Grossman & Rhodes, 2002), and the likelihood that youth with mental health problems may be at greater risk for these experiences, researchers must be vigilant regarding this possibility when designing trials. Well-established multicomponent interventions have not conducted dismantling studies (e.g., randomization to MTFC with or without BSS) to isolate the independent effects of mentorship. This may be particularly important for those interventions that incorporate trained and supervised paraprofessionals utilizing behavioral management principles. Because such interventions are known to be highly effective (e.g., Weisz et al., 1995), it is important to ascertain the incremental benefit of the mentoring relationship. However, in many mentorship approaches, either the mentors are not specifically trained in a particular type of effective mental health intervention or the mentor role is so central that it cannot be meaningfully differentiated from the rest of the intervention (e.g., Check and Connect). Nevertheless, future trials could include measures of the quality and quantity of youth contacts with mentors to evaluate the potential role of this component in positive and negative outcomes. Nonexperimental analysis of whether the intensity or duration of mentoring among those who received it showed different outcomes (i.e. “dosage effects”) would be relatively easy to conduct. Research on mentoring youth with chronic disabling psychiatric problems is primarily descriptive, but it is an area of expressed need (Epstein et al., 2003; Jivanjee, Kruzich, & Gordon, 2009). The transition to adulthood is particularly challenging for such youth, especially given the oftenabrupt shift from child mental health systems to adult systems. Transition programs exist in which paid mentors provide emotional support and practical guidance regarding educational attainment, job preparation, and living skills (Rosenberg, 2008). These functions are consistent with Rhodes’s (2005) model of how mentors enhance youth outcomes via social-emotional, cognitive, and identity development, as well as the proposed pathway through treatment adherence. We believe it is unrealistic to expect or suggest that traditional mentoring programs could meet these needs. However, specialized mentoring programs that offer a higher standard of training, mentor support, and coordination with a treatment team or case manager may fill an important service and advocacy gap. Controlled research is needed. We offer several other brief research notes and recommendations. First, for problems with a stronger intervention research base (e.g., adolescent

Youth With Mental Health Needs   335 depression), mentorship as an adjunctive condition should be tested; we hypothesize that the effects of mentoring on youth outcomes will be partially mediated by its favorable effects on treatment adherence. Second, naturalistic studies of mentoring and youth psychopathology raise the same questions that Rhodes (2005) raised regarding directionality of relations in nonclinical populations. That is, there may be bidirectional effects between youth functioning and their ability to attract and benefit from mentoring. Researchers could use panel designs to determine when and how participants contribute to positive or negative dyadic process and how such processes could be controlled (e.g., improved, prevented, or interrupted with additional screening or mentor training). Third, and related to these last two points, mentors’ efforts may be affected by stigma. Mentors may react to diagnostic labels for youths’ problems, and view diagnoses as evidence that problems are family based, intractable, or an excuse for inappropriate behavior. These possibilities should be studied, and research in the field of stigma (e.g., Pescosolido, Jensen, Martin, Perry, Olafsdottir, & Fettes, 2008) should be used to guide training and supervision of mentors.

Practice We conclude with practice guidelines for mentorship programs involving youth with mental health needs. A checklist is provided in Table 22.1. At the most general level, we recommend mentoring as a supplemental intervention for youth with significant mental health needs. Evidence-based treatments that target identified mental health problems more directly than do mentoring are indicated, and indirect harm may come from providing nonspecific and likely weaker interventions, particularly if the interventions are suggested or implied to be a replacement for treatment. However, because many youth have unidentified mental health problems and approximately half of those with identified problems have not obtained treatment, most if not all youth mentorship programs are serving youth with mental health needs. As such, we recommend establishing a program infrastructure (see Table 22.1) that incorporates screening and referral processes and ensures that mentors have clear roles and know how to seek additional support. Screening should be conducted with parental permission and participation. A number of screening tools are available for which

Table 22.1   Checklist for Practitioners Topic

Recommendations

Program design and goals

Mentoring should supplement rather than replace evidence-based clinical treatments for youth with mental health needs. Mentoring programs may play an important role in helping youth with mild to moderate mental health needs, particularly through support of treatment adherence, support of new skills and ways of thinking learned in treatment, and the use of evidence-based strategies that have been successfully implemented by parents, teachers, and paraprofessionals (e.g., reinforcing engagement in healthy activities). Following “best practices” in the screening and training of mentors is especially critical for youth with mental health needs, to maximize benefit and avoid harm. Group mentoring approaches are not currently recommended, especially for youth presenting with symptoms sensitive to peer reinforcement (e.g., antisocial behavior). Strong, naturally occurring prosocial peer networks can provide needed support. However, formal (i.e., program-based) mentoring by peers is not currently recommended for youth with identified mental health concerns. For youth with chronic and impairing conditions who are transitioning to adult services and independent living, trained mentors in specialized programs (i.e., not mentors from traditional programs) could provide needed emotional and pragmatic supports.

Program practices

Screen youth for mental health concerns, encourage mentors to report youth needs to program staff, and recommend professional evaluation and treatment, as appropriate.

(Continued)

336  SPECIAL POPULATIONS Table 22.1  (Continued) Topic

Recommendations Learn about available mental health treatment resources in the community and share this information with youth and their caregivers, as appropriate. Facilitate youth entry into effective treatment by learning about such treatments (see Society of Clinical Child and Adolescent Psychology, 2010) and by developing a referral list of mental health providers who offer these services. Include at least one mental health professional on the mentoring program staff or arrange for consultative time from a community mental health professional. This will be helpful in making decisions about the screening and referral protocol (including how to link youth with effective treatments), and in addressing concerns about specific youth. Provide mentors with a clear delineation of their role, including how it differs from that of a mental health professional. Provide mentors with crisis contact information. The number of local services for general medical or psychiatric emergencies (including 911), and contact numbers for the mentoring program director or the delegated individual responsible for addressing mentor concerns.

implementation would be feasible. We highlight two instruments that assess multiple domains of functioning, are relatively brief, and have good psychometric properties. First, the Pediatric Symptom Checklist (Jellinek, Murphy, & Burns, 1986; Pagano, Cassidy, Little, Murphy, & Jellinek, 2000) is completed by youth and/or caregivers. Second, the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997) has parent and youth self-report versions. Self-reports may be useful for adolescents who may not share their emotional distress with caregivers. Both forms are available online at no charge (see Massachusetts General Hospital, n.d.; Youthinmind, n.d.). As discussed earlier, when families seek treatment for youth mental health needs, the treatment is likely to be time limited due to resource limitations or nonadherence. Moreover, the optimal evidencebased treatment may not be available in the youth’s community, and the treatment obtained may not fully treat or “cure” the presenting problems. As such, we believe that mentoring may play an important role in helping some youth who are at risk for mental health problems or who have mild to moderate problems related to depression, anxiety, interpersonal conflict, and disruptive behavior. Keep in mind, however, that the spectrum of youth mental health problems is broad, and the severity and acuity of problems are highly variable. Relationships with youth with some mental health problems may be challenging to maintain. Thus, youth should be screened to be sure their mental health needs are not

of a type or severity (acutely psychotic, acutely suicidal, substance dependent, severe conduct disorder) for which it may be unrealistic for a mentor to meet the youth’s needs. Still, mentoring programs can serve a valuable community function by referring youth with these more significant needs for professional services. Although further research is needed, we hypothesize that mentoring may be helpful with mild to moderate mental health problems in one or more of the following ways, each of which relates to an evidence-based treatment. First, mentoring may facilitate mental health treatment linkage and treatment engagement. Second, mentors can encourage and positively reinforce adaptive behaviors; specifically, this may occur by helping youth engage in more activities that the youth identifies as pleasurable, which is one component of evidence-based treatments for youth depression (“behavioral activation”), and by encouraging youth to do what they are afraid of and typically avoid, which is a component of evidence-based treatments for anxiety disorders (“exposure”). Third, mentors may support youth through active listening and the facilitation of healthy problem-solving related to current interpersonal struggles and concerns. Fourth, mentors may be trained to implement behavioral principles when responding to problem behaviors. When mentorship is used to encourage and supplement evidence-based mental health treatment, such as the combination of CBT and an antidepressant medication for depressive disorders

Youth With Mental Health Needs   337 among adolescents, mentor supervisors should consult with the youth, parents, and therapists to decide whether it is appropriate to brief mentors on the logic and strategies of the treatment model. Although this has not been studied empirically, such briefing may help mentors support treatment engagement and adherence and avoid contradicting therapists or stigmatizing problems. However, many clinicians do not use evidence-based treatments; assigning mentors an active role in reinforcing nonefficacious treatments may detract from the natural benefits of mentoring. Mentoring programs should improve outcomes and avoid harming youth with mental health needs by following evidence-based practices for youth mentoring as these continue to evolve in the field (see DuBois et al., 2002, 2011; MENTOR, 2009), as articulated by DuBois et al. (2002) and Grossman and Rhodes (2002). As part of this adherence, mentors must be screened carefully (regarding availability and expectations), trained, and supported to avoid exposing youth to further rejection. As listed in Table 22.1, such support might involve a clear delineation of the mentor’s role, easy access to the program director for discussion of concerns, and the availability of a mental health professional for program consultation. Mentors also should be screened for antisocial norms and deviant behavior to avoid negative influences. Even conventional adults should be trained not to reinforce youth’s deviant behaviors and attitudes (e.g., by sharing a humorous story about duping a teacher). This point bears further emphasis, as it may require considerable selfmonitoring and maturity to resist doing so in an effort to establish trust and connection with youth. The YST intervention manual (C. A. King, Kramer, & Preuss, 2000; available from Cheryl King upon request) provides specific guidance and written scripts regarding identifying and screening potentially supportive adults, discussing role expectations, and providing training and ongoing role support to those who agree who serve in the role of a supportive adult within the program. Related to the last two points, mentoring from nonadults may be more likely to expose youth with mental health needs to rejection and encouragement of deviant behavior. DuBois et al. (2011) found that programs that used adult and older peer (nonadult) mentors had similarly positive youth development outcomes. Still, it remains to be seen whether this effect holds true for youth with mental health needs (and see review of C. A. King et al., 2006, above). Thus, currently we cannot recommend mentoring of youth with mental health needs by nonadults.

Next, although general evidence indicates that group mentoring approaches also yield positive effects (DuBois et al., 2011), it is not clear whether this finding would hold within the population of youth with mental health needs. Group approaches may appear to be efficient, but the necessary ingredients of effective relationships (mutuality, trust, and empathy) may not develop if group needs are prioritized, and if the higher risk youth command more adult attention. Also, the aggregation of highrisk youth can reinforce negative behaviors, such as antisocial behavior, self-harm, and eating disorder symptoms (e.g., Dishion & Dodge, 2005).

Conclusion Mentoring for youth with identified mental health needs has only recently become a focus of systematic research. The theoretical basis for mentoring approaches with this population and the emerging research evidence suggest that mentoring is promising as an adjunctive intervention to reinforce or supplement the effects of formal mental health treatments. Despite a growing number of evidencebased treatments for mental health problems, not all youth benefit from these treatments, many of the youth who do benefit do not achieve a full resolution of their mental health problems, and these problems are often embedded in an environmental context of risk that the mental health treatment does not address. Although mentoring is not recommended as a substitute for formal mental health treatments, it could have direct positive benefits for youth with mild to moderate mental health problems that do not involve psychosis, acute suicide risk, substance dependence, or other severe disorders. We believe that systematic and controlled research in this area is a high priority, and we have highlighted several promising and important scientific directions. Many of the most fundamental research questions regarding mentorship with this population remain unanswered. Nevertheless we can support mentoring approaches that follow the practice recommendations regarding the importance of quality mentorship relationships based on mutuality and trust, and the careful screening, training, support, and supervision of mentors.

References American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Author.

338  SPECIAL POPULATIONS August, G. J., Hektner, J. M., Egan, E. A., Realmuto, G. M., & Bloomquist, M. L. (2002). The Early Risers Longitudinal Prevention Trial: Examination of 3-year outcomes in aggressive children with intentto-treat and as-intended analysis. Psychology of Addictive Behaviors, 16, 27–39. Barrett, P., Farrell, L., Pina, A., Peris, T., & Piacentini, J. (2008). Evidence-based psychosocial treatments for child and adolescent obsessive-compulsive disorder. Journal of Clinical Child and Adolescent Psychology, 37, 131–155. Beam, M. R., Gil-Rivas, V., Greenberger, E., & Chen, C. (2002). Adolescent problem behavior and depressed mood: Risk and protection within and across social contexts. Journal of Youth and Adolescence, 31, 343–357. Bernat, D. H., August, G. J., Hektner, J. M., & Bloomquist, M. L. (2007). The Early Risers preventive intervention: Testing for six-year outcomes and mediational processes. Journal of Abnormal Child Psychology, 35, 605–617. Brown, R. T., Antonuccio, D. O., DuPaul, G. J., Fristad, M. A., King, C. A., Leslie, L. K.,  . . . Vitiello, B. (2008). Childhood mental health disorders: Evidence base and contextual factors for psychosocial, psychopharmacological and combined interventions. Washington, DC: American Psychological Association. Burns, B. J., Schoenwald, S. K., Burchard, J. D., Faw, L., & Santos, A. B. (2000). Comprehensive communitybased interventions for youth with severe emotional disorders: Multisystemic therapy and the wraparound process. Journal of Child and Family Studies, 9, 283–314. Cavell, T. A., Elledge, L. C., Malcolm, K. T., Faith, M. A., & Hughes, J. A. (2009). Relationship quality and the mentoring of aggressive, high-risk children. Journal of Clinical Child and Adolescent Psychology, 38, 185–198. Cavell, T. A., & Hughes, J. A. (2000). Secondary prevention as context for assessing change process in aggressive children. Journal of School Psychology, 38, 199–235. Chamberlain, P., & Smith, D. K. (2003). Antisocial behavior in children and adolescents: The Oregon Multidimensional Treatment Foster Care Model. In A. E. Kazdin & J. R. Weisz (Eds.), Evidence-based psychotherapies for children and adolescents (pp. 282–300). New York: Guilford Press. Clarke, G. N., DeBar, L. L., & Lewinsohn, P. M. (2009). Cognitive-behavioral group treatment for adolescent depression. In A. E. Kazdin & J. R. Weisz (Eds.), Evidence-based psychotherapies for children and adolescents (pp. 120–134). New York: Guilford Press. Costello, E. J., Mustillo, S., Erkanli, A., Keeler, G., & Angold, A. (2003). Prevalence and development of psychiatric disorders in childhood and adolescence. Archives of General Psychiatry, 60, 837–844. David-Ferdon, C., & Kaslow, N. (2008). Evidence-based psychosocial treatments for child and adolescent

depression. Journal of Clinical Child and Adolescent Psychology, 37, 62–104. Dishion, T. J., & Dodge, K. A. (2005). Peer contagion in interventions for children and adolescents: Moving towards an understanding of the ecology and dynamics of change. Journal of Abnormal Child Psychology, 33, 395–400. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring program for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Karcher, M. J. (2005). Youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 2–12). Thousand Oaks, CA: Sage. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., & Silverthorn, N. (2005a). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524. DuBois, D. L., & Silverthorn, N. (2005b). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92. Eby, L. T., Allen, T. D., Evans, S. C., Ng, T., & DuBois, D. L. (2008). Does mentoring matter? A multidisciplinary meta-analysis comparing mentored and non-mentored individuals. Journal of Vocational Behavior, 72, 254–267. Eddy, J. M., & Chamberlain, P. (2000). Family management and deviant peer association as mediators of the impact of treatment condition on youth antisocial behavior. Journal of Consulting and Clinical Psychology, 68, 857–863. Epstein, M. H., Nordness, P. D., Kutash, K., Duchnowski, A., Schrepf, S., Benner, G. J., & Nelson, J. R. (2003). Assessing the wraparound process during family planning meetings. Journal of Behavioral Health Services & Research, 30, 352–362. Eyberg, S. M., Nelson, M. M., & Boggs, S. R. (2008). Evidence-based psychosocial treatments for child and adolescent with disruptive behavior. Journal of Clinical Child and Adolescent Psychology, 37, 215–237. Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38, 581–586. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Henggeler, S. W., & Lee, T. (2003). Multisystemic treatment of serious clinical problems. In A. E. Kazdin & J. R. Weisz (Eds.), Evidence-based psychotherapies for children and adolescents (pp. 301–322). New York: Guilford Press.

Youth With Mental Health Needs   339 Hidalgo, R., Tupler, L., & Davidson, J. (2007). An effectsize analysis of pharmacologic treatments for generalized anxiety disorder. Journal of Psychopharmacology, 21, 864–872. Hughes, J. N., Cavell, T. A., Meehan, B., Zhang, D., & Collie, C. (2005). Adverse school context moderates the outcomes of two selective intervention programs for aggressive children. Journal of Consulting and Clinical Psychology, 73, 731–736. Husky, M. M., Sheridan, M., McGuire, L., & Olfson, M. (2011). Mental health screening and follow-up care in public high schools. Journal of the American Academy of Child & Adolescent Psychiatry, 50, 881–891. Jackson, Y. (2002). Mentoring for delinquent children: An outcome study with young adolescent children. Journal of Youth and Adolescence, 31, 115–122. Jellinek, M. S., Murphy, J. M., & Burns, B. J. (1986). Brief psychosocial screening in outpatient pediatric practice. Journal of Pediatrics, 109, 371–378. Jent, J. F., & Niec, L. N. (2006). Mentoring youth with psychiatric disorders: The impact on child and parent functioning. Child and Family Behavior Therapy, 28, 43–58. Jent, J. F., & Niec, L. N. (2009). Cognitive behavioral principles within group mentoring: A randomized pilot study. Child and Family Behavior Therapy, 31, 203–218. Jivanjee, P., Kruzich, J. M., & Gordon, L. J. (2009). The age of uncertainty: Parent perspectives on the transition of young people with mental health difficulties to adulthood. Journal of Child and Family Studies, 18, 435–446. Joiner, T. E., Jr. (2005). Why people die by suicide. Cambridge, MA: Harvard University Press. Kazdin, A. E., Siegel, T. C., & Bass, D. (1990). Drawing on clinical practice to inform research on child and adolescent psychotherapy: Survey of practitioners. Professional Psychology: Research and Practice, 21, 189–198. Kercher, A., & Rapee, R. M. (2009). A test of a cognitive diathesis-stress generation pathway in early adolescent depression. Journal of Abnormal Child Psychology, 37, 845–855. King, C. A., Klaus, N., Kramer, A., Venkataraman, S., Quinlan, P., & Gillespie, B. (2009). The YouthNominated Support Team for suicidal adolescents– version II (YST-II): A randomized controlled efficacy trial. Journal of Consulting and Clinical Psychology, 77, 880–893. King, C. A., Kramer, A., & Preuss, L. (2000). YouthNominated Support Team intervention manual. Department of Psychiatry, University of Michigan. King, C. A., Kramer, A., Preuss, L., Kerr, D. C. R., Weisse, L., & Venkataraman, S. (2006). Youth-Nominated Support Team for suicidal adolescents (version 1): A randomized controlled trial. Journal of Consulting and Clinical Psychology, 74, 199–206. King, K. A., Vidourek, R. A., Davis, B., & McClellan, W. (2002). Increasing self-esteem and school connect-

edness through a multi-dimensional mentoring program. Journal of School Health, 72, 294–299. Massachusetts General Hospital. (n.d.). Pediatric Symptom Checklist (PSC) forms. Retrieved from http://www2.massgeneral.org/allpsych/psc/psc_ forms.htm MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. Retrieved from http://www.mentoring.org/program_resources/ elements_and_toolkits Murphy, R., Cooper, Z., Hollon, S. D., & Fairburn, C. G. (2009). How do psychological treatments work? Investigating mediators of change. Behaviour Research and Therapy, 47, 1–5. Murray, C., & Greenberg, M. T. (2001). Relationships with teachers and bonds with school: Social emotional adjustment correlates for children with and without disabilities. Psychology in Schools, 38, 25–41. Pagano, M. E., Cassidy, L. J., Little, M., Murphy, J. M., & Jellinek, M. S. (2000). Identifying psychosocial dysfunction in school-age children: The Pediatric Symptom Checklist as a self-report measure. Psychology in the Schools, 37, 91–106. Pelham, W., & Fabiano, G. (2008). Evidence-based psychosocial treatments for attention-deficit/hyperactivity disorder. Journal of Clinical Child and Adolescent Psychology, 37, 184–214. Pescosolido, B. A., Jensen, P., Martin, J. K., Perry, B., Olafsdottir, S., & Fettes, D. (2008). Public knowledge and assessment of child mental health problems: Findings from the National Stigma Study-Children. Journal of the American Academy of Child and Adolescent Psychiatry 47, 339–349. Rhodes, J. E. (2005). A model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30–43). Thousand Oaks, CA: Sage. Rosenberg, L. (2008). Building a meaningful future for young people with mental illness. Journal of Behavioral Health Services & Research, 35, 362–364. Shaffer, D., Gould, M. S., Fisher, P., & Trautman, P. (1996). Psychiatric diagnosis in child and adolescent suicide. Archives of General Psychiatry, 53, 339–348. Silverman, W., & Hinshaw, S. (2008). The second special issue on evidence-based psychosocial treatments for children and adolescents: A 10-year update. Journal of Clinical Child and Adolescent Psychology, 37, 1–7. Silverman, W., Pina, A., & Viswesvaran, C. (2008). Evidence-based psychosocial treatments for phobic and anxiety disorders in children and adolescents. Journal of Clinical Child and Adolescent Psychology, 37, 105–130. Sinclair, M. F., Christenson, S. L., Lehr, C. A., & Anderson, A. R. (2003). Facilitating student engagement: Lessons learned from Check & Connect longitudinal studies. California School Psychologist, 8, 29–41. Sinclair, M. F., Christenson, S. L., & Thurlow, M. L. (2005). Promoting school completion of urban

340  SPECIAL POPULATIONS secondary youth with emotional or behavioral disability. Exceptional Children, 71, 465–482. Sirey, J., Bruce, M., Alexopoulos, G., Perlick, D., Friedman, S., & Meyers, B. (2001). Stigma as a barrier to recovery: Perceived stigma and patientrated severity of illness as predictors of antidepressant drug adherence. Psychiatric Services, 52, 1615–1620. Society of Clinical Child and Adolescent Psychology. (2010). Evidence-based mental health treatment for children and adolescents. Retrieved from http:// www.effectivechildtherapy.com/sccap Stice, E., Ragan, J., & Randall, P. (2004). Prospective relations between social support and depression: Differential direction of effects of parent and peer support? Journal of Abnormal Psychology, 113, 155–159. Suter, J. C., & Bruns, E. J. (2009). Effectiveness of the wraparound process for children with emotional and behavioral disorders: A meta-analysis. Clinical Child and Family Psychology Review, 12, 336–351. Treatment for Adolescents With Depression Study Team. (2004). Fluoxetine, cognitive-behavioral therapy, and their combination for adolescents with depression. JAMA, 44, 1137–1144. U.S. Public Health Service. (1999). The Surgeon General’s call to action to prevent suicide. Rockville, MD:

U.S. Department of Health and Human Services. Weisz, J., & Jensen, A. (2001). Child and adolescent psychotherapy in research and practice contexts: Review of the evidence and suggestions for improving the field. European Child and Adolescent Psychiatry, 10, 12–18. Weisz, J. R., Jensen-Doss, A., & Hawley, K. M. (2006). Evidence-based youth psychotherapies versus usual clinical care: A meta-analysis of direct comparisons. American Psychologist, 61, 671–689. Weisz, J. R., Weiss, B., Han, S. S., Granger, D. A., & Morton, T. (1995). Effects of psychotherapy with children and adolescents revisited: A meta-analysis of treatment outcome studies. Psychological Bulletin, 117, 450–468. Wickrama, K. A. S., Conger, R. D., Lorenz, F. O., & Jung, T. (2008). Family antecedents and consequences of trajectories of depressive symptoms from adolescence to young adulthood: A life course investigation. Journal of Health and Social Behavior, 49, 468–483. Woo, S. M., & Keatinge, C. (2008). Diagnosis and treatment of mental disorders across the lifespan. Hoboken, NJ: Wiley. Youthinmind. (n.d.). Information for researchers and professionals about the Strengths & Difficulties Questionnaires. Retrieved from http://www.sdq info.com

23  YOUTH IN FOSTER CARE Preston A. Britner, Kellie G. Randall, and Kym R. Ahrens

Introduction During the past decade, mentoring of children and adolescents who have been placed in foster care or other out-of-home placements, such as residential facilities or group homes, has begun to receive more attention. Some formal mentoring programs focusing exclusively on foster youth have been created, whereas other foster youth continue to be served by mentoring programs that include youth from a variety of circumstances. Still other instances of mentoring take place informally as naturally occurring mentors aid youth in care. We consider all of these approaches as we discuss the mentoring issues unique to this population. Based on findings reported in their meta-analysis of mentoring program effects, DuBois, Holloway, Valentine, and Cooper (2002) suggested that youth from backgrounds of greater social and economic risk both have the most to gain and tend to benefit the most from youth mentoring, especially when best practices are employed and strong relationships are formed. Findings of a meta-analysis of more recent research (DuBois, Portillo, Rhodes, Silverthorn, and Valentine, 2011) supported a similar conclusion, although benefits were somewhat diminished for programs that focused on samples of youth who were rated as relatively high in terms of both individual and environmental risk, as arguably could be the case for youth in foster care. In interviews with foster youth, foster parents, and child welfare professionals, all three groups report long-term relationships to be extremely important, perhaps more so than formal services (Geenen & Powers, 2007). Anecdotally “resilient” former foster youth have also emphasized mentoring and supportive relationships as important to their success as young adults (Hass & Graydon,

2009). In this chapter, we present the current state of theory and (more limited) research supporting mentoring initiatives for youth in foster care. We review current program models and propose several recommendations for improving mentoring practice with foster youth. As background for our discussion, the remainder of this introduction provides a brief overview of youth in foster care and issues pertinent to this population. Foster Care Population In 2008, approximately 463,000 U.S. children were in foster care. The average length of stay for these children was 27.2 months. A large number (47%) were placed in a nonrelative foster family home (U.S. Department of Health & Human Services, 2009). Whereas many children find themselves in just one out-of-home placement, others bounce through a series of placements. Outcomes for the 285,000 children exiting foster care in 2008 varied greatly, from reunification with parent(s) or primary caregivers (52%) to adoption (19%) to emancipation (10%) (U.S. Department of Health and Human Services, 2009). Situations of abuse and neglect are the most common reasons why children enter the foster care system. Other less common reasons for children’s placement into foster care include parental imprisonment (see Eddy et al., this volume, Chapter 25), death, and incapacitation (U.S. Department of Health and Human Services, 2009). Co-occurrence With Child Abuse and Neglect. Most youth in foster care have been maltreated; conversely, most maltreated children do not enter foster care. In 2008, 3.3 million referrals were made to child 341

342  SPECIAL POPULATIONS protective service agencies in the United States regarding the welfare of approximately 6 million children. Most were investigated, and approximately 772,000 children were found to be victims of psychological or emotional abuse, physical abuse, sexual abuse, and/or neglect. Compounding the nature of this social problem, 80% of perpetrators were parent(s) (U.S. Department of Health and Human Services, 2010). Abuse by parents may damage youths’ faith in others and sense of security more severely than when perpetrators are unfamiliar adults (Rohner & Britner, 2002). Older Youth Transitioning Out of Care. Nearly 30,000 youth transitioned out of foster care in 2008, either because they reached the age of majority or were emancipated early (U.S. Department of Health and Human Services, 2009). These youth face a unique set of issues as they are preparing to transition to adulthood. This is a time when most youth are facing critical decisions about education, careers, relationships, and other life circumstances. Many foster youth encounter great difficulties during this transitional period (Collins, Spencer, & Ward, 2010; Courtney & Dworsky, 2006). These youth frequently face this developmental transition perceiving less family-based support than their general population peers (Farruggia, Greenberger, Chen, & Heckhausen, 2006). Given this potential deficit in familial adult support and guidance for transitioning foster youth, mentoring is a particularly salient topic. A recent focus of research in this area has been on this segment of the population to see how mentor relationships can help youth in foster care transition into adulthood and their independent lives. Historical Background and Current Developments A majority of children in foster care display emotional or developmental problems while in care (Garland et al., 2001). They are also at greater risk of many problems as they “age out” of the system, including unemployment, homelessness, unintended pregnancy, substance abuse, and costly adult forms of dependency, such as welfare and social services utilization, psychiatric commitment, and incarceration (Collins et al., 2010; Courtney & Dworsky, 2006; Dworsky & Courtney, 2009; Hafton, English, Allen, & DeWoody, 1994; Hass & Graydon, 2009). Although these outcomes paint a bleak picture, there have been some bright spots on the policy landscape. The Chafee Foster Care Independence Act of 1999 (PL 106-169) provides health care coverage to former foster children until age 21 and doubled the funding for independent living programs for youth in

state custody, including mentoring, life skills, housing, and college tuition assistance programs. More recently, the Fostering Connections to Success and Increasing Adoption Act of 2008 (PL 110-351) gives states the ability to extend Title IV-E foster care to the age of 21 and requires states to have a transition planning process in place for youth leaving care. Unsuccessful attempts have been made to establish and support statewide mentoring programs through a Foster Care Mentoring Act by every Congress since 2002 (MENTOR, 2011; Spencer, Collins, Ward, & Smashnaya, 2010). Nonetheless, much of the formal mentoring that occurs with maltreated youth takes place when the youth are in state custody (Britner & Kraimer-Rickaby, 2005). Sometimes, mentoring is viewed as an option only after the child is in a stable placement, whereas in other instances of “foster care drift” (i.e., an all-too-common series of removals and/or returns to the family), a mentor may be assigned (or a natural mentor supported) in an attempt to present a rare “constant” in a youth’s life. Opportunities and Challenges A shift in focus from a risk-based to more strengths-based (e.g., resilience, positive youth development) and family-focused approaches in interventions with youth in foster care presents new opportunities for the study, support, and evaluation of mentoring relationships (Alpert & Britner, 2009; Spencer et al., 2010). Enthusiasm for the possible benefits of mentoring also must be tempered by concerns about the risks of failed mentoring relationships (Spencer et al., 2010). The potential for mentoring programs and relationships to have negative effects appears to be heightened for youth who enter programs with significant personal vulnerabilities, such as may stem from experiences of maltreatment and foster care (e.g., Grossman & Rhodes, 2002).

Theory Several theoretical models of the processes of youth mentoring seem particularly useful as frameworks for guiding research on mentoring programs and relationships for youth in foster care. Other theories point to techniques and cautions that are unique to this population. Established Mentoring Frameworks Rhodes’s Model. The model of youth mentoring proposed by Rhodes (2002) suggests that there is a

Youth in Foster Care   343 crucial step to mentoring in which mentor and youth protégé form an emotional bond. Theoretically, only then can the mentor influence the youth’s developmental outcomes by (a) enhancing social skills and emotional well-being, (b) improving cognitive skills through dialogue, and (c) promoting identity development through reflected appraisals. Rhodes also proposed several factors that may moderate the impact of youth mentoring. When this model is applied to youth in foster care, some of these factors reflect domains in which enduring and pronounced deficits are evident, such as interpersonal history and social competencies. Past experiences of trauma or disrupted relationships with caregivers early in life, and resultant maladaptive approaches to relationships later in life (e.g., Dozier & Rutter, 2008), may make it more difficult for foster youth to form a social bond with a mentor. Relationship duration, another moderator proposed by Rhodes, may be influenced by out-of-home placement as well, in that placement changes may truncate even high-quality mentoring relationships and thus decrease their impact. In sum, this theoretical model highlights youth in foster care, as a population, as having characteristics and experiences that likely make it more difficult for mentoring relationships or programs to achieve positive effects on youth. A second feature of Rhodes’s (2002) model that merits attention is the manner in which the impact of the mentor-youth relationship is expected to be mediated (at least in part) through improvements that occur via mentoring in the youth’s parental, peer, and other close relationships. Given the nature and quality of their prior relationships, this type of mediational pathway may be constrained for youth in foster care in at least two ways. First, important others (e.g., parents, peers, siblings) may be inherently limited in their capacity to enter into more favorable relationships with the youth; this is likely especially true with respect to parents, because the relationship has been disrupted by the child’s removal from the home. Second, foster youth by virtue of their interpersonal histories may be less open to investing effort in attempts at change or growth in their relationships. In short, there may be more obstacles in the path to positive outcomes of mentoring for youth in foster care. Despite the preceding considerations, Rhodes’s (2002) model also points toward the potential for certain processes that could facilitate significant effects of mentoring. Of particular note is the potential moderating role of the youth’s ecological circumstances. It could be that in the absence of family connections and consistent external support (e.g.,

due to multiple foster placements), the life circumstances of these youth are often such that the introduction of a mentor may represent a notably more salient and potentially influential positive source of change in the youth’s social resources and selfesteem than is the case for most other youth. The lack of family data on foster youth who are engaged in mentoring programs limits our ability to test many of the mediational paths suggested in Rhodes’s (2002) model. Do close relationships with important others (including mentors) protect against familial risks or help individuals to repair insecure “working models” of close relationships? What would be the effects of a mentor when the child-parent relationship is marked by abuse or is no longer intact? These important questions remain unanswered. Self-Esteem. Supportive mentoring relationships have been theorized to be vital to fostering selfesteem and thereby contributing to resilience in high-risk youth, including youth living in foster care (Yancey, 1998). DuBois, Neville, Parra, and Pugh-Lilly (2002) presented a structural model in which mentoring that results in the development of a significant bond between youth and mentors widens youths’ perceptions of their social support networks, which in turn bolsters youth resources, including selfesteem. Youth in foster care would seem particularly likely to be in need of “updated” working models (Bowlby, 1969) of how they see themselves—and how they think that others perceive them—as key foundations for a positive self-concept. Mentors may be able to help meet this need. Broader Theories and Their Relevance Several theoretical orientations outside of the core mentoring literature may be useful in guiding investigations and evaluations of mentoring for youth in foster care. Attachment and Parental Acceptance-Rejection Theories. Human attachment theory (Bowlby, 1969) has proved to be an extremely helpful approach to understanding the nature and influence of close relationships across the lifespan. The theory links the qualities of social interaction and caregiving behaviors within close social relationships to internal representations (working models) of relationships and to a variety of behavioral outcomes. Youth with histories of separations from caregivers, losses, and trauma are more likely to develop insecure (and even disorganized) attachment relationships and to manifest a host of negative behavioral outcomes (Dozier & Rutter,

344  SPECIAL POPULATIONS 2008). The ambiguous loss of family members (i.e., wondering whether parents will return or the family reunite) may also result in children becoming confused about roles and family boundaries (Bocknek, Sanderson, & Britner, 2009). The consequences of early experiences on relationship style have been shown to persist into adulthood (Thompson, 2008). Parental acceptance-rejection (PAR), “a theory of socialization that attempts to explain and predict major antecedents, correlates, and consequence of parental acceptance and rejection the world over” (Rohner, 1986, p. 14), similarly posits the importance of close parental relationships in a variety of outcome areas. Parental rejection has been implicated in an array of developmental, behavioral, and psychological problems among children, adolescents, and adults (e.g., different forms of psychopathology, depression, behavior problems, psychological adjustment problems, substance abuse, attachment disorders, academic problems, and troubled personal relationships); conversely, parental acceptance has been associated with an array of positive outcomes including a sense of happiness, life satisfaction, and low psychological distress across the life course (for a review, see Rohner & Britner, 2002). Both of these theories may have specific implications for mentoring relationships among youth in foster care. They suggest that as these youth establish relationships with mentors, early life disruptions and/or perceived rejections in parental relationships may influence working models and shape how these mentoring and other relationships are viewed. For example, given their interpersonal histories and experiences of being placed outside the home, youth in foster care may perceive themselves as experiencing rejection in mentoring relationships in response to behaviors (e.g., missed meetings) that may do relatively little to shake the security of other youth (Rhodes, 2002). Conversely, a history of experiencing rejection may complicate the initial relationship building; mentors of youth in foster care might encounter more anger or resistance on the part of the child. Both theories may be useful frameworks for the conceptualization and measurement of mentoring processes, relationship disruption, youths’ perceptions of rejection, and youths’ functioning in this specific population. Social Exchange Theory. Social exchange theory is based on the concept of social economics and thus posits that human interactions are transactions in which the aim is to maximize one’s rewards and minimize one’s costs (e.g., Sabatelli & Shehan, 1993). The theory has been used to explain what compels or motivates individuals to volunteer and

sacrifice their time in order to assist others for no apparent tangible rewards. When used as a theoretical basis for understanding volunteer participation and motivation, a social exchange differs from an economical exchange in that, whereas there is a general expectation of some future reward or return, the exact nature of the reward is not stipulated in advance. According to social exchange theory, relationships continue only when rewards outweigh costs (Sabatelli & Shehan, 1993). Applied to mentoring of youth in foster care, it is noteworthy that these youth may be prone to receive relatively high costs compared to rewards when presented with the opportunity to form a relationship with a mentor. Some of these youth—especially those with a history of rejection—may push hard to “test” mentors to see if they are committed to the relationship and to see if they will “stick around for awhile.” Likewise, others may simply be reluctant to engage or invest in relationships, given that few experiences in their past would suggest a high potential for positive payoffs. Theoretically, the perceived costs and rewards of investing in mentoring relationships could be important in mediating outcomes for youth from backgrounds of maltreatment by affecting such factors as the strength and quality of the ties they form with mentors and the likelihood of early termination due to relationship difficulties. Keller’s Contextual Mentoring Model. Using ecological and systems theories, Keller (2005) developed a framework of mentoring that emphasizes the network of relationships surrounding the youth. In this model, the mentor-youth relationship is embedded in a network that also includes the parent or guardian and the mentoring agency; exchanges in any of these relationships can have an impact on the rest of the relationships. For example, exchanges in the mentor-parent/guardian relationship may hinder or promote the mentor-child relationship, depending on the situation. Additionally, relationships between the adults (e.g., mentor and parent/guardian) can also have an impact on the child. When this view is applied to youth in foster care, the network of relationships is likely to expand to include foster parents, social workers, and other adults with whom a child in an out-of-home placement has regular contact. The introduction of a new foster parent, for example, may hinder or promote the role the mentor plays in the youth’s life. Given the increasing complexity of this network of relationships for foster youth, it is important to understand that the mentorchild relationship takes place in this larger context and to consider the resultant implications of and for such a complex system.

Youth in Foster Care   345 Key Theoretical Questions Consideration of these theories leads to some key questions regarding mentoring youth in foster care. What challenges might exist in the initiation and maintenance of relationships with this population? What mentor qualities might be most salient in these relationships? What elements of the relationship are most important in securing positive outcomes for youth in foster care? How can mentors best assist older youth as they face the transition out of foster care? And, are the answers to all of these questions affected by whether the mentoring relationship is a naturally occurring one or a formal, assigned one? This fledgling area of research is raising such questions, and it is beginning to answer a few of them.

Research Although some of the theories discussed in the preceding section have guided research on general youth mentoring, few have been applied directly to mentoring the foster youth population. Prior to publication of the first edition of the Handbook of Youth Mentoring, Rhodes, Haight, and Briggs (1999) reported that “virtually no research has examined the efficacy of mentoring programs with foster youth” (p. 186) or youth with maltreatment histories. This is a population that may be both particularly vulnerable and likely to benefit from mentoring relationships; thus, the dearth of research into this topic was surprising. To determine the current state of the research, a careful search of relevant electronic databases, including PsycINFO, PubMed, Social Work Abstracts, and Sociological Abstracts, was conducted using keywords such as foster care, out of home placement, and mentoring (and a variety of synonyms for each). Additionally, contact was made with researchers in this area of specialization (through listservs and direct correspondence) to ensure that the most up-to-date findings were included. The result was a small, but growing, body of literature. Much of the research is recent, perhaps indicating a growing recognition of the importance of understanding mentoring relationships in this population. Current Research Findings Here we present a brief review of existing research on attitudes toward mentoring, youth outcomes related to both natural and formal mentoring relationships, and the effect of mentoring youth in

foster care on the mentors themselves. We focus on research published since the first edition of the Handbook. Finally, we discuss the current limitations of the literature and suggest areas of focus for future research. Youth Attitudes Toward Mentoring. Whereas mentoring has the potential to have a positive effect on youth in foster care, questions remain about youths’ openness to and interest in developing mentoring relationships. Diehl, Howse, and Trivette (2011) asked 54 youth in foster care about their attitudes toward mentoring. Participants were youth between 10 and 17 years of age who were attending a campbased program for youth in foster care, which had the goal of finding the youth adoptive placements. Youth attitudes toward mentoring were measured by a four item, four-point Likert scale that asked about the youth’s interest in and readiness for an adult mentor. In addition to asking youth about their attitudes toward mentoring, researchers also collected information on youth asset scores (calculated based on a 20-item self-report scale given to youth with items such as, “I feel loved and supported by my caregivers,” and “People in my community respect me and my friends”), youth risk factor scores (based on intake interviews and including items on behavioral problems, mental health problems, and delinquent behaviors), youth perceptions of control, as well as demographic variables. Overall, the youth in the study had positive attitudes about mentoring; most youth in the study reported being interested in having a mentor in their life (Diehl et al., 2011). The only demographic variable that was associated with attitudes toward mentoring was race, with Caucasian youth having more positive attitudes about mentoring than African American youth. Age and gender were not associated with attitudes toward mentoring. Youth asset scores (the number of self-reported developmental assets) positively predicted attitudes toward mentoring. Yet youth risk factor scores were not related to attitudes toward mentoring; youth tended to have positive attitudes toward mentoring, whether they had high or low levels of risk. Finally, youth who reported higher perceived control in their lives had less positive attitudes toward mentoring. This might be because youth who perceive a high level of control over their lives feel they do not need a mentor, or they might be more likely to already have adults serving in mentor roles. Alternatively, youth who tend toward an avoidant attachment style (i.e., who have a high level of discomfort in relying on others, and therefore a strong desire for control over their own lives) may have had less positive attitudes about mentoring.

346  SPECIAL POPULATIONS Naturally Occurring Mentors. Natural mentors are nonparental adults who provide guidance and support to a young person; examples of natural mentors are teachers, neighbors, and extended family. These relationships have been shown to promote healthy behaviors and positive outcomes for many youth (see Keller & Blakeslee, this volume, Chapter 9, for a discussion of mentoring in relation to youths’ social networks). Several studies published since the first edition of the Handbook have examined natural mentoring relationships among foster youth; most suggest a positive association between these relationships and outcomes. For example, in a study of “successful” former foster youth, Hass and Graydon (2009) questioned 44 young adults who had experienced out-of-home placements and had subsequently succeeded in postsecondary education. Mentors and other sources of social support were credited, at least partially, with helping these young adults beat the odds. In another study, Ahrens, DuBois, Richardson, Fan, and Lozano (2008) used data from Waves 1–3 of the National Longitudinal Study of Adolescent Health (Add Health) and compared foster youth with (n = 160) and without (n = 150) mentors. Youth were considered mentored when they reported a relationship lasting at least 2 years with a nonparental adult beginning between ages 14 and 18 years. Despite demographic comparability, the mentored foster youth were more likely than the nonmentored foster youth to report favorable overall health and were less likely to report suicidal ideation, having received a diagnosis of a sexually transmitted infection, and having hurt someone in a fight in the prior year. Greeson, Usher, and Grinstein-Weiss (2010) also analyzed Add Health data and used the same definition of mentoring as Ahrens et al. (2008). They compared how these natural mentors might be associated with increased assets available to foster and nonfoster youth. The authors pointed out that natural mentors may be particularly important as role models for youth in foster care, who might lack other (family-of-origin-based) figures to serve in this capacity. The presence of a natural mentor who served as a role model was associated with greater likelihood of having a bank account for foster youth. Additionally, the presence of a natural mentor who “acted like a parent” was associated with increased earnings expectations. In a study using a dataset of 96 former foster youth, 66 (69%) reported having an important, nonparent adult who had taken a special interest in

them, who they could count on, and who inspired them (Collins et al., 2010). Having such a “mentor” (versus not) was associated with higher educational achievement and lower rates of homelessness. Using a longitudinal survey of youth transitioning out of the Missouri foster care system, Munson and McMillen (2009) studied how the presence of a natural mentor might have positively influenced older youth. About two-thirds of the 339 youth in the study reported having some type of mentor; however, only one-third reported having a long-term relationship. The presence of a mentor at age 18, along with the length of the relationship, was associated with a variety of favorable psychosocial outcomes. Having a mentor was associated with self-reported lower stress and higher life satisfaction; long-term mentoring relationships were also associated with fewer depressive symptoms and lower likelihood of being arrested at age 19. Although evidence indicates that the presence of a natural mentor is associated with a variety of positive outcomes for youth in foster care, until recently there has not been much research on how these relationships function or on which aspects are particularly effective. Several recent studies have attempted to address this gap, and most support the importance of one or more of the theories described in the previous section. For example, in a continuation of the previous study, researchers used qualitative interviews to determine the mentoring relationship qualities and forms of support offered that mattered most to these youth as they transitioned out of foster care (Munson, Smalling, Spencer, Scott, & Tracy, 2010). The study included 19 year olds (n = 189) who reported having a natural mentor in their final interview in the Munson and McMillen (2009) study. Certain qualities of the mentors themselves were mentioned in many of the interviews; mentors were reported to be approachable, easy to be with, and understanding. Additionally, some youth emphasized the similarities between themselves and their mentors (e.g., enjoying the same activities, being close in age) as factors that enhanced the relationship. In looking at qualities of the relationships, consistency and longevity were reported to be significant factors. Often, the mentoring relationship was one of the most stable and longest lasting in the youth’s life. Other important relationship qualities were authenticity, respect, trust, and empathy (the latter two of which are key qualities in the Rhodes model). Finally, support received from the mentors included advice, emotional support, tangible and instrumental support, and helping to hold the youth accountable.

Youth in Foster Care   347 These findings are consistent with those of another study addressing the quality and important aspects of mentoring relationships. Greeson and Bowen (2008) conducted interviews with seven female foster youth of color and focused on their relationships with natural mentors. The age range of this small sample was 13–20 (mean = 16.3 years), so this study included some younger foster children, in contrast to Munson et al.’s (2010) study of 19 year olds. Several themes emerged from the data. In general, the relationship characteristics reported as mattering most were trust, love and caring, and being like the relationship between parent and child, reinforcing the theoretical importance of attachment or acceptance (versus rejection). The youth also gave examples of the support they received from their mentor, including availability to talk, giving advice, sharing their point of view, and providing needed material items. Additionally, the youth discussed how their mentors helped them make positive changes, such as bettering academic performance or improving relationships with others in their lives. Ahrens and colleagues (2011), drawing on qualitative interviews with 23 former foster youth (aged 18–25), found that natural mentor relationships were varied in their nature, duration, and impact, but they were generally concordant with the Rhodes (2002) model. For example, themes that characterized the best relationships echoed other findings in this section (e.g., being caring, respectful, and authentic). Also in line with attachment and PAR theories, some young adults described having difficulties forming and/or maintaining a bond with nonparental adults and related these difficulties to prior (difficult) experiences with caregivers and/or other adults. Others described mentors who were able to overcome the youth’s initial rejection and/or fear of emotional risk. Some youth also reported being more open to mentoring during times of vulnerability, such as transitions out of foster care, suggesting that timing is an important factor in the initiation of mentoring relationships. Consistent with Keller’s contextual mentoring model (2005), Hirsch, Mickus, and Boerger (2002) noted difficulties in recruiting formal mentors and asserted that “[a]ttention to potential natural mentors in the youth’s social network, and among staff of youth development organizations” (p. 302) may prove more fruitful than trying to recruit formal mentors for a variety of youth at risk. Greeson et al. (2010) speculated that the gradual building of the natural mentoring relationship is less pressure filled than might be the case in a formal mentoring program and that relationships with natural mentors

may be more likely to be maintained over time because youth and mentor already share a social network. Ahrens et al. (2011) also suggested that professionals associated with the child welfare system may be particularly effective natural mentors for this population, in that they are likely to understand these youths’ unique challenges. Formal Mentors. Some mentoring programs are specifically tailored to working with foster youth, but there is little research on the effectiveness of these programs. An exception is a recent and rigorous evaluation of the Kempe Fostering Healthy Futures program (Taussig & Culhane, 2010). Researchers used a randomized controlled trial design to test the effectiveness of the intervention with 9- to 11-yearold children in foster care. A control group (n = 77) received only an assessment of cognitive, educational, and mental health functioning. The intervention group (n = 79) received the same assessment, a manualized skills curriculum that combined traditional cognitive-behavioral skills group activities with process-oriented material (led by licensed clinicians and graduate student trainees), and a 30-week mentoring program with graduate social work students. Immediately postintervention, the intervention group reported a better quality of life. At a 6-month follow-up, the intervention group—relative to the control group—had fewer mental health problems and symptoms of dissociation, and they were also less likely to have received mental health treatment (Taussig & Culhane, 2010). Other studies have evaluated formal mentoring programs for foster youth using less rigorous designs (i.e., not using a randomized controlled design). These studies have also suggested a positive impact. Advocates to Successful Transition to Independence is a program designed specifically to provide older foster youth with a mentor to help them transition successfully out of foster care; an evaluation of this program included surveys, individual interviews, and focus groups with youth who had been assigned a mentor through the program (Osterling & Hines, 2006). All youth felt there had been improvements in their lives since they began working with their mentors, particularly in relation to interpersonal skills. There were also increases in independent living skills; mentors helped with a variety of tasks such as “obtaining a job, opening a bank account, saving money, completing tax forms and completing their education” (p. 249). However, youth were more likely to report acquiring psychoemotional/ social skills than independent living skills, which, in light of the economic and housing-related problems

348  SPECIAL POPULATIONS these youth frequently possess during young adulthood (Courtney & Dworsky, 2006), suggests that a greater focus on teaching these skills may be needed to best support a successful transition. VISIONS is a community-based human development program for youth in foster care that focuses on building the skills needed for self-sufficiency (Uzoebo, Kioko, & Jones, 2008). The program offers a variety of resources ranging from money management training to information on how to access social support services to health and nutrition education; each youth is also matched with a mentor in the community. Participants reported skill improvement at follow-up, and the relationship with the mentor was a significant predictor of skills mastery (Uzoebo et al., 2008). A different intervention (Johnson, 2009) compared foster youth who received counseling and support through a System of Care program with foster youth who additionally received therapeutic mentoring. After 6 months, the youth who received mentoring had more favorable outcomes on measures of family and social functioning, and school behavior and achievement. Those who remained in mentoring relationships up to 18 months also demonstrated a reduction of the expression of stress symptoms associated with trauma (Johnson, 2009). Further research is needed for both models, but the findings are suggestive of the role mentors may play in complementing or enhancing other services for youth in foster care. In a cross-sectional study of the Connecticut Department of Children and Families One-on-One Mentoring program, Britner and Kraimer-Rickaby (2005) compared 143 foster youth, including waiting list control subjects, those youth currently involved in a formal mentoring relationship, and those for whom the formal mentoring relationship had ended within the first 6 months. Similar to findings in other populations (Grossman & Rhodes, 2002), youth with disrupted matches reported higher rates of delinquent externalizing behaviors, but not anxious/withdrawn internalizing behaviors, as assessed by Youth SelfReport subscales, than did youth in intact matches or the youth in the control group (waiting list). Some of the problems staff reported about matches that had ended included poor or inconsistent contact between mentor and protégé, and mentors who did not “feel a connection” to their youth, observations that are consistent with theory highlighting these aspects of relationships as barriers to positive outcomes for mentoring (Parra, DuBois, Neville, Pugh-Lilly, & Povinelli, 2002). Some of the strengths noted for intact matches included their more consistent/stable contact and that the mentors and youth reported enjoying each others’ company to a greater degree

(Britner & Kraimer-Rickaby, 2005), findings that are consistent with the literature on natural mentors. One factor that might increase the success of a mentor match is the mentor’s having similar experiences as youth in foster care. In a study of a campusbased learning program for transitioning foster youth, Kirk and Day (2011) reported the inclusion of a positive, peer-based mentoring model between the participating youth and the program counselors. Participating youth were aged 15–19 and were either currently in foster care or had previous involvement in the child welfare system. Undergraduate students served as counselors in the program; most were foster care alumni. Focus group discussions with youth suggested that of the many positive role models in the program, those with the greatest impact had been through the foster care system themselves. Consistent with attachment theory, the youth participants felt they could easily identify with, trust, and build relationships with those who had experiences similar to their own. Turning to larger comprehensive mentoring programs that serve foster youth (though they are not specifically designed to do so), we found some mentoring research that directly addressed maltreated youth and youth in foster care. Grossman and Rhodes (2002) examined data from 487 mentored youth and 472 control youth in the Public/Private Ventures impact study of urban Big Brothers Big Sisters programs in which youth were assessed at baseline and at 18-month follow-up. Within the mentored group, youth who had experienced emotional, sexual, or physical abuse were more likely than other youth (risk ratio of 1.53, p < .05) to have had their mentor relationship end prematurely. These findings are consistent with PAR and attachment theories suggesting that maltreated youth may be more susceptible to early termination of their mentoring relationships in formal programs and, relatedly, less likely to experience the same degree of benefits from such ties as other youth. In this sample, relationships that terminated in less than a year failed to show positive impacts evident for longer term relationships; youth in relationships that lasted less than 6 months showed deficits relative to matched controls (i.e., youth with no mentors; Grossman & Rhodes, 2002). Using the same dataset, Rhodes et al. (1999) studied 90 mentored foster youth (78 in relative care, 12 in nonrelative care) in comparison to (a) nonfoster youth with mentors and (b) foster youth without mentors. At the 18-month follow-up, mentored foster youth were more likely to demonstrate improved social skills and comfort/trust interacting with others based on parent reports, in comparison to (mentored) nonfoster youth. Relative to the second control

Youth in Foster Care   349 group, there was evidence of improvements in selfesteem and prosocial support from friends (e.g, “Would your friends agree to do you a favor if you asked?”), whereas declines on these outcomes were apparent for foster youth without mentors. This study is useful because of its longitudinal, experimental design. Findings are limited, however, by the small sample size, by the limited range of outcomes evaluated, and by a lack of information regarding the maltreatment and placement histories of foster youth. Thus, it was not possible to examine the potential role of such history as a moderator of outcomes in this population, as has been previously suggested in theory (Rhodes, 2002). Nonetheless, this research overall provides support for Rhodes’s model, suggesting that gains in social support received in other relationships and self-esteem may be important proximal outcomes of positive mentoring relationships for youth from backgrounds of maltreatment (DuBois, Neville, et al., 2002; Rhodes, 2002). Effect of Mentoring on Mentors. Although little research has examined the effect of mentoring relationships on the mentors of foster youth, an exception came out of the Fostering Healthy Futures program discussed earlier. The program was aimed at children aged 9–11 who were maltreated and placed in foster care; the two main intervention components were mentoring and a therapeutic skills group (Taussig & Culhane, 2010). Taussig et al. (2010) administered online surveys to the graduate students who had served as mentors to youth in foster care as part of an internship or a practicum. Almost all mentors reported the program to have provided helpful training in working within communities and in working with high-risk children and families, communities, multiple systems, other professionals, and diverse cultures. Several themes emerged in mentors’ responses to qualitative questions. These included feeling supported in their work, gains in clinical skills and confidence in working with families and systems, feeling that the work was powerful and meaningful, and learning a lot about themselves that furthered their professional development. This research suggests that mentors can also receive benefits from the relationship; further understanding of these benefits may have implications for recruiting and preparing mentors to work with youth in foster care.

attended the webinar, demonstrating the interest in and importance of this topic (MENTOR, 2011). However, the research base on mentoring outcomes and best practices with this population remains limited. With few exceptions, most of the research reviewed consisted of rich qualitative data from small, selected samples or limited investigations of mentoring within larger quantitative studies. Due to these design limitations, we caution readers that potential sample selection effects (i.e., who participated) and a reliance on interviews or self-reports of experiences and outcomes call for some caution in the interpretation of the findings. Future longitudinal research should take into account the heterogeneity among youth who have experienced foster placements by testing for differences across potential moderators such as type and severity of abuse or neglect (as applicable) as well as type, duration, number, chronicity, and quality of foster placements (all of which vary widely). In addition, such research should compare natural and formal mentoring. Because youth in foster care are represented disproportionately in other special populations (e.g., in this volume, see Sánchez et al., Chapter 10, on race and ethnicity; Chan & Henry, Chapter 21, on juvenile offenders; and Kerr & King, Chapter 22, on mental health needs), mentoring usually occurs in conjunction with other programs. Greater integration of services is needed to eliminate “support overload” or redundant or conflicting messages from multiple adult authority figures (e.g., social workers, caregivers, parents, foster parents, mentors). We must also study the overlapping protocols and samples within larger child welfare systems in order to disentangle risks, services, and outcomes related to mentoring youth in foster care (Britner, Balcazar, Blechman, Blinn-Pike, & Larose, 2006). Studies of individual differences among foster youth and those transitioning out of care (e.g., Keller, Cusick, & Courtney, 2007) would also add to the mentoring discourse.

Practice In this section, we describe program delivery models for mentoring foster youth, discuss some of the ways programs have been tailored, and conclude with some practice recommendations informed by theory and research.

Summary and Recommendations for Future Research

Program Delivery Models

A recent webinar was hosted by MENTOR: The National Mentoring Partnership on the topic of mentoring youth in foster care; over 500 practitioners

Most programs are one-on-one, state-sponsored programs, perhaps due to protective concerns and custody issues for the youth (Britner &

350  SPECIAL POPULATIONS Kraimer-Rickaby, 2005). Some formal programs targeting youth in foster care are beginning to employ elements of recently emerging program models such as e-mentoring. For example, the Orphan Foundation of America (OFA) was established by the National eMentoring Institute to meet the technical assistance and mentoring program development needs of agencies specializing in fostering transition and permanency for foster youth. E-mentoring may encourage interactions beyond those possible in face-to-face meetings (see Shpigelman, this volume, Chapter 17). Youth in state custody, in particular, may benefit from e-mentoring as a means to keep in contact with a mentor following a move across different placements or between placements and home. For older foster youth who are attending college or aging out of care, e-mentoring may allow their contact to continue when in-person activities are no longer convenient or as routine. Although mentoring programs designed specifically for foster youth exist in several U.S. states, few have published process or outcome evaluation results. One promising model is the Fostering Healthy Futures program, discussed earlier. This research-based prevention program for preadolescent children in foster care serves as an excellent example of a multiple component approach to working with this population. Foster children receive an intensive prevention curriculum aimed at reducing adolescent risk behaviors, attend a 30-week therapeutic skills group, and get frequent mentoring and advocacy from graduate students in social work (Taussig & Culhane, 2010). The program is comprehensive, intensive, and grounded in the use of current mentoring best practices; it is also more clinically oriented than many other models we have identified. Tailoring Mentoring Practices to the Population of Foster Youth What makes mentoring work? Elements of effective practice proposed by leading policy and advocacy organizations in the mentoring field, such as MENTOR (2009), as well as those suggested by research (e.g., DuBois, Holloway, et al., 2002), should be applicable in the design, implementation, and evaluation of mentoring programs for youth in foster care. However, some accommodations in the practices within formal mentoring programs may be necessary, as is suggested (but not greatly expanded upon) in the current version of the MENTOR (2009) practice guidelines. Given the potentially adverse consequences of yet another failed relationship and greater susceptibility to

this occurring when vulnerable youth are the focus of mentoring efforts, it may be particularly important that mentoring programs that serve these youth have a solid infrastructure that fosters the development of enduring, effective mentoring relationships and prepares mentors (through additional and specialized training) to understand and address some of the experiences, concerns, and issues that are more likely to arise when mentoring foster youth (Britner & Kraimer-Rickaby, 2005). Programs must be selective and careful in this process. Several mentoring organizations and government umbrella organizations provide links to resources and recommendations for how mentoring programs should tailor training and support services when running mentoring programs for foster youth. The National Mentoring Center at Education Northwest (Portland, Oregon) has a very helpful “Specialized Mentoring Models” page (National Mentoring Center, n.d.), including links to its own materials, like the Senior Corps’ “Mentoring Children in Foster Care” handbook, and to other sites of relevance to mentoring youth in foster care, like the New York City Administration for Children’s Services’ “Best Practice Guidelines for Foster Care Youth Mentoring” site. Whereas the recommendations on these vetted sites appear reasonable, they tend not to have references to any rigorous research or evaluations of such tailored mentoring practices for the population of youth in foster care. Practice Considerations for Mentoring Youth in Foster Care Table 23.1 presents recommendations that we believe can facilitate the application of the theory and research we have reviewed to the realm of practice. These considerations fall into three categories: training, informed practice, and the utilization of natural mentors. Training. How can your mentoring program learn from (and possibly even coordinate with) foster parent training programs? Dozier et al. (2009), in a randomized clinical trial, reported that their foster parent training program was effective in helping caregivers provide nurturance to youth in foster care—even when such nurturance did not come naturally to the foster parents, and/or the youth did not elicit such nurturance. Mentors could benefit from such training, in terms of learning how to build relationships with foster youth but also in understanding how to manage relationships with biological and foster parents. During recruitment, screening, training, and periods of ongoing support, programs should provide

Youth in Foster Care   351 Table 23.1   Checklist for Practitioners Topic

Recommendations

Training

Provide mentors with training on how to build relationships with youth in foster care as well as with their biological and foster parents. Provide mentors with realistic expectations and information on the difficulties (e.g., initial distrust of others) and benefits (e.g., potential for significant impact) of working with foster youth during recruitment, screening, training, and ongoing support. Provide mentors with comprehensive training and detailed information about the case histories of youth in foster care, keeping in mind that program staff, mentors, and foster parents all request more—not less—training and information to prepare themselves to help youth in foster care succeed. Consider the following topics when planning trainings for mentors who may work with youth in foster care: trust issues, life skills, anger management, boundaries, and relevant theories (i.e., Rhodes’s model, self-esteem, attachment and parental acceptance-rejection theories, social exchange theory, Keller’s contextual model).

Informed practice

Provide youth in foster care with information on their child welfare case status to the degree that it is appropriate for their developmental level. Utilize assessments grounded in attachment theory to help youth in foster care better understand how their attachment styles or family histories might influence their relationships with mentors and other important persons in their lives. Minimize transitions and disruptions of mentoring relationships for foster youth, such as by prioritizing matching with mentors likely to be able to sustain longer-term relationships, because these youth are already in transition and vulnerable to rejection, and take into account any anticipated or potential transitions in plans for mentoring services. Carefully screen and train mentors before the match is made to try and ensure successful and stable matches for youth in foster care. Provide ongoing support to mentors of youth in foster care after the match is made to reduce avoidable terminations, because these youth may be particularly susceptible to harmful effects of relationship disruptions. Help mentors of youth in foster care understand the significance of their potential impact on youth to help them persist through difficult times during their relationships with them. When termination of a mentoring relationship is unavoidable for youth in foster care, develop a termination ritual to aid the youth and mentor in the transition.

Natural mentors

Mobilize and incorporate natural mentors more systematically into services for youth in foster care (e.g., care coordination and transition planning), particularly for older youth for whom formal, programmatic relationships may be potentially less effective or difficult to establish.

mentors with realistic expectations and ideas about working with youth in foster care. Program staff, mentors, and foster parents all routinely request more training and information about youths’ case histories in order to prepare themselves to help the youth succeed (Redding, Fried, & Britner, 2000). Training topics might include trust issues, life skills, anger management or emotion regulation, crisis intervention, boundaries, cultural awareness, and the theories reviewed in this chapter.

Informed Practice. How can you engage youth and mentors in the development of their relationship and their understanding of the youth’s involvement with the child welfare system? Assessments grounded in attachment theory have been used in other mentoring evaluations (e.g., Westhues, Clarke, Watton, & St. Claire-Smith, 2001) and should be incorporated with attachment-based discussions and modules into mentoring programs for youth in foster care. Further, youth should be provided with as much

352  SPECIAL POPULATIONS information about their child welfare case status as is appropriate for their developmental level and understanding, so that they are aware of the decisions that affect them. What can you do to minimize youths’ feelings of rejection? Some unavoidable systemic issues may cause a match to terminate (e.g., youth moving from placement, youth voluntarily leaving the system, caseworker turnover). However, programs should be careful to minimize transitions and disruptions of relationships for youth in foster care, who are already in transition and vulnerable to rejection (Spencer et al., 2010). Careful screening and training before the match, followed by ongoing support to persist through difficult times or the mentor’s lack of understanding of his or her impact on the youth (Ahrens et al., 2011; Britner & Kraimer-Rickaby, 2005), may reduce avoidable terminations. When termination is required, follow a procedure for a termination ritual (see Lakes & Karcher, 2005) so as to help the youth and mentor with the transition. Natural Mentors. How can you identify, involve, and support each youth’s natural mentors? Child welfare services need to identify and incorporate natural mentors more systematically into services, especially for older youth in foster care, for whom formal, programmatic relationships may be less likely to have the intended effects (Hirsch et al., 2002; Munson & McMillen, 2009).

Conclusion Theories from both within and outside the mentoring literature suggest that youth in foster care have the potential to benefit from mentoring relationships and programs. These theories are equally clear in highlighting a range of potential barriers to effective mentoring for this population, as well the potential for unintended harmful effects when relationships fail and programs are not adequately designed to address their needs. Heeding the 5th century bce charge of Hippocrates, we should be certain that we are doing no harm as we seek to do good in the lives of these vulnerable youth. A growing literature finds that youth in foster care who have natural mentors are better off than those who do not. Of course, youth with the social skills and personal characteristics to attract natural mentors may be different from those who lack these skills. Fewer studies have followed youth in programs with formal, assigned mentors, but the outcomes have generally been promising, provided

that mentors can be recruited and retained. Efforts to help youth identify and cultivate informal relationships, like the recruitment, training, and support of mentors in formal programs for foster youth, require further development and research. In both natural and formal mentoring, meaningful and caring relationships that persist with some regularity over time are usually viewed positively by youth and are associated with favorable social, behavioral, health, educational, and vocational outcomes. Mentoring programs and practices tailored for youth in foster care have developed only over the past decade, so research on these practices and their outcomes is limited (but is emerging quickly).

References Ahrens, K. R., DuBois, D. L., Richardson, L. P., Fan, M. Y., & Lozano, P. (2008). Youth in foster care with adult mentors during adolescence have improved adult outcomes. Pediatrics, 121(2), e246–e252. Ahrens, K. R., Garrison, M., Spencer, R., DuBois, D. L., Lozano, P., & Richardson, L. P. (2011). Qualitative exploration of the adult mentoring relationships in the lives of former youth in foster care. Children and Youth Services Review, 33, 1012–1023. Alpert, L. T., & Britner, P. A. (2009). Measuring parent engagement in foster care. Social Work Research, 33, 135–145. Bocknek, E. B. L., Sanderson, J., & Britner, P. A. (2009). Ambiguous loss and posttraumatic stress in schoolage children of prisoners. Journal of Child & Family Studies, 18, 323–333. Bowlby, J. (1969). Attachment and loss. Vol. I: Attachment. New York: Basic Books. Britner, P. A., Balcazar, F. E., Blechman, E. A., Blinn-Pike, L., & Larose, S. (2006). Mentoring special youth populations. Journal of Community Psychology, 34, 747–763. Britner, P. A., & Kraimer-Rickaby, L. (2005). Abused and neglected youth. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 482–492). Thousand Oaks, CA: Sage. Collins, M. E., Spencer, R., & Ward, R. (2010). Supporting youth in the transition from foster care: Formal and informal connections. Child Welfare, 89, 125–143. Courtney, M. E., & Dworsky, A. (2006). Early outcomes for young adults transitioning from out-of-home care in the USA. Child and Family Social Work, 11, 208–219. Diehl, D. C., Howse, R. B., & Trivette, C. M. (2011). Youth in foster care: Developmental assets and attitudes towards adoption and mentoring. Child & Family Social Work, 16, 81–92. Dozier, M., Lindhiem, O., Lewis, E., Bick, J., Bernard, K., & Peloso, E. (2009). Effects of a foster parent training

Youth in Foster Care   353 program on young children’s attachment behaviors: Preliminary evidence from a randomized control trial. Child & Adolescent Social Work Journal, 26, 321–332. Dozier, M., & Rutter, M. (2008). Challenges to the development of attachment relationships faced by young children in foster and adoptive care. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 698–717). New York: Guilford Press. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. New Directions for Youth Development, 93, 21–57. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Dworsky, A., & Courtney, M. E. (2009). Homelessness and the transition from foster care to adulthood. Child Welfare, 88, 23–56. Farruggia, S. P., Greenberger, E., Chen, C., & Heckhausen, J. (2006). Perceived social environment and adolescents’ well-being and adjustment: Comparing a foster care sample with a matched sample. Journal of Youth and Adolescence, 35, 349–358. Garland, A. F., Hough, R. L., McCabe, K. M., Yeh, M., Wood, P. A., & Aarons, G. A. (2001). Prevalence of psychiatric disorders in youths across five sectors of care. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 409–418. Geenen, S., & Powers, L. E. (2007). “Tomorrow is another problem”: The experiences of youth in foster care during their transition to adulthood. Children and Youth Services Review, 29, 1085–1011. Greeson, J. K. P., & Bowen, N. K. (2008). “She holds my hand”: The experiences of foster youth with their natural mentors. Children and Youth Services Review, 30, 1178–1188. Greeson, J. K. P., Usher, L., & Grinstein-Weiss, M. (2010). One adult who is crazy about you: Can natural mentoring relationships increase assets among young adults with and without foster care experience? Children and Youth Services Review, 32, 565–577. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Hafton, N., English, A., Allen, M., & DeWoody, M. (1994). National health care reform, Medicaid, and children in foster care. Child Welfare, 73, 99–115. Hass, M., & Graydon, K. (2009). Sources of resiliency among successful foster youth. Children and Youth Services Review, 31, 457–463. Hirsch, B. J., Mickus, M., & Boerger, R. (2002). Ties to influential adults among Black and White adolescents:

Culture, social class, and family networks. American Journal of Community Psychology, 30, 289–303. Johnson, S. B. (2009). Therapeutic mentoring: Outcomes for youth in foster care. Doctoral dissertation. Available from ProQuest Dissertations and Theses database (UMI No. 3387412). Keller, T. E. (2005). A systemic model of the youth mentoring intervention. Journal of Primary Prevention, 26, 169–188. Keller, T. E., Cusick, G. R., & Courtney, M. E. (2007). Approaching the transition to adulthood: Distinctive profiles of adolescents aging out of the child welfare system. Social Service Review, 81, 453–484. Kirk, R., & Day, A. (2011). Increasing college access for youth aging out of foster care: Evaluation of a summer camp program for foster youth transitioning from high school to college. Children and Youth Service Review, 33, 1173–1180. Lakes, K., & Karcher, M. J. (2005). Mentor/mentee termination ritual. In How to build a successful mentoring program using the Elements of Effective Practice: A stepby-step tool kit for program managers (pp. 157–158). Alexandria, VA: MENTOR. Retrieved from http:// www.mentoring.org/downloads/mentoring_413.pdf MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. MENTOR. (2011). Mentoring youth in foster care: Emerging research and lessons from the field [Webinar]. Retrieved from http://www.mentoring.org/ downloads/mentoring_1338.pdf Munson, M. R., & McMillen, J. C. (2009). Natural mentoring and psychosocial outcomes among older youth transitioning from foster care. Children and Youth Services Review, 31, 104–111. Munson, M. R., Smalling, S. E., Spencer, R., Scott, Jr., L. D., & Tracy, E. M. (2010). A steady presence in the midst of change: Non-kin natural mentors in the lives of older youth exiting foster care. Children and Youth Services Review, 32, 527–535. National Mentoring Center. (n.d.). Specialized mentoring models. Portland, OR: Education Northwest. Retrieved from http://educationnorthwest.org/resource/652 Osterling, K. L., & Hines, A. M. (2006). Mentoring adolescent foster youth: Promoting resilience during developmental transitions. Child and Family Social Work, 11, 242–253. Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. Redding, R. E., Fried, C., & Britner, P. A. (2000). Predictors of placement outcomes in treatment foster care: Implications for foster parent selection and service delivery. Journal of Child and Family Studies, 9, 425–447. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J., Haight, W. L., & Briggs, E. C. (1999). The influence of mentoring on the peer relationships of

354  SPECIAL POPULATIONS foster youth in relative and nonrelative care. Journal of Research on Adolescence, 9, 185–201. Rohner, R. P. (1986). The warmth dimension: Foundations of parental acceptance-rejection theory. Newbury Park, CA: Sage. Rohner, R. P., & Britner, P. A. (2002). Worldwide mental health correlates of parental acceptance-rejection: Review of cross-cultural and intracultural evidence. Cross-Cultural Research: The Journal of Compara­ tive Social Science, 36, 16–47. Sabatelli, R. M., & Shehan, C. L. (1993). Exchange and resource theories. In P. G. Boss, W. J. Doherty, R. LaRossa, W. R. Schumm, & S. K. Steinmetz (Eds.), Sourcebook of family theories and methods: A contextual approach (pp. 385–411). New York: Plenum Press. Spencer, R., Collins, M. E., Ward, R., & Smashnaya, S. (2010). Mentoring for youth leaving foster care: Promise and pitfalls. Social Work, 55, 225–234. Taussig, H. N., & Culhane, S. E. (2010). Impact of a mentoring and skills group program on mental health outcomes for maltreated children in foster care. Archives of Pediatrics and Adolescent Medicine, 164, 739–746. Taussig, H. N., Culhane, S. E., Raviv, T., Schnoll Fitzpatrick, L. E., & Wertheimer Hodas, R. (2010). Mentoring children in foster care: Impact on graduate student mentors. Educational Horizons, 89, 17–32.

Thompson, R. A. (2008). Early attachment and later development: Familiar questions, new answers. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 348–365). New York: Guilford Press. U.S. Department of Health and Human Services. (2009). The AFCARS report: Preliminary estimates as of October 2009. Washington, DC: Author. Retrieved from http://www.acf.hhs.gov/programs/cb/stats_ research/afcars/tar/report16.htm U.S. Department of Health and Human Services. (2010). Child maltreatment 2008. Washington, DC: Author. Retrieved from http://www.acf.hhs.gov/programs/ cb/pubs/cm08/cm08.pdf Uzoebo, V., Kioko, M., & Jones, R. (2008). Deconstructing youth transition to adulthood services: Lessons learned from the VISIONS program. Vulnerable Children and Youth Studies, 3, 37–41. Westhues, A., Clarke, L., Watton, J., & St. Claire-Smith, S. (2001). Building positive relationships: An evaluation of process and outcomes in a Big Sister program. Journal of Primary Prevention, 21, 477–493. Yancey, A. K. (1998). Building positive self-image in adolescents in foster care: The use of role models in an interactive group approach. Adolescence, 33, 253–267.

24  IMMIGRANT AND REFUGEE YOUTH Dina Birman and Lyn Morland

Introduction In this chapter, we summarize the theories, research, and practices relevant to creating and providing mentoring programs for immigrant and refugee youth. Karcher, Kuperminc, Portwood, Sipe, and Taylor (2006) discuss three aspects of mentoring programs: context (site based vs. field based), goals (instrumental and developmental), and multiple structures that mentoring programs can take (see also Karcher & Nakkula, 2010). Guided by this framework, we draw out implications for mentoring from the theoretical, research, and practice literature. We do this by presenting a model that relies on youths’ degree of acculturation and the availability of culturally similar community members to determine specific recommendations for mentoring immigrant and refugee youth. To provide a context for this discussion, we first offer an overview of demographic and background characteristics of current immigrants and refugees. The United States is one of the largest immigrantreceiving countries, with over 1 million immigrants obtaining legal permanent status each year (U.S. Department of Homeland Security, 2009a). Tremen­ dous diversity exists among the arriving immigrants with respect to their demographic characteristics and migration backgrounds. The majority are economic migrants who come from a variety of countries for better work and an improved standard of living. Most undocumented migrants arrive through the southern U.S. border, originating from Mexico and Central America. Refugees are a relatively small, but particularly vulnerable group fleeing war and persecution. Approximately 75,000 refugees were resettled in the United States in 2009 (U.S. Department of Homeland Security, 2009b). Most

recently, the U.S. refugee admissions policies have emphasized warehoused populations that have been living for extended periods of time in refugee camps with no foreseeable hope of repatriation or resettlement (Negash, 2010). Immigrant groups are tremendously diverse with regard to race, ethnicity, socioeconomic status, and other background variables and therefore vary considerably in the degree of their acculturation. In addition to cultural and socioeconomic differences, the migration histories—including the reasons for leaving and the length and characteristics of their particular journeys—are important to understand. For example, refugees have frequently endured forced uprooting, and are less likely to have extended family or coethnic communities for support in this country. While some groups come from educated backgrounds, others have experienced disruptions in education and some have not been exposed to literacy. Data on the adjustment of immigrants in resettlement are somewhat contradictory. On one hand, research suggests that immigrant youth are at higher risk for psychological and behavioral problems including anxiety disorders, depression, posttraumatic stress disorder (PTSD), substance abuse, and conduct and eating disorders (Pumariega, Rothe, & Pumariega, 2005). Refugees, in particular, are at risk for PTSD and related disorders, having been exposed to traumatic events while fleeing war-torn areas, in refugee camps, and during the migration process (Birman, Ho, et al., 2005). Undocumented immigrants arriving from Mexico also report experiencing high rates of violence (Catholic Relief Services, 2010). After arrival in the United States, many immigrants reside in high-poverty areas, where they are vulnerable to crime, and experience difficulties 355

356  SPECIAL POPULATIONS adjusting to underfunded schools (Jaycox et al., 2002). As a result, many immigrant children are thought to have serious mental health and socialemotional adjustment needs, not all of which can be addressed by mentors and may require multiple additional support services. On the other hand, research has documented the “immigrant paradox,” which suggests that immigrants are more resilient than their U.S.-born counterparts. Studies have found that immigrants are healthier (Harris, 1999) and perform better in school (Fuligni, 1997) than their U.S.-born peers. Some observers have attributed this advantage to “immigrant optimism” (Portes & MacLeod, 1996) that motivates immigrants to make meaning out of the difficulties of migration. Suarez-Orozco, Rhodes, and Milburn (2009) suggested that parents strike an implicit “immigrant bargain” with their children, justifying their own sacrifices with the reward of their children succeeding in school. For these families, academic success is an important goal. Mentoring programs can engage immigrant families in services by prioritizing this goal in their programming. Although both risk and protective factors may be operating, immigrant groups vary tremendously in the resources they can draw on to aid in their adjustment. Ironically, while public health concerns generally focus on larger groups such as Latinos, the large numbers of immigrants from Latin America in the United States create large ethnic communities with extensive resources sensitive to cultural concerns and members who are competent in the language. By contrast, many very small refugee communities do not have access to existing ethnic resources in the United States, making it nearly impossible to find teachers, service providers, and community leaders who are sensitive to their cultural backgrounds and competent in their native language. Thus, the challenges of adjustment for these diverse groups of immigrants are varied, and different kinds of mentoring programs may be most helpful for specific groups of immigrants, depending on their circumstances. In the remainder of this chapter, we consider the implications of theory, research, and practice for mentoring immigrant and refugee youth. In the next section, we explain why ecological theory provides a useful perspective on the adjustment of immigrant youth, as well as how acculturation theory, which explains a major process and stressor confronting newly arriving immigrants, may be the key to providing more fine-tuned youth mentoring programming. Next, we review published research and what is known about the practice of mentoring with these

populations. Throughout, we use our conceptual model (see Figure 24.1) to consider the diversity of the immigrant and refugee experience in terms of background experiences and acculturation status to provide specific recommendations for the goals, context, and structure (Karcher et al., 2006) of mentoring programs for immigrants and refugees.

Theory Overview Our review of the mentoring literature suggests that attention to diversity has been focused on the experience of ethnic minority youth. In fact, scholars of mentoring emphasize its usefulness for “highrisk” populations, often referring wholly or in part to youth from minority backgrounds (Jucovy, 2002). Some scholars consider mentors as “parent surrogates” able to step in when a child’s own parent is unable to provide the needed support, supervision, and guidance (Ainsworth, 1989, cited in Rhodes, Bogat, Roffman, Edelman, & Galasso, 2002). Such a focus can inadvertently promote a deficit view of the youth and their families as in need of “fixing” rather than addressing the youths’ needs within the makeup of their larger social and cultural ecology (Rhodes et al., 2002). Further, many mentoring programs have difficulty identifying mentors who can be racially matched with minority youth, particularly boys (Jucovy, 2002). Prior studies have not consistently found race matching to contribute to outcomes, and recent evidence may indicate the contrary (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). In general the literature continues to stress that ethnic minority mentors may be best able to serve as role models and provide strengthbased support to ethnic minority youth (Jucovy, 2002). We provide this prelude to the research on cultural similarity in youth mentoring only to suggest that the question may not be whether or not to emphasize cultural similarity in matching, but rather for whom may similarity be helpful? Newly arriving immigrants and refugees have needs and concerns that are distinct from those of members of U.S.-born minority groups, and these differences have implications for mentoring. Not all immigrants and refugees are ethnic minorities. Some, such as Jews from the former Soviet Union or Bosnians from the former Yugoslavia faced ethnic persecution and discrimination in their countries of origin, but they are visibly similar to U.S.-born Whites. In contrast, many immigrants from Africa, Asia, and Latin America are racialized in the United

357

Migration Background - Reasons for migration - Immigration status - Trauma endured - Missed schooling - Years in refugee camp - Loss of key family members - Age at arrival

Resettlement Context – Size of coethnic community – Cultural resources

Demographic Variables – Race/ethnicity – Socioeconomic status – Formal education, literacy – English proficiency – Family composition

LOW Acculturation to Native Culture

HIGH Acculturation to U.S. Culture

HIGH Acculturation to Native Culture

LOW Acculturation to U.S. Culture

YOUTH BACKGROUND DEGREE OF FACTORS ACCULTURATION

Mentor from same/similar culture and language background

Mentor from majority culture or immigrant with longer length of residence in U.S.

Cultural Competency Training

MATCH

STRUCTURE: Family focused to support parents and reduce acculturation gap

CONTEXT: Field based to support development beyond school and to expose youth to U.S.

GOAL: (Developmental) Support to promote ethnic identity and biculturalism

STRUCTURE: Cross-age, developing mentees into mentors

CONTEXT: School based to assist with initial school adjustment

GOAL: (Instrumental) Support with new language, culture, school adjustment

MENTORING GOAL, CONTEXT, STRUCTURE

Figure 24.1   Conceptual Model of Application of Acculturation Theory to Mentoring Refugee and Immigrant Youth

YOUTH POSITIVE INTEGRATION

ULTIMATE GOAL

358  SPECIAL POPULATIONS States and become racial minorities. Thus, some of the assumptions of the mentoring literature may not be applicable for immigrants who are not visible minorities (Ogbu, 1990). For immigrants, the central aspect of their adjustment process involves acculturation, which in our view is best viewed from an ecological perspective.

A formalized mentoring program—one that is available to all youth but that targets immigrant youth or one that gives immigrant youth opportunities to receive and provide mentorship—can help to shift the ecology of the school toward a climate more supportive of immigrant youth. Acculturation Theory

Ecological Theory An ecological perspective on acculturation and adaptation of immigrants and refugees suggests that these processes unfold in interaction with various aspects of the communities and institutions of resettlement (Birman, Trickett, & Buchanan, 2005). Ecological perspectives on human behavior emphasize the impact of multiple layers of the surrounding context on individuals, and suggest that behavior cannot be understood independent of that context (Bronfenbrenner, 1977; Trickett, 1996). The settings of greatest importance to youth are the family and the school (Bronfenbrenner, 1977). An ecological perspective stresses that the adjustment of children can be understood only in the context of the family and the surrounding community. As they struggle with the new language and culture, parents may experience difficulties acquiring employment, providing for the family, and understanding relevant social, medical, and other necessary service systems. Parents may turn to their children for help with language and culture brokering, a practice that has been linked to family conflict (e.g., Jones, Trickett, & Birman, 2012). Parents may also know little about the lives of their children at school, not see it as their role to interfere in what they see as the school’s responsibility, and/or doubt their parenting skills in the new culture (Birman & Ryerson-Espino, 2007). Further, extended networks of friends and family that parents may have relied on for assistance are disrupted through migration. Thus, mentoring programs for immigrant youth may need to help support parents and family communication and adjustment. School is the setting in which immigrant youths’ acculturative struggles unfold, as it is where they first encounter the new culture and language, and form relationships with peers (Birman, Weinstein, Beehler, & Chan, 2007). From an ecological perspective, interventions that are most effective target changes in the school culture and climate rather than address individual concerns alone (Trickett & Birman, 1989). Schools can ease the process of acculturation and adjustment for newly arriving students through specialized programming that makes the school environment more hospitable (Trickett & Birman, 1989).

Theories of acculturation consider the ways in which immigrants and refugees undergo cultural change as a result of interacting with the new society. This process includes the extent to which they are connected to the new culture, as well as the native culture. Therefore, understanding immigrant youths’ levels of acculturation to the United States as well as to their own culture is essential to effectively using acculturation theory to guide program planning. Phases and Dimensions of Acculturation. Accultur­ ation occurs along different dimensions that become salient during different phases of the acculturation and resettlement process. As new immigrants first come into contact with the new culture, they begin to learn the language and how to behave according to its norms (Gordon, 1964). After the initial adjustment period, immigrants may begin to consider the strength of their relative attachment to the new and the native cultures and engage in cultural identity exploration. Thus, the more instrumental aspects of acculturation (language and behavioral) have been found to occur relatively quickly, while identity exploration continues to unfold for many years to come (Birman & Trickett, 2001). Differentiating the instrumental and developmental acculturation processes is important. The process of acculturation also differs as a function of age at arrival. Because their cognitive abilities are still undergoing profound changes, school-aged children quickly acquire the new language. However, while they use English at school to learn complex and abstract concepts, at home they are more likely to discuss household events and less complex topics in their native language. Thus, native language proficiency will not develop in immigrant children automatically, but requires special efforts to promote and maintain it. For some youth, providing support for native language fluency may be one of the most important roles a mentor can play. Discovering one’s identity as one separates from parents is an important expectation for American adolescents. For immigrant adolescents, this process involves not only distancing themselves

Immigrant and Refugee Youth   359 from their parents—a practice that may be foreign in many of the more collectivistic countries from which immigrants come—but also, in this process, distancing themselves from their native culture, as it is transmitted to them by the parents. Yet, if immigrant adolescents don’t feel accepted by their U.S. peers, they may embrace their native ethnic identity in a process of “reactive identification” (Portes & Zhou, 1993). Thus, the identity exploration that is characteristic of adolescence is particularly complex for immigrants, and they can benefit from role models and guidance from mentors to help them navigate this complex cultural process. By contrast, although many adult immigrants may be successful in learning the new language, their native language is likely to remain primary (Birman & Trickett, 2001). Behaviorally, adults are also more likely than youth to remain attached to their ethnic community, family, and friends. Issues of cultural identity, so salient for youth, are likely to be less critical for adults, resulting in differences in processes of acculturation for children and adults.

culture of origin has been particularly important for immigrant children in maintaining positive relationships with family (Birman, 2006; Ho & Birman, 2010) and coethnic friends (Birman, Trickett, & Buchanan, 2005). Because social networks are disrupted by migration, immigrant youth can benefit from mentoring that helps them maintain cultural ties and links them to sources of support in both the host culture and the ethnic community. Implications of Acculturation and Ecological Theories for Mentoring Ecological and acculturation theories have many implications for goals, contexts, and structures of mentoring programs, as outlined by Karcher et al. (2006). We explain these implications through a framework we describe further in this section (see Figure 24.1).

Family Acculturation Gaps. As a result of such generational differences in acculturation, gaps are thought to develop in families, with parents being more attached to the native culture than their children and children being more acculturated to the American culture than their parents (e.g., Birman, 2006). These gaps have been shown to contribute to conflict between immigrant children and parents (Birman, 2006), as children are faced with navigating different cultural worlds at home and at school. Programs have been designed to intervene to reduce such gaps (Szapocznik, Rio, Perez-Vidal, Kurtines, & Santisteban, 1986) when they occur, usually in later stages of the resettlement process. Mentoring programs need to be mindful of such gaps, and ensure that they help support the family and not exacerbate them.

Mentoring Goals. As suggested by acculturation theories, recently arrived immigrants and refugees often struggle with behavioral and linguistic acculturation and with learning the norms of the school and the surrounding society. For them, instrumental mentoring can provide tangible assistance and help guide the youth through the initial adjustment process. By a year or two after resettlement, schoolchildren have mastered conversational English and behavioral aspects of the culture, but issues of identity and belonging become more salient. Over time, the acculturation gap between children and parents widens. For these reasons, nurturing the connection with their native culture and ethnic community can become more important as resettlement proceeds. Developmental mentoring provided by older immigrant role models who may be farther along in their acculturation journey may be particularly helpful as youth struggle to hold on to their ethnicity and develop a positive ethnic identity.

Bicultural Adjustment. Current acculturation theories stress the importance of bicultural adjustment, as immigrant children are seen to benefit from adopting the new culture while remaining attached to their native culture. Acculturation to both the native and the new culture is seen as potentially beneficial for immigrants, in part because it provides access to different sources of social support (e.g., Oppedal, Roysamb, & Sam, 2004). Acquiring an understanding of the new culture helps immigrant youth succeed at school because knowing the language, the rules, and the norms is critical for academic success and provides access to social support from U.S.-born peers. At the same time, retaining one’s

Mentoring Context. Site-based programs situated in schools—addressing immigrant students’ struggles with language, academics, and understanding peer norms and pressures—may be particularly helpful for newer arrivals. School-based programs also allow interventions with the whole ecology of the classroom or school, as advocated by ecological theory (Trickett & Birman, 1989). As noted earlier, by focusing on individual-level variables such as mentor/mentee characteristics, mentoring programs can ignore the surrounding context that may be contributing to the difficulties experienced by the youth. School-based mentoring programs can be designed in ways that not only offer assistance to individual

360  SPECIAL POPULATIONS students, but also intervene with the school ecology to help integrate the immigrant students into the life of the school (Birman et al., 2007). When the goals of mentoring shift to developmental issues, fieldbased programs may help immigrant youth get to know adult role models in the contexts of their own lives to give them a sense of possibilities for how lives can develop beyond school.

can allow the mentor to assist the family across a range of settings such as school, community, and parents’ work. The family focus can help the parents retain the connection with their child and discover new parenting strategies compatible with life in the new culture, rather than be replaced with an alternative parental figure.

Mentoring Structures. Many different mentoring structures, described by Karcher et al. (2006), may be useful for immigrant youth, including individual, group, cross-age, or other forms of mentoring. In school-based programs, cross-age mentoring may be particularly beneficial for immigrants. Programs that involve slightly older youth as peer mentors can help immigrants feel connected to others. These peers may be U.S. born or immigrants who are farther along in their acculturation process. Being known on a personal level to school peers can help immigrant youth to develop a sense of school belonging, which has been noted as an important factor linked to positive psychological adjustment for immigrant and refugee youth (Kia-Keating & Ellis, 2007). Further, in such cross-age peer mentoring programs, immigrant students can “graduate” to become mentors to newer arrivals, thus increasing the availability of culturally based communities of support in the school. Because new immigrants may experience a low sense of efficacy as they struggle in the new language and culture, giving them opportunities to be experts and helpful to someone else can be particularly empowering. As pointed out by Rhodes et al. (2002), mentoring can be particularly beneficial for the mentors, and immigrant youth can benefit from participating in the larger school community by playing such leadership roles. Field-based programs oriented toward “mentoring as a family strengthening strategy” (Family Strengthening Policy Center, 2004; see also Taylor & Porcellini, this volume, Chapter 31) can be effective given the challenges created by acculturation gaps. Unlike some traditional mentoring programs in which the emphasis is on introducing another adult, perhaps a father figure, into the life of the child, this approach emphasizes working with the whole family to help them bridge the cultural gap with the outside world and within the family itself. In this type of program, the mentor does not step in as a parental figure but serves more as a cultural ambassador (Sánchez & Colón, 2005). The mentor helps provide instrumental support to the family and to the parents in their parenting role and may also provide developmental guidance to the youth as well. The field-based context of such a program

Research A review of the mentoring literature revealed few studies that focused on immigrant and refugee youth, and none examined the link between the stage of resettlement or style of acculturation and mentoring program impacts. This literature does include several studies that (a) examined important sources of informal and formal support and guidance with respect to academic success, and (b) described and evaluated formal mentoring programs for immigrant youth. Most studies focused on Latinos as a group without differentiating first-generation immigrants from those born in the United States or reporting on length of residence in the United States. Thus, these studies cannot provide empirical evidence of whether and how acculturation may be related to mentoring. However, descriptions and limited evaluations of programs provide some suggestions for how to structure mentoring programs for these populations. Studies of Sources of Support Around Academic Success Several studies explored the question of whether and how mentoring may be helpful for academic success of Latino youth, and most reveal the importance of family member mentors. Though many studies suggested the importance of both family members and nonfamily adults in providing support and guidance and serving as role models for students, among several studies of Latino high school (e.g., Sánchez, Esparza, & Colón, 2008) and college (e.g., Sánchez, Reyes, & Singh, 2006) students, most identified family members as mentors who helped them succeed academically. Their mentors included parents, older siblings, aunts and uncles, and other members of the extended family. Notably nonfamily adults identified by youth as having provided significant mentorship were of the same race and ethnicity (Sánchez et al., 2008), and were better educated than family mentors, lending support to the importance of nonfamily mentors in providing instrumental support that helped them succeed in school. Similarly, Cooper, Jackson, Azmitia, Lopez, and Dunbar (1995) interviewed minority students participating in a variety of college

Immigrant and Refugee Youth   361 “bridge” programs for African American and Latino students. Although formal mentoring relationships with sibling-like peers and adults were a component of all of the programs considered, students consistently mentioned the importance of informal mentorship provided by family members, particularly siblings, in helping them traverse the cultural divide between the worlds of home and school. Other studies highlighted the importance of mentorship of nonfamily adults such as teachers, counselors, program directors, as well as mentors in formal mentoring programs. For example, Ceballo (2004) reported on an interview study of 10 successful Latino undergraduates at an Ivy League university, all of whom noted the importance of mentorship provided by both their parents and teachers. In a sample of academically successful Mexican American undergraduates, all reported having had a nonfamily adult mentor or role model who had taken a particular interest in them and helped them (Arellano & Padilla, 1996). Santon-Salazar and Urso Spina (2003) also reported on a qualitative study of Mexican-origin high school students in San Diego who identified the importance of informal nonfamily mentors in the community such as a friend, coordinator of a youth group, and others. These mentors provided emotional support as well as functioned as “institutional agents,” defined as transmitters of institutional knowledge and resources about how educational institutions worked. Ceballo (2004) also highlighted the importance of nonfamily role models who may have a less involved relationship but serve as an example of how to succeed despite obstacles. One large longitudinal study, focused on newly arrived immigrants from diverse groups (Suarez Orozco et al., 2009), found that “relational engagement” with school adults such as teachers and administrators was particularly important for academic success. Such relationships were seen as “protective,” providing the youth with “compensatory attachments, safe contexts for learning new cultural norms and practices, and information that is vital to success in schools” (Suarez Orozco et al., p. 156). Further, the importance of supportive peers was also noted in helping these immigrant students succeed at school. Formal Mentoring Programs Several studies reported on the characteristics of mentoring programs for immigrant and refugee youth. For example, Griffith, Sawrikar, and Muir (2009) described a range of field and site-based, individual and group mentoring programs provided

to newly arrived refugees from the Horn of Africa in Victoria, Australia. They analyzed interviews with mentors, program designers, community members, and refugee youth. Findings emphasized a number of specific concerns when creating programs for newly arrived refugee youth. First, findings suggested the importance of cultural competency training for mentors and mentoring organizations. Second, stakeholders mentioned the importance of waiting until after the initial settlement needs are met (e.g., 3 months after arrival) before engaging youth in a mentorship program. Third, there was no evidence in interviews that being matched on race with a mentor had advantages over nonmatching. In fact, many youth reported that being relatively new to Australia, there were benefits in nonrace matches in linking the mentees with information, services, and networks, and in providing opportunities for cross-cultural learning. This finding is consistent with meta-analytic findings of mentoring programs outside the immigration context (DuBois et al., 2011), which found smaller effects for programs that matched by racial or ethnic similarity. Fourth, Griffith et al.’s (2009) findings suggested the importance of engaging the entire refugee family in a mentoring program, given the collectivistic and family-oriented nature of the refugee communities. The authors explained, “this finding contradicts typical mentoring programs, where engaging families is not common practice, and highlights the importance of tailoring programs to be culturally appropriate” (p. 37). This range of findings was echoed in the remaining studies in the literature on mentoring programs for immigrants and refugees. School-Based Programs. In the literature, the majority of mentoring programs for immigrants that provided nonfamilial support were site-based programs that were located at schools and targeted academic success. For example, Ching, Yeh, Siu, Wu, and Okubo (2009) paired five Chinese immigrant high school students with graduate students in counseling psychology (ethnicity of mentors was not specified). The mentees were participating in a larger participatory action research project and kept track of their experience using open-ended weekly journal entries. Based on qualitative content analysis of journal entries, the authors concluded that the mentees felt they gained skills, personal growth, and guidance with the college application and admission process. Also important student participants (mentees) described a desire to serve as mentors to other immigrant students to provide comfort and social support. Diversi and Mecham (2005) reported on a mentoring program with 50 Latino high school students

362  SPECIAL POPULATIONS at risk for dropping out of school. Here a formal after-school program oriented toward academic support matched these mentees with 20 mostly Caucasian college students. Interviews and written reflections of mentors, mentees, and program coordinators were analyzed for themes. Program coordinators reported improvement in mentees’ grades and problem behaviors over the course of the program. The mentees reported that they enjoyed the program and that mentors were particularly helpful to them in completing homework and influenced them positively in their personal lives. Mentors were satisfied with the relationships they had established with the students but mentioned a number of challenges and tensions surrounding their ability to bridge cultural differences and develop appropriate boundaries around being supportive but not “enabling” around homework. The study draws attention to the cultural differences between the mentors and mentees, speculating that more bicultural competency training for mentors may have reduced cultural misunderstandings and discomfort that arose in some of the relationships. In contrast, four school-based mentoring programs that matched mentees with mentors of the same immigrant and/or ethnic background were positively viewed by participating youth. In the Shadduck-Hernandez (2006) study, refugee and immigrant undergraduate mentors from diverse ethnic groups were matched with youth from similar backgrounds. The mentors, enrolled in a community service-learning course, worked to collaboratively construct a community project with Cambodian and Vietnamese refugee middle and high school students, culminating in an exhibit of photography produced by the mentees. Analyses of ethnographic data suggested that the experience was meaningful, instructive, and identity affirming for the undergraduate mentors, bridging their personal experience with academic learning. Shadduck-Hernandez also stressed that, unlike typical service-learning programs that pair privileged students with minority mentees, the experience of working with familiar and culturally similar mentees was particularly meaningful for the mentors. Similarly, Kaplan, Turner, Piotrkowski, and Silber (2009) assessed 37 Hispanic middle school girls participating in Club Amigas, a school-based program in which they were matched with Hispanic college students who provided weekly individual and/or group activities and held workshops for parents. The girls were assessed at the beginning of and 1 year after the start of the program. Over the course of the year, measures of the girls’ self-esteem, commitment to use of Spanish, and strength of identification with their ethnic groups increased.

Yeh, Ching, Okubo, and Luthar (2007) also used a pre/post design, reporting on a school-based peer mentoring program. Twenty-three high school students who were recent immigrants from mainland China were matched with four high school student mentors. The mentors were of the same ethnic background as the students and were supervised by graduate students. During the semester-long program, mentors met weekly with their mentees individually and in groups. Mentees were found to have significantly higher scores at posttest on peer attachment—trust and peer attachment—need for closeness. However, no significant change was found with respect to college and career self-efficacy, academic self-efficacy, or social connectedness measures, and like the Kaplan et al. (2009) study, this study included no comparison group. Finally, Torres Campos et al. (2009) assessed the impact of a mentoring program for Latino students relative to a nonintervention group, though the groups were not equivalent at baseline. Eleven Latino college freshmen at risk for academic problems received mentoring from 5 Latina graduate students, and were compared to 22 Latino freshmen who were not identified as at risk and received no mentoring. The mentoring included academic advisement, friendship/support, providing knowledge and resources, and role modeling. The low-risk group scored significantly more positively than the high-risk group on all measures at baseline. In the spring, the intervention group was found to improve relative to the low-risk comparison group on measures of academic and university motivation and self-efficacy. Family-Focused Program. One example of a familyfocused mentoring relationship with a refugee youth was provided by Fritzberg and Amaleyhu (2004). This self-ethnography described a mentoring relationship between a professor of education and an Ethiopian refugee youth. The relationship was focused on helping the youth through high school, but evolved into a multiyear “mutual mentoring” relationship involving both families. The authors stressed that “one cannot overstate the importance of engaging the parents when mentoring a young person” (p. 307). Summary and Implications of Research Literature In sum, the studies reviewed here described a number of mentoring programs for immigrants and refugees. With respect to context, the majority were school based with the exception of Fritzberg and Amaleyhu’s (2004) field-based program, as well as

Immigrant and Refugee Youth   363 other field-based programs noted in the Griffith et al. (2009) review. With respect to goals, all schoolbased programs addressed some instrumental goals focused on success at school. However, even when the focus was academic, all programs went beyond tutoring and emphasized the importance of emotional and developmental aspects of the mentoring relationship. Griffith et al. stressed that mentoring may not be appropriate for newly arrived immigrants and refugees, affirming the importance of understanding immigrant youth’s acculturation stage when developing mentoring programs. The studies also stressed the value of cultural understanding when mentees and mentors are ethnically matched (Shadduck-Hernandez, 2006), and the importance of bicultural training in cross-ethnic matches (Diversi & Mecham, 2005). Structurally these programs included group and individual mentormentee matches. Cross-age mentoring emerged as the most frequently used mentoring structure in the school-based programs reviewed. Findings suggested the importance of mentors serving in a sibling-like role (Cooper et al., 1995), and the benefits of such structures for mentors who feel empowered through the process (Shadduck-Hernandez, 2006). This suggests the value of creating programs in which mentees can assume leadership roles and become mentors to more newly arrived immigrants. Finally, the importance of engaging the entire family in a mentoring relationship was emphasized (Fitzburg & Amaleyhu, 2004; Griffith et al., 2009). Although the literature described a range of programs that varied in structure, goals, and context, only three studies of mentoring programs for immigrant or refugee youth could be identified that assessed program participants before and after program enrollment (Kaplan et al., 2009; Torres Campos et al., 2009; Yeh, Ching, Okubo, & Luthar, 2007) and only one of these (Yeh et al., 2007) was focused on first-generation immigrants. These studies are important in suggesting benefits of mentoring programs for immigrant and refugee youth, but the literature is underdeveloped in terms of clarifying the specific factors in such mentoring programs that may contribute to program success. For these reasons, we turn to descriptions of mentoring programs and practices to identify promising directions for mentoring programs for immigrant and refugee youth.

Practice Many mentoring programs that target refugees and immigrants can serve as resources to practitioners searching for feasible models and lessons learned.

Unfortunately, the vast majority of these have never been formally documented or evaluated. Bridging Refugee Youth and Children’s Services (BRYCS), the national technical assistance provider on refugee and immigrant children, youth, and families for the Federal Office of Refugee Resettlement, maintains a Promising Practices Database that offers program descriptions (accessible from the BRYCS home page at www.brycs.org). A brief review of these programs revealed that approximately 20 offered mentoring as a component of more comprehensive youth or family services. The majority of these followed a “traditional” mentoring structure, matching refugee and immigrant youth with U.S.born adults from the larger community and combining instrumental and developmental goals regardless of how long the youth had been in this country. Programs Several programs attend to specific needs of immigrants and refugees depending on how long they have lived in the United States. For example, the Go-Betweener Mentoring Program (Culture Connect, 2010) in Atlanta, Georgia, serves immigrants and refugees who have been in this country for a longer period of time. The goal of this program is to help youth build positive relationships with adults who have themselves successfully navigated the acculturation experience and can support and guide youth as they learn to “go between” cultures. Mentors may work with individual youth or with sibling groups; all mentors are encouraged to connect with and support the youth’s family. This program aims to develop competence in the new culture while cultivating a sense of identity and “roots” in the native culture by providing an experienced guide who can help the youth navigate the process of identity development. Several programs use peer mentoring and some provide the opportunity for mentees to become mentors and to take leadership roles in the program. For example, the African Immigrant Mentoring (AIM) program (Immigrant and Refugee Community Organization, 2010) in Portland, Oregon, matches young African immigrant and refugee women (aged 14–23) who are facing adjustment challenges with adult mentors from the community. Mentors commit to a minimum of 1 year, and all youth participate in a service-learning curriculum. Youth who complete a year of mentoring and who graduate from the service-learning course may become peer mentors to new youth participants. In addition, these program graduates are provided with leadership opportunities on the agency’s African Youth Advisory Council.

364  SPECIAL POPULATIONS This program not only assists refugee and immigrant youth with their developmental needs, but also “empowers” them in the process and prepares these young people to step into leadership roles themselves. Resources A number of resources that directly address mentoring refugee and immigrant youth have recently been developed, chief among them Mentoring Immigrant and Refugee Youth: A Toolkit for Program Coordinators (MENTOR, 2007). The toolkit represents the most extensive resources available to date on this topic. It covers current backgrounds and statistics for immigrant youth, how to conduct a needs assessment that includes engaging immigrant communities and the agencies that serve them, integrating cultural competence into mentoring program management, cultural competency training for mentors, and building and sustaining mentoring relationships with immigrant youth. The BRYCS Clearinghouse (www.brycs.org) contains thousands of resources for agencies serving refugee and immigrant youth. BRYCS has published several resources that can assist mentoring programs, including the recent brief, “New Directions in Mentoring Refugee Youth” (BRYCS, 2010). A resource aimed at those developing programs for refugee and immigrant youth, “Growing Up in a New Country: A Positive Youth Development Toolkit for Working With Refugees and Immigrants” (BRYCS, 2006) includes a “Toolbox” on mentoring refugee and immigrant youth. Both the MENTOR and the BRYCS publications provide links to additional resources. Below we summarize our recommendations. Listed in Table 24.1 and depicted in Figure 24.1, we present some key issues confronted by refugees and immigrants. We present questions for program staff to ask about their programs and provide suggestions for mentoring programs that follow from the model illustrated in Figure 24.1. Given the state of the research literature, these suggestions are more theoretically based than empirically based, and we hope future research and evaluation efforts will test whether or not the tenets of the model hold. Know Your Local Community It is important that developers of mentoring programs know and understand the different refugee and immigrant populations and the network of agencies that serve them. The ecological perspective described above suggests the importance of learning about existing community resources and structures

and building on them when creating a new program. The process of learning about the cultural community must focus on strengths—of the individual, the family, and the ethnic community/culture (Morland, 2007). Mentors must receive training in understanding specific aspects of the cultures with which they may be working, as well as a general orientation to sensitivities involved when working across cultures. Take the Process of Acculturation Into Account When Designing Programs and Services The stage of the acculturation and resettlement process may dictate whether and how mentoring services are provided (see Figure 24.1). It may be useful to match recently arrived immigrants and refugees with mentors from the majority culture or from prior immigration waves. In these situations, it is particularly important that program staff and mentors receive sufficient training in cultural competency. Mentoring can be focused on instrumental goals, such as learning English, gaining academic and social skills, and negotiating the school system. Such an instrumental focus may be particularly important early on, whereas the relationship may shift to encompass more social-emotional/developmental goals after the youth’s immediate adjustment needs are met. School-based cross-age programs may be particularly useful in helping the newly arrived youth to function academically and in enhancing “school belonging.” For immigrant and refugee youth who have been in this country long enough to be relatively proficient in English and U.S. culture, mentors from a similar ethnic background or race may be particularly helpful as role models for integrating their multiple cultural identities and negotiating both the native and U.S. cultures. Use Co- or Mutual Mentoring A particularly useful approach for immigrant and refugee youth may involve co-mentoring or cross-age mentoring programs that develop mentees into mentors. Such an empowering approach can increase feelings of inclusion and competence. It gives youth opportunities to feel that their skills and culture are valued and invites them to become better integrated into the life of their new community and school. Engage the Family as a Whole For all of the reasons outlined in the preceding sections, a “family mentoring” approach can

Immigrant and Refugee Youth   365 Table 24.1   Checklist for Practitioners Key Question

Recommendations

How do you address the acculturation gap between parents and youth?

Consider family-strengthening mentoring strategies such as assigning an individual mentor to an entire family or pairing with a mentor-family. Focus initially on instrumental mentoring to facilitate trust and communicate to parents support of their own parenting practices. Explore ways to support native language learning/maintenance for the youth. Explore ways to assist parents in understanding U.S. schools and peer norms to support their parenting practices.

What are you doing to serve the unique needs of newly arrived immigrants and refugees?

Consider prioritizing an instrumental mentoring approach. Consider pairing with U.S.-born mentors to provide “institutional agents” to help with orientation to new settings such as school. Consider cross-age pairing with mentors who are immigrants but have lived in the U.S. longer, and provide opportunities for mentees to later become mentors for newer arrivals.

What are you doing to serve the unique needs of immigrant and refugee youth who have lived in the country longer?

Consider prioritizing a developmental mentoring approach, particularly in dealing with issues of cultural/ethnic identity. Adult mentors from the same ethnic/immigrant group, or from a similar immigration background, may be particularly useful as role models for how to integrate cultures and identities. Structure mentoring so that it helps support the family, to alleviate the negative impact of acculturation gaps that develop over time.

How are you addressing the problem of interrupted schooling?

Promote a school-based instrumental mentoring approach that emphasizes academic support and tutoring. Couple mentoring with academic support for extended periods of time (several years after arrival), particularly for those with backgrounds of no prior literacy or schooling. Utilize a field-based family-focused mentoring approach if literacy is an issue for the entire family.

How are you addressing the problem of traumatic experiences?

Consider developmental mentoring from adults with good training in trauma-informed services to help youth develop a trusting relationship.

How can you compensate for the lack of local ethnic community?

Utilize cross-ethnic mentoring to provide additional sources of instrumental, emotional, and social support. Require cultural competency training for staff and mentors.

strengthen relationships between the youth and family, and ensure that the mentoring relationship does not inadvertently increase the acculturation gap. At the same time, helping youth develop pride in their parents’ and their cultural backgrounds can help promote a positive sense of identity. Models

can include matching an adult mentor with a youth while including the family in the mentoring activities and process; or matching a mentor family with a mentee family, enabling both cross-age and cross-generational mentoring among the different family members. A focus on instrumental goals

366  SPECIAL POPULATIONS can be particularly useful in helping build trust and reassuring the parents that their parenting practices and cultural values are not being questioned by the mentors.

Conclusion In this chapter, we reviewed demographic and background characteristics, ecological and acculturation theories, research, and current practice with respect to mentoring of immigrant and refugee youth. Despite a relative lack of research on the effectiveness of mentoring programs with these populations, we identified a number of issues that can guide practitioners in developing such programs and highlighted the value of understanding the diversity of immigrant groups and their experiences when conducting research and planning programs. Given this tremendous diversity, a one-size-fits-all approach simply cannot work, and providers need a range of tools and approaches at their disposal to address varied needs. Further, greater attention must be given to describing the participants in mentoring research in terms of their specific heritage, generational status, and acculturation level in order to better understand youths’ needs and program effectiveness.

References Arellano, A. R., & Padilla, A. M. (1996). Academic invulnerability among a select group of Latino university students. Hispanic Journal of Behavioral Sciences, 18, 485–507. Birman, D. (2006). Acculturation gap and family adjustment: Findings with Soviet Jewish refugees in the U.S. and implications for measurement. Journal of Cross-Cultural Psychology, 37, 568–589. Birman, D., Ho, J., Pulley, E., Batia, K., Everson, M. L., Ellis, H., . . . Gonzalez, A. (2005). Mental health interventions for refugee children in resettlement. White Paper II, Refugee Trauma Task Force, National Child Traumatic Stress Network, Chicago, IL. Birman, D., & Ryerson-Espino, S. (2007). The relationship of parental practices and knowledge to school adaptation for immigrant and non-immigrant high school students. Canadian Journal of School Psychology, 22, 152–166. Birman, D., & Trickett, E. J. (2001). The process of acculturation in first generation immigrants: A study of Soviet Jewish refugee adolescents and parents. Journal of Cross-Cultural Psychology, 32, 456–477. Birman, D., Trickett, E., & Buchanan, R. (2005). A tale of two cities: Replication of a study on the acculturation

and adaptation of immigrant adolescents from the former Soviet Union in a different community context. American Journal of Community Psychology, 35, 87–101. Birman, D., Weinstein, T., Beehler, S., & Chan, W. (2007). Immigrant youth in U.S. schools: Opportunities for prevention. Prevention Researcher, 14, 14–17. Bridging Refugee Youth and Children’s Services (BRYCS). (2006). Growing up in a new country: A positive youth development toolkit for working with refugees and immigrants. Washington, DC: Author. Retrieved from http://www.brycs.org/documents/ upload/GrowingUpInANewCountry-Web.pdf Bridging Refugee Youth and Children’s Services (BRYCS). (2010). New directions in mentoring refugee youth. Washington, DC: Bridging Refugee Youth and Children’s Services/Migration and Refugee Services/ U.S. Conference of Catholic Bishops. Retrieved from http://www.brycs.org/documents/upload/BRYCSBRIEF-Mentoring-Summer-2010.pdf Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psycho­logist, 32, 513–531. Catholic Relief Services. (2010). Child migration: The detention and repatriation of unaccompanied Central American children from Mexico. Baltimore, MD: Author. Ceballo, R. (2004). From barrios to Yale: The role of parenting strategies in Latino families. Hispanic Journal of Behavioral Sciences, 2, 171–186. Ching, A., Yeh, C., Siu, W. Y., Wu, K., & Okubo, Y. (2009). Evaluation of a school-based internship program for Chinese immigrant adolescents in the United States. Adolescence, 44, 601–620. Cooper, C. R., Jackson, J. F., Azmitia, M., Lopez, E. M., & Dunbar, N. (1995). Bridging students’ multiple worlds: African American and Latino youth in academic outreach programs. In R. F. Macias & R. G. Garcia Ramos (Eds.), Changing schools for changing students: An anthology of research on language minorities (pp. 211–234). Santa Barbara: University of California Linguistic Minority Research Institute. Culture Connect. (2010). Go Betweener Mentoring Program. Retrieved from http://www.brycs.org/ promisingpractices/program.cfm?ID=3246 Diversi, M., & Mecham, C. (2005). Latino(a) students and Caucasian mentors in a rural after-school program: Toward empowering adult-youth relationships. Journal of Community Psychology, 33, 31–40. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Family Strengthening Policy Center. (2004). Mentoring as a family strengthening strategy: Policy brief no. 4. National Human Services Assembly. Accessed from: http://www.nydic.org/fspc/practice/documents/ Brief4.pdf

Immigrant and Refugee Youth   367 Fritzberg, A., & Amaleyhu, A. (2004). Mutual mentoring: Co-narrating an educative friendship between an education professor and an urban youth. Urban Review, 36, 293–308. Fuligni, A. J. (1997). The academic achievement of adolescents from immigrant families: The roles of family background, attitudes, and behavior. Child Development, 68, 351–363. Gordon, M. (1964). Assimilation in American life: The role of race, religion, and national origins. New York: Oxford University Press. Griffith, M., Sawrikar, P., & Muir, K. (2009). Culturally appropriate mentoring for Horn of Africa young people in Australia. Youth Studies Australia, 28, 32–40. Harris, K. M. (1999). The health status and risk behaviors of adolescents in immigrant families. In D. J. Hernandez (Ed.), Children of immigrants: Health, adjustment, and public assistance (pp. 286–347). Washington, DC: National Academies Press. Ho, J., & Birman, D. (2010). Acculturation gap in Vietnamese refugee families: Impact on family adjustment. International Journal of Intercultural Relations 34, 22–33. Immigrant and Refugee Community Organization. (2010). Young Women’s Equity Project. Portland, OR. Retrieved from http://www.brycs.org/promisingpractices/program.cfm?ID=32 Jaycox, L. H., Stein, B. D., Kataoka, S. H., Wong, M., Fink, A., Escudero, P., . . . Zaragoza, C. (2002). Violence exposure, posttraumatic stress disorder, and depressive symptoms among recent immigrant schoolchildren. Journal of the American Academy of Adolescent Psychiatry, 41, 1104–1110. Jones, C., Trickett, E., & Birman, D. (2012). Determinants and consequences of child culture brokering in families from the former Soviet Union. American Journal of Community Psychology, 50(1–2), 182–196. Jucovy, L. (2002). Same and cross-race matching. Northwest Educational Regional Library. Retrieved from http://www.ppv.org/ppv/publications/assets/26 _publication.pdf Kaplan, C., Turner, S., Piotrkowski, C., & Silber, E. (2009). Club Amigas: A promising response to the needs of adolescent Latinas. Child and Family Social Work, 14, 213–221. Karcher, M. J., Kuperminc, G. P., Portwood, S. G., Sipe, C. L., & Taylor, A. S. (2006). Mentoring programs: A typology to inform program development, research, and evaluation. Journal of Community Psychology, 34, 709–725. Karcher, M. J., & Nakkula, M. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. New Directions for Youth Development, 2010 126, 13–32. Kia-Keating, M., & Ellis, H. (2007). Belonging and connection to school in resettlement: Young refugees, school belonging, and psychosocial adjustment. Clinical Child Psychology and Psychiatry, 12, 29–43.

MENTOR. (2007). Mentoring immigrant youth: A toolkit for program coordinators. Alexandria, VA: Author. Retrieved from http://www.mentoring.org/find_resources/ immigrant_toolkit/ Morland, L. (2007). Promising practices in positive youth development with immigrants and refugees. Prevention Researcher, 14(4), 18–20. Negash, E. (2010). State Letter #10-09. Office of Refugee Resettlement Guiding Principles, Office of Refugee Resettlement, Administration for Children and Families, U.S. Department of Health and Human Services. Retrieved from http://www.acf.hhs.gov/ programs/orr/policy/sl10-09.htm Ogbu, J. A. (1990). Mentoring minority youth: A framework. New York: Columbia University, Teachers College, Institute for Urban and Minority Education. Oppedal, B., Roysamb, E., & Sam, D. L. (2004). The effect of acculturation and social support on change in mental health among young immigrants. International Journal of Behavioral Development, 28, 481–494. Portes, A., & MacLeod, D. (1996). Educational progress of children of immigrants: The roles of class, ethnicity, and school context. Sociology of Education, 69, 255–275. Portes, A., & Zhou, M. (1993). The new second generation: Segmented assimilation and its variants. Annals of the American Academy of Political and Social Science, 530, 74–96. Pumariega, A. J., Rothe, E., & Pumariega, J. B. (2005). Mental health of immigrants and refugees. Community Mental Health Journal, 41, 581–597. Rhodes, J. W., Bogat, A., Roffman, J., Edelman, P., & Galasso, L. (2002). Youth mentoring in perspective: Introduction to the special issue. American Journal of Community Psychology, 30, 149–155. Sánchez, B., & Colón, Y. (2005). Race, ethnicity, and culture in mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 191–204). Thousand Oaks, CA: Sage. Sánchez, B., Esparza, P., & Colón, Y. (2008). Natural mentoring under the microscope: An investigation of mentoring relationships and Latino adolescents’ academic performance. Journal of Community Psychology, 36, 468–482. Sánchez, B., Reyes, O., & Singh, J. (2006). A qualitative examination of the relationships that serve a mentoring function for Mexican American older adolescents. Cultural Diversity and Ethnic Minority Psychology, 12, 615–631. Santon-Salazar, R., & Urso Spina, S. (2003). Informal mentors and role models in the lives of urban Mexicanorigin adolescents. Anthropology and Education Quarterly, 34, 231–254. Shadduck-Hernandez, J. (2006). Here I am now! Critical ethnography and community service-learning with immigrant and refugee undergraduate students and youth. Ethnography and Education, 1, 67–86. Suarez-Orozco, C., Rhodes, J., & Milburn, M. (2009). Unraveling the immigrant paradox: Academic engagement and disengagement among recently

368  SPECIAL POPULATIONS arrived immigrant youth. Youth & Society, 41, 151–185. Szapocznik, J., Rio, A., Perez-Vidal, A., Kurtines, W. M., & Santisteban, D. (1986). Family effectiveness training (FET) for Hispanic families. In H. P. Lefley & P. B. Pedersen (Eds.), Cross-cultural training for mental health professionals (pp. 245–261). Springfield, IL: Charles C Thomas. Torres Campos, C., Phinney, J., Perez-Brena, N., Kim, C., Ornelas, B., Nemanim, L., . . . Ramirez, C. (2009). A mentor-based targeted intervention for high-risk Latino college freshmen: A pilot study. Journal of Hispanic Higher Education, 8, 158–178. Trickett, E. J. (1996). A future for community psychology: The contexts of diversity and the diversity of contexts. American Journal of Community Psychology, 24, 209–234.

Trickett, E. J., & Birman, D. (1989). Taking ecology seriously: A community development approach to individually based preventive intervention in schools. In B. Compas & L. Bond (Eds.), Primary prevention in the schools (pp. 361‑390). Newbury Park, CA: Sage. U.S. Department of Homeland Security. (2009a). Yearbook of immigration statistics: 2008. Washington, DC: U.S. Department of Homeland Security, Office of Immigration Statistics. U.S. Department of Homeland Security. (2009b). Yearbook of immigration statistics: 2009, refugees and asylees. Retrieved from http://www.dhs.gov/files/statistics/ publications/YrBk09RA.shtm Yeh, C., Ching, A., Okubo, Y., & Luthar, S. (2007). Development of a mentoring program for Chinese immigrant adolescents’ cultural adjustment. Adole­ scence, 42, 733–747.

25  CHILDREN OF INCARCERATED PARENTS J. Mark Eddy, Jennifer Cearley, Joseph Bergen, and Jenny Stern-Carusone

Introduction Over the past several decades, social and political changes in the United States have led to a substantial increase in the number of incarcerated adults (Western, 2006; Western & Wildeman, 2009). The majority of prisoners are the parents of minors. Between 1991 and 2007, the number of parents held in prisons increased by 79% (Glaze & Maruschak, 2009); thus, the number of children affected by incarceration also increased significantly. Today, over 2% of children have a parent in prison (a state or federal incarceration facility), and many more have a parent in jail (a local place of incarceration, usually operated by the county) or who was in prison or jail (Maruschak, Glaze, & Mumola, 2010). Due to potential exposure to a variety of situations fraught with risk, including parent criminal behavior before incarceration and parent-child separation during incarceration, the children of incarcerated parents have often been considered a vulnerable population (e.g., Bernstein, 2007; Gabel & Johnston, 1998). Most prominently, these children are two times more likely than their peers to display antisocial behavior (Murray, Farrington, & Sekol, 2012; Murray, Loeber, & Pardini, 2012), a predictor of delinquency during adolescence and adulthood (e.g., Lipsey & Derzon, 1998). Risk for other types of problems is unclear. Despite concerns for the well-being of this population, few preventive interventions have been developed or adapted for the children of incarcerated parents. From a life-course development perspective, a preventive approach would first address parenting and factors that can significantly impair the quality of parenting, such as a parental substance abuse, parent employment, and availability and stability of housing

(e.g., Eddy, Kjellstrand, Martinez, & Newton, 2010). However, the most popular intervention, by far, has been much more limited in scope, namely, volunteer mentoring. While mentoring the children of incarcerated parents is an idea that had been noted in the research and practice literature over the past several decades (e.g., van Nijnatten, 1997; Weissman & LaRue, 1998), only recently did it come to be viewed as an intervention of exceptional significance. In the first edition of the Handbook of Youth Mentoring, the idea of mentoring the children of incarcerated parents was mentioned only in passing (i.e., Maton, Domingo, & King, 2005). Here, we provide a gateway for practitioners to the professional literature on mentoring the children of incarcerated parents. We first highlight the characteristics of children affected by parental imprisonment, discuss why mentoring became a popular intervention for these children, and present the theoretical underpinnings of such work. We then describe the limited research on outcomes related to mentoring the children of incarcerated parents, and summarize the more prominent “recommended” practices literature, which is based primarily on anecdotal experiences and on studies of “at-risk” children in general. To place the available research and practice information in context, we describe two types of youth mentoring programs that serve children of incarcerated parents and illustrate how these programs use recommended practices through case examples. We conclude with recommendations and considerations for future work. Population Characteristics Over 1.7 million minor children in 2007 had a parent in prison (Maruschak et al., 2010). About 369

370  SPECIAL POPULATIONS 25% of these children were under the age of 4 years, 25% were aged 4–9, and 50% were 10 or older (Maruschak et al., 2010). Because most (93%) inmates are men, 91% of children affected by incarceration have an incarcerated father (West & Sabol, 2008). Minority children are affected differentially by parental imprisonment. For example, in 2007, whereas only 0.9% of non-Hispanic White children had a parent in prison, 2.4% of Latino children and 6.7% of African American children had a parent behind bars for extended periods of time. This equates to 484,100 White children; 362,800 Latino children; and 767,400 African American children. Disparities in the experience of parental incarceration increase dramatically when race/ethnicity, education, and cumulative risk are considered simultaneously. By the age of 14, for example, the lifetime risk of having an incarcerated father is vastly different for an African American youth whose father did not graduate from high school (50.5% experienced parental incarceration) versus a White youth whose father did not graduate from high school (7.2%) or an African American or White youth whose father attended at least some college (13.4% and 1.7%, respectively; see Wildeman, 2010). Notably, when asked about their life circumstances before their arrest and/or incarceration, over 50% of parents reported living with their children and about 50% of parents reported providing the primary financial support for their children (Glaze & Maruschak, 2009). History It is not uncommon for community leaders to get together and discuss possible solutions to local social problems, but it is rare that such work helps ignite a national movement. This was the case in Philadelphia. Concern for youth growing up in struggling, inner-city neighborhoods, and particularly for those affected by parental incarceration, led to a series of meetings among city leaders in the local political, academic, nonprofit, private foundation, governmental, and faith communities (Husock, 2003). One outcome of these interactions was the creation of Amachi, a faithbased mentoring program designed specifically for the children of incarcerated parents (Goode & Smith, 2005; Jucovy, 2003). Amachi was built on knowledge gained through the landmark randomized controlled trial of the Big Brothers Big Sisters (BBBS) mentoring program, conducted by Public/Private Ventures, a Philadelphia research firm (see Tierney & Grossman, 1995). The ideas at the core of Amachi were that volunteer mentors

would be recruited through church congregations; screened, trained, and matched with youth; and monitored and supported by professionals from BBBS. Early in President George W. Bush’s administration, Amachi caught the attention of the president. As a candidate, Bush had highlighted his beliefs in the importance of faith and volunteerism in addressing social problems. Mentoring, and programs like Amachi, encapsulated these ideas. Bush expressed his support for youth mentoring during his inaugural address, and immediately took steps that would eventually lead to an increase in federal funding for mentoring. One of his first actions as president was to establish a White House Office of Faith-Based and Community Initiatives. He appointed as its director John J. Dilulio, a professor from Philadelphia who had been involved in the work that created Amachi. Through the new faith-based office, Bush visited the program during the next summer, and several months later signed into law amendments to the Safe and Stable Families program that established a competitive grant program to support the mentoring of children of incarcerated parents. During the subsequent winter, the U.S. Department of Health and Human Services (DHHS) sponsored a national policy conference that focused on the needs of children of incarcerated parents and their families (Travis & Waul, 2003), and plans to broaden the program were announced. In his State of the Union address the following year, Bush highlighted a $450 million youth mentoring initiative that included funding for programs targeting children of prisoners. The Administration for Children and Families, a branch of the DHHS, brought the program into fruition with a request for applications for the Mentoring Children of Prisoners (MCP) program. The request did not mention Amachi by name, but encouraged applications from programs that had similar characteristics. Over the remainder of the Bush administration and during the first 3 years of the Obama administration, MCP grants were awarded to community and faith-based organizations, state and local governments, and tribes to provide the children of incarcerated parents with volunteer mentors. Grant recipients were required to screen mentors, provide training, create “suitable” mentor-youth matches, provide ongoing support and oversight of matches, and monitor youth progress and outcomes. MCP mentors were asked to commit to spending at least 1 hour a week with a child for at least 1 year. Although funding for the program ended recently, MCP brought mentoring

Children of Incarcerated Parents   371 into the limelight as “the” preventive intervention for the children of incarcerated parents.

Theory The justification for mentoring the children of incarcerated parents is typically as follows. Ongoing, positive relationships with nonfamilial adults can improve outcomes for some children from high-risk backgrounds (e.g., DuBois & Silverthorn, 2005; Werner, 1989; Zimmerman, Bingenheimer, & Notaro, 2002). Although such relationships occur naturally during childhood for many children, the complex life situations of some children of incarcerated parents, including frequent household and school moves and/or caregiver changes, may make it more difficult for these children to establish and maintain close and enduring relationships with adults. In these cases, artificial mechanisms may be needed to ensure that mentoring relationships occur (Jucovy, 2003). Unfortunately, there are no studies documenting whether the children of incarcerated parents are either more or less in need of assistance in forming close adult relationships than are other youth. More broadly, the general theory proposed by Rhodes (2002) on how youth mentoring influences youth outcomes seems quite applicable to the population of children of incarcerated parents. As would be expected, the theory highlights the importance of the bond between the mentor and a youth. The development of a strong interpersonal relationship based on mutuality, trust, and empathy is hypothesized to lead to youth gains in social, emotional, cognitive, and identity development. Positive changes in these areas are thought to lead to the improved social functioning of youth in both family and peer relationships. These changes, combined with other advances due to the mentoring relationship, are hypothesized to lead to a variety of positive youth and young adult outcomes. The potentially special significance of this theory to the children of incarcerated parents is related to gains in social and emotional development (Shlafer, Poehlmann, Coffino, & Hanneman, 2009). Separation from a parent during incarceration, as well as relationship disruptions that occurred prior to incarceration, may contribute to a child’s developing a set of cognitive expectations about relationships that negatively affect the child’s ability to form new social bonds. Attachment theorists (e.g., Ainsworth, 1989; Bowlby, 1979) refer to these expectations as the child’s “working model.” Insecure working models may arise from a history of relationship separation and disruption, and children who develop such models tend to view

themselves as unworthy, and others as untrustworthy. These cognitions place children at risk for a variety of adjustment problems (e.g., Deklyen & Greenberg, 2008). The development of a bond with a mentor, and the maintenance of that relationship over time, may modify a child’s working model, and may open up new possibilities for constructive social relationships and other positive outcomes (Carlson & Sroufe, 1995; Rhodes, Haight, & Briggs, 1999). Various contextual factors, which in Rhodes’s (2002) model are referred to as moderators or mediators of mentoring outcomes, may be of particular importance when considering other ways in which mentoring might benefit the children of incarcerated parents. These include family relationships, peer relationships, economic circumstances, and neighborhood characteristics. By definition, a child does not qualify for a mentoring program for the children of incarcerated parents until his or her parent is or has been in prison or jail. Many difficulties may have occurred in the life of a child by this point in time. Figure 25.1, adapted from Parke and Clarke-Stewart (2003), illustrates a “transactional” model that takes “preincarceration conditions” (i.e., contextual factors present before an incarceration) into account and then synthesizes a variety of viewpoints pertinent to understanding the lives of the children of incarcerated parents and how they may, or may not, respond to mentoring. Perspectives embedded in the model include the aforementioned attachment theory (e.g., Bowlby, 1973), as well as lifespan theory (e.g., Elder, 1998), social learning theory (e.g., Eddy & Reid, 2003), systems theory (e.g., Minuchin, 2002), risk and resiliency theory (e.g., Rutter & Sroufe, 2000), and models of cumulative risk (e.g., Sameroff, Bartko, Baldwin, Baldwin, & Seifer, 1998). Within the transactional model, for families in which a parent played a significant role in the family prior to incarceration, parental incarceration can initiate a chain of effects that are influenced by the preincarceration context of the child and family, such as the individual characteristics of the child and caregivers, family interaction patterns, and family economic circumstances. When parents are incarcerated, new stresses confront the family, and the composition of the family may change beyond just the absence of the incarcerated parent. At the same time, old stresses may be gone. The incarcerated parent, may, for example, have been domestically violent, or may have exposed children to high-risk individuals and situations. Such changes in family circumstances may influence caregiver distress and the social support the family receives in positive or negative ways, depending on the issue. Further, these factors may, in turn, interact with a variety of other issues, such as

372  SPECIAL POPULATIONS

Figure 25.1

Transactional Model of Mentoring Relationships in the Life Contexts of Children of Incarcerated Parents

Child adjustment

Preincarceration conditions

Caregiver distress Stressful life experiences

Incarceration of parent

Social support

Caregiving processes

Changes in family composition Contact with incarcerated parent

Mentoring relationship

Source: Adapted from Parke and Clarke-Stewart (2003). the location of a prison or jail and transportation options, to determine the extent to which a child and family have contact with an incarcerated parent. Most notably, and based on a broad set of studies of parental influences on children, what appears most important to child adjustment in the face of a stressor like parental incarceration is the quality of the parenting that a child receives on a day-to-day basis (see Patterson, 1982; Patterson, Reid, & Dishion, 1992; Reid, Patterson, & Snyder, 2002). The shaded triangle in Figure 25.1 highlights the primary context in place when a caregiver allows his or her child to be mentored. Once a mentoring relationship is established, either through a mentoring program, or naturally, youth changes initiated and/or fostered by the mentor-youth relationship may reverberate within and change the relationships of the child with his or her caregivers and incarcerated parent(s). Ideally, these changes would be positive, improve the quality of these relationships, and provide new opportunities for a child to experience positive outcomes in his or her life.

Research To ensure adequate coverage of the research literature on outcomes due to mentoring the children of incarcerated parents, we searched two academic

databases (PsycINFO and Academic Search Premier), conducted internet searches through Google Scholar and Bing, and queried mentoring researcher and practitioner colleagues. We identified four peer-reviewed articles (i.e., Bruster & Foreman, 2012; Laakso & Nygaard, 2012; Shlafer & Poehlmann, 2010; Shafler et al., 2009), each of which reported outcomes for participants from MCP-funded programs. In these studies, measures of child adjustment or other issues of interest were taken at several points in time for a group of mentored children (i.e., a single-group repeated-measures study). We also found several similar unpublished reports (e.g., Andrade et al., 2008). Only one (unpublished) study compared children who were mentored to children who were not (i.e., ICF International, 2011). Finally, several other papers described programs but did not report child outcome data (e.g., Bocknek, Sanderson, & Britner, 2009; Jucovy, 2003; van Nijnatten, 1997; Weissman & LaRue, 1998). All of these studies and reports focused on volunteer mentoring with elementaryschool- or middle-school-aged children. Most programs were sponsored through MCP funding and delivered by BBBS. To illustrate findings in a typical single-group repeated-measures study, in Shlafer and colleagues (2009), children who were mentored were reported to exhibit behavior problems at higher rates than

Children of Incarcerated Parents   373 normative samples and to exhibit many of those behaviors within the context of relationships, particularly with family members and peers. Most children had little in-person contact with their incarcerated parent, with most communication taking place through letters and phone calls. Mentors and children met an average of 3 times per month and spent 2.5 hours together per meeting. More than one-third of the mentor-youth matches created during the study lasted less than 6 months. The primary reasons for match termination were scheduling conflicts, personal or family issues, residential mobility, mentors underestimating the commitment, and match incompatibility. For those matches that continued to meet for at least 6 months, children who had more contact with their matches were reported to exhibit fewer “externalizing” (e.g., antisocial, impulsive, inattentive behaviors) and fewer “internalizing” (e.g., depressive, anxious, obsessive behaviors) symptoms. The only comparison group study of mentoring children of incarcerated parents (ICF International, 2011) contrasted outcomes for children randomly assigned to a ready-to-match list (i.e., the “mentoring” group) or to an 18-month wait before being ready to match (i.e., the “control” group). Although 1,051 children were eligible for the study, only 26% of these children and their parents/caregivers participated in the initial assessment, only 21% participated in the 6-month follow-up assessment, and only 13% participated in the 18-month follow-up assessment. Few mentors participated in assessments. At the 6-month followup, when 81% of mentor-child matches were still active, several statistically significant differences were found between children in the mentoring group versus the control group. Namely, children in the mentoring group reported more positive relationships with parents/caregivers, higher selfesteem, and a more positive sense of the future. Several similar differences also were observed at the 18-month follow-up, again favoring the mentoring group. No differences were found between the groups on school-related outcomes or child risk behaviors at any point in time. The study’s problematic design characteristics, such as the low initial participation rate and the high attrition rate, and the failure to address missing data in the analyses, suggest caution in interpreting these findings. A key point embedded in several of these reports is the variety in the past and present life situations of the children of incarcerated parents (see Eddy & Poehlmann, 2010; Poehlmann & Eddy, in press). For example, Davies, Brazzell, La Vigne, and Shollenberger (2008) interviewed mentors who

worked with children of incarcerated parents in BBBS programs in Baltimore, Milwaukee, and Washington, D.C. Seven focus groups were conducted, and mentors were guided through a series of questions about their mentees and families. Follow-up interviews were then conducted with staff members at each program to gain further understanding. The major conclusion was that there was “considerable variation” in circumstances, and presumed risk, within the population of children of incarcerated parents, and that a “one size fits all” approach to working with these children was not appropriate. Some children have supportive and engaged adults in their lives already and may not need or benefit from a mentor; other children are in very different situations. Conclusion and Recommendations Insufficient information is available to draw conclusions about the value of mentoring for the children of incarcerated parents, at least in terms of research findings. No information is available on other important and related topics, such as the prevalence or impact of natural mentors on the children of incarcerated parents. The children of incarcerated parents appear to come from a variety of backgrounds and have a wide range of intervention needs, including no need. Unfortunately, the relations between various background characteristics, needs, and mentoring outcomes for the children of incarcerated parents have not been examined. An area of research that is particularly lacking is investigation on the implications of racial and ethnic disparities for mentoring practice. For example, no studies were found on culturally competent mentoring practices and how they affect youth outcomes. High-quality research studies are sorely needed on each of these topics. Of greatest research interest at present is whether the numerous mentoring programs that now exist for the children of incarcerated parents make a real difference for youth outcomes, and, if the impact is small or nonexistent, what needs to be changed so that these programs can generate positive impacts. To accomplish this work, researchers need to develop collaborative partnerships with practitioners; conduct studies that use scientifically rigorous practices for evaluations of program efficacy and effectiveness; examine mediators and moderators of outcomes if mentoring programs for this population do, in fact, “work”; and disseminate information on research findings in user-friendly formats to youth, parents/caregivers, researchers, practitioners, administrators, and policymakers.

374  SPECIAL POPULATIONS

Practice Despite the limited research literature on mentoring the children of incarcerated parents, a burgeoning practice literature exists. A number of published articles describe mentoring programs. The typical program uses community-based volunteers (see Sipe, 2005) and tends to have characteristics similar to BBBS, such as a requirement that mentors commit to at least 1 year of service. The degree of structure (i.e., mentor screening, training, support) for programs varies. As highlighted in the prior section, some practice-focused articles provide information on the children served and information on outcomes. Of higher profile, however, are the recent recommended practice reports on mentoring the children of incarcerated parents. A number of national organizations have produced documents in this regard, and have actively disseminated their conclusions through professional conferences and the internet. To ensure that we covered this literature fully, we used the same search strategies here as above. From this search, we identified seven practitioner-focused publications on mentoring the children of incarcerated parents, five of which included specific practice recommendations (i.e., Adalist-Estrin, 2004; Bilchik, 2006; FASTEN, 2004; Goode & Smith, 2005; LEARNS, 2004). Citations within these publications were limited, so presumably justification for recommendations came from the collective practice experiences of the authors and their colleagues. To collate the practices recommended in these documents, two researchers read each document and independently generated a list of practice recommendations from the text. The lists were then compared, and different points of view about any particular recommendation were discussed and resolved. This process was repeated for each document. At the end, each researcher created a summary list of recommendations, and cross-checked the list with the original documents. These lists were compared, discussed, and differences resolved to create a final list of commonly recommended practices that appeared in a majority of the documents (see Table 25.1). The first 6 of these practices apply to mentoring any youth, and overlap with 10 of the 11 “theoretically supported” and recommended practices for mentoring noted in the DuBois, Holloway, Valentine, and Cooper (2002) meta-analysis. DuBois et al. (2002) derived their list from prior recommendations on the characteristics of “effective” mentoring programs (i.e., Freedman, 1992; Hamilton &

Hamilton, 1992; National Mentoring Working Group, 1991; Saito & Blyth, 1992) and found that programs that utilized a greater number of these practices had greater positive impacts on youth (DuBois et al., 2002). The other two practices address specific needs that arise when working with any defined “at-risk” population, namely the development of partnerships that help the program connect with the designated service population and that provide complimentary services to meet the other needs of that population. Recommended Practices in Context Table 25.1 offers ideas on how each of these recommended practices might be delivered with children of incarcerated parents. However, actual practices are shaped by a variety of forces, including program characteristics. To illustrate this, we briefly describe two very different community-based mentoring programs. Each has had extensive experience over the past several decades working with the children of incarcerated parents. The first is a short-term program that uses volunteer, part-time mentors; the second is a long-term program that employs paid, full-time mentors. Volunteer Mentoring. Committed Partners for Youth–Big Brothers Big Sisters of Lane County (CPY), a nonprofit mentoring organization founded in 1991, focuses on youth aged 6–18 years who are affected by a variety of “high-risk” circumstances, including parental incarceration. To fulfill their mission to “inspire the development of personal identity and healthy relationships while connecting youth to meaningful opportunities in their community,” CPY relies on volunteer mentors who work one-on-one with youth within community settings. Although the program was not designed for the children of incarcerated parents, due to the advent of MCP funding and a decrease in other funding streams, by 2010, 100% of the youth enrolled in CPY mentoring services were (a) youth currently or formerly incarcerated in the juvenile justice system, (b) youth with a parent, caregiver, or close relative previously or currently incarcerated, and/or (c) youth living in neighborhoods with high rates of incarceration. To identify eligible youth, CPY actively partners with schools, public agencies, and nonprofits, and conducts outreach through state prisons and local jails. About 37% of referrals come directly from families, many of whom learned of the program via word of mouth. The majority of the youth in the program are male (75%), and more racial and ethnic minority youth participate (i.e., 14% Latino, 8%

Children of Incarcerated Parents   375 Table 25.1   Checklist for Practitioners General Practice

Applications for Children With Incarcerated Parents

The mentoring program has clear goals and protocols

The roles of the mentor are specified (e.g., the mentor is not a parent or caregiver to the child, the mentor is not a social worker or therapist, the mentor does not relay messages between and among family members). The program has procedures to actively inform incarcerated parents about the mentoring program.

Procedures are specified for volunteer recruitment, screening, and training

Mentor training includes information on the impacts of incarceration on children, including issues such as potential stigma and shame, traumarelated behaviors, and attachment disturbances.

Procedures are specified to facilitate the creation of good matches

Supervisors are provided training in effective ways to support mentors who are challenged with meeting the needs of children of incarcerated parents.

Mentors and youth meet weekly for at least 1 year

The program offers emotional and practical support to assist mentors in continuing to meet with children of incarcerated parents during family transition times, such as when an incarcerated parent is released from prison.

Mentors work with parents, and support and honor the parent-child relationship

Training and support to mentors is provided in terms of building and maintaining trust with both caregivers and inmate parents.

Mentors receive ongoing training in culturally respectful and competent interactions with children of incarcerated parents as well as their caregivers and incarcerated parents.

Programs have clear policies and procedures that are well known to mentors regarding how to address the issue of parental incarceration with a child, based on caregiver and inmate wishes. Supervisors regularly discuss with mentors specific ways to appropriately honor the child’s relationship with his or her incarcerated parent.

Matches are monitored, supported, and reconsidered and changed if needed

Supervisors work with mentors of children of incarcerated parents to solve additional and unexpected workload issues that arise because of extenuating circumstances, such as helping transport a youth to a prison for parental visitation or adjusting to numerous changes in the youth’s living situation.

The mentoring program partners with families and other agencies to best engage the children of incarcerated parents and their families

Program administrators engage families with incarcerated parents in helping to shape program characteristics, procedures, and protocols.

The mentoring program partners with other agencies and programs to connect children and their families with other needed services and supports

Program staff members periodically engage with caregivers and incarcerated parents to determine the types of services and supports needed by the children and families served by the program.

Program administrators join and participate in community networks and taskforces that work on issues related to the families of adults involved with the criminal justice system.

Program administrators know what local agencies provide services needed by children of incarcerated parents and their families and make sure supervisors and mentors know not only how to connect children and families to these agencies but also how to advocate for service provision.

376  SPECIAL POPULATIONS African American, 5% Pacific Islander, and 2% Native American) than would be expected from local population demographics. CPY uses screening practices that are congruent with recommended practices. In addition, mentors must be at least 20 years old, and must complete a 6-hour prematch training before beginning a mentoring relationship. The training includes modules on the effects of incarceration on youth and families, culturally competent practice, and youth development and attachment theory. In a typical year, most volunteers are White, collegeeducated, financially stable individuals who do not have children or whose children are grown. Mentors and youth are matched based on individual characteristics, including preferences, residential proximity, ethnic and gender similarities, and corresponding interests and backgrounds. Male mentors are never matched with female youth. Families in which English is a second language (in all current cases, this is Spanish) are always matched with bilingual mentors. On average, 90% of enrolled youth remain in matches through the initial year, close to 40% continue for a second year with the same mentor, and 1% continue into a third year or beyond. Only about 20% of mentors choose to match with new youth after completing their initial commitment. The primary goals of the CPY program are to enhance youth academic performance, increase youth self-worth, improve youth relationships with family members, and enhance the youth’s connection to the community. Mentors and youth mutually commit to their relationship for at least 1 year, agreeing to weekly phone contact and at least three outings each month. To better meet the needs of youth affected by incarceration, several program changes were made and training components were added. A parent/caregiver component was added to increase contact between mentors and parents/caregivers and to connect families to community resource information and specific resources on incarceration. Families and mentors were invited to participate in a monthly dinner and activity night that includes an interactive educational presentation and facilitated discussion for caregivers. To supplement this work, outreach strategies were created to recruit new community partners involved with incarcerated parents and their families, including the state Department of Corrections and the court system. In addition, to keep in better touch with other key partners relevant to the children of incarcerated parents, CPY became an active member of a local Community Partner Network comprising governmental, nonprofit, educational, and justice agencies, as well as key local leaders and individuals affected by incarceration.

Case Illustration: Volunteer Mentoring. Mentor Justin and youth Max were matched through the CPY MCP program almost 14 months ago. At the time of the match, Justin was a 34-year-old single White male working in a financial investment firm. He chose to become a mentor for two primary reasons: to contribute to his community by spending time with someone who was not as fortunate as he had been in his childhood, and to get a better understanding of the current youth culture. Max was a 10-year-old White male who lived with his mother and three younger siblings. Max’s father was incarcerated in a state prison, and did not have any contact with Max. His father lived with the family prior to his incarceration over 3 years ago. In the mentor training, Justin learned about the potential impacts that incarceration can have on youth and their families, and thus, he had felt prepared to be supportive to Max regarding his father’s incarceration. However, after spending about 12 hours together each month for 10 months, Max still hadn’t talked much with Justin about his father. The comments Max had made seemed inconsequential, such as on his 11th birthday, when Max told Justin about a time when his dad gave him an old pocket knife for his 5th birthday, or on an outing, when Max noticed that a truck that drove by looked just like the one his father used to drive. At home, Max’s mother had been struggling to stay employed and keep the bills paid, often working multiple jobs and at odd hours. Frequently, Max’s mother would call Justin at the last minute to cancel or change their outing plans because of changes in her work schedule. In such cases, instead of participating in an outing, Max would end up being cared for at the home of a family friend or a next-door neighbor. Sometimes Max’s mother wouldn’t call, and Justin would arrive to find the house empty. Furthermore, Max’s mother’s phone had been disconnected multiple times. Because of his erratic home life, Max was often tired when he did go on an outing. Additionally, he was failing many of his classes because of missing assignments and frequent absences. About 3 months ago, Max’s mother became ill and got so far behind on her bills that she couldn’t care for Max and his siblings, so she sent them to live with his grandparents in a neighboring small town, about 15 minutes away. Since moving in with his grandparents, Max has had a more structured lifestyle, but seems more withdrawn. The move was intended to be temporary and his mother planned to visit the kids every weekend, but gradually she began canceling her weekend visits and not calling on a regular basis. Justin heard rumors from the

Children of Incarcerated Parents   377 grandparents that she had been using drugs, which had been an issue for her in the past. Initially, Max’s more stable home life enabled him to have more consistent outings with Justin. However, Max began to make inappropriate requests of Justin, such as asking to drive his car, and to talk about things that Justin found troubling, such as wanting guns and knives. Max also started to regularly tell stories about his dad, which usually focused on what an “awesome” person his dad was, and how Max wanted to grow up to be just like his dad. Although Max’s work at school improved initially following the move, he has recently started getting into trouble at school, being accused of picking fights with other kids and with teachers. Justin has begun to express concerns to CPY staff that his relationship with Max is weakening and that he wasn’t sure if he even matters to Max, such that if he stopped seeing Max tomorrow, Max might not even notice. Last week, the grandparents shared concerns about Max’s younger siblings beginning to act out as well and asked if they could be matched with mentors too. The case coordinator is currently working with Max, his grandparents, Justin, and the school to identify ways to strengthen Max and Justin’s relationship. They hope to build in more contact time by arranging opportunities for Justin to interact with Max at school, and are searching for extracurricular activities for the pair to engage in together, such as a karate class. CPY’s strong presence in the community is helping to bring these types of plans to fruition. CPY has active partnerships with a variety of local nonprofits, schools, and governmental agencies, and these partnerships assist mentors such as Justin in connecting youth and families to needed opportunities and services. CPY staff members continue to encourage Justin in his work, reminding him of the positive impact he is making on Max’s life simply by being a consistent, trusted friend; by providing fun ways for Max to connect to his community; and by building Max’s self-esteem and hope for the future. Justin recently reflected that without the training, support, and encouragement he has received from CPY staff, he would not have been able to navigate the challenges he has faced in mentoring Max. Professional Mentoring. Friends of the Children (FOTC), a nonprofit mentoring organization established in 1993, also focuses on youth who are exhibiting behavior problems at an early age and are living in “high risk” neighborhoods and with difficult family circumstances (Campbell, 2003). The FOTC program is delivered through five independent service

delivery “chapters” around the United States that are supported by a national chapter. The program begins when children are in kindergarten or 1st grade, and provides a caring relationship with a paid, supervised, professional mentor until high school graduation. Mentors receive extensive training before working with youth, and then receive ongoing continuing education on key topics, including culturally competent mentoring and working with children of incarcerated parents. Each FOTC chapter partners with local public elementary schools situated in high-poverty neighborhoods. Partner schools tend to have elevated percentages of students receiving free or reducedprice lunch, low educational achievement as indicated by 3rd-grade benchmark test scores in reading and math, and few resources and limited community support for educating disproportionate numbers of students who have high needs. Within these schools, a 6-week selection process is conducted (usually in the spring) to determine which children are most in need of a mentor. FOTC invites into the program children who are demonstrating adjustment problems (e.g., acting out, aggressive, socially withdrawn, depressed) and are currently living in high-risk (i.e., numerous obstacles to success, such as having an incarcerated parent) and low-protection (i.e., limited support for overcoming such obstacles) environments. By design, the program enrolls equal numbers of boys and girls. The chapter where the program was developed, FOTC-Portland, served 378 youth during the 2011 program year (51% girls) in kindergarten through 12th grade. Each youth is paired with a same-sex mentor (called a Friend), a full-time program employee who works with 8–12 youth, depending on the age of the youth. Friends must have a minimum of 2 years of experience working with youth prior to hire and a minimum of an associate’s degree. Currently, 87% of Friends have bachelor’s degrees, 50% are men, and 39% are racial and/or ethnic minority. Friends make an initial 3-year commitment to serve as mentors, and many stay on for much longer periods of service. Reflecting the demographics of the neighborhoods in which the program operates, youth come from a variety of racial and ethnic backgrounds, including African American (45%), Caucasian (26%), multiracial/other (15%), and Latino (14%). The typical youth lives in the context of multiple risk factors. For example, of current youth in the program, approximately 59% have at least one parent with a history of incarceration, 69% have at least one parent with a history of substance abuse, and 65% have been affected by domestic violence

378  SPECIAL POPULATIONS in their lifetimes. Compounding these risks, 58% of youth have a parent who was under 18 years of age when their first child was born, 54% have at least one parent who did not complete high school, and 39% have had a report of abuse or neglect filed on their behalf (e.g., see NPC Research, 2009). The intermediate goals for all youth involved in the FOTC program are to demonstrate positive social and emotional development, to make good choices, to succeed in school, to have a positive plan for the future and to have the skills needed to carry out that plan, and to have adequate health care. The long-term goals for youth in the program are graduating from high school or receiving a GED, avoiding early parenthood, and avoiding involvement with the juvenile justice system. In addition, each youth has unique goals relevant to his or her abilities, interests, and life situation. Friends support youth in meeting these goals through in-school visits and outings in the community as well as through advocacy work in the educational, child welfare, and physical/mental health-care systems to help meet specific youth needs. Friends are supposed to spend an average of 4 hours a week with each youth, and to work to intentionally promote the development of academic, social, emotional, life, and career skills. During elementary school, the majority of time spent with the youth is in one-on-one activities. During adolescence, these experiences continue, but most activities are group based, involving other FOTC youth and their Friends. Friends continue to work with youth as long as they live within a 30-mile radius of the FOTC home office (e.g., in Portland, this encompasses the entire metropolitan area). Case Illustration: Professional Mentoring. Tina entered FOTC as a kindergartner and has been matched with Friend Marianne for the past 6 years. Both are African American. During the kindergarten classroom visits conducted during the initial selection process, Tina exhibited poor work habits, social withdrawal, and signs of depression. She was absent from school on several of the days when Friends were observing students, and Tina’s teacher confirmed that this was typical. When she did come to school, she frequently had not eaten breakfast. During class, her teacher reported that Tina tended to be disruptive and to need frequent redirection. Of further concern, Tina’s mother, her sole caregiver, had been incarcerated since Tina was 4 years old. Tina has lived with her grandmother and her two older brothers since that time, and the family has been living in impoverished conditions. Tina was one of eight girls who Marianne began to mentor 6 years ago. Marianne joined

FOTC after working for 2 years in an after-school program with children from low-income families. She was interested in the program not only because she enjoyed working with children but also because of the opportunity FOTC provided to do so through long-term, intensive relationships with a few children. Marianne views being a Friend as an opportunity to “give back” to her community through being a caring and consistent adult to children who have little family support to overcome the challenges in their lives. When Tina and Marianne first met, Tina would not talk about her incarcerated mother and did not show any positive emotion or warmth for Marianne. Tina was reluctant to show that she looked forward to going on outings with her Friend and would frequently pretend that she did not see Marianne when she visited Tina’s school. This was difficult for Marianne. However, she kept trying to connect. Marianne had been trained to expect new relationships to go through a period of “testing.” Tina was clearly testing to see if Marianne was going to come and leave the way that many adults had done in her life. In a required continuing education seminar focused on working with the children of incarcerated parents, Marianne learned about the grief process that Tina was likely to be experiencing. She figured that Tina might be at a stage in that process in which she was particularly reluctant to form bonds with other adults. This knowledge provided additional comfort that perhaps things would eventually change in their relationship. Subsequently, Marianne addressed Tina’s hesitancy to develop a strong relationship with her by using key strategies of the FOTC program. First, she sought to be consistent and predictable, ensuring that she made contact with Tina twice a week and always followed through with what she and Tina agreed to do. Second, she strove to be authentic, always being open, honest, and genuine with Tina during their interactions. Third, she tried to be patient and persistent, and to keep in mind that building a relationship with a child takes time and effort, particularly when the child’s relationship with the primary caregiver has been disrupted by something like incarceration. After about a year of using these strategies, Tina began to show a few signs of attachment to her Friend, for example, by greeting Marianne as she walked into her classroom, or by saying, “I can’t wait to see you next week,” now and then when they concluded an outing together. Behaviors like these indicated that Tina had begun to bond with Marianne. As an employee at FOTC, Marianne attended weekly team meetings with six other Friends to

Children of Incarcerated Parents   379 discuss current challenges, successes, and frustrations in mentoring work with children. Team meetings are facilitated by a supervisor, called a Team Leader, who guides the meeting and provides support, celebrates achievements, reflects on current practices, and encourages group brainstorming on new approaches to challenges. As Marianne frequently spoke of feeling rejected by Tina, her teammates kept reminding her to “stick with it,” and assured her that eventually she would develop a strong relationship with Tina that would lead to a lasting impact in Tina’s life. Tina’s mother was incarcerated about a halfhour drive from her home. Phone calls were infrequent, but she stayed in contact through occasional letters, and Tina’s aunt would take Tina to visit her mother in person about once a month. As the relationship between Marianne and Tina deepened, Tina began to share little things about her mother, such as her favorite movies and foods. Behaviors such as this indicated to Marianne that Tina was getting ready for their relationship to include a deeper consideration of Tina’s experiences and feelings about her mother. Taking Tina’s lead, Marianne began to insert brief activities into their time together that gave Tina the opportunity to express what she was feeling without asking her directly. For example, Marianne asked Tina to draw a picture of her family, and then asked her to identify each person and say one thing that she loved about each of them. With small attempts such as this, over the next several outings, Tina began to open up more about her current feelings. This work gave Tina the room she needed to feel comfortable talking with Marianne, and she now openly discusses memories or current interactions she has with her mother. Tina’s home situation remains chaotic today. Her mother returned from prison, but was unable to regain custody of Tina and her brothers, and eventually had her visitation rights revoked. During this same period of time, Tina’s grandmother had major health problems and was no longer able to care for Tina and her brothers. For the past several years, Tina has lived with her aunt, who has had trouble holding a job and has had financial troubles. Tina, although very bright, continues to struggle in school. Her difficulties at school have been compounded by frequent moves. As a result, she has attended four different schools in the past 2 years, diminishing her abilities to form healthy relationships with peers and to have a consistent relationship with a teacher. Marianne has stayed with Tina through her caregiver changes and various moves. She has provided specific supports for Tina’s transitions by

arranging outings that acquaint Tina with her new neighborhood, school, and teachers. Further, with support from her Team Leader and teammates, Marianne has provided Tina with opportunities for prosocial involvement in her school and local community, and sought out ways for Tina to learn the skills necessary to be successful in such involvement. Throughout the changes, she has created opportunities for Tina to continually foster her individual strengths and interests. For example, Tina likes singing and running, and with each move, Marianne has linked up Tina with local choirs and track clubs, and has helped her get to practices each week. Through FOTC group activities, Tina has received recognition for her involvement and successes in these endeavors. Conclusions and Recommendations To date, there is consensus in the practitioner literature on recommended practices for mentoring programs for the children of incarcerated parents. These practices are congruent with other general statements on recommended practices in mentoring (e.g., DuBois et al., 2002). To further the quality of work in the field, mentoring programs not only should attempt to use the practices that are thought to be best, but also should document their use and their impact on youth outcomes. Along these lines, new program innovations that enhance or expand on these recommended practices are likely being tried in mentoring the children of incarcerated parents each day. Mechanisms are needed to document these innovations, to find out whether they work, and to share them with others. To assist in documentation, affordable, reliable, and user-friendly database software is needed that mentoring programs can easily adapt and that would enable disparate agencies to track and appropriately share important information, such as the characteristics of mentors and youth, the characteristics of mentoryouth interaction, practice innovations, and youth outcomes. While most mentoring programs for children of incarcerated parents have been relatively short term (up to 1 year) and utilized volunteers, programs that last for longer periods and employ practices to keep mentors engaged over the long run seem worthy of consideration for addressing some of the complex needs of some of the members of this population. However, this population has some needs that mentoring is not well suited to address. Partnerships are needed between mentoring agencies and other agencies that serve children and families to help meet these needs. Partnerships are also needed among practitioners

380  SPECIAL POPULATIONS and researchers not only to assist in the design and conduct of evaluations that can be useful to programs seeking to obtain grant and donor funding, but also to contribute rigorous scientific information that could be used to improve program effectiveness. Of particular interest would be the testing of hypotheses across multiple programs, a task that would be immensely easier if programs shared common protocols and databases. Finally, practition­ ­­ers, as much as researchers, need to find efficient and effective ways to publicly share and widely disseminate information on their programs, program outcomes, and important lessons learned.

Conclusion The children of incarcerated parents represent a growing population that is at risk for (but not guaranteed to have) childhood antisocial behavior problems and that may be at risk for a variety of other concurrent and future problems. Due to a convergence of interests, mentoring has rapidly become the preferred intervention for these children. Scholarship focused on mentoring this population is only just beginning. Rhodes’s (2002) theory of how mentoring influences youth is a potentially useful framework for future investigations, particularly if considered within the context of the broader life circumstances of the child. Parental incarceration can have far-reaching impacts on a child and family, but whether and how incarceration affects a child depends on preincarceration factors, as well as the extent to which caregivers, inmates, and other adults nurture and care for the child during the incarceration. Unfortunately, little is known about how mentoring affects the children of incarcerated parents in the short run, and almost nothing is known about whether mentoring decreases their risk for developing antisocial behaviors or other problems over the long run. In sum, the impacts of mentoring on outcomes for this population are unclear. Probably the most intriguing finding to date regarding the children of incarcerated parents is that wide variability exists in their circumstances and in their needs for intervention. Many children of incarcerated parents do live in complex and challenging situations. Others do not. Given this finding, intervention seems likely to be most effective if tailored to the specific needs of a particular child and his or her family. Funding agencies are beginning to recognize the importance of such tailoring. For example, after 7 years of funding MCP programs, the Administration for Children and Families adjusted

its funding announcement for fiscal year 2010 to better reflect the complex contextual factors that may exist for a given child. To be eligible for funding, programs had to demonstrate the capacity to evaluate and monitor the needs of each of the match partners (i.e., mentor, mentee, and caregiver) and respond appropriately, including linking a partner to services external to the agency. Although the MCP no longer exists, federal recognition of the complexity of the lives of families that include incarcerated parents was significant. It is hoped that future efforts regarding mentoring the children of incarcerated parents will take such a broad point of view, in which mentoring becomes one of a set of interventions that can be used to meet the needs of specific children of incarcerated parents and their families. Within such multimodal interventions, issues related to the race, ethnicity, and culture of children and families involved must be considered. Despite the disparities in terms of the numbers of minority children and families affected by incarceration, the existing literature on mentoring the children of incarcerated parents is unclear as to the implications of such disparities for decisions that are made about the characteristics of programs. In this and other areas, it is time to conduct well-designed research studies that help practitioners understand the conditions under which mentoring is helpful, for which children, and in what ways.

References Adalist-Estrin, A. (2004). Mentoring children of prisoners. Palmyra, VA: Family & Corrections Network. Ainsworth, M. D. S. (1989). Attachments beyond infancy. American Psychologist, 44, 709–716. Andrade, N., Fett, R. O., Fitzhugh, S., Green, C., Lockett, J., Marquardt, S., . . . Lawton, L. (2008). Program evaluation of OreMi: Mentoring children with incarcerated parents. Unpublished report prepared by students in Sociology 190.7: Community Based Research. University of California, Berkeley. Bernstein, N. (2007). All alone in the world: Children of the incarcerated. New York: New Press. Bilchik, S. (2006). Mentoring: A promising intervention for children of prisoners. In J. E. Rhodes (Ed.), Research in Action Series (Issue 10, pp. 1–28). Alexandria, VA: MENTOR. Retrieved from http:// www.mentoring.org/downloads/mentoring_391.pdf Bocknek, E. L., Sanderson, J., & Britner IV, P. A. (2009). Ambiguous loss and posttraumatic stress in schoolage children of prisoners. Journal of Child and Family Studies, 18, 323–333. Bowlby, J. (1973). Attachment and loss, volume II: Separation: Anxiety and anger. New York: Basic Books.

Children of Incarcerated Parents   381 Bowlby, J. (1979). The making and breaking of affectional bonds. London: Tavistock. Bruster, B. E., & Foreman, K. (2012). Mentoring children of prisoners: Program evaluation. Social Work in Public Health, 27, 3–11. Campbell, D. (2003). Friends of the Children: An innovative mentoring program for seriously at-risk children. In K. W. McClusky & A. M. Mays (Eds.) Mentoring for talent development (pp. 143–161). Sioux Falls, SD: Reclaiming Youth International. Carlson, E. A., & Sroufe, L. A. (1995). Contribution of attachment theory to developmental psychopathology. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathology, Vol. 1: Theory and methods (pp. 581–617). Oxford, UK: Wiley. Davies, E., Brazzell, D., La Vigne, N. G., & Shollenberger, T. (2008). Understanding the experiences and needs of children of incarcerated parents: Views from mentors. Washington, DC: Urban Institute Justice Policy Center. Deklyen, M., & Greenberg, M. T. (2008). Attachment and psychopathology in childhood. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 637–665). New York: Guilford Press. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Silverthorn, N. (2005). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92. Eddy, J. M., Kjellstrand, J. M., Martinez, C. R., Newton, R. (2010). Theory-based multimodal parenting intervention for incarcerated parents and their families. In J. M. Eddy & J. Poehlmann (Eds.), Children of incarcerated parents: A handbook for researchers and practitioners (pp. 237–261). Washington, DC: Urban Institute Press. Eddy, J. M., & Poehlmann, J. (Eds.) (2010). Children of incarcerated parents: A handbook for researchers and practitioners. Washington, DC: Urban Institute Press. Eddy, J. M., & Reid, J. B. (2003). The adolescent children of incarcerated parents. In J. Travis & M. Waul (Eds.), Prisoners once removed: The impact of incarceration and reentry on children, families and communities (pp. 233–258). Washington, DC: Urban Institute Press. Elder, G. H. (1998). The life course as developmental theory. Child Development 69, 1–12. FASTEN. (2004). People of faith mentoring children of promise: A model partnership based on service and community. Retrieved from http://www.fastennetwork .org/Uploads/2C16CC3C-DB30-4C26-9D54-0596 C409E1AA.pdf Freedman, M. (1992). The kindness of strangers: Reflections on the mentoring movement. Philadelphia: Public/Private Ventures.

Gabel, K., & Johnston, D. (Eds.). (1998). Children of incarcerated parents. New York: Lexington Books. Glaze, L. E., & Maruschak, L. M. (2009). Parents in prison and their minor children. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice. Goode, W. W., & Smith, T. J. (2005). Building from the ground up: Creating effective programs to mentor children of prisoners. Philadelphia: Public/Private Ventures. Hamilton, S. F., & Hamilton, M. A. (1992). Mentoring programs: Promise and paradox. Phi Delta Kappan, 73, 546–550. Husock, H. (2003). Starting Amachi: The elements and operation of a volunteer-based social program (No. C16-03-1710.0). Cambridge, MA: John F. Kennedy School of Government, Harvard University. ICF International. (2011). Mentoring children affected by incarceration: An evaluation of the Texas Amachi Program. Washington, DC: Office of Juvenile Justice and Delinquency Prevention. Jucovy, L. (2003). Amachi: Mentoring children of prisoners in Philadelphia. Philadelphia: Public/Private Ventures and the Center for Research on Religion and Urban Civil Society. Retrieved from http://www .ppv.org/ppv/publications/assets/21_publication.pdf Laakso, J., & Nygaard, J. (2012). Children of incarcerated parents: How a mentoring program can make a difference. Social Work in Public Health, 27, 12–28. LEARNS. (2004). Mentoring children of incarcerated parents: A toolkit for Senior Corps Directors. Portland, OR: Northwest Regional Educational Laboratory. http://www.nationalserviceresources.org/files/legacy/ filemanager/download/learns/MCIP_Senior_Toolkit .pdf Lipsey, M. W., & Derzon, J. H. (1998). Predictors of violent or serious delinquency in adolescence and early adulthood: A synthesis of longitudinal research. In R. Loeber & D. Farrington (Eds.), Serious and violent juvenile offenders: Risk factors and successful interventions (pp. 86–105). Thousand Oaks, CA: Sage. Maruschak, L. M., Glaze, L. E., & Mumola, C. J. (2010). Incarcerated parents and their children: Findings from the Bureau of Justice Statistics. In J. M. Eddy & J. Poehlmann (Eds.) Children of incarcerated parents: A handbook for researchers and practitioners (pp. 33–54). Washington, DC: Urban Institute Press. Maton, K. I., Domingo, M. R., & King, J. (2005). Faithbased organizations. In D. L. DuBois, & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 376–391). Thousand Oaks, CA: Sage. Minuchin, P. (2002). Looking toward the horizon: Present and future in the study of family systems. In J. P. McHale & W. S. Grolnick (Eds.), Retrospect and prospect in the psychological study of families (pp. 259–278). Mahwah, NJ: Erlbaum. Murray, J., Farrington, D. P., & Sekol, I. (2012). Children’s antisocial behavior, mental health, drug use, and educational performance after parental incarceration:

382  SPECIAL POPULATIONS A systematic review and meta-analysis. Psychological Bulletin, 138, 175–210. Murray, J., Loeber, R., & Pardini, D. (2012). Parental involvement in the criminal justice system and the development of youth theft, depression, marijuana use, and poor educational performance. Criminology, 50, 255–302. National Mentoring Working Group. (1991). Mentoring: Elements of effective practice. Washington, DC: National Mentoring Partnership. NPC Research. (2009). Annual report on the Friends of the Children program. Portland, OR: Author. Parke, R. G., & Clarke-Stewart, K. A. (2003). The effects of parental incarceration on children: Perspectives, promises, and policies. In J. Travis & M. Waul (Eds.), Prisoners once removed: The impact of incarceration and reentry on children, families, and communities (pp. 189–232). Washington, DC: Urban Institute Press. Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castalia. Patterson, G. R., Reid, J. B., & Dishion, T. J. (1992). A social interactional approach: Antisocial boys: Vol. 4. Eugene, OR: Castilia. Poehlmann, J., & Eddy, J. M. (Eds.) (in press). Risk and resilience: Children of incarcerated parents. Monographs of the Society for Research in Child Development. Reid, J. B., Patterson, G. R., & Snyder, J. (Eds.) (2002). Antisocial behavior in children and adolescents: A developmental analysis and model for intervention. Washington, DC: American Psychological Association. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E., Haight, W. L., & Briggs, E. C. (1999). The influence of mentoring on the peer relationships of foster youth in relative and nonrelative care. Journal of Research on Adolescence, 9, 185–201. Rutter, M., & Sroufe, L. A. (2000). Developmental psychopathology: Concepts and challenges. Development and Psychopathology, 12, 265–296. Saito, R. N., & Blyth, D. A. (1992). Understanding mentoring relationships. Minneapolis, MN: Search Institute. Sameroff, A. J., Bartko, W. T., Baldwin, A., Baldwin, C., & Seifer, R. (1998). Family and social influences on the development of child competence. In M. Lewis

& C. Feiring (Eds.), Families, risk, and competence (pp. 161–185). Mahwah, NJ: Erlbaum. Shlafer, R. J., & Poehlmann, J. (2010). Attachment and caregiving relationships in families affected by parental incarceration. Attachment & Human Develop­ ­ment, 12, 395–415. Shlafer, R. J., Poehlmann, J., Coffino, B., & Hanneman, A. (2009). Mentoring children with incarcerated parents: Implications for research, practice and policy. Family Relations, 58, 507–519. Sipe, C. L. (2005). Toward a typology of mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 65–80). Thousand Oaks, CA: Sage. Tierney, J. P., & Grossman, J. B. (1995). Making a difference: An impact study of Big Brothers/Big Sisters. Philadelphia: Public/Private Ventures. Travis, J., & Waul, M. (2003). Prisoners once removed: The impact of incarceration and reentry on children, families, and communities. Washington, DC: Urban Institute Press. van Nijnatten, C. (1997). Children in front of the bars. International Journal of Offender Therapy and Comparative Criminology, 41, 45–52. Weissman, M., & LaRue, C. M. (1998). Earning trust from youths with none to spare. Child Welfare League of America, 77, 579–594. Werner, E. E. (1989). High-risk children in young adulthood: A longitudinal study from birth to 32 years. American Journal of Orthopsychiatry, 59, 72–81. West, H., & Sabol, W. (2008). Prisoners in 2007. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. Western, B. (2006). Punishment and inequality in America. New York: Russell Sage Foundation. Western, B., & Wildeman, C. (2009). The Black family and mass incarceration. Annals of the American Academy of Political and Social Science, 621, 221–242. Wildeman, C. (2010). Mass imprisonment, social policy, and the future of American inequality. In J. M. Eddy & J. Poehlmann (Eds.), Children of incarcerated parents: A handbook for researchers and practitioners (pp. 303–318). Washington, DC: Urban Institute Press. Zimmerman, M. A., Bingenheimer, J. B., & Notaro, P. C. (2002). Natural mentors and adolescent resiliency: A study with urban youth. American Journal of Community Psychology, 30, 221–243.

PART VII Practice and Programmatic Considerations

26  PROGRAM FUNDING Susan G. Weinberger

Introduction Managers of youth mentoring programs are required to balance multiple tasks in order to ensure that their programs run effectively and efficiently. Among the priorities are recruiting, screening, and training mentors and mentees; providing ongoing staff support to the matches; designing marketing and public relations tools to create program awareness; and conducting yearly outcomes evaluations. As if these responsibilities were not enough, from the very first day of a mentoring initiative, those responsible for supporting its day-to-day operations and sustainability must begin the process of planning to fund the program adequately for the future. Even if a mentoring program has been lucky enough to receive federal, state, or local grants, dollars awarded have beginning and ending dates. If the grant is on a 3-year cycle, for example, planning for beyond the grant period must begin early in the first year of the award. The focus of this chapter is on assisting practitioners throughout the mentoring field with raising funds to support their programs and initiatives. The terms program funding and fundraising are used interchangeably throughout this chapter. So too are the terms agency and organization with nonprofit status. The broad and varied terminology used in referring to practitioners, furthermore, is intended to reflect the reality that significant responsibility for program funding in any given mentoring organization may fall variously under the auspices of staff who have titles such as manager, coordinator, executive director, and with relevance to larger programs with greater budget capability, a chief development officer who has direct responsibility for carrying out the program’s fundraising efforts.

Likewise, references to a program’s board of directors should be taken as synonymous, for present purposes, with titles such as board of trustees, board of governors, advisory council, and the like. The overarching aim of this chapter, which is organized into two major sections, is to provide practitioners with helpful guidelines for the development and implementation of a fundraising plan for their mentoring programs. The first section discusses internal planning. The second section examines the key program sources of funding and how each one can be embraced by a program in order to seek multiple levels of financial support. The chapter also contains two tables with key questions that may guide the application of the concepts and techniques described here. The chapter concludes with a brief synthesis. It also includes a call for future study that could help to establish a stronger evidence-based foundation for fundraising in youth mentoring.

Developing and Implementing a Fundraising Plan Internal Planning Assessing Fundraising Capabilities. Fundraising is generally defined as the solicitation of voluntary financial support for a specific cause. It is widely undertaken by charitable organizations, educational institutions, and political groups to acquire sufficient funds to support their activities. To ensure adequate funding for a mentoring program’s long-term sustainability, practitioners have to engage in multiple activities. The cost of running a youth mentoring program, coupled with budget shortfalls in many states (Lav & 385

386   Practice and Programmatic Considerations McNichol, 2009), has made it increasingly difficult for program leaders to access sufficient resources. In my experience, the cost to run a mentoring program is typically figured based on the number of matches. This can range from $500 per mentormentee match yearly to as much as $1,500 per match depending on such considerations as office expenses, program location (site based or community based), and level of staff support. Based on a survey of 1,000 mentoring organizations, more than half (53.8%) of the programs responding were concerned they will have to shrink in size or end their operations because of a lack of funding (MENTOR, 2009). Programs that place a priority on developing and implementing a comprehensive fundraising plan will reap the benefits of long-term sustainability. When considering program viability, the focus is on building mentoring programs that are alive and organizationally healthy so that they can serve youth over the long term. Successful funding of programs and a healthy organization go hand in hand (Webster, 1999). Securing Nonprofit Status. A first step in planning to fund a mentoring program is to ensure that the program has secured appropriate nonprofit status. Forming a nonprofit corporation and applying for tax-exempt status under Internal Revenue Code section 501(c)(3) is a requirement before beginning any fundraising activities. It is not particularly difficult to complete the required forms, but the program manager may wish to call on a tax advisor, ideally one who serves on the agency’s board of directors and can offer pro-bono assistance. Before making donations to agencies, many corporations, foundations, and individuals require proof that the agency has nonprofit status and is able to provide a receipt for tax-deduction purposes. By their very nature, public schools in the United States have tax-exempt status but they are not 501(c)(3) entities. Therefore, their exempt status is often not sufficient for donors who may seek a tax deduction for their gifts to a mentoring program. Even if a mentoring program is connected to or operated by a public school district, it is a good idea for the program to seek its own 501(c)(3) status in order to satisfy potential donors. Some programs find it more convenient to partner with agencies that already have nonprofit status. For step-by-step instructions on incorporating a nonprofit and applying for tax-exempt status, see Wrobel (2010), on how to start a 501(c)(3) nonprofit organization, or Mancuso (2009), on how to form a nonprofit corporation. Internal Revenue Service publications on related tax rules can be downloaded from www.irs.gov.

To apply for nonprofit status, a mentoring program must have formulated a mission statement, formed a board of directors, filed articles of incorporation, drafted bylaws, developed a budget, and put in place a record-keeping and accounting system. In addition, most states require such programs to register with their state’s attorney general before soliciting funds and reporting on them. The details may take a considerable amount of effort and time. This is also the best time to assess how the agency is doing. The mentoring program must have not only a clear, written mission statement but also defined goals and objectives, successful outcomes, and a financial statement of health. Potential funding sources want to support agencies that demonstrate strong signs of stability before they invest. These materials must be in order before an organization asks for money to sustain its efforts. It is also the time to examine the program’s current sources of funding revenues. Multiple and diverse funding sources are indicators of the agency’s health. Consideration must be given to money being raised that can be relied on year after year. The more financial supporters an organization has, the more stable it is. It is generally agreed that no more than 30% of an organization’s revenue should come from a single source, and even less is better. A thousand people donating $10 each is healthier for a mentoring program than a single $10,000 grant (Webster, 1999). Involving the Board of Directors Most mentoring programs do not have the luxury of having a paid fund development professional on staff. This responsibility rests with the program manager, but the leadership function does not rest with the manager alone. Managers must reach out to members of their board of directors and invite other volunteers connected to the organization to join in the planning for a successful fundraising effort. Managers should allocate adequate time to support the planning process (Fulop, 2005). The focus should be on creating a team capable of sharing ownership and responsibility for carrying out the fundraising effort, based on their expertise, reputations, and relationships (Kotter, 1996). By law, every nonprofit must have a board of directors. This is usually a volunteer group of about 12–15 people, whose responsibilities include overseeing the organization and being accountable for its compliance with legal and other requirements (Bray, 2005). To build and expand an effective nonprofit board, members should be completely

Program Funding  387 comfortable with their duties and responsibilities. These include determining the mission, goals, and objectives of the organization; selecting the executive director and assessing that person’s performance; monitoring the agency’s programs; ensuring that there are adequate resources and legal and ethical integrity; and assessing the board’s own performance. A board serves four purposes: (a) It contributes to the agency’s annual campaign, via individual members’ personal financial support, (b) it opens doors to recruit mentors among members’ colleagues and contacts, (c) it provides guidance and advice to the program, and (d) it assists with all fundraising efforts by identifying potential sources and by arranging for members to attend in-person meetings with their contacts along with the program manager. Some managers think that asking board members for donations seems odd. After all, they are already volunteering their time to the organization. But board members will set an example and be much more successful at soliciting large gifts from others if they themselves give. It demonstrates their commitment to the cause and confidence that every donation will be well spent (Bray, 2005). Typically a corporation or foundation representative will ask the mentoring program manager, “What percentage of your board of directors gives to the organization and its mission?” The answer must be 100%. Gifts made should be at a level that is comfortable for the individual board member, but all members are expected to give. A major consideration in forming an effective board is the diversity of its members. Many experts in board development describe the right balance of a board as the three Ts, which stand for time, talent, and treasure. Time refers to members’ commitment to give of their own time to open doors using their contacts in the community and to roll up their sleeves. They provide talent to help the program succeed and offer their own professional expertise. Finally, treasure refers to members’ ability and willingness to support the program financially. Many years ago, I suggested using the three Ws (wealth, wisdom, and workers). Regardless of whether you use the three Ts or the three Ws, thinking in these terms will help you to balance your board properly. Board member representation should have the experience and contacts to assist with fundraising efforts. Potential board members could include the following: business representatives; community organizers; financier-accountants (banking, trusts, investments); fundraisers/developers; grant writers; endowment/ foundation representatives; members of the media; special events planners; insurance agents; attorneys;

marketing/public relations/communications experts; human resources directors; politicians (elected officials who are respected in the community, who are popular, and who will be an advocate for mentoring when state and federal grants are available); real estate agents; educators; members of the faith-based community; members of the arts; technology/web designers; philanthropists; and physicians. Multiple members from one group are not as effective as one chosen from each of several categories. Creating a Fundraising Plan The time to develop a comprehensive fundraising plan for a mentoring program is not when current funding is drying up and/or the program is in its final year of a mentoring grant. Plan development should not be a solution to a crisis, but rather a comprehensive effort over 365 days of the year. Mentoring managers who have vision, passion, and leadership skills can direct a successful planning process and secure program funding. They must believe in the cause for which they are raising money; they must maintain that belief during defeats, tedious tasks, and times of financial insecurity; and they must possess high hopes and low expectations and faith in the goodness of people (Klein, 2004). Resource development planning should set aside adequate time to accomplish the effort and to solicit support from the board and volunteers. Identifying board members and other individuals in the community who have expertise in sustainability planning is key. Although professional fundraising consultants exist, most mentoring programs have to rely on volunteers because budgets virtually never include funds to hire consultants. Throughout the planning process, partnering with the media will allow the community to learn about the agency’s efforts and be ready to assist. Having a member of the media as part of the program’s board of directors will ensure achievement of this result. For instance, the media representative may be willing to report on an ongoing basis on the program’s progress toward its goal. The resource planning process includes conducting both an internal and an external assessment of the organization. Fulop (2005) provides a 10-step planning process, starting with allocating resources and assembling a planning team to developing and implementing an action plan. This process will ensure that planning to fund a mentoring program will include all of the required elements for a successful initiative.

388   Practice and Programmatic Considerations In addition to the development plan, consideration should be given to writing a code of ethics related to fundraising activities. Fundraising practices should reflect the beliefs and values of the mentoring program. I believe it is both unadvisable and unacceptable to consider shifting a program’s mission to satisfy a potential donor or to accept donations from a company that produces goods and services that may be illegal or potentially dangerous for youth. These views can be stated as guidelines to follow in your program. Table 26.1 provides a checklist of guiding questions to help managers plan to launch a successful fundraising effort for their mentoring program (see also Rhodes et al., this volume, Chapter 35). Once the planning process is in place, it is time for the planning team to appoint

a chair to lead the fundraising effort and for team members to receive their individual assignments and assist the chair with the initiative.

Key Sources of Program Funding To ensure the sustainability of mentoring programs, multiple funding sources must be considered. In this section, I outline the major categories. Fundraising from major donors and trusts are the most cost-effective methods. Some of the fundraising methods that seem to be the most popular are not the ones that get the best returns (Charity Buzz, 2009). This is especially the case with direct mail. Compared to some of the other methods, direct mail

Table 26.1   Checklist for Practitioners: Planning Phase of Funding a Mentoring Program Does the agency currently have 501(c)(3) tax-exempt status? If not, who is an expert on your board or in the community that you could rely on to assist in completing the paperwork to achieve that status? If yours is a school-based program that is operated totally or in part by the school district, how much of a challenge will it be to convince them that the district’s public education tax-exempt status is not sufficient for fundraising purposes and that your program needs to seek its own 501(c)(3) status? What are the requirements of the state in which your agency operates regarding soliciting funds and reporting on fundraising expenditures and revenues? Does your mentoring program have a written mission statement, bylaws, record-keeping and accounting systems, and a board of directors in place? Does the program have defined goals and objectives, successful outcomes, and a strong financial statement of health? How much of the revenue coming in to the agency has conditions attached to how it is spent? How much of the revenue is unrestricted? Do you receive funding from many sources rather than just one such as a federal grant? How much money can you rely on as secured funding year after year? Does the board of directors represent a broad cross-section of community members with expertise and contacts? Has each member indicated how he or she can assist in the fundraising process, when, and with whom? Do you have any members of the board who do not have term limits (e.g., any who have served for 20 years)? How will you now bring on new members and retire those who are not doing the job required of an effective board member? If 100% of the members on your board do not give yearly to the program, how will you change this? If you do not like to ask your board for money, do you have a board chair who can take on this important assignment after making his or her own donation? Are you able to identify a team of volunteers and board members who have the time, interest, and expertise to assist the program manager with the agency’s fundraising efforts? Have you conducted an internal and an external assessment of your program before examining potential funding streams?

Program Funding  389 has comparatively low returns per dollar of investment. Consider carefully which sources yield the best results. Institutions Government Funding. Mentoring programs will be one step ahead if they have staff, board members, or volunteers who are familiar with how to seek government funding and keep a vigilant eye on available grants. Funding is available on a frequent basis through federal, state, and local governments. There are formula and block grants that both states and local communities and school districts can receive and spend. These usually meet certain program goals. There are also discretionary and project grants that fund program activities and are usually awarded through a competitive process. The best way for a mentoring program to know when funding is available from the federal government is by monitoring the Federal Register website, which lists requests for proposals and their submission deadlines in many formats (e.g., print, web, PDF). A state’s discretionary grant funding programs can usually be accessed on each state agency’s website. Before applying for funding from a government agency, whether local, state, or federal, make sure that your mentoring program has the capacity to track and respond to proposal announcements in a timely manner. Systems must be in place to meet the reporting requirements. Statistics are hard to come by, but nonprofits nationwide have been estimated to get about 12% of their revenues from the government (Bray, 2005). Government grants can be unreliable, because they are often affected by the political goals of the current administration. Applications can be intimidating to read and understand and are often announced with a short turnaround time for completing and submitting the application to the appropriate agency. Recipients of federal grants are required to attend annual trainings with other grantees and learn how to complete the agency’s data collection and reporting requirements. Nevertheless, receiving a local, state, or federal mentoring grant is a prestigious accomplishment. Receiving a grant awarded through a peer review process demonstrates that you beat out many others in a competitive process. This, in turn, often provides leverage for receiving additional funding in the future. Staff and assigned volunteers should read every program announcement and determine whether the grant aligns with your program’s mission. Programs should never compromise on their

purposes because of grantor requirements. When completing a local, state, or, most important, federal Request for Proposal, do not deviate from the application criteria, be sure to address each section, and follow guidelines carefully. Keep in mind that an outstanding proposal was disqualified recently because it exceeded the length requirement by only one paragraph! Pay close attention to every question, provide details in the budget section, and never exceed the stated length limit for the application. Many federal grant applications require a logic model that describes how program activities will lead to intended outcomes. Agencies that can clearly define how their successes will be evaluated and achieved—and how the information to reach those results will be collected and analyzed—will present a strong proposal to the government. Each proposal must diligently address all of the outlined criteria in the Request for Proposal. It is a good idea to have someone outside of your mentoring program read the proposal and provide objective input before you submit it. Corporate Giving. Many large corporations grant money to nonprofits through their own foundations. Corporate foundations are discussed later in this section. They may also grant smaller gifts through the company’s discretionary fund. The person directly responsible for making decisions about corporate giving is typically the head of community affairs, marketing, or community or public relations for the foundation. Among the ways that a corporation helps to fund a mentoring program are through employee volunteer and matching gifts programs; in-kind donations of services and goods; sponsorship of events in the community, including those with your mentoring program as the recipient (e.g., a golf tournament or bowl-a-thon); and direct cash donations. To begin the process of approaching a company to support your efforts, determine first who among your board members, volunteers, and staff may have close contacts with individuals inside the company. If the company has already provided work release for its employees to serve as mentors in your program, one or several of the mentors can be ideal champions to approach their top management for a company donation. Involving mentors in your program before asking for funding can almost guarantee that they understand your need for financial support. They may then provide testimony to their employers about your need or help you to gain access to the right people in their companies.

390   Practice and Programmatic Considerations In my widespread experience with corporations throughout the United States, I find that most would prefer to give to a specific cause or need rather than to a mentoring program’s general operating budget. Some specific causes or needs include a special after-school activity, a summer program, a group activity, donation of equipment such as furniture or computers to the school or agency for use by youth and mentors in the program, transportation to an activity such as a visit to a museum or zoo or to a baseball game, or donations to the program’s scholarship program for mentees heading to postsecondary education (Weinberger, 2005a). A great way to reach corporations for support is by starting with the local Chamber of Commerce. No doubt one or more of the members of your board are also Chamber members. Even though a mentoring program is a nonprofit, it should spend the money to join the Chamber. It is a wonderful opportunity to rub shoulders with the businesses in your town or city; act like a “big business,” which I believe a mentoring program is or can become; volunteer to serve on committees; and get involved. The Chamber of Commerce publishes a list of its members, which are businesses of many sizes, and shares the list with all its members. From the list, your potential donors will evolve. Some Chambers offer a lower membership rate for nonprofits. It is worthwhile to ask about this possibility. Becoming a Chamber member adds value to your program. The first schoolbased mentoring program I directed in the early 1980s (Weinberger, 2005b) was funded for many years as a line item of the school district budget. When I recommended that the district’s Board of Education become a member of the Chamber, I was met with resistance because of both the cost of membership and a lack of understanding about its potential value. It took a little coercing, but the Chamber membership paid off through the number of contacts made, as well as through the recruitment of additional mentors and the funding received after joining the organization. Get to know the company you wish to approach for program funding. Begin by doing research to learn about the company’s policies for giving to nonprofits. Many corporations have published guidelines that indicate where they spend their money and what types of requests they will receive and honor. Do your homework before you send a request. Read the company’s annual report to get to know more about it in advance. Practice your presentation “ask” over and over again in front of the mirror, to your family, colleagues, friends, and

even the family dog before you make an appointment to visit the company. Weinberger (2005a) suggests 10 steps to success in soliciting corporate giving. Matching Gifts. Many mentors work for companies that have matching gift programs. A manager would not know this unless they inquire. If mentors give to their favorite charity, in this case your mentoring program, their employers may match their gifts up to as much as 100%. For a mentoring program, the process for soliciting matching gifts is not complicated. First, have two questions on each mentor application that ask, “Do you work for a company? If so, does your company have a matching gift program?” Subsequently, during your campaign, track mentors who give financial contributions and identify those whose workplaces have matching gift programs. Ask them to furnish their company’s one-page matching gift form. You will need to verify that you qualify as a 501(c)(3) and indicate the amount of the mentor’s confirmed gift on the form. The mentoring program manager mails the form to the company. Your program will receive the matching gift directly. This process translates into little effort and potentially big results. Public and Private Foundations. Public and private foundations can support mentoring programs. Grants come in the form of sponsoring activities and events, purchasing equipment and supplies, offering professional development, and funding for program evaluation. Typically the grants are time limited, providing funding for 1, 2, or 3 years, and they rarely can be used for operating expenses (Anuszkiewicz, Salomon, Schmid, & Torrico, 2008). Foundations, like corporations, prefer to award a specific program or event. Some foundations will offer matching gifts and even assist the mentoring program to raise additional funds. Other foundations may partner with the mentoring program, offering training and technical assistance to improve the organization’s capacity. Like mentoring programs, foundations are covered by section 501(c)(3) of the Internal Revenue Code. The good news is that most foundations give out money not because they want to, but rather because, depending on their tax status, they must spend a portion of their assets (at least 5% for private foundations) on philanthropy each year (Bray, 2005). Managers can examine the tax reports that both public and private foundations file with the Internal Revenue Service by using a foundation finder also known as a widget. It also enables managers to find potential funders in their area (Foundation Center, 2010).

Program Funding  391 Foundations are categorized as public or private in terms of the way they operate. Independent foundations are private grant-making organizations that derive their endowment funds from a single source. Many are family foundations with family members serving on the board of trustees. Typically they are very specific about their areas of interests. Program managers should learn about the mission of independent foundations before they apply for funds. Corporate foundations are private grantmaking organizations whose endowment and annual contributions come from a profit-making corporation. Since their giving is geared toward the communities where the corporations operate and reflects the corporations’ areas of interest, they can be a great source of funding, especially if the mentoring program already has mentors volunteering from those corporations. In contrast to independent and corporate foundations, community foundations are public charities. They derive their support from diverse sources, which may include foundations, individuals, and government agencies (Foundation Center, 2010). Community foundations have many donors who pool their trust funds into a single entity and benefit from the investment of their funds, and sharing of staff and management. These foundations limit their giving to their communities and are popular sources of funding for mentoring programs. Managers of mentoring programs must do their homework to determine which foundations have the right fit for their mission and goals. The community foundation that gives grants only in its own community is not one to approach for funding if it is not located where your mentoring program operates. Each foundation has a website that should be reviewed carefully to determine where the foundation makes its grants. If you win the grant, plan an event to announce the grant via a press conference or at a mentoring site, and invite the most important people from the foundation as well as your board members and other key individuals. Make sure the media are invited so that they can announce your award in the local press. Individuals Individual Giving. Managers planning to ask individuals for money need quickly to acquire the philosophy that asking for money is fun and not a difficult or painful process. As I always say, the worst thing that can happen when you ask someone for money is that they will say no. You are also asking for a very good cause—children and youth.

There are good reasons why mentoring managers should concentrate on individuals and their potential to give to your program. According to Bowman (2005), who cites the American Association of Fundraising Counsel and Giving Institute, a total of 83% of all giving comes from individuals. Citing a report by the Independent Sector, Bowman suggests more people donate money than vote in national elections and that nearly 9 out of 10 U.S. families make charitable donations. This is good news. Learning why individuals give and especially why they would give to your program is worthwhile to consider. People seem to give when they think they will get something in return, something that satisfies them on a deeply personal level (Bray, 2005). Mentors often give to the mentoring program if they are satisfied with their mentoring match and level of program support. The individuals who have an interest in your program are typically those you already know. If each member of your board of directors can identify two individuals who they can approach for donations to your organization, this will increase your base of giving. Personal contacts are still the best way to make money. Person-to-person, face-to-face contact is a refreshing approach to asking for money rather than the overwhelming barrage of direct-mail letters from people you do not know. It is a misconception that only individuals with wealth give to mentoring programs. The good news is that most annual charitable giving comes from middle-income, working-class, and poor people. Given that these data represent most of the nation’s population, your universe of prospective donors is huge (Bowman, 2005). Ask your team to brainstorm all the people they know who might want to give to your individual campaign, and you may be amazed at the length of the list. Asking them once a year to contribute to what you may call your annual campaign will typically yield results. You can cosign the letter, but make sure that one of your volunteers who knows the individual you are trying to reach also signs and writes a personal note on the letter. Nothing compares to a personal request. Giving patterns do not differ significantly by age, race, or gender. Klein (2004) cites a study by Arthur Blocks of the Maxwell School of Citizenship and Public Affairs at Syracuse University that showed that 19% of families living on welfare give away an average of $72 a year. Consider the best time of year to conduct an annual campaign to solicit money from individuals. Numerous requests take place around the Christmas holidays. The rationale that December is closing in on the end of the year

392   Practice and Programmatic Considerations and individuals need to make their tax-deductible donations before year end may have some appeal. But most people receive multiple requests at the end of the year. Many of these are thrown in the garbage. Thanksgiving presents a good opportunity to set the stage to ask for money for mentoring. It is a nearly universal holiday in the United States that is not associated with a religion. Therefore, no one is offended. Just before one of the most popular, warm, and family-oriented holidays of the year, send a letter to all sponsors, mentors, donors, volunteers, and others from the mentoring program. Do not ask for money. Rather, include only a message, such as, “During this time of Thanksgiving, we in the XYZ Mentoring Program give thanks for your continuing support and belief in our efforts to make a difference in the lives of youth.” That will go a long way to prepare for individual giving efforts during your annual campaign. Determine the month that might be the right time to solicit gifts on a yearly basis. Maybe the request should come during January, National Mentoring Month, or during the summer just before the beginning of a new year of mentoring for many school-based programs. Make it the same date every year so that potential donors can anticipate your request. Planned Giving. Mentoring programs that have been in existence for a long time, that have a strong donor base, and that keep accurate tracking records on their mentors and alumni may be able to use planned giving as another way to raise funds for the organization. This strategy focuses on the motivation of many donors to achieve some mortality. Your mentoring program is a means for individuals who care deeply about your mission and have been involved in the organization to consider a lasting gift. You may need some help from a professional to establish a planned giving program. Preparing to receive bequests will take time and require some evidence that your agency will be around for a long time to come. Small, new, or unstable organizations should be discouraged from establishing a planned giving program until they become stronger financially and have long-term success rates with the health and management of the mentoring program. The first section of this chapter discussed the health of a mentoring organization. Program Events and Direct Marketing Special and Third-Party Events. In addition to the funding streams discussed in this section, special and third-party events, while often yielding fewer results in terms of money, should be part of any

fundraising campaign. These events are a perfect opportunity to create awareness about the mission of your program. Typically special events can raise both restricted and unrestricted funds for your operations or a particular initiative for which you are seeking support. Popular events include “taste of [your town],” celebrity dance marathon, black-tie gala, bowl-a-thon, walkathon, car wash, crafts fair, lecture, or jail-a-thon. Some programs even invite guests to a nonevent. The guests pay to stay home, and your organization reaps the benefits of an event that eliminates staff time and associated costs. Third-party events are defined as those where another organization schedules the event and your mentoring organization is the recipient of the profits. This is the kind of event that takes the least amount of staff and volunteer time yet can yield considerable results in dollars. One example that has been popular for mentoring programs is a golf tournament. Another organization in the community runs the event, handles all the publicity, and designates the XYZ Mentoring Program as the beneficiary. Key to benefiting from third-party events is researching where and when they take place in your community. Most likely you will have to apply to be the beneficiary long in advance. Offering to have some of your volunteers assist with ticket sales or selling raffle tickets at the event, for example, shows your appreciation for the donor’s generosity in selecting your program. Some programs benefit year after year. When special and third-party events are being planned, managers must consider any potential disasters that might occur. Injuries, accidents, or someone robbing your cash draw are rare occurrences, but it is always a good idea to plan for the unforeseen. Research the laws that apply to the event and any permits you may need. Events should be staffed well, and staff need to review in advance where they can go if a problem occurs. If necessary, ask someone knowledgeable about your mentoring program’s current insurance policy to determine if additional insurance will be needed for these events. It is surely better to be safe than sorry. An insurance agent or attorney serving on your board can assist with these matters. Direct Mail. Direct mail, as referenced earlier in this chapter, is the least effective way to raise funds for your program. Soliciting new supporters through direct mail usually involves buying mailing lists from another organization or professional marketing firm that specializes in selling lists to nonprofits. In addition, you will need to purchase stationery and stamps for the direct-mail campaign. Unless your mentoring

Program Funding  393 program has a well-recognized name, such as comes from being part of a national organization (e.g., Boys & Girls Clubs, Big Brothers Big Sisters of America), all of your mail requests may be thrown away. People have mixed feelings about direct-mail fundraising (Bray, 2005). Managers will need to weigh the benefits of direct-mail campaigns before spending a great deal of time on such a project. E-mail, internet banners, and a mentoring program’s website are forms of direct donor solicitation. These are easy to create and are used to raise awareness and funds for the organization. A separate page on your website dedicated to giving and information on how money is funding worthwhile programs and projects can go a long way. Newer forms of solicitation, including social media such as Facebook, are popular ways to appeal and receive funds for your program. Users of your website can be tracked easily. Keeping site traffic statistics enables your program to learn not only how many people are visiting your site but also where they come from and where on your site they are looking for information. If you have a web designer on your board, he or she can assist with this effort. Local Initiatives. Most communities have affiliates of social, fraternal, and civic organizations whose mission it is to help local nonprofits. These need to be researched to determine if their interests and yours are compatible. In addition, local affiliates of the United Way of America have a long history of funding nonprofits in most communities. If your program resides in a rural area, typically there is a regional United Way office that serves your location. United Way conducts fundraising campaigns each year and distributes grants to support local programs. The organization has specific guidelines about the programs that they prioritize, and the recipients of their donations reflect those priorities. The United Way also has a strict policy regarding a black-out period during its own community fundraising campaign. If you agree to be one of the organization’s recipients, you will most likely not be able to hold significant fundraisers during the months when the United Way’s efforts to raise dollars are heightened. You will need to review the organization’s policies to see if the United Way is a good fit for your organization. If you have a board member who also serves on the United Way’s board, you can ask to secure time at one of the United Way’s board meetings to introduce its members to your mentoring program and its goals. United Way also has a designated donations program. Although the organization may prefer to raise unrestricted dollars for distribution to its

member agencies in need of funds, any individual can make a United Way contribution to the 501(c) (3) organization of his or her choice. A mentoring program can be a recipient as the charity of choice of mentors. The program has to complete a designated donation form provided by the United Way and verify that it has the proper nonprofit status. Capitalizing on Success Fees for Program and Product Services. Successful mentoring programs are replicating their models and selling them and associated materials to other programs for a fee. This can be an excellent revenue source for stable and well-financed programs. Before replicating your program, it is important to reflect on its age, strong and diverse revenue streams, and continuous expansion of mentoring matches. It is also important to ensure that programs, prior to efforts at commercial replication, have suitably rigorous outcome evaluations that indicate positive results (see DuBois, this volume, Chapter 33). Some programs offer to assist smaller programs by holding training sessions for their mentors, by providing a training curriculum, or by handling all of smaller programs’ criminal background checks. These services add to the base of funding for programs. No mentoring program should go in this direction unless it closely adheres to the Elements of Effective Practice (MENTOR, 2005) regarding long-range planning, recruitment, screening, training, evaluation, and ongoing program support. Using Program Evaluation to Support the Need. Regardless of the sources of funding for your mentoring program, potential funders will be more interested in giving if you can provide proof that your program is operating effectively. The best way I know to maintain and grow funding support is by providing evidence of program quality (formative evaluation data) and effectiveness (outcome data). If you are achieving defined goals and objectives (e.g., number of new matches, retention of matches, accurate records of mentor and mentee contacts, feedback surveys of training events) and can track short-, intermediate-, and long-term program outcomes (e.g., improvement in mentee attendance, academic performance, and peer engagement), your funders will be impressed and more convinced that their dollars will be used effectively. Responding to Actual and Potential Contributors Thanking Donors. The Chronicle of Philanthropy connects the nonprofit world with news, jobs, and ideas. Weekly the magazine addresses fundraising, giving,

394   Practice and Programmatic Considerations causes, facts and figures, and the day’s news in the nonprofit sector. It is a valuable resource for mentoring programs wanting to stay current with the nonprofit sector. Recently, the Chronicle featured an article about Kivi Leroux Miller, who reported on what she called the pitiful response to her “What I Got When I Gave” experiment (Wasley, 2009). She donated $20 to each of 10 national nonprofits through the Capital One Giving site. One month after sending the gifts, she had heard from only 3 of the 10 organizations. How can nonprofits expect to thrive off the kindness of others, Miller asks, when the kindness of a simple thank you note to an unsolicited donation is too much to ask? It appears that nonprofits that show appreciation for gifts, both large and small, by sending a thank-you note are in the minority. Regardless of who donated money, goods, or services to your program, once you receive the gift you must thank the donor promptly. In the case of monetary donations, write the thank-you note before you deposit the check. In this technological world, many managers opt for writing e-mail thank-you notes. Notes do not have to be long or fancy, but nothing replaces a personal note written on the mentoring program’s official stationery. The donor can then use the note as proof of their gift for tax purposes. Accepting Rejection. Regardless of how much you and your fundraising team have tried, there are

times when a potential donor is not willing to support your mentoring program. You may think that a gift would be an important investment in youth, but not everyone is positioned to believe the way you do. Managers need to learn how to accept refusals graciously. Do not try to make the individual or the organization’s representatives feel guilty, but rather keep in mind that there is always a next time. The refusal may have nothing to do with your program. Send the person or organization that declined to make a donation a note of thanks for considering your program and keep them updated on your successes throughout the year. Table 26.2 provides some guiding questions to consider when seeking sources of funding for mentoring programs.

Conclusion This chapter addressed the most important planning processes that must be in place prior to launching a comprehensive fundraising plan. To begin, it is critically important that the mentoring organization has a strong infrastructure and is healthy, and it must have a clear written mission statement, compelling and defined goals and objectives, a statement of critical need, successful outcomes, a detailed budget, data-collection procedures, and a financial statement of growth patterns. Funding should come from

Table 26.2   Checklist for Practitioners: Funding Sources for Mentoring Programs What challenges will you encounter in identifying board members and other volunteers to help the manager to seek multiple sources of funding? Who will monitor the federal register for Requests for Proposals for federal grants? What kind of tracking system will you need to develop to collect data and to monitor contacts with revenue sources and results of your efforts? Do you track alumni of your mentoring program—mentors and mentees—and ask them to help fund your program? Do you have a champion among local, state, or federal legislators? Do you have a list of companies with a strong history of support for your program (e.g., offering work release for mentors) that could now provide you with program funding? Is your agency a member of the Chamber of Commerce? Will you encounter any resistance to joining from your board or others? Do you have an attorney and an insurance agent on your board who can provide professional advice in areas such as insurance for special events and planned giving? What policies will you need to put in place around fundraising? How do you rank in terms of your timeliness for saying “thank you” to donors? What will you need to do to prepare for rejection?

Program Funding  395 multiple and diverse sources. The organization must have a passionate, qualified manager and a strong, active, experienced, and enthusiastic board of directors. All board members must be willing to roll up their sleeves, work on a fundraising campaign, have the knowledge and expertise to make the right contacts and ask for money, and give annually to the mentoring program themselves. To yield results, multiple funding streams must be pursued. No manager can do this alone. Board members and volunteers play an important role as partners. It is not so much what you know but who you know that will elevate the fundraising campaign to the results you anticipate. All fundraising streams mentioned in this chapter help to create awareness of your mentoring program and your need. Managers and their fundraising committees must research well to determine the best approach in each category of giving potential. A deliberate and long-term process will bring great results. Finally, many important questions remain for which little or no evidence is available to guide fundraising efforts. I hope future researchers will examine these questions empirically and that program staff will address these issues pragmatically as best they can in the meantime: What are the most appropriate data for program managers to use to ensure that mentoring programs are effective? How can we increase understanding of resistance on the part of managers of mentoring programs to conduct process and outcome evaluations that will add credibility to their programs? How can we determine the most effective ways to raise funds in order to ensure long-term sustainability for mentoring programs?

References Anuszkiewicz, B., Salomon, N., Schmid, W., & Torrico, R. (2008). Finding resources to support mentoring programs and services for youth. Retrieved from www .financeproject.org

Bowman, C. (2005). Individual giving: Building a powerful constituency. In M. Garringer (Ed.), Sustainability planning and resource development of youth mentoring programs (pp. 85–116). Portland, OR: Education Northwest. Bray, I. (2005). Effective fundraising for nonprofits. Realworld strategies that work. Berkeley, CA: Nolo Press. Charity Buzz. (2009). Direct mail among least costeffective fundraising techniques. Retrieved from www.charitybuzz.net Foundation Center. (2010). Foundation finder. Retrieved from www.foundationcenter.org Fulop, M. (2005). The resource planning process. In M. Garringer (Ed.), Sustainability planning and resource development of youth mentoring programs (pp. 5–34). Portland, OR: Education Northwest. Klein, K. (2004). The ten most important things you can know about fundraising. Grassroots Fundraising Journal, 25, 1–4. Kotter, J. P. (1996). Leading change. Boston: Harvard Business School Press. Lav, I., & McNichol, E. (2009). 29 states faced total budget shortfall of at least $48 billion in 2009. Washington, DC: Center on Budget and Policy Priorities. Retrieved from http://www.cbpp.org/1-15-08sfp.htm Mancuso, A. (2009). How to start a nonprofit corporation. Berkeley, CA: Nolo Press. MENTOR. (2005). Elements of effective practice (2nd ed.). Alexandria, VA: Author. MENTOR. (2009). Save federal funding for mentoring fact sheet. Retrieved from http://www.mentoring.org Wasley, P. (2009, March 13). Charities flunk the gratitude test. Chronicle of Philanthropy. Retrieved from https:// beta.philanthropy.com/blogAuthor/Prospecting /15/2/Paula-Wasley/159/ Webster, B. (1999). Sustaining program viability. Sacramento, CA: EMT Group. Weinberger, S. (2005a). Direct corporate support. In M. Garringer (Ed.), Sustainability planning and resource development of youth mentoring programs. (pp. 41–49). Portland, OR: Education Northwest. Weinberger, S. G. (2005b). Mentoring a movement: My personal journey. Norwalk, CT: Author. Wrobel, T. (2010). Starting 501c3 nonprofits and expert 501c3 advice. San Francisco: Nonprofit Legal Center. Retrieved from http://nonprofitlegalcenter.com

27  MENTOR RECRUITMENT AND RETENTION Arthur A. Stukas, E. Gil Clary, and Mark Snyder

Introduction Despite the long and healthy tradition of formal volunteering in developed nations, nonprofit organizations often still struggle to find the right volunteers. Organizations that run mentoring programs for youth may face additional hurdles, because they must find volunteers who can first pass important screening tests and who will then commit to a sustained relationship with a young person. But volunteer coordinators know that not all of the volunteers who meet these minimal criteria will end up having a sustained or effective match with their youth. Currently, organizations use a variety of approaches to attract the best possible mentors. For example, as Garringer (2006) points out, programs seeking mentors may make formal presentations to a variety of community and professional groups (e.g., businesses, churches, fraternities, universities), they may ask staff and current volunteers to make personal appeals to friends and associates, they may engage in media campaigns, and they may make use of volunteers available through programs run by the Corporation for National and Community Service in the United States, among many other strategies. In the previous version of the Handbook of Youth Mentoring, Stukas and Tanti (2005) focused on features of persons that might attract them to, or make them suitable for, volunteering broadly and youth mentoring specifically. Their review focused on demographic characteristics, personality traits, experience in a related role, and motivational differences that have been associated with participation. But these authors were also careful to point out that features of persons often interact with features of situations (or tasks, organizations, environments) to

predict successful recruitment and retention of volunteers. For example, the kinds of opportunities (or affordances) available in volunteer or mentoring activities might attract only those volunteers with particular motives or personal characteristics. Stukas and Tanti suggested, therefore, that more successful recruitment might result from appealing to these particular potential volunteers by marketing to them the features of mentoring they find most personally relevant and attractive. Moreover, sustaining mentors in their role was seen to be a function of the ongoing experiences that mentors had, including the formation of close and positive relationships with their mentees, and ongoing satisfaction of their own needs and motivations for volunteering. Proper orientation, ongoing training, supervision, and contact with organization staff and coworkers were seen as necessary to help mentors forge strong and stable bonds with their young partners. In this chapter, we do not retread the content from Stukas and Tanti (2005), although we use a similar organizational framework that focuses on features of persons and features of situations. Where appropriate, themes from the chapter in the previous edition of the Handbook are updated, but mostly this new chapter provides new information on the theoretical and empirical literature on recruitment and retention of volunteers.

Theory Theorizing about volunteerism often focuses more on who volunteers and why than on who volunteers well. Perhaps this is because volunteer effectiveness is often difficult to measure, either practically or politically. However, when it comes to youth 397

398   PRACTICE AND PROGRAMMATIC CONSIDERATIONS mentoring, youths’ benefits and improvements provide a useful outcome measure by which to judge mentor effectiveness. Little work has related mentor characteristics directly to youth outcomes, but given Grossman and Rhodes’s (2002) important finding that the length of the relationship between mentors and youth is an important predictor of outcomes, the most effective mentors may be those who attend to relationships in a consistent and enduring fashion. In this section, we focus on personal characteristics that may lead people to volunteer and that may make them more likely to sustain their volunteerism. We focus in particular on those features of the situation or the environment that organizations supporting mentoring can control: the strategies they use to recruit and to retain volunteers. Features of the Person Many psychological approaches to studying human behavior separate personal predictors of action into two classes: those that focus on motivations and preferences and those that focus on skills and abilities. With regard to the specific task of youth mentoring, volunteers need not only to be inclined toward the act, but also to be able to manage the specific requirements of working with youth who may come from very different, and even difficult, backgrounds. As we discuss below, both relevant motivations and requisite skills may predict a mentor’s continuity in the match. Skills and Abilities. For many theorists, the starting point for understanding whether or not a person will volunteer help for a person in need is whether the potential helper can take the other person’s perspective and feel empathy for his or her plight. For example, Batson’s empathy-altruism model (Batson, 1991) posits that help will be given more often to someone with whom we can identify and therefore imagine things from their viewpoint; in his model, this perspective-taking leads to a sharing of, or sympathetic feeling for, the other’s affective experience, which is empathy or empathic concern. Batson suggests that empathy can be elicited by aspects of the person in need (e.g., their similarity) or their plight (e.g., having personally experienced it) and may spontaneously occur when we explicitly value the welfare of the person in need (Batson, Eklund, Chermok, Hoyt, & Ortiz, 2007). Other theorists have suggested that perspective-taking and empathy are skills or abilities that different people possess in varying degrees (e.g., Davis, 1980; a well-known scale is available at http:// www.eckerd.edu/academics/psychology/iri.php).

However, the predictive value of dispositional measures of these skills for helping behaviors has received a mixed pattern of support in the literature. The capacity for empathy may be associated with prosocial behavior broadly, but measures of dispositional empathy may relate only weakly to specific types of prosocial behavior, such as volunteering. For example, Eisenberg’s (2010) work suggests that individual differences in emotional regulation and self-regulation are stable from childhood to adulthood and do predict both the capacity for empathy and the tendency to behave prosocially. But, in a representative survey of Americans, Einolf (2008) reported only small correlations between Davis’s (1980) dispositional measure of empathic concern and volunteering; this, however, may make sense when you consider the multiple influences on the decision to volunteer that have been documented. The same mixed pattern exists when examining whether the capacity for empathy predicts effective helping or ongoing commitment to helping and community service. For example, Stolinski, Ryan, Hausmann, and Wernli (2004) found that dispositional empathic concern and perspective-taking influenced people’s intentions to continue serving as a buddy to a person with AIDS, with these effects mediated by how important and challenging the volunteers felt their work to be; Stürmer, Snyder, and Omoto (2005) found that empathy predicted people’s actual helping activities (including volunteer longevity) toward people with AIDS but only when helpers and recipients were from the same “ingroup” (conversely, helping for an “outgroup” was facilitated by liking for the person in need). Clary and Orenstein (1991) found that prospective volunteers for a telephone-based crisis-counseling service were more likely to be screened out of the program if they had lower dispositional perspective-taking ability, suggesting that organizations may be able to see real differences in volunteer capabilities and performance as a result of these dispositional tendencies. A range of other dispositional traits and abilities have also been examined as predictors of helping. For example, Carlo, Okun, Knight, and de Guzman (2005) examined relationships between the well-known Big Five personality traits (sometimes assessed with scales at http://ipip.ori.org/ ipip/; Goldberg, 1992) and measures of volunteer processes, finding that Agreeableness and Extraversion were the most viable candidates. Indeed, Agreeableness and Extraversion may have their effects on volunteering by promoting stronger motivation to express prosocial and humanitarian values (Carlo et al., 2005).

Mentor Recruitment and Retention   399 Others have focused on attachment styles stemming from early childhood relationships with parents, showing that insecure attachment styles can result in less positive adult relationships and less interest in prosocial behavior (Fraley, Waller, & Brennan, 2000; Gillath et al., 2005; a scale is available at http://www.psych.illinois.edu/~rcfraley/ measures/ecrritems.htm). With regard to volunteering, Gillath et al. found that being high in attachment avoidance was associated with volunteering less, both in terms of number of activities and time spent, and with lowered motivations to volunteer, especially the motivations to express prosocial values and to gain understanding about other people; whereas being high in attachment anxiety was associated with greater motivations to volunteer in order to boost one’s own self-esteem, to protect oneself from guilt, and to fit in to the social expectations of others. Nevertheless, when it comes to mentoring, which requires positive relationships to be formed, attachment security seems all the more important. Thus, organizations may wish to enhance their pool of effective or interested mentors by searching for, screening in, or otherwise enticing certain types of volunteers. However, not every community will be able to provide a ready stock of volunteers with necessary or desirable skills. That is, some communities may have a deeper pool of appropriate mentors than others. Communities may also contain people who have more or less supportive values and attitudes, thus making recruitment an easier or more difficult task for organizations. For example, Kemmelmeier, Jambor, and Letner (2006) reported higher rates of charitable giving and volunteering in U.S. states that were more highly individualist, suggesting that helping strangers is associated with the individualistic value of personal responsibility. Aside from the proportion of prosocially oriented people within them, communities may also vary in the degree and quality of connections between and among people. That is, some communities also have more social capital than others, the social connections that build trust through the giveand-take of reciprocal relationships. For example, Oishi et al. (2007) found that communities marked by more residential stability, and therefore more long-standing connections between people, engaged in more civic and prosocial behavior. This effect of residential stability on community participation may be mediated by the sense of identification that people feel with their communities (Oishi et al., 2007). A sense of community may be leveraged to recruit volunteers or may make people more civic minded (Omoto & Snyder, 2010); moreover,

identification with others in one’s community might be an outcome from participating in volunteer service. Stukas, Daly, and Cowling (2005) found that volunteers for two community charities (one health focused and one poverty focused) reported higher levels of trust and sense of community after volunteering, but only when their volunteer activities satisfied their motivations for volunteering. Indeed, benefits such as increased sense of community may accrue only when mentors find that mentoring activities satisfy their personal goals and motives, an issue we turn to next. Motivations and Preferences. Stukas and Tanti (2005) reviewed the tenets of the functional approach to volunteerism (Clary et al., 1998), which suggests that different volunteers may have different reasons or motivations for volunteering, even if performing the same task, and that satisfaction and intentions to continue are a result of volunteers having the opportunity to fulfill their motives in their volunteer tasks. Work by Stukas, Worth, Clary, and Snyder (2009) suggests that volunteers who both have multiple reasons for getting involved (as assessed by Clary et al.’s Volunteer Functions Inventory; VFI; found at http://gener osityresearch.nd.edu/assets/13636/clary_snyder _volunteer_function_inventory_scale.pdf) and who find that their volunteer opportunity allows them to fulfill these multiple motives may ultimately wind up more satisfied and more likely to intend to stay in a volunteer position than volunteers who have fewer fulfilled motives. Among the multiple motives measured by the VFI, volunteers interested in youth development may be most motivated toward expressing their humanitarian values and understanding more about the world and other people than other types of volunteers, as was reported by Clary, Snyder, and Worth (2003) in their survey of nearly 1,400 Minnesotan volunteers, 82 of whom were working with youth. Thus, programs could attract mentors by highlighting opportunities for value expression and learning in their recruiting efforts and could maximize satisfaction by ensuring that such opportunities are salient and available. Other research on motivation has differentiated between intrinsic and extrinsic reasons for helping. For example, Finkelstein (2009) suggested that intrinsic orientations and extrinsic orientations toward work, including volunteer work, might function like stable dispositions rather than vary with the particular activity. She demonstrated that students with a more extrinsic orientation were more likely to see career benefits as persuasive reasons to volunteer, as compared to students with more intrinsic orientations who preferred more “internal”

400   PRACTICE AND PROGRAMMATIC CONSIDERATIONS reasons, such as value expression or understanding. Other researchers (Gebauer, Riketta, Broemer, & Maio, 2008) have argued that prosocial motivation may stem from either pleasure (intrinsic motives) or pressure (extrinsic motives); only greater pleasurebased motivation to help others was associated with life satisfaction, positive affect, self-esteem, and self-actualization, suggesting that motivations based on pressure might not be as healthy. Nevertheless, these considerations point to the continued value of assessing the motivations and reasons that volunteers give for getting involved and of ensuring that the benefits they seek from volunteer activities are available to them, as a way of increasing satisfaction and long-term commitment to service (e.g., Clary et al., 1998; Konrath, Fuhrel-Forbis, Lou, & Brown, 2012). More data may be needed to determine whether all motivations are created equal or whether some motivations (e.g., those arising from external pressure) result in poorer outcomes or effectiveness. Features of the Situation The situational or environmental factors that might lead someone to volunteer as a youth mentor may differ from those that might influence her or him to continue mentoring for a sustained period. For this reason, we first review recruitment strategies and then retention strategies, focusing on the ways that organizations might improve their processes. Recruitment Strategies. As discussed by Stukas and Tanti (2005), volunteers often report that being asked to participate was the reason they began their involvement. This method can successfully leverage personal relationships to recruit new mentors, but not every current volunteer may be willing or able to do this. As well, organizations may need to bear in mind that not every potential volunteer is the same and that different people may be better recruited using different methods. For example, we (Stukas, Snyder, & Clary, 2008) examined strategies that could be used with people already prone to volunteer, people resistant to volunteering, and the large group of those people who are open to a good offer. We suggested that those who are already prone to volunteering might be recruited using a simple educational campaign, explaining how and when to sign up. These people may also be most receptive to “the ask.” In keeping with our functional approach, we also suggested that people open to a good offer might be persuaded to volunteer if the benefits available from the activity (as described

in a recruitment message) matched up with their goals and motivations. In line with best practice in the field of social marketing, we also advised recruiters to fully understand the product they are offering, the price (time, emotions, opportunity lost) it will extract from volunteers, and the place in which activities must take place. Providing enough information at recruitment will allow volunteers to self-select into activities that best match their motivational impulses and to engage in cost-benefit calculations to determine whether some factors (such as distance from home) make serving more or less desirable. The most difficult group to recruit is those currently resistant to volunteering. Sundeen, Raskoff, and Garcia (2007) found that the major barriers to volunteering reported by 48,168 nonvolunteers in a U.S. national survey were lack of time (43.4%), lack of interest (27.1%), and ill health (14.0%). Although genuine lack of time and ill health seem difficult to overcome, recruiting those who are simply not interested in volunteering is sometimes achieved with more heavy-handed recruitment strategies. For example, we (Stukas et al., 2008) discussed government-imposed requirements that have mandated community service for high school students in some American states and similar requirements instituted by colleges and universities. Requirements may serve to get hands to work and thus can result in a real contribution to the community. Additionally, advocates suggest that those who might not have been attracted to volunteering initially may find benefits that they did not anticipate and thus may become converts to lifelong volunteering (see Metz & Youniss, 2005). However, some empirical data suggest that requirements to serve can undermine future intentions to volunteer, which in the case of mentoring could mean shortened matches, a negative for youth (Grossman & Rhodes, 2002). For example, volunteers in a required university program calling for 40 hours of service across a single semester had lowered intentions to volunteer in the future when they felt extrinsically controlled by the requirement, as compared to those who retained their feelings of autonomy despite being required (Stukas, Snyder, & Clary, 1999). Planty, Bozick, and Regnier (2006) tracked high school volunteers for 8 years, discovering that those who were required but maintained that they would have volunteered even without the requirement were still serving at higher than average rates in the future, whereas those who felt their participation was dictated by the requirement and who would not have volunteered otherwise did not sustain their participation over 8 years.

Mentor Recruitment and Retention   401 Although requirements for participation might seem unusual in the case of youth mentoring, where organizations must be careful about who is involved, there have been suggestions that youth mentors could be influenced to volunteer by way of another external pressure, in the form of monetary payment (as exemplified by a 2009 call for proposals by the U.S. Office of Juvenile Justice and Delinquency Prevention to study the effects of compensating mentors). However, such rewards have also been shown to undermine motivation. According to Deci, Koestner, and Ryan (1999), to the extent that rewards are perceived as reducing a person’s sense of autonomy rather than reinforcing their competence, rewards will undermine both interest and intrinsic motivation. They suggested that tangible rewards, such as money, can be more damaging to intrinsic motivation than intangible rewards, such as verbal praise or the kind of benefits afforded volunteers, including opportunities to learn new skills, meet new people, feel good about oneself, or foster career advancement. Indeed, simply reminding people of money seems to reduce their helpfulness (Vohs, Mead, & Goode, 2006). Retention Strategies. As Stukas and Tanti (2005) noted, ensuring that volunteer activities actually provide the benefits that volunteers seek is an important factor in the retention of volunteers. Indeed, to assess whether volunteers are successfully meeting their needs, Stukas et al. (2009) suggested that organizations could calculate motivational match indices that are the result of a volunteer’s initial motivation score on a VFI subscale, multiplied by the benefits that the volunteer perceived in their actual activity, as assessed by questions yoked to the original VFI subscale. When the multiplicative scores across all six subscales (career, enhancement, social, values, understanding, protective) were summed, this “total match index” was shown to predict overall satisfaction better than either motivation or affordance (benefit) factors alone. Consequently, success in retaining mentors would involve either finding volunteers who possess motives that link to established benefits or modifying activities to guarantee the presence of benefits that mentors are motivated to obtain. However, Stukas et al.’s (2009) findings also suggest that, for some organizations, there are other factors beyond matching task-related benefits to volunteers’ motivations that may help to promote retention. The total match index was not as strong a predictor of overall satisfaction in those organizations that were more highly “structured,” which was operationalized here as having a volunteer coordinator, engaging in performance evaluations, and

having volunteers interact with clients as well as other volunteers and paid staff. For organizations with less structure, however, matching motives to activities was more important in understanding satisfaction (as discussed in the previous paragraph). One wonders then, given equivalent levels of satisfaction in high and low structured organizations, what contributes to the enjoyment and retention of volunteers in these more structured organizations, if meeting their original goals is less important. Aside from the possibility that these better structured organizations are larger and have better resources to allow volunteers to experience a uniformly satisfying set of activities, these potentially “name brand” organizations may foster greater identification with being a volunteer. Indeed, Boezeman and Ellemers (2007) demonstrated that volunteers’ perceptions of the emotional and taskrelated support they received from an organization was a predictor of their perception that the organization respects them and their contribution, which subsequently predicted commitment to the organization. In the same model, volunteers’ perceptions of the importance of their volunteer work was a predictor of their pride in being part of the organization, which again predicted commitment. More structured organizations may be better at communicating the importance of their mission and at supporting their volunteers than less structured organizations, due to their higher profile and resources. Potentially then, pride and respect may contribute to volunteers’ commitment to these more structured organizations and might be more difficult to generate in smaller, less structured organizations. In the smaller programs, volunteers may need to find their own way to satisfaction and commitment, by ensuring that their motivational needs are fulfilled.

Research Little research has been conducted on the recruitment processes that are best for recruiting youth mentors or on the characteristics of mentors that are best suited for the challenges of the tasks. Stukas and Tanti (2005) reviewed a number of important studies conducted up until 2003 or 2004. In this short section, we extend their review but focus only on those areas of particular interest in this chapter. Features of the Person Skills and Abilities. Few studies have investigated the relations between mentors’ dispositional traits and abilities and their interest in mentoring or their

402   PRACTICE AND PROGRAMMATIC CONSIDERATIONS ability to forge a lasting relationship with a youth. More work is needed in this area, using better instruments and methods. For example, using an observational coding method called the Group Assessment of Interpersonal Traits (which did not prove to be very reliable), Dicken, Bryson, and Kass (1977) found that college student mentors with greater empathy formed better quality relationships with children aged 6 to 13, as judged by psychologists who rated the mentors’ visit reports. Herman and Usita (1994) examined the 16 PF personality inventory for 11 Big Brothers and Big Sisters who had failed to complete their 1-year commitment as compared to 11 matched volunteers who had remained as mentors for 2 years or more. They identified a discriminant function suggesting that committed volunteers were more likely to be abstract thinkers, conscientious, and imaginative. This sample may be too small to allow valid conclusions, however. Moreover, as Roaf, Tierney, and Hunte (1994) reported, personality tests are used predominantly by mentoring organizations as a requirement imposed on them by insurers to aid in the process of screening out mentors with undesirable characteristics (e.g., child abusers) rather than as a way of identifying the best mentors. At this stage in the research on mentor characteristics, little conclusive evidence has emerged about which mentor traits and skills are best linked to mentoring success. The evidence summarized by Stukas and Tanti (2005) suggesting that certain demographic characteristics (e.g., being young, unmarried, economically better off, or a member of a helping profession) are associated with greater commitment to mentoring may still form the best basis for selecting mentors based on empirical research specifically focused on mentors to youth. Regarding other personal characteristics reviewed earlier, there is little reason to suspect that empirical studies of other types of volunteers would not apply to the case of youth mentoring. One might expect the challenges of youth mentoring to require mentors of youth to have even higher levels of the traits and abilities discussed earlier (e.g., empathy). Motivations and Preferences. We know of only one study of youth mentors that has explicitly examined the motivations underlying participation of mentors since the last version of the Handbook. As part of the 3-year Study of Mentoring in the Learning Environment (SMILE), Karcher (2004) administered Clary et al.’s (1998) VFI to 151 mentors, finding that women were more motivated by opportunities for esteem-enhancement and value expression, whereas men were driven primarily by social

concerns. Yet this social motivation was higher among Anglo men than Hispanic men; Hispanic men were also less likely to see mentoring as a way to express important values. Finally, Anglo women had the highest expectations that mentoring would help their careers, whereas Anglo men were lowest on this scale. A clear implication of Karcher’s results, then, is that women and men from different cultural backgrounds may be recruited and sustained through different methods, focused on their differing reasons for mentoring. Although mentors with a variety of motives may all have positive impacts on youth, data suggest that those with positive attitudes toward young people may yield better outcomes for mentees (Karcher, Davidson, Rhodes, & Herrera, 2010). Features of the Situation Recruitment. In 2005, MENTOR: The National Mentoring Partnership commissioned a new survey of 1,000 randomly selected Americans to determine who mentors and why (O’Connor, 2006). The survey asked respondents to indicate up to three (out of five) reasons that led them to become a mentor, with “to help young people succeed” (82%) and “to make a difference in someone’s life” (76%) far outstripping “to give back to the community” (43%), “religious and spiritual reasons” (27%), and “someone helped [me] when [I] was young” (22%). Interestingly, none of the reasons targeted by the survey designers focused on the benefits achievable for mentors directly, as in the functional approach (Clary et al., 1998), although value expression and esteem enhancement may be implicit. When it comes to how people became mentors, the 2005 survey (O’Connor, 2006) again found “the ask” to be the number-one reason, with 50% of all respondents currently mentoring indicating that someone personally asked them to mentor (whether formally or informally). The next highest frequency responses, through membership in an organization or through an existing mentor, also show the extent to which personal connections are responsible for most mentors initiating their service. Nevertheless, 35% of mentors responded that they actively sought out the activity, suggesting that they could be “open to good offers.” Indeed, the survey’s findings from those not currently mentoring suggest a social marketing approach (e.g., Stukas et al., 2008) focused on product, place, and price might pay dividends. Nonmentors who would seriously consider mentoring (29% of those surveyed) sought the convenience of seeing their mentee near their own home or workplace (83%) with smaller groups hoping to mentor online (33%) or at their own workplace (30%) or at

Mentor Recruitment and Retention   403 least with time off from work in order to mentor (70%). Nonmentors also wanted the “product” to include support from experts (83%) and orientation or training (73%). In addition, nonmentors who were not considering becoming a mentor also nominated “lack of time” as the chief barrier, alongside what might be termed “lack of knowledge” about mentoring. A third of this “resistant to mentoring” group indicated that they volunteered in other ways, whereas 5% indicated that they had previously had a bad experience mentoring. When asked, mentors can and do report benefits that they receive. Herrera and her colleagues (Herrera, Kauh, Cooney, Grossman, & McMaken, 2008), in their evaluation of 10 Big Brothers Big Sisters (BBBS) programs (assisting over 1,100 youth in 71 schools), found that both high school mentors and adult mentors reported increases in their interpersonal (communication, patience) and personal skills (responsibility, reliability, organization) as well as increases in leadership ability, knowledge about youth development, interest in social issues, and respect for other cultures and religions. High school volunteers often reported larger increases than adult volunteers, although adults may have less to learn in these areas. Similarly, Karcher (2009) found that, after 8 months, teen mentors reported gains in academic connectedness and self-esteem relative to their nonmentoring peers. Empirical studies that compare mentor recruitment messages that highlight these benefits to messages that do not are needed to demonstrate the potential efficacy of this strategy, bearing in mind that potential mentors may differ in their receptiveness depending on the benefits they seek. Most youth mentors freely choose to become involved in mentoring; however, some are offered monetary incentives or course credit at a university in exchange for their participation, or to help cover actual costs. In their 2002 meta-analysis, DuBois et al. reported no differences between programs that offered compensation to mentors and programs that did not in terms of youth outcomes; whether recruitment and retention of mentors is affected by any extrinsic rewards for mentoring is unknown. Experimental or quasi-experimental studies that examine the effects of compensation on mentors’ outcomes (e.g., the length of their relationships with youth), such as one initiated by the U.S. Office of Juvenile Justice and Delinquency Prevention, also may clarify whether such rewards reduce intrinsic motivation. Retention. Given that early termination of mentoring relationships can have negative effects on youth

(Grossman & Rhodes, 2002), research examining the factors that influence mentors to uphold their commitment is extremely important. To find out why they terminated, Spencer (2007) conducted in-depth semi-structured interviews with 20 BBBS mentors and 11 mentees whose relationships did not last the full 12-month commitment. Mentors’ unrealistic expectations about the needs and capabilities of their youth partners were a common theme in these failed relationships; some mentors began with a romantic idea of the mentoring process that lost its sheen when the challenges facing their mentee became apparent. Other mentors may not have been able to transcend their own cultural backgrounds in order to properly understand or engage with a youth from a different background; although this suggests that a certain degree of dispositional empathy or training in perspective-taking would be helpful, other research suggests that liking a person from another group can be sufficiently motivating (Stürmer et al., 2005). Finally, a few mentors from failed relationships mentioned that they expected to get more out of mentoring (e.g., a warm feeling) and that this wasn’t achieved. Using a quantitative method, Madia and Lutz (2004) surveyed 95 BBBS mentors to examine whether perceived similarity in interests, demographic characteristics, and personality traits with their mentees was a significant predictor of intentions to remain in the relationship. Surprisingly, of these variables, only similarity in levels of extraversion predicted intentions to continue. This finding may suggest that mentors and mentees who share preferences for certain types of activities or conversational styles may fit better together, although it is difficult to generalize from a single study. In addition, Madia and Lutz found that mentors’ retrospective reports that the reality of the mentoring experience was different from the idealized expectations they held when first recruited was also a significant predictor of intentions to remain. Although these results are intuitively believable, a longitudinal study that examines mentors’ initial expectations before they begin mentoring, and then follows volunteers forward in time to assess changing perceptions and their effects on relationship length and termination, could provide more valid results. No study has explicitly linked mentors’ perceptions of the benefits they receive from mentoring to the length of their relationships with youth or tenure with a mentoring program. However, in the 3-year SMILE, Karcher et al. (2006) examined 151 mentors’ reports of outcomes from school-based mentoring, based on the motivational benefits described by Clary et al.’s (1998) VFI. They found that mentors

404   PRACTICE AND PROGRAMMATIC CONSIDERATIONS who reported engaging in developmental activities with their mentees also reported receiving more benefits from their service, as compared to mentors who reported engaging in instrumental activities. Developmental mentoring activities were defined as those that focus on the relationship itself and include engaging in creative activities, listening and learning from each other through social interaction, and discussing the relationship. Conversely, instrumental activities included discussions of the youth’s behavior, attendance at school, and other social issues. As Stukas and Tanti (2005) reviewed, developmental activities have also been shown to be better predictors of youth outcomes than instrumental activities. Karcher et al.’s findings indicate that developmental activities also may lead mentors to experience enhanced self-esteem, assistance with their own problems, greater understanding of other people and the world, and opportunities to express important values and to meet social needs. As such, programs that identify mentors who are predisposed to use developmental activities or that use training and support to guide mentors toward such activities might help mentors achieve their own goals and increase satisfaction and retention. As research demonstrating that youth mentoring does provide benefits to youth, especially to the extent that best practice is followed (Rhodes & DuBois, 2008), has reached a critical mass, we hope that researchers will turn their attention to research focused on outcomes for mentors and the factors that might serve to keep them involved (or lead them to terminate a relationship). Research of this type, especially grounded in solid theory and relevant empirical studies, can provide a better basis for identifying, recruiting, and sustaining the most important ingredients of mentoring programs.

Practice We agree with Rhodes and DuBois’s (2008) analysis that practices used by organizations that engage adults and youth in mentoring relationships should be supported by theoretical principles and research evidence, and this includes practices designed to recruit and retain mentors. Below we offer some recommendations, based on our reading of theory and research on volunteerism, generally, and on youth mentoring, more specifically, for three constituent groups involved in mentoring: (a) adults who are considering becoming mentors for youth, (b) organizations that support and supervise mentoring relationships, and (c) communities that wish

to develop habits and traditions of youth mentoring in their areas. But first, writing from our own theoretical perspective, the functional approach to volunteerism (e.g., Clary et al., 1998; Stukas et al., 2008), we are encouraged that our recommendations for practice (also detailed in the first edition of the Handbook; Stukas & Tanti, 2005) have been picked up by many organizations (U.S. Department of Education, 2006). Simply put, our advice is to recruit mentors by targeting the motivations held by potential mentors and then to increase their satisfaction and retention by ensuring that their experiences allow mentors to achieve desired benefits. For the potential mentor, this means understanding one’s own motivations in order to more efficiently and effectively have them fulfilled. For the organization, it means assessing incoming mentors so that motive-relevant benefits can be provided and to allow a strategy of understanding current, successful mentors in order to recruit new mentors with a similar profile. Finally, for the community, it means identifying predominant motives in the local population as a first step in developing a marketing plan or even in designing a new mentoring program with structures in place to enable mentors to meet their primary needs. We now expand these recommendations to incorporate lessons learned from the broader theory and research on mentoring and volunteering. Recommendations for Potential Mentors Some mentors will be better prepared than others for the challenges of youth mentoring. As Walker (2007) pointed out, as many as one in four mentors may not be adept at establishing relationships with youth. In addition, he estimates that establishing an effective relationship would require a mentor to commit 100 hours across a 1-year period, a significant barrier for many would-be mentors. So, before committing themselves, those considering mentoring should think carefully about the potential costs vis-à-vis the potential benefits of mentoring. This is a particularly important consideration to make given the substantial negative effects of relationships that end early (Grossman & Rhodes, 2002). Mentors may also need to consider their intended approach for mentoring for a given context (e.g., school vs. community based) or population of youth (e.g., older vs. younger mentees). Although both goal-oriented (instrumental) and developmental approaches have a place, the extant evidence suggests the latter may lead more clearly to benefits for mentors and mentees (Karcher et al., 2006;

Mentor Recruitment and Retention   405 Karcher, Herrera, & Hansen, 2010). Organizations may offer advice about developmental and instrumental activities in their orientation and training schemes, but only mentors themselves may know whether they feel comfortable focusing primarily on befriending youth and not on actively trying to change them. In addition, as Spencer’s (2007) work shows, some mentors may find it difficult to bridge cultural divides, to see things from their youth’s perspective, or to feel empathy for the youth’s situation. Finally, mentors may wish to consider why they are getting involved in mentoring and whether they seek benefits for themselves or are getting involved purely because of pressure from others, which may not be as healthy (Gebauer et al., 2008). Recommendations for Organizations After suggesting that there may be an upper limit on the number of people willing and able to become involved as youth mentors, especially of the most challenging youth, Walker (2007) highlighted several innovative recruitment practices. For example, organizations can form partnerships with corporations, schools, and community or faithbased organizations. As Murphy and Ensher (2006) suggested, there are benefits all around from such partnerships, with corporations improving their images, developing staff morale and abilities, and fulfilling their responsibilities to the communities that support them. Schools and communities also benefit from the improved circumstances of the youth who live and work within them. Walker added that holding mentoring meetings at a workplace can reduce the costs of mentoring (in line with social marketing principles focused on “price”) and that such meetings could even involve one or more mentors with one or more youth to increase efficiency or reduce anxiety. Walker (2007) suggested that partnerships with faith-based organizations may also provide a route to recruiting mentors, perhaps particularly for youth with challenging characteristics, such as some children of prisoners. But he also presented the case of Friends of the Children, a Portland, Oregon, organization that provides mentors to troubled 1st graders, which decided that volunteers were not sufficient to provide support to these kids and instead decided to pay its mentors. As indicated on the organization’s website (http://www .friendschildren.org/portland/friend.html), these full-time paid mentors take on a caseload of eight children, seeing each child 4 hours per week. Such an option does not seem cost-effective for most mentoring programs.

MENTOR (2009) lists as a benchmark goal that a mentoring program “engages in recruitment strategies that realistically portray the benefits, practices, and challenges of mentoring in the program” (p. 4). This meshes with the limited but intuitively believable research suggesting that mentors with unrealistic expectations may have failed matches (Madia & Lutz, 2004; Spencer, 2007), and we recommend ensuring that potential mentors are fully aware of the challenges that they may face. A hardsell to capture ambivalent mentors might be successful in the short term but disastrous in the long run, particularly if a child experiences yet another failed relationship. In his guidebook for mentoring programs associated with the U.S. Department of Education’s Office of Safe and Drug-Free Schools, Garringer (2006) offered many suggestions for recruitment strategies, with a strong emphasis on engaging in continual recruitment, with all staff, mentors, youth, parents, and teachers talking up the program at every opportunity. He also recommended public service announcements, corporate presentations, notices before movies, posters, flyers, websites, and targeting groups (churches, fraternities, businesses) that might be a conduit for recruiting. Not unexpectedly, Garringer advised mentoring organizations to be persistent, as they will need to contact many people to find just a few interested and appropriate mentors. Ballasy (2004), under the auspices of the Northwest Regional Educational Library, took a more explicit social marketing approach in her workbook for mentoring programs. She emphasized marketing as facilitating an exchange, by highlighting what mentors will receive from mentoring as well as what they will give to organizations and youth. Best practice in this area involves careful planning in stages (preplanning, planning, and postplanning) with the involvement of important stakeholders (e.g., mentors themselves, the board of directors). Ballasy offered a large amount of practical advice to help with the launch of a recruitment campaign, although the link to research that undergirds good practice was not explicit. Both Garringer (2006) and Ballasy (2004) discussed the importance of branding as a tool to help organizations recruit mentors. Branding was defined as “the intentional actions of a program or organization to build recognition and reputation with its customers” (Ballasy, p. 48). Establishing a well-known brand can involve espousing a clear and consistent mission supported by the program name, logo, slogans, and graphics. Organizations that are recognizable in the community, with clear “brands” and positive reputations, may be better

406   PRACTICE AND PROGRAMMATIC CONSIDERATIONS able to communicate their overall mission and its importance to their volunteers in a way that fosters volunteers’ pride in and commitment to the organization (Boezeman & Ellemers, 2007). Recommendations for Communities In the context of her social marketing approach, Ballasy (2004) also suggested that mentoring organizations should conduct an “environmental analysis” before beginning a marketing campaign, focusing on both internal factors (program reputation, resources, and capacity) and external factors (community attitudes and demographics, strength and activities of the nonprofit sector) that may affect recruitment. These latter concerns may loom larger for some communities than for others, because just as not every person is right for youth mentoring, not every community will be able to sustain a youth mentoring program. Given Omoto and Snyder’s (2010) point that having a strong sense of community can increase willingness to volunteer, organizations situated in communities that lack a strong sense of cohesion and identity, perhaps due to residential mobility (e.g., Oishi et al., 2007), may need

to engage in community-building before the recruitment of a stable pool of mentors can be successful. Wandersman et al. (2006) also suggested that a coalition of community stakeholders conducts a needs assessment to determine whether local organizations have the capacity and resources to meet the identified needs of youth (whose specific population must also be specified). Wandersman et al. pointed to resources that are readily available to assist with the development and implementation of mentoring programs (e.g., from the MENTOR or Mentoring Resource Center websites), but suggested that new programs will need to rely on increases in capacity (skills and motivation) that come from an iterative or cyclical process involving feedback from evaluations of program processes and access to research that indicates which mentoring practices lead to the best outcomes for youth (and for mentors and the community, as well). In Table 27.1, we present a checklist of guidelines that mentoring programs may wish to follow to incorporate lessons from theory and research on volunteerism, generally, and youth mentoring, specifically, into their practices. We summarize these points in the conclusion.

Table 27.1   Checklist for Practitioners Topic

Recommendations

Recruiting mentors

Use a range of approaches, differentiating between those volunteers who are already prone to volunteer, those who are open to “good offers,” and those who are currently resistant to volunteering. Match recruitment messages to the motivations of potential mentors. Present the challenges and difficulties of mentoring (as well as the benefits) to prevent disillusionment, disappointment, and resentment. Consider screening in mentors for positive characteristics, such as for relevant skills and abilities, rather than only screening out those with negative characteristics. Maintain knowledge of the relevant community, the potential mentors within it, and the connections between people (social capital, sense of community) that may be leveraged to support programs.

Retaining mentors

Use training and orientation sessions to guide mentors toward actions and activities that are known to predict positive outcomes, such as taking the perspective of a youth or engaging in developmental activities. Encourage mentors to identify with the program and build a sense of community, perhaps through an organizational brand and a clear mission statement, as a way to encourage pride and commitment. Offer sufficient support to mentors to allow them to feel respected. Evaluate and ensure that mentors’ own needs and goals are satisfied by mentoring as well as any other activities that the organization offers to support mentors and their relationships with youth.

Mentor Recruitment and Retention   407

Conclusion We recommend that mentoring programs think carefully about the target audience for their recruitment efforts and whether messages are directed to people who are already interested in becoming mentors. Volunteer mentors may come with a range of motivations; however, some research suggests that prosocial behavior engaged in because of pressure rather than pleasure may be less healthy for volunteers. Similarly, being required to volunteer or rewarded monetarily for doing so may undermine future motivation. Fortunately, the majority of volunteers, including mentors, report benefits from their activities, and we suggest that recruitment strategies target the benefits that mentors are motivated to obtain (which may be different for different volunteers), as long as efforts are made to ensure that those benefits are achievable. Organizations should do all they can to ensure that potential mentors do not have unrealistic expectations when they sign up. Organizations may also wish to focus on identifying mentors with positive skills and abilities that will best promote long and satisfying relationships with mentees. However, the literature on volunteerism tells us more about who volunteers than about who makes an effective volunteer. Research looking at mentors specifically, where outcomes and effectiveness may be easier to identify, is still sparse. Perspective-taking and empathy, secure attachment style, and possibly agreeableness, extraversion, and individualism (or personal responsibility) seem possible contenders for positive predictors of mentoring success, along with attitudes and values that demonstrate concern for and valuing of youth in need. To ensure commitment of volunteers, we recommend that organizations build a sense of community within the program that allows mentors to identify with the organization’s mission (and brand), to feel bonded with other mentors, and to feel pride in their association with a valuable endeavor. Guaranteeing that mentor support services leave volunteers feeling respected should also improve retention. Finally, we encourage mentoring organizations to be aware of their local communities and the fact that different communities may be differentially able to offer strong and committed mentors. Community assessments can form the basis of better strategy for mentor recruitment, suggesting methods that involve active partnerships with corporations, schools, and faith-based and community organizations. In closing, we note that, to date, research dollars have often flowed to those who are engaged in demonstrating the effectiveness of mentoring programs for youth. However, now that it is becoming

clearer that mentoring can work and there is evidence for specific practices that are likely to produce the best outcomes, we hope that researchers will turn their attention toward better understanding the features of persons that attract them to mentoring and that make them suitable for forming solid bonds with youth, as well as the situational and environmental features that can allow mentoring to blossom and grow in our communities.

References Ballasy, L. (2004). Marketing for the recruitment of mentors: A workbook for finding and attracting vol­ unteers. Portland, OR: Northwest Regional Educational Library. Retrieved from http://educa tionnorthwest.org/webfm_send/178 Batson, C. D. (1991). The altruism question: Toward a social psychological answer. Hillsdale, NJ: Erlbaum. Batson, C. D., Eklund, J. H., Chermok, V. L., Hoyt, J. L., & Ortiz, B. G. (2007). An additional antecedent of empathic concern: Valuing the welfare of the person in need. Journal of Personality and Social Psycho­ logy, 93, 65–74. Boezeman, E. J., & Ellemers, N. (2007). Volunteering for charity: Pride, respect, and the commitment of volunteers. Journal of Applied Psychology, 92, 771–785. Carlo, G., Okun, M. A., Knight, G. P., & de Guzman, M. R. T. (2005). The interplay of traits and motives on volunteering: Agreeableness, extraversion, and prosocial value motivation. Personality and Individual Differences, 38, 1293–1305. Clary, E. G., & Orenstein, L. (1991). The amount and effectiveness of help: The relationship of motives and abilities to helping behavior. Personality and Social Psychology Bulletin, 17, 58–64. Clary, E. G., Snyder, M., Ridge, R. D., Copeland, J., Stukas, A. A., Haugen, J., & Miene, P. (1998). Under­ standing and assessing the motivations of volunteers: A functional approach. Journal of Personality and Social Psychology, 74, 1516–1530. Clary, E. G., Snyder, M., & Worth, K. (2003). The volun­ teer organization environment: Key dimensions and distinctions. Report to the Aspen Institute Nonprofit Sector Research Fund. Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85. Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A metaanalytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125, 627–668. Dicken, C., Bryson, R., & Kass, N. (1977). Companionship therapy: A replication in experimental community psychology. Journal of Consulting and Clinical Psychology, 45, 637–646. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for

408   PRACTICE AND PROGRAMMATIC CONSIDERATIONS youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. Einolf, C. J. (2008). Empathic concern and prosocial behaviors: A test of experimental results using survey data. Social Science Research, 37, 1267–1279. Eisenberg, N. (2010). Empathy-related responding: Links with self-regulation, moral judgment, and moral behavior. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior: The better angels of our nature (pp. 129–148). Washington, DC: American Psychological Association. Finkelstein, M. (2009). Intrinsic vs. extrinsic motivational orientations and the volunteer process. Personality and Individual Differences, 46, 653–658. Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item-response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. Garringer, M. (2006). Effective mentor recruitment: Getting organized, getting results. Folsom, CA: Mentoring Resource Center. Retrieved from http:// educationnorthwest.org/webfm_send/172 Gebauer, J. E., Riketta, M., Broemer, P., & Maio, G. R. (2008). Pleasure and pressure based prosocial motivation: Divergent relations to subjective well-being. Journal of Research in Personality, 42, 399–420. Gillath, O., Shaver, P. R., Mikulincer, M., Nitzberg, R. E., Erez, A., & Van Ijzendoorn, M. H. (2005). Attach­ ment, caregiving, and volunteering: Placing volunteerism in an attachment-theoretical framework. Personal Relationships, 12, 425–446. Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assess­ ment, 4, 26–42. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Herman, K. C., & Usita, P. M. (1994). Predicting Big Brothers/Big Sisters volunteer attrition with the 16 PF. Child and Youth Care Forum, 23, 207–211. Herrera, C., Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, M. (2008). High school students as mentors: Findings from the Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Karcher, M. J. (2004, March). The motivations of Hispanic mentors and the activities they use in their mentor­ ing of Hispanic youth. Presentation at the Latino Research Initiative Symposium on Youth Mentoring, University of Nebraska-Lincoln. Karcher, M. J. (2009). Increases in academic connectedness and self-esteem among high school students who serve as cross-age peer mentors. Professional School Counseling, 12(4), 292–299. Karcher, M. J., Benne, K., Gil-Hernandez, D., Allen, C., Roy-Carlson, L., Holcomb, M., & Gomez, M. (2006, June). The Study of Mentoring in the Learning Environment (SMILE): A functional approach to predicting mentor satisfaction from mentoring

interactions. Paper presented at the 14th annual meeting of the Society for Prevention Research, San Antonio, TX. Karcher, M. J., Davidson, A., Rhodes, J. E., & Herrera, C. (2010). Pygmalion in mentoring: The role of teenage peer mentors’ attitudes in shaping their mentees’ outcomes. Applied Developmental Science, 14, 1–16. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. New Directions in Youth Development, 126, 51–70. Kemmelmeier, M., Jambor, E. E., & Letner, J. (2006). Individualism and good works: Cultural variation in giving and volunteering across the United States. Journal of Cross-Cultural Psychology, 37, 327–344. Konrath, S., Fuhrel-Forbis, A., Lou, A., & Brown, S. (2012). Motives for volunteering are associated with mortality risk in older adults. Health Psychology, 31, 87–96. Madia, B. P., & Lutz, C. J. (2004). Perceived similarity, expectation-reality discrepancies, and mentors’ expressed intention to remain in Big Brothers/Big Sisters programs. Journal of Applied Social Psychology, 34, 598–623. MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. Retrieved from http://www.mentoring.org/find_ resources/elements_of_effective_practice/ Metz, E. C., & Youniss, J. (2005). Longitudinal gains in civic development through school-based required service. Political Psychology, 26, 413–437. Murphy, S. E., & Ensher, E. A. (2006). Best practices of prosocial organizations in youth development. In E. G. Clary & J. E. Rhodes (Eds.), Mobilizing adults for positive youth development: Strategies for closing the gap between beliefs and behaviors (pp. 117–135). New York: Springer. O’Connor, R. (2006). Mentoring in America 2005: A snap­ shot of the current state of mentoring. Alexandria, VA: MENTOR. Retrieved from http://www.mentoring .org/start_a_program/evaluation/2005_national_poll/ Oishi, S., Rothman, A. J., Snyder, M., Su, J., Zehm, K., Hertel, A. W., . . . Sherman, G. D. (2007). The socioecological model of procommunity action: The benefits of residential stability. Journal of Personality and Social Psychology, 93, 831–844. Omoto, A. M., & Snyder, M. (2010). Influences of psychological sense of community on voluntary helping and prosocial action. In S. Stürmer & M. Snyder (Eds.), The psychology of prosocial behavior: Group processes, intergroup relations, and helping (pp. 224–243). Oxford, UK: Wiley. Planty, M., Bozick, R., & Regnier, M. (2006). Helping because you have to or helping because you want to? Sustaining participation in service work from adolescence to young adulthood. Youth & Society, 38, 177–202. Rhodes, J. E., & DuBois, D. L. (2008). Mentoring relationships and programs for youth. Current Directions in Psychological Science, 17, 254–258.

Mentor Recruitment and Retention   409 Roaf, P. A., Tierney, J. P., & Hunte, D. E. I. (1994). Big Brothers/Big Sisters: A study of volunteer recruitment and screening. Philadelphia: Public/Private Ventures. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Stolinski, A., Ryan, C. S., Hausmann, L. R. M., & Wernli, M. A. (2004). Empathy, guilt, volunteer experiences, and intentions to continue volunteering among buddy volunteers in an AIDS organization. Journal of Applied Biobehavioral Research, 9, 1–22. Stukas, A. A., Daly, M., & Cowling, M. J. (2005). Volun­ teerism and social capital: A functional approach. Australian Journal on Volunteering, 10(2), 35–44. Stukas, A. A., Snyder, M., & Clary, E. G. (1999). The effects of “mandatory volunteerism” on intentions to volunteer. Psychological Science, 10, 59–64. Stukas, A. A., Snyder, M., & Clary, E. G. (2008). The social marketing of volunteerism: A functional approach. In C. P. Haugtvedt, P. Herr, & F. Kardes (Eds.), Handbook of consumer psychology (pp. 959–979). Mahwah, NJ: Erlbaum. Stukas, A. A., & Tanti, C. (2005). Recruiting and sustaining volunteer mentors. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 235–250). Thousand Oaks, CA: Sage. Stukas, A. A., Worth, K. A., Clary, E. G., & Snyder, M. (2009). The matching of motivations to affordances

in the volunteer environment: An index for assessing the impact of multiple matches on volunteer outcomes. Nonprofit and Voluntary Sector Quarterly, 38, 5–28. Stürmer, S., Snyder, M., & Omoto, A. M. (2005). Prosocial emotions and helping: The moderating role of group membership. Journal of Personality and Social Psychology, 88, 532–546. Sundeen, R. A., Raskoff, S. A., & Garcia, M. C. (2007). Differences in perceived barriers to volunteering to formal organizations: Lack of time versus lack of interest. Nonprofit Management and Leadership, 17, 279–300. U.S. Department of Education. (2006). Mentoring fact sheet # 8. Retrieved from http://educationnorthwest .org/webfm_send/170 Vohs, K. D., Meade, N. L., & Goode, M. R. (2006). The psychological consequences of money. Science, 314, 1154–1156. Walker, G. (2007). Mentoring, policy and politics. Philadelphia: Public/Private Ventures. Retrieved from http://www.ppv.org/ppv/publications/assets/224_ publication.pdf Wandersman, A., Clary, E. G., Forbush, J., Weinberger, S. G., Coyne, S. M., & Duffy, J. L. (2006). Community organizing and advocacy: Increasing the quality and quantity of mentoring programs. Journal of Commu­ nity Psychology, 34, 781–799.

28  MENTOR SCREENING AND YOUTH PROTECTION Sarah E. Kremer and Becky Cooper

Introduction The primary duty of youth-service programs is to care for the well-being, healthy development, and growth of young persons. Mentoring programs seek to fulfill this duty by pairing each participating youth with one or more specific adults (or, in some instances, older peers). These mentors, most of whom are volunteers, are typically expected to serve as models for healthy relationships, show youth that they have valuable gifts to share with others, and illustrate through action that someone cares deeply about their well-being. These relationships carry with them a potential for deep and lasting enrichment to young people and their development. At the same time, because of their direct and close relationshipbased approach to working with youth, mentoring programs present risks for youth to be deliberately harmed that are not necessarily found in other types of youth services or programs. In some instances, for example, child molesters have specifically sought out involvement in youth mentoring programs, presumably in part because such programs provide unsupervised access to vulnerable youth. Organizations in this way may inadvertently assist perpetrators by helping them to gain the trust of parents and by serving to legitimize the relationship (Arevalo, Cooper, & Smith, 2006), as many parents or caregivers are inclined to trust the program without asking many questions. Along with the well-demonstrated power of mentoring programs to benefit youth through close and often intimate relationships (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011) comes a fundamental ethical responsibility to protect the youth involved from harm (for a broader discussion of ethics in youth mentoring, see Rhodes et al., this volume, Chapter 35).

Media articles over the years have documented numerous substantiated instances of sexual abuse of youth by adults with whom they have been paired in formal mentoring programs (Terry & Tallon, 2004). Such instances of youth being abused in mentoring programs are generally assumed to be relatively infrequent. Reliable estimates of their prevalence are not available. Moreover, as is discussed in this chapter, there is remarkably little research in general to inform mentor screening and monitoring practices and policies directed toward youth safety in mentoring programs. In addition to this lack of evidence, programs face numerous other obstacles in creating policies and procedures in this area. These include a broad denial of the potential for child sexual abuse within communities, especially with regard to the possibility that well-known and well-respected individuals such as those volunteering to serve as mentors can be perpetrators (van Dam, 2001, 2006). Likewise, given such perceptions, victims or those responsible for their care are often reluctant to report instances of abuse. One survey of adults who were sexually victimized by adults as children reported that 91% did not disclose the abuse when it was happening (Arevalo et al., 2006). Programs, too, may feel pressure to conceal incidents of potential sexual abuse of mentees for fear of potential liability and loss of credibility. Furthermore, as programs strive to ensure youth safety, they face a need to balance protective measures with other important programmatic concerns. There may be a tension, for example, between potentially useful mentor screening steps and the ability to complete the application process within a reasonable time before interest is lost and the potential mentor has moved on to another opportunity 411

412   PRACTICE AND PROGRAMMATIC CONSIDERATIONS (Roaf, Tierney, & Hunte, 1994; Schmiesing & Henderson, 2001). Agency staff must balance their own cautious attitude and potential suspicions toward applicants with the openness necessary to build positive relationships and trust with mentors who ultimately are approved (K. C. Herman, 1995; Roaf et al., 1994). Programs, as they develop their protocols, also need to determine when to encourage staff to be open and nonjudgmental and when to trust staff’s informed intuition regarding potential risks to youth safety. When engaging youth and families, programs may often find themselves walking a fine line between communicating to parents or caregivers the seriousness of the possibility of child sexual abuse occurring in their program despite their stringent screening standards while at the same time conveying a reassuring attitude about the benefits of mentoring (Schmiesing & Henderson, 2001). Finally, but not least significantly, programs must balance the pressure to serve sufficient numbers of youth, often required by funding sources, with the priority of ensuring that only safe mentors and relationships are allowed. From these challenges emerge the opportunity to strengthen screening and monitoring processes and develop mandatory standards for the field to use effectively. With such standards, programs can then customize tools and processes for their particular settings and populations. Opportunity also lies in establishing standards that do not reduce mentor supply, divert agency resources, or require programs to engage in unethical practices (such as excluding mentors based on lifestyle preferences), all common misperceptions among professionals. These considerations underscore the importance of taking a proactive approach to ensuring youth safety within mentoring or any program intended to benefit young people: Youth safety “should not be about abuse response . . . . We cannot afford to wait until the child is victimized to take action” (Patterson, 2008). Thorough screening and monitoring demonstrate a program’s commitment to youth and their families; it shows that an agency takes youth safety seriously and emphasizes that the priority of services is focused on the youth, not the mentors. Issues of mentor screening and youth protection, furthermore, are issues applicable to all types of programs and settings in which adults (or older peers) have contact with young persons for purposes of mentoring—e.g., not only communitybased, but also site-based mentoring—as all provide an introduction and access to youth that have the potential to be exploited for harm. Child sexual abuse, for the purposes of this chapter, is defined as sexual activities involving a

child for sexual stimulation and includes a violation of a trusting relationship with unequal power. Abusive activities are carried out using secrecy, force, or trickery (Arevalo et al., 2006). The initiators of this kind of sexual activity with children or youth are referred to as perpetrators (Olson, Daggs, Ellevold, & Rogers, 2007), whether they are adults or youth themselves. In the majority of instances, the perpetrators have been male. The information presented in this chapter focuses on male perpetrators who sexually abuse boys, as this is the dynamic for which the most research is available, due to its prevalence in practice. In this chapter, we examine mentor screening and other practices intended to help ensure youth safety in mentoring programs. We begin with an overview of relevant background information as well as literature from allied fields that addresses safety considerations when bringing both paid and unpaid adults into contact with youth. Next, we review research on mentor screening and youth protection in mentoring programs. Drawing also on what has been learned from practitioner experience, we then offer recommendations for practice and policies in the area of youth safety in mentoring programs. This chapter does not address the negative effects that can stem from prematurely terminated and/or low-quality relationships that can result from ineffective, but not necessarily unsafe mentors (for a discussion of issues relating to relationship termination, see Spencer & Basualdo-Delmonico, this volume, Chapter 32). Finally, we encourage readers to keep in mind throughout our discussion that no process is guaranteed to remove the risk of unsafe mentors (Roaf et al., 1994). The recommendations offered in this chapter, while intended to provide a useful foundation for youth protection in mentoring programs, should not be regarded as sufficient for fully realizing this goal, nor do they apply to every program or every state’s legal requirements.

Theory Child sexual abuse occurs far too often in our culture. Estimates state that 1 in every 4 or 5 girls and 1 in every 9 or 10 boys has been sexually abused (Abel & Harlow, 2001; Finkelhor, Ormrod, & Chaffin, 2009; Terry & Talon, 2004). In the vast majority of child sexual abuse incidents, the perpetrator is known to the youth; estimates range from 74% to 90% (Arevalo et al., 2006; Olson et al., 2007; Terry & Tallon, 2004). There is a preponderance of evidence linking childhood sexual abuse to a variety of psychiatric, physical, and developmental

Mentor Screening and Youth Protection   413 issues including posttraumatic stress disorder, sexually transmitted diseases, teen pregnancy, increased drug and alcohol use, obsessive-compulsive disorder, depression, personality disorders, increased suicidality, increased prevalence of parenting difficulties, increased utilization of health and social services, high cost of lost work days, psychiatric issues, and interference with the normal development of adult sexual behavior (Abel & Harlow, 2001; Arevalo et al., 2006; Middlebrook & Audage, 2008). The financial cost to society is also great: A 2007 report conservatively estimated the annual cost of all types of child abuse and neglect as $103.8 billion (Prevent Child Abuse America, 2007). The U.S. Department of Education (Shakeshaft, 2004) issued research concerning the occurrence of child sexual abuse in school settings, finding that nearly 10% of students in grades 8–11 reported educator sexual misconduct involving children. Sullivan and Beech (2002) examined nearly 700 incidents in the United Kingdom that involved “professional perpetrators,” those who use their work—both paid and voluntary—as a cover for targeting and sexually abusing children. Terry and Talon (2004) reviewed findings on child sexual abuse from several youthserving organizations, including the Boy Scouts of America (updated by Boyle, 2010), Big Brothers Big Sisters (BBBS; the only mentoring-specific organization included), and the YMCA. This survey found incidents with more similarities than differences across these organizations: Sexual abuse occurred in isolated settings (while camping or at the perpetrator’s home), and the perpetrators targeted and groomed—seduced using specific techniques— emotionally vulnerable youth, were known in the community, and had multiple victims, including those outside of the agency in question. In every case of child sexual abuse from these studies, perpetrators were not identified by background checks (or their criminal histories were never obtained) and they found vulnerable children through youth-service programs; groomed their communities, organizations, parents or caregivers, and victims; sought opportunities for unmonitored access and isolated activity; and formed relationships over a significant period of time before beginning the abuse. These findings are, unfortunately, not surprising: “A percentage of predators will target child-service groups because they provide access to samples of highly vulnerable children and often there are opportunities for isolated access. Many of these children have already been molested, making them more vulnerable to the predator” (Arevalo et al., 2006, p. 34). Perpetrators cannot be identified by simple comparison with a profile or completion of a

checklist, or, as discussed previously, a criminal background check. They are diverse in their socioeconomic backgrounds, levels of education, religious preferences, ethnic heritage, and age (Arevalo et al., 2006). The work of Groth (1979), a pioneer in the scientific study of male child sexual abuse perpetrators, was a starting point for researchers in developing theory. Many “red flags,” based on behavioral characteristics from Groth’s research, have been cited by researchers and law enforcement officials over the years. Behaviors and traits that have been observed in identified perpetrators include employment in menial work; a passive personality; little or no social contact with peers; child pornography collecting; relating better to children; feeling misunderstood and discriminated against by society for their beliefs regarding a love of children; having been sexually abused as a child; and a history of alcoholism, depression, and frequent moving. Theory and available research suggest that no single red flag in and of itself indicates that an applicant to a mentoring program is a perpetrator. Theoretically, however, a combination of several indicators could be useful as a basis for concern. Such indicators might include a lack of balance in one’s life (e.g., few or no peer relationships), extreme behavior (e.g., impatience with or attempting to manipulate or circumvent the application process or being overly cooperative), inappropriate behavior (e.g., expressing discomfort about information shared or concern about other applicants), overinvolvement with children and under-involvement with adults, a focus on personal needs (e.g., wanting a child-focused relationship to fill an empty place in his life), unhealthy attitudes (e.g., children being allowed to make their own decisions about sexual activity), problematic personal interests (e.g., activities that involve isolated access, photographing children), and problematic background indicators (e.g., a job requiring extensive travel, family history of alcoholism) (Arevalo et al., 2006). According to a new theory illustrating a process used by perpetrators, child sexual predators use a communicative process of entrapment to lure and then manipulate their victims into an ongoing sexual relationship (Olson et al., 2007). This research identified several characteristics that the authors anticipate child sexual abusers would exhibit, including low self-esteem, interpersonal inadequacy, a lack of empathy, a fear of intimacy, an inability to form relationships with adults, and a history of poor parent-child relationships. This research-informed theory echoes the strategic process that other researchers (Arevalo et al., 2006; Terry & Tallon,

414   PRACTICE AND PROGRAMMATIC CONSIDERATIONS 2004; van Dam, 2001, 2006) have proposed many perpetrators to employ. This process involves grooming using desensitization and reframing, physical and mental isolation, and approaching the victim in an attempt to see if sexual contact is possible by creating an environment of secrecy, threats, or bribes. The heart of effective mentoring programming is engaging in relationship-building activities with the goal of forming a close relationship. This parallels a process that also occurs in the perpetrator practice of grooming to gain trust, referred to as “deceptive trust development” (Olson et al., 2007, p. 236). Theoretically, youth-serving agencies may be able to implement effective processes to screen out potential child sexual abuse perpetrators by applying knowledge from the amalgamation of research and evidence-based findings on strategies perpetrators employ, along with general behavior characteristics and patterns. Through a careful analysis, program staff may be able to identify characteristics that potentially indicate an unsafe mentor (Arevalo et al., 2006; Lanning, 2001; Olson et al., 2007; van Dam, 2001, 2006). One approach suggested to be particularly effective in screening involves creating a holistic portrait of the applicant through a confluence of using written materials submitted by the applicant, external documentation, observations and impressions by staff specifically compared to established indicators of concern, and “informed intuition” or staff judgment (Arevalo et al., 2006; Roaf et al., 1994). Intuition, gut feelings, hunches, or unconscious intelligence has been defined as a judgment that appears quickly in consciousness even if one is not fully aware of the underlying reasons but it is strong enough to act upon (Gigerenzer, 2007). Informed intuition is a similar reaction to a person, in conjunction with having relevant knowledge (Arevalo et al., 2006). As a law enforcement official with extensive experience researching child sexual abuse perpetrators states, “I have investigated hundreds of child predator cases involving thousands of victims. In the case of every single victim, there was a woman—the mother, agency staff, a neighbor, a teacher, an aunt—who looked back and said, ‘I thought something wasn’t right. I had a funny feeling about him’” (Arevalo et al., 2006, p. 34). Recommendations for screening and monitoring in youth mentoring programs focus primarily on a highly trained staff because educating professionals about preventing child sexual abuse has been found to be the most effective strategy, as compared to only educating children and/or parents or caregivers (Arevalo et al., 2006; Finkelhor et al., 2009). However, forming a “prevention partnership” with

parents or caregivers, going beyond mere education, should also theoretically decrease the risk of child sexual abuse. Programs can extend safety and protection efforts by including parents or caregivers as monitors and educators themselves (Wurtele & Kenny, 2010), though no research recommends any specific educational curriculum or protocol.

Research A literature search for research on mentor screening and youth protection from 1980 to 2010 as it relates to mentoring program participation was conducted in Medline, Web of Science, Cochrane Database, and PsycINFO using keywords such as youth-serving organizations, child abuse, maltreatment, big brothers and big sisters, and screening tools for child abuse. Additionally, reference lists of retrieved articles were reviewed manually for more studies. Literature was also reviewed from Friends for Youth, Inc.’s resource library, which contains articles and books used to support publication of SAFE (Screening Applicants for Effectiveness): Guidelines to Prevent Child Molestation in Mentoring and Youth-Serving Organizations (Arevalo et al., 2006). Research Studies Only three studies focusing on screening and/or monitoring tools used within youth mentoring programs were identified. One study focused on a specific measure used for identifying potential physical abusers (K. C. Herman, 1995). Another study focused on a survey of screening steps, including the use of psychological tests, in identifying inappropriate mentor candidates (Roaf et al., 1994). Another tool, the Diana Screen, is designed to help programs better identify applicants who may be of sexual risk to children. There is currently no published research on the use of the tool within youth mentoring programs or other contexts. It is included in this review, however, because it illustrates the types of measures that are candidates for future investigation. K. C. Herman (1995) investigated the Child Abuse Potential Inventory (CAP) as a screening tool in one BBBS agency. The CAP consists of an abuse scale, two special scales (ego strength and loneliness), six subscales (distress, rigidity, unhappiness, problems with child and self, problems with family, and problems with others), and three validity indexes. Studies evaluating the CAP found that it was useful in distinguishing between physical abusers and nonabusers. The CAP scores in this study were thought to distinguish mentors who would be

Mentor Screening and Youth Protection   415 more nurturing and those who “possess an array of personality characteristics that may be dangerous or inappropriate in BBBS volunteers” (p. 94), though, as the author noted, the CAP was not designed to screen out child sexual abusers. This sample consisted of 73 White, college-educated applicants and volunteers, ranging in age from 18 to 70 years. The volunteer applicant sample was compared to the norm population for the CAP and also compared to evaluations by program staff who rated volunteers as “excellent” or “bad.” Eight “bad” and 12 “excellent” volunteers were identified within the sample based on staff ratings. The mean score for excellent volunteers was similar to scores for nurturing parents, and the mean abuse score for applicants who dropped out of the screening process was comparatively high. A second study examined 21 males in a sexual offender treatment center who were administered the CAP. The average CAP abuse score was significantly higher than the norm scale from the manual but was still below the cut-off score indicating a potential physical child abuser or someone with future physical child abuse potential. The authors noted several shortcomings of the study, including the lack of a matched control group of non-sexualoffending males. In summary, the authors concluded that though the CAP “may screen some sexual offenders against minors, it is insufficient.” They recommended incorporating the CAP into an existing screening process with other tools. In the second example (Roaf et al., 1994), eight BBBS agencies participated in a general volunteer recruitment and screening study. These agencies collected information from applicants over a 6-month period and then recorded each applicant’s progress over the following 4 months. The authors visited four agencies and interviewed staff from all eight agencies. Additionally, the authors conducted focus groups of applicants. The psychological tests reviewed in this study (Child Abuse Potential Inventory, Minnesota Multi-Phase Inventory, California Psychological Inventory, Hogan Personality Inventory, Clinical Analysis Questionnaire, and a variant of PSI-5S) were required by the BBBS’s primary insurer; alternatively, agencies could provide child abuse education and training. Three agencies required the psychological tests and one agency used the test results as a permanent disqualifier; all three rejected 27.4% of applicants solely on the basis of test scores. Though a fairly high percentage of applicants were rejected because of a psychological measure, the results from this research indicated that staff thought these tools were less useful and that applicants were often confused or bothered by having to complete the measures involved.

The Diana Screen (Abel Screening, 2009, n.d.), an online test designed to eliminate high-risk adults and reduce sexual risk to children in a program, was in development for 18 years, and its questions have been tested on over 100,000 previous test takers. Program staff are trained on protocol and must pass an exam in order to administer the Diana Screen. Most clients use the tool in hiring employees to work with children. In pilot research, the authors examined the results of 200 prospective hires: 174 applicants received passing scores, and 26 received failing scores. Those with failing scores provided answers that indicated previous sexual involvement with children or answers thought to indicate a higher sexual risk to children, such as a poor understanding of adult/child sexual contact boundaries, based on comparison data from child sexual abusers who concealed their activities. Summary and Recommendations for Future Research The dearth of research on mentor screening and youth protection in the context of mentoring may partly reflect obstacles to conducting research in this area. Significant ethical problems are involved with creating control and treatment populations involving known inappropriate applicants (K. C. Herman, 1995), and any sharing of identifying information about rejected volunteer applicants could violate confidentiality and place a program at risk of possible legal action. Despite such challenges, several directions for future research might be useful. Researchers could survey staff from a diverse group of programs about their practices to learn about their rationale for using specific tools and processes. Keeping extensive records and analyzing these data on a regular basis is recommended. For example, a task force formed for Big Brothers Big Sisters of Canada (K. Shaver, personal communication, March 22, 2011) conducted a Child Safety Audit, reviewing internal case files of known child sexual abuse cases. As a result, the agency has already made significant changes to its system and has recommended an additional 25 changes to its national standards. Among the findings, the agency identified trends that parallel research from child sexual abuse perpetrators: The perpetrating mentor will have groomed agency staff first, the abused child is highly unlikely to disclose to agency staff during the life of the match, and the behavior of the mentor and mentee will change in significant ways. What is sorely needed, however, is for the results of such efforts to be published, ideally after scientific peer review, contributing to the

416   PRACTICE AND PROGRAMMATIC CONSIDERATIONS meager base of rigorously established knowledge in the area of mentor screening and youth protection within mentoring (and other) programs for youth. Research interviews with child sexual abuse perpetrators could also be useful in generating more focused “profiles” of offenders or high-risk offender-mentee pairs. As Olson et al. (2007) stated, “[T]he more we know about the process . . . used by perpetrators of child sexual assault, the better equipped we are to recognize who these individuals are before they have a chance to abuse” (p. 232). Additionally, a national databank or national center designed to keep rejected applicant information and confidential incident information, similar to Big Brothers Big Sisters of Canada’s national registry of volunteers, could help the field by tracking and analyzing occurrences of child sexual abuse. These suggestions are just some activities that could prove useful in establishing a stronger foundation of knowledge on which to base mentor screening and protection efforts in mentoring programs for youth. It is more important for research to meet conventional standards for rigor and transparency than it is for research to cover any particular topic or method in this area. This responsibility falls not only to researchers, but also to practitioners, policymakers, and funders in strategic positions to support such investigations (DuBois et al., 2011).

Practice Based on the principle of “do no harm,” it is essential that mentoring relationships start with applicants who have passed stringent safety and effectiveness criteria, a necessarily intrusive process, in order to fulfill this duty (Grossman & Furano, 2002; Rhodes, Liang, & Spencer, 2009). As many program staff understand, insufficient screening and monitoring procedures may leave a mentee in harm’s way, potentially leading to a mentor’s sexually abusing a child or youth, resulting in multiple adverse outcomes for the child or youth (Abel & Harlow, 2001; Middlebrook & Audage, 2008; Prevent Child Abuse America, 2007), as well as the mentoring agency (Arevalo et al., 2006). Because of the limited availability of research, mentoring programs have by necessity had to rely on practitioner knowledge and experience more than formal research in developing their screening and safety policies. Many practices, demonstrating face validity at least, have developed through this method. Such methods also may be more likely to be used than research- or theorybased techniques that are generated outside programs, as program staff who have many years of

experience may feel more confident in their own assessments of both applicants and the progress of a mentoring relationship. However, tools and processes should not be used just because of policy or staff preferences (K. C. Herman, 1995; Roaf et al., 1994), but rather should be carefully chosen through a rational process that involves both initial and ongoing review of all available evidence from research and practice. In this section, we address practice considerations in the area of youth safety and mentor screening. After providing a broad overview of legislative and related developments, we discuss tools and processes that are aligned with theory and available research. Historical Background and Current Status The first federal legislation addressing child safety was enacted in 1993 (Office of the Attorney General, 2006). Screening guides first began appearing in the 1990s, starting with volunteers in general (Graff, 2003; M. L. Herman, Jackson, & Oliver, 2006; Saul & Audage, 2007) and, then, for those who work with children or other vulnerable populations (National Collaboration for Youth, 1997; Office of Juvenile Justice and Delinquency Prevention, 1998; Public Safety Canada, 2008). Recommendations specifically for mentoring programs arrived around the same time, emphasizing the importance of screening for safety and effectiveness (MENTOR, 2009; White, Patterson, & Herman, 1998). However, the first specific guidelines and training devoted to screening volunteer mentors arrived about a decade later (Arevalo et al., 2006; Oliver & MacRae, 2005; Pearson, 2009). Insurance companies have also supported agencies’ standardization of volunteer screening, producing their own recommendations (M. Herman, 2000; Indiana Insurance Company, 2000). Having access to applicants’ criminal history records is one important component to mentor screening. The federally funded pilot of SafetyNET in 2003 allowed youth mentoring programs to receive applicants’ federal-level fingerprint-based background results through the National Center for Missing and Exploited Children for a low fee and with a quick response time. SafetyNET significantly reduced steps in accessing criminal histories and allowed access to the FBI’s database, to which many states do not allow access. Additionally, the SafetyNET pilot produced data that affirmed the importance of background checks: In the program’s first 5 years, over 2,000 mentor applicants (over 6%) had criminal records of concern (MENTOR, 2010). In 2010, the pilot program ended, suspending this

Mentor Screening and Youth Protection   417 access, while legislators consider making SafetyNET a permanent program. Currently, the youth mentoring field lacks mandatory standards and practices, consistent recommendations, and accountability in regard to implementing screening protocols. Even within the largest national youth mentoring program, according to published research, there are inconsistencies in screening and monitoring processes across different agencies (Roaf et al., 1994). The National Child Protection Act/ Volunteers for Children Act allows state governmental agencies, without requiring specific state statutes, to conduct background checks for volunteers who work with children. However, the patchwork of state and federal statutes that allows for this screening also makes mandatory standards and practices difficult. What is easily accessible in one state is not necessarily so in another (Office of the Attorney General, 2006). Even if access were streamlined, there are known errors and inconsistencies in reporting to databases (National Association of Professional Background Screeners, 2005), and most significantly,

clearance from background checks does not always indicate an applicant’s propensity for safety (Arevalo et al., 2006; Finkelhor et al., 2009). One study found that only one in 150 deviant episodes actually led to arrest; in another, fewer than 5% of 647 child sexual molestations studied were ever reported to police (Arevalo et al., 2006). Additionally, if a program uses youth as mentors, they must modify their process because of legal and experience issues (Kremer & Johnson, 2009), but not lessen it, as more than onethird of identified child molesters are themselves under the age of 18 (Finkelhor, 2009). Recommendations by Manuals Other literature in this area consists of surveys of tools or processes without evaluation of their effectiveness (Arevalo et al., 2006; Grossman & Furano, 2002; Kremer & Johnson, 2009; MENTOR, 2009; Schmiesing & Henderson, 2001). Table 28.1 lists several potential tools and processes for mentor screening and youth protection that theory and

Table 28.1   Tools and Processes for Mentor Screening and Youth Protection in Mentoring Programs Number of Research Citations

Number of Manual Citations

Citation Referencesa

Minimum Standards

Staff training and supervision

3

3

5, 6, 7, 9



Applicant database

0

0

Written eligibility criteria/ policy

2

4

1, 7, 8, 9, 10



Distinct process for youth mentorsb

1

1

6, 8



Mentee training

0

1

9



Parent/caregiver training

4

1

1, 5, 6, 7, 9



Mentor position description

1

1

4, 8



Commitment statement

2

5

5, 7, 8, 9, 10



Eligibility, screening, monitoring processes

1

3

5, 8, 9



Orientation

3

3

1, 5, 6, 8, 9

Highly Recommended





(Continued)

418   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Table 28.1  (Continued) Number of Research Citations

Number of Manual Citations

Written application

6

In-person interview

Citation Referencesa

Minimum Standards

5

1, 3, 4, 5, 6, 7, 8, 9, 10



7

5

1, 2, 3, 4, 5, 6, 7, 8, 9, 10



Home visit

5

1

1, 3, 4, 5, 6

Criminal history background–state

7

4

1, 2, 3, 4, 5, 6, 7, 8, 9



Criminal history background–federal

7

5

1, 2, 3, 4, 5, 6, 7, 8, 9, 10



Sex offender registries

3

3

5, 6, 7, 8



Child abuse registries

3

4

5, 6, 7, 8, 10



Driving record

4

3

3, 4, 5, 6, 8, 9



Internet/social media searches

0

0

Character references

7

5

1, 2, 3, 4, 5, 6, 7, 8, 9, 10

Psychometric/ psychological tests

3

1

1, 2, 3, 8

1

8



Health screeningb

Highly Recommended



  

Prematch mentor training

4

4

1, 5, 6, 7, 8, 9



Final decision

2

1

5, 6



Ongoing mentor check-ins

3

4

5, 6, 7, 8, 10



Ongoing mentee check-ins

3

4

5, 6, 7, 8, 10



Ongoing parent/caregiver check-ins

3

3

5, 6, 7, 8, 10



Rescreening continuing mentors

1

0

6



Note: aCitation references are as follows: (1) Roaf et al. (1994); (2) K. C. Herman (1995); (3) Tierney & Grossman (2000); (4) Schmiesing & Henderson (2001); (5) Arevalo et al. (2006); (6) Kremer & Johnson (2009); (7) MENTOR (2009); (8) Oliver & MacRae (2005); (9) White et al. (1998); and (10) Council on Accreditation (2008). b

Recommended only when specified by the agency’s or collaborators’ policies.

Mentor Screening and Youth Protection   419 available research suggests might be useful for this purpose when applied in mentoring programs for youth. As indicated in the table, each of these tools or processes was cited in the literature we reviewed for this chapter, including available manuals. No one manual, however, covers all potentially helpful steps. Some address the topic with a big-picture approach to assist in creating policy and procedures (Council on Accreditation, 2008; MENTOR, 2009; White et al., 1998), whereas others offer the nuts and bolts of specific tools to use (Arevalo et al., 2006; Oliver & MacRae, 2005). Taken together, a program may create a solid screening process. Most programs, however, do not have the luxury of staff resources to research different manuals and experiment with different approaches. Additionally, all of the manuals describe the balance between gathering information from potential applicants and then interpreting the data, though the exact process of interpretation is not always defined precisely. Many program staff would like to see a “checklist” of desirable attributes and characteristics to follow literally, as well as what could be grounds for rejection, but, as the literature available on perpetrators shows, there is no actual profile in terms of an applicant’s background. Established research on behaviors and characteristics that can be interpreted as indicators of concern combined with staffinformed intuition, however, can stand in the place of a rigid profile. Probably the most widely known guide for mentoring programs is the Elements of Effective Practice produced by MENTOR: The National Mentoring Partnership. The third edition (MENTOR, 2009) represents an important step in the field because it combines criteria with consideration of research to support recommendations. Benchmarks for mentor screening include the most popularly cited steps, including a written application, a commitment agreement, an in-person interview, reference checks, and a comprehensive criminal background check. Enhancements include a more specific national, fingerprint-based FBI background check and, for school-based programs, an assessment of interest in continuing a mentoring relationship over summer months. The justifications for screening benchmarks list research from the SafetyNET pilot program for the background check recommendations but no evidence-based research for other steps, aside from noting that they are generally recommended by other sources. The guide does not include other important steps and procedures, such as a distinct process for youth mentors, or allowing all staff to participate in a final decision process, because of the lack of research proving their value to a screening

and monitoring process. However, it does include several suggestions for a mentee screening process, though these steps are not necessarily related to safety. In a separate section on monitoring and support, there are benchmarks for ongoing and regular assessments of the mentoring relationship, but these benchmarks do not include any information regarding youth safety (e.g., noticing indicators of concern during participant check-ins). Arevalo et al. (2006) presented strong support for the connection between recommended tools and how to analyze the data collected from the perspective of safety; however, this resource did not address an overall risk management strategy. The guide presented in Oliver and MacRae (2005) was comprehensive, though less assertive in its suggestions, because it was designed to appeal to a variety of programs. White et al. (1998) provided crucial information from a policy creation and risk management perspective, with little information on how to implement a screening and monitoring process. Council on Accreditation (1998), while based on recommendations from other manuals and research, was the least comprehensive and could leave a program at great risk for selecting unsafe mentors. Two tools not covered in the manuals described here, internet and social media searches and rescreening, have grown in popularity only recently. Internet and social media searches in particular may not have been in existence, as well understood, or as easily accessible when these manuals were published, and the potential effectiveness of these tools and recommendations regarding their use remain unaddressed. General Principles As programs design their mentor screening process, it is crucial for them to remember that youth, not the mentors, are their clients. Although focusing on applicants in the selection process and in supporting the relationship is important, keeping youth safe is the ultimate priority. The goal of mentor screening is to stop perpetrators from entering programs, to prevent youth from becoming abused and, possibly, becoming abusers themselves (Groth, Hobson, & Gary, 1982). Overall, it is not an agency’s responsibility to investigate or “prove” that an applicant may be a perpetrator; the agency must only prevent him or her from becoming a mentor. However, if there is a preponderance of indicators of concern and elevated feelings of unease, programs may wish to consult with local law enforcement, while being mindful of confidentiality and legal issues.

420   PRACTICE AND PROGRAMMATIC CONSIDERATIONS The ability to accept and reject candidates is entirely at the discretion of each agency, though it is recommended that reasons be kept confidential. Consistency in the process (i.e., ensuring all mentors complete every step) is important, as is concluding the entire process before a final decision is made by all staff. All staff must feel comfortable expressing their opinions and be supported in discussing their intuitive feelings, especially if they disapprove of a candidate. Programs must also take into consideration local and state laws, especially as they pertain to any mandatory steps and the confidentiality of application materials. Finally, thorough and effective screening and monitoring processes are imperative, no matter the mentoring or youth-service model. Even site-based models, such as in schools and community centers, provide perpetrators with what they are seeking: access to vulnerable youth. Staff may be able to monitor mentor-youth activities at the site, but they cannot control further interactions and relationship development off site. Setting clear eligibility criteria with permanent disqualifiers (“nonnegotiables”) and mitigating circumstances (Arevalo et al., 2006; Oliver & MacRae, 2005) is an important first step in establishing screening policies and procedures. For example, many programs would not accept any applicant with a history of crimes against children or violent or sexual assault, but a conviction of driving under the influence from 40 years earlier might not be an automatic disqualifier, if it is considered along with other factors from the entire application, indicating the applicant no longer engages in that kind of behavior. Specific Recommended Practices Based on experience in the field, interaction with other agencies, and existing research on child sexual abuse perpetrator characteristics and patterns, we recommend several tools and processes be used for screening and monitoring throughout the cycle of mentor contact with a program. These steps are categorized as (a) minimum standards, which every program should employ, and (b) highly recommended, as additional steps to add to the basic process (see Table 28.1). Steps may be mandatory for a particular program, depending on insurance or setting requirements. It should be noted, as well, that programs are encouraged to develop a thorough screening process for mentoring program staff, including criminal history background checks, thorough interview, and references, as well as following state employment requirements for organizations working with youth.

Program Design. Training, supportive development, and supervision of staff are key in counteracting high turnover rates in direct service positions, which result in losses of agency history, expertise gained in screening and monitoring, and relationships formed with all program participants, including parents or caregivers. Staff members trained on the prevalence and effects of abuse, information about victims and perpetrators, the dynamics of abuse, sample case studies, red flags and warning signs, and screening and monitoring steps that prevent abuse have a greater chance of identifying potential risk. Without this education, staff members may lack the knowledge necessary to recognize warning signs or to validate their intuitive feelings about a particular applicant. There will be times when staff become uncomfortable with an applicant and have a negative reaction. Quite often they can’t put their finger on a specific reason. This intuitive reaction to a candidate may be a first impression or emerge at a later date, such as when an applicant asks inappropriate questions during a training session or becomes overly uncomfortable during the interview, or if a reference is reluctant to share information. It may even take place after a mentor is matched with a mentee. As an agency, it is important for leadership to validate gut reactions and empower staff to make judgments, especially when used to disqualify and reject an applicant, based on informed intuition. Mandatory reporting of child abuse, including sexual abuse, exists in every state and should also be included in staff training. Laws vary to the extent to which mentors are designated mandatory reporters and to whom the abuse needs to be reported, but educating staff, mentors, and parents or caregivers on these procedures can be expected (although not yet confirmed by research) to help ensure that abuse is reported and investigated. Maintaining an applicant database, even if it exists on paper or in a basic spreadsheet, is necessary for tracking rejected candidates and identifying those who reapply after waiting for a fresh start because of staff turnover. Having eligibility criteria or an eligibility policy in writing removes pressure from staff who may be inexperienced in screening. These criteria clearly define nonnegotiables—like having a record that involves crimes against children or making a staff member feel uneasy—that will result in an applicant’s disqualification, as well as examples of mitigating circumstances (Oliver & MacRae, 2005). A distinct application process should be used to ensure that youth/older peer mentors, who are often not thought of as perpetrators, are thoroughly screened (for a discussion of cross-age peer mentoring, see Karcher, this volume, Chapter 16).

Mentor Screening and Youth Protection   421 Preparing Mentees and Parents or Caregivers. Parent/caregiver training, education, and empowerment can increase the chance of preventing child sexual abuse, as family adults may feel more comfortable in validating their own observations and then sharing them with program staff. By engaging parents or caregivers as members of the team to support their children and protect them from harm, programs can not only decrease the risk of child sexual abuse but also increase the involvement of the mentee’s support system with the program. Additionally, mentee training (Roaf et al., 1994) can help prepare youth for the relationship and provide education on personal safety. Program staff can also take the opportunity to continue to build relationships with mentees, assuring them of confidentiality in discussing concerns about the mentoring relationship. Mentor Recruitment. Key steps that we recommend occur during recruitment include creating a mentor position description that clearly lists the expected commitment and necessary qualities required of a successful mentor. By clearly stating the expectations for a mentor, programs introduce the idea that there are standards to be met and not everyone will meet such standards. A more thorough description of acceptable and unacceptable actions should be contained in a code of conduct or responsibilities contract that appears later in the process. Application Process. Compiling data throughout the application process from all available mentor screening tools and processes is much like creating a “paint-by-number” portrait. As an applicant moves through the screening process, colors are added to the portrait. A vast number of experiences, attitudes, behaviors, activities, interests, emotions, and relationship dynamics can be envisioned as tiny sections of the portrait. Those that are related to any one specific theme (e.g., social, career, family, extracurricular) would be represented as gradations of one color. It is the role of the mentoring staff to paint each section of the picture as the screening process unfolds, ideally leading to a finished portrait. Each paint-bynumber segment should be filled, the pieces should fit well together, and a rainbow of colors should be present, indicating a healthy balance and a variety of facets to the candidate. An agency has cause for concern about a candidate’s motivation and appropriateness for working with youth if the portrait of the candidate is incomplete (e.g., a segment of time is missing in the applicant’s life or the applicant does not wish to discuss significant relationships in his or her own life), if it is inconsistent (e.g., the applicant’s account of a particular event does not agree

with what a reference describes or the applicant exhibits responses in training that do not match what was expressed in the interview), or if it is too monochromatic (e.g., the applicant spends an inordinate amount of time socializing with children as opposed to peers or the staff notices an inflexible attitude) (Arevalo et al., 2006). During the orientation, staff can make note of inappropriate questions, comments, or social interactions with other applicants. Programs are encouraged to clearly communicate their eligibility, screening, and monitoring processes, as child perpetrators are less likely to pursue participation in a program that clearly indicates its focus on extensive screening and ongoing monitoring (Arevalo et al., 2006). Not only does this “self-selection” process increase overall safety, but program staff also spend fewer resources on inappropriate candidates. The written application serves as a starting point for learning about an applicant’s life history, values, and motivations. It includes work history, education, references, previous volunteer or paid work with children, and questions regarding past legal history. All questions should be answered and the application and release forms signed. Gaps and inconsistencies become important in assessing an applicant’s appropriateness. An in-person interview, which may include a home visit, offers the opportunity to gather extensive knowledge about a candidate’s life history, family relationships, hobbies, motivation to mentor, current social support networks, and history of crises (e.g., abuse, grief, alcoholism) and subsequent resolution or treatment. Observations of concern to staff should be noted, such as if the applicant is attempting to control the process, is uncomfortable discussing certain issues, or omits information that spans long stretches of time. The staff’s use of informed intuition is most prominent in this intimate step of the screening process, as staff members use the existing research on child sexual abuse perpetrators and red flags, noting responses that may cause concern. For example, if an applicant describes a desired match specifically, including gender, age range, body type, or hair or eye color, this information alone may point to an inappropriate mentor. We recommend that applicants submit to a number of background checks, including both federaland state-level criminal histories, which report an applicant’s criminal history of indictments and convictions of child abuse, sexual offense charges, violent crimes, arson, and drug charges. Fingerprintbased identification is more accurate than name-based checks; programs need to investigate their state’s statutes in this regard. The use of third-party vendors that provide background checks may be an option,

422   PRACTICE AND PROGRAMMATIC CONSIDERATIONS although it is advised that programs thoroughly understand their obligations to the applicant by using a paid service (Oliver & MacRae, 2005). Staff can easily access sex offender registries online, though these databases are not considered current. Programs in some states will have access to child abuse registries, though other programs will not unless they collaborate with a state agency. The use of any kind of background check is hampered by the voluntary nature of the National Child Protection Act/Volunteers for Children Act, inconsistencies in the databases themselves along with states’ use of the law, and the fact that most perpetrators have not been caught or do not have an existing record. Therefore, although background checks are widely regarded in the practice community and sometimes required by insurers as a must to prevent gross oversight and possible legal culpability, an equally strong consensus is that they are not sufficient to guarantee the safety of children. Programs that allow for mentor transportation of a mentee should include submission of a driving record. Even if driving participants is not allowed, the driving record can give a perspective of potential poor judgment or a lack of safety. By creating a program identity or agency presence on current social media platforms, programs make it possible to search for applicants’ profiles via an internet or social media search. Using search engines to research all variations of an applicant’s name can also lead to important information, such as membership in questionable groups, inappropriate materials posted to accounts, and photographs that may reveal unsuitable activities or interests. Although having staff check personal and professional character references can feel like the most tedious part of the process, this step has the potential to reveal superficial or inappropriate connections, excessive enthusiasm about an applicant’s relationships with children, or inconsistencies with information provided by the applicant. Stressing confidentiality of the reference check and providing a staff member’s name and contact information can encourage responses and open sharing of pertinent information. We recommend the use of psychometric inventories/psychological tests only when there is a clear rationale for the results to be useful in the application process. The results of such tests or assessments should be analyzed only by a professional with appropriate qualifications as described by the test developer or publisher. The Diana Screen, the previously described online questionnaire designed to screen out applicants who are a sexual risk to children, is an example of one potential

instrument that could be used, although insufficient research exists on this or any other available tool to recommend its use. Health screening, like a tuberculosis test, may be mandatory depending on the program’s setting (e.g., public school site) and must be undertaken to satisfy requirements. Prematch training, a designated group setting in which agencies provide data, training, role playing, and feedback related to situations that candidates may encounter during their mentoring experience, can reveal a wealth of information about an applicant, including possession of poor boundaries and lack of common sense. Staff are also advised to take note of other participants’ responses to a questionable applicant. During training, programs are encouraged to present a Mentor Code of Conduct that addresses a range of topics related to the safety and well-being of mentees, including gift giving, self-disclosure, alcohol and drug use, discussions about sex, and overnight visits. As the last step in the application process, a final decision involves all staff in contact with the applicant. Promoting a culture of safety can support staff in being honest about their impressions and informed intuition, allowing for an open dialogue about the safety of each applicant before the applicant is accepted or rejected. In assessing the applicant’s entire “portrait,” it is important to remember that no one indicator of concern defines the applicant as a child sexual abuser, but in the context of the entire application, a significant number of red flags may be grounds for rejection (Arevalo et al., 2006). Postmatch. As thoroughly as an agency may implement its screening processes, it is still possible for a perpetrator to infiltrate a program. There are, in fact, websites that teach how to appropriately present oneself as a safe applicant to youth-serving agencies (Arevalo et al., 2006), including how to “beat” the Diana Screen (Abel Screening, n.d.). Ongoing monitoring that involves direct communication, observations, and checking in with the mentor, mentee, and parent/caregiver is essential in preventing child sexual abuse. More frequent monitoring in the earlier relationship development period is recommended to help mentor-mentee matches get established in a positive manner. However, longterm monitoring is highly recommended as well, given that perpetrators take time to groom their victims. Questions to all program participants are meant to uncover any problems in the development of the relationship, such as not meeting regularly or experiencing communication problems, unnecessary gift-giving by the mentor, a lack of boundaries or excessive contact, nonconformance with agency

Mentor Screening and Youth Protection   423 policies, or a participant’s attempting to manipulate the agency. Weinberger, Garringer, and MacRae (2005) and Oliver and MacRae (2005) list a number of questions to ask mentors, mentees, and parents or caregivers during check-ins. Of particular safety importance are questions about the kinds of activities that matches do together, if the mentor has asked the mentee to keep secrets, and, to parents or caregivers, if there are any concerns about the match or changes in their child’s behavior. If child sexual abuse happens, programs must terminate matches without delay and provide resources for immediate clinical help for youth (Arevalo et al., 2006). For matches that terminate early, care should be taken in rematching mentors. The circumstances of the match’s ending should be assessed thoroughly for red flags, such as if either the mentor or the mentee ended the match suddenly and was not forthcoming about the reason. For programs with mentors who return each year to be matched with new youth, rescreening steps can reveal any new information that may be found to be out of line with the services provided or make the applicant ineligible. Steps may be modified or shortened to cover only information that may have changed since the first application process. For example, if the program hasn’t changed significantly, there is little reason for the returning mentor to attend either an orientation or prematch training. However, if the returning mentor will be

working with new staff, a brief interview to renew the relationship and learn more about recent life events would be recommended. Agencies are advised to research their state’s ability to provide subsequent criminal history records without resubmitting fingerprints, as reviewing any recent criminal history is recommended. Internet searches can also reveal more detailed information about a returning mentor. Development and Ongoing Refinement of Practices for Mentor Screening and Youth Protection The development of a single source of recommendations or the development of standards on what to include in policies and procedures, based on the evidence from research and practice, can significantly impact youth safety in mentoring programs. Such a protocol could then be adopted by entities such as insurance carriers and licensing/accreditation providers. Ultimately, however, each program will need to develop policies and practices related to mentor screening and youth protection that are tailored to its specific characteristics and resources. Such decisions, rather than being static, will need to be reevaluated on a periodic or ongoing basis in the context of new developments in research and practice. With these considerations in mind, in Table 28.2 we have provided a set of guiding questions for

Table 28.2   Checklist for Practitioners Topic

Recommended Questions

Staff training

What tools and processes are currently in use for mentor screening and youth protection, and to what extent are program staff consistently and completely using them? Are staff sufficiently trained in identifying child sexual abuse perpetrators to use their informed intuition throughout the application and monitoring process?

Screening

How useful are existing tools and processes for our mentor screening process?

Tools and processes

What tools and processes for mentor screening and youth protection are essential to include in the program (see Table 28.1 for specific recommendations)? What other tools and processes could we add, given our resources and mentor selection pool? What resources are needed to add tools or processes? What barriers (other than resources) exist to adding any specific tools or processes, and how can we address these barriers? How can we use the information that is being obtained to strengthen the process of screening and monitoring mentors and their relationships with youth?

424   PRACTICE AND PROGRAMMATIC CONSIDERATIONS programs to use in developing or strengthening processes used for mentor screening and youth protection.

Conclusion Several key points bear emphasizing. First, it should be clear from our brief review that practice in the area of mentor screening and youth safety has outpaced research. Previous research has focused on highlighting tools and processes that have been in use (or should be in use) by the field, based on the premise that they will serve a protective function. Most of these suggestions appear sound but have not been tested empirically. More research is needed to determine what steps and procedures are most effective in screening mentors, as there is a significant lack of information to assess what really works and which tools or processes will be the most effective while requiring the fewest resources. Second, the responsibility to protect youth from child sexual abuse clearly lies with the mentoring organization. In addition to preventing adverse outcomes, a commitment to safety shows the program’s ultimate dedication to each youth participant and parent or caregiver. The need for mentoring continues to grow in a society in which youth receive decreased amounts of positive adult time. More programs are emerging to fill this need, sometimes as an add-on component to existing services, due the success of mentoring in the healthy development of youth. Therefore, the need for appropriate safeguards to protect youth in these types of programs is greater than ever. Finally, there is tremendous opportunity for the mentoring field to take a leadership role in keeping youth safe in the wide range of formal youth-serving programs. The field can, in our view, best assume this role through a collaborative partnership among researchers, practitioners, funders, and policymakers. Existing theory, research, and practice point to the importance of youth mentoring programs acknowledging that molestation is still a possibility; being open about program processes and incidents, as well as feeling safe in sharing program or agency records; thoroughly educating staff on effective screening and monitoring practices; partnering more with parents/caregivers through education and training; creating a position statement; and setting mandatory safety practices based on research that apply to all youth mentoring programs. The need for advancement of research-based knowledge in this area is great, however, and remains a critical undertaking in need of united attention and resources.

References Abel, G., & Harlow, N. (2001). The stop child molestation book: What ordinary people can do in their everyday lives to save three million children. Bloomington, IN: Xlibris Corporation. Abel Screening. (2009). The Diana Screen pilot results: 200 prospective hires. Atlanta, GA: Author. Abel Screening. (n.d.). The Diana Screen and the Diana Comprehensive: Scientific support and an issue regarding publication. Atlanta, GA: Author. Arevalo, E., Cooper, B., & Smith, M. (2006). SAFE (Screening Applicants for Effectiveness): Guidelines to prevent child molestation in mentoring and youth-serving organizations. Redwood City, CA: Friends for Youth. Boyle, P. (2010, April 1). Boy Scout confidential. Youth Today. Retrieved from http://www.youthtoday.org/ view_article.cfm?article_id=3878 Council on Accreditation. (2008). Private agencies standards for volunteer mentoring services, 8th ed. New York: Author. Retrieved from http://www.coastandards.org/ standards.php?navView=private&core_id=595 DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Finkelhor, D. (2009). The prevention of childhood sexual abuse. Preventing Child Maltreatment, 19(2), 169–194. Finkelhor, D., Ormrod, R., & Chaffin, M. (2009, December). Juveniles who commit sex offenses against minors. Juvenile Justice Bulletin. Washington, DC: U.S. Government Printing Office. Gigerenzer, G. (2007). Gut feelings: The intelligence of the unconscious. New York: Viking Press. Graff, L. (2003). Better safe . . . Risk management in volunteer programs and community service. Ontario, Canada: Linda Graff and Associates. Grossman, J., & Furano, K. (2002). Making the most of volunteers. Philadelphia: Public/Private Ventures. Groth, A. N. (1979). Men who rape: The psychology of the offender. New York: Plenum Press. Groth, A. N., Hobson, W., & Gary, T. (1982). The child molester: Clinical observations. In R. Conte & A. Shore (Eds.), Social work and child sexual abuse (pp. 129–144). New York: Haworth Press. Herman, K. C. (1995). Appropriate use of the Child Abuse Potential Inventory in a Big Brothers/Big Sisters agency. Journal of Social Service Research, 20(3-4), 93–103. Herman, M. (2000). Managing volunteers: Balancing risk and reward. Santa Cruz, CA: Nonprofits’ Insurance Alliance of California. Herman, M. L., Jackson, P. B., & Oliver, B. B. (2006). No surprises: Harmonizing risk and reward in volunteer management (4th ed.). Washington, DC: Nonprofit Risk Management Center. Indiana Insurance Company. (2000). Guidelines for the screening of persons working with children and

Mentor Screening and Youth Protection   425 vulnerable adults. Waukesha, WI: Author. Retrieved from http://www.indianains.com/omapps/Content Server?pagename=IndianaInsurance/Views/Indiana Insurance&ft=6 Kremer, S., & Johnson, K. (2009). Going beyond the background check: The status of volunteer screening in San Mateo County youth-serving organizations. Redwood City, CA: Friends for Youth. Retrieved from http://www.friendsforyouth.org/files/MIpdfs/ SAFE_2009Report.pdf Lanning, K. (2001). Child molesters: A behavioral analysis. Washington, DC: National Center for Missing and Exploited Children. MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. MENTOR. (2010). Overview of the SafetyNET criminal background check pilot fact sheet for mentoring organizations. Washington, DC: Author. Retrieved from http://www.mentoring.org/downloads/mentoring _1281.pdf Middlebrook, J. S., & Audage, N. C. (2008). The effects of childhood stress on health across the lifespan. Atlanta, GA: Centers for Disease Control and Prevention. Retrieved from http://www.cdc.gov/ ncipc/pub-res/pdf/childhood_stress.pdf National Association of Professional Background Screeners. (2005). The National Crime Information Center: A review and evaluation. Morrisville, NC: Author. National Collaboration for Youth. (1997). Screening volunteers to prevent child sexual abuse: A community guide for youth organizations. Washington, DC: National Assembly of National Voluntary Health and Social Welfare Organizations. Office of Juvenile Justice and Delinquency Prevention. (1998). Guidelines for the screening of persons working with children, the elderly, and individuals with disabilities in need of support. Washington, DC: U.S. Government Printing Office. Office of the Attorney General. (2006). The Attorney General’s report on criminal history background checks. Washington, DC: U.S. Government Printing Office. Retrieved from http://www.justice.gov/olp/ ag_bgchecks_report.pdf Oliver, B., & MacRae, P. (2005). Guide to screening and background checks. Folsom, CA: Mentoring Resource Center in cooperation with the U.S. Department of Education, Office of Safe and Drug-Free Schools. Retrieved from http://educationnorthwest.org/ webfm_send/182 Olson, L., Daggs, J., Ellevold, B., & Rogers, T. (2007). Entrapping the innocent: Toward a theory of child sexual predators’ luring communication. Communi­ cation Theory, 17(3), 231–251. Patterson, J. (2008, March). The essence of youth protection: Rules, reporting, and consequences [eNews Brief]. Retrieved from http://nonprofitrisk.org/ library/enews/208/enews030508.htm Pearson, J. (2009). Preventing and dealing with sexual abuse of children in one to one mentoring programs. Retrieved from http://educationnorthwest.org/resource/360

Prevent Child Abuse America. (2007). Total estimated cost of child abuse and neglect in the United States. Chicago: Author. Public Safety Canada. (2008). Best practice guidelines for screening volunteers. Canada: Volunteer Canada. Retrieved from http://www.publicsafety.gc.ca/res/ cor/rep/_fl/volunteer_screening_e.pdf Rhodes, J., Liang, B., & Spencer, R. (2009). First do no harm: Ethical principles for youth mentoring relationships. Professional Psychology: Research and Practice, 40, 452–458. Roaf, P., Tierney, J., & Hunte, D. (1994). Big Brothers/Big Sisters: A study of volunteer recruitment and screening. Philadelphia: Public/Private Ventures. Saul, J., & Audage, N. C. (2007) Preventing child sexual abuse within youth-serving organizations: Getting started on policies and procedures. Atlanta, GA: Centers for Disease Control and Prevention. Retrieved from http://www.cdc.gov/violenceprevention/pdf/Pre ventingChildSexualAbuse-a.pdf#page=1 Schmiesing, R., & Henderson, J. (2001). Identification of volunteer screening practices for selected Ohio youth organizations. Journal of Extension, 39(1). Retrieved from http://www.joe.org/joe/2001february/a2.php Shakeshaft, C. (2004). Educator sexual misconduct: A synthesis of existing literature. Washington, DC: U.S. Department of Education. Retrieved from http://www2.ed.gov/rschstat/research/pubs/miscon ductreview/report.pdf Sullivan, J., & Beech, A. (2002). Professional perpetrators: Sex offenders who use their employment to target and sexually abuse the children with whom they work. Child Abuse Review, 11, 153–167. Terry, K., & Tallon, J. (2004). Child sexual abuse: A review of the literature. New York: John Jay College. Retrieved from http://old.usccb.org/nrb/johnjay study/litreview.pdf Tierney, J. P., Grossman, J. B., & Resch, N. (2000). Making a difference: An impact study of Big Brothers/Big Sisters. Philadelphia: Public/Private Ventures. van Dam, C. (2001). Identifying child molesters: Preventing child sexual abuse by recognizing the patterns of the offenders. Binghamton, NY: Haworth Press. van Dam, C. (2006). The socially skilled child molester: Differentiating the guilty from the falsely accused. Binghamton, NY: Haworth Maltreatment and Trauma Press. Weinberger, S., Garringer, M., & MacRae, P. (2005). Going the distance: A guide to building lasting relationships in mentoring programs. Folsom, CA: Mentoring Resource Center in cooperation with the U.S. Department of Education, Office of Safe and Drug-Free Schools. Retrieved from http://educationnorthwest.org/webfm_send/166 White, L., Patterson, J., & Herman, M. (1998). More than a matter of trust: Managing the risks of mentoring. Washington, DC: Nonprofit Risk Management Center. Wurtele, S., & Kenny, M. (2010). Partnering with parents to prevent child sexual abuse. Child Abuse Review, 19, 130–152.

29  MENTOR AND YOUTH MATCHING Julia Pryce, Michael S. Kelly, and Sarah R. Guidone

Introduction The process involved in successfully matching an older person to a younger person, particularly in program-based mentoring relationships, is critically important yet remains minimally explored and understood. Despite limited research on how best to conduct the matching process, practitioners facilitate the formation of matches every day. This process aims to identify, secure, and ultimately support the pairing of two people in such a way as to, per the words of many mentoring organizations, “change the lives” of both participants. This matching process is complicated, due in part to the individual idiosyncrasies of both participants within any given match. To complicate the process further, it is critical to acknowledge that many of these relationships are formed between two people who may differ in significant aspects of their life experiences. These differences may be in social class, gender, generational differences in experience and perspectives, and/or race and ethnicity (see, in this volume, Deutsch et al., Chapter 12; Liang et al., Chapter 11; Taylor & Porcellini, Chapter 31; and Sánchez et al., Chapter 10, respectively). In large part, mentoring programs aim to simulate through their programbased matching the protective processes of resilience (Rhodes & Lowe, 2008) that take place in naturally occurring mentoring relationships (DuBois & Silverthorn, 2005). However, the “matching” and building of a relationship, even among naturally occurring mentoring relationships, is only beginning to be understood, and is often not evaluated in the literature with any explicit reference to theory. Thus, this area of

mentoring provides an opportunity for creative practice and innovative research and scholarship. Given the limited empirical attention devoted to the process of matching within mentoring relationships, this chapter aims to apply what we know from positive youth development and other fields to achieve a more informed understanding of this important process. In this chapter, our first goal is to illuminate ways by which multiple theories can inform the matching process across mentoring domains. Second, we examine the research associated with individual- and program-level factors associated with matching. Finally, we explore current practices and identify those that appear most promising based on available research.

Theory The literature on mentoring has borrowed from numerous theories in order to support various research and practices. For the purposes of this chapter, we focus on three key frameworks based on the literature: attachment theory, the “voice and choice” framework associated with positive youth development, and probability matching theory. The first two of these theories have been applied to the youth mentoring literature, including specific applications to the matching process. The last framework, probability matching theory, while not used currently in youth matching, is widely used in the more developed field of matching within romantic relationships. Our hope is that future work focused on matching can incorporate these and other theoretical frameworks to guide a more intentional approach to the matching of prospective mentors and mentees.

427

428   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Attachment Theory The most prominent theory used to date in explaining youth mentoring pairings is attachment theory. Attachment theory has been applied to myriad adult-youth and mentor-mentee relationships, including therapeutic relationships such as those employed in the psychotherapeutic context (Bernier & Dozier, 2002). In the context of mentoring, researchers hypothesize that mentors can provide their mentees with relationship experiences that are strongly “disconfirming” of their existing attachment representations. These experiences then can spur a reevaluation of existing relationship models, and can create some amount of positive change in maladaptive relationship patterns (Bernier & Dozier, 2002). In the domain of therapeutic relationships, attachment theory is often used to help explain how complementary differences within a pairing may serve to provide a “corrective emotional experience” for the client (Alexander & French, 1946, p. 294). This “correction” results from the shift in the internal attachment schema maintained by the client, which occurs in response to a new, more affirming attachment experience with one’s therapist. Within psychodynamic theory, this corrective emotional experience is considered a key factor in therapeutic change. Attachment theory places particular weight on how the adult-child relationship is influenced by the child’s beliefs about adults, including nonparental adults such as teachers, therapists, and mentors. However, controversy exists concerning the extent to which models for parent relationships generalize to influence youth’s interactions with other nonparental adults. This controversy stems in part from questions regarding how stable, and how resistant to change, one’s attachment constructs remain over time (Cassidy & Shaver, 1999). Attachment theory focuses on emotional closeness, response to conflict, dependency, and ways by which an emotionally close or “secure” relationship between a child and an adult can assist the child in emotional regulation (Cassidy & Shaver, 1999). Research has developed a strong basis for considering the role of an “insecure” attachment to the primary caregiver as predictive of vulnerability to a host of difficulties across youth, adolescent, and adult development. Attachment theory has also been applied to the field of romantic relationships, and has led to questions about whether complementary or similar attachment styles most benefit dyadic relationships. In this context, the focus is often on partner behavior and attitudes, such as jealousy, overall satisfaction, and proximity maintenance. Findings from this

research on attachment theory suggest that similarity in attachment style is not necessarily common among couples (e.g., Brennan & Shaver, 1995). Historically, emphasis on similarities among successful couples supported the importance of similarity in relationships (Burleson & Denton, 1992) However, more recent analysis of the dynamics within the couple’s relationship provides evidence of the importance of complementarities in close relationships (Markey & Markey, 2007). To date, it remains unclear how attachment theory might apply to older youth in the formation of mentoring relationships. However, it remains a possibly helpful lens in considering the ways that similar and complementary styles may influence mentoring matches. Along the same lines, the similarity-attraction paradigm (Byrne, 1971) is also worth brief mention here. This paradigm suggests that those who have more in common (such as race, gender, etc.) would experience more successful relationships than would those of disparate characteristics. Although this paradigm has not received consistent empirical support to date, it is also conceptually useful as we consider the complexity of the role of similarity and difference in strong mentoring matches. Voice and Choice The “voice and choice” framework refers to youth involvement and shared decision making as applied in youth development programs to multiple programming aspects, such as the matching process in mentoring. The framework has been associated with increased youth engagement, motivation, and personal skill and problem-solving development (e.g., High/Scope Educational Research Foundation, 2008). It involves youth representation in both structural and content components (High/Scope Educational Research Foundation, 2008). Structural program components include organizational policies, program scheduling or offerings, and generally the way youth “spend their time.” Content components exist within designated program structures, such as “topics covered, tools or materials used in an activity, or how end products may be presented” (p. 18). The presence of voice and choice within youth mentoring programs could reflect soliciting youth opinions about how and why matches are created (structural component), and then providing opportunities for content choices within the match process. Other examples of voice and choice in mentoring, including matching “mixers” or “meet-ngreet” sessions with prospective mentors (Karcher, 2007; Karcher & Santos, 2011) and youth-initiated

Mentor and Youth Matching   429 mentoring (YIM; Millenky, Bloom, Muller-Ravett, & Broadus, 2011), are discussed later in this chapter. Probability Matching Theory Probability matching theory is based on mathematical modeling designed to explain the process of interpersonal attraction and to predict the probability that specific matches or relationships will occur (Takeuchi, 2006). The theory is used in mate selection to mathematically predict the likelihood that one person will choose another based on “playby-play” scenarios of similar physical attractiveness and overall market value (e.g., other qualities that the seeker can offer a potential target). Mathematical algorithms have been applied extensively in matchmaking on online dating sites (e.g., P. Schwartz, 2006). These sites use personality assessments (similar to the Myers-Briggs Type Inventory) to classify people according to a set of personality traits and preferences (P. Schwartz, 2006). Each individual combination, or personality profile, could potentially be compatible with a number of other combinations. Schwartz’s Duet system has initial matches suggested for each personality type, but also calculates user preferences over time. It is unclear which matching dimensions are given more weight than others as the systems do not divulge their formulas and algorithms. Most online dating systems involve a lengthy list of categories for each individual in the match. For example, OKCupid’s classifications are formed through questions answered by users about opinions and lifestyles, such as desire for children, smoking habits, political affiliation, or desire for commitment. One of the most distinguishing features of the romantic matching model is the continual refinement and reassessment of potential matches. Current youth programs do not reassess and rematch mentors and mentees, but perhaps benefits of “refining the match decision” might be worth considering. A challenge to the use of this approach in youth mentoring is that many individuals have not given the mentoring relationship much thought before they enter a program, and having little prior experience mentoring, many people may not be able to verbalize their preferences for a mentoring match. In addition, few programs provide the opportunity for mentors and mentees to “try on” matches in such an iterative way. Conclusion There appears to be an unstated assumption in much of the mentoring literature that mentors and

mentees need to find someone with whom they “click” as a result of similar interests (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011); personality traits; or racial, ethnic, or gender characteristics. Yet, significant and positive alliances can be formed even between mentors and mentees who differ in many ways. More theoretical and conceptual work is needed to identify the characteristics that should be and that need not be similar between mentors and mentees. The three theories described here reveal new opportunities for research and creative practices in mentor-mentee matching. In relation to attachment theory, we need to know how to match participants in a way that increases the likelihood the mentoring relationship will provide a corrective emotional experience (Bernier & Dozier, 2002). Building on research on couples’ attachment styles in successful relationships, to what extent should similar interests, backgrounds, or interpersonal styles take priority in the matching process? In addition, the process of incorporating participant choice in the context of youth mentoring relationships raises questions about what happens when a youth or mentor changes his or her mind. Given that prospective mentors and mentees, like those pursuing romantic relationships, learn more about one another well after the initial meeting, what room can the field make for “trial” periods in relationships, and how does choice affect the process of rematching?

Research Although progress has been made regarding our understanding of the degree to which mentoring affects youth (i.e., DuBois et al., 2011; Wheeler, Keller, & DuBois, 2010) and who makes the best mentors (Keller & Pryce, 2010; Pryce, 2012), there has been limited specific empirical inquiry into the process of matching itself. Some early research investigated more general characteristics as applied to matching, such as the role of outgoing personality traits on mentoring relationships among “disadvantaged youth” (Dicken, Bryson, & Kass, 1977), but such examinations were rare and preliminary in scope. Rather, the majority of mentoring research has focused on understanding why matches fail (Spencer, 2007) or succeed, without necessarily focusing on the matchmaking process. However, the link between the initial matching process and the likelihood of success later in the relationship has begun to receive more attention from researchers (Rhodes & DuBois, 2006).

430   PRACTICE AND PROGRAMMATIC CONSIDERATIONS To better understand matching in youth mentoring, we engaged in an evidence-based process by systematically searching for the best evidence on effective matching processes in the recent empirical literature (Kelly, Raines, Stone, & Frey, 2010). This evidence-based practice process is often used to identify what is known about an issue or intervention. In so doing, researchers can appraise the depth and quality of the research evidence for a specific question relevant to practice; thus, this process differs from a more conventional narrative literature review (Kelly et al., 2010). Specifically, we were looking for research on youth mentoring programs that involved a matching procedure that was detailed in the article, as well as (in some cases) some information about how that matching process itself was evaluated for its success in making effective mentor-mentee matches. We searched for literature on youth mentoring programs (as opposed to mentoring of employees, for example). For this review, a database search was conducted using Ovid (Medline, EBM Reviews, and PsycINFO) and the educational database ERIC for all mentoring research focused on youth and adolescents published between 2000 and 2010, including the search terms mentoring, randomized controlled trial, program, school, and afterschool or community setting. From the studies we found, we narrowed our search by combining the above search terms, and then searched abstracts and articles by hand to identify articles that fit the scope of this chapter. The search identified 10 studies involving mentoring matches with youth and adolescents, of which 6 included specific detail on the matching process itself. The literature indicates three different methods of matching. The first method uses interest surveys to match on similarity, the second employs the use of meet-n-greet events to give youth and mentors a voice in the matching process, and the third aims to offer some combination of the first two. In this section, we provide examples within each of these three approaches. Research on Matching Procedures Several studies used interest surveys as a way to match mentors and mentees. Of the studies using interest surveys, one focused on matching at-risk high school students to mentors who were school staff (Converse & Lignugaris-Kraft, 2009), one focused on adolescents being mentored in the Big Brothers Big Sisters program (Rhodes, Grossman, & Resch, 2000), and another focused on middle school students matched with community-based mentors as part of a violence prevention program (Rollins, Kaiser-Ulrey, Potts, & Creason, 2003).

Two studies employed a meet-n-greet format with older youth mentoring younger children (Karcher, Nakkula, & Harris, 2005), and with faculty working with 9th graders at risk for school failure (Holt, Brenna, & Johnson, 2008). These studies provided sufficient details regarding how matches were made and the process of matching that their approaches could be replicated by others. However, they did not evaluate the effect of meet-n-greet on outcomes. Another study, involving adults mentoring Latino youth in a school-based program, used a hybrid process of an interest survey and a program meet-n-greet session, which allowed mentors and mentees to have input on their match but also let program staff finalize the matches in terms of shared interests. Within this study, 14% of the matches (n = 45) were made by direct choice between the mentors and mentees, while 86% (n = 275) were made based on staff input and schedule compatibility (Karcher, 2008). Subsequent analyses compared the 45 mentees who participated in the meet-n-greet on reports of mentoring relationship quality, according to whether the mentor, the mentee, or both chose the person with whom they were matched (Karcher & Santos, 2011). In 14 matches, neither listed the other; in 16 matches, only one listed the other; and in 14 matches, both listed each other. The matches in which only one person was nominated by the other scored highest and scored significantly higher than matches in which no one selected the other on three of six match-quality ratings: feeling valued by the mentor (mean differences in scores = .48, p < .05), motivated by the mentor (mean differences = .51, p < .05), and feeling they mattered to their mentors (mean difference = .42, p < .05). The mutually nominated matches also scored higher than the nonomination group, but only on feeling they mattered to their mentor (mean differences = .43, p < .05). Of note, they did not score higher on any outcomes than the group in which only one person selected the other, but the findings suggest nevertheless that when one or more partner has a choice, relationship benefits may ensue. Beyond our search of programs focused on youth and adolescents, work within the mentoring field with college students helps to further illuminate the process by which choice and perception within the matching process may impact relationship outcomes. In work with college students, Kendall (2007) hypothesized that the smaller effects of formal mentoring relationships may result from the mentee’s lack of flexibility in choosing the mentor. To test this hypothesis, Kendall attempted to simulate informal mentoring relationships between

Mentor and Youth Matching   431 undergraduates by allowing first-year mentees a choice in selecting their older college student mentors. Kendall hypothesized that if mentees were offered a choice in their mentors, they would be more motivated to get to know their mentors and engage in a longer relationship. A total of 123 mentor-mentee dyads of undergraduate peer mentors were assigned randomly to one of three groups of dyads that involved different levels of “choice” for mentees in regard to matching: The first group was allowed to pick their mentors from online profiles (labeled “choice match”), the second was told that they would be matched by project staff based on similarities in interests and career goals based on an online survey (labeled “perceived similarity match”), and the third group was matched with a mentor based on similar availability/schedule of mentors and mentees (labeled “convenience match”). Online mentor and mentee profiles consisted of information on gender, ethnicity, college major, career goals, life obstacles that were overcome, and outside interests. Kendall’s study yielded support for the importance of voice and choice in mentor selection. Such findings comport with an early investigation of the ChalleNGe mentoring program, which is a national program that relies on young people nominating their own mentors during the program application process. ChalleNGe initiates the mentoring relationship part way through the Residential Phase, after the staff screen and train the mentors. The staff then maintain contact with both the program’s graduates and their mentors at least monthly during the Post-residential Phase to help solve problems and to report on youths’ progress. Through the ChalleNGe program, youth (called cadets) and their parents are asked to nominate mentors. To qualify, prospective mentors must be at least 21 years old, be the same gender as the cadet, and live within reasonable geographic proximity. They must not be from the cadet’s household or immediate family and cannot be a member of the ChalleNGe staff or a staff member’s spouse. As part of the application process, most programs require applicants to identify two potential mentors. Programs attempt to have a prospective mentor for each participant at the beginning of the program, with the match formalized by week 13. The matching ceremony includes a meeting with the case manager, mentor, and cadet and the signing of a written agreement (Bloom, Gardenhire-Crooks, & Mandsager, 2009). Some recent research has sought to examine the ways by which mentor choice is correlated with long-term relationship maintenance and quality

(S. Schwartz, in preparation). As hypothesized, at both 21- and 38-month follow-up surveys, youth who selected their mentors “mostly on [their] own” were significantly more likely to be in contact with their mentors than those whose mentors were selected by parents, ChalleNGe staff, or some other method (S. Schwartz, in preparation). At the 38-month follow-up, selection of mentors by youth remained an important predictor of subsequent contact (odds ratio = 1.82, p < .05). In addition, mentors and youth who were of the same race were more likely to be in contact (odds ratio = 3.44, p < .01). No other baseline characteristics significantly predicted contact at the later follow-up point. These findings, while preliminary, suggest that youth who chose their mentors mostly on their own were more likely to be in enduring relationships, and provide additional support for the idea that choice matters in mentoring relationships. Also, per the Kendall (2007) work, it seems that if mentors know that their mentees selected them, they may be more invested in and committed to the relationship and may maintain a greater sense of self-efficacy as a mentor. Research on the Role of Similarity in Individual Characteristics Given the limited research on matching procedures and their impact on outcomes, we decided to work inductively by culling hints from existing research on match quality about what characteristics of relationship are most strongly associated with relationship quality or program outcomes. This may help to identify the key variables that researchers and practitioners should focus on in their efforts to garner empirical evidence based on best matching practices. In other words, what can the empirical literature tell us about how similar personality types; racial, gender, or socioeconomic similarities and differences; and other factors play into making a mentoring match successful? This section briefly reviews some of the complexities inherent in this question. Meta-analyses of the Effects Associated With Programs That Match According to Similarity. Two meta-analyses of mentoring programs have advanced our insights regarding the importance of specific mentor-mentee characteristics that may contribute to a successful match. In a meta-analysis of 55 independent evaluations of mentoring programs published through 1998 (DuBois, Holloway, Valentine, & Cooper, 2002), the association between program effectiveness and programs’ systematic efforts to match mentors and youth in general, or

432   PRACTICE AND PROGRAMMATIC CONSIDERATIONS their practices of matching on the basis of gender, race/ethnicity, or interests, respectively, did not reach statistical significance. However, higher scores on the former, a systematic approach to matching, were predictive of stronger effects, and systematic matching was retained in the authors’ set of theory-based practices, given the general consensus in the field of its importance. In their more recent meta-analysis of 73 independent evaluations of youth mentoring programs published between 1999 and 2010 (DuBois et al., 2011), and consistent with the trend found in their earlier work, DuBois and colleagues found that program effects were significantly stronger when there was evidence that programs included an emphasis on matching youth and mentors based on similarity of interests. Matching emphasizing similarity of interests emerged as a statistically significant predictor of outcomes in a “best fitting” model, based on stepwise regression analysis that considered the independent contributions of different program practices. However, as in the earlier meta-analysis, matching on the basis of gender or race/ethnicity was not positively associated with effectiveness. Rather, matching on the basis of race/ethnicity also was included in the final model, but a programmatic emphasis on matching by race was associated with weaker program effects. Because these findings related to program-level findings, this does not mean that matches of the same race did poorer, but rather only that programs that emphasized same-race matching had weaker effects. Perhaps programs that emphasized race in matching gave less attention to matching by interests, and this is why the negative association was found. Single-Study Findings on the Effects Associated With Individual Similarities Among Matches. By focusing on the evaluation of programs that use specific matching approaches, these meta-analyses differ from most other research that has looked at the role of similarities in race, gender, and interests. Most research has also been applied to understanding how specific individual characteristics (e.g., personality traits, race, gender) and perceived sense of shared characteristics or experience may impact eventual match success for individual matches. We highlight some of the research found in comparisons of matches that do and do not match on these three characteristics. More detailed analyses can be found in other chapters of this volume (see Sánchez et al., Chapter 10; Liang et al., Chapter 11; and Deutsch et al., Chapter 12). Some evidence indicates that interest similarity and personality similarity are associated with stronger matches. One study found that, when

selecting a match, undergraduates and faculty mentors emphasized shared academic discipline, more than any racial, gender, or ethnic considerations (Campbell & Campbell, 2007). Madia & Lutz (2004) looked at how mentors matched with their mentees for 1–11 months viewed their shared qualities as extroverts. Findings suggested that viewing one’s mentee as similarly extroverted was associated with longer matches. This study was one of the first to report that mentors’ perceived similarity with their mentees was related to the overall length and quality of the mentor-mentee relationship. Research has also investigated attachment styles of mentors and mentees. For example, in collegelevel academic mentoring, contrasting attachment orientations (i.e., ambivalent and avoidant) were found to lead to more positive outcomes (Larose, Bernier, & Tarabulsy, 2005). Similarly, in psychotherapy research, contrasting attachment orientations between therapist and patient have been found to predict better working alliances and more positive outcomes (Tyrrell, Dozier, Teague, & Fallot, 1999). There also is some evidence that racial similarity is associated with stronger matches. Schippers (2008) revealed preliminary information to suggest that mentees in formal mentoring programs may prefer mentors of their same race. An experimental study of preferences of undergraduate students (mentees) toward hypothetical mentors also found that mentees may be more likely to prefer same-race than cross-race mentoring relationships (Hu, Thomas, & Lance, 2008). On the other hand, Karcher, RoyCarlson, and Benne (2006) found that ethnic similarity in school-based mentoring with Latino youth was not significantly related to change in most program outcomes, such as social skills and self-esteem, but rather that cross-race matches yielded benefits in connectedness to culturally different youth while same-race matches were associated with improvements in self-esteem. Still other work has emphasized the personal role that race plays in the lives of mentees and youth, suggesting that attending to individual differences when considering these factors is key to strong match relationships (Syed, Goza, Chemers, & Zurbriggen, 2012). Although cultural mistrust has also been identified as a core challenge to relationship formation in cross-race matches (Sánchez et al., this volume, Chapter 10), racially dissimilar mentors have reported feeling they are just as emotionally and instrumentally supportive and close as those in samerace matches (Herrera, Sipe, & McClanahan, 2000). Therefore, there is inconsistent evidence to suggest that racial similarity in matching is necessary to foster the emotional closeness that can support an

Mentor and Youth Matching   433 “emotionally corrective” experience as described in attachment theory. Mentors, mentees, and parents may also perceive the role of race, and therefore their preferences, differently as it relates to matching (Spencer, Lewis, & Basualdo-Delmonico, 2009). In addition to cultural mistrust, students’ initial beliefs about adult mentors have been explored as an important factor in the process of relationship matching and development. Using an experimental design with a sample of African American high school students (N = 94), Linnehan, Weer, & Uhl (2005) examined the determinants of students’ trust beliefs about adult mentors prior to their encounter with a mentor. Results indicated that both beliefs in the larger program structure (i.e., structural assurance beliefs) and youth dispositions toward trust were positively associated with their beliefs about an adult mentor (in terms of mentor benevolence, honesty, competence, and predictability). Ethnic identity of the student was found to moderate the relation between two of these beliefs (competence and predictability) and racial similarity between the mentor and student. In other words, African American students with low ethnic identity believed that a White adult mentor would be more competent and predictable than African American students with high ethnic identity. This finding highlights the complexity inherent in the building of trust, and how the perceptions of one another are so critical to the matching process between mentors and youth. Findings on matching by gender were also mixed. Although the majority of youth development programs match based on gender, this is typically done as a means of reducing the risks associated with cross-sex matches. For example, in a study of Mexican American high school students (Flores & Obasi, 2005), although gender differences were found on several outcome variables, these differences were not based on the sex or ethnicity of the mentor match. This finding contradicted the researchers’ hypothesis that students who reported a natural mentoring relationship with someone of the same sex or ethnic background might score higher on measures of interest (i.e., career interest, career selfefficacy) than those students whose mentor was not of the same sex or ethnicity. In short, research remains inconclusive regarding the capacity of female and male mentors to respond equally well to the unique needs of boys and girls (Liang et al., this volume, Chapter 11). These studies of racial or gender similarity in matching must be viewed as preliminary, as none experimentally varied matching such that some youth randomly received mentors with similar characteristics. Although these studies examined

individual-level factors as applied to relationships, program-related processes tended to demonstrate a more robust association with early match success (Liang & Grossman, 2007), thereby inviting consideration of the program context alongside individuallevel factors in research on matching. Preferences likely vary by setting (i.e., school based, community based) and developmental stage, suggesting that this area remains important for future exploration. Although research is not yet conclusive, mentor skills, such as mentor attunement, or the ability to respond flexibly to verbal and nonverbal cues (Pryce, 2012), seems to contribute as much or more to relationship quality than do demographic similarities. Recommendations for Future Research Given the limited empirical evidence available on the matching process itself, in-depth qualitative inquiry regarding the role or importance of the matching process on relationship development is recommended as a next step in the development of empirical evidence regarding this important area of mentoring. Close study of the matching process itself is also warranted due to the possibility that variation in program characteristics may provide an alternative explanation for the quality of matches over time (DuBois et al., 2011). Further, given that recent researchers have begun to incorporate standardized measures of relationship preferences and personality type into their research agenda (with seemingly little disruption to the mentoring process), it bears asking whether it is time for mentoring researchers to consistently demonstrate in their article submissions that they (a) ground their matching process in a well-supported theoretical framework, (b) critically evaluate the options for how to match mentors and mentees, and (c) employ psychometrically sound instruments to help facilitate the matching process. Such advances could provide needed continuity and rigor to an area of mentoring that remains minimally researched and understood. In short, though the issue of matching is commonly acknowledged as important in the mentoring literature, published studies do not, as a rule, typically explain the procedures for these matches. It is also significant that, to date, no single overarching theoretical framework is used within the field explicitly to inform the matching process. Finally, it is worthwhile to consider ways by which the power of mentoring’s intuitive appeal may have outstripped our knowledge in this area. Matching is an underdeveloped part of the infrastructure for mentoring programs, and one that could affect the viability of many mentoring programs.

434   PRACTICE AND PROGRAMMATIC CONSIDERATIONS

Practice Best practices seem to point to consideration of program aims as well as characteristics of the mentor and mentee (e.g., interests, geographic proximity, availability, age, gender, race, ethnicity, personality, and expressed preferences) when making a match (MENTOR, 2009). The Elements of Effective Practice for Mentoring (MENTOR, 2009) standards reinforce the importance of an initial meeting between the mentor and mentee with program staff present before decisions about the match are made. Research and conceptual work also highlight the importance of monitoring the match immediately following training, as the initial phase of the relationship is critical to success (Rhodes, 2005). Matching practices within mentoring programs tend to assume four different shapes within the field. First, some mentoring programs assume an “administrator-assigned” model, in which an administrator assigns mentors to mentees. Second, some use a “choice-based” or “youth-initiated mentoring” model (e.g., National Guard Youth Challenge Program; Bloom et al., 2009; Millenky et al., 2011), in which each participant has some input regarding the choice of their partner. Specifically, this model uses a guided process for youth wherein they select potential mentors who meet certain criteria (e.g., geographic proximity, similar experience of life challenges and/or vocational interests, established trust and comfort) from a list generated by the young

person. A third matching model, often used within corporate or educational settings, employs more formal assessments, such as the Myers-Briggs Type Inventory or PeopleMatch. Fourth, still others use program-initiated meetn-greets (Karcher, 2007; Karcher & Santos, 2011) and other facilitative tools to help prospective mentors and mentees get to know one another informally prior to matching. Regarding such program tools, staff are advised to avoid advertising the event in any way that explicitly references matching. Instead, the gathering is presented as a simple “get to know you” event wherein participants engage in name games, icebreakers, and other activities intended to bring out individuals’ traits and preferences. At the conclusion of each activity, participants are invited to write on index cards the names of “those people s/he enjoyed meeting” (Karcher, 2007; Karcher & Santos, 2011), so as to mitigate any confusion or disappointment about who an appropriate match might be for each participant. Table 29.1 provides recommendations for practitioners, including the tools and suggestions associated with the meet-n-greet format. Table 29.2 offers guiding questions for practitioners that relate to matching and to the application of theory and research to program practices within youth mentoring organizations. The field of technology offers yet another home for many of the current innovations to matching processes (Headlam-Wells, Craig, Gosland, &

Table 29.1   Recommendations for Practitioners Topic

Recommendations

Youth engagement/ choice

Solicit youth opinions about how and why matches are created (e.g., how youth see mentoring as it relates to their goals). Provide opportunities for youth choice within the match process (i.e., host a meetand-greet event, consider ways that youth can initiate mentors as part of your recruitment). Take advantage of potential relationship matches between youth and the existing nonparental adults in their lives.

Individual characteristics in matching

Try using different assessment tools (i.e., Myers-Briggs, Strong Interest Inventory) to match youth with adults who share their interests or preferences; alternatively, try matches with different interests or complementary personality traits.

Ongoing match support

Consider ‘trial period’ of match, during which match can explore fit and adjust or rematch as necessary while minimizing emotional consequences. Consider reassessing the match relationship between youth and mentors periodically to determine whether the match is as good a fit as originally decided.

Mentor and Youth Matching   435 Table 29.2   Guiding Questions for Practitioners Theory How might underlying principles from attachment theory, voice and choice, or probability matching theory inform your agency’s assumptions about matching and matching process? For example, what do staff think about the importance of differences and complementarity in matching? What structures do you have for receiving feedback from participants during the relationship that could inform understanding of important considerations in matching decisions? Research What information could be gathered to inform systematic evaluation of the matching process within your program? How could interviews or other qualitative techniques (e.g., systematic observation) be used to inform the ways by which matching impacts relationship outcomes in your program? In what ways is your matching process aligned with available research evidence? How could it potentially be strengthened using research as a guide? For example, are matching processes aligned with what research suggests about the role of race, gender, and socioeconomic status in the matching process? Practice What does your “ideal” matching procedure involve? How do you consider components of the youth’s system (e.g., parent, school personnel) as one component of creating successful “matches”? What innovations are you currently engaging in as part of your matching process, and how might this creative approach be monitored for its effectiveness?

Holdsworth, 2007). Although the use of technology in matching is more prevalent in the business world, youth development practitioners also have available to them a range of technological tools to facilitate mentoring programs and the relationship match more specifically. As an example, iMentor Interactive (iMi), a program developed by iMentor between 2004 and 2006, is a web-based platform that assists mentoring organizations and schools around the world. iMi provides customizable interfaces for mentors, mentees, and program staff in order to facilitate monitored online communication, as well as program processes such as matching. Some programs provide profiles through iMi that prospective participants can complete prior to the match. These profiles include questions informed by the program’s emphasis (e.g., if the program is focused on vocational development, then questions regarding work experience and/or professional interests will be asked). These (and other values) are weighted and responses are processed through an algorithm that results in a matching or compatibility “score,” which the programs can use to facilitate the match. Other packages (e.g., MentorMatch and Coaching &

Mentoring Network Mentor Matching Software; www .learningobservatory.com/uploads/publications/ 1455.doc) also aim to match mentors and mentees electronically, for both in-person and e-mentoring relationships. Similar to matching systems used for romantic relationships, many of these systems ask for participant updates during the match, therefore employing the best practice of checking back to see how well the match worked. To obtain additional information on current innovations in the practice world, we spoke with key informants in youth mentoring programs to learn more about variations within the applied side of mentoring. Programs may vary in strategies used to match mentors and mentees based on the needs of the youth, approaches to match supervision and support, screening practices, and program setting (e.g., school or community based). Through these conversations, it became clear that beyond the core concepts assumed to be meaningful in match relationships (e.g., gender, race, shared interests), programs learned through their own experiences about the other constructs critical to consider in matching their youth.

436   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Researchers and practitioners alike have emphasized the importance of matching motivations with volunteer opportunities associated with mentoring (see Stukas et al., this volume, Chapter 27; Madia & Lutz, 2004). To obtain this information, an agency can incorporate projective questions through which they ask the mentor to imagine the relationship 3 months, 6 months, and 1 year down the road. These questions reveal what one program described by Stukas et al. calls the “personal timeline” of the mentor. In other words, if the young person does not accommodate the mentor’s implicit expectations (e.g., such as at what point the youth will share personal information with the mentor), mentor motivation is more likely to wane. The practitioners with whom we spoke challenged some of the common assumptions regarding matching. For example, staff we interviewed in two of the three programs asserted that mutual interests are of limited importance, despite recent evidence that program impacts are larger in programs that use interest-based matching procedures (DuBois et al., 2011). Instead, personality characteristics, such as level of physical activity and the way each participant processes information, were highlighted as important considerations within community-based matches. Similarly, “willingness to try new things” was identified as a consideration in the matching process. In addition to considerations of individual-level characteristics of mentors and youth, in every agency we interviewed staff who emphasized the role of the parent and larger system (e.g., school personnel) in the matching process (see also Keller, 2005). One program emphasized the importance of matching youth to mentors who do not resemble the youth’s parent in terms of age and stage of life. This can help the youth to conceive of the mentor as a separate relationship from his or her parent, and can reduce the likelihood that the parent will feel threatened by the mentor. It is also important to consider the way that matching is framed to the participants in the process. For example, framing to the participants that the match is built on participant strengths communicates a different message than that of meeting the youth’s weaknesses. Because mentoring is inherently built on the concept that the youth needs an additional adult in his or her life, it is critical to counteract these assumptions by highlighting youth strengths.

Conclusion This chapter draws together a diverse and somewhat disparate literature regarding the understanding and maintenance of effective match pairings in a variety

of relationships. Though a range of theoretical frameworks exist to inform mentoring matches, much mentoring research fails to consistently cite a theory as the basis for the matching process in which the examined programs have engaged, and few projects seem to be attempting to build on a theory as they refine their specific mentoring interventions. We now offer several implications for theory, research, and practice drawn from our consideration of current literature and practice. Closing the Research-Practice Gap This chapter has illuminated the importance of the matching process while outlining the relatively new and undeveloped nature of knowledge in this area. This combination invites researchers and practitioners to work together to create an innovative and evidence-informed approach to bringing prospective mentors and mentees together. Practitioners have many valuable ideas that they share regularly with colleagues, but it is striking how many of these ideas are not yet operationalized in the mentoring matching literature. The development of a mentoring “match scale,” which could incorporate the theoretically important traits outlined in this chapter with the practical suggestions, could enable a variety of mentoring programs to build on one another’s lessons as we seek to understand the matching process better. Additionally, the need to include members of the larger system (i.e., parents, school personnel) in the matching process emerged as a persistent theme in the theory, research, and practice literatures. Such incorporation opens another possibility for using practice wisdom to inform future mentoring research. For example, examination of how the parenting styles with which mentees are raised may impact the match process could develop into an interesting literature. Similarly, input from school personnel could inform the matching process, particularly as it relates to school-based mentoring programs in which teachers or other school staff may have special insight into the needs of the student. Approaching the Matching Process Intentionally and Systematically It is our hope that the information consolidated in this chapter will help practitioners and researchers begin to address the issue of mentor-youth matching in a more systematic and intentional way. While there are several lead researchers and practitioners who have broken ground in this area with creative and thoughtful innovations, many of these have not been

Mentor and Youth Matching   437 operationalized or measured. In other cases, matching continues to present as a somewhat forgotten process of the mentoring relationship, the outcomes of which are left up to providence and good luck. As mentioned earlier in this chapter, the use of technology may offer one innovative way to make the matching process more systematic. Such application could be informed by the theories outlined in this chapter. As an example, programs could create a password-protected, fully secure mechanism whereby youth could sign on and learn about potential mentors (their interests, goals, likes, and dislikes) and have that “search” inform their choice regarding a possible matching process. This process could empower youth by giving them a more direct say in connecting with prospective mentors. Such an approach also mirrors some of the recent literature from the client-directed, outcome-informed psychotherapy literature (Sparks, Duncan, & Miller, 2008) that attempts to privilege the client’s view of the therapy relationship, and may hold promise as we seek to enhance, and develop, this critical and largely unexplored aspect of mentoring relationships.

References Alexander, F., & French, T. M. (1946). Psychoanalytic therapy: Principles and applications. New York: Ronald Press. Bernier, A., & Dozier, M. (2002). The client-counselor match and the corrective emotional experience: Evidence from interpersonal and attachment research. Psychotherapy: Theory/Research/Practice/Training, 39, 32–43. Bloom, D., Gardenhire-Crooks, A., & Mandsager, C. (2009). Re-engaging high school dropouts: Early results of the National Guard youth ChalleNGe program evaluation. New York: MDRC. Brennan, K. A., & Shaver, P. R. (1995). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21, 267–283. Burleson, B. R., & Denton, W. H. (1992). A new look at similarity and attraction in marriage: Similarities in social-cognitive and communication skills as predictors of attraction and satisfaction. Communication Monographs, 59, 268–287. Byrne, D. (1971). The attraction paradigm. New York: Academic Press. Campbell, T. A., & Campbell, D. E. (2007). Outcomes of mentoring at-risk college students: Gender and ethnic matching effects. Academy of Management Learning & Education, 5(1) 25–37. Cassidy, J., & Shaver, P. R. (Eds.) (1999). Handbook of attachment: Theory, research, and practice. New York: Guilford Press.

Converse, N., & Lignugaris-Kraft, B. (2009). Evaluation of a school-based mentoring program for at-risk middle school youth. Remedial and Special Education, 30(1), 33–46. Dicken, C., Bryson, R., & Kass, N. (1977). Companionship therapy: A replication in experimental community psychology. Journal of Consulting and Clinical Psychology, 45(4), 637–646. DuBois, D. L., Holloway, B., Valentine, J., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30(2), 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., & Silverthorn, N. (2005). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92. Flores, L. Y., & Obasi, E. M. (2005). Mentors’ influence on Mexican American students’ career and educational development. Journal of Multicultural Counseling & Development, 33, 146–164. Headlam-Wells, J., Craig, J., Gosland, J., & Holdsworth, J. (2007). Mentoring for management development: An evaluation of new communication models for e-mentoring [Research Memorandum]. Hull, UK: University of Hull. Herrera, C., Sipe, C., & McClanahan, W. (2000). Mentoring school-age children: Relationship development in community-based and school-based programs. Philadelphia: Public/Private Ventures. High/Scope Educational Research Foundation. (2008). Voice and choice. Ypsilanti, MI: Author. Holt, L. J., Brenna, J., & Johnson, V. L. (2008). Enhancing school engagement in at-risk, urban minority adolescents through a school-based, adult mentoring intervention. Child & Family Behavior Therapy, 30, 297–318. Hu, C., Thomas, K., & Lance, C. (2008). Intentions to initiate mentoring relationships: Understanding the impact of race, proactivity, feelings of deprivation, and relationship roles. Journal of Social Psychology, 148(6), 727–744. Karcher, M. (2007). Meet-n-greet: A mentor-mentee matching approach for increasing the prevalence of naturally self-selected mentoring partners in program-based matches. Unpublished manuscript, University of Texas at San Antonio. Retrieved from http://utsasmile.org Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9, 99–113. Karcher, M. J., Nakkula, M. J., & Harris, J. (2005). Developmental mentoring match characteristics: Correspondence between mentors’ and mentees’ assessments of relationship quality. Journal of Primary Prevention, 26, 93–110.

438   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Karcher, M. J., Roy-Carlson, L., & Benne, K. (2006). The role of mentor-mentee ethnicity similarity in mentees’ connectedness to other cultures. Poster symposium at the 14th annual meeting of the Society for Prevention Research, San Antonio, TX. Karcher, M. J., & Santos, K. (2011, March 30). Natural selection: Exploratory test of the meet-n-greet approach to mentor-mentee matching in youth mentoring programs. Paper presented at the 2011 Biennial Meeting of the Society for Research in Child Development, Montreal, Quebec, Canada. Keller, T. E. (2005). A systematic model of the youth mentoring intervention. Journal of Primary Prevention, 26, 169–188. Keller, T., & Pryce, J. (2010). Mutual but unequal: Mentoring as a hybrid of familiar relationship roles. New Directions in Youth Development, 126, 33–50. Kelly, M. S., Raines, J. C., Stone, S., & Frey, A. (2010). School social work: An evidence-informed framework for practice. New York: Oxford University Press. Kendall, D. L. (2007). Does choice matter? The impact of allowing proteges to select their own mentors. Unpublished dissertation, University of Central Florida. Larose, S., Bernier, A., & Tarabulsy, G. M. (2005). Attachment state of mind, learning dispositions, and academic performance during the college transition. Developmental Psychology, 41(1), 281–289. Liang, B., & Grossman, J. M. (2007). Diversity and youth mentoring relationships. In T. D. Allen & L. T. Eby (Eds.), The Blackwell handbook of mentoring: A multiple perspectives approach (pp. 239–258). New York: Wiley. Linnehan, F., Weer, C., & Uhl, J. (2005). African-American students’ early trust beliefs in work-based mentors. Journal of Vocational Behavior, 66, 501–515. Madia, B. P., & Lutz, C. J. (2004). Perceived similarity, expectation-reality discrepancies, and mentors’ expressed intention to remain in Big Brothers/Big Sisters programs. Journal of Applied Social Psycho­ logy, 34, 598–623. Markey, P. M., & Markey, C. N. (2007). Romantic ideals, romantic obtainment, and relationship experiences: The complementarity of interpersonal traits among romantic partners. Journal of Social and Personal Relationships, 24, 517–533. MENTOR. (2009). Elements of effective practice for mentoring. Retrieved from http://www.mentoring.org/ downloads/mentoring_1222.pdf Millenky, M., Bloom, D., Muller-Ravett, S., & Broadus, J. (2011). Staying on course: Three year results of the National Guard youth challenge Evaluation. Retrieved from http://www.mdrc.org/publications/599/over view.html. Pryce, J. (2012). Mentor attunement: An approach to successful school based mentoring relationships. Child

& Adolescent Social Work Journal, 1-21. doi: 10.1007/s10560-012-0260-6 Rhodes, J. E. (2005). Theoretical model of youth mentoring. In D. L. DuBois, & M. J. Karcher (Eds.), Handbook of Youth Mentoring (pp. 30–43). Thousand Oaks, CA: Sage. Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20(3), 3–19. Rhodes, J. E., Grossman, J. B., & Resch, N. L. (2000). Agents of change: Pathways through which mentoring relationships influence adolescents’ academic adjustment. Child Development, 71, 1662–1671. Rhodes, J., & Lowe, S. R. (2008). Youth mentoring and resilience. Child Care in Practice, 14(1), 9–17. Rollins, S. A., Kaiser-Ulrey, C., Potts, I., & Creason, A. H. (2003). A school-based violence prevention model for at-risk eight grade youth. Psychology in the Schools, 40, 403–416. Schippers, M. (2008). Doing difference/doing power: Negotiations of race and gender in a mentoring program. Symbolic Interaction, 31, 77–98. Schwartz, P. (2006). Finding your perfect match. New York: Berkley Publishing. Schwartz, S. (in preparation). Youth initiated mentoring: Investigating a new approach to working with vulnerable adolescents. Unpublished dissertation, University of Massachusetts at Boston. Sparks, J., Duncan, B., & Miller, S. (2008). Common factor in psychotherapy: Common means to uncommon outcomes. In J. Lebow (Ed.), 21st Century psychotherapies. New York: Wiley. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22(4), 331–354. Spencer, R., Lewis, T. O., & Basualdo-Delmonico, A. (2009). Mentoring across differences: How race and class shape the youth mentoring process. Paper presented at the Society for Social Work Research annual conference, New Orleans, LA. Syed, M., Goza, B. K., Chemers, M. M., & Zurbriggen, E. L. (2012). Individual differences in preferences for matched-ethnic mentors among high-achieving ethnically diverse adolescents in STEM. Child Development, 83(3), 896–910. Takeuchi, S. A. (2006). On the matching phenomenon in courtship: A probability matching theory of mate selection. Marriage & Family Review, 40(1), 25–51. Tyrrell, C. L., Dozier, M., Teague, G. B., & Fallot, R. D. (1999). Effective treatment relationships for persons with serious psychiatric disorders: The importance of attachment states of mind. Journal of Consulting and Clinical Psychology, 67, 725–733. Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials of schoolbased mentoring. Social Policy Report, 24(3), 1–21.

30  MENTOR TRAINING Janis B. Kupersmidt and Jean E. Rhodes1

Introduction Mentoring programs have experienced tremendous growth in the past decade, fueled in part by an increased availability of funding for mentoring initiatives (Rhodes & DuBois, 2006) and an increased federal marketing effort to increase volunteerism. Rarely acknowledged in the midst of this expansion and enthusiasm, however, is the importance of training mentors to successfully work with youth. As mentoring programs continue to expand, it is important to evaluate the training available to volunteers and to optimize training opportunities by incorporating evidence-based practices into the development and evaluation of training materials (Rhodes & DuBois, 2006). This chapter explores the theory, research, and practice guidelines that support the inclusion of prematch and postmatch training in all mentoring programs. Importance of Mentor Training and Ongoing Support Training mentors to successfully work with youth is critically important for obtaining positive child outcomes and for preventing the potential harm that can result from unsuccessful mentoring relationships. Ineffective mentoring relationships, which include poor quality and prematurely terminating matches, have been implicated in a range of negative youth outcomes. Indeed, in two, large

random assignment studies, youth in prematurely terminating mentoring relationships showed increases in problem behaviors (such as increases in alcohol consumption) relative to randomly assigned control groups (Grossman & Rhodes, 2002; Herrera et al., 2007) and a recent qualitative study described the profound disappointment reported by youth abandoned by their mentors (Spencer, 2007). By contrast, children and adolescents who are “effectively mentored” (as measured by the quality and length of the relationship) have better outcomes than those who are not mentored. Likewise, a 2002 meta-analysis of mentoring program effects emphasized the importance of appropriate training and support to aid the development of committed and effective mentors. In fact, training of mentors and ongoing support directly affect mentor retention and menteerelated outcomes (DuBois, Holloway, Valentine, & Cooper, 2002; Herrera et al., 2007). The effectiveness of mentor training appears to be manifested through its documented impact on mentors’ feelings of closeness, support, satisfaction, and effectiveness (DuBois, Holloway et al., 2002; Herrera, Sipe, & McClanahan, 2000). These perceptions positively influence both outcomes and duration, suggesting the lasting importance of mentor training and the mentor’s self-perceptions of efficacy to be a mentor for youth outcomes (DuBois & Neville, 1997; Parra, DuBois, Neville, Pugh-Lilly, & Povinelli, 2002).

1

The authors wish to thank Rebecca Stelter, Alison Parker, Johanna Greeson, Ashley Craig, Adrienne Banny, Sable Watson, Sharon Daura, and Graig Meyer for their help with literature reviews, practice recommendations, or comments on this chapter. This chapter was supported in part by grant R43HD057674 from the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD) to the first author. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NICHD or the National Institutes of Health.

439

440   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Findings from related fields also underscore the need for training. A recent meta-analysis of the effectiveness of afterschool programs indicated that empirically-based skills programs involving teacher training had a more positive effect on child outcomes than did afterschool programs that provided no training or skills training for children that did not have a research basis (Durlak & Weissberg, 2007). Another study found that after-school staff members in community-based programs had a stronger intention to continue working in the youth development field when they received professional development training, learned aspects of their job from more experienced staff, and had adequate supervision and support (Hartje, Evans, Killian, & Brown, 2008). Likewise, meta-analyses of the effectiveness of psychotherapy have underscored the importance of therapist training as a means of advancing their understanding of the change process, developing clinical skills, and improving the mental health outcomes of clients (Joyce, Wolfarrdt, Sribney, & Aylwin, 2006). Taken together, these findings suggest that programs need to reliably, consistently, and competently implement evidencebased training methods. Ongoing training is just one form of support that programs must provide to their mentors. Supervision through onsite and match-specific case management and crisis intervention also is essential but is not addressed directly in this chapter. Rather, we focus on key concepts and skills that are essential to provide to all mentors during the life of the match through prematch and ongoing training. Some mentors may need specific forms of training to bolster their skills or to support their work with specific types of youth, but this is not the same as the day-to-day, on the ground, and one-on-one support that must be provided through mentor supervision efforts. However, to date, even less is known from a research perspective about supervision of mentors. For this reason and to do adequate justice to issues involved with mentor training, the present chapter is limited to this topic.

explain the growing difficulties with volunteer retention, a particularly troubling trend given the adverse effects associated with breakdowns of relationships (Grossman & Rhodes, 2002). Indeed, high rates of volunteer attrition represent a major drain on staff and financial resources in mentoring programs, particularly given the effort involved in recruiting, screening, training, and matching volunteers. Carefully designed, empirically tested prematch training of potential mentors and ongoing support to postmatch mentors, particularly early in the relationship, may address the problem of unanticipated volunteer attrition by helping to retain mentors and, in turn, supporting effective relationship development.

Theory The theoretical framework used for explaining the role of training in promoting longer-term, more satisfying mentoring relationships for mentors is presented in Figure 30.1. Training mentors in the cognitions and behaviors needed to cultivate successful mentor-mentee relationships is expected to improve the satisfaction and quality of the relationship, which should create the foundation for increasing the length of the relationship. In turn, these qualities are expected to increase positive outcomes for mentees. In other words, the effect of prematch training on positive outcomes for youth is hypothesized to be mediated by the quality and length of the mentor-mentee relationship. Briefly we consider these two mediators: relationship quality and match length. Then we summarize some of the key elements of training that we think influence relationship quality and match length. In the next section, on research, we review empirical findings on the importance of the training elements we’ve identified. Finally, in the practice section at the end of the chapter, we summarize the frequency with which training topics are addressed in a review of 15 program training manuals.

Need for Mentor Training Programs Programs that rely predominantly on volunteers’ natural skills may miss important opportunities for improving program effectiveness. Particularly given the linkages between volunteer training and youth outcomes, all mentoring programs should include purposeful, innovative, and consistently implemented training content and methods. Regrettably, however, there are few evidence-based training programs for volunteer mentors. This gap might

Mentor-Mentee Relationship Quality Beneficial effects of youth mentoring depend on the extent to which the mentor and youth manage to forge a strong connection characterized by mutuality, trust, and empathy. As Levinson (1979) observed, “Mentoring is not a simple, all-or-none matter” (p. 100), and if a bond does not form, youth and mentors may disengage from the match before the relationship lasts long enough to have a positive

Mentor Training  441 Figure 30.1   Model of Mentor Training and Youth Outcomes

Mentor-Mentee Relationship Quality Positive Outcomes (e.g., mentee’s grades, emotional well-being, prosocial behavior)

Mentor Training

Relationship Length; Mentor Retention

impact on youth. Theoretically, a meaningful connection becomes possible only to the extent that the mentee is willing to share his or her feelings and self-perceptions and is actively engaged in the relationship (Csikszentmihalyi & Rathunde, 1998). Dworkin, Larson, and Hansen (2003) described this process in terms of both motivation and concentration, wherein the youth is “involved in actively constructing personal change” (p. 17). This emphasis on the mentor’s empathic responding to the youth’s sharing of thoughts and feelings does not imply that every moment needs to be packed with profundity and personal growth. It seems more likely that successful youth mentoring is characterized more often by a series of small wins that emerge sporadically over time. Yet these mundane moments, which might be laced with boredom, humor, and even frustration, can help to forge a connection from which the mentee may draw strength in moments of vulnerability or share triumph in moments of accomplishment. Mentors must also be engaged in the relationship and attuned to the particular qualities and circumstances of their mentees. Research has indicated the benefits of attunement in adult-youth relationships (Pianta, 1999). Allen et al. (2003), for example, found that more empathic and supportive parenting was predictive of attachment security among adolescents. Such parents appeared to be better able to provide the sort of safe haven that

youth needed to take on challenges and cope with emotional stress. By the same token, mentors who are attuned to their mentees are likely to be in a better position to handle discussions around vulnerable topics without undermining the youngsters’ sense of self-confidence. Specific dimensions of relationships that indicate the presence of a high-quality relationship include positive companionship (sharing activities and time together), instrumental aid (seeking and accepting assistance), trust, mutual respect, intimacy (sharing of personal thoughts and feelings), low conflict, and high overall satisfaction (Furman & Buhrmester, 1985). These relationship qualities, in turn, are hypothesized to be associated with positive outcomes in mentored children. Relationship quality may also be manifested through mentors serving as positive role models and providing mentees with mutual respect and opportunities for sharing thoughts and feelings. In doing so, mentors can help mentees broaden and build their personal resources and learn to approach negative experiences as opportunities for intimacy and learning. By providing training in active listening, empathy, and problem-solving skills to mentors in the context of a mentoring relationship, the quality of the mentoring relationship may be enhanced. Through training, mentors should demonstrate an improved ability to provide appropriate types and levels of intimacy, effective opportunities for

442   PRACTICE AND PROGRAMMATIC CONSIDERATIONS conflict resolution, and general relationship satisfaction. In addition, exposure to these beneficial experiences with mentors may generalize to other relationships, enabling youth to interact with peers and other adults more effectively. Relationship Length and Mentor Retention If the positive impact of mentoring occurs, over time, through the context of an ongoing supportive relationship with a caring adult, then there may be a critical relationship length needed to achieve positive child outcomes. However, if a bond does not form between the mentor and mentee, then one or the other member of the pair may disengage from the match before the relationship lasts long enough to have a positive impact. The proposed model suggests that training of mentors, which includes ongoing support, may be the key strategy needed for increasing mentor retention and decreasing premature termination in the practice field. Some evidence supports this hypothesis in that training was found to be a predictor of relationship length and to directly affect mentor retention and mentee-related outcomes (DuBois, Holloway et al., 2002). Thus, the premature failure of mentoring relationships may result, in part, from inadequate training of mentors. Training: Duration, Content, Timing, and Mode of Delivery The content and methods of mentor training include several factors that are related but distinct. One critical factor includes the amount of time that should be devoted to training, specifically the frequency, dose, duration, and timing of training. For example, practitioners seek guidance in terms of whether training should occur prematch and/or postmatch, as well as how many hours of training are needed to competently and effectively prepare mentors for each phase of the mentoring relationship. Another factor is the content of training. The content of training needs to address each of the research findings, safety concerns, and agency policies associated with being an effective mentor. A third factor concerns the modality of training that is optimal for achieving successful outcomes; the mode can include in-person, written, or webbased delivery, or blended learning that incorporates some combination of these approaches. In the next section, we review the research addressing each of these dimensions as well as findings from a survey of mentoring agency procedures related to training.

Research In this section, we review research on the prevalence of mentor training as well as the research supporting the theorized mediators of program impact—relationship quality and length. We also provide information on the most commonly covered, research-based methods and content areas in youth mentor training. We organize this content using some conceptual heuristics from the field and also suggest areas of mentoring that we believe need greater attention both in research and in practice. Training Prevalence Two surveys suggest that the prevalence of prematch and ongoing training provided by programs is less than universal. In a somewhat dated survey of a diverse set of over 700 mentoring programs, most programs (more than 90%) reported providing their volunteers with at least some orientation or training; however, surprisingly, 14% required an orientation, but no actual training (Sipe & Roder, 1999). In addition, we know from 53 mentoring programs that were included in a meta-analysis on the effectiveness of mentoring that prematch training or orientation was provided to mentors in 71% of the studies and ongoing training was provided to postmatch mentors in only 23% of the studies (DuBois, Holloway et al., 2002). The programs included in this review were evaluated rigorously and may not be representative of current training practices in the field, possibly even inflating rates higher than the average service provider in the mentoring field. More recently, a survey was conducted of 131 Big Brothers Big Sisters (BBBS) agencies in the United States providing community-based mentoring programs (Wheeler & DuBois, 2009). Of these programs, 83% provided prematch volunteer orientation/training on an individual basis and 48% provided such training on a group basis. Hence, although training to be a mentor is supported by research, this practice has not been adopted throughout the field. Relationship Quality We cover research on the importance of relationship quality only briefly, because many program staff assume this is a given essential element of effective mentoring relationships, but also because elements of relationship quality are covered in Nakkula and Harris (this volume, Chapter 4). The importance of “the relationship” in youth mentoring matches has been the focus of much research. After

Mentor Training  443 examining survey data on more than 600 mentoring pairs in community- and school-based programs, Herrera et al. (2000) concluded, “at the crux of the mentoring relationship is the bond that forms between the youth and mentor” (p. 72). Consistent with this hypothesis, a national evaluation of BBBS mentoring programs found that levels of trust and closeness in mentoring relationships predicted positive academic and behavioral child outcomes above and beyond the effects of relationship length (Rhodes, Reddy, Roffman, & Grossman, 2005). Thus, the quality of the mentor-mentee relationship clearly influences child outcomes. The extent to which mentors and youth establish a strong connection is influenced by the dynamics of their interactions with each other (Rhodes & DuBois, 2006). In general, close and enduring ties are fostered when mentors adopt a flexible, youthcentered style in which they emphasize the young person’s interests and preferences rather than focusing predominantly on their own agendas or expectations for the relationship. Thus, although a structured or goal-oriented dimension to a mentor’s activities with a young person can be beneficial, these advantages may be compromised if the overall approach is not sufficiently collaborative in its orientation.

attendance, employment), mental health (e.g., selfesteem, life satisfaction), problem behavior (e.g., gang membership, physical fighting, risk taking), and health (e.g., exercise, birth control) (DuBois & Silverthorn 2005). Taken together, these studies suggest that longevity is critical for achieving positive child outcomes. Training Length Training length is important for producing positive match and mentee outcomes. A combination of more hours of training and supervision implemented with a coherent approach (e.g., interpersonal, behavioral) has been associated with increased mentor effectiveness, as compared to mentors who received fewer hours of training implemented with a nonspecific approach (Davidson & Redner, 1988; Davidson, Redner, Amdur, & Mitchell, 1990). In fact, mentors who received less than 2 hours of prematch training reported the lowest levels of closeness with their mentees, as compared to mentors who received at least 6 hours of training, who reported the highest levels of closeness (Herrera et al., 2000; Herrera et al., 2007). Training Timing

Relationship Length The importance of relationship length is also commonly accepted by most program staff, but it also is founded on considerable empirical evidence. In fact, this hypothesized relationship between the longevity of the mentoring relationship and child outcomes has been demonstrated in several studies. For example, Grossman and Rhodes (2002) found that positive effects on youth outcomes became progressively stronger as relationships persisted for longer periods of time and were greatest when relationships lasted at least 1 year. However, for youth in relationships that terminated prematurely within the first 6 months of the match, there were no clear benefits. Likewise, in an evaluation of a schoolbased mentoring program, mentor attendance significantly predicted changes in children’s social skills and self-esteem (Karcher, 2005). Similarly, a recent study analyzed data from the National Longitudinal Study of Adolescent Health (DuBois & Silverthorn, 2005), involving a nationally representative sample of adolescents who were first assessed in grades 7–12. In this case, compared to teens who did not have a stable mentor, those who reported having experienced a mentoring relationship since the age of 14 exhibited better outcomes in education/work (e.g., high school completion, college

Training needs will likely vary according to the stage of the mentoring relationship. Murphy and Ensher (2006) described mentoring relationships as progressing through five general phases. First, in the introductions phase, participants look for similarities and make judgments. Second, in the relationship-building phase, mentees and mentors engage in positive activities that give them things to talk about and remember and opportunities to spend time together. Third, in the growth phase, the mentor uses relatively open communication and role modeling to provide emotional and instrumental support to the mentee. Fourth, in the maturation phase, the pair focuses on the mentee’s goals, and the mentor begins to derive benefits as well. Fifth, and finally, in the transition phase, the relationship declines or is redefined. Similarly, Keller (2005) describes four stages of a mentoring relationship including the contemplation stage, which occurs prematch when the mentor is learning about mentoring and is anticipating meeting his or her mentee; the initiation stage, when the mentor and mentee are matched, and they begin to get acquainted with one another; the growth and maintenance stage, when they meet regularly and build their mentoring relationship; and either the closure or the redefinition stage, when the relationship ends or

444   PRACTICE AND PROGRAMMATIC CONSIDERATIONS evolves such that the relationship is renegotiated or rejuvenated. Prematch training coincides with the contemplation stage of mentoring relationship development (Keller, 2005) and, hence, should focus on this delicate time during which the mentor is anticipating and preparing for the relationship. One overarching purpose of prematch training is to increase readiness to mentor and a sense of self-efficacy as a mentor to prepare for both the inception of the match and the early stages of the relationship, during which time the bond between mentor and mentee is developing. Consistent with this idea, mentors’ self-efficacy has been found to affect the quality of the mentoring relationship as well as youth outcomes (DuBois & Neville, 1997; DuBois, Neville, Parra, & Pugh-Lilly, 2002; Hirsch, 2005; Karcher, Nakkula, & Harris, 2005; Morrow & Styles, 1995). For example, Parra and colleagues (2002) found that a mentor’s initial belief that she or he could perform well in the relationship was related to an array of positive outcomes, including greater contact between the pair, greater involvement in program-related activities, and fewer obstacles to relationship development. Thus, a key goal of prematch training, among other things, is increasing knowledge and skills as well as optimism and enthusiasm in the service of increasing perceptions of self-efficacy to be an effective mentor. These stages can serve as a roadmap for the outline of a training program, forecasting periods of instability, and providing benchmarks for evaluating progress. Caseworkers often describe a sequence in which youth in relatively new relationships sometimes make a dramatic “leap to health” in their attitudes and behaviors, followed by a setback and an unraveling of earlier gains. The setback often coincides with the transition from the relationshipbuilding stage of the relationship to the growth stage. This transition is the point when mentors begin to leverage the trust and good will that has been built up to focus on points of vulnerability. This stage can be unsettling to mentees, who may begin to miss meetings or withdraw from discussions. Training can play an important role in helping mentors and mentees understand setbacks and maintain or restore momentum. Training Content Mentor training plays an important role in guiding mentors’ beliefs about their role as mentors. A series of exploratory interviews with volunteers and youth in successful and terminated relationships revealed that unfulfilled expectations, pragmatic concerns, and common frustrations in the

early, vulnerable stages of relationship development were more common in the least successful relationships (Spencer, 2006). Mentors who are not trained on the realities of working with youth may experience disappointment and, in turn, may terminate the mentoring relationship prematurely. Broadly speaking, difficulties arise from four major failings: the misuse of power (e.g., exploitation; political or religious proselytizing), inappropriate boundaries (e.g., breaching confidentiality; improper disclosures), communication breakdowns (e.g., breaking commitments), and early or unexpected termination of the mentoring relationship (Grossman & Rhodes, 2002; Kalbfleisch, 2002). Training in the goals of mentoring; the roles, behaviors, and expectations of mentors; the ethical issues facing mentors; and issues unique to the type of child to be mentored all can reduce the likelihood that these difficulties undermine relationship quality and length. Introduction to Mentoring. All mentoring programs provide prematch mentors with basic information regarding the mission and goals of the program as well as expectations of mentors for the frequency and length of contact. This information may be provided in person or in writing, but it is fundamental to setting up clear program expectations to avoid confusion or disappointment. In addition, mentors may not have a “professional” view of the practice of mentoring and could benefit from information about the rationale for requiring prematch training and postmatch support. Providing this type of explanation can lower resistance to training requirements, especially in-person training, which mentors may find challenging to complete. In contrast, providing a clear, cogent rationale for training may increase mentors’ commitment to engaging in training and support activities that typically do not involve direct interactions with mentees. Motivation. The goals and personal motivations of mentors are especially influential during the early stages of the relationship. A mentor’s motivation may influence contemplation activities, including obtaining information about mentoring, planning for future activities, and forming expectations about the relationship (Keller, 2005). For example, Karcher et al. (2005) found that mentors’ motivations predicted relationship quality early in the match, and that by the end of 1 year, the association between motivation and relationship quality was no longer significant. Therefore, addressing mentor motivation is a critically important training topic at prematch, a period of time involving anticipation and preparation for the relationship. It may be helpful

Mentor Training  445 to raise mentors’ awareness of their motivation to mentor to prevent the kinds of unrealistic motivations that can negatively affect relationship quality or length (Spencer, 2007). Many factors inspire people to become mentors (Gehrke, Jenkins, Miskovetz, & Wray, 2006), and most mentors have more than just one motivation (Stukas, Daly, & Clary, 2006). Having more than one goal allows mentors to sustain their motivation, because if one goal is not achieved, then others may be. Common goals of mentors include giving back to society, the intrinsic value of spending time with a child, and personal life enhancement (Stukas et al., 2006). Stukas, Clary, and Snyder address motivations in greater detail elsewhere in the Handbook (Chapter 27). Mentors and mentees may experience difficulties when their goals for the relationship do not match. In fact, congruency in goals among the mentor, the mentee, and parents was predictive of less conflict in the mentor-mentee relationship (Meissen & Lounsbury, 1981). Educating mentors about this important finding will provide them with an incentive to learn about their mentees’ goals for the relationship as well as the goals held by their mentees’ parents. If the goals between these individuals are not in alignment with one another, then compromises may be needed to reduce discrepancies between the goals of different pairs of individuals in the service of sustaining motivation to continue to build and develop the mentoring relationship. Expectations. It is critical to help mentors establish realistic expectations for the mentoring relationship. Expectations for mentoring will likely be shaped at least somewhat by the stated goals, guidelines, requirements, and training practices of the mentoring program. For example, communication of clear program expectations regarding mentors’ frequency of contact and anticipated duration of the mentormentee relationship has been found to be associated with successful mentoring programs (DuBois, Holloway et al., 2002). Mentors also have personal expectations about what the mentoring relationship will entail (Keller, 2005). Mentors’ unfulfilled expectations often contribute to the demise of mentoring relationships (Spencer, 2007). In fact, Spencer reported that many mentors enter mentoring relationships with romanticized expectations of “making a difference,” but that these expectations often prove to be more challenging than they had anticipated. When imagined rewards are not realized immediately or take a different form, mentors may decide that the relationship is not what they had bargained for and may end the match prematurely.

Training can help the mentor and youth calibrate their expectations. Both parties are likely to have preconceptions about the relationship that influence how they judge one another over its course. One of the challenges for case managers is to uncover and correct unrealistic expectations. Mentors could receive training in the stages of development of the mentoring relationship as well as knowledge of the mentee’s socioeconomic and cultural contexts (Hirsch, 2005). Increased knowledge about the effects of risk and protective factors on children’s development may contribute to the development of more realistic expectations for the early stage of the relationship. Thus, in general, an important prematch and early-match training focus is on having modest expectations of a mentee and encouraging patience in the process associated with developing a mentoring relationship with a child or an adolescent. First Meeting Between Mentor and Mentee. Mentors need training on what to expect regarding the match process and during the first meeting with their mentee and their mentee’s family. They also need to know what to expect from their mentoring organization at the first meeting with their mentee, including whether or not someone from the organization will be present, where the meeting will be held, and the length of the visit. Roles. Rhodes, Liang, and Spencer (2009) discussed the importance of role clarity in mentoring relationships. Individuals with prior experience in helping roles or occupations (DuBois, Holloway et al., 2002) have been found to be able to establish closer, more effective mentoring relationships. Perhaps one reason for the success of this group of people is that they have prior training regarding boundaries in relationships and the roles that a mentor should and should not play in the life of their mentee. It would likely be helpful for mentors to analyze the possible negative consequences of trying to fulfill roles that are not appropriate and could be harmful to the mentee and/or the relationship. For example, trying to be a mentor to the family, acting like a social service or health provider to the child or family, or trying to be the mentee’s parent or caregiver all involve boundary violations. Also, Darling, Hamilton, and Shaver (2003) cautioned mentors about seeing themselves as a substitute for adequate parent and community support, and that, in particular, they should not try to provide the same level and type of emotional support as their mentees’ caregivers. One of a mentor’s primary roles is typically to serve as a role model to the mentee. Role models are usually defined as adults who are worthy of imitation

446   PRACTICE AND PROGRAMMATIC CONSIDERATIONS in some area of life (Pleiss & Feldhusen, 1995). Mentees frequently express admiration for their mentors and a desire to be like them (Liang, Spencer, Brogan, & Corral, 2008). Mentors have many opportunities to be a positive role model for mentees. For example, mentors have been helpful to mentees vocationally by serving as role models for the types of behaviors and skills needed for achieving positive job performance in the workplace (Hamilton & Hamilton, 2005). Interpersonally, mentors can provide a model of effective communication skills and they can help their mentees to better understand, express, and regulate both positive and negative emotions (Pianta, 1999). Mentors who openly display positive emotions, particularly under difficult circumstances, actively model the process of using positive emotions constructively (Denham & Kochanoff, 2002; Spencer, 2007). In doing so, mentors can help mentees to broaden and build their personal resources and to approach negative experiences as opportunities for developing healthy intimacy in relationships, self-discovery, and self-growth. Mentors also need to be alerted to the fact that they may serve as a negative role model to mentees by engaging in risky or unhealthy behaviors such as substance use (Beam, Gil-Rivas, Greenberger, & Chen, 2002). Research suggests that even just talking in a neutral or joking way about these types of behaviors in the context of peer relationships can serve to glamorize or reinforce risky and unhealthy behaviors such as delinquency and substance abuse (Dishion, Spracklen, Andrews, & Patterson, 1996). In addition to serving as a role model, mentors should be a trusted adult friend. A trusted adult friend provides a mentee with positive companionship and shared activities, but is distinct from an adult, peer friend. For example, members of a pair in a peer adult friendship will likely share their concerns, successes, and failures. Although it is likely beneficial to the child and the relationship for mentors to be empathic and good listeners such that the mentee is comfortable sharing his or her thoughts and feelings, the challenges and issues faced by the mentor may not be appropriate to share with the mentee. It also takes the focus of the relationship away from the mentee and promoting the mentee’s development. Sharing of personal stories by the mentor should be done in a strategic way and not to fulfill a personal need. Many mentors need training to understand why it is not an entirely equal and reciprocal relationship. Another important role for the mentor is to be a nurturer of possibilities. The mentor can be trained and supported to tune in to the mentee’s interests and abilities and to help the mentee develop skills, by

serving as a liaison to important resources. It is important for mentors to create new opportunities for positive growth by introducing their mentees to new experiences, skills, places, people, and ideas. Behaviors of Successful Mentors. After reviewing the many behavioral characteristics included in mentor training programs, we proposed in our Building the Foundation web-based mentor training course that the three truly essential behaviors of successful mentors are the “3 B’s of Mentoring: Be trustworthy. Be empathic. Be authentic.” Successful mentoring relationships are characterized by mentees as consisting of mutual trust (Liang et al., 2008). Research has corroborated this sentiment, as high levels of trust within mentoring relationships are associated with positive academic and behavioral outcomes above and beyond the effects of relationship length (Rhodes et al., 2005). As a characteristic of trust, mentor dependability is related to mentees’ increased self-esteem, social skills, and behavioral competence. The lack of dependability is related to declines in mentees’ selfesteem and social skills (Karcher, 2005). Youth describe beneficial mentor relationships as including the development of a positive, consistent, available, trusting, and caring relationship (Dallos & ComleyRoss, 2005). Based on these findings, mentors can be trained to adopt a gradual trust-building approach with their mentees in which they learn a variety of ways to demonstrate their trustworthiness. Empathy, described as the ability of individuals to recognize themselves in someone else, is also a critical characteristic of effective mentors. One of the positive benefits of empathy is that it promotes close emotional connections between mentors and mentees (Spencer, 2006). Mentors who are attuned to their mentees are likely to be in a better position to handle sensitive discussions around vulnerable topics without undermining the youth’s selfconfidence. Training could include demonstrations of empathic expressions in interactions with mentees, while highlighting the elements of the communication that demonstrate empathic expression. Authenticity also fosters close emotional connections between mentors and their mentees (Spencer, 2006). Therefore, it is important that mentors be themselves by being honest about their interests, abilities, and limitations. “Be authentic” emphasizes the importance of authenticity in the development of the mentor-mentee relationship. Even though there may be a desire to be perceived by a mentee as “cool,” mentors who are willing to be themselves are better able to express genuine feelings toward their mentees.

Mentor Training  447 Having Fun. The cornerstone of successful mentoring relationships has been described by mentees as mentoring support within the context of or as enhanced by shared activities that are fun (Liang et al., 2008). Having fun is not just an important part of relationship-building, it also provides youth with valuable opportunities that may not be feasible due to family situations (Sipe, 2002). In fact, overly focusing on academic activities may detract from match success (Karcher, 2007), whereas rapportbuilding activities have been associated with better child outcomes (Karcher, Kuperminc, Portwood, Sipe, & Taylor, 2006). Mentors may benefit from training on the importance of getting input and feedback regarding what a mentee finds fun. Population-Specific Content Some programs may need to develop programspecific supplemental content to reflect the unique needs and issues faced by special populations of youth (Spencer, 2006). Issues associated with mentoring children of prisoners, children in the foster care system, children in the juvenile justice system, children who have dropped out of school, or immigrant children may be important to address in tailored training content. As described by Birman and Moreland (this volume, Chapter 24), immigrant children constitute a population that may face challenges that are unique to their situation, including stress related to discrimination, poverty, and separation from family members in their home country (MENTOR, 2009b). The training may need to be tailored for mentors of immigrant children to raise their awareness of the challenges that their mentees may be facing, as well as to heighten their cultural sensitivity. Training on the negative impact of early termination may be particularly important, given that such terminations can be deleterious for these children, especially if they have experienced the loss of family members during the process of migration (Suarez-Orozco & Suarez-Orozco, 2001). As suggested by Eddy et al. (this volume, Chapter 25), mentors would also benefit from training related to the challenges faced by children of prisoners. For example, children with incarcerated parents may have difficulty trusting adults and feelings of social stigma (Adalist-Estrin, 2006). Because they have an incarcerated parent with a criminal record, they may believe that no one trusts them and they may have issues with trust themselves due to their unstable family situations (Adalist-Estrin, 2006). Issues associated with trust, including consistency, open-mindedness, and dependability, can be

especially important content to include in training these mentors. Mentees may also be embarrassed about a parent’s incarceration, and mentors should be introduced to skills to respond effectively in the event that these feelings are disclosed to them (Morrow & Styles, 1995). Likewise, mentors should be aware of the role of gender in shaping relationships for reasons explained by Liang and Bogat (this volume, Chapter 11). Findings from diverse disciplinary perspectives shed light on how gender might affect adultyouth mentoring relationships. Recent findings (Rhodes, Chan, & Lowe, 2010), for example, suggest that for girls, the quality of their attachment relationships with mentors generalizes to the quality of their relationships with their parents and peers. Mentoring relationships do not generalize in the same way among boys. Boys may not yet be as oriented to nonparental relationships and are generally more likely to stress their independence in and across relationships (Crosnoe & Elder, 2004). Taken together, such gender difference might affect mentoring relationships, including their duration and perceived importance and helpfulness. Prematch and/or Postmatch Training and Support Topics The topics addressed in this section can and should be introduced to mentors before the match occurs; however, they will become increasingly salient and relevant to mentors as the mentoring relationship develops. Hence, these topics represent excellent subjects to be included as part of postmatch training. Building a Positive Mentoring Relationship. The quality of each mentoring relationship depends on the extent to which the mentor and mentee come to know, respect, and trust each other. Effective relationship development, including both communication skills (e.g., active listening, use of humor), needs to be explored. Also, conflict resolution skills (e.g., negotiating, compromising, distracting) need to be examined as well as the types of situations in which these types of skills would most appropriately and effectively be deployed. For example, problems arising from the mentor’s feeling unappreciated; from the mentor’s concern that the mentee is not sharing feelings, initiating conversation, or calling the mentor; and from the mentor’s thinking that the mentee is not interested in the match or showing motivation to improve his or her functioning should be addressed. Related to the previously discussed topics, learning about the pros and cons of different orientations to

448   PRACTICE AND PROGRAMMATIC CONSIDERATIONS the mentoring relationship can be particularly beneficial. One overarching framework that has been proposed concerns whether mentoring interactions and activities are primarily youth-centered, sometimes called a developmental relationship, versus driven by the mentor’s goals and interests, referred to as a prescriptive relationship. Developmental relationships have been associated with greater relationship quality and duration than prescriptive relationships (Herrera et al., 2000; Morrow & Styles, 1995; Styles & Morrow, 1992). However, the most positive outcomes appear to emerge when mentees report experiencing structure as well as support in their mentoring relationships (Langhout, Rhodes, & Osborne, 2004), suggesting a need for mentors to be more than simply “good friends.” Karcher and Hansen (this volume, Chapter 5) provide an integrated framework for thinking about the importance of mentor and mentee input, the balancing of play and learning, and the need to both move toward goals but also appreciate the present. Thus, helping mentors to translate and operationalize these complex research findings into practice is an important part of training and support. Closing the Mentoring Relationship. Terminations, even planned ones that follow successful relationships, can evoke conflicting emotions and defensive reactions. The termination of a relationship with a child has not been given adequate attention in many mentor training programs (see Spencer and Basualdo-Delmonico, this volume, Chapter 32). For at least one child outcome (i.e., alcohol use), Grossman and Rhodes (2002) found a significant increase in problems among mentees who experienced premature terminations. Their findings indicate that primacy of longevity as a key factor in the success of mentoring and that the effects of premature termination are not benign. Premature termination may be interpreted by the mentee as a personal failure or personal rejection, resulting in an iatrogenic effect of mentoring or the lack of effective mentoring. Unfortunately, as many as half of volunteer mentoring relationships end prematurely and most end at the request of the volunteer (Rhodes, 2002). This rate is disturbingly high and reveals the importance of using training strategies to directly address methods for increasing mentor retention, particularly during the early, vulnerable stage of the relationship. Although the processes used in the termination of mentor-mentee relationships are rarely discussed, how these types of relationships end can color the ways that mentees think about their entire experience with a mentor. In addition to preventing feelings of abandonment and loss, a well-handled termination

can provide a healthy model for sharing feelings around other losses as well. For children who have not had many positive experiences with prosocial adults, the successful termination of a relationship with a mentor can teach a child that he or she is capable of having warm, positive relationships with an adult and has the skills to do it again. Mentors who anticipate an impending termination should give their mentees ample warning. Mentors should be provided with clear guidelines regarding how to end the relationship, and mentors should be held accountable to them. Terminations of mentoring relationships should always be planned, the steps should be clear, and the child and his or her family should be given ample notice and explanation. Activities should be designed to help mentors review the mentoring relationship, highlight accomplishments, address issues of loss, and reframe the ending of the relationship in a positive light with their mentees. Ethical and Professional Issues. Unfortunately, ethical issues have been largely absent from the field of mentoring until recently, when Rhodes et al. (2009) proposed a set of ethical principles for youth mentoring relationships. The relative absence of discussion of this topic is alarming, particularly when we consider the sheer number of volunteer mentors who are entering into relationships with youth. Training should introduce how moral quandaries arise while building mentoring relationships and address strategies for resolving these moral dilemmas. The ethical imperatives built on the principles proposed by Rhodes and colleagues as well as from the ethical codes of the American Psychological Association (APA) and the National Association of Social Workers (NASW), and grounded in fundamental moral principles, are to do no harm, to do good, to respect others, and to treat all individuals honestly and fairly (Fisher, 2003). Although the APA code tends to focus primarily on issues pertaining to individual responsibility (e.g., confidentiality), the NASW code also addresses the complex issues that can arise within diverse social and cultural contexts. Therefore, it will be important to consult both professions’ codes of ethical conduct in developing basic principles of training of mentors regarding ethics in mentoring, as described by Rhodes et al. (this volume, Chapter 35). Training Format We located no research that provided information about the specific formats or methods of delivery currently used in mentor training programs. Presumably, most mentor training is currently

Mentor Training  449 delivered in person and in small groups. In addition, the 3rd edition of the Elements of Effective Practice for Mentoring (MENTOR, 2009a) has set a 2-hour minimum benchmark for in-person training as one of the two evidence-based standards for mentoring programs. However, given the time and cost constraints associated with providing more extensive in-person training, the incorporation of web-based mentor training into standard training protocols for mentoring organizations has a number of distinct advantages. The use of an on-demand, web-based training program for provision of mentor education allows for place and time independence (Vrasidas & McIsaac, 2000). This type of computer-mediated communication allows for interactive, self-paced learning and facilitates self-reflection (see Shpigelman, this volume, Chapter 17). Volunteers can complete individual lessons on-demand without having to complete a whole training in one sitting. In addition, online training will appeal to busy volunteers, who can access the training at times that suit their schedules and locations. Whereas some programs might opt to rely entirely on in-person training, others may wish to conduct blended training by supplementing their regular in-person training with this type of enhancement. Another strength associated with the use of a web-based platform is that it allows for universal updates of the training content as new research findings become available. Several examples of mentor training opportunities are currently available on the web, some of which we describe in the next section. However, none has been evaluated rigorously to date.

Table 30.1

In the survey by Wheeler and DuBois (2009), questions were included on both the presence of and the technology used in providing postmatch training and support. As can be seen in Table 30.1, the most common form of match support is individualized, which is required by the majority of BBBS community-based agencies. In addition, more than half of the agencies offer group training sessions that mentors are encouraged (but not required) to attend. In fact, attendance at group training is required by a negligible number of agencies. Peer learning, found to be important in many preventive intervention programs with youth, is rarely required, but is offered by more than half of the participating agencies. Web 2.0 technology (e.g., Facebook, forums) for communication among mentors or with staff is rarely required and is unavailable in more than half of agencies. Postmatch online training is rarely required or even available. Taken together, these results suggest that both prematch mentor training and postmatch mentor support are common, but they are not universal practices across mentoring agencies. Summary Mentor training affects mentor retention and youth outcomes. However, there are few, if any, evidence-based training programs. Many mentors manage to seamlessly connect with mentees despite having had very little training or supervision. But mentoring does not come easily to everyone, and difficulties sometimes arise that overtax even the most skilled individuals. As mentoring programs continue to proliferate or expand, it is important to evaluate pre- and postmatch training efforts and to

Percentage of Big Brothers Big Sisters Agencies Offering Different Types of Postmatch Training and Support (N = 80) Agency Response

Type of Postmatch Training Support

Required

Recommended

Unavailable

No Response

Individualized match support

88

10

 0

 2

Group training sessions

10

65

23

 2

Peer learning

 2

63

28

 7

Web 2.0 communication among mentors

 1

25

60

14

Web 2.0 communication with staff

 1

27

56

16

Online training

 2

10

72

17

Source: Based on data reported by Wheeler & DuBois (2009).

450   PRACTICE AND PROGRAMMATIC CONSIDERATIONS optimize those efforts. We need to know more about the core critical elements of successful training and how these might vary as a function of the needs and characteristics of mentees. Such issues, which are complex and tap into the very heart of what constitutes a helping relationship, could be incorporated into evidence-based training practices.

Practice Overview of Training Curricula and Evaluations Appropriate training and structure must be in place to support the development of committed and effective mentors. The most widely used resource by mentoring programs, Elements of Effective Practice for Mentoring (EEPM), was revised and updated in 2009 by an advisory committee of MENTOR: The National Partnership for Mentoring (led by the authors), and the 3rd edition was distributed to more than 35,000 constituents through the organization’s website (MENTOR, 2009a). This new edition provides not only excellent guidance regarding the administrative and business aspects of creating a new mentoring organization or program, but also specific research-based advice regarding program operations including the training of individual mentors. Training Length Practices and Recommendations Primarily based on Herrera et al. (2000), the new EEPM has set 2 hours of training as one of the benchmarks for meeting the Training Standard. Given the added beneficial effects observed for longer mentor training, the EEPM suggests a program enhancement goal of 6 hours of mentor training (MENTOR, 2009a). Although Sipe and Roder (1999) reported in their survey of 700 mentoring programs that 81% of programs required training, the amount of time required varied greatly across programs, from 15 minutes to 126 hours. The median amount of time required was 3 hours, and 49% of programs offered 2 hours or more of training. In a more recent survey of 131 BBBS agencies, the directors of program development were asked about the length of prematch training for communitybased programs (Wheeler & DuBois, 2009). The length of prematch volunteer orientation/training varied dramatically across programs: 34% provided 30–60 minutes, 34% provided 1–2 hours, 17% provided 2–4 hours, and 1% provided more than 4 hours. According to these surveys, many mentoring programs in the United States are not meeting the basic training length benchmark set by MENTOR

in the EEPM. Training Content Practices and Recommendations: A Review of Current Training Practices Best practice wisdom in training has been difficult to gather. Thus, we conducted a content analysis of 15 mentor training manuals from the United States and Canada in order to examine practices used in the training of prematch mentors. Training manuals were obtained directly from mentoring organizations as well as through internet searches. Consistent with the topics reviewed previously, each manual was read and coded for the following topics: cognitions of mentors, behavioral characteristics of successful mentors, roles that mentors should and should not play in the lives of mentees, the first meeting with the mentee, development of the mentoring relationship, ethical issues in youth mentoring, and closure of the match. In addition, the pedagogic method used to teach each topic or training modality was coded (e.g., role play, lecture, game, written document). Results of the content analysis can be seen in Table 30.2. Of the 15 training programs reviewed, all included some training on how to establish a positive, personal relationship with a mentee. Communication skills were especially emphasized, with most of the training programs including information about using active listening skills as a means of building trust within the relationship. Other popular lessons on relationship building included information about the stages of the mentoring relationship, the importance of being persistent in terms of maintaining contact with the mentee, and the critical role of being trustworthy. Several manuals specifically discussed the first meeting between mentor and mentee. Discussions of ethical responsibilities were also relatively popular. Issues dealing with confidentiality, relationship boundaries, and child safety were the most popular ethics topics. Somewhat surprisingly, match closure was not included in all of the training manuals. The topic of closure consisted of ending the relationship in a positive manner and what to talk about during the closure process. Notably, the content analysis revealed that most topics were presented in a list or narrative text format, with little to no use of interactive exercises. Training Format Practices and Recommendations In-Person Training and Resources. We identified two clearly research-based training programs. The first program, Across Ages (LoSciuto, Rajala, Townsend, & Taylor, 1996), derived its training from empirically

Mentor Training  451 Table 30.2   Number of Training Manuals Including Each Training Topic (N = 15) Training Topic Communication skills

Number 13

Stages of mentoring relationship

7

Persistent contact with mentee

8

Being trustworthy Initial meeting Ethics •• Confidentiality •• Boundaries •• Safety Closure •• Ending relationship in a positive way •• What to talk about during closure

validated strategies and placed considerable emphasis on sustaining relationships. This program has expanded to over 30 sites in 17 states; it reports relatively low volunteer attrition and match durations that greatly exceed national averages; and it focuses on behavioral, academic, and psychosocial outcomes. Another example of a training program that was derived from theory and based on research is the Mentor’s Training Guide developed for the Cross-Age Mentoring Program (CAMP) for Children with Adolescent Mentors (Karcher, 2012). This guide provides three formats for training, including a 2-hour orientation, a 2-day prematch training workshop, and several postmatch training workshops. Also offered is an alternative format that involves a 5-hour prematch training workshop followed by shorter postmatch booster sessions. The trainings, the handbook for mentors, and evaluation materials are all oriented toward promoting the same two programmatic goals of building perspective-taking skills and school connectedness. Web-Based Training and Resources. Two readily available web-based mentor training programs were located in an internet search focusing on courses using combinations of search words including youth, mentor training, program, course, webbased, and online. MENTOR has web-based training available on their website, www.mentoring.org, under the

10 6 10 9 8 6 12 10 6

Find Resources tab in the E-mentoring Clearinghouse under E-Mentor Training. The format of the course is similar to a slide presentation, and it does not include rich media (e.g., audio, video) nor allow interactivity. The content covers six main topics related to becoming a mentor and the stages of the mentoring lifecycle. The training is grounded in practice wisdom and is consistent with research findings, but it doesn’t reference research findings. The course includes a downloadable workbook as well as a quiz at the conclusion of the training. No evaluation of the course was located. Big Brothers Big Sisters of America (BBBSA) commissioned Michael Karcher to create a 1-hour web-based training website to provide training to their High School Bigs. The website includes nine text-based lessons, each based on research on mentoring, in general, on the High School Bigs program, or on CAMP (Karcher, 2008). The training materials include an opening page on which participants complete Crandall’s social interest scale. On the second web page, the trainee’s score is generated and, based on prior research using the scale (Karcher & Lindwall, 2003), the trainee is encouraged to consider how other teens with similar scores have excelled and struggled. Subsequent web pages address the importance of staying with the mentee, making sure the mentee feels important and special, and engaging in a playful and encouraging

452   PRACTICE AND PROGRAMMATIC CONSIDERATIONS manner with mentees. Based on prior research by Karcher (2005), the final section of the training emphasizes the importance of terminations and effective closure. BBBSA has allowed other mentoring programs to have unrestricted access to the site. To date, over 1,700 youth have completed the training. A link to this site is located permanently at the CAMP website, www.crossagepeer mentoring.com. Finally, a web-based mentor training course, Building the Foundation, developed by Janis Kupersmidt, Jean Rhodes, and Rebecca Stelter contains 12 multimedia, interactive lessons aligned with the research on mentoring and the EEPM. The topics include an introduction to mentoring; common expectations and motivations of mentors; key behavioral characteristics of successful mentors; roles that mentors should and should not play; the importance of having fun in the mentoring relationship; initiation of the relationship; ethics, safety, and boundary issues; growth and maintenance of the relationship; ways of approaching mentoring relationships; setting goals and collaborative planning of activities; and closure or relationship redefinition. The course includes resources useful for mentors and agency staff. For example, lessons include many downloadable handouts with tips of information and skills to remember and research summaries that justify recommendations. Journal pages are interspersed throughout the lessons to encourage reflection and track attitudes. Quiz questions and a final test are included to assess knowledge acquisition. Agency administrators have access to a dashboard in the web software application that allows for enrollment and tracking of mentors’ progress in the course as well as downloading of individual or aggregate reports and data. The course, hosted by Mentoring Central at www.mentoringcentral.net, has been examined in a small feasibility study of 23 mentors and 22 program staff from a wide range of programs and locations across the United States. Almost all (95%) of the mentor participants were current first-time mentors; one mentor had previously mentored but was about to be matched with a new child. The average length of time that these participants had been in a mentoring relationship with a child was about 2 years. A little more than half of the mentor participants reported having received mentor training previously. Participants, on average, provided very positive ratings of the course across a variety of consumer satisfaction variables. In fact, one case manager wrote: “I think that this is awesome. It is a universal training module. This allows grassroots mentor programs to provide effective training for its

mentors and mentees.” A match coordinator also said, “I did find the course useful and interesting. I think it is a great way for volunteers to complete their training.” Kupersmidt, Rhodes, and Stelter have been working with the Learning and Development Department at the National Office of BBBSA to create customized pre- and postmatch, web-based training courses based on the Building the Foundation course. The customized course, Making You the Best Big Possible: Starting Something Big, is currently being piloted to evaluate consumer satisfaction with the web-based training format and content as well as to determine how to integrate it into their service delivery model. Based on surveys from 1,008 mentors, 98% reported that they were satisfied or very satisfied with the course. Most important, across 10 areas of preparedness (e.g., ready to have an impact on a Little, having realistic expectations about the match and my Little, properly defining boundaries, selecting ageappropriate match activities, knowing who and when to contact agency staff), an average of 84% reported feeling more or much more ready. At least 70% of respondents reported that the course answered most of their questions and provided information they did not know they needed in order to be a mentor. In addition, 76% reported that they planned to use the tools provided when they were matched with a mentee. In terms of usability, an average of 92% reported the program was good or excellent in ease of use, convenience, quality of media, and appeal. In terms of feasibility, only 5% reported preferring in-person training to online training. An average of 99% of prospective mentors rated the content as good or excellent in terms of thoroughness, amount, accuracy, and topic appropriateness. Finally, 94% rated the online training course as good or excellent for being motivational and inspiring and 97% reported being excited to begin mentoring. These promising findings provide support from the practice world for the usability, feasibility, and potential costsavings associated with a well-designed webbased training course. Checklist of Recommendations Table 30.3 lists recommendations for providing high-quality mentor training, based on the theory and research reviewed in this chapter. These recommendations were also developed in alignment with the standards, benchmarks, and justifications published in the new Elements of Effective Practice for Mentoring.

Mentor Training  453 Table 30.3   Guiding Questions for Practitioners Is the prematch training at least 2 hours in length or greater (ideally at least 6 hours in length)? Does some prematch training take place in person? Is the in-person training interactive? Does the training address the following content areas? a. b. c. d. e. f. g. h. i.

Program rules Mentors’ goals and expectations for the mentor-mentee relationship Mentors’ obligations Roles that mentors should and should not play The first meeting with the child Relationship development and maintenance Ethical issues that may arise related to the mentoring relationship Effective closure of the mentoring relationship Sources of assistance available to support mentors, especially the importance of maintaining close communication with program staff

Does the program use evidence-based materials? Does the program provide additional prematch training opportunities beyond the 2-hour, in-person minimum? Does the program address other relevant topics in the training? a. b. c. d. e. f.

Youth development process Cultural issues Gender issues Issues of poverty and social class Opportunities associated with mentoring specific populations of children Challenges associated with mentoring specific populations of children

Does the program use training to continue to screen mentors for suitability and develop techniques for early troubleshooting should problems be identified? Does the program offer postmatch training opportunities? Does the program offer postmatch support to mentors? Source: These recommendations are based in part on the Elements of Effective Practice for Mentoring (MENTOR, 2009a).

Conclusion This chapter focused entirely on the training and support of mentors; however, training of mentees as well as the parents or guardians of mentees is also important for the development and maintenance of the mentoring relationship. A similar, but reduced, set of topics including program orientation and requirements, expectations, motivation, key behavioral characteristics, roles and responsibilities, and communication skills could form the basis for training of these individuals. Training of mentees, in particular, would need to be developmentally

appropriate and culturally relevant, and is greatly needed. The cornerstone, however, of producing both longevity and quality in the mentoring relationship is mentor training, and as this chapter illustrates, sufficient resources are available, in text and web formats, for all programs to provide thorough training at minimal additional cost. What remains most needed in the field, now, appears to be research on the effectiveness of different training content and formats in facilitating longer and stronger matches to produce more effective mentoring relationships and, ultimately, more positive outcomes for youth.

454   PRACTICE AND PROGRAMMATIC CONSIDERATIONS

References Adalist-Estrin, A. (2006). Providing support to adolescent children with incarcerated parents. The Prevention Researcher, 13, 7–10. Allen, J. P., McElhaney, K. B., Land, D. J., Kuperminc, G. P., Moore, C. W., O’Beirne-Kelly, H., & Kilmer, S. L. (2003). A secure base in adolescence: Markers of attachment security in the mother-adolescent relationship. Child Development, 74(1), 292–307. Beam, M. R., Gil-Rivas, V., Greenberger, E., & Chen, C. (2002). Adolescent problem behaviour and depressed mood: Risk and protection within and across social contexts. Journal of Youth and Adolescence, 31, 343–357. Crosnoe, R., & Elder, G. H., Jr. (2004). Family dynamics, supportive relationships, and educational resilience during adolescence. Journal of Family Issues 25, 571–602. Csikszentmihalyi, M., & Rathunde, K. (1998). The development of person: An experiential perspective on the ontogenesis of psychological complexity. In W. Damon (Series Ed.) & R. M. Lerner (Vol. Ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (5th ed., pp. 635–684). New York: Wiley. Dallos, R., & Comley-Ross, P. (2005). Young people’s experience of mentoring: Building trust and attachments. Clinical Child Psychology & Psychiatry, 10, 369–383. Darling, N., Hamilton, S. F., & Shaver, K. H. (2003). Relationships outside the family: Unrelated adults. In G. R. Adams & M. D. Berzonsky (Eds.), Blackwell Handbook of Adolescents (pp. 349–370). Malden, MA: Blackwell Publishing. Davidson, W. S., & Redner, R. (1988). The prevention of juvenile delinquency: Diversion from the juvenile justice system. In R. H. Price, E. L. Cowen, R. P. Lorion, & E. J. Ramos-McKay (Eds.), Fourteen ounces of prevention: Theory, research, and prevention (pp. 123–137). New York: Pergamon Press. Davidson, W. S., Redner, R., Amdur, R., & Mitchell, C. M. (1990). Alternative treatments for troubled youth: The case of diversion from the justice system. New York: Plenum Press. Denham, S., & Kochanoff, A. T. (2002). Parental contributions to preschoolers’ understanding of emotion. Marriage & Family Review, 34, 311–343. Dishion, T. J., Spracklen, K. M., Andrews, D. W., & Patterson, G. R. (1996). Deviancy training in male adolescent friendships. Behavior Therapy, 27, 373–390. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs: A meta-analytical review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Neville, H. A. (1997). Youth mentoring: Investigation of relationship characteristics and perceived benefits. Journal of Community Psychology, 25, 227–234. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. In

G. G. Noam (Ed. in chief) & J. E. Rhodes (Ed.), A critical view of youth mentoring (New Directions for Youth Development: Theory, Research, and Practice, No. 93, pp. 21–57). San Francisco: Jossey-Bass. DuBois, D. L., & Silverthorn, N. (2005). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524. Durlak, J. A., & Weissberg, R. P. (2007). The impact of after-school programs that promote personal and social skills. Chicago: Collaborative for Academic, Social, and Emotional Learning. Dworkin, J. B., Larson, R., & Hansen, D. (2003). Adolescents’ accounts of growth experiences in youth activities. Journal of Youth and Adolescence, 32, 17–26. Fisher, C. B. (2003). Decoding the ethics code: A practical guide for psychologists. Thousand Oaks, CA: Sage. Furman, W., & Buhrmester, D. (1985). Children’s perceptions of the personal relationships in their social networks. Developmental Psychology, 21, 1016–1024. Gehrke, G. E., Jenkins, Q. D. H., Miskovetz, S. A., & Wray, P. F. (2006). Why mentor?: A qualitative study of men’s and women’s reasons for becoming mentors. Praxis: Where Reflection & Practice Meet, 6I, 25–34. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring programs. American Journal of Community Psycho­ logy, 3, 199–219. Hamilton, M. A., & Hamilton, S. F. (2005). Work and service-learning. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 348–363). Thousand Oaks, CA: Sage. Hartje, J. A., Evans, W. P., Killian, E. S., & Brown, R. (2008). Youth worker characteristics and self-reported competency as predictors of intent to continue working with youth. Child Youth Care Forum, 37, 27–41. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Herrera, C., Sipe, C. L., & McClanahan, W. S. (2000). Mentoring school-age children: Relationship development in community-based and school-based programs. Philadelphia: Public/Private Ventures. (Published in collaboration with MENTOR, Alexandria, VA.) Hirsch, B. J. (2005). A place to call home: After-school programs for urban youth. Washington, DC: American Psychological Association; Teachers College Press. Joyce, A., Wolfaardt, U., Sribney, C., & Aylwin, A. S. (2006). Psychotherapy research at the start of the 21st Century: The persistence of the art versus science controversy. Canadian Journal of Psychiatry, 51, 797–809. Kalbfleisch, P. J. (2002). Communicating in mentoring relationships: A theory for enactment. Communication Theory, 12, 63–69. Karcher, M. J. (2005). The effects of school-based developmental mentoring and mentors’ attendance on mentees’ self-esteem, behavior, and connectedness. Psychology in the Schools, 42, 65–77.

Mentor Training  455 Karcher, M. J. (2007). Cross-age peer mentoring. Youth Mentoring: Research in Action, 1(7), 3–17. Karcher, M. J. (2008). The Cross-Age Mentoring Program (CAMP): A developmental intervention for promoting students’ connectedness across grade levels. Professional School Counseling, 12(2), 137–143. Karcher, M. J. (2012). The Cross-Age Mentoring Program (CAMP) for Children with Adolescent Mentors: Program manual. San Antonio, TX: Developmental Press. Karcher, M. J., Kuperminc, G. P., Portwood, S. G., Sipe, C. L., & Taylor, A. S. (2006). Mentoring programs: A framework to inform program development, research, and evaluation. American Journal of Community Psycho­ logy, 34, 709–725. Karcher, M. J., & Lindwall, J. (2003). Social interest, connectedness, and challenging experiences. What makes high school mentors persist? Journal of Individual Psychology, 59, 293–315. Karcher, M. J., Nakkula, M. J., & Harris, J. T. (2005). Developmental mentoring match characteristics: Correspondence between mentors’ and mentees’ assessments of relationship quality. Journal of Primary Prevention, 26, 93–110. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Langhout, R. D., Rhodes, J. E., & Osborne, L. N. (2004). An exploratory study of youth mentoring in an urban context: Adolescents’ perceptions of relationship styles. Journal of Youth and Adolescence, 33, 293–306. Levinson, D. J. (1979). The seasons of a man’s life. New York: Ballantine. Liang, B., Spencer, R., Brogan, D., & Corral, M. (2008). Mentoring relationships from early adolescence through emerging adulthood: A qualitative analysis. Journal of Vocational Behavior, 72, 168–182. LoSciuto, L., Rajala, A. K., Townsend, T. T., & Taylor, A. S. (1996). An outcome evaluation of Across Ages. Journal of Adolescent Research, 11, 116–129. Meissen, G. J., & Lounsbury, J. W. (1981). A comparison of expectations of volunteers, children, and parents in a Big Brother Big Sister program. Journal of Community Psychology, 9, 250–256. MENTOR. (2009a). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. MENTOR. (2009b). Mentoring immigrants & refugee youth: A toolkit for program coordinators. Alexandria, VA: Author. Morrow, K. V., & Styles, M. B. (1995). Building relationships with youth in program settings: A study of Big Brothers/ Big Sisters. Philadelphia: Public/Private Ventures. Murphy, S. E., & Ensher, E. A. (2006). Lessons from research on prosocial organizations for mobilizing adults for positive youth development. In E. G. Clary & J. E. Rhodes (Eds.), Mobilizing adults for positive youth development: Strategies for closing the gap between beliefs and behaviors (Search Institute Series on Developmentally Attentive Community and Society) (pp. 117–135). New York: Springer.

Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. Pianta, R. C. (1999). Enhancing relationships between children and teachers. Washington, DC: American Psychological Association. Pleiss, M. K., & Feldhusen, J. F. (1995). Mentors, role models, and heroes in the lives of gifted children. Educational Psychologist, 30, 159–169. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E., Chan, C., & Lowe, S. R. (2010). Pathways of influence in youth mentoring: Understanding the role of gender. Manuscript under review. Rhodes, J. E., & DuBois, D. L. (2006). Social policy report: Giving child and youth development knowledge away. Society for Research in Child Development, 10, 3–19. Rhodes, J. E., Liang, B., & Spencer, R. (2009). First do no harm: Ethical principles for youth mentoring relationships. Professional Psychology: Research and Practice, 40, 452–458. Rhodes, J. E., Reddy, J., Roffman, J., & Grossman, J. (2005). Promoting successful youth relationships: A preliminary screening questionnaire. Journal of Primary Prevention, 26, 147–167. Sipe, C. L. (2002). Mentoring programs for adolescents: A research summary. Journal of Adolescent Health, 31, 251–260. Sipe, C. L., & Roder, A. E. (1999). Mentoring school-age children: A classification of programs. Washington, DC: U.S. Department of Education, OERI. Spencer, R. (2006). Understanding the mentoring process between adolescents and adults. Youth Society, 37, 287–315. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Stukas, A. A., Daly, M., & Clary, E. G. (2006). Lessons from research on volunteering for mobilizing adults to volunteer for positive youth development. In E. G. Clary, & J. E. Rhodes (Eds.), Mobilizing adults for positive youth development: Strategies for closing the gap between beliefs and behaviors (pp. 65–82). New York: Springer Science + Business Media. Styles, M. B., & Morrow, K. V. (1992). Understanding how youth and elders form relationships: A study of four Linking Lifetimes programs. Philadelphia: Public/Private Ventures. Suarez-Orozco, C., & Suarez-Orozco, M. M. (2001). Children of immigration. Cambridge, MA: Harvard University Press. Vrasidas, C., & McIsaac, M. S. (2000). Principles of pedagogy and evaluation for web-based learning. Education Media International, 37, 105–111. Wheeler, M., & DuBois, D. (2009). Analysis of responses to agency practices survey for Big Brothers Big Sisters of America’s community-based mentoring program. Philadelphia: Big Brothers Big Sisters of America.

31  FAMILY INVOLVEMENT Andrea S. Taylor and Lorraine Porcellini

Introduction Early research on mentoring relationships suggests that the likelihood of positive and effective relationships may be greater when a youth’s mentor and parent(s) are better acquainted and share a similar understanding about mentoring and expectations regarding the match (Meissen & Lounsbury, 1981; Sipe, 2002). When a mentor has a strong relationship with a family and is part of a “team” that supports the child, the outcomes for the child are usually improved (Grossman & Rhodes, 2002) but not always (Cavell, Elledge, Malcolm, Faith, & Hughes, 2009). In an early meta-analysis of mentoring program evaluations, parental involvement was found to be one of the program practices associated with positive outcomes for youth (DuBois, Holloway, Valentine, & Cooper, 2002). A more recent metaanalysis, however, did not find it to be a significant moderator of program effects (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). This latter meta-analytic finding could have resulted from a general increase in the involvement of parents across multiple programs as a result of earlier findings revealing its importance, or it could be that these more recent program evaluations either did not track parental involvement or don’t consider it a formalized (i.e., reported) program characteristic. What is clear, however, is that there is much yet to learn about the role of parents. Because parents are a primary gatekeeper to the child, familial support of the mentoring relationship is critical and, therefore, it seems logical for programs to encourage family involvement in order to support effective, sustained relationships. Nevertheless, an essential tension that appears in the youth mentoring literature often focuses on the

perception that inadequate parenting may be the reason a child needs a mentor to begin with and also that nonsupportive parents may try, deliberately or unconsciously, to sabotage the mentor-youth relationship by not respecting appropriate boundaries between the mentor and youth (Miller, 2007; Styles & Morrow, 1992). Some researchers have suggested that it is preferable not to actively engage parents in the mentoring process in order to minimize the risk of disrupting the developing relationship (Miller, 2007; Philip, Shucksmith, & King, 2004). For example, when mentors overstep family boundaries, such as by sharing information about the child with the parent or vice versa (whether solicited by the youth and parent or not), children may express disappointment that the mentor’s attention had shifted away from them or is less private and become more reluctant to confide in the mentor in the future (Morrow & Styles, 1995; Taylor, LoSciuto, Fox, & Hilbert, 1999). The inclusion of family in the mentoring process may pose challenges. From a broader perspective, it also is important to consider the role that culture plays when thinking about family involvement. Youth from collectivist cultures, such as African American and Latino youth, may be more likely to have natural mentors who are immediate or extended family members (Rhodes, Contreras, & Manglesdorf, 1994; Sánchez & Reyes, 1999), suggesting the importance of understanding the family member’s role as mentor to the child. In summary, to date the benefits of parental involvement in youth mentoring are not clear, nor are there accepted best practices for involving family to strengthen mentoring relationships. The starting point for this chapter on involving families is that, regardless of whether a child is in a 457

458   PRACTICE AND PROGRAMMATIC CONSIDERATIONS mentoring relationship, parental involvement, encouragement, and consistency in communication are essential for positive child development (Harris & Wimer, 2004; Henderson & Mapp, 2002). If we use this as a beginning point, assuming that effective parenting is key to fostering healthy child development, then a logical corollary is that parental interaction with other stakeholders in their children’s lives can increase all adults’ positive influence on youth, and that in some cases more effective family functioning might result from a child’s participation in a mentoring program. Indeed, this is a positive outcome worth pursuit in its own regard. Stemming from these assumptions, several key questions about the possible consequences of family involvement arise. The first question we consider is whether or not the presence of a mentor can jumpstart or facilitate higher-quality family experiences and communication between parents and children. Second, in what ways might involving the parents in their children’s mentoring relationship increase the quality of the children’s mentoring experience, yield benefits within the home or school, or both? Benefits could be to either the family or the youth, and could happen through increased parental engagement in the children’s lives through increased monitoring, more effective parent-child communication, and/or improved parenting as a result of lessons learned from the program staff or mentor. A third important question addresses a possible confound or spurious relationship between family involvement and youth outcomes from mentoring—namely, are families that typically provide high-quality “family environments” the same families who tend to become involved in their children’s matches such that any positive effects of parental involvement in mentoring programs may more likely be a proxy for the presence of a higherquality “family environment” in the lives of those mentees to begin with? The theoretical and empirical evidence cited in this chapter point to three types of program models that engage and involve families. Assessing the impacts of family involvement resulting from each model may help answer such “chicken-and-egg” questions about what strategic, program-prescribed family involvement adds to youth mentoring, compared to programs conducted without such strategic family involvement. These three strategic approaches are as follows: •• Youth and family mentoring (programs that target the entire family and provide mentoring, family support, and access to community resources)

•• Youth-only mentoring + skill building for families (programs that provide mentoring and support to an identified youth but also engage family members in strategies to strengthen parents’ skills in working with their children) •• Youth-only mentoring + family activities (programs that provide youth with one-onone mentoring and engage families in some of the program activities) Specific program practices and activities associated with each of these models are described in this chapter, and a set of best practices is proposed based on the research. To date, little has been written about how best to facilitate family involvement in formal youth mentoring relationships and what is considered best practices in parental involvement. This chapter attempts to organize and critique such efforts as a first step toward the identification of useful approaches to family involvement in mentoring. The first edition of the Handbook of Youth Mentoring omitted a chapter on this topic because at the time there was insufficient research to support such a chapter. The available research remains quite limited. The source material for this chapter was drawn from systematic reviews of a number of peerreviewed journals in the field of education, social work practice, primary prevention, and child/adolescent development. Key search terms included family involvement in mentoring, engaging families, best practices in family involvement and parental support. Searching the education, social work, prevention, and child development literatures using these keywords revealed several theories, programs, and specific research findings that we describe in the next sections. Although the existing literature and research shed some light on the value of engaging families and the caveats of doing so, there is still considerable need for more research, especially in the arena of family involvement in youth-only mentoring. Therefore, following these summaries of available research, we turn to the issue of best practices and are in the unfortunate but exciting circumstance of having to propose, albeit tentatively, a core set of practices for consideration by future researchers and practitioners alike.

Theory More knowledgeable and engaged families produce better outcomes for youth. Theories about why family involvement in youth programming generally

Family Involvement  459 and mentoring specifically appear to yield significant benefits to youth yield three approaches or explanations for the importance of the role of mentoring programs: (a) it can strengthen the family’s interaction and skills, (b) it can foster parent-child connectedness by increasing the mentor’s awareness of the nature of the family’s interactions, stressors, and ecological demands and how to respond to them with the mentee, and (c) it may encourage parent/caregiver support for the mentoring relationship and foster the mentor’s ability to effectively interact and collaborate with the parents to create a “team.” Three theories have been put forth to explain the mechanisms at play in each of these approaches. Parental Acceptance Theory Parental acceptance theory (Rohner & Britner, 2002) provides a framework for understanding why parental support for the mentoring relationship contributes to efficacious outcomes for youth. The theory suggests that parental acceptance is associated with various positive outcomes, including prosocial behavior in children and the capacity for effective relationships in adolescence, because, drawing from the literature on natural mentoring, a strong attachment to parents (Bowlby, 1969) may increase a youth’s capacity to seek out and solicit social support from other adults, including relationships with nonrelated adults (Barrera & Bonds, 2005; Soucy & Larose, 2000). Parental acceptance is key to the youth’s willingness to seek support from adults outside the family. The development of a stronger mentoring relationship may be fostered by parents who convey acceptance to their children, encourage their children’s expectations of success outside the family, and facilitate the child’s receptivity to the mentor (Noam & Malti, this volume, Chapter 7; Zimmerman, Bingenheimer, & Behrendt, 2005). Family Systems Theory Theories that support the first two types of programs we describe, youth and family mentoring and youth-only mentoring + skill building for families typically focus on a systems approach. Family systems theory suggests that the family is an emotional unit in which family members are interdependent and function in reciprocal relationships with one another. Therefore, the functioning of one member cannot be understood completely if taken out of the context of the functioning of the people closely involved (Kerr & Bowen, 1988). As suggested by Bronfenbrenner’s (1979) ecological systems theory,

when youth live in a challenged family or community environment, the positive effects of mentoring alone may be diluted when they occur in isolation from the child’s larger ecology. Understanding the family system from a cultural perspective reveals an important variable in the role of families and family involvement in mentoring. In collectivist cultures, the needs and perspectives of the group take precedence over the individual (Marin & Marin, 1991). Family members are seen as having an obligation to one another such that the upbringing of the children is often the responsibility of multiple people, including immediate and extended family members (Sánchez & Colón, 2005; Sánchez et al., this volume, Chapter 10). These theories provide support for the program practice of ensuring that mentors are made aware of how key culturally based factors in the child’s relationships might influence the match, insight that may foster the mentor’s capacity to engage more effectively with both the child and the family. Systemic Mentoring Theory Another theory, which emerges from systems theory, suggests that the overall effect of the mentoring intervention may result from a cohesive alliance of multiple caring adults (mentor, parent/ guardian(s), and case manager) who collectively support the child and his or her development (Keller, 2005). This theory proposes that the success of the mentoring relationship is related to the “pattern and content of interactions that occur within and across the relationship subsystems represented by the model” (p.172; see also Keller & Blakeslee, this volume, Chapter 9), not just to the mentor’s or program staff’s awareness of the larger social ecology in which the youth lives. To date, this model has not been tested empirically, yet it would appear likely that mentees would benefit significantly more from matches in which parents are knowledgeable and supportive of the mentor’s role and where there is mutual understanding and reciprocal communication between the mentee and the significant adults involved in the match than from matches in which the mentor has only been informed about the others in the mentee’s life. These theories suggest that there are positive outcomes for youth when programs enhance parents’ skills in engaging with their children (e.g., parenting) and where communication exists (and is solicited) between families and mentors/staff. They differ, however, in the mechanisms by which family involvement in mentoring relationships affects the youth.

460   PRACTICE AND PROGRAMMATIC CONSIDERATIONS

Research The research on family involvement in youth mentoring is slim, and with the exception of a recent qualitative investigation that used interview data from parents themselves (Spencer, BasualdoDelmonico, & Lewis, 2011), parents’ roles in the mentoring relationship have been understood primarily from the perspective of agency staff, mentors, and youth (Morrow & Styles, 1995; Styles & Morrow, 1992). The results of the Spencer et al. (2011) investigation suggest that efforts to keep parents at a distance may not be in the best interest of the mentor-youth pair, and that understanding what types of parental involvement, and under what circumstances, may help to facilitate sustained, effective relationships. Spencer et al. (2011) interviewed 13 parents, whose children were enrolled in a community mentoring program, for the purpose of exploring the parents’ understanding of the mentoring process, their expectations, and their role in the development of the youth-mentor relationship. In some cases, parents reported that the parent and mentor worked together to facilitate interactions and address issues that arose for the youth. In other cases, the parent described playing a more protective role, such as when the mentor failed to follow through. Overall, the parents in this study felt better about the mentor-youth relationship if they had a connection with the mentor and seemed to be more anxious and uncertain if they did not, suggesting that Keller’s (2005) systemic approach of a united front may be preferable to parents (compared to the other theoretically prescribed approaches that help either parents or mentors directly but separately). Consistent with our characterization of the three theories described earlier, the existing research on parent involvement can be organized according to three specific strategies: youth and family mentoring, youth-only mentoring + skill building for families, and youth-only mentoring + family activities. We therefore organize the research accordingly. Youth and Family Mentoring Family mentoring is not always focused directly on a given youth but is intended to strengthen the entire family system, often with the goal of helping a family cope more effectively with external issues such as economic fluctuations or community conditions (Hess, Barr, & Hunt, 2009). These interventions attempt to help families change both external conditions (such as being unemployed or residing in an unsafe neighborhood) and the way family members relate to these conditions. They appear to have

at least two essential elements in common: (a) mentoring by a well-trained adult volunteer and (b) professional case management. Other components may include group activities for families, including skill building, advocacy, and access to resources. Advocacy as a mentoring practice has emerged recently as one of a select group of program moderators associated with significantly larger programmatic effects (DuBois et al., 2011). Although empirical data are available that focus on advocacy models for families facing specific crisis situations (as cited in Hess et al., 2009), perhaps the only study focusing more specifically on youth and parent outcomes is an evaluation of the Family Mentoring Program (BarronMcKeagney, Woody, & D’Souza, 2001). The program was developed in response to increased violent criminal activity, low educational attainment, and poverty among immigrant Latinos living in South Omaha, Nebraska. In this study, the program provided mentoring for 49 10-year-old Latino children and included a family mentoring component consisting of family assessments, counseling, education, and participation in community activities. The program used mentors who were recruited through neighborhood churches and from local colleges and universities. Families were linked to a range of educational, governmental, mental health, substance abuse, and medical services. Barron-McKeagney, Woody, and D’Souza’s (2002) evaluation of the program examined its impact on two main outcomes, the mothers’ perceptions of the parent-child relationship and of family strength as measured by the Parent-Child Relationship Inventory (PCRI) and the Family Hardiness Index (FHI). Their analysis focused on three factors expected to affect the outcomes (mothers’ participation in the parent group education programs, children’s levels of involvement in the program, and children’s beginning levels of problematic behaviors). Stepwise linear regression results indicated that only the number of hours spent in parent education explained a statistically significant proportion of variability in outcomes. Hours of parent education explained approximately 30% of the variability in scores on two of the dependent measures—support (adjusted R2 = .285) and communication (adjusted R2 = .297). T-test results revealed that, compared to the standardized samples used in the development of the outcome measures, mothers, at the beginning of the program, scored significantly lower on six of seven PCRI subscales and on the FHI. By the end of the program, all of the mothers’ scores had improved and the scores on the “support” and “limit setting” measures no longer differed significantly from the standardized norms. Despite the absence of

Family Involvement  461 an equivalent comparison group, the overall findings suggest that parents of the children in this mentoring program needed and could benefit from both services and parent education. However, the study provided no information or explanation on how such parent education may moderate youth outcomes. Youth-Only Mentoring + Skill Building for Families Programs that use youth-only mentoring and skill building for families are often intended to address an identified academic or behavioral issue among targeted youth and typically involve the child in a one-on-one mentoring relationship and the parent in a tandem specific curriculum or workshop series intended to enhance their knowledge and skills in building an effective relationship with their child. The programs described in this section involved different settings (high adversity schools/communities) as well as types and intensity of family engagement. In general, the studies focused on outcomes for youth, which tell us very little about the effect of practices specifically for families on family functioning. Two of these programs are Project START, Support to Affirm Rising Talents (Moon & Callahan, 2001) and Early Risers “Skills for Success” (August, Hektner, Egan, Realmuto, & Bloomquist, 2002). Project START was designed to investigate the impact of mentoring, parental involvement, and multicultural curricula on student achievement and students’ likelihood of being accepted into talented/gifted programs. A total of 120 children were randomly selected and assigned to one of three conditions: (a) modification of classroom activities + a family outreach program, (b) modification of classroom activities + a family outreach program + mentoring, or (c) a comparison group. No statistically significant effects were found on any of the measures. However, students who received curricula modifications and participated in family outreach, whether or not they had a mentor, were referred and placed more often into the district gifted and talented program than students who did not receive these benefits (31% of the treatment groups versus 12% of the control group for grades 2 and 3 combined). Though this finding lends support to the positive effect of curricula modifications and family outreach, the independent contribution of family outreach was not investigated. Early Risers “Skills for Success” was a comprehensive 5-year program that included parent education and skills training as well as a mentoring component that targeted aggressive behavior in elementary-school-age children. The program featured two complementary components, CORE (which

promoted the child’s behavioral self-regulation, social competence, and academic achievement domains) and FLEX (which promoted family competence through family support, education, and empowerment). An evaluation was conducted in the 3rd year of the program (August et al., 2002) in 20 demographically comparable elementary schools in two semirural midwestern areas characterized primarily by Caucasian families of low to low-middle socioeconomic status. Half the schools were assigned to the intervention group and half to the control group. In the sample of 245 children whose parents agreed to participate, there were roughly twice as many boys as girls, and the mean age was 6.6 years. After 3 years in the program, 199 children, with equal numbers in each condition, remained in the program. Groups were comparable at baseline, and retained children did not differ from those who dropped out. The goal of the FLEX component of Early Risers was to reduce family stress factors that undermined parents’ ability to provide nurturance, support, and direction for their children and thereby to improve children’s academic, behavioral, and social competencies. The program had an effect on academic competence (the size of this effect, d, was 0.26) but not on behavioral competence. Increases in behavioral competence were reported only by parents of severely aggressive youth (d = .70). More specifically, parents with highly aggressive children reported more improvements in self-regulation, aggression, hyperactivity, and impulsivity than control families with similar children. The only effect on family functioning (parental distress), however, was confounded by dosage differences. Highly aggressive youth who received more than half of the program activities had parents who reported lower distress after the program than aggressive intervention youth whose parents attended less than half of the activities (d = .56). Regarding the proposition made in the theory section, that family involvement would be lowest in families with struggling youth (and conversely “more parent involvement” in mentoring would correlate with better outcomes for youth not because of family involvement in the program but because of higher family involvement in all domains of the child’s life), the authors reported “parents of the most severe children showed the lowest levels of investment in their children” (p. 621); only parents who attended more than half of the sessions reported improved parenting practices. PrimeTime and Lunch Buddy. Hughes and colleagues (Hughes, Cavell, Meehan, Zhang, & Collie,

462   PRACTICE AND PROGRAMMATIC CONSIDERATIONS 2005) compared PrimeTime to the Lunch Buddy program. PrimeTime is a multicomponent intervention that combined community-based mentoring with consultation for parents and teachers and problem-solving skills training for children. Lunch Buddy involved mentoring alone, occurring weekly during lunch with a different college student mentor each of the three semesters. In the comparison, 2ndand 3rd-grade children (N = 174) in 13 schools with a high percentage of children of low socioeconomic status, who met teacher- and peer-rated criteria for heightened aggression, were randomly assigned to each program. Both programs spanned three academic semesters, relied on college student mentors, and used course credit and class grades as contingencies to ensure consistent visits and to eliminate the risk of early mentor termination. In an evaluation comparing the two programs, Hughes, Cavell, and Mehan (2001) found no differential treatment effects immediately following the intervention; however, differences were found a year later, and interestingly, these effects favored the lessintensive Lunch Buddy program. Subsequent analyses involving children who stayed in the same school throughout the intervention period (n = 86) revealed that school context significantly moderated treatment outcome (Hughes et al., 2005). A hierarchical linear regression analysis revealed a significant interaction on aggression level between treatment condition and school adversity, F(1,85) = 5.72, p = .04, effect size correlation = .27. Specifically, Lunch Buddy mentoring was particularly effective for aggressive children in schools marked by high levels of playground aggression, economic disadvantage, and family mobility. In contrast, PrimeTime was more effective in schools with low adversity. One interpretation of these findings is that mentoring plus parent interventions may be less effective when youth are aggressive, poor, and highly mobile. This finding suggests that for some parents and youth, parent involvement may be less helpful. Youth-Only Mentoring + Family Activities In programs that involve youth-only mentoring and family activities, youth are matched with mentors and parents are invited to participate in occasional activities. Studies of these programs typically do not target or measure behavioral changes in the parents. A study of Big Brothers Big Sisters (Wheeler & DuBois, 2009) was designed to examine agency practices in relation to performance indicators for community-based mentoring programs, including match quality and duration, 6- and 12-month retention

rates, and 3-month child and volunteer strength of relationship scores. Responses on a three-item measure of parental engagement practices—(a) parent input in matching process, (b) inviting parents to agency events, and (c) presence of a parent/family liaison on staff—were found to correlate positively with strength of relationship scores at 3 months into the match. Negative associations with other performance measures included inviting parents to waitlist activities and providing parents with specific activities/recommendations on how they can support the matches. That is, engaging parents in those practices was associated with poorer outcomes. The Cross-Age Mentoring Program (CAMP) is a peer-mentoring program in schools in which teenage mentors meet regularly with a child mentee, includes a weekend component for parental involvement in which the parents participate in activities with their children and their children’s mentors (Karcher, 2008, 2012). These events, called Super Saturdays, occur three to five times each year. In the first study of the program, Karcher, Davis, and Powell (2002) found that program impacts on spelling achievement were mediated by improvements in mentees’ connectedness to parents. This finding implies that children in the program became more connected to their parents (communicated more, valued parents more) as a function of program involvement and that gains in connectedness to parents likely facilitated academic achievement gains. Evidence that program participation led to increased connectedness was revealed by parent reports that, in addition to the time they spent interacting with their children’s mentors at the Super Saturdays, the time they spent in transit (taking their children to and from the program, or waiting for the bus to pick up their children) allowed parents an opportunity to discuss with their children the program and their experience with the mentors. As has been found from community-based mentoring (Grossman & Tierney, 1998), these reflections may have served to both reinforce the curricular content of the mentormentee interactions as well as help foster a stronger bond between parent and child through close conversations. Such interactions may explain why participation in the school-based CAMP program had large and statistically significant association with improvement in youths’ connectedness to parents (eta2 = .22) and why these changes fully mediated or explained the improvement in academic achievement demonstrated by program youth compared to youth in the control group (Karcher et al., 2002). Across Ages is a comprehensive, intergenerational mentoring program designed to reduce adolescent drug use among 9–13 year olds (Taylor, 2000).

Family Involvement  463 An evaluation of Across Ages (LoSciuto, Rajala, Townsend, & Taylor, 1996) involved youth assigned to one of three conditions: family involvement, life skills training, and community service (PS); family involvement, life skills training, and community service plus mentoring (MPS); or a comparison group (C). The original sample (N = 538) comprised urban African American, Latino, Asian, and Caucasian youth, aged 10–13, living in Philadelphia, Pennsylvania. Family involvement consisted of monthly social and recreational activities such as bowling, youth-run talent shows, attendance at sporting or theater events, and potluck dinners. During these activities, program staff provided information about community resources and engaged the parents in learning effective ways of communicating with their children. Mentors frequently called their youths’ family members to encourage attendance and often followed up by offering to assist with transportation. Parent attendance was tracked for each event, and parents were interviewed regarding their knowledge/understanding of the program, satisfaction with their children’s participation, and perception of the program’s benefit (n = 331). Parents who were interviewed whose children were in the MPS full-treatment group (n = 176) more often expressed satisfaction with the program when their child’s mentor reached out as described. Mentors also reported increased satisfaction when they had access to the parents and were able to share the children’s successes and challenges with them. Families of mentored youth reported increased participation in school-related activities and engagement in social/recreational activities as a family, increased awareness of community resources, and more positive strategies for communicating with their children (LoSciuto et al., 1996). These findings seem to support Spencer et al.’s (2011) observation that the parent’s connection to the mentor enhances positive feelings about the relationship and the program. As with many of the studies cited, a limitation is that it did not attempt to connect outcomes for individual youth to the level of parent involvement. Summary and Recommendations Despite some evidence that parental involvement in mentoring boosts program impact (DuBois et al., 2002), and some programs deliberately include parents in the three ways described earlier, to our knowledge, no studies have been able to separate the possible effects of parental involvement from other program elements. Therefore, no hard evidence indicates that parental involvement

alone, in any of the three ways described earlier, is directly related to program outcomes. For example, DuBois et al.’s (2002) earlier meta-analysis compared programs with and without stated parental involvement components. Although the metaanalytic techniques for data analysis accounted for many moderators that might serve as rival explanations for the benefits of programs with versus those without parental involvement, there may be factors they did not measure that differentiate these programs and better explain the benefits of those programs with parental involvement. Stated more simply, programs that involve parents may do a host of other things that also bolster program impact that individually or collectively account for their relatively greater effect, compared to programs without parental involvement (and concomitantly, without the other components in the other programs). As noted earlier, DuBois et al.’s (2011) subsequent meta-analysis did not find that parental involvement moderated program impacts. Perhaps all programs evaluated in the past decade incorporated parental involvement, such that the presence or absence of parental involvement no longer differentiated high and low performing programs. But the reason for the absent effect of parental involvement in differentiating more and less effective programs in the meta-analysis is unclear. To be sure, more definitive and focused research on the best ways to involve families, and the ways in which the involvement facilitates change, is needed.

Practice A review of existing programs that incorporate family involvement in mentoring relationships identified three primary practices for doing so, although not all have empirical evidence of the value added from this component. The first practice, youth and family mentoring, includes mentoring by a volunteer and case management, often provided by a social worker. The programs most relevant to our exploration of the benefits for youth mentoring specifically, however, are those that provide mentoring for an identified youth (as opposed to mentoring an entire family) and include case management and access to resources and services for families. Although only the Family Mentoring Program (Barron-McKeagney et al., 2001) reports evidence of the overall efficacy of this model, the combination of one-on-one mentoring of youth (weekly meetings for the purpose of enjoying social activities and academic enrichment), plus information and access to resources for parents, appears beneficial for both youth and parents. For

464   PRACTICE AND PROGRAMMATIC CONSIDERATIONS many parents, however, knowledge about resources is only the first step. Accessing these resources means providing parents more support and guidance, requiring a “here is how you do it” approach. Mentors in Across Ages, for example, often helped parents negotiate public transit so that they could participate in family activities, or accompanied parents to meetings with teachers to facilitate parental involvement in making decisions for their child’s academic future (Taylor et al., 1999). The second practice involves youth mentoring plus intentional skill building opportunities for family members. Two types of program models fall under this descriptor: the first are those engaging adult family members in workshops that educate them about effective strategies for working with their children (e.g., school advocacy, listening and communication, conflict resolution, and strategies for avoiding risky behavior); the second are those that often follow a specific curriculum designed to address family support needs, such as teaching parents how to prevent aggressive behavior or other parenting skills. The Families and Mentors Involved in Learning With Youth (F.A.M.I.L.Y) project (Weinberger, 1992) involves the family with mentors in order to support the youth, whose interests they have in common. This approach is most consistent with Keller’s (2005) systemic model that attempts to strengthen the ties among key stakeholders in youth’s lives. Created by Susan Weinberger as a component of her school-based mentoring initiative in Norwalk, Connecticut, the F.A.M.I.L.Y project provides a series of workshops for mentors and families on topics such as stress management, AIDS prevention, communication skills, and conflict resolution. Initially the program began as a way to help parents understand what the mentor was trying to accomplish and to get them involved in similar strategies that would support the child’s success in school. These shared learning experiences have a twofold benefit, that of enhancing the skills of the significant adults in the child’s life and encouraging respect and understanding between the mentor and the family members. Over time, it became apparent to Weinberger that the families themselves were interested in continuing education and employment skills to meet their own needs. Consequently, skill-building opportunities for parents were an added component (Family Strengthening Policy Center, 2004; personal communication with S. Weinberger, 2010). Family skill-building programs are often focused on addressing a specific problem area, such as reducing aggression or preventing substance abuse. These models appear to utilize mentors in a

variety of ways. In some programs, mentors are paid staff who function as case managers and school-based family advocates (Early Risers), while other programs engage mentors as communitybased volunteers (PrimeTime and Lunch Buddies). Families are also involved in multiple ways, including attending curriculum-specific workshops that address issues such as parenting skills and conflict resolution and participating in family-focused support and empowerment groups. These programs also provide skill-building sessions for targeted youth on topics such as emotional regulation and social problem solving (August, Egan, Realmuto, & Hektner, 2003). The practice of offering one-on-one mentoring to youth while engaging families in occasional activities typically involves families in periodic weekend or evening activities, such as social or cultural events, that help family members better understand how the program works and the nature of the interaction between the child and the mentor. Family involvement of this type extends along a continuum from more to less intense, including program activities that are regularly scheduled and ongoing, to meetings restricted to pick-up and drop-off times. Across Ages and CAMP engage parents in similar ways: first, through a baseline orientation to the program and, second, in weekend family activities that are designed for fun and socialization and occasional but optional skill-building workshops. Table 31.1 provides information regarding resources specific to programs involving families in one of the three ways identified in this chapter. Table 31.2 provides some guidance, both anecdotal and empirical, for recommendations regarding likely-to-be-effective strategies for family involvement in mentoring programs. We organized Table 31.2 using the four topic areas that emerged repeatedly (when we reviewed the existing research and program practices) as strategies most likely to contribute to effective engagement and support for parents and families.

Conclusion The studies, programs, and practices described in this chapter reveal many ways that parents can be engaged in programs in which their children are involved and mentored. The data, both empirical and anecdotal, suggest the generally positive benefits of doing so, although almost no programs have isolated and tested specific types of family involvement, rendering it possible that other program components are responsible for overall program impacts.

Family Involvement  465 Table 31.1   Resources for Mentoring and Family Involvement Strategy

Resources

Youth and family mentoring

Family Strengthening Policy Center, National Human Services Assembly, Policy Brief No. 4: www.nationalassembly.org/fspc/documents/PolicyBrief/Brief4.pdf Harvard Family Research: www.hfrp.org/family-involvement/publicationsresources/focus-on-families!-how-to-build-and-support-family-centered-practicesin-after-school National Collaboration for Families: www.nassembly.org/fspc/Resources.aspx Solutions for America: www.solutionsforamerica.org/healthyfam/index.html Ohio Department of Education: www.ode.state.oh.us/GD/Templates/Pages/ODE/ ODEDetail.aspx?page=3&TopicRelationID=1828&ContentID=78604&Cont ent=94244

Youth-only mentoring + skill building for families

Early Risers Skill Building Program: www.psychiatry.umn.edu/research/ earlyrisers/home.html

Youth-only mentoring + family activities

Across Ages: www.acrossages.org

Lunch Buddy Mentor Manual: http://crav.uark.edu/PSP_Mentor_Manual_.pdf

Cross-Age Mentoring Program (CAMP): http://www.crossagepeermentoring.com National Mentoring Partnership: www.mentoring.org

Table 31.2   Checklist for Practitioners Topic

Recommendations

Understanding the program’s goals and mentor roles

Conduct intensive outreach to parents to engage them in orientation sessions and to address barriers that might prevent their participation (e.g., transportation, language, feelings of intimidation). Provide opportunities in orientation sessions and one-on-one conversations with program staff for parents and youth to articulate expectations for the mentoring relationship. Include, in an orientation session, a reflection for parents on a mentoring relationship they may have had as a way to identify and clarify mentor roles. Encourage communication between program staff and parents by providing intentional opportunities for dialogue.

Building trust between mentors and parents

Assist mentors and parents in appreciating their mutual interest in and responsibility to the youth. Train mentors in effective communication and provide specific strategies for gaining the parent’s confidence while respecting boundaries. Explore and address fears and challenges that mentors and parents may anticipate or experience in working together. Educate mentors in working with the families of special populations, including immigrant youth (for a resource, see MENTOR, 2009) and those with incarcerated parents (for suggestions, see Adalist-Estrin, 2006). (Continued)

466   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Table 31.2  (Continued) Topic

Recommendations

Enhancing parents’ skills and confidence

Provide opportunities for parents to participate in workshops that increase their knowledge and skills in areas such as parenting, communication, health topics, and resume writing. Conduct orientation for parents and mentors together to facilitate interaction and shared knowledge. Engage parents in training that will help them deal more effectively with their children’s challenging behavior. Provide home visiting coaching and counseling.

Connecting families to resources

Identify available community resources needed by families. Connect families to case managers or family counselors.

We do know that youth, especially those living in poverty and high-stress situations, appear to benefit when their parents understand the availability of community resources that enhance their quality of life (e.g., Barron-McKeagney et al., 2001) but do not benefit as much from comprehensive parenting interventions coupled with youth mentoring (Cavell et al., 2009). Similarly, youth challenged by behavioral issues may appear to benefit the most when their parents/caregivers have skills that promote positive parenting and healthy coping strategies relating to stressful life situations. The parents/caregivers of these young people may, however, be the least likely to make the investment in the program that would yield such benefits (August et al., 2003). Youth in community- or school-based mentoring programs also appear to benefit when parents/caregivers understand the goals of the mentoring programs in which they are involved, can support the activities that will help to achieve those goals, and have a relationship with their children’s mentors that affords opportunity for collaborative conversations (Spencer et al., 2011). Compelling evidence supporting one of the reviewed theories about how parental involvement affects families, parents, or youth relative to one another was not found, but this may simply reflect the neophyte status of the research on this topic. The challenge we face is that no studies have disentangled the findings on parental involvement from the myriad program activities that engage youth and their parents. It is still not clear exactly what constitutes “the right amount and the right kind” of parent involvement, nor have we yet developed a set of best practices that can guide programs in coaching mentors, supporting parents, and training staff to ensure the optimal contact necessary to build relationships between mentors and parents. Nor do we know what

amount and types of mentor-parent contact best respect boundaries, particularly with regard to youth-mentor confidentiality, for given families and types of youth. Missing from the literature is attention to issues affecting parental involvement for particularly challenged populations, such as refugee and immigrant youth (MENTOR, 2009), children of prisoners (Adalist-Estrin, 2006), and those from multiply stressed family situations. One of the major challenges with regard to future practice is that nearly all suggestions involve additional training of mentors and even more vigilant contact with parents and youth. Given the fact that most mentoring programs are already working with limited resources, it is hard to imagine how many staffs could easily add this additional layer of responsibility to their existing workload. Clearly, programs will need to be creative to find the resources and appropriate approaches for supporting greater family involvement in their programs.

References Adalist-Estrin, A. (2006). Providing support to adolescent children with incarcerated parents. Prevention Researcher, 13(2), 7–10. August, G., Egan, E., Realmuto, G., & Hektner, J. (2003). Four years of the Early Risers early-age targeted preventive intervention: Effects on aggressive children’s peer relations. Behavior Therapy, 34, 453–470. August, G., Hektner, J., Egan, E., Realmuto, G., & Bloomquist, M. (2002). The Early Risers Longitudinal Prevention Trial: Examination of 3-year outcomes in aggressive children with intentto-treat and as-intended analyses. Psychology of Addictive Behaviors, 16, S29–S39.

Family Involvement  467 Barrera, M., & Bonds, D. (2005). Mentoring relationships and social support. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 133–142). Thousand Oaks, CA: Sage. Barron-McKeagney, T., Woody, J. D., & D’Souza, H. J. (2001). Mentoring at-risk Latino children and their parents: Impact on social skills and problem behaviors. Child and Adolescent Social Work Journal, 18, 119–136. Barron-McKeagney, T., Woody, J. D., & D’Souza, H. J. (2002). Mentoring at-risk Latino children and their parents: Analysis of the parent-child relationship and family strength. Families in Society, 83, 285–292. Bowlby, J. (1969). Attachment and loss: Attachment (Vol. 1). New York: Basic Books. Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press. Cavell, T. A., Elledge, L. C., Malcolm, K. T., Faith, M. A., & Hughes, J. N. (2009). Relationship quality and the mentoring of aggressive, high-risk children. Journal of Clinical Child & Adolescent Psychology, 38, 185–198. DuBois, D. L., Holloway, B., Valentine, J., & Cooper, H. (2002). Effectiveness of mentoring programs: A metaanalytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. Family Strengthening Policy Center. (2004). Mentoring as a family strengthening strategy (Policy Brief No. 4). Retrieved from http://www.nationalassembly.org/ fspc/documents/PolicyBriefs/Brief4.pdf Grossman, J., & Rhodes, J. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work? An impact study of the Big Brothers/Big Sisters program. Evaluation Review, 22, 403–426. Harris, E., & Wimer, C. (2004). Engaging with families in out of school time learning. Harvard Family Research Project. Harvard School of Education (Research Brief # 4). Henderson A. T., & Mapp, K. L. (2002). A new wave of evidence. The impact of school, family and community connections on student achievement. Austin, TX: National Center for Family and Community Connections With Schools. Hess, J. Z., Barr, S. C., & Hunt, G. D. (2009). The practice of family mentoring and advocacy: A theoretical investigation of critical issues. Families in Society: The Journal of Contemporary Social Services, 90, 189–195. Hughes, J. N., Cavell, T. A., & Meehan, B. T. (2001). Preventing substance abuse in regressive children: Final report of grant R01-DA10037. Unpublished manuscript.

Hughes, J. N., Cavell, T. A., Meehan, B. T., Zhang, D., & Collie, C. (2005). Adverse school context moderates the outcomes of two selective intervention programs for aggressive children. Journal of Consulting and Clinical Psychology, 73, 731–736. Karcher, M. J. (2008). The Cross-Age Mentoring Program (CAMP): A developmental intervention for promoting students’ connectedness across grade levels. Professional School Counseling, 12, 137–143. Karcher, M. J. (2012). The Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors: Program manual. San Antonio, TX: Developmental Press. Karcher, M. J., Davis, C., & Powell, B. (2002). The effects of developmental mentoring on connectedness and academic achievement. School Community Journal, 12(2), 35–50. Keller, T. (2005). A systemic model of the youth mentoring intervention. Journal of Primary Prevention, 26, 169–188. Kerr, M. E., & Bowen, M. (1988). Family evaluation: The role of the family as an emotional unit that governs individual behavior and development. New York: Norton. LoSciuto, L., Rajala, A., Townsend, T., & Taylor, A. (1996). An outcome evaluation of Across Ages: An intergenerational mentoring approach to drug prevention. Journal of Adolescent Research, 11, 116–129. Marin, G., & Marin, B. (1991). Research with Hispanic populations. Newbury Park, CA: Sage. Meissen, G. J., & Lounsbury, J. W. (1981). A comparison of expectations of volunteers, children and parents in a Big Brothers/Big Sisters program. Journal of Community Psychology, 9, 250–256. MENTOR. (2009). Mentoring immigrant and refugee youth: A toolkit for program coordinators. Alexandria, VA: Author. Miller, A. (2007). Best practices for formal youth mentoring. In T. D. Allen & L. T. Eby (Eds.), The Blackwell handbook of mentoring: A multiple perspectives approach (pp. 307–324). Malden, MA: Blackwell. Moon, T. R., & Callahan, C. M. (2001). Curricular modifications, family outreach, and a mentoring program: Impacts on achievement and gifted identification in high-risk primary students. Journal for the Education of the Gifted, 24, 305–321. Morrow, K., & Styles, M. (1995). Building relationships with youth in program settings. Philadelphia: Public/ Private Ventures. Philip, K., Shucksmith, J., & King, C. (2004). Sharing a laugh? A qualitative study of mentoring interventions for young people. York, UK: Joseph Rowntree Foundation. Rhodes, J. E., Contreras, J. M., & Manglesdorf, S. C. (1994). Natural mentoring relationships among Latina adolescent mothers: Psychological adjustment, moderating processes and the role of early parental acceptance. American Journal of Community Psychology, 22, 211–227.

468   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Rohner, R. P., & Britner, P. A. (2002). Worldwide mental health correlates of parental acceptance-rejection. Review of cross cultural and intracultural evidence. Cross Cultural Research, 36, 16–47. Sánchez, B., & Colón, Y. (2005). Race, ethnicity, and culture in mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 191–204). Thousand Oaks, CA: Sage. Sánchez, B., & Reyes, O. (1999). A descriptive profile of the mentoring relationships of urban Latino adolescents. Journal of Community Psychology, 27, 299–302. Sipe, C. (2002). Mentoring programs for adolescents: A research summary. Journal of Adolescent Health, 31, 251–260. Soucy, N., & Larose, S. (2000). Attachment and control in family and mentoring contexts as determinants of adolescent adjustment to college. Journal of Family Psychology, 14, 125–143. Spencer, R., Basualdo-Delmonico, A., & Lewis, T. O. (2011). Working to make it work: The role of parents in the youth mentoring process. Journal of Community Psychology, 39, 51–59.

Styles, M., & Morrow, K. (1992). Understanding how youth and elders form relationships: A study of four Linking Lifetimes programs. Philadelphia: Public/ Private Ventures. Taylor, A. S. (2000). Mentoring across generations: Partnerships for positive youth development. New York: Springer. Taylor, A., LoSciuto, L., Fox, M., & Hilbert, S. (1999). The mentoring factor: An evaluation of Across Ages. In V. S. Kuehne (Ed.), Intergenerational program research: Understanding what we have created (pp. 77–100). Binghamton, NY: Haworth Press. Weinberger, S. (1992). How to start a student mentor program. Bloomington, IA: Phi Delta Kappa Education Foundation. Wheeler, M., & DuBois, D. L. (2009). Analysis of responses to agency practices survey for Big Brothers/Big Sisters of America’s community-based mentoring program. Unpublished report. Zimmerman, M. A., Bingenheimer, J. B., & Behrendt, D. E. (2005). Natural mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 143–157). Thousand Oaks, CA: Sage.

32  TERMINATION AND CLOSURE OF MENTORING RELATIONSHIPS Renée Spencer and Antoinette Basualdo-Delmonico

Introduction “We live our lives, forever taking leave”—Rilke (1923/1939) When do youth mentoring relationships end? Mentoring, by definition, evokes a sense of timelessness. Indeed, some mentoring relationships do last a lifetime—whether in the actuality of sustained interpersonal contact, however infrequent, or in the spirit of present-day successes being attributed to the influences of trusted guides from long ago. Perhaps it should not be surprising then that endings in youth mentoring relationships, whether natural or formal, have received little concerted attention in the empirical literature to date (Keller, 2005; Spencer, 2007). We seem to prefer to think of mentoring relationships as going well and continuing indefinitely in one form or another or simply naturally fading over time. Many agencies do in fact hope that the relationships established through their programs will grow into ties that are more natural and sustained over time without the support of the agency. Unfortunately, research suggests otherwise, with fewer than half of relationships established through formal programs lasting to their initial time commitment (Bernstein, Dun Rappaport, Olsho, Hunt, & Levin, 2009; Grossman & Rhodes, 2002). Further, evidence suggests that relationships that end prematurely may do more harm than good (Grossman & Rhodes, 2002; Herrera et al., 2007), perhaps due in part to the greater salience of negative experiences in interpersonal relationships (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001). At the same time, some evidence suggests that short, but highly consistent

formal relationships with frequent contact and planned endings may not run these same risks (e.g., Cavell, Elledge, Malcolm, Faith, & Hughes, 2009). Endings are a normal part of all relationships that, when handled well, can reinforce the positive aspects of the relationship and help build expectations or “working models” (Shaver, Mikulincer, Leary, & Hoyle, 2009) that can be applied productively to future relationships. Engaging in a closure process can model healthy endings for youth who have experienced poorly handled endings in prior relationships with adults. Conversely, poorly ended relationships can reinforce negative working models of relationships and diminish optimism that things will go well or differently in future relationships. Mentoring programs tend to serve vulnerable youth and natural mentoring relationships are thought to be protective for vulnerable youth as well (e.g., Werner, 1995). Although the intention of mentoring is to provide significant support to such youth and to contribute positively to their development, their vulnerabilities can both make it more challenging to form a meaningful relationship and heighten the potential negative effects of a poor relationship. As such, there can be unintended consequences of not ending mentoring relationships well. Further, when a mentoring relationship is not going well, an expedient and wellhandled closure may offer important opportunities. A disinterested youth can be freed up to pursue other interests and the mentor can be matched with another child; or a low-quality match (whether due to poor fit, mentor inconsistencies, etc.) may be replaced by one that is more beneficial to the young person. In this chapter, we review theory within youth mentoring on the importance of endings and their 469

470   PRACTICE AND PROGRAMMATIC CONSIDERATIONS place in the mentoring process as well as theory and research from related fields of study: higher education and workplace mentoring and termination of psychotherapy relationships. We then review existing research on endings in mentoring relationships, identified through the use of major search engines including PsycINFO, ProQuest Social Sciences, and ERIC, with the following keywords: youth mentoring, formal mentoring, mentoring, closure, ending, and termination. This review is followed by recommendations for future research. Finally, we discuss current program practices and argue for the integration of attention to closure into all stages of the mentoring process. In these efforts, we pay closest attention to formal youth mentoring relationships, as the scant literature that does exist on endings in youth mentoring is focused almost exclusively on formal relationships. We incorporate relevant literature on naturally occurring mentorships and discuss endings in those relationships, when possible.

Theory Mentoring Phase models of mentoring that describe the “life cycle” or developmental stages of mentoring relationships—whether youth, workplace, or academic—all include a termination or dissolution phase, recognizing that endings are a normative part of the mentoring process (Johnson, 2007; Keller, 2005; Kram, 1988; Taylor & Bressler, 2000). Keller calls this phase in youth mentoring “decline and dissolution” and draws a distinction between these two types of endings. Decline is the more passive drifting apart over time that accompanies reductions in the importance and level of closeness in the relationship, whereas dissolution is a more intentional or active termination of the relationship that may be marked by a clear event. In addition to a more natural fading away over time, relationships may either decline or dissolve as a result of conflict, betrayal, or disappointment, or from lack of attention to actively maintaining the connection. In both workplace and academic mentoring relationships, the separation phase is thought to optimize the benefits of mentoring, as the hierarchical nature of the relationship shifts in response to greater autonomy, increasing self-confidence, and a more independent sense of professional identity on the part of the mentee (Johnson, 2007; Kram, 1988). Ideally, the decision to end a mentoring relationship should be mutual and result in a positive and growth-promoting experience for the protégé.

However, this is often not the case. Relationships may be ended by the mentor, the youth, the youth’s parent or guardian, the mentoring program, or any combination of these and may result from negative experiences, such as dissatisfaction and disappointment or even abandonment (Spencer, 2007). Kram (1988) conceptualized endings in workplace relationships as having both structural and psychological dimensions. Structural endings include promotions or other changes in assignments or departments that end the formal relationship between mentor and mentee. Psychological endings include diminished need on the part of the mentee for the guidance offered or the mentor’s diminished capacity or availability to offer meaningful guidance. Kram considered relationship development and the interplay between structural and psychological endings, noting that both the timing and the type of ending can make a difference. A well-timed structural ending can encourage more of a psychological separation, enabling the mentee to try out functioning with less guidance and support. Premature structural endings may provoke feelings of abandonment and anxiety as a mentee is forced to function more independently before feeling ready to do so. A structural ending that occurs later than a psychological one may lead to feelings of resentment on the part of both mentee and mentor as the relationship no longer suits the needs of one or both parties. In all of these cases, endings may be functional or dysfunctional. Functional well-timed endings may play an important role in the mentee’s development. Taylor and Bressler (2000), in their explication of the Across Ages youth mentoring program, asserted succinctly, “How a relationship ends is key to how mentors and especially youth think about and value their experiences together” (p. 79). While even the positive ending of a relationship involves loss, it may also open up possibilities for new relationships. The literatures on both workplace and academic mentoring purport a need for multiple mentoring relationships, as different mentors can meet different needs and may be more and less useful at different points in the development of the mentee (Burlew, 1991; Kram & Bragar, 1992). It is also possible that an individual who effectively mentors a young child may struggle to mentor a teen, so in some cases a change to or the addition of a new mentor may be warranted as a child progresses through different developmental periods. In both workplace and academic mentoring, some consideration has been given to identifying mentoring relationships that are not going well and perhaps should be ended. Unfortunately, to date, little of this type of conceptual work has been done

Termination and Closure of Mentoring Relationships    471 on endings in youth mentoring. A dysfunctional academic mentoring relationship has been defined as one that is “no longer functioning effectively for one or both partners” (Johnson & Huwe, 2002, p. 45). This may be due to one or both partners’ needs not being met, costs outweighing the benefits, or experiencing distress due to being in the mentorship. Relationship dysfunction is believed to result from poor matching, mentor incompetence, mentor neglect and abandonment, relational conflict, boundary violations, and mentee traits and behaviors, such as personality types and emotional and behavioral issues (Johnson & Huwe, 2002). Cross-gender and cross-race matching can also be problematic when factors such as stereotyping and differing socialization practices interfere with the mentoring process. Eby, McManus, Simon, and Russell (2000) developed a taxonomy of negative experiences in workplace mentoring relationships based on qualitative accounts of mentees’ experiences. Five groups of experiences were identified: (a) mismatches in mentor and mentee personality, values, and work styles, (b) lack of mentor expertise, (c) distancing and manipulative behavior on the part of the mentor, and (d) poor attitude and/or (e) personal problems on the part of the mentor. Eby and Allen (2002) found associations between negative mentoring experiences and self-reports of poor outcomes—such as low job satisfaction, thoughts of quitting, and stress. Mentees also indicated that negative experiences in formal relationships had a greater effect on thoughts about quitting and workplace stress than did such experiences in informal mentoring relationships. Eby and Allen (2002), drawing on the social psychology literature addressing close relationships more generally, speculated that mentees have higher expectations for formal mentoring relationships and that such relationships tend to be shorter in duration and also less deep and intimate and thus less resilient in the face of difficulties. Perhaps the influence of early terminations also varies according to relationship type and length. Participants in programs that strive for longer-lasting relationships may hold higher expectations for the relationship and feel more disappointed when a relationship ends early than those in programs offering shorterterm relationships. Dashed expectations may be of special significance to the many youth served by mentoring programs who have experienced significant disruptions in their primary caregiving relationships, whether due to parental separation, incarceration, or transfer to foster care. Research on attachment has found that youth who experience disruptions in attachment early in life tend to have difficulty developing

secure attachments later (Shaver et al., 2009). These youth tend to expect rejection and unpredictability in their relationships, whereas a more securely attached young person will expect availability and the meeting of their needs. Unfortunately, due to these kinds of disruptions in interpersonal relationships, poorly attached youth may be especially hard to engage in a mentoring relationship and at higher risk for early termination. Psychotherapy The work of termination in psychotherapy involves looking back and reviewing progress made during therapy, looking forward to areas of continued growth and plans for the future after therapy, and saying goodbye in therapy (Gelso & Woodhouse, 2002). When handled well, it is believed that the termination process can solidify gains made, resolve issues that have arisen in the psychotherapy relationship itself, and prepare the client for maintaining healthy functioning without the support of treatment (Joyce, Piper, Ogrodniczuk, & Klein, 2007). Poor endings, in contrast, hold the potential to undo some of these gains when the process of ending the relationship stirs up and does not adequately address issues and conflicts previously raised in the treatment and any unresolved issues associated with loss and separation the client may have carried into treatment (Joyce et al., 2007). Further, positive or appropriate terminations are thought by many to contribute to the work of therapy continuing after the treatment has ended through what has been called “afterwork,” or the internalization of the therapy, such as clients imagining the questions the therapist might ask and then asking these questions of themselves (Bernard & Drob, 1989). Ideally, termination of a psychotherapy relationship occurs when symptoms have abated and the person is functioning well (Holmes, 1997). However, as with mentoring, early terminations are quite common. Some estimates suggest about 40%– 60% of youth end psychotherapy early (Fabricius & Green, 1995; Kazdin, 1996). Distinctions are made between different types of early terminations. Premature terminations are those initiated by the client, and forced terminations are those initiated by the therapist (Joyce et al., 2007). Some therapistinitiated terminations result from life events, such as the short-term nature of training internships, a move, retirement, or death. Others arise out of what in the psychotherapy literature is called countertransference issues (i.e., the therapist’s own emotional responses) that contribute to the therapist’s feeling she or he is not able to be helpful to the

472   PRACTICE AND PROGRAMMATIC CONSIDERATIONS client (Joyce et al.). Forced terminations tend to be harder for clients than natural ones (Gelso & Woodhouse, 2002) and can provoke a range of responses, including feelings of anxiety and loss, sadness, self-blame, and anger or a complex mix of these (Penn, 1990). Whatever the reason for the termination, it is considered to be the therapist’s responsibility to manage the termination process in a way that enhances rather than undoes whatever progress has been made and prepares the client either to function well without the continued support of psychotherapy or to build a productive relationship with a new therapist. Unfortunately, one study found that therapists tend to engage in fewer of the generally accepted termination practices with unsuccessful cases than with successful ones (Quintana & Holahan, 1992). A number of factors influence the importance of termination and the issues that may need to be addressed. Joyce and colleagues (2007) grouped these factors into three major categories: those associated with (a) the specific therapy case (duration of treatment, who initiated the ending, and treatment satisfaction), (b) the treatment approach (time limited or open-ended, degree of structure and emphasis on skill-building, and the role of the relationship in the treatment process), and (c) therapist or client characteristics (e.g., client level of dependence on therapist, therapist and client history with loss, especially during critical moments in development). For example, the nature of the issues raised by and addressed in the termination process would be quite different for a client who has suffered multiple traumatic losses and who has been in a multiyear therapy relationship that is being terminated by the therapist than for a client coming to the end of a predetermined 10-week course of therapy focused on developing specific skills to manage feelings of anxiety resulting from an automobile accident. Some links have been made between termination activity and client outcomes. In a comparison of more and less successful short-term (4–15 sessions) counseling cases (Quintana & Holahan, 1992), terminations in less successful cases included less discussion of closure and review of the counseling process and preparation of the client for the end of the counseling relationship. The clients in these cases expressed more frustration about the ending of the relationship and were more devaluing of the therapy experience than were the clients in more successful cases. Some research has found associations between termination experiences and aspects of the client, therapist, treatment approach and course of treatment. In one study (Marx & Gelso, 1987), the

importance of exploring feelings about the end of therapy was greater among those clients who had a history of significant loss, such as death or divorce of parents. Another (Roe, Dekel, Harel, Fennig, & Fennig, 2006) found that clients who experienced premature terminations of their therapy relationships and those who indicated not processing the termination with their therapist tended to report more negative feelings about termination. Others have found that clients also have many positive responses to the ending of treatment, especially when the ending was planned and mutual. These include feelings of joy and pride associated with a sense of accomplishment and moving toward greater independence (Marx & Gelso, 1987; Roe at al., 2006). One qualitative study of therapy relationships with interns found that all of the clients informed about the intern’s planned departure at the outset of treatment completed the full course of treatment, whereas some of the clients who were not informed until near the end of the year did not return for subsequent sessions (Gould, 1978). Client reports indicate a desire for clear endings (e.g., Marx & Gelso, 1987), and it is commonly accepted practice wisdom that abrupt endings have the potential to be harmful in both the near and long terms (Gelso & Woodhouse, 2002).

Research Although there has been little dedicated empirical study of closure and termination of mentoring relationships, information about endings can be gleaned from studies of youth mentoring focused on different aspects of the mentoring process. Evidence is mounting that youth in closer, more enduring relationships tend to benefit more from mentoring than do youth in shorter, less connected matches (Rhodes & DuBois, 2006). Herrera and colleagues (2007), in their national study of school-based mentoring relationships, found that youth whose relationships continued into a second school year benefited the most from their mentoring relationships but also that the gains made by youth who were mentored for only one school year faded by the fall of the second year. Notably, studies of natural mentoring relationships associated with positive outcomes have found many of these ties to be quite long lived, with some middle and late adolescents reporting on relationships begun in early and middle childhood (DuBois & Silverthorn, 2005; Rhodes, Contreras, & Mangelsdorf, 1994). At the same time, research on formal mentoring programs indicates that premature endings are a

Termination and Closure of Mentoring Relationships    473 frequent occurrence. National studies of both community- and school-based programs report that fewer than half of the relationships in these studies lasted to their initial time commitment, whether a calendar or school year (Bernstein et al., 2009; Grossman & Rhodes, 2002). In the study of school-based mentoring relationships by Herrera and colleagues (2007), which followed matches for 15 months, only 52% of the youth were continuing to meet with a mentor in the late fall of the second year and fewer than 18% of the youth in the study met with the same mentor for all three semesters. Thus, about half of the youth participating had experienced one closure and nearly a fifth experienced two, many of these premature. Shorter, time-limited (i.e., 10–20 weeks) and structured programs may do better in this regard, with studies of such programs reporting negligible rates of early closure (Anderson et al., 2006; Cavell & Hughes, 2000; T. A. Cavell, personal communication). There has been some examination of the effects of early termination on youth, with studies of programs promising youth a mentor for a minimum of either a school or calendar year reporting no gains in youth functioning (Grossman, Chan, Schwartz, & Rhodes, 2012) or associations between early match closure and decrements in youth functioning (Britner & Kraimer-Rickaby, 2005; Grossman & Rhodes, 2002). Increasing match length by simply rematching youth whose relationships end early may not rectify this problem. The only study to date comparing youth receiving mentoring for roughly the same amount of time, but with one group in intact matches and the other in a second match after their first ended early, found that only the youth in intact matches demonstrated improvements in academic functioning (Grossman et al., 2012). Further, match length does not appear to be the only critical factor. One study (Karcher, 2005) also found a link between mentor attendance and youth outcomes, with youth whose mentors attended sporadically reporting decrements in self-esteem and physical attractiveness, suggesting that mentor consistency may play an important role. Other research has examined factors that contribute to premature endings. In one qualitative interview study of early-ending matches in traditional community-based mentoring programs (Spencer, 2007), the following appeared to play a role in the early ending of these relationships: (a) mentor or mentee abandonment, (b) perceived lack of mentee motivation, (c) unfulfilled expectations, (d) deficiencies in mentor relational skills, including the ability to bridge cultural divides, (e) family interference, and (f) inadequate agency support.

Unfortunately, multiple youth interviewed for this study described being abandoned by mentors who simply stopped returning phone calls or did not show up for a scheduled visit and never resurfaced. In other cases, challenges within the relationship, such as lack of interest on the part of the youth or mentors’ expectations for the relationship not being met, seemed to contribute to difficulties building a relationship or the erosion of an initial connection over time. Likewise, Shlafer, Poehlmann, Coffino, and Hanneman (2009), in a mixed-methods study of mentoring relationships among youth whose parents were incarcerated, found five primary reasons for early match termination (< 6 months): (a) scheduling conflicts, (b) personal or family issues (e.g., financial burden of being a mentor, youth seeming not to want a mentor), (c) residential mobility (e.g., changes in caregivers), (d) mentors underestimating the commitment, and (e) match incompatibility (e.g., child viewing mentor only in instrumental terms). Both of these studies noted that many mentors seemed to feel unprepared for the challenges they encountered and reported having strong personal reactions to the differences between their own lives and those of their youth mentees. Beyond these reasons for match closure, some characteristics of youth and mentors associated with early match closure have been identified. Youth age at time of match, gender, and risk status have been found to play a role. Youth age may interact with program type. Community-based matches with older youth (13–16 years of age) tend to be of shorter duration than those with younger youth (Grossman & Rhodes, 2002). Relationships with youth who have more complex problems (e.g., history of emotional, sexual, or physical abuse), or who are referred to a mentoring program in response to psychological or educational difficulties, tend not to last as long (Grossman & Rhodes, 2002). In addition, female matches tend to end earlier (Grossman & Rhodes, 2002), and one study (Rhodes, Lowe, Litchfield, & Walsh-Samp, 2008) found that girls were significantly less satisfied than boys in shortand medium-length mentoring relationships, suggesting that girls may also be more sensitive to early terminations. On the volunteer side, one study found that adults who had a lower income tended to have shorter matches than did adults who were married and in their late 20s (Grossman & Rhodes, 2002). Summary and Recommendations Too little attention has been paid to how and why mentoring relationships end and the consequences to the youth participants of different types

474   PRACTICE AND PROGRAMMATIC CONSIDERATIONS and approaches to relationship endings. This is especially true for natural mentoring relationships, as to our knowledge there has been no focused study of endings in these relationships. The lack of attention to endings in youth mentoring may inadvertently contribute to false expectations that most relationships will be long term if not lifelong, when current evidence suggests otherwise. Although some evidence suggests that earlyending relationships may be harmful to youth, it is not clear whether this is due to the relationships being short lived or the endings being unplanned. Can planned short-term mentoring relationships be effective? Some research indicates they can (Cavell et al., 2009). However, research that focuses systematically on the influence of planned versus unplanned endings on youth outcomes is needed to begin to disentangle the effects of relationship duration and planned versus unplanned terminations. The development of a typology of mentoring relationship endings would be useful for examining the associations between how relationships end and youth outcomes. Discerning whether unplanned endings due to changes in life circumstances (e.g., moving out of the area) versus some type of challenge or difficulty in the relationship (e.g., mentor inconsistency, mentor or youth abandonment) differentially influence youth outcomes is critical and would have significant implications for program practices for match closures. So, too, would research on the effects of the way that endings are handled. Do planned endings marked by discussions of the ending and reflections on the gains made or key relationship milestones enhance or solidify the benefits of the relationship, as the psychotherapy literature would suggest? Conversely, do poorly planned, unprocessed endings have the potential to diminish the efficacy of the experience? Finally, we are in critical need of clear guideposts for when a relationship should be terminated early; that is, when continuing the relationship in its present form may be more harmful than ending the relationship early. Karcher’s (2005) finding that inconsistent attendance on the part of school-based mentors was associated with decrements in youth functioning highlights the need to look beyond duration and closeness for additional potential markers of quality that could be monitored as the relationship develops and progresses. The contribution of specific program practices to rates of early relationship endings and the effects of these on participants are other areas ripe for research. Are different types and amounts of mentor training associated with higher and lower early match closure rates? Do some approaches to match

endings, such as having specific structures for how and over what amount of time to say goodbye, yield better outcomes for youth, whether by enhancing the influences of a relationship that has gone well or mitigating the negative effects of a poor experience? These and many other questions addressing endings in youth mentoring relationships warrant attention.

Practice As the evidence for the greater effectiveness of closer, more enduring mentoring relationships has grown (Rhodes & DuBois, 2006), so too has an emphasis on sustaining matches made through formal programs for longer periods of time. The hope is that longer matches will result in higher quality, more impactful mentoring. This is a welcome counterpoint to the push from funders and even agency governing boards to make ever more matches rather than ensuring the matches already made do well and last. But focusing on match length as a benchmark of quality may have the unintended consequence of discouraging agencies from closing relationships when termination may be warranted. In such cases, timely intervention to end the match in as positive a manner as possible may better serve the child’s interests. Agencies may also be inclined (and rewarded) to keep matches on their rolls well beyond the point that the relationship is active or productive, which can interfere with developing a better understanding of the true average relationship length in different types of programs. Unfortunately, such practices may also mean that many youth served in mentoring programs have experienced relationships that ended without proper closure. Moreover, parents or guardians may be reluctant to push for closure of a relationship that has fizzled as their child may have been on a waiting list for a rather long time. Hoping that the relationship gets reignited may seem preferable to having to endure a long wait for a new mentor or, in programs that do not rematch youth, being certain to have no mentor at all. Still, some inroads have been made in the practice realm. Match closure is more prominent in the most recent edition of the Elements of Effective Practice (MENTOR, 2009) than it has been previously, and a number of readily available tools can be adapted to fit the needs of a variety of programs. Closure is conceptualized as the third stage of mentoring relationships (following the development of trust and rapport and the growing of the relationships and reaching goals) in the tool kit for program managers published by MENTOR (2005). Relatively

Termination and Closure of Mentoring Relationships    475 recent guidebooks for mentoring program managers and staff (Garringer & McRae, 2008; MENTOR, 2005; Weinberger, Garringer, & MacRae, 2005) include recommended program policies and practices, such as having clearly defined procedures for deciding when to end a match, handling both planned and unplanned endings, and conducting exit interviews with all mentors and youth. The MENTOR tool kit (2005) contains a “termination ritual” that considers the different ways a relationship may end and constructive ways of handling these (Lakes & Karcher, 2005). Garringer and McRae also recommend having clear policies, and even a signed contract, regarding future contact between mentors and youth once the program has ended. One of these guides (Garringer & MacRae, 2008) was notable in its treatment of closure for the inclusion of parents and guardians. This resource notes that parents may feel angry in response to the match closing, especially if the mentor has decided to end the relationship and the family feels let down by the mentoring agency. In our qualitative study of endings in youth mentoring relationships, we found parents to have significant insight into what contributed to the dissolution of the relationship and the consequences of poorly handled endings for their child (Spencer, Basualdo-Delmonico, & Walsh, 2011). This leads us to concur that involving parents and guardians at the early signs of relationship struggles may be helpful in determining the best course of action.

Beyond a handful of other fact sheets and summaries, we were unable to locate more in-depth treatments of the practice issues related to termination and closure of youth mentoring relationships. Below we offer a set of recommendations extending existing practice recommendation in ways that are informed by the conceptual and empirical literatures reviewed in the previous sections of this chapter. These recommendations are summarized in Table 32.1. Fully Integrate Closure Into the Service Delivery Model Start Ending From the Beginning. Closure should be introduced at the outset of every mentoring relationship and revisited throughout the life of the match. Although it may seem to be at odds with the goal of building a new relationship, especially perhaps one that all parties hope will last indefinitely, waiting until trouble arises or even simply waiting until the last planned meeting of a match to discuss closure and say goodbye may leave vulnerable youth in a position of experiencing another painful loss. Raising the issue of closure at the beginning introduces endings as a normal phase in the mentoring process and lays the groundwork for participants to know what to expect and to prepare for this phase— whether it occurs sooner or later. It also offers the opportunity for agencies to inform all participants

Table 32.1   Checklist for Practitioners Topic

Recommendations

Is closure fully integrated into the service delivery model?

Introduce endings from the start of the match. Make explicit closure policies and procedures. Communicate these clearly to all participants. Prepare mentors for the day-to-day experiences of mentoring and the challenges they may encounter. Monitor all matches regularly for quality and assess whether they should be allowed to continue or be closed. Conduct exit interviews with mentors, youth, and families.

Are clear structures and supports for the closure process provided to all participants?

Coach parties in how to engage in positive closure. Provide closure activities. Support all participants throughout the closure process according to individual needs and circumstances. Establish and communicate clear guidelines for contact once the match has ended.

476   PRACTICE AND PROGRAMMATIC CONSIDERATIONS about the expectations regarding closure and the importance of planning and preparing for ending and saying goodbye. Develop and Communicate Explicit Closure Policies and Procedures. Programs should have clear policies regarding when relationships should end, including when a relationship should be ended before the agreed-upon initial time commitment, and procedures for tracking relationships to ensure that each match is ended formally or, when applicable, continues but no longer as a part of the mentoring program. When and how endings are to take place should be communicated clearly to all program participants (mentors, youth, families) and collaborating institutions (e.g., schools) at the outset of the match. Such policies can be flexible enough to accommodate individual match circumstances while also offering sound structure. For example, even when a program intends for matches to continue for multiple years, there can still be clear time points at which the progress of the relationship is reviewed and discussions take place about whether the relationship should continue. Procedures for handling both planned and unplanned endings should be in place and include clear policies regarding whether and, if so, under what conditions youth will be rematched. These, along with the roles that mentors, youth, and families are expected to play in the closure process should be delineated at the outset of the relationship. These discussions should include what the agency will do in the event that one party refuses to participate in the closure process. For example, in cases where mentors cannot be reached or refuse to say goodbye to the young person, agency staff would need to step in to complete the closure process and meet with the youth and a parent or guardian to process feelings about and responses to the ending of the relationship. In addition, whether and under what circumstances a child would be matched with another mentor upon the ending of one match should be communicated clearly to youth and their families. Given the personal nature of mentoring, closure should be handled through in-person meetings whenever possible. Participants, including parents, are often expected to make significant investments of time in the screening and matching process at the start of the relationship. Similar care should be given to the ending. Set Expectations and Prepare Volunteer Mentors. Research on early match closures indicates that volunteer expectations play a role in these, with mentors describing feelings of disappointment and even

dejection when the experience of mentoring a child is quite different from what they had imagined (Shlafer et al., 2009; Spencer, 2007). Public service announcements and recruitment pitches tend to emphasize the positive aspects of mentoring and highlight exceptional relationships. It is important to ensure that volunteer mentors are fully informed about the nature of the commitment they are making, what is expected of them and what will happen if they do not meet these expectations, common challenges that can arise when building a relationship with a young person, and the program supports available should they encounter these difficulties. So, too, can youth and families be prepared for what to expect from the mentor and the mentoring program. Regularly Monitor and Evaluate Mentoring Matches. Regular and careful monitoring of matches makes it possible to identify relationships that are not going well and creates opportunities for intervention, whether by offering additional support to shore up a struggling match with potential or taking action to end a match that is unlikely to improve and may even be harmful. Programs should have guidelines for when matches should be retained and when closure should be considered. Regular follow-up may also reveal when a mentor has decided to end a match but is waiting until the last visit to tell the youth, creating the opportunity for staff to avert an abrupt and potentially harmful ending and facilitate a more full closure process. Involving parents in monitoring activities and in the planning of match endings is important as well, as they have a particular vantage point on their child’s needs and satisfaction with the match and can serve as valuable resources throughout the mentoring process (Spencer, Basualdo-Delmonico, & Lewis, 2011). Conduct Exit Interviews. Once the match has ended, exit interviews should be conducted with mentors, youth, and families. Such interviews can serve many purposes. They further mark the clear end of the mentoring relationship or, in cases where matches “graduate” from the program but continue on, the end of the relationship with the agency. Programs can receive feedback about their services, identify potential areas for improvement, and learn from relationships that have gone poorly. Exit interviews also ensure that some closure process takes place, with or without the mentor’s participation, and provide a mechanism for monitoring whether youth have been adversely affected by the termination. Parents may be especially helpful informants in these arenas.

Termination and Closure of Mentoring Relationships    477 Provide Clear Structure and Support for the Closure Process Coach All Parties in Positive Closure. Programs should not assume that either mentors or youth have the skills to handle ending the relationship well, even under the best of circumstances. For example, in the Cross-Age Mentoring Program (CAMP; Karcher, 2008, 2012), in response to observations that participants rarely handled goodbyes in a way that felt satisfactory despite the use of a structured “Termination Ritual,” a practice goodbye session was instituted, as were periodic structured times for the pair to reflect on the relationship along the way. In this program, which considers endings in mentoring relationships to be rich opportunities for growth and learning, planning for the end from the beginning is built into the program structure, as is training of the participants in how to say goodbye and chances to practice doing so. Mentors may struggle with their own feelings around relationship endings and can sometimes respond to these in ways that are hurtful to youth, such as avoiding saying goodbye or making promises to the young person they will not be able to keep. Ascertaining during recruitment how mentors have handled relationships endings in their own lives may provide insight about the kinds of issues that could surface during termination or indications of their fit for the kind of youth mentoring they would be providing. Provide Closure Activities. Saying goodbye can be difficult, even when the ending is positive, and many people have not had much experience processing relationship endings in a productive manner. Programs should provide rituals, activities, ideas for ways to celebrate endings, and questions to facilitate mentors and youth reviewing their relationship (e.g., “What did you like best about the time we spent together?” “What one thing do you think we should have done differently?”). In the most unfortunate cases, when either the mentor or youth refuses to participate in a closure process or the match is closing because one party has abandoned the relationship, programs should engage in a closure process with the remaining participant. Support Participants Throughout the Closure Process. Program staff should be trained and prepared to coach and support mentors, mentees, and families; offer opportunities to practice saying goodbye; and guide all parties through the closure process. Participants may be reluctant to participate in a closure process and may need strong encouragement and support from program staff to do so (Karcher,

2006). When the relationship has gone poorly for the mentee, both the youth and the parent or guardian may need additional support as they deal with the fallout from the negative mentoring experience. Staff should also clearly communicate next steps to each of the parties, such as whether the youth is to be rematched (and, if so, when) or the volunteer reassigned, and convey any recommendations for other programs and resources that might be beneficial to the youth. Set Clear Guidelines for Post-Program Contact. In some cases, the end of the program may coincide with that of the mentoring relationship, whereas in other cases it may not. Some matches made through formal programs may continue after participation in the program has ended. Indeed, many programs may see such continuation as a sign of program success, as the mentoring relationship has now become a meaningful, worthwhile, and lasting relationship for both mentor and youth. However, even in these cases, clarity about the end of the role of the program in the mentoring relationship and the expectations of all parties moving forward is needed. General promises to “keep in touch” are too often broken, as the business of life or uncertainties about what type of contact and when it would be welcomed by the other party interfere with good intentions. When the relationship is to end and no further contact is expected, being clear about that provides all parties with an opportunity to say goodbye. End-of-match meetings can provide opportunities to discuss and set expectations regarding postprogram contact and facilitate clear goodbyes—even in cases where the only party leaving the relationship is the mentoring program.

Conclusion As Rilke’s quote at the opening of this chapter poignantly reminds us, endings are an inevitable part of living, and we would do well to pay much greater attention to endings in youth mentoring relationships. Johnson (2007) in his guidebook for faculty mentors wrote, “[L]eave taking is always healthiest when done deliberately” (p. 101). Although there is not a clear research evidence base upon which to form such a firm conclusion regarding youth mentoring relationships, research on other forms of mentoring relationships and in the related field of psychotherapy suggest that we would be wise to heed Johnson’s words until proved otherwise. At the heart of the mentoring process is the personal relationship between mentor and youth, and the kind of care and

478   PRACTICE AND PROGRAMMATIC CONSIDERATIONS attention paid to the formation of these ties should also be brought to bear on their closure. When youth and their families are trying to get by with extremely limited resources, it can be easy to slip into thinking that offering something would be better than doing nothing. However, greater caution is warranted in mentoring, given that the tool of change is a human relationship, which by its very nature holds the potential to both promote growth and cause suffering. Once a relationship has been introduced into the life of a child, programs must do everything they can to ensure that the relationship is handled responsibly— from beginning to end.

References Anderson, T., Lipman, E., Mills, B., Metz, H., Teram, E., Elbard, M., . . . Sanford, M. (2006). The Recreation Mentoring Program: A community engagement initiative for children. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 15(2), 59–63. Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psychology, 5, 323–370. Bernard, H. S., & Drob, S. (1989). “Afterwork”: A clinicalphenomenological report. Psychiatric Quarterly, 60, 359–369. Bernstein, L., Dun Rappaport, C., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S. Department of Education’s Student Mentoring Program (NCEE 2009-4047). Washington, DC: National Center for Educational Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education. Britner, P. A., & Kraimer-Rickaby, L. (2005). Abused and neglected youth. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 482–492). Thousand Oaks, CA: Sage. Burlew, L. D. (1991). Multiple mentor model: A conceptual framework. Journal of Career Development, 17, 213–221. Cavell, T. A., Elledge, L. C., Malcolm, K. T., Faith, M. A., & Hughes, J. N. (2009). Relationship quality and the mentoring of aggressive, high-risk children. Journal of Clinical Child and Adolescent Psychology, 38, 185–198. Cavell, T. A., & Hughes, J. N. (2000). Secondary prevention as context for assessing change processes in aggressive children. Journal of School Psychology, 38, 199–235. DuBois, D. L., & Silverthorn, N. (2005). Characteristics of natural mentoring relationships and adolescent adjustment: Evidence from a national study. Journal of Primary Prevention, 26, 69–92. Eby, L. T., & Allen, T. D. (2002). Further investigations of protégés’ negative mentoring experiences: Patterns

and outcomes. Group and Organization Manage­ ment, 27, 456–479. Eby, L. T., McManus, S. E., Simon, S. A., & Russell, J. E. A. (2000). The protégé’s perspective regarding negative mentoring experiences: The development of a taxonomy. Journal of Vocational Behavior, 57, 1–21. Fabricius, J., & Green, V. (1995). Termination in child analysis: A child-led process? The Psychoanalytic Study of the Child, 50, 205–226. Garringer, M., & MacRae, P. (2008). Foundations of successful youth mentoring. Washington, DC: Hamilton Fish Institute on School and Community Violence and the National Mentoring Center and Northwest Regional Educational Laboratory. Gelso, C. J., & Woodhouse, S. S. (2002). The termination of psychotherapy: What research tells us about the process of ending treatment. In G. S. Tryon (Ed.), Counseling based on process research: Applying what we know (pp. 344–369). Boston: Allyn & Bacon. Gould, R. P. (1978). Students’ experience with the termination phase of individual treatment. Smith College Studies in Social Work, 48(3), 235–269. Grossman, J. B., Chan, C. S., Schwartz, S. E. O., & Rhodes, J. E. (2012). The test of time in schoolbased mentoring: The role of relationship duration and re-matching on academic outcomes. American Journal of Community Psychology, 49, 43–54. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring programs. American Journal of Community Psycho­ logy, 30, 199–219. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Holmes, J. (1997). “Too early, too late”: Endings in psychotherapy—An attachment perspective. British Journal of Psychotherapy, 14, 159–171. Johnson, W. B. (2007). On being a mentor: A guide for higher education faculty. Mahwah, NJ: Erlbaum. Johnson, W. B., & Huwe, J. M. (2002). Toward a typology of mentorship dysfunction in graduate school. Psychotherapy: Theory/Research/Practice/Training, 39(1), 44–55. Joyce, A. S., Piper, W. E., Ogrodniczuk, J. S., & Klein, R. H. (2007). Termination in psychotherapy: A psychodynamic model of processes and outcomes. Washington, DC: American Psychological Association. Karcher, M. J. (2005). The effects of school-based developmental mentoring and mentors’ attendance on mentees’ self-esteem, behavior, and connectedness. Psychology in the Schools, 42, 65–77. Karcher, M. J. (2006). What happens when high school mentors don’t show up? In L. Golden & P. Henderson (Eds.), Case studies in school counseling (pp. 44–53). Alexandria, VA: ACA Press. Karcher, M. J. (2008). The Cross-Age Mentoring Program (CAMP): A developmental intervention for promoting students’ connectedness across

Termination and Closure of Mentoring Relationships    479 grade levels. Professional School Counseling, 12, 137–143. Karcher, M. J. (2012). The Cross-Age Mentoring Program (CAMP) for Children With Adolescent Mentors: Program manual. San Antonio, TX: Developmental Press. Kazdin, A. E. (1996). Dropping out of child psychotherapy: Issues for research and implications for practice. Clinical Child Psychology and Psychiatry, 1(1), 133–156. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Kram, K. E. (1988). Mentoring at work: Developmental relationships in organizational life. Lanham, MD: University Press of America. Kram, K. E., & Bragar, M. C. (1992). Development through mentoring: A strategic approach. In D. H. Montross & C. J. Shinkman (Eds.), Career development: Theory and practice (pp. 221–254). Springfield, IL: Thomas. Lakes, K., & Karcher, M. J. (2005). Mentor/mentee termination ritual. In How to build a successful mentoring program using the Elements of Effective Practice (pp. 157–158). Alexandria, VA: MENTOR. Marx, J. A., & Gelso, C. J. (1987). Termination of individual counseling in a university counseling center. Journal of Counseling Psychology, 34(1), 3–9. MENTOR. (2005). How to build a successful mentoring program using the Elements of Effective Practice. Alexandria, VA: Author. MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA.: Author. Penn, L. S. (1990). When the therapist must leave: Forced termination of psychodynamic therapy. Professional Psychology: Research and Practice, 21, 379–384. Quintana, S. M., & Holahan, W. (1992). Termination in short-term counseling: Comparison of successful and unsuccessful cases. Journal of Counseling Psychology, 39, 299–305. Rhodes, J. E., Contreras, J. M., & Mangelsdorf, S. C. (1994). Natural mentor relationships among Latina adolescent mothers: Psychological adjustment, moderating processes, and the role of early parental acceptance. American Journal of Community Psycho­ logy, 22, 211–227.

Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20(3), 3–19. Rhodes, J. E., Lowe, S. R., Litchfield, L., & Walsh-Samp, K. (2008). The role of gender in youth mentoring relationship formation and duration. Journal of Vocational Behavior, 72, 183–192. Rilke, R. M. (1923/1939). Duino elegies (J. B. Leishman & S. Spender, Trans.). New York: Norton. Roe, D., Dekel, R., Harel, G., Fennig, S., & Fennig, S. (2006). Clients’ feelings during termination of psychodynamically oriented psychotherapy. Bulletin of the Menninger Clinic, 70(1), 68–81. Shaver, P. R., Mikulincer, M., Leary, M. R., & Hoyle, R. H. (2009). Attachment theory and attachment styles. In R. Leary & R. H. Hoyle (Eds.), Handbook of individual differences in social behavior (pp. 62–81). New York: Guilford Press. Shlafer, R. J., Poehlmann, J., Coffino, B., & Hanneman, A. (2009). Mentoring children with incarcerated parents: Implications for research, practice, and policy. Family Relations, 58, 507–519. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Spencer, R., Basualdo-Delmonico, A., & Lewis, T. O. (2011). Working to make it work: The role of parents in the youth mentoring process. Journal of Community Psychology, 39, 51–59. Spencer, R., Basualdo-Delmonico, A., & Walsh, J. (2011, March). The good, the bad and the ugly: A qualitative interview study of endings in youth mentoring relationships. Symposium paper presented at the Society for Research in Child Development biennial meeting, Montreal, Canada. Taylor, A. S., & Bressler, J. (2000). Mentoring across generations: Partnerships for positive youth development. New York: Kluwer Academic. Weinberger, S., Garringer, M., & MacRae, P. (2005). Going the distance: A guide to building lasting relationships in mentoring programs. Washington, DC: U.S. Department of Education Mentoring Resource Center. Werner, E. E. (1995). Resilience in development. Current Directions in Psychological Science, 4, 81–85.

33  PROGRAM EVALUATION David L. DuBois

Introduction Mentoring programs for young people typically involve several interrelated components and processes. Mentors must be recruited, screened, and trained; youth who fit the parameters and goals of the program need to be identified and engaged successfully; and youth and mentors then need to be strategically paired and their relationships provided with ongoing oversight and support. A breakdown in any of these areas can derail even the most promising of programs. Gathering and analyzing the types of data necessary to diagnose and remedy such problems in a timely and efficient manner is important not only to continued viability of a program, but also to the safety and well-being of the young persons it serves. The desired outcomes of programs include strong, often long-lasting mentoring relationships as well as improvements in the adjustment of participating youth in diverse areas such as mental health, involvement in problem behavior, parent and peer relationships, academic achievement, and employment and career development. Programs also often have good reason to expect benefits for those who serve as mentors and others who are in the orbit of the youth’s life, such as parents, classmates, and members of the surrounding community. Gauging outcomes accurately and doing so in ways that can be attributed persuasively to program involvement is a challenging undertaking. Adding further complexity, policymakers and funders may place a premium on having a careful accounting of program costs and whether these expenditures are justified in relation to demonstrated benefits. Many of the outcomes that are central to this type of analysis, however, may not be realized fully until long after a youth’s participation in a program has ended.

The preceding considerations suggest that evaluation can make important contributions to the effectiveness, efficiency, safety, and sustainability of mentoring programs for youth, thus strengthening their overall quality and capacity to enhance the lives of participating youth and others (DuBois, 2007). In support of this conclusion, a meta-analysis of youth mentoring program effectiveness found that estimated positive impacts on youth outcomes were significantly greater when programs included systematic procedures for monitoring and evaluation (DuBois, Holloway, Valentine, & Cooper, 2002). It is equally apparent that highly formidable challenges will be encountered when attempting to undertake a rigorous and informative evaluation of a youth mentoring program. This chapter explores these challenges and potential solutions to them. The chapter begins with a brief overview of evaluation standards. The remainder of the chapter is devoted to the presentation of a general framework for program evaluation with emphasis on its application to evaluation of youth mentoring programs in particular. The framework is organized according to the widely accepted distinction between process and outcome evaluation, sometimes referred to as formative and summative evaluation, respectively. The first edition of the Handbook of Youth Mentoring featured a superb, but fairly technical treatment of issues involved in youth mentoring program evaluation (Grossman, 2005). To complement this earlier presentation, the emphasis in the present chapter is on practical strategies for evaluation and on presenting information in a manner that will be accessible to those with limited background or training in research. In line with this goal, the chapter’s key recommendations for good practice in mentoring program evaluation are summarized in Table 33.1 for easy reference. 481

482   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Table 33.1   Recommendations for Good Practice in Evaluation of Youth Mentoring Programs Topic

Recommendations

Standards and ethics

Plan, conduct, and report evaluations in accordance with accepted standards for ethical and high-quality evaluation as described by the Joint Committee on Standards for Education Evaluation (2012). Obtain institutional review board approval for all evaluation activities.

Planning

Decide on an appropriate scope and purpose for the evaluation, and plan and design it accordingly. For example, if a program is in an early stage of development, it may be most appropriate to focus on process evaluation and pursue outcome evaluation later. Ensure that the evaluation will be supported with adequate resources and persons with appropriate technical expertise. To help meet these needs, it may be possible to arrange for the evaluation to be conducted in partnership with a university-based researcher (faculty member or graduate student). Evaluations should be planned with foresight such that they will be able to conform to established standards for rigor in evaluation research (e.g., Standards of Evidence of the Society for Prevention Research).

Process evaluation

Assess both fidelity of the program’s implementation (extent to which it was implemented according to plan and design) and dosage (levels that individuals such as youth and mentors participated in or otherwise were exposed to different elements of the program). Keep in mind that fidelity likely will be appropriate to assess with respect to both staff- and mentor-implemented activities and that dosage will likely be important to assess with respect to services provided to both youth and their mentors. Examine all relevant aspects of the program implementation process, including adherence (whether intended services are provided according to plan or program model), exposure and quality of delivery (how much and how well services are provided), responsiveness (how clients as well as those tasked with program delivery respond to and experience the services), and (as applicable) program differentiation (the degree to which services provided can be distinguished from those associated with other similar programs or services).

Outcome evaluation

To help ensure accurate and unbiased estimates of the program’s effects, design evaluations to include a comparison group of youth who do not participate in the mentoring program. Especially when random assignment is not possible, special care should be taken to select the comparison group so that these youth are as similar as possible to those participating in the mentoring program. Obtain data from a large enough sample of youth to make it likely that program effects, if present, will be detectable. Select participating youth as well as program sites that are representative of the larger populations of youth and sites to which the results of the evaluation are intended to be generalized.

Program Evaluation  483

Topic

Recommendations Select outcomes to assess based on a careful review of the program’s logic model or theory of change, ensuring that both proximal and immediate and relatively more distal and indirect effects of the program are assessed. Utilize measures that have demonstrated evidence of being both reliable (producing consistent or dependable scores) and valid (measuring the constructs or outcomes that they are intended to assess). Consult available compendia to facilitate the search for psychometrically sound measures (for a list of available compendia and their online locations, see Wilson-Ahlstrom et al., 2011, p. 5). Whenever feasible, assess program outcomes at the conclusion of the minimum expected duration of the youth’s involvement as well as at later points in time that will be informative with respect to effects of longer periods of program involvement as well as those that may endure or emerge only after program participation is completed. Analyses of evaluation data should address several technical considerations, including (a) control for baseline (preintervention) differences between program and control or comparison group participants, (b) appropriate handling of missing data and potential bias introduced by attrition (loss of participants over the course of the evaluation), (c) use of statistical procedures that are aligned with the distribution or response scale of each outcome measure, (d) adjustment for any nonindependence that exists across observations for different youth (e.g., as a result of attending the same school), and (e) correction for chance findings that may arise due to conducting multiple tests of statistical significance across several outcomes. Analyses should test for possible differences in program effects across different subgroups of youth (e.g., male and female; i.e., moderation) as well as whether program effects are sequenced in a manner consistent with the program’s theory of change (i.e., mediation).

Linking process and outcome evaluation

Evaluations involving multiple program sites should investigate possible associations between differences in implementation and program effects. Associations between dosage and outcomes should be considered where feasible using the most rigorous methods possible. In all instances, such findings must be reported and interpreted with due consideration given their limitations and potential sources of bias.

Reporting and utilizing findings

Findings of evaluations, regardless of results, should be shared with all appropriate stakeholders. Reports should comprehensively present relevant aspects of findings, provide appropriate qualification when describing findings and their potential implications, offer alternative explanations for results, and highlight limitations of the evaluation as well as recommended next steps for future research. Reports of experimental and quasi-experimental evaluations should adhere to reporting guidelines of the CONSORT and TREND statements, respectively. Evaluation findings should be used to guide ongoing efforts to strengthen program processes and practices.

484   PRACTICE AND PROGRAMMATIC CONSIDERATIONS

Evaluation Standards It is essential that all evaluations adhere to accepted principles in the field. In the United States, the most influential of these are the Program Evaluation Standards established by the Joint Committee on Standards for Educational Evaluation (JCSEE). These standards were first published in 1981 and were most recently revised and updated in 2011. The standards are organized into five areas (Yarbrough, Shulha, Hopson, & Caruthers, 2011; see also the website of the JCSEE, 2012, for a summary of the standards). Utility standards are intended to ensure that an evaluation will serve the information needs of intended users. For example, the Attention to Stakeholders standard calls for attention to be given to the full range of individuals and groups invested in a program or affected by a program, while the Relevant Information standard states that care should be taken to ensure that evaluation information (e.g., data collected) is aligned with identified and emergent needs of such stakeholders. Feasibility standards are intended to ensure that an evaluation will be realistic, prudent, diplomatic, and frugal. For example, the evaluation procedures should be practical and responsive to the way the program operates, so as to keep disruption to a minimum while needed information is obtained. Propriety standards are intended to ensure that evaluations are conducted in ways that are proper, fair, legal, and just and with due regard for the welfare of those involved in the evaluation as well as those affected by its results. In general, for example, participation in evaluation activities should be voluntary and participants should be informed of all potential risks associated with their involvement. Another key standard in this area is that evaluations should openly and honestly identify and address any real or perceived conflicts of interest that could compromise the results of the evaluation. Most universities have an institutional review board (IRB) that is charged with reviewing and approving proposed research in relation to ethical and legal standards. Researchers based out of these settings must submit a description of their proposed research to the IRB, including procedures for ensuring informed consent of all participants, and may not initiate the research until approval is obtained. Less well recognized is the need for community-based organizations to obtain IRB approval for evaluation activities that they may undertake independently, if these are supported by state and/or federal funds (Bronte-Tinkew, Allen, & Joyner, 2008). In such situations, the organization likely will need to contract with an external IRB for the needed services.

Aside from the ethical imperative to obtain this type of external oversight of evaluation activities, failure to do so may leave a program or organization vulnerable to legal action (Bronte-Tinkew et al., 2008). Bronte-Tinkew and colleagues (2008) provide excellent practical guidance for programs on working with IRBs. Accuracy standards are intended to ensure the dependability and truthfulness of the content and findings of evaluations, especially those that bear on the quality or merit of the program being evaluated. The standard of Sound Designs and Analyses, for example, addresses the need for evaluations to utilize technically adequate designs that are aligned with their goals or purposes, while the standard of Communication and Reporting calls attention to the need for all reports and other communications about the evaluation to be inclusive of pertinent information and findings and to guard against misconceptions, biases, distortions, and errors. Finally, Evaluation Accountability standards focus on the importance of adequate documentation of evaluations and the need for accountability and ongoing efforts at improvement in evaluation processes and products. A thorough review of the JCSEE standards is beyond the scope of the chapter. It should be readily apparent, however, that they merit careful consideration throughout the process of conducting any evaluation of a youth mentoring program.

Evaluation Framework As noted, one of the most widely emphasized distinctions within frameworks for program evaluation is between process (or formative) and outcome (or summative) evaluation. Process evaluation is generally concerned with the extent and quality of a program’s implementation, including how it is experienced by participants, whereas outcome evaluation is focused on the impact that a program has on its participants and others. Clearly, the two are related. If a program is not implemented well or received positively by participants, it has few prospects for achieving its desired benefits. Accordingly, following separate overviews of key considerations in process and outcome evaluation, strategies for ensuring a strong relationship between them are addressed as well. It is important to note, however, that an evaluation need not encompass both process and outcome evaluation. In the early stages of a program’s development and implementation, for example, it may be advisable to focus solely on process evaluation (DuBois, Doolittle, Yates,

Program Evaluation  485 Silverthorn, & Tebes, 2006). Indeed, a key first step in any evaluation is to determine its primary purposes and to determine whether adequate resources exist to carry out an evaluation that will meet these goals. Throughout the presentation of the framework, the reader is referred to more in-depth treatment of different topics. Typically, the content of these resources is more technically demanding. There are, however, numerous other guides that offer more detailed nontechnical introductions to program evaluation both generally (see Basic Guide to Program Evaluation [Including Outcomes Evaluation], http:// managementhelp.org/evaluation/program-evaluationguide.htm, for one such resource and a listing of others) and for mentoring programs in particular (see the websites of the National Mentoring Center, http://educationnorthwest.org/nmc, and MENTOR: The National Mentoring Partnership, http://www .mentoring.org/). Interested readers without substantial background in evaluation or research methods are strongly encouraged to consult these resources. Process Evaluation Several frameworks have been described for use in guiding the process component of a program evaluation. These frameworks overlap extensively in their recommendations. It is widely recommended, for example, that attention be given to carefully assessing both program fidelity and dosage. Fidelity refers to the extent to which a program is implemented according to plan and design, whereas program dosage refers to the levels at which individuals participate in or otherwise are exposed to different elements of the program (Steckler & Linnan, 2002). A concern with understanding how the program is experienced by participants and others involved is a further common point of concern. It should be noted, however, that the process evaluation literature has been plagued by inconsistent terminology and definitions (Steckler & Linnan, 2002). At present, the best antidote to this situation is to carefully define evaluation terms and concepts rather than assume they will be understood in the same way by all audiences. Dane and Schneider (1998) advanced one widely referenced framework that addresses several core elements of process evaluation. These authors distinguished among five different components of the program implementation process: adherence,

1

quality of delivery, exposure, responsiveness, and program differentiation. Each of these components contributes important information to an evaluation of program process, including whether intended services are provided according to plan or program model (adherence), how much (exposure) and how well (quality of delivery) services are provided, how clients as well as those tasked with program delivery1 respond to and experience the services (responsiveness), and the degree to which services provided can be distinguished from those associated with other similar programs or services (program differentiation). Following Steckler and Linnan (2002), we add one further dimension, dose received. For present purposes, dosage is taken to refer to the extent to which individual participants actively engage with and use program resources (the authors’ definition is somewhat broader, encompassing receptiveness of participants, which is already covered under the concept of responsiveness). Several additional process-oriented facets of programs are frequently important to evaluate but are not addressed in depth here due to space constraints. These include (but are not necessarily limited to) aspects of the broader social, political, and economic environment that may influence program implementation; the program’s reach in terms of the proportion of the larger intended targeted audience or population (e.g., all youth meeting program eligibility criteria in a given school) that receives the intervention; and program costs (Steckler & Linnan, 2002). The interested reader is encouraged to consult chapters in this volume on program sustainability (Weinberger, Chapter 26), volunteer recruitment and retention (Stukas et al., Chapter 27), and economic evaluation (Foster, Chapter 34) for discussion of issues pertinent to evaluation of programs in these areas. Typically, it is necessary to collect and analyze information from several different sources in order to comprehensively and reliably assess all relevant components of program process. This is the case in part because different types of measures are likely to be best suited to evaluating different components. Also, multiple measures of the same process provide a key means of converging on a more accurate and comprehensive assessment of it (sometimes referred to as triangulation across measures). In keeping with research noted previously (DuBois et al., 2002), procedures for collecting program process data should ideally be an

Whereas Dane and Schneider (1998) focused on clients when considering responsiveness, we expand the concept here to include those responsible for program implementation, such as staff and mentors.

486   PRACTICE AND PROGRAMMATIC CONSIDERATIONS integral part of the program itself rather than “addon” activities that take place only in the context of a separate evaluation. In this way, data can be used to monitor and adjust program processes on an ongoing and “real-time” basis, thus facilitating continuous quality improvement. Adherence. Adherence involves assessing whether the activities intended to occur in the program actually take place. It is critical in this regard that the specific activities that comprise a given program be identified. This is an especially pertinent consideration for youth-serving mentoring programs given that these vary widely in their design, practices, and expectations for mentors. Ideally, a written “blueprint” of programmatic activities will be available against which to gauge adherence. For example, many mentoring agencies will have a policies and procedures manual to guide staff in program delivery as well as some form of written guidelines for mentors to utilize. As this suggests, for mentoring programs it is useful to distinguish between adherence to activities that are the responsibility of staff and those that fall within the role expectations of mentors. This assumes that mentors are not themselves program staff, which is typically the case. One criticism of mentoring programs for youth has been that the specific nature of expected mentor-youth activities has often not been specified adequately—that is, the so-called “black box” problem (see Cavell & Elledge, this volume, Chapter 3, and Tolan, Henry, Schoeny, & Bass, 2008). Developing welloperationalized standards for adherence as part of an evaluation plan is one way that programs can overcome this limitation. In general, evaluation of adherence involves a focus on whether a minimum threshold of delivery has been met, with assessments of exposure and quality being reserved for more refined and in-depth assessments. For example, whereas a program may be found to be adherent in the area of mentor training by virtue of offering a training with targeted content, the training as delivered may be relatively brief or offered only infrequently (exposure concerns) and/or may be delivered with little skill or enthusiasm (quality concerns).2 In terms of adherence as it relates to staff, relevant activities may include recruitment efforts,

training sessions, scheduled check-ins with mentors, youth, and parents, and agency-sponsored events, although clearly others are possible as well. For mentors, one key adherence consideration would be whether mentors, once paired with youth, make a concerted effort to engage them in activity, as it is not necessarily unusual for relationships to fall short of even this minimum criterion. Similarly, whether mentors continue to engage youth in activity over the minimum duration (e.g., 1 year) prescribed by the program is an important consideration (Grossman & Rhodes, 2002). For purposes of evaluating adherence, these types of data would be examined in terms of trends at the level of the overall program. As such, they can be distinguished from evaluations of dosage, which although often involving similar sources of data are focused on the experiences of individual participants. Optimally, procedures should be in place to objectively track the occurrence of key staff-led activities as a routine part of program operations. In this way, more accurate information can be obtained and timely adjustments can be made to bring practices in greater alignment with program design. Likewise, although information on mentor adherence is often not sought until the end point of an evaluation (e.g., Bernstein, Dun Rappaport, Olsho, Hunt, & Levin, 2009), these data should be obtained at regular intervals throughout the program, if possible. Several programs have had mentors submit logs of their activities with youth on a set schedule (e.g., monthly) for this purpose. Exposure and Quality of Delivery. Assessments of exposure and quality of delivery move beyond more fundamental considerations of adherence to consider the amount of services delivered and how well they are implemented, respectively. Within the context of youth mentoring program evaluations, these facets of program delivery are again likely to be important to assess with respect to both staff- and mentor-led activities. As in most programs, exposure may be assessed in a relatively straightforward manner by tracking the duration and the frequency with which activities are delivered. Illustratively, for staff-led activities, this could include keeping a record of how often mentors are contacted by staff

2 In practice, the dividing line between assessments of adherence and those focused on more refined issues of exposure and quality may be fuzzy. More important than how such judgment calls are made is that relevant data be collected and analyzed.

Program Evaluation  487 and how long these support contacts typically last.3 Likewise, mentors can be asked to share information periodically about the frequency and duration of their activities with youth. Accurately gauging the quality of service delivery is typically a more challenging undertaking. This is in part because judgments of quality are always subjective, thus placing a premium on triangulating across multiple sources of data. The data needed, furthermore, may require significant demands on staff and evaluators as they must go beyond relatively straightforward assessments of whether or not an activity occurred. Important aspects of delivery quality can include provider preparedness, use of relevant examples, enthusiasm, interaction style, respectfulness, confidence, and ability to respond to questions and communicate clearly (James Bell Associates, 2009). For youth mentoring programs, it is desirable to assess the quality with which all of the activities noted above as part of adherence (e.g., training sessions) are delivered. For staff-led activities, live observation or review of audio- or videotape recordings often may be feasible for assessing the quality of implementation. Review of written records, such as case notes, also may be useful. Big Brothers Big Sisters of America (BBBSA), for example, utilizes a structured tool to audit the quality of both enrollment and ongoing match support activities within its programs. Typically, resources will permit only a representative sample of activities to be assessed for quality in these ways. The BBBSA audit system, for example, involves taking a random sample of 10% of the matches that are under the purview of a given staff member. A further source of information on quality of program implementation that should not be overlooked are the reports of those on the receiving end of different activities, such as youth, mentors, and parents. Surveys of mentors, for example, may be a valuable source of information regarding

3

whether program staff demonstrate professionalism, offer constructive guidance, and convey appreciation in their contacts with mentors. Having staff report themselves on such aspects of their service delivery may be helpful as well not only for the data yielded, but also because doing so can facilitate useful forms of self-monitoring and reflection that contribute to improved services over time. Given the inherent potential for some degree of bias and distortion in these types of reports, it seems unwise to rely on them as a sole or primary source of information for gauging quality of program delivery. Yet, in the case of mentors, it may be a practical necessity to rely to a considerable extent on their own reports of how well they have delivered intended services and activities. To minimize potential for bias and incompleteness, such data ideally will be collected relatively close in time to when the activities took place. Furthermore, obtaining “thick” descriptions of activities and interactions may be more revealing and facilitate more accurate assessments than relying only on more structured “forced choice” forms of reporting. In keeping with the standards discussed previously, the timing and extent of data that a program attempts to collect from mentors in this area will need to reflect a level of burden that is prudent and feasible. As recipients of mentors’ efforts, youth also may be valuable sources of information pertaining to quality of delivery. This could include, for example, having youth report on the extent to which their mentors engaged with them in a collaborative (i.e., nonprescriptive) manner when selecting activities to do together (Keller, 2005). Clearly, staff, parents, and others (e.g., school staff) often may be able to shed light on such concerns. In view of the likely limitations of the preceding types of information sources, every effort should be made to capitalize on opportunities for direct observation of mentors’ interactions with youth. These may be relatively abundant

Some researchers also have surveyed mentors and other program participants to gain insight into the extent to which different types of staff-led program activities have occurred. Herrera and colleagues (Herrera, Kauh, Cooney, Grossman, & McMaken, 2008), for example, used aggregated reports from mentors to assess the degree to which training was provided by different Big Brothers Big Sisters of America (BBBSA) school-based mentoring programs. This approach may be less than ideal for at least two reasons. First, the information obtained will also address levels of participant engagement in activities, a separate component of program process (i.e., dose received), thus making it more difficult to distinguish reliably between fidelity (i.e., whether an activity was offered by the program) and dosage (i.e., whether participants took advantage of the opportunity to take part in an activity). Second, participants may not have accurate recall of their experiences or be able to identify specific activities that they have experienced. In the research by Herrera et al. (2008), for example, some mentors may not have recognized or registered their receipt of training because in many BBBSA agencies training is integrated into different aspects of the enrollment process (e.g., volunteer interviews) rather than conducted as a separate, stand-alone activity.

488   PRACTICE AND PROGRAMMATIC CONSIDERATIONS in the case of site-based programs in which staff are routinely present. Other possibilities may also present themselves, such as agency-sponsored events. In some instances, it may even prove feasible for a staff member to accompany a mentor on a community-based outing with a youth, thus providing the opportunity for a firsthand assessment of issues relating to program delivery quality from a mentor standpoint.

typically have with mentors and youth and, in many instances, with parents as well. If these contacts are experienced as overly intrusive or without a useful purpose, it may be difficult to execute this facet of a program successfully. The preceding considerations illustrate the potential for important interplay across different areas of concern within process evaluations of mentoring programs for youth, such as responsiveness and quality of service delivery.

Responsiveness. As noted earlier, responsiveness in the present discussion encompasses the opinions and feelings that both clients and those tasked with implementation have about different facets of a program. Key considerations include feasibility (whether intended users can carry out the activities, taking into account potential situational and other constraints), acceptability (whether activities are regarded by those implementing them and those to whom they are directed as being appropriate and consistent with their interests and priorities), and utility (the extent to which activities are viewed as useful and beneficial). In addition to overall ratings of the relative ease or difficulty of implementing different activities, assessment of feasibility may be enriched by asking about specific factors that facilitated or impeded implementation. In the context of youth mentoring program evaluations, the latter might involve asking staff whether they felt adequately trained to carry out different activities (e.g., mentor screening) and about situational factors that made implementation of these activities more or less feasible (e.g., competing job responsibilities, supervisor support). Similar questions could be asked of mentors with appropriate tailoring to their expected roles and (likely) volunteer status. Even if activities are generally regarded as feasible to implement, this is of course no guarantee that they will be or that, if they are, the quality of implementation will be high. For sufficient motivation to exist for implementation goals to be realized, it may be important for activities to be perceived as both being acceptable and having utility or value. Mentoring program staff, for example, who view a training curriculum as culturally insensitive may be reluctant to implement it. Likewise, mentors may be significantly more inclined to invest time and energy in carrying out suggested activities with youth if they expect the activities will strengthen mentoring relationships or foster desired youth outcomes. In addition to staff and mentors who are responsible for implementing various activities, it is likely to be informative to ask those on the receiving end for their views concerning acceptability and utility. Consider, for example, the support contacts that program staff

Program Differentiation. As noted previously, a key concern in the area of program differentiation is the extent to which the services being evaluated can be distinguished from other similar programs or services. This is likely to be of particular concern when the focus is on a program or set of practices that is intended to represent an enhancement or alternative to an existing program and its practices. Such evaluations typically fall under the umbrella term of “comparative effectiveness” studies. These types of evaluations can be important for the field’s development because they move beyond the basic question of “do mentoring programs work?” to the more refined or second-generation question of “which types of mentoring programs work best?” Illustratively, DuBois and colleagues (2008) described their evaluation of a mentoring program for girls, GirlPOWER!, that was designed to enhance the effectiveness of the BBBSA community-based mentoring program. Matches were assigned randomly either to participate in GirlPOWER! (and also receive all standard BBBSA services) or to receive only BBBSA standard programming. One key feature of the new program was the participation of girls and their mentors in a series of workshops that addressed topics important to the development of girls as they transition to adolescence, such as self-esteem and maintaining healthy peer relationships. To address program differentiation, it would be important in such an evaluation to assess not only whether the workshops were implemented and thus delivered to those matches assigned to participate in GirlPOWER! (i.e., adherence), but also the degree to which similar workshops or activities may have been offered as a part of the agency’s typical practices and thus available to matches in the comparison group. Furthermore, where such workshops occurred as part of standard programming, it would be useful to examine the extent of overlap in content and approach and other potentially important characteristics (e.g., whether they were delivered to the same group of matches over time as in GirlPOWER!). Program differentiation also may be important to factor in to an evaluation even when the

Program Evaluation  489 program or practices serving as a point of comparison do not constitute what would typically be regarded as formal youth mentoring. This is particularly likely to be the case when there is reason to expect that mentoring activity and/or relationships, although not an explicit or planned feature of the other program, could nonetheless be a salient part of what occurs in that intervention. Such potential may seem most evident for programs in which adults are paired to work with youth in one-on-one or small-group contexts, such as tutoring or skills training. Yet, fostering supportive ties and interactions between adults and youth is a goal common to many programs. It appears, furthermore, that such experiences are often a significant factor in accounting for program effectiveness (see, for example, Mekinda & Hirsch, this volume, Chapter 15, with respect to this potential within after-school programs). These considerations suggest that program differentiation should be investigated across a broad range of interventions that might be the focus of comparative effectiveness research involving youth mentoring programs. More generally, it should be kept in mind too that substantial numbers of youth in “no treatment” control or comparison groups may be exposed to mentoring through their ad hoc participation in programs with mentoring elements as well as through more informal circumstances. Illustratively, in the large-scale randomized trial evaluation of the U.S. Department of Education’s Student Mentoring Program, roughly one-third of youth in the control group (35%) reported participating in a formal mentoring program at some point during the study (Bernstein et al., 2009); a similar portion (34%) of youth in the control group of the recent randomized trial evaluation of the BBBSA school-based mentoring program (Herrera, Grossman, Kauh, Feldman, & McMaken, 2007) reported meeting with “an adult or older student mentor, ‘buddy’ or ‘big’” during the study. Such findings suggest that it has become relatively commonplace for youth in our society to be exposed to some form of mentoring. For purposes of informing assessments and understanding of program effects, evaluations of youth mentoring programs should take careful stock of such experiences among all participating youth—that is, both those in control or comparison groups and those receiving the program under study (Grossman, 2005; Wheeler, Keller, & DuBois, 2010). Dosage. As discussed earlier, the final element of process evaluation to be addressed, dosage, refers to the levels of involvement and exposure that

individual participants have in relation to different program activities. The types of data likely to be required to assess levels of dosage are the same as those described previously in the context of considering evaluation of adherence, exposure, and quality of service delivery. Because they exist at the level of individual participants, dosage indices can be useful for purposes of exploring potential variation in the extent to which program services are being delivered to (or taken advantage of by) different subgroups of participants. As is discussed later, measures of dosage also have valuable potential applications in the context of efforts to link process and outcome components of evaluations. For youth mentoring programs, dosage is likely to be important to consider from the perspective of not only youth, but also mentors and any other recipients of services (e.g., parents). Illustratively, for youth, relevant indicators of dosage could include measures of the frequency and amount of mentor-youth contact, relationship duration, and extent of engagement in targeted types of activities with mentors. For mentors, dosage indicators might include amount of training received preand postmatch, frequency of support contacts from staff, and extent of coverage of pertinent content in such activities. More refined indices that take into account quality of delivery also may be valuable. This could be accomplished through a variety of approaches, such as weighting more objective indices (e.g., support contacts with staff) by the assessed quality with which the activities involved were implemented. Outcome Evaluation Typically, outcome evaluation is concerned with accurately gauging the extent to which a program has influenced one or more outcomes for participants in the intervention. Rigorous outcome evaluation requires careful attention to myriad methodologic concerns (Grossman, 2005). These issues can be classified broadly within the domains of study design and sampling, measurement, data analysis, and reporting and interpretation of findings (DuBois & Silverthorn, 2005). Carrying out an evaluation that successfully navigates this terrain will almost inevitably require a substantial investment of resources as well as leadership from one or more persons with appropriate technical expertise (e.g., doctoral- or specialized master’slevel training in evaluation methods). The following sections provide a high-level overview of key considerations within each of the major areas noted above. The presentation is by intention relatively

490   PRACTICE AND PROGRAMMATIC CONSIDERATIONS brief and nontechnical.4 More detailed and technical discussions of the issues covered here as they relate to evaluation of youth mentoring programs in particular can be found elsewhere (DuBois et al., 2006; DuBois & Silverthorn, 2005; Grossman, 2005). Study Design and Sampling. Essential considerations in the area of study design and sampling revolve around the issues of how best to structure the evaluation to ensure the potential for valid, accurate, and generalizable conclusions about program effects. With regard to validity (more specifically, what researchers typically refer to as internal validity), to achieve a desirable level of rigor it typically will be necessary to include in the study design a comparison group of youth who do not receive the intervention under consideration. Grossman (2005) summarized the logic behind this principle as follows: A program’s impact can be gauged accurately (i.e., be internally valid) only if one knows what would have happened to the participants had they not been in the program. This hypothetical state is called the “counterfactual.” Because one cannot observe what the mentees would have done in the absence of the program, one must identify another group of youth, namely a comparison group, whose behavior will represent what the participants’ behavior would have been in the absence of the program. Choosing a group whose behavior accurately depicts this hypothetical no-treatment (or “counterfactual”) state is the crux to getting the right answer to the effectiveness question, because a program’s impacts are ascertained by comparing the behavior of the treatment or participant group to that of the selected comparison group (pp. 255–256). Grossman (2005) illustrated the relevance of this principle to youth mentoring program evaluations by pointing out that the landmark Public/ Private Ventures (P/PV) evaluation of the BBBSA community-based mentoring program, had it not included a comparison group of nonmentored youth, would have likely indicated the program to be ineffective (and perhaps even harmful) with regard to affecting substance use due to approximately 7% of the mentored youth (aged 10–14 years) reporting initiation of substance use during the 18-month period of the evaluation. The comparison group

4

data, which revealed a larger rate of drug use onset for nonmentored youth (11%), allowed the evaluators to detect a statistically significant and beneficial effect of the program in this area, suggesting that it was able to stem some of the naturally occurring increases in drug use that occur as youth transition into adolescence. It is critically important for comparison groups to be similar to intervention participants prior to initiation of their involvement in the program under evaluation, especially with regard to any characteristics that could influence outcomes that are the focus of the evaluation. Random assignment of participants to the intervention and control groups (referred to as an experimental design) is widely agreed to be an ideal method of ensuring this type of comparability (Grossman, 2005). When the comparison group is selected or determined through an alternative process (referred to as a quasi-experimental design), the evaluator must demonstrate that the group is sufficiently similar to the intervention group for a comparison of outcomes across the two groups to provide a basis for valid inferences about program impact. Systematically selecting comparison group participants to be “matched” to those in the intervention group is a common strategy for working toward group equivalence. Differences on characteristics of the groups that are measured at baseline (preprogram) also can be controlled statistically. Still, there often will be a potential for consequential differences to exist in other areas that were not measured (and thus were not factored in to a matching process), leaving study inferences subject to unknown degrees (and directions) of bias. Suppose, for example, that instead of random assignment, the P/PV evaluation discussed earlier had utilized a comparison group of youth who were matched with intervention group youth on factors such as age, gender, race/ethnicity, socioeconomic status, and baseline levels of key outcomes of interest such as problem behavior involvement and self-esteem. Although laudable, such matching would still leave open the potential for consequential differences between the two groups in other areas. Of particular note is that youth whose parents demonstrate the motivation to engage them in a mentoring program and those youth who convey the requisite level of openness to this type of experience to be approved for services may differ in meaningful, but difficult to

As we have noted, it may be useful and important in evaluation of youth mentoring programs to examine evidence of impact on mentors and others such as parents. Due to space constraints, however, the present discussion is limited to consideration of evaluation of outcomes for youth, which will be of primary concern in most circumstances.

Program Evaluation  491 measure ways from those who otherwise may appear similar to them (e.g., greater levels of parental commitment or youth motivation for personal growth). In the special case of comparative effectiveness studies in which the focus is on the relative effectiveness of two or more alternative forms of mentoring, a further important concern is whether groups are determined so that their program experiences are comparable to one another in all areas other than those that are intended to differ. This often will need to entail taking steps to ensure that the mentors assigned to youth in different groups have similar backgrounds and abilities. One strategy for achieving “mentor equivalence” is to randomly assign youth to study conditions after they have been matched with the best available mentor. This is the approach taken in the evaluation of the GirlPOWER! program discussed previously (DuBois et al., 2008). Even with this type of safeguard, though, complications may arise. Consider, for example, that a youth’s relationship with his or her initially assigned mentor may end at some point during the study and a new mentor must be found. It may be impractical in such instances to shield program staff from the study condition to which the youth is assigned. This introduces a risk for selection bias and thus systematic differences in the mentors who are assigned to youth across conditions. It should be noted, too, that the preceding concerns are equally relevant with regard to the program staff who are assigned to support the delivery of different forms of mentoring to youth. Thus, when comparing a new form of mentoring to “business as usual” (standard services), it may be desirable to randomly assign which staff will be responsible for implementation of the new practices. Clearly, as this discussion illustrates, evaluation designs in comparative effectiveness studies for youth mentoring may become quite involved. In an evaluation that the author of this chapter is currently co-leading, for example, of the incorporation of practices based on the Step-It-Up-2-Thrive model (http://www.stepitup2thrive.org/) into the community-based mentoring program of BBBSA, random assignment (to the Thrive model or standard practices) is occurring at the levels of both participating youth, after matching with best available mentors, and at the level of staff, so as to ensure that differences in their backgrounds and abilities do not distort the evaluation’s findings. It is also important for samples (i.e., the groups of participants involved in evaluations) to be of adequate size and representativeness (DuBois & Silverthorn, 2005). Sample size has to do with the concept of what researchers refer to as statistical

power, which refers to the likelihood that a statistically significant result will be obtained when testing an association or effect that is present in the larger population. When power is limited, there is an increased risk of concluding that a program effect does not exist when in fact it does (referred to commonly as a Type II error). Because the effects of mentoring programs tend to be relatively small by most standards, there typically will be a need for substantial numbers of youth to be involved in an evaluation (e.g., several hundred) in order to have a high level of confidence that impacts on outcomes of interest will not be missed (for further discussion, see DuBois & Silverthorn, 2005). Sample representativeness has direct implications for the extent to which the findings of an evaluation are likely to generalize to a larger population of interest (referred to by researchers as external validity). Illustratively, returning to the P/ PV evaluation, youth participants who met various eligibility requirements, such as age, were sampled from eight participating BBBSA affiliates (Tierney, Grossman, & Resch, 1995). The findings obtained may thus not be representative of the program’s effects on younger or older youth served by the program or those with characteristics or backgrounds (e.g., rural) not typical of the youth served by the participating affiliates. It is important to keep in mind, too, that issues of representativeness extend to programmatic considerations (Valentine & Cooper, 2008). Within the realm of youth mentoring, for example, concern has been raised regarding the extent to which findings of evaluations such as those of the BBBSA community- and schoolbased mentoring programs can be generalized to the much larger and quite diverse range of mentoring programs (Boyle, 2007; Wheeler et al., 2010). Furthermore, because, as is often the case, participating affiliates were not selected at random to be involved in these studies, the findings obtained may not be representative even of those that occur within the broader network of BBBSA affiliates. We could add to this discussion the concern that the results of an evaluation may not generalize to future points in time due to shifts in youth served by program or program practices. In the P/PV evaluation of the BBBSA community-based mentoring program, for example, 6 of the 8 participating affiliates required mentors to meet with their mentees 3 hours or more on a weekly basis over the first year of the relationship. Today, expectations appear to have shifted noticeably downward (e.g., one outing every other week), thus potentially lessening the dosage of mentoring that youth receive in the program (Herrera, DuBois, & Grossman, 2013) and, in turn,

492   PRACTICE AND PROGRAMMATIC CONSIDERATIONS its effects on their outcomes. Simply put, even when other aspects of representativeness have been attended to carefully in the design of the research, it should be kept in mind that evaluation findings may not necessarily have a long “shelf life” due to developments that occur outside the context of the research itself. One challenge specific to evaluation of mentoring programs with respect to sample size and representativeness is that youths’ mentors too often may need to agree to be part of the research. In the context of a comparative effectiveness evaluation, for example, mentor consent may be necessary because mentors too stand to be exposed to new (“experimental”) program practices. Mentor consent also may be ethically required in an evaluation because youth will be asked to report on the behaviors of their mentors or because information about mentors will be collected from agency records. Under such circumstances, study participation may be restricted to those youth whose mentors also elect to be part of the research. If substantial numbers of youth must be excluded for this reason, it may become difficult to recruit a sufficiently large and representative sample of participants for an evaluation. Measurement. For measurement, key concerns for outcome assessment include the psychometric strength, coverage, and timing of measures (DuBois & Silverthorn, 2005). Measures, first and foremost, must have adequate psychometric properties to support their reliability and validity. Reliability refers to the extent to which a measure yields consistent scores when expected to do so (e.g., two administrations over a brief period of time), whereas validity pertains to whether the scores on a measure can be taken as reflective of the underlying construct (e.g., social skills) that it is intended to assess. There are multiple forms of both reliability and validity, consideration of which is beyond the scope of the present discussion (a nontechnical overview of different types of reliability and validity and their importance appears in Wilson-Ahlstrom, Yohalem, DuBois, & Ji, 2011, which is available online). An important consideration is the nature of the samples and methods that were used to generate evidence of the reliability and validity of a measure. To the extent that these are not mirrored in the design of any given evaluation, it should not be assumed that previously reported psychometrics will apply. Illustratively, a scale assessing prosocial behavior used in a mentoring program evaluation may have been “validated” with teenagers using research assistants to administer the measure, whereas the evaluation may call for having younger children complete the measure

when administered by program staff. Similarly, although it is common practice to adapt measures in various ways (e.g., using only selected items, rewording items or instructions), such modifications necessarily introduce an element of uncertainty into claims of reliability and validity. In view of the preceding types of considerations, it is advisable (and often required by journal editors!) to also collect and report data that address the psychometric properties of measures within the framework of their use within a particular evaluation. At a minimum, this will typically involve reporting estimates of internal consistency reliability (a measure of reliability that focuses on the consistency of responses across items comprising a given scale) for the overall sample used in one’s evaluation as well as relevant demographic subgroups of youth within the sample (e.g., male and female). A final note regarding psychometrics is that it is highly desirable, whenever feasible, to obtain multiple measures of a given outcome of interest in an evaluation (DuBois & Silverthorn, 2005). As noted earlier, triangulation across multiple measures can provide a valuable means of converging on a reliable and valid assessment of a construct. This is especially likely to be the case when measures differ from one another along dimensions such as informant (e.g., youth vs. teacher) and administration format (e.g., survey vs. objective test) such that when considered together their strengths can be complementary and their limitations offsetting. With respect to coverage, it is important for measures used in evaluations to address the full range of outcomes of interest. These should include both outcomes that are hypothesized to be relatively proximal and direct effects of the program as well as those that are distal and indirect. Assessing the full range of outcomes in this manner will facilitate analyses that provide tests of a program’s “theory of change.” It is also the case that program effects on outcomes that are expected to be a relatively direct result of program participation will typically occur sooner and be of relatively larger magnitude, thus enabling them to be detected more easily and on a shorter timetable. On the other hand, measurement of more distal outcomes may be critically important for advancing understanding of the extent to which a program is having an impact in areas that are of particular interest and concern to those in policy and funding roles. A relative lack of attention to these types of outcomes has been highlighted as a limitation of most youth mentoring evaluations carried out to date (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). Developing a program logic model is one useful strategy for ensuring

Program Evaluation  493 that the appropriate range of outcomes is targeted for assessment within an evaluation of a youth mentoring program (for a useful guide on constructing a logic model, see Guidelines and Framework for Designing Basic Logic Model, http://managementh elp.org/evaluation/logic-model-guide.htm; for illustration of a logic model for a youth mentoring program, see Tools to Establish Evaluation Methods and Criteria on the MENTOR/National Mentoring Partnership website, http://www.mentoring.org/pro gram_resources/elements_and_toolkits/tool_kit/ eval/). Typically, logic models distinguish among immediate, intermediate, and longer-term outcomes, thus facilitating attention to the full range of potentially relevant outcomes for a program. A further important consideration relating to coverage involves measurement of factors that could be expected to moderate (i.e., change the size and/or direction of) program effects. Rhodes (2005) has proposed a model in which a range of different facets of the backgrounds and characteristics of the youth, including interpersonal history, social competencies, developmental stage, and family and community context, have the potential to moderate the effects of mentoring relationships on outcomes (e.g., grades, emotional well-being, behavioral). The specific factors assessed in any given evaluation should be informed by consideration of the program’s theory of change. In many instances, this list will need to be prioritized so as to avoid undue costs and respondent burden. Still, judicious attention to measurement of plausible moderators will set the stage for analyses to probe beyond analyses of overall impact to test for differences across subgroups of youth. Such analyses are increasingly appreciated as important within the broader area of prevention science (Flay et al., 2006; Wang & Ware, 2013) and also have been demonstrated to be informative in several evaluations of youth mentoring programs specifically (see, e.g., Karcher, 2008; Schwartz, Rhodes, Chan, & Herrera, 2011). Timing of measurements is also an important consideration. At a minimum, where feasible, outcomes should be assessed at least twice in any evaluation, once at a preintervention baseline and then again following program completion. Unlike the majority of programs undergoing evaluation, however, for youth mentoring programs there is often not an unambiguous “end” of the program. Consider, for example, programs in which mentors are asked to commit to a minimum duration of time (e.g., 1 year), but in which they are encouraged and supported in continuing their relationships with youth for much longer periods of time (e.g., until youth age out of the program). Complicating matters

still further, youth whose initial relationships with their mentors end often are then rematched with new mentors, thus extending the duration of their involvement with the program. At present, there are no agreed-upon approaches with regard to how such situations should be handled. Assessing outcomes only at the end of the minimum desired time period for mentoring relationship duration or program involvement may fail to capture the effects of longerterm mentoring experiences that youth may have in the program. On the other hand, delaying the postprogram assessment runs the risk of not capturing effects of the program that may dissipate over time in the absence of mentoring for those who experience only the minimum dosage. Similarly, delayed assessments run the risk of loss of participants to follow-up (i.e., referred to as attrition), thus potentially compromising results of the evaluation. In view of these competing considerations, it is advisable to include both types of assessments, when feasible. Illustratively, in the P/PV evaluation of the BBBSA school-based mentoring program (Herrera et al., 2007), youth were assessed both at the end of the initial school year (minimum targeted length of program involvement/mentoring relationships) and then again the following school year. Beyond structuring timing of assessments with respect to these types of considerations, it is also highly desirable to include one or more assessments that occur far enough in the future to provide insight into whether program effects endure (or, in some cases, even emerge) following the end of program participation. As noted earlier, follow-up assessments such as these also may be essential for informing an understanding of the extent to which programs are having an effect on outcomes with greatest interest to policy and funding audiences (e.g., high school graduation, employment, health outcomes during adulthood). Several compendia describing measures for possible use in evaluations of youth programs have been developed (for one such resource and a listing of several others, see Wilson-Ahlstrom et al., 2011). Such resources can be of tremendous value, greatly reducing the significant time and effort that frequently is required to identify suitable measures for use in an evaluation. As the preceding discussion makes clear, however, measurement decisions ultimately need to be anchored to a careful consideration of the goals and needs that are specific to any given evaluation. Data Analysis. Rigorous, state-of-the-art data analysis within an outcome evaluation of a program typically requires application of advanced statistical procedures. Increased appreciation of the need

494   PRACTICE AND PROGRAMMATIC CONSIDERATIONS for application of state-of-the-art methods has advanced in this area and is encompassed by the concept of statistical conclusion validity (Valentine, 2009). In the present context, this form of validity would refer primarily to whether data are analyzed in a manner that yields accurate inferences about a program’s effects on outcomes. It calls attention to the reality that regardless of the strength or rigor of all other aspects of an evaluation (e.g., use of a randomized controlled design, valid measures), an evaluation can become flawed if appropriate statistical procedures are not used to analyze the data that are obtained. Among the issues that may be important in this regard are control for baseline (preintervention) differences between program and control or comparison groups, appropriately taking into account missing data5 and attrition (loss of participants over the course of the evaluation), use of statistical procedures that are aligned with the observed distribution or response scale of each outcome measure, adjustment for any nonindependence that exists across observations for different youth, and correction for chance findings that may arise due to conducting multiple tests of statistical significance. Illustratively, in the P/PV evaluation of the BBBSA schoolbased mentoring program (Herrera et al., 2007), analyses controlled for baseline differences between intervention and control groups on the baseline score for the outcome measure being predicted, a range of youth demographic and background variables (e.g., gender, race/ethnicity, free/reduced lunch status, number of stressful life events experienced in the past 6 months) and for an additional outcome measure (substance use) that was found to differ significantly between control and intervention groups at baseline. With regard to missing data and attrition, youth for whom data could not be collected at follow-up assessments (“attriters”) were significantly “needier” at baseline (e.g., higher levels of reported stress, poorer grades) than those with follow-up data (“non-attriters”). It is important to note, however, that the types of youth who attrited from the control group did not differ from those who attrited from the treatment group at baseline (at least on the measures that were available). The lack of evidence of “differential attrition” suggested that conclusions about effects of the program on outcomes would not be biased due to a selective of loss of some types of youth from either the intervention or the control group. Different statistical proce-

5

dures—ordinary least squares regression or logistic regression—were used depending on whether the outcome being tested was continuous (scores took on a range of values) or dichotomous (e.g., yes/no), respectively. Analyses adjusted for the fact that youth within the evaluation sample were grouped by school by including a school-level error component, such that scores on outcome measures for different youth within a given school were not assumed to be fully independent from one another. Finally, tests of significance for outcomes within a given domain (e.g., academic performance) were adjusted (using the Benjamini-Hochberg procedure) so that the risk of incorrectly concluding that a program effect was present (referred to as a Type I error) within each set of outcomes was limited to 10%. As alluded to previously, it is also important for data analyses to test for potential differences in program effects across different subgroups of youth (i.e., moderation) as well as whether program effects are sequenced in a manner consistent with the program’s theory of change (i.e., mediation). In the P/PV evaluation (Herrera et al., 2007), analyses tested for differences in program impact as a function of youth gender, race or ethnicity, grade in school, and academic performance. In a subsequent report, structural equation modeling analyses was used to test a model in which program effects on academic outcomes were mediated by more immediate impacts in other areas (e.g., reported availability of support from a significant nonparental adult; Silverthorn, DuBois, Herrera, & Kauh, 2010). Linking Process and Outcome Evaluation As noted earlier, it is also important to consider opportunities for linking process and outcome components of program evaluation. In a multisite evaluation, for example, it may be useful to consider the ways in which estimated effects of the program vary in association with indicators of implementation fidelity or quality. In the P/PV evaluation of the BBBSA school-based mentoring program, for example, effects on youth outcomes were found to be stronger at those program sites where mentors reported receiving greater levels of staff support and greater access to school resources and space (Herrera et al., 2007). Such associations between implementation differences and program effects are correlational and thus do not establish

Approaches to handling missing data in studies is a large and involved topic in itself, even the basics of which are beyond the scope of this chapter. For an excellent primer, see Enders (2010).

Program Evaluation  495 the implementation indicator involved as having caused the differences in program effects across sites. These types of results can nonetheless serve as a valuable source of evidence in guiding ongoing efforts to improve mentoring programs and their implementation. Variation in indices of dosage at the level of individual participants also may be useful to examine in relation to youth outcomes. Yet observed differences in dosage may reflect variations not only in program delivery efforts (e.g., staff and mentors), but also in characteristics and behaviors of the youth involved. Consider, for example, that lower levels of mentor adherence to program protocols for frequency of contact and engaging in different types of activities with mentees may, in part, reflect challenges that youth and their surrounding life circumstances bring with them. Grossman and Rhodes (2002) found that several different characteristics of youth and their family backgrounds (e.g., youth being referred after program intake for psychological testing, number of family moves) predicted a greater likelihood of mentoring relationships terminating. Consequently, any dose-outcome associations that are found may reflect partly, or possibly even entirely, factors other than the “dose” of implementation. Similar potential for bias occurs when comparing outcomes between those with a given dosage level in the program group to control group participants. Grossman (2005) provided an excellent discussion of these types of “suspect comparisons” as well as two approaches to minimizing the risk for bias in such comparisons: two-staged least squares (TSLS) and propensity score analysis (see also Gennetian, Magnuson, & Morris, 2008, for recent work on the use of random assignment status as an instrumental variable in TSLS). Within the mentoring literature, an illustrative application of TSLS can be found in Grossman and Rhodes (2002) and similarly for propensity score analysis in Herrera et al. (2013). Reporting and Utilization of Evaluation Findings The final stage of any evaluation involves reporting and making appropriate use of the findings. A number of concerns must be addressed in the reporting process. These include comprehensively presenting relevant aspects of findings, providing appropriate qualification when describing findings and their potential implications, offering possible alternative explanations for results, and highlighting potential limitations of the evaluation as well as recommended next steps for future research. Lack of

optimal reporting in any of these areas can have significant consequences. Illustratively, in the area of youth mentoring, considerable confusion resulted from the original presentation of findings from P/PV’s evaluation of the BBBSA community-based mentoring program (Tierney et al., 1995) because information regarding the magnitude of the observed program effects (referred to as effect size) was not reported in the format necessary for “benchmarking” against available interpretive guidelines and results of other evaluations. When ultimately gauged in this manner, the effects of the program appeared to be relatively modest in magnitude (see DuBois et al., 2002, footnote 5). Several sets of guidelines have been developed to promote more uniform and comprehensive reporting of the methodology and findings of program evaluations. Most notably, these include the CONSORT (Consolidated Standards of Reporting Trials) Statement (Schulz, Altman, Moher, & CONSORT Group, 2010; see also Moher et al., 2010), which provides minimum criteria for reporting of findings from randomized controlled trials (see also the companion TREND [Transparent Reporting of Evaluations with Nonrandomized Designs] Statement [Des Jarlais, Lyles, Crepaz, & the TREND Group, 2004] for reporting of nonrandomized evaluations). Embedded in such guidelines are both implicit and explicit standards for carrying out a high-quality evaluation. Accordingly, evaluators are well advised to consult these guidelines carefully not only at the point of report writing, but also in the planning stages of an evaluation so as to ensure that all relevant design and methodology requirements have been anticipated. Other resources likely to prove valuable for purposes of both evaluation planning and reporting include the Society for Prevention Research’s Standards of Evidence (Flay et al., 2006) and the Study Design and Implementation Assessment Device (Valentine & Cooper, 2008). It is important that the findings of an evaluation be put to use in a responsible manner (Yarborough et al., 2011). Too often, results may simply be taken as a thumbs-up (or thumbs-down) endorsement of a program’s quality or effectiveness. To do so overlooks the much richer and more nuanced information routinely produced by a well-conducted evaluation. These details often provide insights into the specific program aspects that are being implemented more or less well, which participants are receiving the most and least adequate dosages of the program’s activities, the ways in which linkages or pathways across different types of outcomes do or do not conform to the program’s theory of change, and a host of similar considerations.

496   PRACTICE AND PROGRAMMATIC CONSIDERATIONS The high-profile evaluations of the BBBSA community- and school-based mentoring programs provide an exemplar of the uses of evaluation results within the field. The results of the earlier evaluation of the organization’s community-based program seem to have held greatest sway in terms of providing a rationale for accelerated growth of mentoring both within and outside the organization (Boyle, 2007). Ironically, these efforts to “go to scale” entailed a watering down of various program elements (e.g., mentor training and support) that there is now good reason to believe may have been key conditions underlying the program’s observed effectiveness in the evaluation trial (Rhodes & DuBois, 2006). In contrast, although the more recent evaluation of the school-based program model yielded similarly encouraging results, the use of the findings (at least as far as BBBSA is concerned) has been quite different. Specifically, instead of focusing on the results as validation of the program’s effectiveness, the organization has devoted most of its attention and resources to using the findings to strengthen the program’s core practices and the procedures that need to be in place to ensure that these are implemented with optimal fidelity (Wheeler et al., 2010).6

this chapter, too, that the process of evaluating a youth mentoring program brings with it a set of additional requirements and challenges that are not typically encountered in evaluations of other types of programs. In view of the foregoing considerations, organizations that are interested in mounting a full-scale evaluation of one of their mentoring programs are typically well advised to look externally, not only to secure needed resources (e.g., grant funding) but also to arrange for the involvement of researchers or evaluators with appropriate technical expertise. At the same time, it would be a mistake to overlook viable opportunities for smaller scale, and thus more feasible, evaluation efforts (e.g., those focused solely on process evaluation data). Such efforts can be expected to yield a strong return on investment on several fronts. For one thing, such activities can be instrumental in laying the foundation necessary to support more ambitious evaluation initiatives in the future. Perhaps of even more fundamental significance is the promise that well-conceived and properly executed evaluation activities hold for fostering a culture of organizational learning and improvement in which candid, data-based reflection on program processes and practices is the norm.

Conclusion

References

An overarching goal of this chapter was to convey practical guidance and recommendations for evaluation of youth mentoring programs and to do so in terms that would be accessible to readers without specialized training in evaluation or research methods. As the presentation has made clear, however, the process of designing, conducting, and reporting an evaluation is in many ways an inherently complex, multifaceted, and technically demanding undertaking. A range of formidable obstacles— including those relating to capacity, required partnerships, ethical and effective engagement of participants in evaluation activities, measurement of outcomes that often defy easy operationalization, and ensuring that a receptive environment exists for making use of results—frequently make successful evaluation a challenge within the nonprofit sector (Gilbert, 2012). It can be seen from

Bernstein, L., Dun Rappaport, C., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S. Department of Education’s Student Mentoring Program (NCEE 2009-4047). Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education. Big Brothers Big Sisters of America. (2009). Enhanced school-based mentoring preliminary findings and high school Bigs. Unpublished document, Big Brothers Big Sisters of America. Boyle, P. (2007, January 1). The study that ignited (or diluted) mentoring: How the “Big Brothers” report became the most influential, most useful and most misused evaluation in youth work. Youth Today. Retrieved from http://www.youthtoday.org Bronte-Tinkew, J., Allen, T., & Joyner, K. (2008). Institutional review boards (IRBs): What are they and why are they important? (Publication 2008-09). Washington, DC: Child Trends. Retrieved from

6 Informed by findings of the evaluation, BBBSA recommended that its affiliates consider adopting a range of practices aimed at enhancing the program’s effectiveness in areas such as volunteer training, mentoring relationship duration, and support for relationships during the summer months. These recommendations were subsequently incorporated into a revised school-based program model that was piloted and evaluated in 23 BBBSA agencies (BBBSA, 2009).

Program Evaluation  497 http://www.childtrends.org/Files/Child_Trends -2008_02_19_Evaluation7IRBs.pdf Dane, A. V., & Schneider, B. H. (1998). Program integrity in primary and early secondary prevention: Are implementation effects out of control? Clinical Psychology Review, 18, 23–45. Des Jarlais, D. C., Lyles, C., Crepaz, N., & the TREND Group. (2004). Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions: The TREND statement. American Journal of Public Health, 94, 361–366. DuBois, D. L. (2007). Effectiveness of mentoring program practices. In J. E. Rhodes (Ed.), Research in Action Series (Issue 2, pp. 1–15). Alexandria, VA: MENTOR/ National Mentoring Partnership. Retrieved from http://www.mentoring.org/access_research/research_ in_action/research_in_action_series/ DuBois, D. L., Doolittle, F., Yates, B. T., Silverthorn, N., & Tebes, J. K. (2006). Research methodology and youth mentoring. Journal of Community Psychology, 34, 657-676. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., & Silverthorn, N. (2005). Research methodology. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 44–64). Thousand Oaks, CA: Sage. DuBois, D. L., Silverthorn, N., Pryce, J., Reeves, E., Sánchez, B., Silva, A., . . . Takehara, J. (2008). Mentorship: The GirlPOWER! program. In C. W. Leroy & J. E. Mann (Eds.), Handbook of prevention and intervention programs for adolescent girls (pp. 325–365). Hoboken, NJ: Wiley. Enders, C. (2010). Applied missing data analysis. New York: Guilford Press. Flay, B. R., Biglan, A., Boruch, R. F., González Castro, F., Gottfredson, D., Kellam, S., . . . Ji, P. (2006). Standards of evidence: Criteria for efficacy, effectiveness and dissemination. Prevention Science, 6, 151–175. Gennetian, L. A., Magnuson, K., & Morris, P. A. (2008). From statistical associations to causation: What developmentalists can learn from instrumental variables techniques coupled with experimental data. Developmental Psychology, 44, 381–394. Gilbert, L. (2012). Five hurdles to non-profit performance assessment. Cambridge, MA: Center for Effective Philanthropy. Retrieved from http://www .effectivephilanthropy.org/blog/2012/09/five-hur dles-to-nonprofit-performance-assessment/ Grossman, J. B. (2005). Evaluating mentor programs. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of

youth mentoring (pp. 251–265). Thousand Oaks, CA: Sage. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Herrera, C., DuBois, D. L., & Grossman, J. B. (2013). The role of risk: Mentoring experiences and outcomes for youth with varying risk profiles. New York: A Public/Private Ventures project published by MDRC. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J. (with Jucovy, L. Z.). (2007). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Herrera, C., Kauh, T. J., Cooney, S. M., Grossman, J. B., & McMaken, J. (2008). High school students as mentors: Findings from the Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia: Public/Private Ventures. James Bell Associates. (2009, October). Evaluation brief: Measuring implementation fidelity. Arlington, VA: Author. Retrieved from http://www.jbassoc.com/ reports/documents/evaluation%20brief%20-%20 measuring%20implementation%20fidelity_ octob%E2%80%A6.pdf Joint Committee on Standards for Education Evaluation. (2012). Program evaluation standards statements. Retrieved from http://www.jcsee.org/programevaluation-standards/program-evaluation-stan dards-statements Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized study of the effectiveness of school-based mentoring. Prevention Science, 9, 99–113. Keller, T. E. (2005). The stages and development of mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 82–99). Thousand Oaks, CA: Sage. Moher, D., Hopewell, S., Schulz, K. F., Montori, V., Gøtzsche, P. C., Devereaux, P. J., . . . CONSORT Group. (2010). CONSORT 2010 explanation and elaboration: Updated guidelines for reporting parallel group randomised trial. Journal of Clinical Epidemiology, 63, e1–e37. Rhodes, J. E. (2005). A Model of youth mentoring. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 30-43). Thousand Oaks, CA: Sage. Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20(3). Schulz, K. F., Altman, D. G., Moher, D., & CONSORT Group. (2010). CONSORT 2010 statement: Updated guidelines for reporting parallel group randomised trials. PLoS Med, 7(3), e1000251. Schwartz, S., Rhodes, J. E., Chan, C., & Herrera, C. (2011). The impact of school-based mentoring on youth with different relational profiles. Develop­ mental Psychology, 47, 450–462.

498   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Silverthorn, N., DuBois, D. L., Herrera, C., & Kauh, T. (2010). Investigation of a model of the impact of school-based mentoring on academic outcomes. Unpublished manuscript. Steckler, A., & Linnan, L. (2002). Process evaluation for public health interventions and research. San Francisco: Jossey-Bass. Tierney, J. P., Grossman, J. B., & Resch, N. L. (1995). Making a difference. An impact study of Big Brothers Big Sisters. Philadelphia: Public/Private Ventures. Tolan, P., Henry, D., Schoeny, M., & Bass, A. (2008). Mentoring interventions to affect juvenile delinquency and associated problems. Campbell Systematic Reviews, 16. Retrieved from http://www.campbellcol laboration.org/lib/download/238/ Valentine, J. C. (2009). Judging the quality of primary research for research synthesis. In H. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis (2nd ed., pp. 129–146). New York: Russell Sage Foundation.

Valentine, J. C., & Cooper, H. (2008). A systematic and transparent approach for assessing the methodological quality of intervention effectiveness research: The Study Design and Implementation Assessment Device (Study DIAD). Psychological Methods, 13, 130–149. Wang, R., & Ware, J. H. (2013). Detecting moderator effects using subgroup analyses. Prevention Science, 14, 111–120. Wheeler, M. E., Keller, T. E., & DuBois, D. L. (2010). Review of three recent randomized trials of schoolbased mentoring: Making sense of mixed findings. Social Policy Report, 24(3). Wilson-Ahlstrom, A., Yohalem, N., DuBois, D., & Ji, P. (2011). From soft skills to hard data: Measuring youth outcomes. Washington, DC: Forum for Youth Investment. Retrieved from http://www.forumfyi. org/content/soft-skills-hard-dataYarbrough, D. B., Shulha, L. M., Hopson, R. K., & Caruthers, F. A. (2011). The program evaluation standards: A guide for evaluators and evaluation users (3rd ed.). Thousand Oaks, CA: Sage.

34  ECONOMIC EVALUATION OF MENTORING PROGRAMS E. Michael Foster1

Introduction The tools of economic evaluation have been applied to a broad range of policy issues over the past 50 years. These tools are rooted in welfare economics, the branch of economics that attempts to determine the best allocation of resources for a society. One can find economic evaluations in many fields, such as development economics (e.g., clearing landmines in war-torn nations; Gibson et al., 2007); environmental economics (e.g., the control of feral hogs on public lands; Engeman et al., 2007), and agricultural policy (e.g., weed control by Bangladeshi potato farmers; Khan, Hossain, Mahmud, Howlader, & Rahman, 2009), to name just a few. (For others, see Schmitz & Zerbe, 2008.) These fields all have a long history of developing the tools in the economic evaluator’s tool box to fit specific substantive problems. For example, cost-benefit (or equivalently, benefit-cost) analysis depends on a measure of society’s willingness to pay for a natural resource. Environmental economists solicit such measures using contingent valuation methodology, and they have developed various strategies to solicit sensible and consistent values from informants (Kenkel, Berger, & Blomquist, 1994). Economists have used contingent valuation to estimate the value of public parks (e.g., the mountains of Nepal; Baral, Stern, & Bhattarai, 2008) and how that value is affected by public and private decisions (Orens & Seidl, 2009). The application of economic evaluation in health policy also has stimulated the development

of the tools of economic evaluation. The National Health Service in Great Britain uses cost-effectiveness analysis to allocate the nation’s health expenditures, and health economists there lead the world in developing the necessary methodology for such analyses. Health economists, for example, have developed new ways for characterizing the statistical uncertainty surrounding a measure of cost-effectiveness. In the past, researchers ignored the fact that measures of program benefits and costs are estimated from samples—i.e., they ignored the possibility that the apparent cost-effectiveness of one program relative to another might be a chance finding. Health economists have developed tools like the costeffectiveness acceptability curve (discussed below) to characterize this uncertainty. Over the past 30 years, interest in adapting the tools of economic evaluation to spending on programs for children and youth has increased dramatically. A landmark study fueling this growth involved the Perry Preschool, an intensive early education program for disadvantaged children (Aos, Lieb, Mayfield, Miller, & Pennucci, 2004; Barnett, 1985; Belfield, Nores, Barnett, & Schweinhart, 2006a; Schweitzer, 2003). Overstating the importance of this study in shaping public policy toward early childhood education would be difficult. The net result of this and related studies (e.g., Abecedarian project; Barnett & Masse, 2007) has been to reframe the very notion of early childhood education as an investment (Babcock & Bilotti, 1994). Many advocates, policymakers, and researchers clearly believe that early childhood

1

The author would like to thank the editors, Kim McCombs and Katherine Magnuson, for their helpful comments.

499

500   PRACTICE AND PROGRAMMATIC CONSIDERATIONS education generates a substantial return on the public’s investment. Over time, the methods of economic evaluation have been applied to other programs affecting children, youth, and families, such as nursing home visits (Olds, Henderson, Phelps, Kitzman, & Hanks, 1993), drug treatment (Cartwright, 2000), specialized mental health services (Schoenwald, Ward, Henggeler, Pickrel, & Patel, 1996), and others. Mentoring programs provide a natural extension for this work. In this chapter, we consider the economics of mentoring programs. That discussion begins with the economist’s rationale for mentoring programs— i.e., for the support of mentoring by public sources and private philanthropies. Why is such support needed? In other words, why is there not a market for mentoring where parents can buy those services for their children? Such a market might function much like a market for music lessons or other activities from which children benefit. We argue that the basis for mentoring is much like that for other programs for children and youth, such as early childhood education. This discussion leads naturally to the rationale for economic evaluation of such programs: If public funds are used to support mentoring programs, naturally one wants to know that these programs accomplish their aim—improving society’s wellbeing. Policymakers need to know not only whether mentoring programs are a good investment in some general sense but also how such programs stack up against alternative uses of the funds. This assessment needs to incorporate not only the average return on that investment but also the risk and uncertainty involved. We then consider different types of economic evaluation that might be used to evaluate mentoring programs and review past evaluations of mentoring programs by the standards of health economics and other areas of social policy as outlined above. To date, the available studies of mentoring do not meet the standards of health economics. These studies could be better characterized as back-of-the-envelope calculations that are suggestive or speculative. As discussed below, they lack key elements considered part of good practice in economics. More fundamentally, these studies reflect a conceptual framework—that of the costeffectiveness study in health or medicine—that may not fit the question that policymakers face. We conclude with a set of guiding questions and recommendations for policymakers and practitioners. Before proceeding, we note what the chapter is not. Our emphasis is not on how to conduct economic evaluation in terms of procedures. That information is available elsewhere for economic

evaluation in general (Drummond & Mcguire, 2001; Drummond, O’Brien, Stoddart, & Torrance, 1997; Gold, Russell, Siegel, & Weinstein, 1996; Zerbe & Dively, 1994) and for mentoring in particular (Yates, 2005). Our goal here is to provide a broader context and standards for past and future economic evaluations of mentoring and to identify future directions for such work.

Why Support Mentoring Programs With Public Funds? Economic evaluation attempts to identify the types and amounts of programs that maximize social well-being. Such solutions are known as efficient. When economists use the term efficiency, they refer not just to working hard or in a well-organized way. Rather, efficiency refers to whether the amounts produced, bought, and sold produce the highest level of well-being among buyers and sellers, given that the resources could be used in other ways. An employee might work very hard yet the production be inefficient because the amount of the good produced is not socially optimally. For example, one might argue that the number of casinos in a community is not efficient even if the employees there use the latest gaming technology and work very hard. The lack of efficiency refers to the fact that too many of society’s resources are devoted to gaming relative to the benefits to society. If markets are perfectly competitive, the market produces the distribution of resources that is best for society. This notion lies at the heart of economics. A well-functioning market naturally maximizes social well-being as if guided by an “invisible hand” (Smith, 1776). Each individual or firm acting in his, her, or its own interest naturally produces the socially optimal (or efficient) distribution of resources. The benefits to consumers represents the amount by which their willingness to pay for the good exceeds what they pay (i.e., their consumer surplus generated from the good rather than other goods to which they might have devoted their spending). Economists have devoted more than a century to determining the circumstances under which markets produce the best outcome for society. These include, for example, the opportunity for firms to start businesses in markets where firms are unusually profitable. The notions of social wellbeing and willingness to pay are guiding concepts underlying the tools of economic evaluation. Economists recognize that free (or unregulated) markets do not maximize social well-being in all

Economic Evaluation of Mentoring Programs    501 instances. Indeed, in some instances a market may not exist. For example, a private market for national defense is difficult to imagine. Some individuals would refuse to purchase their share of defense knowing that they would benefit from others’ expenditures on defense even if they do not contribute. (This is the so-called free-rider problem.) Another possible market failure is that transactions between buyers and sellers may not capture the effect on all individuals (e.g., pollution or casinos) (Samuelson, 1951). These broader effects are known as externalities—these are effects that spill over onto individuals other than the buyer and seller directly involved in the transaction. Externalities represent one justification for government investment in public education. No doubt, education has a substantial private return, but it also creates many benefits for others (Haveman & Wolfe, 1984; Wolfe & Haveman, 2001). For example, society arguably benefits from an educated citizenry. Because private transactions (i.e., the decisions of parents) do not capture these public benefits, economists believe that children and families underinvest in education if left on their own. Do these ideas offer a basis for public support of mentoring programs? Like education, one could argue that mentoring programs have many public benefits and so a private solution—i.e., leaving families to purchase mentoring as they see fit—will result in underconsumption of mentoring. For that reason, policymakers, acting on behalf of society, could potentially intervene, offering mentoring services and potentially improve social well-being. However, economists recognize that public programs may be adopted for a range of reasons and may not function efficiently (in either sense of the word). For that reason, they have developed the tools of economic evaluation to gauge whether programs improve social well-being in situations when a well-functioning market cannot be relied on to allocate resources. Economic evaluation includes several tools, and one must pick the type best suited to the question of interest. We consider that choice in the context of mentoring later in this chapter.

Overview of Economic Evaluation This section provides an overview of two of the main forms of economic evaluation. Benefit-Cost Analysis As noted, the goal of an economic evaluation is to determine whether an investment of society’s

resources generates a positive social net benefit or profits—i.e., whether the benefits of the program exceed its costs. A program with a positive net social benefit is a good use of society’s resources. The calculation of net social benefits depends on at least two key issues. First, one must address the question of profit to whom? The costs and benefits of any program will generate profits for some groups and not others; the costs will be allocated across the groups involved differently. Reflecting the grounding of these tools in welfare economics, economic evaluation emphasizes that costs and benefits be gauged from the social perspective. In the case of benefitcost analysis, the latter involves a measure of societal willingness to pay for those benefits. Note that economists define social benefits in a specific way— all individuals count equally. A dollar transferred from billionaire Warren Buffett to a homeless man leaves society no better or worse off. This choice of perspective is important because costs and benefits differ depending on the perspective from which they are assessed, and this variation can be informative. Consider the prototypical example of a job training program. That program may move unemployed workers into jobs and reduce welfare payments. In that case, the latter represent costs to participants—they give up those payments. On the other hand, the reduction represents benefits to taxpayers. From a social perspective, the gain and loss offset each other. The only net effect involves the costs of processing the welfare payments and running the welfare program. Other, narrower perspectives might confuse cost-shifting with true social benefit. For example, nursing home visits might reduce future delinquency among the children born to participating mothers. The visits themselves represent costs to the public health system. The benefits would accrue to the juvenile justice system. Each perspective offers only an incomplete picture. Eliminating the program might save money for the public health system yet represent a poor choice from a societal perspective. Having selected a perspective from which to assess the costs and benefits of a program, a second key issue involves the handling of the distribution of costs and benefits over time. Any investment likely requires spending today and generates a stream of future benefits and perhaps more costs. To compare the magnitude of costs and benefits, one needs to express both in terms of today’s resources (Zerbe & Bellas, 2006). For that reason, economists employ discounting—the translation of future expenditures into current dollars. That adjustment is made partially for inflation but also because money

502   PRACTICE AND PROGRAMMATIC CONSIDERATIONS now is inherently worth more than money later. The former, for example, can be invested and generate a return. That process, known as discounting, effectively works like accrual of interest in reverse. Money today is worth more than money in the future because the money can be invested and interest earned. Similarly, a future sum is equivalent to a lower amount today: A consumer is indifferent between a given amount today and more tomorrow because he or she knows that interest can be earned. Having discounted both costs and benefits, one can calculate the difference as net present benefits (Zerbe & Bellas, 2006). Economists cannot agree on the appropriate discount rate, and standard practice is to calculate the net benefits under alternatives (e.g., 3%, 5%, and 10%). A reader familiar with prominent economic evaluations, like the Perry Preschool (discussed below; Barnett, 1985, 1996; Belfield, Nores, Barnett, & Schweinhart, 2006b), may wonder about the role of the benefit-cost ratio, the ratio of the benefits of the intervention to its costs. That ratio is frequently used to compare programs for children and youth (Aos et al., 2004). (Another popular measure, return on investment, simply subtracts one from the benefit-cost ratio and has the same limitations.) Though common, the benefit-cost ratio is generally an inferior measure of a program’s merits. The primary reason is that the ratio is an unreliable guide to picking among competing programs in some situations. In particular, if two programs cost different amounts and you can choose only one, then the one with the higher benefit-cost ratio may not be the better investment. Suppose projects A and B cost $5,000 and $500, respectively. The former generates benefits of $10,000; the latter $1,500. The benefit-cost ratios are 2 and 3, respectively, yet project A should be preferred because it generates the largest net benefits (Birch & Donaldson, 1987; for discussion, see Zerbe & Bellas, 2006). Whether the benefitcost ratio leads to the right choice depends on other features of the policymaker’s choice, an issue we consider in more detail later in this chapter. An irony of the prevalent use of benefit-cost ratios for social programs is that they discard critical information on program scale. In these applications, program scale is an important issue and, as discussed below, far more complex than in the standard application in health and medicine. (Other bases for decisions are commonly used yet incorrect. These include the internal rate of return and the payback period. This issue is not hypothetical, and it is easy to find examples. For a discussion, see Zerbe & Bellas, 2006.)

A second problem with the benefit-cost ratio involves cost savings and their role in calculating the ratio (Zerbe & Bellas, 2006). For example, suppose an intervention costs $500 and is delivered at school. The intervention generates $1,000 in benefits (e.g., expected reductions in future delinquency) and also reduces the costs of other school services by $300. One could calculate the rate of return in two plausible ways. The first includes the cost sav1000 + 300 = 2.6. ings in the benefits: That ratio is 500 Another possibility is to calculate the denominator 1000 = 5.0. One could as the net school costs: 500 − 300 imagine a fair number of alternatives situated between these two possibilities, leaving the policymaker uncertain as to which options to pursue. As noted by Birch and Donaldson (1987), this choice can be made rather arbitrarily or even inconsistently in the same study. Cost-Effectiveness Analysis Benefit-cost analysis is common in some areas of economic evaluation (e.g., environmental economics) but not others. In health care, for example, such analyses are rare. A major barrier in health care is that the most straightforward way to value program inputs or outputs involves the market prices of those goods. However, in the case of health services, market imperfections drive a wedge between market prices and consumers’ willingness to pay (Zweifel & Telser, 2009). Rather, health economists rely on a second form of economic analysis, cost-effectiveness analysis. A key feature of cost-effectiveness analysis is that the program benefits are not expressed in dollars—rather they remain in their natural metric. The differences in costs and benefits between two programs are used to calculate an incremental cost-effectiveness ratio (ICER), or the additional costs per unit of improvement in the outcome. One might wonder why ratios are appropriate here but not in the case of benefitcost analysis. The difference is that the choice at hand is typically more narrow—involving, for example, mutually exclusive programs targeting the same ends for homogeneous groups. As a result, the scale of the analysis is clear: a single patient. By picking among alternative means to accomplish the same ends, one can use cost-effectiveness ratios to attain an efficient allocation of resources. For example, one might compare two cardiovascular treatments and calculate the ICER of one treatment versus the other. Treatment A might be $100 more expensive per heart attack averted. In that case,

Economic Evaluation of Mentoring Programs    503 treatment A is preferred if policymakers are willing to pay at least $100 to avert a heart attack. Otherwise, treatment B is preferred. (Of course, if the two programs cost the same, the analyses reduce to an analysis of effectiveness; if they have the same effect, an analysis of costs. If A is more effective than B and costs less, then A is said to “dominate” B.) That cost-effectiveness analysis does not monetize benefits is the method’s principal advantage and disadvantage. Estimating the benefits of a program, especially those extending far into the future, can be quite resource intensive. For example, one of the key benefits of mentoring may be improved educational outcomes and earnings years after the program. Tracking study participants and collecting accurate information on these outcomes can be quite expensive. On the other hand, in many instances, an intervention may have multiple indicators of effectiveness, and the relative cost-effectiveness of alternative treatments may depend on the outcome selected. For example, psychostimulants looked especially cost-effective as a treatment for ADHD when the outcome measure was symptoms; their advantage over behavior therapy was smaller for broader measures of functioning. Benefit-cost analysis avoids problems of this sort because all outcomes are expressed either in willingness to pay or some proxy.

Economic Evaluation of Mentoring Programs An economic analysis of a mentoring program could address several economic questions, and the nature of the latter would dictate the form of the former. First, one might consider whether a mentoring program is a good investment for a single individual. In that case, one would need a measure of program costs. Assuming the program existed already, that estimate would be the marginal costs of including one more child in the mentoring program. Such an analysis could involve either a benefit-cost or a cost-effectiveness analysis. The latter might involve the incremental costs of the program divided by the effect on a single, key outcome, such as the likelihood of becoming a chronic offender. If the program cost $1,000 and reduced the chance of that outcome by 5 percentage points, the program effectively cost $1,000/.05 or $20,000 per chronic offender avoided. This is the ICER for the program. This ratio would reveal only whether the program was a good choice for this child (versus doing nothing) if we had a good

measure of the willingness of society to pay to reduce chronic offending. In the absence of such a figure, one might use an estimate of the costs such offenders impose on society and argue that that figure represents the minimum policymakers would be willing to pay. (These social costs are labeled “costs of illness” in the context of health and medicine and are generally considered a lower bound on willingness to pay; Kenkel, 1994; Kenkel et al., 1994). These cost reductions as well as increased earnings are treated as the measure of the program’s benefits in studies in early childhood education. The latter represent an increase in society’s productive capacity. Whether conducting a cost-effectiveness analysis or some other form of economic evaluation, one should remember that the necessary calculations involve estimates of the program impact on a key outcome. As a result, the researcher or analyst should present the policymaker not only with the estimated ICER or net benefits but also with an indication of its uncertainty, such as a costeffectiveness acceptability curve (CEAC). The CEAC shows the probability of cost-effectiveness (i.e., that the program cost per outcome exceeds the measure of program benefits). A second question involves comparing two alternative programs, mentoring and another program. In many situations, the presumption is that one program is more costly and more effective than the other. The question is whether the added costs are worthwhile. As before, both are assumed to exist, and so the costs of establishing the program are ignored: They will not be changed by enrolling an additional child and in that sense are “sunk,” not part of the marginal cost. One calculates the ICER as before except that no treatment or treatment as usual is replaced by the lower-cost, less-effective program. Again, the proper presentation of the resulting ICER or net benefits should be accompanied by a CEAC or an indication of uncertainty. This situation is most like that characterizing the standard costeffectiveness analysis in health medicine. A variant of this basic question involves the comparison of offering mentoring to one child versus offering a different program to another. This choice strikes at a major feature (or weakness) of economic evaluation more generally: These analyses typically ignore equity issues entirely. As mentioned earlier, a dollar gained by one individual is equivalent to that gained by anyone else: A dollar transferred from one person to another leaves society’s well-being unchanged. This issue does not arise when one is thinking about alternative programs for a single child; however, in comparing

504   PRACTICE AND PROGRAMMATIC CONSIDERATIONS different combinations of programs and recipients, these issues move to the fore. In health economics, this issue is typically ignored. One individual might benefit from one treatment but another would not. Since the subjects have the same illness, this variation averages out. Even when comparing treatments for different illnesses (treatments for prostate and breast cancers), this issue is typically ignored (in spite of significant issues of gender equity). These issues become even more important in comparisons of rather disparate programs, such as a mentoring program for disadvantaged youth and an after-school arts program for all youth. This issue is one of several that distinguish economic questions surrounding mentoring from the standard application in health economics. Others are apparent especially when one thinks of mentoring in the context of public health. As noted, economic evaluation in health and medicine typically ignores startup costs. However, a public health policymaker may be considering a program that is currently unavailable to a (large) group of children. The costs of establishing the program may not be irrelevant—indeed they may be the costs that matter. In other words, marginal costs still matter, but the margin is not a single kid but an entire program. There are also complicated questions of scale. Increasing the scale of the program may reduce the average costs (per child) but also influence the benefits as well. The standard cost-effectiveness analysis treats the scale of the treatment as a single patient; even if the treatment is going to be offered to a large group of people, the estimated ICER or other measure is presumed to apply to each individual (on average). One could argue that health economists should not ignore this variation, but it seems likely to be even larger outside of medicine. The variation in providers and structure of mentoring programs is quite substantial, so policymakers’ decisions about scale likely influence the return as well. (This is a form of risk, discussed below.) The fit between the standard methods of health economics and mentoring and social programs is even worse when one considers a public health perspective more generally. In that situation, policymakers consider not whether to implement one program but rather how to identify the best mix of programs for children and youth in the community. This issue requires a reframing of the entire economic issue to that of the portfolio. We discuss this broader vision later in this chapter. Before doing so, however, we consider the nature and quality of existing economic evaluations in mentoring.

Existing Economic Evaluations of Mentoring Programs The literature on the economics of mentoring is tiny, but our search did identify a handful of studies that one might label “economic.” The earliest is that of Fountain and Arbreton (2000), which offers an “analysis” in a loose sense. As discussed below, the findings of most of these studies could be considered only suggestive, indicating the possibility that mentoring is cost-effective but offering no strong evidence that this is true and certainly no evidence that resources should be diverted from alternatives into mentoring. Strongest among the economic studies is the study of program costs by Fountain and Arbreton (2000). The report generally conforms to the practices of economic evaluation—that is, the study seeks to calculate the marginal cost for offering mentoring to a youth. The authors report calculations from a sample of 50 mentoring programs selected from a database of 720 programs that had participated in an earlier study. The average mentoring program cost an estimated $1,114 per participant. As is the norm in health economics (Gold et al., 1996), Fountain and Arbreton (2000) estimate the costs of mentoring from a social perspective. Primarily, this perspective means that the authors attempt to capture off-budget (or donated) resources as well as budget costs. (The foundation for doing so in welfare economics is solid, but the authors provide additional justification—namely, future mentoring programs may not be able to depend on community sources to provide these resources.) Primarily, these resources include the value of the volunteers’ time—the authors use the average wage for mentoring staff at programs where mentors are paid. This strategy seems reasonable as long as volunteer and paid mentors do not differ substantially in terms of their education and other predictors of the value of their time. These costs and other donated resources (such as space) are estimated to represent half of the overall costs. (There is a large literature in health economics about how to value patient time, and the issues it raises are relevant here. One concern in the former case is that wages reflect market imperfections such as discrimination. In that case, one would value the time of Black mentors less than White mentors. This is an ongoing controversy within economic evaluation.) A key finding in this study is the enormous variation in costs per participant across programs of various types and approaches. Clearly, programs are working from different budgets to implement different strategies to deliver mentoring. The large

Economic Evaluation of Mentoring Programs    505 variation in the size of the programs only adds to further variation in overall costs per program. Belfield (2003) combines these figures with data from a well-known evaluation of Big Brothers Big Sisters (BBBS) programs (Grossman & Tierney, 1998). The latter involved eight BBBS programs across the country and randomly assigned over 1,000 youth to the BBBS programs or to a waitlist control condition. That study reported a range of key outcomes, such as antisocial behavior, academic outcomes, among others. The original study, however, did not attempt to monetize the program benefits in order to calculate net benefits. Although Belfield’s (2003) analysis has weaknesses (e.g., it seems to confuse the taxpayer and social perspectives), it has strengths as well. The study attempts to capture a fuller range of the benefits of education, including the broader social returns (Haveman & Wolfe, 1984). In essence, the study attempts to link a series of observed outcomes to unobserved future outcomes to which dollar values can be attached. For example, the author uses data from the National Longitudinal Survey of Youth to link truancy to long-term earnings. The author then uses that relationship to express the impact of the BBBS program on truancy in dollar terms (i.e., future earnings). Combining data in this way is a frequent practice in economic evaluation. It is indeed difficult to obtain all of the information needed for an economic evaluation from a single study or evaluation. The primary problem here, however, is that each of these projections introduces potential errors of one sort or another. For example, even with a large dataset, projecting future earnings from education is an uncertain endeavor. That uncertainty introduces further uncertainty into the calculated economic return. Like studies in early childhood education, Belfield (2003) neglects this issue entirely. In sum, this study is suggestive of the potential for mentoring and cost-effectiveness, but at this point the findings are only suggestive. They are not a valid basis for allocating public funds. Two additional studies provide calculations involving programs and dollars but do not represent actual economic evaluations. Both represent “back-of-the-envelope” calculations. Blechman, Maurice, Buecker, and Helberg (2000) calculate cost-effectiveness ratios for two programs offering mentoring services. However, they use the amount of a state block grant for program costs; as discussed earlier, this likely underestimates true program costs. No confidence intervals are provided for the cost-effectiveness ratios, and the study suffers from other flaws. Another study in this vein

(Moodie & Fisher, 2009) does not involve outcome data from an actual outcome study. Rather it estimates the magnitude of benefits that would be needed for a program to break even. A study by Aos and colleagues (2004) combines data on the estimated costs and benefits of several different mentoring programs. These authors follow much the same methodology as those described above, but they do draw on more data sources. They also report not just the benefit-cost ratio but also the net present value. All in all, however, the study has many of the problems of earlier studies. For example, the study lacks essential elements of an economic analysis, such as any sense of the uncertainty with which the costs and benefits are estimated. For example, the study relies on estimates of the link between test scores and future earnings. The increase in earnings represents the largest benefit of the program, but the original effect of the program used is barely statistically significant (p = .10). Converting the effect of test scores into earnings depends on the rate of return to test scores from another study in the literature. Of course, that rate of return is an estimate and suffers from other forms of uncertainty having to do with whether the association between test scores and earnings is truly a causal relationship (i.e., whether improving test scores leads to improved earnings).

Discussion: What Do We Know and What Do We Need To Know The literature on the economics of mentoring is just beginning. What do we know at this point? Not much. Relatively crude informal calculations suggest that mentoring may be a good investment. The existing studies do not meet the standards of health economics, and for that reason, they generally have not been published in the peer-reviewed literature. Even if policymakers consider such outlets to be the impractical realm of scholars, publication in scholarly journals does provide an important source of quality control. Furthermore, peer-reviewed publishing also would increase the degree of objectivity— several of the reports involved were prepared for organizations that provide or advocate mentoring services. Whether an investment in mentoring should be preferred to alternatives is currently unknown. One could compare the net benefits of mentoring to that of other investments in children and youth (Aos et al., 2004), but that comparison would be incomplete and potentially misleading. Comparing one ballpark estimate with another is an uncertain enterprise, to

506   PRACTICE AND PROGRAMMATIC CONSIDERATIONS say the least. One might or might not prefer mentoring to other programs at this point for various reasons, but one could not ground that argument in any type of economic reasoning. This chapter argues that there is a great deal of loose talk currently about “investments” in children and youth. The presumption here, however, is that the term investments is not a metaphor but that policymakers want to seriously assess these programs as investments much the way a private investor would assess alternative uses of real money. Such an investor would apply a series of rigorous tools to determining how to invest those funds. Neither a practitioner nor a researcher would want his or her retirement income invested based on evidence of the strength that exists for mentoring or early childhood education as economic investments. The questions raised by investments in children and youth are even more difficult than those faced by a private investor. Investment in alternatives such as early childhood education and mentoring, for example, may be even more closely connected or interdependent than, say, investment in a public utility and in mortgage-backed financial derivatives. In the case of investments in children and youth, different investments may interact in various ways, reinforcing the effects of one on another. At this point, knowing that the average rate of return for mentoring is higher or lower than that for some other program (such as early childhood education) is not really useful, given that both figures are calculated using small and/or unrepresentative samples served in model programs. Research in health economics currently brings to bear much, much more data on arguably much easier (or circumscribed) problems such as providing statins to reduce cholesterol. Those data do not address all of the economic questions, but they do at least offer more reliable answers to a narrow set of questions. What do we need to know about the economic returns to mentoring? As in early childhood education, the starting point is indeed a series of small high-quality studies. The results of those studies need to be interpreted in light of broader questions raised. For that reason, the studies need to point toward the portfolio framework and assess risk and uncertainty. The tools required to do so are available in the health economics literature, and they should be applied. It is hard to understand why recent reports from the Perry Preschool do not report confidence intervals for the benefit-cost ratio. That interval would be enormous and would likely temper arguments that early childhood is an investment.

For now, all one can say about the economics of mentoring is that such programs are conceivably a good use of society’s resources. The existing literature neglects key issues, such as the role of uncertainty, and thus cannot be used as a reliable source for judging whether this is indeed the case. By providing sound cost estimates, the Fountain and Arbreton (2000) and Herrera, Grossman, Kauh, Feldman, and McMaken (2007) studies provide a solid foundation with which to begin. That these studies use appropriate methodology (e.g., including donated resources) is particularly promising. A comprehensive understanding of the economics of mentoring would begin with this foundation and pass through phases. In a first phase, high-quality economic evaluations would be conducted for a handful of studies. These studies would provide the opportunity to establish a standardized framework for economic evaluation in this area. That framework would attend to issues that are irrelevant and/or ignored in the context of health policy. As discussed earlier in this chapter, these issues include the costs of starting and establishing programs. Health economists may take it for granted that a hospital has an emergency room; that a community has a high-quality mentoring program in place is far from certain. More generally, the issues raised by mentoring and other social programs simply do not fit the standard cost-effectiveness framework. The public health policymaker is responsible for all children in his or her community. The programs involved are not mutually exclusive—the key questions are not whether to provide early childhood education or mentoring but how much of each and to whom. Neither are the returns to these programs independent. Having had early childhood education may make mentoring (or other social programs) more effective. Indeed, the reason a child needs mentoring may reflect (at least partially) the absence of quality early childhood education. The right framework for these questions is that of a portfolio of investments. Health economists themselves recognize that such a framework characterizes many key decisions in health policy (Bridges, Stewart, King, & Van Gool, 2002; Dow & da Costa Werlang, 1992; O’Brien & Sculpher, 2000; Sendi, Al, & Zimmermann, 2004; Xiaoxia, 2010). Such a framework incorporates the costs of establishing programs as well as issues of program scale. Equity issues could be better addressed as well. For example, the decision about how to spend money on adolescents could be understood in the context of past spending on those youth at younger ages. Issues of uncertainty also become quite important. Policymakers may want to “hedge their bets”—

Economic Evaluation of Mentoring Programs    507 some investments may be more uncertain. A portfolio perspective would consider not only the expected return on an investment but also the variability associated with it. Such a framework would require shifts in the research focus in mentoring and in other fields. For example, a portfolio perspective adds another reason to prefer effectiveness studies to those focused on efficacy. Moving programs from research settings into the “real world” has real relevance for economic analyses. Programs for which evidence of impact is efficacy research only suffer from added uncertainty. Such programs should suffer in comparisons of what programs to include in a public health portfolio of programs.

Conclusion As this chapter makes clear, estimating the economic returns to investing in mentoring is a challenging enterprise. The discussion has highlighted a set of questions researchers and policymakers should consider in assessing existing and future economic evaluations of mentoring programs (see Table 34.1). At the very least, an economic evaluation requires rigorous evaluation(s) of mentoring programs. Cost-effectiveness analysis can be no better than the data used to evaluate effectiveness. For that reason, economic evaluation is appropriate only for a mature field of research, so the first question one can pose for policy and practice is whether the field is ready for economic evaluation. If there is not a solid research base for mentoring programs, then economic evaluation should be postponed. A second question is related: Can we accept lower-quality evidence for the effectiveness and cost-effectiveness of mentoring? Clearly, the field has a long way to go to meet the standards of health

economics for gauging the economic return on mentoring programs. One possibility is just that those standards are too high and unreasonably so. The evaluation of many medical treatments occurs in tightly controlled settings involving a rather homogeneous group of patients with medical conditions that are well defined. In contrast, mentoring programs exist in the real world, a messy place where children and youth are diverse and the programs are underresourced. Perhaps lower-quality evidence is good enough. In that case, the evidence to date—that mentoring might be cost-effective—is good enough. Another question is whether economic evaluation is really necessary at all. One problem with existing research on the economics of social programs is that it treats children as if they were savings accounts in which one could deposit money and generate a certain return. As this chapter makes clear, that view is naïve and perhaps misleading. Children are not banks, and perhaps more fundamentally, the notion that we must earn a profit on them is rather offensive. Perhaps this entire enterprise is misguided. Of course, a health economist such as the author is unlikely to endorse the idea that less health economics is needed. Indeed, it seems unlikely that public budgets will ever be large enough to justify all programs that someone feels is the “right thing to do” or “worth it for its own sake.” In that light, the idea that we might use those funds in ways that are best for society seems worthwhile. But if that is the aim for economic analysis, we are a long, long way from having the necessary information to allocate public funds across alternative uses. Such a goal requires comprehensive information not on just one use of public funds but on all possible uses. What seems rather sub-optimal is making decisions based on an incomplete or sub-standard research.

Table 34.1   Questions to Consider in Assessing Research on the Economics of Mentoring Programs Is the analysis grounded in a sound evaluation of program impact? Do the authors clearly state the economic question being examined? Is that question the one that policymakers need to have answered? Does the analysis of costs and benefits reflect a social perspective? Does the presentation of findings present a confidence interval or some indication of the statistical uncertainty surrounding the measure of economic impact? If the analysis uses other data sources to estimate future benefits and costs of the program, does it include the related uncertainty?

508   PRACTICE AND PROGRAMMATIC CONSIDERATIONS A more reasonable goal, however, might be to use economic evaluation to make the best use of funds we decide to spend on mentoring for other reasons. In that light, the costs and benefits of a mentoring program represent a form of continuous quality improvement. The real contribution of economic evaluation might be to help us make sure that the money spent on mentoring produces the greatest return. Mentoring programs might benefit from a good dose of economic thinking. Perhaps the greatest contribution of economics is not to be found in ex post economic evaluations of programs already implemented. Rather, the real contribution of economics may lie in improving the tracking and management of resources used for mentoring.

References Aos, S., Lieb, R., Mayfield, J., Miller, M., & Pennucci, A. (2004). Benefits and costs of prevention and early intervention programs for youth. Seattle: Washington State Public Policy Institute. Babcock, L., & Bilotti, A. (1994). Defining and measuring health over life. In D. Kenkel & R. Fabian (Eds.), Valuing health for policy: An economic approach (pp. 205–220). Chicago: University of Chicago Press. Baral, N., Stern, M. J., & Bhattarai, R. (2008). Contingent valuation of ecotourism in Annapurna conservation area, Nepal: Implications for sustainable park finance and local development. Ecological Economics, 66, 218–227. Barnett, W. S. (1985). The Perry Preschool Program and its long-term effects: A benefit-cost analysis. Ypsilanti, MI: High/Scope Educational Research Foundation. Barnett, W. S. (1996). Lives in the balance: Age-27 benefit-cost analysis of the High/Scope Perry Preschool Program. Ypsilanti, MI: High/Scope Educational Research Foundation. Barnett, W. S., & Masse, L. N. (2007). Comparative benefitcost analysis of the Abecedarian program and its policy implications. Economics of Education Review, 26, 113–125. Belfield, C. (2003). Estimating the rate of return to educational investments: A case study using the Big Brothers Big Sisters mentoring program. New York: National Center for the Study of Privatization in Education. Belfield, C. R., Nores, M., Barnett, S., & Schweinhart, L. (2006a). The High/Scope Perry Preschool Program. Madison: University of Wisconsin Press. Belfield, C. R., Nores, M., Barnett, S., & Schweinhart, L. (2006b). The High/Scope Perry Preschool Program: Cost-benefit analysis using data from the age-40 followup. Journal of Human Resources, 41, 162–190. Birch, S., & Donaldson, C. (1987). Applications of costbenefit analysis to health care: Departures from

welfare economic theory. Journal of Health Economics, 6, 211–225. Blechman, E. A., Maurice, A., Buecker, B., & Helberg, C. (2000). Can mentoring or skill training reduce recidivism? Observational study with propensity analysis. Prevention Science, 1, 139–155. Bridges, J. F. P., Stewart, M., King, M., & Van Gool, K. (2002). Adapting portfolio theory for the evaluation of multiple investments in health with a multiplicative extension for treatment synergies. The European Journal of Health Economics, 3, 47–53. Cartwright, W. (2000). Cost benefit analysis of drug treatment services: Review of the literature. Journal of Mental Health Policy and Economics, 3, 11–26. Dow, J., & da Costa Werlang, S. R. (1992). Uncertainty aversion, risk aversion, and the optimal choice of portfolio. Econometrica: Journal of the Econometric Society, 60, 197–204. Drummond, M. F., & Mcguire, A. (Eds.). (2001). Economic evaluation in health care: Merging theory with practice. Oxford, UK: Oxford University Press. Drummond, M. F., O’Brien, B., Stoddart, G. L., & Torrance, G. W. (1997). Methods for the economic evaluation of health care programmes (2nd ed.). New York: Oxford University Press. Engeman, R. M., Constantin, B. U., Shwiff, S. A., Smith, H. T., Woolard, J., Allen, J., & Dunlap, J. (2007). Adaptive and economic management methods for feral hog control in Florida. Human-Wildlife Conflicts, 1, 178–185. Fountain, D. L., & Arbreton, A. (2000). The costs of mentoring. In J. B. Grossman (Ed.), Contemporary issues in mentoring (pp. 48–65). New York: National Center for the Study of Privatization in Education. Gibson, J., Barns, S., Cameron, M., Lim, S., Scrimgeour, F., & Tressler, J. (2007). The value of statistical life and the economics of landmine clearance in developing countries. World Development, 35, 512–531. Gold, M. R., Russell, L. B., Siegel, J. E., & Weinstein, M. C. (Eds.). (1996). Cost-effectiveness in health and medicine. New York: Oxford University Press. Grossman, J. B., & Tierney, J. P. (1998). Does mentoring work?: An impact study of the Big Brothers Big Sisters program. Evaluation Review, 22, 403–403. Haveman, R. H., & Wolfe, B. L. (1984). Schooling and economic well-being: The role of nonmarket effects. Journal of Human Resources, 19, 376–407. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., & McMaken, J. (with Jucovy, L. Z.). (2007). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Philadelphia: Public/Private Ventures. Kenkel, D. (1994). Cost of illness approach. In D. Kenkel & R. Fabian (Eds.), Valuing health for policy: An economic approach (pp. 42–71). Chicago: University of Chicago Press. Kenkel, D., Berger, M., & Blomquist, G. (1994). Contingent valuation of health. In D. Kenkel & R. Fabian (Eds.), Valuing health for policy: An

Economic Evaluation of Mentoring Programs    509 economic approach (pp. 72–104). Chicago: University of Chicago Press. Khan, A. S. A., Hossain, M. A., Mahmud, A. A., Howlader, M. I. A., & Rahman, M. A. (2009). Integrated weed management in potato at Munshigonj. Bangladesh Journal of Agricultural Research, 33, 647–654. Moodie, M. L., & Fisher, J. (2009). Are youth mentoring programs good value-for-money? An evaluation of the Big Brothers Big Sisters Melbourne program. BMC Public Health, 9, 41. O’Brien, B., & Sculpher, M. (2000). Building uncertainty into cost-effectiveness rankings: portfolio risk-return trade-offs and implications for costeffectiveness decision rules. Medical Care, 38, 460–468. Olds, D. L., Henderson, C. R., Phelps, C., Kitzman, H., & Hanks, C. (1993). Effect of prenatal and infancy nurse home visitation on government spending. Medical Care, 31, 155–174. Orens, A., & Seidl, A. (2009). Working lands and winter tourists in the Rocky Mountain West: A travel cost, contingent behaviour and input output analysis. Tourism Economics, 15, 215–242. Samuelson, P. A. (1951). Economics (2nd ed.). New York: McGraw-Hill. Schmitz, A., & Zerbe, R. O. (2008). Applied benefit-cost analysis. Cheltenham, UK: Edward Elgar. Schweitzer, L. J. (2003, April). Benefits, costs and explanation of the High Scope Perry Preschool Program. Paper presented at the 70th biennial meeting of the Society for Research in Child Development, Tampa, Florida.

Sendi, P., Al, M. J., & Zimmermann, H. (2004). A riskadjusted approach to comparing the return on investment in health care programs. International Journal of Health Care Finance and Economics, 4, 199–210. Schoenwald, S. K., Ward, D. M., Henggeler, S. W., Pickrel, S. G., & Patel, H. (1996). Multisystemic therapy treatment of substance abusing or dependent adolescent offenders: Costs of reducing incarceration, inpatient and residential placement. Journal of Child and Family Studies, 5, 431–444. Smith, A. (1776). An inquiry into the nature and causes of the wealth of nations. London: A. M. Kelley. Wolfe, B., & Haveman, R. (2001). Accounting for the social and non-market benefits of education. In J. Helliwell (Ed.), The contribution of human and social capital to sustained economic growth and well being (pp. 221–250). Paris: OECD. Xiaoxia, H. (2010). Portfolio analysis: From probabilistic to credibilistic and uncertain approaches. New York: Springer. Yates, B. (2005). Cost-benefit and cost-effectiveness analysis. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 525–543). Thousand Oaks, CA: Sage. Zerbe, R. O., & Bellas, A. S. (2006). A primer for benefitcost analysis. Cheltenham, UK: Edward Elgar. Zerbe, R. O., Jr., & Dively, D. D. (1994). Benefit-cost analysis in theory and practice. New York: Harper Collins College Publishers. Zweifel, P., & Telser, H. (2009). Cost-benefit analysis for health. In R. J. Brent (Ed.), Handbook of research on cost-benefit analysis (pp. 31–53). Cheltenham, UK: Edward Elgar.

35  ETHICAL ISSUES IN YOUTH MENTORING Jean E. Rhodes, Belle Liang, and Renée Spencer

Introduction Adults who volunteer to serve as mentors generally enter the relationships with a strong desire to make a positive difference in the lives of young people. Typically armed with a limited amount of training and a set of program rules and guidelines, these volunteers set out to forge meaningful relationships with young people to whom they have no prior connection. Unfortunately, the ethical implications of pairing unrelated youth and adults in this way has received little systematic consideration in the field of youth mentoring, and there are few guidelines to address the ethical responsibilities and obligations of adults serving youth in this capacity. This lack of attention is alarming, particularly in light of both the large numbers of volunteers and youth who each year are paired in formal mentoring relationships and the fact that substantial proportions of these relationships terminate prematurely or unfold in other ways that raise important ethical considerations (Rhodes, 2002). The goal of this chapter is to review some of the ethical dilemmas common to mentoring relationships for youth, to present a set of ethical principles, and to discuss research and practice implications with an eye toward facilitating enhanced attention to the ethical responsibilities involved in mentoring of youth. We focus on mentors serving in formal youth programs rather than on natural mentors or other more informal ties between youth and adults that occur outside of youth-serving programs or settings. Volunteer mentors cannot be held accountable to a set of suggested guidelines the way a therapist or nurse, for example, is beholden to a professional code of ethics. Nonetheless, mentoring programs can

sensitize mentors to ethical issues and clearly communicate their expectations of volunteers. Some of the ethical dilemmas we consider overlap with legal issues, to which programs and mentors should be sensitized. Although the boundary between the moral and legal is not always clear, discussion and training around these issues may help to allay volunteers’ concerns about personal liability and risk. Indeed, potential volunteers have raised concerns around such issues, including the possibility of being accused of inappropriate behavior (Scales, 2003). Volunteers may also struggle with the legal and ethical limits of confidentiality (e.g., when a protégé discloses potentially illegal or injurious intentions). The guidelines we propose, although not a blueprint, are designed to provide a stronger framework than has heretofore been available for addressing ethical issues in mentoring relationships and programs for youth. It is often assumed that mentoring programs are universally beneficial to youth. As such, the emphasis of many programs has been on recruiting mentors and forming new matches. Funders reinforce this tendency, often measuring the number of new matches, as opposed to their quality or sustainability, as the indicator of program success (Rhodes & DuBois, 2006). Moreover, because volunteer recruitment is often the rate-limiting factor in program growth (see Stukas et al., this volume, Chapter 27), many mentoring programs for youth have relaxed minimum volunteer screening, commitment, and training requirements. Unfortunately, these trends are at odds with the types of practices that research indicates are needed to establish and sustain high-quality mentoring relationships (Rhodes & DuBois, 2006). The only real danger associated with placing youth in the care of unrelated adults, in the thinking

511

512   PRACTICE AND PROGRAMMATIC CONSIDERATIONS of many agencies, is child molestation—a very troubling but comparatively infrequent occurrence. Meanwhile, the more subtle and frequent forms of sadness and disappointment youth may experience in connection with their experiences in mentoring programs have gone largely unattended. Yet, because a personal relationship lies at the heart of mentoring as an intervention strategy, inconsistencies, misunderstandings, and terminations may touch on youths’ vulnerabilities in ways that other, less personal, approaches do not. To the extent that protégés have identified with their mentors and have begun to value the relationship, they may feel profound disappointment, rejection, and even betrayal when problems arise (Rhodes, 2002). Such feelings, in turn, may contribute to negative emotional, behavioral, and/or academic outcomes (Downey, Lebolt, Rincon, & Freitas, 1998). Indeed, in two large random assignment studies, youth in prematurely terminating mentoring relationships showed increases in problem behaviors relative to youth in control groups (Grossman & Rhodes, 2002; Herrera, Grossman, Kauh, Feldman, McMaken, & Jucovy, 2007). Qualitative research has similarly documented the profound feelings of disappointment that youth may experience when abandoned by their mentors (Spencer, 2007). Although volunteer mentors are not bound by any professional codes of ethics, they are obliged to act responsibly and, as we argue, ethically, in their roles with youth. In this chapter, we begin by drawing on theory and research to present a broad set of ethical guidelines and in doing so exploring the application of each guideline to youth mentoring relationships. Next, we provide an overview of the implications of the guidelines for mentoring program practice. We conclude with a discussion of the contexts in which the guidelines might be used to improve practice and in this context also discuss needs for additional research. Although our guidelines and examples are most applicable to volunteer, face-to-face mentoring programs, the issues discussed could nonetheless be relevant to mentors who are paid and/or to those who connect with youth through online programs.

Theory In this section, we provide a set of five ethical principles to guide the development, maintenance, and closure of adult-youth mentoring relationships. These guidelines were adapted from the American Psychological Association (APA, 2002)

Ethical Principles of Psychologists and Code of Conduct. Ethical Guidelines as Building Blocks for Strong Relationships Engagement in ethical guidelines is expected to improve mentoring of youth by facilitating closer, more enduring relationships that, in turn, are associated with better youth outcomes. By attuning volunteer mentors to the subtle ways in which their own ethical behaviors and stances can affect the dynamics of their interactions with youth, a codified set of ethical guidelines is expected to strengthen mentoryouth relationships. Effective mentoring depends on the extent to which the mentor and youth manage to forge a strong connection that is characterized by mutuality, trust, and empathy. Without such connection, the youth and/or mentor may disengage from the relationship before it lasts long enough to have a positive impact on the young person. What Constitutes an Ethical Approach to Mentoring? The proposed ethical principles were developed recently by the authors of this chapter to guide volunteer mentors and the program staff who advise them as they strive to build meaningful and growthpromoting relationships with their protégés (Rhodes, Liang, & Spencer, 2009). As noted previously, the principles are based on ethical principles that have been established for psychologists, many of whom are engaged in direct practice as psychotherapists. Although there are many limits to the clinical analogy, both mentoring and therapeutic relationships involve a human connection whose explicit goal is to foster the positive development of one of the partners. The ethical principles that serve as the foundation for the APA (2002) Ethics Code thus may be a helpful point of reference from which to consider ethics in youth mentoring. More specifically, we have drawn on the five general principles of the APA Ethics Code—(a) beneficence and nonmaleficence, (b) fidelity and responsibility, (c) integrity, (d) justice, and (e) respect for people’s rights and dignity—to develop a set of corollary principles for youth mentoring relationships. We view these principles as relevant to the wide range of program formats in which formal youth mentoring may occur (e.g., face-to-face, online, group) as well as potentially to other voluntary adult-youth relationships that can be forged in various types of programs and contexts (e.g., sports teams, youth groups, camp).

Ethical Issues in Youth Mentoring   513 Promote the Welfare and Safety of the Young Person (Beneficence and Nonmaleficence) Beneficence implies that mentors should work to benefit their youth protégés. At face value, this may appear to be the most obvious and least controversial of all principles. In ethical theory, the concept of beneficence, on which this principle is based, encompasses behavior that both benefits the good of another and avoids harm (Fisher, Hoagwood, & Jensen, 1996). The philosopher Hume held beneficence (which he contrasted with egoism) as a central principle in his moral psychology, close to the essence of morality (Beauchamp & Childress, 2008). Thus, a mentor’s primary responsibility is to take positive action to promote the welfare of the young person and to refrain from any action that may cause harm. Yet morality is not always that clear and simple when it comes to youth mentoring. What if there are competing ideas about “what is best” for the young person? Volunteer mentors must be sensitive to subtleties in protégés’ family dynamics, culture, and social ecology. For example, a parent might do something (e.g., discourage participation in a competitive summer math camp so that the child can work) that, at face value, may not seem to be in the best interests of the youth. It may, however, make perfect sense when considered within the broader social ecology of the family. More generally, volunteer mentors, the majority of whom are White, middle-class students and professionals (MENTOR, 2006), may find their values to be at odds with those of the lower-income parents of the youth who are typically served in programs. Yet research suggests that the success of a mentor match often depends on the degree of cooperation between mentors and family members (Rhodes & DuBois, 2006). Promoting the welfare of a young person will, in many cases, require that mentors build rapport not only with their protégés, but also with the protégés’ primary caregivers so that they may develop an understanding of the family’s values, belief systems, and expectations for the child. Nonmaleficence refers to avoidance of mentor behavior that results in disappointment and strain as well as more extreme forms of harmful behavior, such as sexual harassment, abuse, and exploitation. Many programs have careful background checks and screening procedures in place, and the incidence of such occurrences appears to be quite rare (Big Brothers Big Sisters of Canada, 2006; see also Kremer & Cooper, this volume, Chapter 28). As such, it seems reasonable to assume that all but a very small fraction of volunteers do not intend to deliberately harm their young charges. Nonetheless,

if volunteers lack skills, knowledge, or caution, other types of difficulties can and often do arise that pose risks to youth well-being. For example, some research (Grossman & Rhodes, 2002; Kalbfleisch, 1997) has suggested that harm to protégés can arise from the misuse of power (e.g., exploitation, heavy-handed persuasion) and inappropriate boundaries (e.g., breaching confidentiality, improper disclosures). Misuse of Power. Although mentoring relationships typically are intended to “empower” youth, they are at the same time susceptible to unacknowledged power inequities. In fact, by virtue of the different roles and ages of the participants, a power differential is inherent in adult-youth mentoring relationships. The gulf may widen when there are also differences in class and racial backgrounds (Spencer, 2007). Mentors may not even be aware of the social inequities driving such differentials or how they can play out in interpersonal relationships (Fisher, 1997; see also Deutsch et al., this volume, Chapter 12). A mentor may express beliefs or opinions that are at odds with the experiences and values of his or her protégé, for example, creating conflict or tension for the young person. With such possibilities in mind, we recommend that mentors refrain from religious or political proselytizing, strive to raise their own awareness of power dynamics in cross-age and cross-cultural relationships, and seek consultation from mentoring program staff to effectively negotiate such differentials. Inappropriate Boundaries. Clear boundaries provide defined limits around relationships and are central to avoiding the misuse of power. Yet, the boundaries surrounding mentoring relationships are far murkier than those governing professional relationships. Mentors fill a niche that lies somewhere between professional and kinship and are thus afforded greater latitude in what constitutes appropriate boundaries. Still, some guidelines for professional helping relationships are easily translated to the mentoring context, such as prohibitions against sexual relationships and placing primacy on the needs and safety of the child (Fisher & Younggren, 1997). For example, although there is nothing inappropriate in a mentor holding hands with his 6-yearold protégé as they cross the street, other instances of physical contact or seemingly benign gestures or comments can be interpreted differently. And, particularly in light of the power differentials, it is not always easy for protégés to set limits or voice concerns. A 7-year-old child might be hesitant to

514   PRACTICE AND PROGRAMMATIC CONSIDERATIONS accept her mentor’s invitation to sit on her lap as they share a computer screen, and a teenager might feel vaguely uneasy by her mentor’s comments about her changing body. In light of these issues, we would conclude that it is best to err on the side of caution, as there are many ways to show affection and closeness that do not involve physical contact or even benignly crossed boundaries. The ethical guidelines presented here also overlap with legal restrictions around child abuse and exploitation. As mentioned, most programs obtain criminal background checks on prospective mentors to reduce the incidence of child exploitation. Screening practices, including face-to-face interviews with prospective mentors as well as reference and background checks, are recommended as a guideline across a wide range of mentoring programs (Miller, 2007). In particular, criminal background checks are a necessary component of screening prospective mentors and should be conducted before initiating any contact between the mentor and the youth. The checks provide a concrete method for mentoring programs to enhance the likelihood that the mentee will be protected and safe with his or her mentor. Analysis conducted by the National Center for Missing & Exploited Children, as part of the PROTECT ACT Criminal Background Check Demonstration Project for potential mentors, indicates that it is important to gain access to records from national and not just state registries. This is because criminals move, and one state registry alone may not provide a complete picture of an individual’s criminal history (Wiley, 2011). For a more in-depth discussion of issues relating to safety screening procedures in youth mentoring relationships, see Kremer & Cooper, this volume, Chapter 26. But beyond these clear legal boundaries there are ways in which the boundaries of the mentoring role are far less defined. Indeed, mentors can complicate the relationship by stepping beyond the customary role of volunteer and engaging in multiple relationships. Just as is the case for therapists for whom it may not be feasible or responsible to avoid social or other nonprofessional contacts with their clients, mentors should still be aware of the potential harmful effects that certain types of multiple roles may have on their relationships with protégés and on their protégés themselves. Entering into a professional, financial, or other relationship with a protégé or one of his or her family members, for example, has the potential to interfere with a mentor’s ability to work effectively with the protégé and could exploit or cause other harm to the protégé. In other instances, mentors may find themselves encroaching on the territory of professionals in the youth’s life

(e.g., tutors, therapists, juvenile probation officers). Maintaining clarity about the boundaries of the mentor’s role is not always easy. Mentors will be best served in our view by being mindful of situations in which their protégés make disclosures that could be better handled by a professional and, in such instances, referring the matter to program staff or other appropriate persons. In doing so, careful consideration should be given to the youth’s right to privacy and the harm that could follow from the youth’s experiencing a perceived break in trust with his or her mentor. Ideally, as we discuss later, to minimize risk for such repercussions, the limits of confidentiality should be communicated proactively by programs to both mentors and protégés. Despite the potential boundary violations that may result from the inherent power inequities within mentoring relationships, mentors’ experiences and resources can also be used for good. From this perspective, a mentor’s and a program’s power (i.e., position, knowledge, experience, interpersonal connections, other material and human resources) can be conceptualized as a form of social capital that can be deployed to the advantage of protégés. Mentors may use their connections, personal and professional experience, and other resources to help their protégés gain desired information, status, position, influence, and other types of personal and professional achievements. Ultimately, a mentoring relationship that is respectful of a youth will be one in which the mentor works both to mitigate against culturally ascribed inequalities and to foster the youth’s development in such a way that inequalities due to age and experience are able to dissipate over time as the youth learns, grows, and matures. Be Trustworthy and Responsible (Fidelity and Responsibility) The principle of being trustworthy and responsible is rooted in the ethical notion of fidelity, or “behaving in a trustworthy manner and keeping one’s promise or word” (Strom-Gottfried, 2008, p. 21). For volunteer mentors, this involves being aware of one’s responsibilities for meeting frequency and match duration, as stipulated by the program. This type of consistency and reliability—which has been associated with more positive outcomes for youth participants (Rhodes & DuBois, 2006)—serves as a cornerstone for trust in the relationship. Young adolescent protégés have reported that honesty, keeping promises, and relationship longevity promote trust (Liang, Spencer, Brogan, & Corral, 2008). Yet volunteers sometimes do things that inadvertently violate that trust. Volunteers may neglect to inform their

Ethical Issues in Youth Mentoring   515 protégés of travel plans, or fail to set up meetings when work or school demands become overwhelming. In one instance known to us, a tax accountant unexpectedly failed to respond to her protégé’s e-mails or phone calls for over a month, only to reemerge in mid-April, unapologetic and ready to resume normal activities. And, unfortunately, as noted at the outset of this chapter, many volunteer mentoring relationships end prematurely. Most often such terminations occur at the request of the volunteer (Rhodes, 2002). Some premature endings are unavoidable, such as when a mentor is unexpectedly relocated or the protégé’s family decides to leave the area. All too often, however, relationships come to a heartbreaking end when the mentor drops the ball and contacts neither the protégé nor the mentoring program to provide an explanation (Rhodes, 2002). This is especially concerning in light of evidence that early terminations can lead to decrements in youth functioning (Grossman & Rhodes, 2002; Herrera et al., 2007; Karcher, 2005; Slicker & Palmer, 1993). In a reanalysis of data from a random assignment study of the Big Brothers Big Sisters (BBBS) community-based mentoring programs, Grossman and Rhodes found evidence that effects on youth outcomes became progressively stronger as relationships persisted for longer periods of time. The largest benefits were apparent for youth in relationships that lasted 1 year or longer. By contrast, youth in relationships that terminated in less than 3 months showed declines in functioning relative to controls. Such trends remained even when considering potential confounding influences such as baseline characteristics of youth that could contribute to increased risk for premature termination. Findings of other investigations have similarly highlighted the importance of match length and consistency as well as the negative consequences of early terminations (DuBois, Neville, Parra, & Pugh-Lilly, 2002; Karcher, 2005; Slicker & Palmer, 1993; Spencer, 2007). Unfortunately, as many as half of mentoring relationships terminate prematurely (Bernstein, Dun Rappaport, Olsho, Hunt, & Levin, 2009; Herrera et al., 2007; Rhodes, 2002). To redress this problem, practitioners often rematch youth—that is, replace prematurely terminated mentors with new volunteers (Keller, 2005). In BBBS school-based programs, for example, nearly 15% of youth are rematched with new volunteers as a result of early, unexpected terminations (Herrera et al., 2007). Yet, recent research casts doubt on that practice. In particular, in an analysis of school-based matches within BBBS programs, researchers found that youth who experienced premature match terminations and were rematched

actually fared worse than their control group peers. A similar difference was not apparent when comparing youth with prematurely ended matches who were not rematched to controls (Grossman, Chan, Schwartz, & Rhodes, 2012). Engaging in multiple truncated matches may ignite a youth’s vulnerabilities to loss or short-circuit the necessary stages of understanding that the termination process can afford. Although preliminary, such findings highlight the importance of mentors being trustworthy and honoring their commitment to the relationships. At the same time, programs should carefully screen for mentors who can honor their commitments and support them in doing so. Another key aspect of volunteer mentors’ obligation to be responsible involves availing themselves of training, enrichment, and supervision provided by the program as well as seeking outside opportunities for enhancing their competence in such areas as ethnic diversity, social class differences, family systems, and adolescent development. Such training and support can assist volunteers in recognizing the boundaries and limits of their expertise and in identifying situations in which it is appropriate to seek assistance from program staff. Act With Integrity (Integrity) The principle of acting with integrity highlights the obligation of mentors to be thoughtful and forthright about the commitments (e.g., time, financial) involved in mentoring and to avoid setting up unrealistic expectations. A mentoring relationship, like any other interpersonal relationship, holds the potential for disappointment, misunderstanding, conflict, and various types of communication breakdowns, which, if not handled well, can have negative consequences for both parties. Families can experience lapses in telephone service or fail to be home at an appointed meeting time. Similarly, mentors may leave phone messages or send e-mails that young people never receive or hold unrealistic expectations regarding the nature, promptness, or frequency of communication they expect to receive. Volunteers may also neglect to inform their protégés of travel plans or cancel meetings when work or school demands become overwhelming. Whereas to an adult mentor such interruptions or cancellations may be regarded as the expectable consequences of a demanding schedule, to a youth protégé they may be experienced as significant disappointments or even as negative indications of the mentor’s regard for him or her and commitment to the relationship. Mentors should be expected to bear the greater responsibility for finding ways to honor plans and

516   PRACTICE AND PROGRAMMATIC CONSIDERATIONS commitments, to effectively and consistently communicate with their younger protégés, and to seek guidance from mentoring program staff should they find that they are unable to do so. Mentors should also conduct themselves with integrity in their protégés’ schools, homes, and communities by being respectful of customs and regularities and by not acting in ways that leave programs having to run interference. Issues of integrity can also surface around the exchange of money. Although there are always exceptions, mentors should be wary of entering into financial arrangements with protégés or their families. It might seem harmless for a mentor to cover one month’s electricity bill, particularly when rationalized in terms of helping the protégé, but this may have the unintended effect of establishing expectations of further (and perhaps greater) assistance. Later unwillingness or inconsistencies in a mentors’ willingness to pay bills could then diminish the integrity of the relationship. There are many ways, however, in which a volunteer can express generosity (e.g., spending time, helping to establish connections to other resources) without complicating the relationship. Promote Justice for Young People (Justice) The principle of promoting justice calls for mentors to exercise good judgment and to take precautions to ensure that the potential biases inherent in their own backgrounds do not lead to prejudicial treatment of their protégés. The standards of the APA (2002) Ethics Code stipulate that psychologists should not “engage in unfair discrimination based on age, gender, gender identity, race, ethnicity, culture, national origin, religion, sexual orientation, disability, socioeconomic status, or any basis proscribed by law” (p. 5). We would argue strongly that the same holds true for mentors. Mentors should not engage in unfair treatment. However, as noted previously, the reality is that the largest proportion of volunteer mentors are White, middleclass students and professionals, whereas protégés tend to be more economically and ethnically diverse (MENTOR, 2006). In one instance that came to the attention of an author of this chapter, a volunteer’s suggestion that his African American protégé serve as a golf caddy in his country club enraged the protégé’s mother. The volunteer’s seemingly generous gesture unwittingly ignited racial tensions. Differences in cultural backgrounds and values may also lead volunteers to unwittingly act on cultural biases. In one instance, a mentor was invited by her protégé to attend a religious ritual to exorcise evil spirits from the apartment of the protégé’s family.

The mentor struggled with her case manager to keep her own beliefs in check so as to remain open and nonjudgmental throughout the ceremony. Volunteers should consistently engage in self-examination and seek out supervision/consultation to avoid making assumptions about mentees that are based on, or insensitive to, the latter’s background in areas such as social class, gender, religion, or disability. Mentoring programs have an obligation to provide training in cultural and gender sensitivity so as to raise volunteers’ awareness of their own biases and blind spots. Programs may assume that once a mentoring relationship has been formed, the strength of the bond will mitigate against potential misunderstandings and miscommunications that may arise as a result of differences in cultural values and backgrounds. The sparse research on same- and cross-race matches does lend some provisional support for the assumption that a strong bond can offset cultural differences (see Sánchez et al., this volume, Chapter 10). Nevertheless, unacknowledged prejudices can subtly affect interpersonal relationships (Cohen & Steele, 2002). Promoting justice can also extend beyond the one-to-one relationship with a protégé. By bringing more adults who have educational and financial resources into the lives of young people with reduced access to such resources, mentoring has the potential to promote social change (see Deutsch et al., this volume, Chapter 12). In developing close personal connections with vulnerable youth, mentors have the opportunity to develop a firsthand understanding of the challenges faced by young people today. Theoretically, this can inspire mentors to engage in efforts to redress social ills and to advocate for social change that could improve the health and well-being of all youth living in difficult circumstances. Respect the Young Person’s Rights and Dignity (Respect for People’s Rights and Dignity) The ethical principle of respecting people’s rights and dignity is rooted partially in the moral principle of self-determination. In the context of youth mentoring, one key implication of this principle is the need for mentors to respect the choices of their protégés (and their families). Although volunteers may sometimes feel they know what is in their protégés’ best interests (e.g., enrolling in college prep courses), and may even want to override what they see as shortsighted decisions, the reality is that they do not have full access to or awareness of the many concerns and values that shape the decisions of their protégés and their families. A well-meaning volunteer once arrived at his

Ethical Issues in Youth Mentoring   517 protégé’s apartment with purchases from a storage and reorganization store, announcing plans to help the child organize what the volunteer perceived as the child’s cluttered homework environment. Although insulted, the child’s mother refrained from intervening. She looked up to the mentor (a college graduate), saw her son’s excitement, and questioned her own concerns. The foregoing considerations lead us to conclude that, except in extreme situations (e.g., abuse, neglect, endangerment), volunteers should seek to understand and respect the decisions and lifestyle of a young person and his or her family (Beauchamp & Childress, 2008). Respect for self-determination involves behaving in ways that enable rather than interfere with the ability of protégés and their families to exercise their own reasoning and judgment. Mentors should seek to understand the youth’s own personal goals, desires, and values and not undermine the young person’s capacity to make his or her own decisions. Such an approach not only conveys respect, it also offers protégés opportunities to experience appropriate power and control in the mentoring relationship, which can promote feelings of self-efficacy and competence (Prilleltensky, Nelson, & Peirson, 2001). Adherence to the rights and dignity principle also includes being attentive to the protégé’s right to privacy and confidentiality. Issues of confidentiality, which abound in youth mentoring relationships, have been given insufficient attention. Youth and parents often disclose deeply personal information to volunteers, sometimes with specific injunctions against sharing it with the other. And, indeed, youth have noted the importance of having a place to share private thoughts and feelings (Spencer, Jordan, & Sazama, 2004). Mentors can serve as important sounding boards, particularly in adolescence when youth are exploring their identities and may experience new forms of conflicts in their relationships with their parents (Allen, Moore, & Kuperminc, 1998). Yet, as noted previously, serving as a confidante to young people can involve difficult decisions about what information can or should be kept confidential and what needs to be disclosed. Despite such complications, volunteers are rarely trained in the nuances of managing sensitive information, including situations in which confidence should not be maintained. Professional helpers are trained to disclose the limits of confidentiality, and mentors and the programs that support them should follow suit. As such, protégés should be informed of their mentors’ obligation to break confidence should they disclose intentions to be harmful to themselves or

others. Similarly, mentors are ethically (and, in some states, also legally) obliged to report to program staff any suspicions that their protégés are the subject of abuse or neglect.

Practice Implications Very little research exists on the specific role that ethics and ethical behaviors play in the youth mentoring process. Indeed, there is a vital need for research on ethical considerations in mentoring. This includes, for example, qualitative research on ethical dilemmas and how they are addressed by mentors and program staff. Moreover, as we advance our knowledge about topics with ethical implications (e.g., volunteer screening, match duration, optimal training), findings will shed additional light on issues with ethical implications. Nonetheless, this chapter focuses on the practical implications of the foregoing guidelines. Avoiding harm requires that mentors be trained to be attuned to the potential ethical problems described earlier in this chapter and in appropriate strategies for preventing and/or mitigating problems as they arise (see Kupersmidt & Rhodes, this volume, Chapter 30). With this in mind, we offer a series of research-informed recommendations for fostering relationally and ethically attuned mentoring relationships for youth (see Table 35.1). Screen and Train Mentors A growing body of research (e.g., Herrera et al., 2007) suggests that close, ethical mentoring relationships require adequate prematch screening and training of mentors. Prescreening may provide information useful for identifying mentors who will need extra support and training in fostering the qualities necessary for relationally and ethically attuned mentoring. Indeed, mentors’ ratings of their prematch efficacy for serving in the role of mentor has been shown to predict increased levels of mentor/youth contact as well as more positive relationship experiences (Parra, DuBois, Neville, Pugh-Lilly, & Povinelli, 2002). In addition to a sense of self-efficacy and relevant past experience, available research suggests a number of skills, attributes, and relational qualities (i.e., authenticity, engagement, empowerment, reciprocity, and an authoritative style) that might be useful to consider in prescreening mentors with regard to their likelihood for success in fostering relationally and ethically attuned mentoring relationships with youth (Rhodes & DuBois, 2006). Training may increase both mentors’ perceived and actual efficacy as they approach morally

518   PRACTICE AND PROGRAMMATIC CONSIDERATIONS Table 35.1   Guiding Questions for Practitioners Are volunteers screened carefully for their appropriateness and capacity to serve as mentors and prepared to engage in ethically responsible relationships with their protégés? Are mentors sensitized to the ethical nature of youth mentoring relationships and the kinds of ethical dilemmas that can arise? Are skilled and knowledgeable program staff readily available to assist mentors confronted with ethical dilemmas in generating possible courses of action and considering their ethical implications? Are the program guidelines and expectations clearly delineated and communicated to all participants, setting the basic frame for ethical conduct in the mentoring relationship at the outset? Are mentor behaviors (frequency, duration, and nature of contacts with youth) and the mentoring relationship closely monitored and supported with specific attention to potential ethical issues and concerns? Are there opportunities to engage mentors in a process of ongoing self-evaluation regarding the ethical dimensions of their relationships with protégés? Do the program’s design and practices ensure that the rights and dignity of the youth and their families are respected?

ambiguous and difficult situations. In addition, research indicates that mentoring styles should be fine-tuned to the characteristics and contexts of the participants. Without training in specific competencies, the most well-intentioned mentors may make critical errors that negatively impact their relationships with youth (Kupersmidt & Rhodes, this volume, Chapter 30). Mentor training with respect to issues relating to gender and race is critical given the shortage of same-sex, same-race mentors, particularly for males and youth of color (Rhodes, 2002). For example, mentors should be aware of the subtle but important differences in the relational styles and needs of boys and girls (see Liang et al., this volume, Chapter 11). Moreover, mentor training should attend to the developmental characteristics of youth populations served. For example, a qualitative study of middle school, high school, and college students found that, although protégés of different developmental stages agreed on the basic attributes of good mentors, they placed different values on self-determination and autonomy in the relationship (Liang et al., 2008). A strong program infrastructure that supports the careful screening and training of mentors is likely to be critical to the formation of close relationships between youth and mentors and, ultimately, to the effectiveness of both current and newly evolving forms of mentoring. No matter how well a mentoring (or any other type of) program is designed and conceptualized, it will not achieve its potential benefits if implementers lack the training and

organizational support needed to carry it out with fidelity (Durlak & DuPre, 2008). Sensitize Mentors to Identify Ethical Issues and to Engage in Ethical Behaviors As mentoring is often portrayed as simply spending time and having fun with a young person, the ethical obligations of serving as a mentor may not be readily apparent to many volunteers. Yet, even when the skills required to be a mentor are downplayed in marketing or recruitment materials, a mentor is expected to form a meaningful bond that “makes a difference” in the life of youth. In the training of professionals engaging in other types of relationship-based interventions, the importance of becoming sensitized to the responsibilities associated with one’s roles, and the fact that one’s actions have ethical consequences that potentially harm as well as help, have long been recognized (Kitchener, 1986). We would urge programs to prepare mentors to become similarly equipped to frame challenging situations as ethical dilemmas. Such awareness, especially if present from the start of a mentoring relationship, may prove pivotal in shaping the nature, course, and impact of the mentor-mentee bond. Indeed, we expect that awareness of and sensitivity toward ethical issues is likely best achieved by incorporating discussion of these issues into the various stages of the mentoring process— recruitment, screening, training, matching, and sustaining the relationship over time.

Ethical Issues in Youth Mentoring   519 Establish and Communicate Clear Expectations and Program Guidelines Given that appropriate ethical behavior depends in part on the nature of the relationship between the parties involved, it is important that programs have clear guidelines regarding the structure of the program and the expectations for mentor, youth, and family participants. To avoid unrealistic expectations and the blurring of role boundaries, explicit definitions should be provided to participants about the role and limits of mentoring relationships. Moreover, mentor training would do well to incorporate role-playing opportunities for mentors so as to provide practice in recognizing and negotiating situations in which there are mismatches between participant expectations or boundary violations. Finally, expectations regarding mentor commitments such as the duration of the relationship should be made clear and mentors made aware of the pitfalls of premature termination. When terminations of mentoring relationships are necessary, they should be planned carefully and the child and his or her family should be given ample notice and explanation (see Spencer & Basualdo-Delmonico, this volume, Chapter 32). Monitor and Support Mentor Behaviors and the Mentoring Relationship A mentor may enter into a new relationship with a young protégé with the best of intentions. However, what may have seemed like a workable commitment at the beginning of the relationship can become challenging to keep over time. Monitoring and supporting mentor behaviors, such as the regularity and duration of meetings and the nature of activities, both conveys the seriousness of the obligation to the young person and creates the opportunity for program staff to identify potential problems as they arise (MENTOR, 2009). Moreover, providing support to mentors throughout the course of their mentoring relationships can serve a powerful, proactive, and preventive purpose, especially for those mentors who are struggling with fatigue or discouragement that might otherwise adversely affect their relationships with protégés. Even with guiding principles, ethical dilemmas and moral ambiguities often arise that are not resolved easily. Because selecting the right course of action is rarely easy or straightforward, a process is needed by which case managers and volunteers can work together to generate and evaluate possible courses of action in light of the ethical principles set forth earlier in this chapter. In addition, ongoing

support may help mentors to better understand and relate to protégés of diverse backgrounds and to create more positive protégé outcomes (Diversi & Mecham, 2005). Engage Mentors in a Process of Ongoing SelfEvaluation Regarding Values and Culture As we emphasized throughout this chapter, the mentor and youth mentee often come to their relationship with different sets of values and cultural experiences. Mentors can help shape the meaning young people make regarding their race, ethnicity, and identities, and how this meaning is used to advance self-understanding (Liang & Grossman, 2007). Guidelines pertaining to multicultural education and training put forth by the APA (2003) caution that even the most thoroughly trained helping professionals are vulnerable to unconscious biases and stereotypic attitudes about people in the “outgroup,” such as individuals in racial or ethnic minority groups. These hidden biases and preconceived notions have the potential to influence the way mentors work with culturally diverse protégés. A key principle of culturally competent mentoring in our view (and that of Sánchez et al., this volume, Chapter 10), therefore, is that mentors are trained to engage in critical self-assessment, as captured in the popular adage “[doctor/counselor/mentor], know thyself” (Sue, Arredondo, & McDavis, 1992; Sue & Sue, 2003). Although such a process of selfreflection can be introduced and begun in prematch training, differences in worldviews, values, and cultural influences can emerge and become prominent over time. Follow-up training that encourages mentors to reflect on their own values and assumptions may help increase their capacity to identify ethical dilemmas and respond to them in more nuanced and relationally attuned ways. Ensure That Program Designs Respect the Rights and Dignity of the Youth Mentoring that is ethically and relationally attuned is fostered in the context of programs designed to be a good fit with youth populations served (Britner, Balcazar, Blechman, Blinn-Pike & Larose, 2006). More specifically, mentoring outcomes reflect the ways in which mentors respond to multiple characteristics of the protégé. Of course, a mentor’s ability to respond appropriately to a youth depends on the context—including the setting and purpose of the mentoring, cultural factors, and the young person’s developmental stage. Such factors shape the relationship context and determine how

520   PRACTICE AND PROGRAMMATIC CONSIDERATIONS certain ethical principles should be applied (Shore, Toyokawa, & Anderson, 2008). For example, gender-specific programs may backfire when adhering too closely to gender stereotypes and conventionality (Bay-Cheng, Lewis, Stewart, & Malley, 2006). Bay-Cheng et al. (2006) described how a program designed to foster “feminist” ideals, such as empowerment and giving voice, unwittingly stifled girls’ voices through stopping, ignoring, and shaping talk (especially talk about sex). Based on their findings, the authors suggested that mentoring programs adopt more “sex-positive” approaches to adolescent sexuality and engage mentors in dialogues of understanding differences in sexual attitudes that may arise through differences in generation and ethnicity. Moreover, these researchers argued that programs should prepare mentors to focus on listening to the interests of youth with less presumption, rather than imposing values.

Conclusion A mentor entering into a personal relationship with an unrelated youth is charged with safeguarding the welfare of that young person. Although youth mentoring shares many of the conventions of more professionalized helping relationships, the informal and voluntary nature of this endeavor releases it from some of the more rigid protections that govern such ties. Nonetheless, mentors often enter into the lives of youth in very personal ways and, as such, should be held accountable for protecting their rights and welfare. Programs have an obligation to sensitize volunteer mentors to the ethical issues that can arise when working with unrelated youth, to offer guidelines for how to support and nurture the positive development of young persons, and to actively foster the fulfillment of these expectations through training and ongoing support. Good intentions alone are not enough to ensure that mentors will build relationships with their youth mentees that are helpful and not harmful. Lack of attention to ethical issues has many pitfalls and can ultimately lead to poor and even harmful decisions. Mentors will be more likely to frame their quandaries as falling within the purview of ethics if they have been sensitized to such issues from the beginning of their relationships (Kitchener, 1986). Within this context, mentors can be prompted to think about whether and how the actions they take may affect youths’ welfare and mentoring relationships. Mentors must be willing to tolerate ambiguity as they determine the best course of action. The complex

nature of youth mentoring relationships, and the many sociodemographic divides that these relationships tend to cross, call for a reasoned, informed, and self-reflective approach. The ethical principles delineated in this chapter are purposely broad. A more detailed set of guidelines would be difficult to apply to the many different types of mentoring relationships in which youth and adults engage and the varying contexts within which they take place. Further, it is our view that more prescriptive guidelines would likely limit mentoring relationships in ways that could inhibit the open and natural qualities that make them special relationships in the lives of youth. As such, we favor an approach in which the intent is not to define for mentors or mentoring programs the parameters of relationships, but rather to provide touchstones to aid in the navigation of the inherently complex and, at times, uncertain terrain of relationships between adult mentors and youth. In light of the expanding modes of forging and maintaining mentoring ties (e.g., electronically mediated relationships, groups), it is impossible to predict the innumerable situations that may arise between mentors and youth. The principles discussed in this chapter and associated recommendations for practice are geared instead to raise ethical awareness and to encourage more studied reflection on the complex situations that inevitably arise when caring adults enter the lives of unrelated youth. We have offered a set of ethical guidelines to help mentors as they strive to build and sustain meaningful relationships with their protégés. However, these guidelines are not enough. We need much more than just a “rule book” based on an agreed-upon morality of youth mentoring as it is currently practiced. Given that a major purpose of ethical guidelines should be to strengthen mentoring relationships in ways that ultimately benefit youth protégés, we need research into the specific factors that foster relationally and ethically attuned mentoring relationships and into the effectiveness of varying methods for translating such research into practice. Moreover, although general guidelines provide needed direction and structure for ethical decision making, they should not lure us into believing that we need not think carefully about the larger questions and issues that might not appear to be ethical and moral ones at first glance. Ethical principles may be applied and interpreted in many situations, but in an infinite number of situations, ethical principles cannot speak—in these cases, mentors in consultation with others must make their own decisions. Moreover, as we stated at the outset, the intent of this chapter is to foster greater attention

Ethical Issues in Youth Mentoring   521 among practitioners and researchers alike regarding the ethical aspects of youth mentoring. It is imperative that we recognize that mentoring work with youth is infused with moral issues and that we cast light on those moral aspects not yet made visible.

References Allen, J. P., Moore, C., & Kuperminc, G. (1998). Attachment and adolescent psychosocial functioning. Child Development, 69, 1406–1419. American Psychological Association. (2002). Ethical Principles of Psychologists and Code of Conduct. Washington, DC: Author. American Psychological Association. (2003). Guidelines on multicultural education, training, research, practice, and organizational change for psychologists. American Psychologist, 58, 377–402. Bay-Cheng, L. Y., Lewis, A. E., Stewart, A. J., & Malley, J. E. (2006). Disciplining “girl talk”: The paradox of empowerment in a feminist mentorship program. Journal of Human Behavior in the Social Environ­ ment, 13, 73–92. Beauchamp, T. L., & Childress, J. F. (2008). Principles of biomedical ethics (6th ed.). New York: Oxford University. Bernstein, L., Dun Rappaport, C., Olsho, L., Hunt, D., & Levin, M. (2009). Impact evaluation of the U.S. Department of Education’s Student Mentoring Program. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education. Retrieved from http://ies.ed.gov/ncee/ Big Brothers Big Sisters of Canada. (2006). Risk management handbook. Retrieved from http://www.ecvo.ca/ Risk_Management_Handbook_2006.pdf Britner, P., Balcazar, F., Blechman, E., Blinn-Pike, L., & Larose, S. (2006). Mentoring special youth populations. Journal of Community Psychology, 34, 747–763. Cohen, G. L., & Steele, C. M. (2002). A barrier of mistrust: How negative stereotypes affect cross-race mentoring. In J. Aronson (Ed.), Improving academic achievement: Impact of psychological factors on education (pp. 303–327). San Diego, CA: Academic Press. Diversi, M., & Mecham, C. (2005). Latino(a) students and Caucasian mentors in a rural after-school program: Towards empowering adult-youth relationships. Journal of Community Psychology, 33, 31–40. Downey, G., Lebolt, A., Rincon, C., & Freitas, A. L. (1998). Rejection sensitivity and children’s interpersonal difficulties. Child Development, 4, 1074–1091. DuBois, D. L., Neville, H. A., Parra, G. R., & Pugh-Lilly, A. O. (2002). Testing a new model of mentoring. In G. G. Noam (Ed. in chief) & J. E. Rhodes (Ed.), A critical view of youth mentoring (New Directions for Youth Development: Theory, Research, and Practice, No. 93, pp. 21–57). San Francisco: Jossey-Bass.

Durlak, J. A., & DuPre, E. P. (2008). Implementation matters: A review of research on the influence of implementation on program outcomes and the factors affecting implementation. American Journal of Community Psychology, 41, 327–350. Fisher, C. B. (1997). A relational perspective on ethics-inscience decision making for research with vulnerable populations. IRB: A Review of Human Subjects Research, 19, 1–4. Fisher, C. B., Hoagwood, K., & Jensen, P. S. (1996). Casebook on ethical issues in research with children and adolescents with mental disorders. In K. Hoagwood, P. S. Jensen, & C. B. Fisher (Eds.), Ethical issues in mental health research with children and adolescents (pp. 135–266). Hillsdale, NJ: Erlbaum. Fisher, C. B., & Younggren, J. N. (1997). The value and utility of the 1992 Ethics Code. Professional Psychology: Research and Practice, 28, 582–592. Grossman, J. B., Chan, C., Schwartz, S., & Rhodes, J. E. (2012). The test of time in school-based mentoring: The role of relationship duration and re-matching on academic outcomes. American Journal of Community Psychology, 49, 43–54. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring programs. American Journal of Community Psychology, 30, 199–219. Herrera, C., Grossman, J. B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a difference in schools: The Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia: Public/Private Ventures. Kalbfleisch, P. J. (1997). Appeasing the mentor. Aggressive Behavior, 23, 389–403. Karcher, M. J. (2005). The effects of school-based developmental mentoring and mentors’ attendance on mentees’ self-esteem, behavior, and connectedness. Psychology in the Schools, 42, 65–77. Keller, T. E. (2005). A systemic model of the youth mentoring intervention. Journal of Primary Prevention, 26(2), 169–188. Kitchener, K. S. (1986). Teaching applied ethics in counselor education: An integration of psychological processes and philosophical analysis. Journal of Counseling & Development, 64, 306–310. Liang, B., & Grossman, J. (2007). Diversity and youth mentoring. In T. Allen & L. Eby (Eds.) Blackwell handbook of mentoring: A multiple perspectives approach (pp. 239–258). Malden, MA: Blackwell Publishing. Liang, B., Spencer, R., Brogan, D., & Corral, M. (2008). Mentoring relationships from early adolescence through emerging adulthood: A qualitative analysis. Journal of Vocational Behavior, 72, 168–182. MENTOR. (2006). Mentoring in America 2005: A snapshot of the current state of mentoring. Alexandria, VA: Author. MENTOR. (2009). Elements of effective practice for mentoring (3rd ed.). Alexandria, VA: Author. Retrieved

522   PRACTICE AND PROGRAMMATIC CONSIDERATIONS from http://www.mentoring.org/program_resources/ elements_and_toolkits Miller, A. (2007). Best practices for formal youth mentoring. In T. D. Allen & L. T. Eby (Eds.), The Blackwell handbook of mentoring: A multiple perspectives approach (pp. 307–324). Malden, MA: Blackwell Publishing. Parra, G. R., DuBois, D. L., Neville, H. A., Pugh-Lilly, A. O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367–388. Prilleltensky, I., Nelson, G., & Peirson, L. (2001). The role of power and control in children’s lives: An ecological analysis of pathways toward wellness, resilience and problems. Journal of Community & Applied Social Psychology, 11, 143–158. Rhodes, J. E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press. Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20(3), 3–19. Rhodes, J. E., Liang, B., & Spencer, R. (2009). First do no harm: Ethical principles for youth mentoring relationships. Professional Psychology: Research and Practice, 40, 452–458. Scales, P. C. (2003). Other people’s kids: Social expectations and American adults’ involvement with

children and adolescents. New York: Kluwer Academic/Plenum. Shore, W. J., Toyokawa, T., & Anderson, D. D. (2008). Context-specific effects on reciprocity in mentoring relationships: Ethical implications. Mentoring & Tutoring: Partnership in Learning, 16, 17–29. Slicker, E. K., & Palmer, D. J. (1993). Mentoring at-risk high school students: Evaluation of a school-based program. School Counselor, 40, 327–334. Spencer, R. (2007). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22, 331–354. Spencer, R., Jordan, J. V., & Sazama, J. (2004). Growthpromoting relationships between youth and adults: A focus group study. Families in Society, 85, 354–362. Strom-Gottfried, K. (2008). The ethics of practice with minors: High stakes, hard choices. Chicago: Lyceum Books. Sue, D. W., Arredondo, P., & McDavis, R. J. (1992). Multicultural counseling competencies and standards: A call to the profession. Journal of Counseling & Development, 70, 477–486. Sue, D. W., & Sue, D. (2003). Counseling the culturally diverse: Theory and practice (4th ed.). New York: Wiley. Wiley, T. (2011). Federal background checks vital for youth programs. Youth Today, 16.

PART VIII Conclusions

36  YOUTH MENTORING: PROGRESS AND PROSPECTS FOR THE 21ST CENTURY David L. DuBois and Michael J. Karcher

Introduction In keeping with our goal as editors to provide a comprehensive treatment of issues of contemporary importance for youth mentoring, the contributions to this handbook cover a great deal of ground. Attempting to summarize or capture even just the highlights of this content in the space available to us in this concluding chapter would, in our estimation, do poor service to the in-depth and nuanced examination that the widely varying topics have received. We do see value, however, in drawing on this material to explore two key questions. First, how can mentoring be made available to as many youth in need as possible? Second, how can the mentoring that youth do experience be made as high quality and effective as possible? These questions call attention, of course, to issues of both quantity and quality, respectively. In his eloquent commentary in the foreword, David Shapiro underscored the potential for mentoring to make a truly meaningful and lasting difference in the life prospects and contributions of current and future generations of youth—in essence, his is an ambitious vision of societal-level impact. Realization of this lofty goal arguably will require, in equal measure, that mentoring be made accessible to those youth who could benefit from it and that the mentoring youth receive is of the highest quality, offering ample prospects for enduring benefit and few, if any, risks for harm. With this understanding, we begin this concluding chapter by taking stock of the state of affairs with respect to both the availability and the effectiveness of mentoring for young people. In doing so, we make our best effort to give balanced

consideration to indications of both promise and concern. We also probe for trends over time as best we can with available data. Our working assumption is that such trends may tell us something about the degree to which efforts to date are yielding desired results and thus the extent to which new or enhanced strategies may be needed to achieve societal-level impact. Informed by this analysis and drawing on the rich and varied contributions to this handbook, we then explore future prospects for the field’s advancement. We organize our discussion around the following four broad areas: the frameworks and perspectives used for understanding the dynamics and significance of mentoring as it is experienced by developing youth, the roles and characteristics of the persons who are actively engaged and supported in mentoring young persons, implications of changes occurring at a societal level, and the bridging of research with practice and policy.

Current Status of the Quantity and Quality of Mentoring For Youth Quantity and Access Point Estimates. America’s Promise—the Alliance for Youth (an outgrowth of America’s Promise, founded by former general Colin Powell in conjunction with the President’s Summit in 1997) includes as one of its five promises a commitment that all children and youth need and deserve support and guidance from caring adults in their families, schools, and communities (America’s Promise Alliance, 2006). Based on interviews conducted 525

526  Conclusions in 2005 and 2006 with a nationally representative sample of 6- to 17-year-old youth and their parents (both youth and parents were interviewed for the 12- to 17-year-old portion of the sample, whereas only parents were interviewed for 6- to 11-year-old youth), it was estimated that roughly one in five young people (20%) in the U.S. “do not have caring adults in their lives” (America’s Promise Alliance, p. 6) and thus the above referenced “promise” is not fulfilled. The criteria used in making this assessment were described and justified by the researchers as follows: Although the literature often refers to the importance of young people “having at least one” caring adult (Scales, Benson, Mannes, Hintz, Roehlkepartain, & Sullivan, 2003), this part of the Promise sets the standard higher. To have the overall Caring Adult Promise, young people must report having at least 2–5 adults in at least TWO of these three contexts [Extended Family, School, Neighborhood/Community]. If they have a MENTOR with whom they spend time at least ONCE A WEEK, this can substitute for not having either school or community adults. . . . These criteria reflect the important principle that redundancy of developmental opportunity throughout the youth’s ecology is related to better outcomes than positive experiences in only one context (see Benson, Scales, Hamilton, & Sesma, 2006; Jessor, Turbin, & Costa, 1998; Eccles & Gootman, 2002; Cook, Herman, Phillips, & Settersten, 2002), as well the reality that different children and youth may have differing paths they take to acquire the developmental resource of caring adults. (Scales et al., 2006a, p. 103). Youth also needed to respond favorably on items asking about parent support and communication to be categorized as having caring adults in their lives. Clearly, one could raise a number of questions about such criteria. To what extent, for example, do we have evidence to support the assumption that a formal mentor can compensate for a lack of supportive ties with adults (e.g., teachers) who are likely to be more deeply woven into the natural ecology of youths’ lives (e.g., school)? Similarly, what do we know about the exchangeability of supportive ties with adults across extended family, school, and community settings? Putting these questions aside, however, it is clear from these data that substantial numbers of young people lack the types of supportive ties with parents and other adults—an estimated 8.5 million youth based on the

foregoing criteria (America’s Promise Alliance, 2006)—that we have good reason to expect increased prospects for healthy development and later life success. Other aspects of these data also bear mention. First, only approximately 8% of youth were reported to have a relationship with a formal mentor (meeting the criterion of at least weekly contact during the past year). In contrast, supportive ties with other nonparental adults within the extended family, at school, and in the neighborhood/community were reported at rates of 88%–92% (Scales et al., 2006a). Such figures, which are consistent with prior research (for discussion, see DuBois & Karcher, 2005), underscore that at present the mentoring to which young people have access is much more likely to be outside the context of formal programs designed for this purpose. Second, although some observers might question whether existing prevalence data reflect a less concerning tendency for young persons to simply seek out and receive mentoring at the levels they desire and need, available data argue against this perspective. In the Voices Study, a mixed-methods investigation of 8–21 year olds that accompanied the National Promises Study, nearly half (42%) of the youth surveyed reported wishing they had more caring adults to whom they could turn for help (Scales et al., 2006a). Furthermore, there exists a wealth of research, much of which is captured in the contributions to this volume, to support the benefits for young persons of having more rather than less access to mentoring. In keeping with this understanding, the caring adults promise has been cast rightly as being of fundamental importance in part because such relationships are seen as “gateways” to the rest of the five promises (i.e., that young people also have safe places, good health care, encouragement in their school work, and chances to volunteer; America’s Promise Alliance, 2006). Youth for whom more promises are fulfilled, in turn, have significantly higher levels of social and academic competence, show better ability to avoid involvement in violence, and are more likely to be “giving back” to their communities through volunteering (America’s Promise Alliance, 2006). Inequities in Access. Data from the previously cited research also suggest that opportunities to access youth mentoring relationships are not distributed equitably across subgroups of young persons (Scales et al., 2006b). As would be expected, socioeconomic differences are salient in this regard. Illustratively, whereas among 12- to 17-year-old

Youth Mentoring: Progress and Prospects for the 21st Century    527 youth from the poorest families (income under $20,000), promise criteria were not met for nearly one in three youth, this was the case for only one in five youth from the highest family income category ($100,000+). Of further note, however, are developmental differences. Whereas the criteria for meeting the caring adults promise was met for over 90% of youth in the age ranges of 6–8 and 9–11 years, this proportion dropped to 81% for youth aged 12–14 and to only 70% for those aged 15–17. Putting these trends together, older youth from families with fewer resources are a group that seems especially likely to be lacking in access to opportunities for mentoring. Clearly, too, differences in access may be relevant to other subgroups, such as those with disabilities (McDonald, Balcazar, & Keys, 2005) and those in certain regions of the world (Goldner & Scharf, this volume, Chapter 13). Trends Over Time. Reliable data on trends over time in the availability of mentoring for young people are difficult to come by. One useful resource, however, are data obtained through the Monitoring the Future study, a recurring national survey of representative samples of students in grades 8, 10, and 12 in the United States (Monitoring the Future, 2012). Since 1992, this survey has included a question asking youth if there is at least one adult, other than a parent, who they would feel able to talk to if they were having problems in their life. Response options are “No,” “Yes–some problems,” and “Yes–all problems.” A parallel question asks youth if they could talk to their parents about problems. To inform our understanding of possible trends over time in youths’ access to mentoring, we examined public-use data­ sets that are available for this research (Substance Use and Mental Health Data Archive, n.d.). We found that, in 1992, 22.9% of responding 8th graders said they did not have a nonparental adult in their lives to whom they could talk about problems they might be having. Furthermore, among those who indicated that they could not talk to their parents about problems, nearly half (48.6%) also reported lacking a nonparental adult who could serve in this capacity. A troubling subgroup of youth, comprising about 10% of the sample, denied having any adult in their lives from whom they could seek help with personal concerns. In 2010, the latest year for which data are available, a slightly greater proportion of youth (25.2%) reported lacking such a nonparental adult, and the more troubling group of youth without either parental or nonparental support had increased by a half, to roughly one in seven youth (14.5%). Whether these data reflect a meaningful and reliable decrement in the availability of relationships

that can offer mentoring to youth is beyond the scope of the present discussion. What does seem clear, however, is that despite the “mentoring movement” (Rhodes & DuBois, 2006) that has unfolded over the past two decades or so, and resulting increases in the number of youth who have access to formal mentors through programs (MENTOR, 2006), the overall proportion of the nation’s young persons who report a lack of ties with adults conducive to mentoring has not been reduced in a manner commensurate with the significant level of societal investment made in this area. The extent to which youth in different regions of the world benefit from the constructive support and guidance of nonparental adults in their lives is not well documented. Yet, viewing the issue of access to mentoring through this type of wider lens likely will be essential when seeking to further a global vision of health and prosperity for young people. Quality and Effectiveness Overall Evidence. As we have noted, independent of issues of quantity and availability, an equally important concern is the quality and effectiveness of the mentoring to which young people have access while growing up. For present purposes, we are especially interested in the extent to which mentoring as it typically occurs in the lives of youth is able to foster healthy development and, in so doing, establish a foundation for enduring well-being, success, and contribution throughout adulthood. The offsetting potential for mentoring to do serious harm to young persons in some circumstances, however, must also be taken into account. As summarized in this volume, numerous studies have found an association between the reported presence of a so-called natural (i.e., typically not assigned by a program) mentor in the lives of children and adolescents and positive outcomes, including mental health, avoidance of involvement in high-risk behavior, better school performance, and indicators of more favorable functioning during and following the transition to adulthood, such as educational attainment and employment (for additional reviews, see Spencer, 2007b, and Zimmerman, Bingenheimer, & Behrendt, 2005). The methodologically strongest of these studies have been prospective, thus linking mentoring relationships with subsequent improvement in outcomes over time (see, e.g., Hurd & Zimmerman, 2011; Kogan, Brody, & Chen, 2011), and based on nationally representative samples of youth (see, e.g., DuBois & Silverthorn, 2005; Erickson, McDonald, & Elder, 2009). As is well documented in contributions to this

528  Conclusions volume, positive outcomes have been linked to youths’ relationships with adults and older peers that offer potential for mentoring within a wide range of contexts, including their extended families, schools, the workplace, and after-school and service-learning programs, and across a range of societal cultures. Cumulatively, this body of literature clearly supports the idea that mentoring relationships as experienced by youth in a range of differing facets of their dayto-day lives are often of high quality and impactful, potentially in ways that may have profound implications for their overall life course trajectories. At the same time, such findings are limited in their ability to shed definitive light on the quality or effectiveness of the mentoring that youth receive, for at least two reasons. The first of these has to do with the ways in which relationships have been assessed and identified. For the most part, studies have focused on just one mentoring relationship and in doing so typically have instructed youth to consider the person most important or significant to them. In addition, the definition or set of criteria that youth are provided often has incorporated an assumption of a high-quality and, in some instances, also positively impactful relationship. Consider, for example, the following instructions for the mentoring assessment from the National Longitudinal Study of Adolescent Health (Add Health), which has been used as the basis for numerous studies: Other than your parents or step-parents, has an adult made an important positive difference in your life at any time since you were 14 years old? . . . If there has been more than one person, describe the most influential. When such elements of selectivity (and even circularity) are introduced into assessments, the associated data should perhaps be regarded as offering insight into the quality and effectiveness of mentoring under ideal rather than necessarily typical circumstances. A second concern has to do with the design of the studies involved. Because naturally occurring mentors are just that (i.e., not assigned experimentally), associations with outcomes may be inflated by confounding characteristics of youth, such as a tendency for those with the highest levels of social competence to be most apt to have forged such relationships (Zimmerman et al., 2005). In view of these limitations, it is difficult to offer more than tentative and qualified statements about the overall quality and effectiveness of mentoring that youth receive from the full spectrum of adults and influential older peers in their lives during their development. Still,

the evidence speaks positively to the potential for the mentoring that youth experience to be both of high quality and impactful. Evaluations of formal mentoring programs for youth provide another useful window into questions relating to mentoring quality and effectiveness. Especially when experimental (random assignment) designs are used to determine which youth in studies participate in the program and thus are assigned a mentor, estimates of impact are relatively free from confounding influences such as those noted above (Grossman, 2005). In the most recent comprehensive meta-analysis of youth mentoring program evaluations (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011), which encompassed 73 independent evaluations published between 1999 and 2010, findings offered support for the effectiveness of mentoring for improving outcomes across behavioral, social, emotional, and academic domains of young people’s development. One of the most common patterns observed was for mentored youth to exhibit positive gains on outcome measures while nonmentored youth exhibited declines, thus suggesting that the effectiveness of mentoring extends to both promotion and prevention aims. The magnitude of the benefits observed, while small or modest by most conventions for gauging intervention impact, was nonetheless within the range reported for other types of “competing” forms of school- and community-based interventions (e.g., tutoring, afterschool programs). Also of note, unlike many of these other types of programs, individual mentoring programs also showed evidence of being able to affect multiple domains of youth functioning simultaneously, supporting the idea that mentoring can be broadly beneficial for any given young person. Overall, available evidence appears to augur well for the quality and effectiveness of mentoring relationships that are made available to youth through formal programs (and, by implication, those that are naturally occurring, but sufficiently resemble these in key characteristics). As with research on naturally occurring mentoring relationships, there are also ample reasons to be tempered and qualified in such assessments (DuBois et al., 2011). These include, for example, a relative lack of attention in evaluations to assessment of key outcomes of policy interest (e.g., educational attainment, juvenile offending, substance use, obesity prevention) or the question of whether benefits that youth derive from mentoring are sustained at later points in their development. DuBois and colleagues noted that such gaps in the current knowledge base “make it difficult, if not impossible, to be confident that interventions relying on

Youth Mentoring: Progress and Prospects for the 21st Century    529 mentoring relationships as their ‘active ingredient’ are capable of reliably producing the types of enduring and transformative results that typically have been central to the arguments of their most enthusiastic supporters.” (p. 74). Variability. Relatively abundant evidence indicates that mentoring relationships, whether occurring more naturally or formed through program assignment, vary considerably in their quality and potential for positive impact (Rhodes & DuBois, 2006). More specifically, it appears that relationships with certain features are more likely to be of benefit for youth. These features include a close emotional bond, a collaborative rather than prescriptive approach on the part of the mentor, and attention to actively fostering the youth’s development through both psychosocial and instrumental forms of support such as active listening, guidance, and advocacy (de Tormes Eby et al., 2012; DuBois et al., 2011; Karcher, Herrera, & Hansen, 2010; Rhodes & DuBois, 2006). The implications of variation in quality extend beyond simple differences in the extent to which desired effects are realized. Available findings point toward negative consequences for relationships under circumstances such as the adult involved serving as a model for engaging in problem behavior (Beam, Gil-Rivas, Greenberger, & Chen, 2002), engaging in high levels of conflictual interaction with the youth (Hirsch, Deutsch, & DuBois, 2011), or, in formal programs, not fulfilling minimum commitments to the young person in terms of regularity or longevity of contact (Grossman & Rhodes, 2002; Karcher, 2008). Trends Over Time. With regard to potential trends over time, the overall level of effectiveness found in the meta-analysis by DuBois and colleagues (2011) is essentially the same (within the margin of error) as that found in a meta-analysis of an earlier generation of program evaluations (published prior to 1999) (DuBois, Holloway, Valentine, & Cooper, 2002). Neither meta-analysis found evidence of a change in effectiveness across the time period encompassed in the review. These considerations thus suggest a lack of discernible improvement in effectiveness in mentoring programs over time. There is, however, a trend across the two reviews toward less variability in effectiveness of the programs evaluated. This finding suggests that unevenness in the quality of mentoring programs may be dissipating over time, perhaps as a result of more consistent use of commonly recommended program practices (e.g., mentor training) (DuBois et al., 2011). Of further note, whereas the earlier review found a mixed record of

effectiveness for programs serving youth with preexisting difficulties (e.g., “individual [behavioral/ emotional] risk”), this was not the case in the metaanalysis of more recent evaluations. For these data, for example, findings indicated greater effectiveness for programs geared toward youth with already existing behavioral problems (DuBois et al., 2011). It may be the case, therefore, that programs also have become better equipped to serve relatively more vulnerable youth. However, with regard to all estimates of mentoring effectiveness, it should be kept in mind that they apply only to those programs that have been subjected to relatively rigorous evaluation. Accordingly, they may not be representative of the quality or impact of the much broader universe of programs in existence at any point in time.

Strategies to Improve the Quantity and Quality of Mentoring Available to Youth Quantity and Access Given some of the data reported earlier in this chapter, it seems we must accept that programs will never be able to fill the needs of all youth for multiple caring adults in their lives by using volunteers in mentoring programs. There can be no doubt that we also need more natural mentors. Until 1900—before the concept of strategic or formal mentoring programs for youth was envisioned by Coulter and Whitmer, two founders of the Big Brothers movement in the first decade of the 20th century (see Baker & Maguire, 2005)—all mentoring was natural. Mentor, the original mentor in Homer’s Odyssey was appointed by King Odysseus, so this “match” was not organic in the way we think of natural mentoring relationships forming. Nor was the relationship between Telemachus and Mentor a youth-initiated relationship. But it was natural in the sense of a parent recruiting known adults to assist them in parenting their youth. Of course, by this line of reasoning, enlistment of program-sponsored mentors by parents who enroll their children in a mentoring program may simply reflect the evolution of the process by which King Odysseus found a Mentor for his son. But there are important differences. Recent research on youth mentoring has not given natural mentors a seat at the table, so to speak, in terms of acknowledging their importance. Nor does the presence of natural mentoring research in the larger literature (as well as in this volume) acknowledge the likelihood that natural mentors may provide a more realistic means of reaching the larger number children in need than do mentors from

530  Conclusions “mentoring programs.” In fact, we should find it surprising that, just a century after the seeds of the Big Brothers Big Sisters program were sown, when we asked mentoring researchers to include research on “natural mentoring” in their chapters for this second edition of the Handbook of Youth Mentoring, many struggled to do so. How can this be? How can the concept of natural mentoring have become subordinate or viewed as ancillary or antiquated, when it has been for centuries the primary medium of mentoring? Quite to the contrary, fostering effective natural mentoring may be where the world needs to go, the next frontier of youth mentoring research, and deserving a more prominent place in the third edition of the Handbook of Youth Mentoring. The study of apprenticeships for adolescents is the one area of youth mentoring research that has retained a close affiliation with natural mentoring for many years. The work of Hamilton and Hamilton (2005; this volume, Chapter 19) and, more recently, Halpern (2005) and Larson (Larson, Hansen, & Moneta, 2006; Larson, Walker, & Pearce, 2005) on ways to help prepare adults in the workplace to identify and mentor apprentices may provide important leads for how to use natural mentors to serve adolescents better in schools and community settings. The evidence that some older adolescents may not respond as favorably as do children to unstructured, friendship-based youth mentoring in schools (e.g., Karcher, 2008; Wood & MayoWilson, 2012) means we need to rethink how best to help teens using mentors in schools. Some evidence indicates that natural mentoring relationships are critically important for older teens and, perhaps especially, for ethnic minority teens (Kogan et al., 2011; Sánchez, Reyes, & Singh, 2006). Therefore, we think that professionals in the field need to develop programs—ones that both inform researcher’s efforts and are informed by researcher’s findings—to foster natural relationships between teachers and teens, between coaches and teens, and perhaps between other school staff and teens. Fortunately, promising programs are attempting to deepen existing ties and cultivate new mentoring relationships between school staff and students in high schools (e.g., Anderson, Christenson, Sinclair, & Lehr, 2004; Holt, Bry, & Johnson, 2008). Interesting programs also are emerging that build bridges between the school and the workplace, and use mentors to help youth navigate the transition from one world to the other. In this way, youth can be exposed to workplace mentors long before they even think about working in a given setting or trade. In short, we feel more work must be done on how to cultivate existing relationships between youth and adults.

Another way to meet the needs of older youth—that is, another way to respond to the aforementioned research that more effective programs are needed in schools for teens, particularly teens whose parents are not likely to seek out programbased mentors for their children—reveals the importance of the paraprofessional movement and lessons we may need to relearn to help mentors better serve youth. One way to serve the developmental needs of older youth is to foster their leadership and responsibility by teaching these teens to be mentors. Karcher’s chapter (this volume, Chapter 16) points out that using teens as mentors is not a new practice but one that has been in development since the first efforts of the paraprofessional movement of the 1960s. That chapter, as well as the one by Karcher and Hansen on mentoring activities (Chapter 5), reveals that much interesting and instructive work was done prior to 2000 on the use of paraprofessionals and the preparation or training of paraprofessionals. The one thing we can be sure about the future is that there will likely never come a day when there are enough programmatic mentors to meet the needs of all youth who could benefit from a mentor. More than 50 years ago, Albee (1959) argued that even if the training of professional helpers increased significantly every year, we would not in our lifetime train enough to serve all those who need help. Mentoring will continue to be one way in which we try to fill the gap in the availability of professional helpers to support youth in need. But perhaps we could do better at this in the future if we looked more closely at the past. Goodman’s (1972) study on companionship therapy is not the only early study of paraprofessionals serving as mentors, but it may be the best. Goodman’s study of communication in match interactions yielded a specific set of “talking tools” that mentors could use in their work. To our knowledge, these tools have never been incorporated into mentor training, yet, ironically, they have been used to train professional helpers. (It is ironic because this research, funded by the National Institute of Mental Health, was intended to help us better understand how to support paraprofessionals who could compensate for the insufficient number of professional clinicians, who ultimately were the only ones to benefit from the study’s findings.) Similar work conducted almost 50 years ago on the use of peers as mentors to younger children (Hamburg & Varenhorst, 1972; Perlmutter & Durham, 1965; Reinherz, 1964)—much on a par in quality with the majority of today’s research on mentoring—is rarely included in literature reviews on mentoring in

Youth Mentoring: Progress and Prospects for the 21st Century    531 general or peer mentoring in particular. The research on paraprofessionals yielded other lessons that have never been brought to bear on the practice of (or research on) youth mentoring. We hope that more researchers will step back and consider what research on paraprofessionals may have to say about their areas of study and will be able to capitalize on these lessons to advance the field. Finally, another “out of the box” way to serve youth through mentoring is to use mentors to support mothers and fathers to better care for those in their charge. Several recent studies have looked at mentoring of parents (Bogat, Liang, & Rigol-Dahn, 2008; Dieterich, Landry, Smith, Swank, & Hebert, 2006), and we might be wise to consider the triadic nature of consulting relationships with adults who mentor parents and who thus help children indirectly. Such mentors also could be used to mentor those natural mentors we noted earlier—coaches, teachers, and other adults who interface with youth. How to mentor the natural mentors of youth represents an important frontier. Such efforts also might be more lasting than the direct effects of those who most often only temporarily mentor youth directly. But access alone will not solve the issue of youths’ needs for caring adults in their lives without access advancing alongside improvements in program quality and overall effectiveness. Quality and Effectiveness Following the breadcrumbs of research may lead to novel practices that could enhance the impact of youth mentoring through higher-quality programming. Innovative practices, from new methods of training as described by Kupersmidt and Rhodes (this volume, Chapter 30) to cuttingedge uses of technology as described by Shpigelman (this volume, Chapter 17), may lead to stronger mentoring programs and mentoring relationships. Better relationships also may increase the pool of available mentors by raising the rates of reenrollment of mentors who have had better experiences. In this way, program improvements may increase not just quality of mentoring but also the quantity of mentors available to youth. This second edition of the Handbook of Youth Mentoring has covered several new program approaches and match support techniques that could boost the quality and quantity of mentoring available to youth. Although the inclusion of new approaches and practices in this edition is largely a result of the body of research on each having grown to the point where there is something to say, “based on the evidence,” there is more to learn.

As one example of a practice included in this volume that may help to foster stronger matches— but on which there was only nominal research to base a chapter—consider the important practice of training mentors in what to say or do with their mentees. A resurgent interest in understanding the mechanisms of change in mentoring relationships has revealed the importance of studying the nature of mentor-mentee interactions. The chapter by Karcher and Hansen reveals the importance of recent typologies of mentoring relationships as well as the influence that the nature and timing of those relationship interactions can have on match quality (Goldner & Mayseless, 2008; Karcher & Nakkula, 2010; Keller & Pryce, 2012). Yet, while most recent research on mentoring styles references efforts to define developmental and instrumental relationship styles identified in the 1990s, we seem somewhat amnesiac about earlier findings, like those reported by Goodman (1972) over 40 years ago. Goodman’s research on youth mentoring revealed the importance of specific mentoring micro-skills (later called “talking tools,” as applied to therapy relationships; see Goodman & Dooley, 1976). There have been some interesting recent efforts to conduct microanalyses of mentoring interactions (Smith-Jentsch, Scielzo, Yarbrough, & Rosopa, 2008; Spencer, 2007a), but more needs to be done. Such efforts can directly inform mentor training and could therefore yield significant gains in programmatic impact once put into use. A second example of how alterations in program practices, informed by research evidence, may improve the quality of matches is addressed in the new chapter on matching procedures (Pryce et al., this volume, Chapter 29). We are only now beginning to accrue basic evidence of the importance of matching procedures (DuBois et al., 2011) and have just barely scratched the surface in our understanding of best practices in mentor-mentee matching. That chapter draws from other fields of research to suggest ways in which technology may enhance matching efforts. Although this high-tech future possibility is exciting, equally interesting is the approach of youth-initiated matches, about which we know almost nothing. These matches, in which youth identify significant adults in their lives, and programs serve to support and strengthen these matches, reflect the time-tested method in which mentees seek out the support of adults whom they see as useful, helpful, or simply interested in them (e.g., Bloom, Gardenhire-Crooks, & Mandsager, 2009; Schwartz, in preparation). This also reveals the ways in which lessons from natural mentoring relationships may help teach us better how to establish and support program-based matches. But the

532  Conclusions converse is also true: Research on natural mentoring may improve the quality of program-based matches, but research on program-based mentoring matches may help improve the quantity of natural mentoring that we can help make available to youth.

Conclusion Despite the proliferation of formal mentoring programs and of federal, foundation, and philanthropic support for establishing, operating, and evaluating such programs, we see little change in the past 20 years in the overall impact of youth mentoring programs. By contrast, the need for such programs has, if anything, increased over this same time period. The consistent program effect sizes may reflect the use of mentoring to work with more challenging youth, such as those whose parents have been incarcerated, youth in schools who would not have been brought to community-based programs but who can now be served in schools, and children in the foster care system. So, it may be that program effectiveness looks like it has not changed, but program quality improvements have occurred alongside delivery of programs to harder-to-reach children. But it seems clear that just as more and potentially increasingly vulnerable youth may need mentoring programs and more mentoring programs are being developed to serve these youth, efforts to enhance the effectiveness of such programs and thereby increase their “return on investment” are equally needed. In line with these concerns, this edition of the Handbook of Youth Mentoring has pointed the light of empirical investigation onto new program models, practices, and youth served. We hope in doing so that we can help usher in more widespread use of innovative and empirically examined practices. We also hope that we can think beyond programs as we have traditionally done so. The “mentoring program” is a relatively recent phenomenon—Big Brothers Big Sisters of America only recently celebrated its centennial. Before that were millennia of natural mentors in the lives of youth, and we are excited to see new ways of facilitating a rapprochement between formal programs and natural mentors. Youth-initiated matches, teachers and coaches as mentors, and other strategic efforts to strengthen and support the natural mentors in youths’ lives all reveal promise. But we know little about the best ways in which to provide these programmatic supports to natural mentors, nor do we fully know the true impact of natural mentors (at least once the eye of empirical scrutiny moves beyond studies of those mentors who youth found exceptional).

There is much to learn about how to support natural mentors, just as there is still much to learn about how to support volunteer mentors in traditional mentoring programs in schools and communities. We need to continue to push on both fronts, advancing research on these practices while we increase the availability of both program-sponsored and other mentors to youth. We hope that, as in the past, researchers will learn as much from practitioners as practitioners can learn from researchers, and that by the next edition of the Handbook, this symbiotic relationship—which is relatively uncommon in the social sciences and nonprofit world alike—will yield important advances that ultimately afford more mentoring, of a higher quality, to a larger number of youth.

References Albee, G. (1959). Mental health manpower trends. Joint Commission of Mental Illness and Health. Monograph Series No. 3. New York: Basic Books. America’s Promise Alliance. (2006). Every child every promise: Turning failure into action. Washington, DC: Author. Retrieved from http://www.ameri caspromise.org/Resources/Partner-Resources/e/ Every-Child-Every-Promise.aspx Anderson, A. R., Christenson, S. L., Sinclair, M. F., & Lehr, C. A. (2004). Check & Connect: The importance of relationships for promoting engagement with school. Journal of School Psychology, 42, 95–113. Baker, D. B., & Maguire, C. P. (2005). Mentoring in historical perspective. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring. (pp. 14–29). Thousand Oaks, CA: Sage. Beam, M. R., Gil-Rivas, V., Greenberger, E., & Chen, C. (2002). Adolescent problem behavior and depressed mood: Risk and protection within and across social contexts. Journal of Youth and Adolescence, 31, 343–357. Bloom, D., Gardenhire-Crooks, A., & Mandsager, C. (2009). Re-engaging high school dropouts: Early results of the National Guard youth ChalleNGe program evaluation. MDRC. Bogat, G. A., Liang, B., & Rigol-Dahn, R. M. (2008). Stages of mentoring: An analysis of an intervention for pregnant and parenting adolescents. Child & Adolescent Social Work Journal, 25(4), 325–341. de Tormes Eby, L. T., Allen, T. D., Hoffman, B. J., Baranik, L. E., Sauer, J. B., Baldwin, S., . . . Evans, S. C. (2012). An interdisciplinary meta-analysis of the potential antecedents, correlates, and consequences of protégé perceptions of mentoring. Psychological Bulletin, 139(2), 441-476. Dieterich, S. E., Landry, S. H., Smith, K. E., Swank, P. R., & Hebert, H. M. (2006). Impact of community mentors on maternal behaviors and child outcomes. Journal of Early Intervention, 28(2), 111–124. DuBois, D. L., Holloway, B. E., Valentine, J. C., & Cooper, H. (2002). Effectiveness of mentoring programs: A

Youth Mentoring: Progress and Prospects for the 21st Century    533 meta-analytical review. American Journal of Community Psychology, 30, 157–197. DuBois, D. L., & Karcher, M. J. (2005). Youth mentoring: Theory, research, and practice. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 2–11). Thousand Oaks, CA: Sage. DuBois, D. L., Portillo, N., Rhodes, J. E., Silverthorn, N., & Valentine, J. C. (2011). How effective are mentoring programs for youth? A systematic assessment of the evidence. Psychological Science in the Public Interest, 12, 57–91. DuBois, D. L., & Silverthorn, N. (2005). Natural mentoring relationships and adolescent health: Evidence from a national study. American Journal of Public Health, 95, 518–524. Erickson, L. D., McDonald, S., & Elder, G. H., Jr. (2009). Informal mentors and education: Complementary or compensatory resources? Sociology of Education, 82, 344–367. Goldner, L., & Mayseless, O. (2008). The quality of mentoring relationships and mentoring success. Journal of Youth & Adolescence, 38(10), 1339–1350. Goodman, G. (1972). Companionship therapy: Studies in structured intimacy. San Francisco: Jossey-Bass. Goodman, G., & Dooley, D. (1976). A framework for help-intended communication. Psychotherapy: Theory, Research & Practice, 13(2), 106–117. Grossman, J. B. (2005). Evaluating mentoring programs. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 251–265). Thousand Oaks, CA: Sage. Grossman, J. B., & Rhodes, J. E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199–219. Hamburg, B. A., & Varenhorst, B. (1972). Peer counseling in the secondary schools: A community mental health project for youth. American Journal of Orthopsychiatry, 42(4), 566–581. Hamilton, M. A., & Hamilton, S. F. (2005). Work and service. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 348–363) Thousand Oaks, CA: Sage. Halpern, R. (2005). Instrumental relationships: A potential relational model for inner-city youth programs. Journal of Community Psychology. Special Issue: Youth-Adult Relationships in Community Programs: Diverse Perspectives on Good Practices, 33(1), 11–20. Hirsch, B. J., Deutsch, N. L., & DuBois, D. L. (2011). After-school centers and youth development: Case studies of success and failure. New York: Cambridge University Press. Holt, L. J., Bry, B. H., & Johnson, V. L. (2008). Enhancing school engagement in at-risk, urban minority adolescents through a school-based, adult mentoring intervention. Child & Family Behavior Therapy, 30(4), 297–318. Hurd, N., & Zimmerman, M. (2011). Natural mentors, mental health, and risk behaviors: A longitudinal

analysis of African American adolescents transitioning into adulthood. American Journal of Community Psychology, 46. Karcher, M. J. (2008). The Study of Mentoring in the Learning Environment (SMILE): A randomized evaluation of the effectiveness of school-based mentoring. Prevention Science, 9(2), 99–113. Karcher, M. J., Herrera, C., & Hansen, K. (2010). “I dunno, what do you wanna do?”: Testing a framework to guide mentor training and activity selection. New Directions for Youth Development, 126, 51–69. Karcher, M. J., & Nakkula, M. J. (2010). Youth mentoring with a balanced focus, shared purpose, and collaborative interactions. New Directions for Youth Development, 126, 13–32. Keller, T. E., & Pryce, J. M. (2012). Different roles and different results: How activity orientations correspond to relationship quality and student outcomes in school-based mentoring. Journal of Primary Prevention, 33(1), 47–64. Kogan, S. M., Brody, G. H., & Chen, Y. (2011). Natural mentoring processes deter externalizing problems among rural African American emerging adults: A prospective analysis. American Journal of Community Psychology, 48(3–4), 272–283. Larson, R. W., Hansen, D. M., & Moneta, G. (2006). Differing profiles of developmental experiences across types of organized youth activities. Develop­ mental Psychology, 42(5), 849–863. Larson, R., Walker, K., & Pearce, N. (2005). A comparison of youth-driven and adult-driven youth programs: Balancing inputs from youth and adults. Journal of Community Psychology. Special Issue: Youth-Adult Relationships in Community Programs: Diverse Perspectives on Good Practices, 33(1), 57–74. McDonald, K. E., Balcazar, F. E., & Keys, C. B. (2005). Youth with disabilities. In D. L. DuBois & M. J. Karcher. (Eds.), Handbook of youth mentoring (pp. 493–507) Thousand Oaks, CA: Sage. MENTOR. (2006). Mentoring in America 2005: A snapshot of the current state of mentoring. Alexandria, VA: Author. Monitoring the Future. (2012). Monitoring the Future: A continuing study of American youth. Retrieved from http://www.monitoringthefuture.org/ Perlmutter, F., & Durham, D. (1965). Using teen-agers to supplement casework service. Social Work, 10(2), 41–46. Reinherz, H. (1964). The therapeutic use of student volunteers. Children, 11, 137–142. Rhodes, J. E., & DuBois, D. L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20, 3–19. Retrieved from http://www .srcd.org/ Sánchez, B., Reyes, O., & Singh, J. (2006). A qualitative examination of the relationships that serve a mentoring function for Mexican American older adolescents. Cultural Diversity and Ethnic Minority Psychology, 12(4), 615–631.

534  Conclusions Scales, P. C., Benson, P. L., Bartig, K., Streit, K., Moore, K. A., Lippman, L., . . . Theokas, C. (2006a). Keeping America’s promises to children and youth: A Search Institute-Child Trends report on the results of the America’s Promise National Telephone Polls of Children, Teenagers, and Parents. Retrieved from http://www.search-institute.org/system/files/2006 NationalPromisesStudy.pdf Scales, P. C., Benson, P. L., Bartig, K., Streit, K., Moore, K. A., Lippman, L., . . . Theokas, C. (2006b). Keeping America’s promises to diverse children and youth: A Search Institute-Child Trends supplemental report on the results of the America’s Promise National Telephone Polls of Children, Teenagers, and Parents. Retrieved from http://www.search-institute .org/system/files/2006NationalPromisesStudy.pdf Schwartz, S. (in preparation). Youth initiated mentoring: Investigating a new approach to working with vulnerable adolescents. Unpublished dissertation, University of Massachusetts, Boston. Smith-Jentsch, K. A., Scielzo, S. A., Yarbrough, C. S., & Rosopa, P. J. (2008). A comparison of face-to-face and electronic peer-mentoring: Interactions with

mentor gender. Journal of Vocational Behavior, 72(2), 193–206. Spencer, R. (2007a). “It’s not what I expected”: A qualitative study of youth mentoring relationship failures. Journal of Adolescent Research, 22(4), 331–354. Spencer, R. (2007b). Naturally occurring mentoring relationships involving youth. In T. D. Allen & L. T. Eby (Eds.), The Blackwell handbook of mentoring: A multiple perspectives approach (pp. 2099–2117). Malden, MA: Blackwell Publishing. Substance Use and Mental Health Data Archive. (n.d.). Browse and download data [Monitoring the Future Series data sets]. Retrieved from http://www.icpsr .umich.edu/icpsrweb/SAMHDA/browse Wood, S., & Mayo-Wilson, E. (2012). School-based mentoring for adolescents: A systematic review and meta-analysis. Research on Social Work Practice, 22, 257–269. Zimmerman, M. A., Bingenheimer, J. B., & Behrendt, D. E. (2005). Natural mentoring relationships. In D. L. DuBois & M. J. Karcher (Eds.), Handbook of youth mentoring (pp. 143–157). Thousand Oaks, CA: Sage.

AUTHOR INDEX

Abel, G., 412-413, 415-416 Aber, J. L., 176 Adalist-Estrin, A., 374, 447, 465-466 Addis, M. E., 160 Adler, H., 122 Ahrens, Kym R., 8, 77, 346-347, 352 Ainsworth, M. D. S., 356, 374 Al-Issa, I., 191-192 Albee, G., 530 Albertini, V. L., 147 Albin, G., 210, 307 Alexander, F., 428 Alexopoulos, G., 327 Alfonso, L., 108 Allen, M., 342 Allen, T., 9, 160, 484, 529 Almeida, D. M., 160 Alpert, L. T., 342 Alster, P., 120 Altman, D. G., 495 Amaleyhu, A., 362-363 America’s Promise Alliance, 525-526 Amichai-Hamburrger, Y., 265 Amico, J., 117, 120 Anderman, L. H., 204 Anderson, A. R., 139, 332, 530 Anderson, C., 147 Anderson, D. D., 519-520 Anderson, T., 473 Andrade, N., 372 Andrews, D. W., 321, 446 Ang, R., 276 Angelov, B., 264 Anton, B. S., 161 AOL, 203 Aos, S., 499, 502, 505 Arbreton, A. J., 46, 203, 221, 223-225, 227-230, 504, 506 Arellano, A. R., 361 Arnold, A., 120 Arevalo, E., 411-414, 416-417, 419-423

Asai, M., 191-192 Ascher, C., 161 Asgari, M., 262, 266 Astone, N. M., 177 Atkinson, D. R., 147 Audage, N. C., 413, 416 August, G. J., 333, 461, 464, 466 Avani, N., 246, 248, 251, 253 Avitable, N., 121 Ayers, M., 160 Ayers, P. M., 87, 207-208, 210 Aylwin, A. S., 440 Azmitia, M., 360 Azuma, H., 189, 192 Babcock, L., 499 Baker, D., 33, 529 Balcazar, Fabricio E., 50, 85, 90, 92-93, 96, 349, 519, 527 Baldwin, A., 371 Baldwin, C., 371 Ballasy, L., 19, 405-406 Baltes, P. B., 19 Baltodano, H. M., 317, 320-321 Bandura, A., 162, 295 Barak, A., 262 Baral, N., 499 Baranik, L. E., 529 Barber, B. L., 205 Bargh, J. A., 262 Barnes, S. B., 261, 265, 267 Barnett, J., 210, 307 Barnett, W. S., 499, 502 Barnoski, R., 318 Barns, S., 499 Barr, S. C., 460 Barrera, Jr., M., 123, 274-275, 459 Barrett, L. F., 162 Barrett, P., 325 Barron-McKeagney, T., 460, 463, 466 Bartels, A., 119 Bartholomew, K., 161 535

536   HANDBOOK OF YOUTH MENTORING Bartig, K., 526 Bartko, W. T., 371 Bass, A., 9, 326, 486 Basualdo-Delmonico, Antoinette, 149-150, 181, 185, 217, 412, 433, 448, 460, 475-476, 519 Batson, C. D., 398 Battjes, R., 275 Baumeister, R. F., 160, 469 Baumgartner, T., 123 Bay-Cheng, L. Y., 168, 520 Beam, M. R., 134, 207, 328, 446, 529 Beardslee, W., 104 Beauchamp, T. L., 513, 517 Becker, J., 125 Beech, A., 413 Beehler, S., 358 Behrendt, D. E., 180, 207, 223, 315, 371, 459, 527 Belfield, C. R., 499, 502, 505 Bellas, A. S., 501-502 Belshaw, S. H., 316, 322 Benard, B., 48, 101 Benford, P., 260-261 Benne, K., 102, 111-112, 149, 404, 432 Benner, G. J., 333-334 Bennett, D., 262, 264 Benson, P. L., 17-19, 21, 68, 234, 526 Bentley, A. C., 175-177 Berardi, Luciano, 134 Berger, M., 499, 503 Bergman, L. R., 170 Bergseth, K. J., 316-317, 320-321 Berlin, L. J., 132 Bernard, Claude, 48, 101, 119 Bernard, H. S., 471 Bernat, D. H., 333 Berndt, T. J., 160 Bernier, A., 304, 306, 308, 428-429, 432 Bernstein, L., 6, 34, 164, 178, 208-209, 469, 473, 486, 489, 515 Bernstein, L., 369 Bessell, A. G., 238 Bhattarai, R., 499 Bierema, L. L., 259, 264 Biglan, A., 493, 495 Bilchik, S., 374 Billig, S. H., 295 Bilotti, A., 499 Bingenheimer, J. B., 180, 207, 223, 315, 371, 459, 527 Birch, S., 502 Birman, Dina, ,358-359, 447 Bjorvell, H., 120 Black, D. R., 246, 253 Black, D. S., 207 Blake, J., 264, 270

Blakeslee, Jennifer E., 8, 295, 346, 459 Blanchard, A., 261, 265, 268 Bland, R., 245 Blankenship, V., 307 Blash, R., 151 Blechman, E. A., 49, 132, 315-316, 322, 349, 505, 519 Blinn-Pike, L., 132, 303, 307-308, 349, 519 Blocker, K. M., 48-49 Blomquist, G., 499, 503 Bloom, D., 428-429, 434 Bloomquist, M. L., 333, 461, 464 Blum, D. J., 90 Blum, R. W., 22-23 Blumberg, F. C., 161 Blumenfeld, P. C., 295 Blyth, D. A., 374 Boat, T., 38 Bobek, D., 20 Bockneck, E. B. L., 344, 372 Bodin, M., 189 Boerger, R., 166, 347 Boezeman, E. J., 401, 406 Bogat, G. Anne, 159, 165, 170, 216, 356, 447, 531 Bogere, S., 193-194 Boggs, S. R., 325 Bonds, D. D., 123, 459 Boniel-Nissim, M., 262 Bontempo, R., 191-192 Booth, A., 120-121 Bopp, J., 315, 316, 322 Borrmann, B., 78, 195 Borden, L. M., 19, 228 Borduin, C. M., 322 Bornstein, H. H., 19 Boruch, R. F., 493, 495 Bosker, R. J., 205 Botell, R. E., 85 Botvin, G., 35 Bouffard, J. A., 316-317, 320-321 Bouchard, C., 189 Bovend’Eerdt, T. J., 85 Bowen, G. L., 204 Bowen, M., 459 Bowen, N. K., 347 Bowers, E., 19-20 Bowlby, John, 100-102, 106, 304, 343, 371, 459 Bowles, J., 3 Boyce, W., 122 Boyle, P., 34-35, 413, 491, 496 Bozick, R., 400 Bradshaw, M., 225 Brady, B., 189 Bragar, M. C., 470 Braithwaite, V. A., 165

Author Index  537 Brandtstädter, J., 20 Bransford, J. D., 296 Bratslavsky, E., 469 Bray, I., 386-387, 389-391, 393 Brazzell, D., 373 Brennan, K. A., 399, 428 Bressler, J., 470 Bridges, J. F. P., 506 Briggs, E. C., 345, 371 Britner, Preston A., 137, 342, 344, 348-350, 351-352, 372, 459, 473, 519 Brittain, A., 23 Broadus, J., 428-429, 434 Brodeur, P., 46, 240, 306 Brody, G. H., 528, 530 Broemer, P., 400, 405 Brogan, D., 48, 58, 77, 109, 165, 446, 514 Bronfenbrenner, U., 19, 21, 65, 132, 189, 292-295, 358 Bronte-Tinkew, J., 484 Brooks, C., 176-177 Brooks-Gunn, J., 18-20, 22 Broomfield-Massey, K., 74 Brotman, L., 121 Brown, A. L., 296 Brown, R., 317, 325, 440 Brown, S., 400 Bruce, C., 264 Bruce, M., 327 Brumovská, T., 189, 193-194 Bruns, E. J., 333 Brunstein, J. C., 84 Bruster, B. E., 372 Bry, B. H., 136, 207, 238, 308, 310, 312, 530 Bryant, A. L., 162, 165 Bryson, R., 402, 429 Buchanan, C., 121, 125 Buchanan, R., 358-359 Buche, M. W., 264 Buecker, B., 505 Buhrmester, D., 56, 160, 441 Buist, K. L., 161-162 Bulle, M. J., 223-224, 229 Bullen, P., 152, 155 Bullis, M., 317, 319 Burchard, J. D., 333 Burger, J. M., 147 Burgstahler, S., 261, 264 Burleson, B. R., 428 Burleson, J., 276 Burlew, L. D., 470 Burns, B. J., 333, 336 Burraston, B., 122 Burt, K. B., 32 Burt, R. S., 131, 133

Bush, K., 192 Buss, C., 119, 123 Byrne, D., 146, 149, 162, 428 Cagampang, H., 295 Callahan, C. M., 461 Callaman, J. M., 152 Cameron, M., 499 Campbell, A., 160-161 Campbell, D. E., 149, 169, 377, 432 Campbell, M., 225 Campbell, T. A., 149, 169, 432 Canon, Walter, 119 Carder, P., 130 Carlo, G., 398 Carpenter, A., 196 Carswell, S. B., 32, 152, 279 Carter, C., 119 Carter, E. W., 243, 245-246, 251 Cartwright, W., 500 Caruthers, F. A., 484 Cassidy, J., 132, 336, 428 Catalano, R. F., 18, 22 Cataldi, E. F., 303 Cavell, Timothy A., 32-33, 35, 37-38, 48, 51, 56, 78, 108, 136, 149, 159, 208, 212, 224, 234-235, 240, 242, 254, 276, 304, 307, 330, 332, 334, 457, 461-462, 466, 469, 473-474, 486 Ceballo, R., 361 Ceci, S. J., 189 Cervantes, R., 48, 51-52, 136, 166 Chaffin, M., 412, 414, 417 Chaloux, N., 309 Chamberlain, P., 333-334 Chan, Wing Yi, 8, 194, 208, 212, 349, 447, 473, 493, 515 Chapman, C., 303 Chassin, L. A., 30-31, 38 Chavous, T. M., 147 Chemers, M. M., 182, 432 Chen, C., 134, 192, 194, 207, 328, 446, 527, 529 Chen, E., 121 Chen, L., 163 Chen, Y., 528, 530 Chermok, V. L., 398 Chew, A., 239 Childress, J. F., 513, 517 Ching, A. M., 244, 342, 361-363 Chiodo, D., 245 Chiu, G., 222-223, 225-228 Choate, R. O., 86 Christenson, S. L., 139, 332, 530 Christenson, T., 249 Chu, J. Y., 160

538   HANDBOOK OF YOUTH MENTORING Cicchetti, D., 29-30, 32 Cintrón, R., 149 Claes, M., 189 Claire, T., 177 Clark, H. B., 242 Clark, M., 160, 163 Clarke, G. N., 327-328 Clarke, L., 351 Clarke-Stewart, K. A., 371-372 Clary, E. Gil, 398-400, 402-404, 445 Coatsworth, J. D., 315-316 Cobbs, G., 160 Cocking, R. R., 296 Coffino, B., 371-372, 473, 476 Cohen, A. B., 175-176 Cohen, D. J., 29-30, 32, 307 Cohen, G. L., 516 Coie, J. D., 29-33, 35, 38 Coleman, M., 75 Coleman, J. S., 131-132 Colley, H., 189, 193 Collier, M. L., 268-269 Collins, E., 152, 155 Collins, M. E., 342, 346 Colón-Torres, Yarí, 145-146, 149, 179-180, 183, 309, 360, 459 Comley-Ross, P., 446 Conger, R. D., 327 Connell, J. P., 204 Constantin, B. U., 499 Contreras, J. M., 161, 457, 472 Converse, N., 308, 311, 430 Cooney, S. B., 6, 178, 212, 234, 403, 487 Cooper, Becky, 9, 11, 50, 64, 77, 135, 166, 175, 190, 209, 224, 235-236, 259, 291, 306, 328, 330, 341, 360-361, 363, 374, 411-414, 416-417, 419-423, 431-432, 439, 457, 481, 491, 495, 513-514, 529 Cooper, Z., 328 Corlett, J., 74-75 Corral, M., 48, 77, 109, 446, 514 Costello, E. J., 325 Coulter, E. K., 68-69, 529 Courtney, M. E., 342, 348-349 Cowling, M. J., 399 Cox, P. D., 274 Cox, R., 245-246 Coyne, J. C., 160 Craft, C., 245 Craig, J., 434-435 Cravens, J., 268 Crawford, L., 264 Creason, A. H., 295, 430 Crepaz, N., 495 Crick, N., 104

Crisp, G., 305, 311 Croizet, J. C., 177 Cronheim, D., 261 Crooks, C. V., 245 Cropley, M., 86 Crosnoe, R., 447 Cross, S., 160 Cross, W., 37 Crouter, A. C., 132 Cruz, I., 305, 311 Crystal, D. S., 189, 192 Csiksizentmihalyi, M., 441 Culhane, S. E., 32, 149, 347, 349-350 Culpepper, D. W., 266-268 Cummings, E., 19, 21 Cummings, L., 273, 278, 283-284 Cunningham, P. B., 322 Curry, P., 244 Cushing, L. S., 246 Cutrona, C. A., 264 Cyrene, D., 46, 240, 306 D’Ambrosio, R., 317, 319 D’Souza, H. J., 460, 463, 466 da Costa Werlang, S. R., 506 Daggs, J., 412-414, 416 Dallos, R., 446 Daly, M., 399, 445 Damon, W., 19 Dane, A. V., 485 Danish, S., 88 Darling, N., 159, 165-166, 192, 194, 207, 262, 268, 292, 445 David-Ferdon, C., 325 Davidson, A., 8, 19, 212, 240, 243, 251, 308, 325, 402 Davidson, J., 325 Davidson, W. S., 443 Davies, E., 373 Davies, G. L., 90 Davies, H., 10 Davis, A., 206 Davis, B., 88, 331 Davis, C., 17, 109, 135, 242, 309, 462 Davis, K., 121, 150 Davis, M. H., 398 Davis, V. T., 77 Dawson, N., 189 Day, A., 348 Day, S. X., 259, 268 de Boer, H., 205 de Guzman, M. R. T., 398 De Kloet, E., 120 de Tormes Eby, L. T., 529 De Wit, D., 193

Author Index  539 DeBar, L. L., 327-328 DeBell, M., 189, 192 Deci, E. L., 190, 196, 305, 401 Degenne, A., 133 Dekel, R., 472 Deklyen, M., 371 Dekovic, M., 161-162 Del Guidice, M., 121 del Prado, A., 147 DeLongis, A., 123 Delpit, L., 207 Demetriou, A. P., 161 Denham, S., 446 Dennis, M., 276 Dennison, S., 89 Denny, G., 210, 307 Denton, W. H., 428 Derzon, J. H., 369 Des Jarlais, D. C., 495 Desai, S., 222-223, 225-228 Deschenes, S. N., 221, 223, 225-226, 228-229 Deutsch, Nancy L., 5, 84, 177, 181-182, 221-222, 225-227, 229-230, 277, 283-284, 427, 432, 513, 516, 529 DeWoody, M., 342 Diamond, J., 3 Dicken, C., 402, 429 Dickerson, S., 120-121 Diehl, D. C., 345 Dies, K., 274, 286 Dieterich, S. E., 531 DiMeo, M., 74 DiRenzo, M. S., 262, 265, 267-269 Dishion, T. J., 34, 236, 273, 276, 286, 321, 337, 372, 446 Ditton, T., 264 Dively, D. D., 500 Diversi, M., 361-363, 519 Dodge, K. A., 30, 32-33, 104, 236-267, 286, 337 Domingo, M. R., 369 Donaldson, C., 502 Donaldson, M. E., 161 Dong, Q., 192 Dong, Y. R., 266 Dooley, D., 64-65, 67, 531 Doolittle, F., 30, 137, 484 Dow, J., 506 Downey, G., 132, 512 Dozier, M., 121, 343, 350, 428-429, 432 Drake, E. K., 318 Drexler, S., 78, 195 Drob, S., 471 Drummond, M., 500

DuBois, David L., 3, 5-6, 8-9, 17, 21, 30, 32-39, 47-48, 50, 64, 73-78, 83, 90, 96, 118, 122-123, 126, 135, 137, 147, 149-151, 154-155, 163-166, 168, 175, 178, 180-183, 190, 203, 205, 207, 209-210, 221, 223-224, 226-227, 235-236, 244-245, 247, 251-252, 254, 259, 261, 264, 273-274, 277-278, 291-292, 294, 303, 306-309, 312, 319, 321-322, 326, 329-330, 332, 334, 337, 341, 343, 346, 348-350, 356, 361, 371, 374, 379, 403-404, 411, 416, 427, 429-433, 436, 439, 442-445, 449-450, 457, 460, 462-463, 472, 474, 481, 484-485, 488-492, 494-496, 511, 513-515, 517, 526-529, 531 Duchnowski, A., 333-334 Dun Rappaport, C., 469, 473, 486, 515 Dunbar, N., 360-361 Duncan, B., 437 Duncan, G. J., 176-178 Dunphy, A., 152, 155 Dupree, D., 177 Durham, D., 236, 530 Durlak, J. A., 35, 161, 210, 440, 518 Dworkin, J., 223, 441 Dworsky, A., 342, 348 Early, M. R., 160 Eberly, S., 37 Ebert, L., 136 Eby, L. T., 9, 294, 303, 307, 329, 471, 529 Eccles, J., 19, 125, 205, 223-224, 526 Eddy, C. L., 244 Eddy, J. Mark, 333-334, 341, 369, 371, 373, 447 Edelman, P., 356 Ederyn, W., 264 Egan, E. A., 333, 461, 464 Egan, S. K., 163 Eggert, L., 283 Ehlert, U., 123 Einolf, C. J., 398 Eisenberg, N., 398 Eklund, J. H., 398 Elder, Jr., G. H., 8, 19, 21-22, 136, 150, 180, 207, 291, 306, 371, 447, 527 Elledge, L. Christian, 32, 35, 48, 224, 304, 330, 334, 457, 469, 486 Ellemers, N., 401, 406 Ellevold, B., 412-414, 416 Elliot, D. S., 31 Elliot, S. N., 31, 38, 87, 91 Ellis, B., 121-122 Ellis, H., 360 Elson, M., 206 Emmerson, G. J., 86 Engeman, R. M., 499

540   HANDBOOK OF YOUTH MENTORING English, A., 342 Ensher, E. A., 149, 261, 265, 268, 405, 443 Entwisle, D. R., 177 Epstein, M. H., 333-334 Erikson, Erik, 8, 99, 100-103, 105-106, 124, 136, 150, 180, 206-207, 291, 306, 309, 527 Erikson, J. M., 101-102, 111-112 Erickson, L., 124 Erkanli, A., 325 Erney, T., 247 Eröz-Tuğa, B., 261 Esparza, P., 134, 149, 309, 360 Espiritu, R., 48, 51-52, 136, 166 Essex, M., 121 Etgen, A., 120 Evans, G., 123 Evans, I. M., 189 Evans, S., 9, 303, 329, 394 Evans, W. P., 317, 440 Eyberg, S. M., 325, 332 Fabiano, G., 325 Fabricius, J., 471 Fahrbach, S., 120 Fairburn, C. G., 328 Faith, M. A., 35, 37, 48, 334, 457, 469 Fallot, R. D., 432 Fan, M. Y., 346 Farrington, D. P., 9, 316, 319, 369 Farruggia, S., 152, 155, 192, 342 Faust, K., 137 Faw, L., 333 Fay, K., 23 Fehr, E., 125-126 Feldhusen, J. F., 445-446 Feldman, A. F., 6, 34, 216, 234, 489, 506, 512 Felner, R. D., 276 Fennig, S., 472 Ferber, T., 19 Ferguson, R., 266 Fernandes-Alcantara, A. L., 10 Ferrell, L., 325 Ferrer-Wreder, L., 190 Fett, R. O., 372 Fettes, D., 335 Feuer, Rachel, 150-151 Field, J., 161 Figg, C., 260 Finckler, J., 244 Fink, A., 356 Finkelhor, D., 412, 414, 417 Finkelstein, M., 399 Finkelstein, N., 295 Finkenauer, C., 469 Fischbacher, U., 125-126 Fischer, K., 136

Fisher, C. B., 448, 513 Fisher, J., 505 Fisher, P., 122, 328 Fitzgerald, L. F., 149 Flanagan, C., 19 Flannery, D. J., 162 Flaxman, E., 161 Flay, B. R., 29, 31, 38, 493, 495 Fletcher, K. R., 235 Flinn, M., 119 Flores, L. Y., 166, 433 Fo, W. S., 65, 234 Folkman, S., 123 Fong, C., 244 Forbush, J., 406 Foreman, K., 372 Forman, Y. E., 19 Forneris, T., 88 Forret, M. L., 109 Forsyth, J., 196 Foster, E. Michael, 485, 499-508 Fountain, D. L., 504, 506 Fox, M., 457 Fraley, R. C., 399 Frank, J. B., 66-67 Frank, J. D., 66-67 Fredriksen, K., 206 Freedman, M., 132, 175, 189, 291, 374 Freitas, A. L., 132, 512 French, T. M., 428 Fresko, B., 190, 193 Frey, A., 430 Fried, C., 351 Fries, A., 121-122 Fritzberg, A., 362-363 Fuhrel-Forbis, A., 400 Fuligni, A. J., 356 Fuller, A., 295 Fulop, M., 386-387 Furano, K., 416-417 Furstenberg, Jr., F. F., 133 Gagne, B., 75 Galama, K., 86-87 Galambos, N. L., 160 Galasso, L., 356 Gale, L., 273 Gamache, P., 242-243 Garate-Serafini, T. J., 85, 88, 92 Garceau, O., 46, 240, 306 Garcia, A. J., 282, 284 Garcia, M. C., 400 Garcia, N., 123 Garcia-Coll, C., 176-177 Gard, G., 152 Gardenhire-Crooks, A., 431, 531

Author Index  541 Garland, A. F., 160, 342 Garmezy, N., 34 Garraway, H., 165 Garringer, M., 397, 405, 423, 475 Garwick, G., 86 Gary, T., 419 Gastic, B., 8, 166, 207 Gatha, R., 58, 165 Geary, D., 119 Gebauer, J. E., 400, 405 Geenen, S., 341 Gelso, C. J., 471-472 Gennetian, L. A., 495 Georgiou, S. N., 161 Gershoff, E. T., 176 Gibson, J., 499 Gidron, D., 189 Gigerenzer, G., 414 Gil-Rivas, V., 207, 328, 446, 529 Gillath, O., 399 Gilligan, C., 160 Gilligan, R., 132 Girdler, S., 120 Gjerde, P. F., 129, 132 Glaze, L. E., 369-370 Godshalk, V. M., 320 Gold, M. R., 500, 504 Goldberg, L. R., 398 Goldner, Limor, 132, 136, 146, 189-190, 192-193 Gomez, L. M., 265, 268-269 Goode, M. R., 401 Goode, W. W., 370, 374 Goodenow, C., 204 Goodlad, S., 235 Goodman, Gerald, 64, 65, 67, 76, 234, 530-531 Gootman, J. A., 19, 223, 526 Gopez, A., 284 Gordon, K., 121 Gordon, L. J., 334 Gordon, M. M., 176 Gordon, M. S., 275, 358 Gore, A., 19, 21 Gorman, B., 275 Gormley, B., 162, 304 Gorski, P. C., 183 Gorter, J. W., 86-87 Gosland, J., 434-435 Gottlieb, G., 19, 124, 118 Gould, M. S., 328 Gould, R. P., 472 Gouley, K., 121 Government of Alberta, 246-248, 252 Goza, B., 182, 432 Graff, L., 416

Graham, J., 161 Granger, D., 120-121, 332 Grannucci, E. A., 242-243 Granovetter, M., 131-134 Grant-Thompson, S. K., 147 Graydon, K., 341-342, 346 Green, M. A., 259, 268 Green, V., 471 Greenberg, G., 19 Greenberg, M. T., 38, 327, 371 Greenberger, E., 134, 207, 328, 446, 529 Greene, M. L., 147, 150 Greeson, J. K. P., 346-347 Greenwood, D., 162 Grenard, J. L., 207 Grewen, K., 117, 120 Griffin, K. W., 35 Griffiths, M., 153, 155, 361 Griller-Clark, H., 321 Griner, D., 148 Grinstein-Weiss, M., 346 Grossman, J. B., 6, 8, 31, 34, 36-37, 46, 49, 51, 64, 71, 89-90, 108, 135, 178, 192, 209, 212, 221, 223-230, 234, 295, 306, 308, 317, 329-330, 334, 337, 342, 348, 370, 398, 400, 403-404, 416-417, 430, 433, 439-440, 443-444, 448, 457, 462, 469, 473, 481, 486-487, 489-491, 495, 505-506, 512-513, 515, 519, 528-529 Groth, A. N., 413, 419 Grusky, D. B., 177 Gunnar, M., 121-123 Gur, M., 166 Haddad, E., 149 Hafton, N., 342 Hagerty, B., 160 Haggerty, R. J., 29-30 Haight, W. L., 345, 371 Halas, J., 196 Hall, G. C. N., 146 Hall, J. C., 189 Halpern, R., 294-295, 530 Hamburg, B. A., 234, 530 Hamilton, Mary Agnes, 68, 71, 89, 110, 132-133, 165-166, 192, 277, 291-293, 195-296, 374, 445-446, 526, 530 Hamilton, Stephen F., 68, 71, 89, 110, 132-133, 165-166, 192, 207, 277, 291-293, 295-296, 374, 445-446, 526, 530 Hammer, T. J., 160 Hamre, B. K., 204 Han, S. S., 332 Handwerk, M., 276 Hanks, C., 500 Hanlon, T. E., 32, 152, 279

542   HANDBOOK OF YOUTH MENTORING Hanneman, A., 371-372, 473, 476 Hansen, Keoki, 30, 50, 73-76, 85, 102, 109, 205, 235, 240, 249, 252, 261-262, 306, 405, 441, 448, 529-531 Hare-Mustin, R., 163 Harel, G., 472 Harlow, N., 412-413, 416 Harrington, C., 161 Harris, E., 458 Harris, John T., 46-47, 50, 56, 75, 84, 89, 91, 224, 260-261, 256, 306, 430, 442, 444, 458 Harris, V. W., 224 Harter, S., 160 Hartje, J. A., 440 Hartmann, T., 177 Hartup, W., 274, 276 Hass, M., 341-342, 346 Hastings, P., 121 Hattie, J., 176 Hauser, R. M., 178 Hausmann, L. R. M., 398 Haveman, R. H., 501, 505 Hawley, K. M., 326 Hayashi, Y., 91 Hayes, E., 85 Headlam-Wells, J., 434-435 Hebert, H. M., 531 Heckhausen, J., 342 Hedges, L. V., 211 Heffer, R. W., 149 Heinrichs, M., 123 Hekter, J. M., 333, 461, 464 Helberg, C., 505 Helms, J. E., 145-147 Hen, L., 262 Henderson, A. T., 458 Henderson, C. R., 500 Henderson, J., 411-412, 417 Hendricks, B. C., 37 Hendry, L. B., 189, 194, 274 Henggeler, S. W., 316, 322, 333, 500 Henrie, J. L., 78, 108, 136, 212, 214, 239-240, 242 Henry, David B., 349, 486 Herman, K. C., 402, 412, 414-416 Hernandez, R., 189 Herrera, Carla, 6, 8, 34, 37, 46, 50, 68, 73, 85, 110, 121, 135, 161, 164, 166, 177-178, 203, 205, 208-213, 215, 217, 221, 234-235, 240-241, 249, 252-253, 273-274, 276, 286, 306-308, 311, 402-403, 405, 432, 439, 443, 448, 450, 469, 472-473, 487, 489, 491-495, 506, 512, 515, 517, 529 Hess, J. Z., 460 Hetherington, L., 261 Hettleman, D., 32, 149

Heun, C., 261, 265, 268 Hewatt-Grant, J., 147 Higginbotham, B. J., 224 Higgins, E. T., 190-191, 196 Higgins, M. C., 132, 134 Hilbert, S., 457 Hildago, R., 325 Hill, J. P., 160 Hines, A. M., 347 Hinshaw, S. P., 206, 325 Hirsch, Barton J., 5, 163, 166, 181-182, 221, 223, 225-228, 276-278, 347, 352, 444-445, 489, 529 Ho, J., 355, 359 Ho, W. C., 192, 194 Hoagwood, K., 513 Hobbs, N., 297 Hobson, W., 419 Hoffman, B. J., 529 Hoffman, M., 264, 270 Holahan, W., 472 Holcomb, M., 237, 404 Holdsworth, J., 434-435 Hollon, S. D., 328 Holmes, J., 471 Holt, L. J., 136, 207, 238, 308, 310, 312, 430, 530 Honey, M., 262, 264 Hopson, R. K., 484 Horton, R. S., 147 Hossain, M. A., 499 House, L. D., 278, 280, 283-284 Houston, A. C., 175-177 Hout, M., 176-177 Howard, A., 273, 282, 284 Howse, R. B., 345 Hoyle, R. H., 471 Hoyt, J. L., 398 Hu, C., 149, 432 Huang, K., 121 Hubbell, R., 152 Hudley, C., 147 Huefner, J., 276 Hughes, C., 245-246, 251 Hughes, J. N., 35-38, 48, 51, 56, 208, 212, 243, 276, 307-308, 332, 334, 457, 461-462, 469, 473 Hughes, M. E., 134 Hunt, D., 6, 34, 164, 178, 208, 460, 469, 473, 486, 515 Hunt, D. E. I., 402, 411-412, 414-417, 421 Hupbach, A., 119, 123 Hupert, N., 262, 264 Hur, M. H., 179-180 Hurd, N., 8, 149-150, 527 Hurn, J., 86

Author Index  543 Husky, M. M., 326 Husock, H., 370 Huwe, J. M., 471 Hymel, S., 33 Insel, T., 120, 123 Irby, M., 19 Jackson, J. F., 360 Jackson, M. A., 161 Jackson, P. B., 416 Jacoris, S., 122 Jambor, E. E., 399 James Bell Associates, 487 Jarrett, R. L., 134, 230 Jaycox, L. H., 356 Jeličić, H., 20 Jensen, P., 335, 513 Jensen-Doss, A., 326 Jent, J., 32, 135, 278-279, 331-332, 334 Jernigan, M., 145-146 Jessor, R., 68, 526 Jessor, S. L., 68, 526 Ji, P., 38, 483, 492-493 Jilnina, J., 105 Jivanjee, P., 334 John, O., 123 Johnson, D., 8, 122, 166, 207, 235 Johnson, J. L., 303-304 Johnson, K., 417-418 Johnson, S. B., 353 Johnson, V. L., 207, 238, 242, 246, 308, 430, 530 Johnson, W. B., 470-471, 477 Johnston, D., 369 Johnston, J., 318 Johnston-Brooks, C., 123 Joiner, Jr, T. E., 329 Jolliffe, D., 9, 316, 319 Jones, A. L., 90 Jones, C., 358 Jones, R., 348 Jonson-Reid, M., 308 Jordan, J. V., 159-160, 165, 205, 517 Jory, A., 189 Joyce, A., 440, 471-472 Joyner, K., 484 Jucovy, L. Z., 234, 311, 356, 370-371, 450, 473, 495, 512 Jung, T., 327 Kafai, Y. B., 222-223, 225-228 Kaiser-Urlery, C., 294, 430 Kakinuma, M., 189, 192 Kalbfleisch, P. J., 513 Kamboukos, K., 121

Kamel Boulos, M. N., 261 Kaminer, Y., 276 Kaplan, A. G., 159, 168 Kaplan, C. P., 151, 362-363, 320 Karcher, Michael J., 6, 9, 17, 20, 31-32, 34, 37, 46-50, 59, 68, 70, 73, 75-76, 78-79, 85, 89, 102, 108-111, 135-136, 139, 149, 163-164, 205, 207-210, 212-213, 216, 223, 234-235, 237, 240-243, 245-246, 248-249, 251-254, 261-262, 273-276, 291, 306, 308-309, 326, 352, 355-356, 359-360, 402-405, 420, 428-429, 430, 432, 434, 443, 444, 447-448, 451-452, 462, 473-475, 477, 493, 515, 526, 529-531 Kaslow, N., 325 Kass, N., 402, 429 Kassinove, H., 275 Kastelic, D., 160 Kataoka, S. H., 356 Kauh, T. J., 6, 34, 178, 212-213, 234, 403, 487, 489, 494, 506, 512 Kazdin, A. E., 30, 32-33, 326, 471 Keatinge, C., 328 Keith, J. G., 19 Keefe, K., 160 Keller, Thomas E., 4, 6, 9, 34-35, 64, 72-73, 75, 83-85, 87, 110, 129, 132-133, 135, 138, 194, 206, 210, 212, 264, 276, 295, 304-305, 311-312, 315, 344, 346-347, 349-350, 429, 436, 443-445, 459-460, 464, 469-470, 487, 489, 515, 531 Kelly, J. G., 322 Kelly, Michael S., 430 Keltner, D., 123 Kemeny, M., 120-121 Kemmelmeier, M., 399 Kempler, T. M., 295 Kendall, D. L., 269, 306, 344, 430-431 Kenkel, D., 499, 503 Kennedy, C. H., 246 Kenny, G. A., 161, 165 Kenny, M. E., 58, 414 Kercher, A., 327 Kerr, David C. R., 349, 459 Kessler, R. C., 30 Ketelaar, M., 86-87 KewalRamani, A., 303 Keys, Christopher B., 50, 85, 92, 527 Khan, A. S. A., 499 Kia-Keating, M., 360 Killian, E. S., 317, 440 Kim, C., 362-363 King, Cheryl A., 8, 20, 88, 189, 330-331, 334, 337, 349, 369, 457, 506 King, J., 369

544   HANDBOOK OF YOUTH MENTORING King, M., 506 Kingston, P. W., 176 Kinnison, K. E., 87 Kioko, M., 348 Kirchner, J., 147 Kiresuk, T. J., 85-87 Kirk, R., 348 Kirkpatrick, B., 119 Kirschbaum, C., 121, 123 Kitchener, K. S., 518, 520 Kitzman, H., 500 Kjellstrand, J. M., 369 Klagholz, D., 48, 51-52, 136, 166 Klaw, E. L., 136, 149 Klein, K., 387, 391 Klein, R. H., 471 Klem, A. M., 204 Klimes-Dougan, B., 121 Kloosterman, V. I., 238 Kneebone, I., 86 Knight, G. P., 398 Knight, P., 264 Kochanoff, A. T., 446 Koestner, R., 401 Kogan, S. M., 528, 530 Kohlberg, L., 100, 105 Kohut, H., 206, 237, 242 Konrath, S., 400 Konstantopoulos, S., 211 Kosfeld, M., 125-126 Kotter, J. P., 386 Kraemer, H. C., 30, 79 Kraimer-Rickaby, L., 342, 348, 350, 352, 473 Krajcik, J. S., 295 Kram, K. E., 132, 134, 470 Kremer, Sarah E., 513-514 Kreye, M., 121 Kritsonis, W. A., 316, 322 Krupa, M., 104 Kruzich, J. M., 334 Kuperminc, Gabriel P., 32, 70, 74, 78, 139, 224, 273, 278, 283-284, 355, 447, 517 Kupersmidt, Janis B., 91, 183, 216, 285-286, 452, 517-518, 531 Kupfer, D. J., 30 Kurain, J., 122 Kurtines, W. M., 359 Kutash, K., 333-334 La Vigne, N. G., 373 Laakso, J., 372 LaBlanc, L., 48, 51-52, 136, 166 Lacourse, E., 189 Laird, J., 303 Lakes, K., 352, 475 Lambert, J., 292

Lance, C. E., 149, 432 Landry, S. H., 531 Langhout, R. D., 71, 75, 85, 448 Lanning, K., 414 Lansford, J. E., 236, 286 Lareau, A., 175-177 Larose, Simon, 90, 132, 161, 205, 240, 305-311, 349, 432, 459, 519 Larson, R., 21, 223-224, 294-295, 441, 530 LaRue, C. M., 369, 372 Lav, I., 385-386 Lave, J., 296 Lawrence, Edith C., 183 Lazarus, R., 123 Lea, M., 264 Leary, M. R., 160, 471 Lebeaux, M., 133 Lebolt, A., 132, 512 Lederer, J., 150, 152 Lederhendler, I., 119 Lee, T. R., 224 Lee, Y., 308 Leh, A. S. C., 261, 265 Lehr, C. A., 139, 332, 530 Leifman, H., 189 Lenear, P. E., 267-269 Lennon, M. C., 176 Lens, W., 87 Lerner, Richard M., 17, 19-23, 69, 216 Letner, J., 399 Levendosky, A. A., 170 Levin, M., 6, 34, 164, 178, 208, 469, 473, 486, 515 Levinson, D. J., 440 Lewin-Bizan, S., 17 Lewinsohn, P. M., 327-328 Lewis, A. E., 168, 520 Lewis, E., 350 Lewis, M., 121, 123 Lewis, T. O., 135, 149, 181, 185, 433, 460, 476 Liabo, K., 189 Liang, Belle, 24, 48, 58, 77, 109, 133, 145, 155, 159, 165, 169, 175, 179, 196, 206, 216, 416, 427, 432-433, 445-447, 512, 514, 518-519, 531 Liberzon, I., 121 Lickliter, R., 118 Lieb, R., 499, 502, 505 Light, K., 117, 120 Lignugaris/Kraft, B., 308, 311, 430 Lim, S., 499 Lindeman, E., 86 Lindenberger, U., 19 Lindhiem, O., 121 Lindsey-Dennis, L. A., 273, 284 Lindwall, J., 243, 248, 241, 251, 253, 451

Author Index  545 Linnan, L., 485 Linnehan, F., 136, 151, 262, 295, 433 Lipman, E., 193, 473 Lippman, L., 526 Lipsey, M. W., 369 Litchfield, L., 161, 211, 320, 473 Liu, C. H. J., 113 Liu, H., 266-267 Liu, W. M., 177 Loevinger, J., 100 Lofstrom, M., 303-304, 311-312 Lombard, M., 264 Long, J. D., 32 Lopez, E. M., 360-361 Lorenz, F. O., 327 Lorenzetti, D., 9 LoSciuto, L., 88, 208, 450, 457, 463 Lou, A., 400 Lounsbury, J. W., 135, 445, 457 Lovallo, W., 121 Lowe, S. R., 17, 22, 161, 190, 211, 224, 320, 322, 427, 447, 473 Lozano, P., 8, 346 Lucas, P., 189 Lucca, N., 191-192 Luckow, D., 189 Lundeberg, T., 120 Lupien, S., 119, 123 Luthar S. S., 244, 342, 361-363 Lutz, C. J., 403, 405, 432, 436 Lydgate, T., 234 Lyles, C., 495 Lynch, M. E., 160 Lyons, N. P., 160 Macintyre, R., 266 MacLeod, D., 356 MacRae, P., 416, 419-420, 422-423 Madia, B. P., 403, 405, 432, 436 Madson, L., 160 Magnuson, K. A., 176-178, 495 Magnusson, D., 170 Maguire, C. P., 33, 529 Magura, S., 244 Mahalik, J. R., 160 Mahmud, A. A., 499 Mahoney, J. L., 224 Maier, G. W., 84 Maio, G. R., 400, 405 Malcolm, K., 33, 35, 48, 334, 457, 469 Malec, J. F., 86 Mallen, M. J., 259, 268 Malley, J. E., 168, 520 Mallory, J. E., 308, 311 Malti, Tina, 105, 113, 459 Mancuso, A., 386

Mandsager, C., 431, 531 Mangelsdorf, S. C., 161, 472 Manka, L., 245 Manni, M., 121 Mansoory, S., 190 Manza, J., 176-177 Mapp, K. L., 458 Maramba, G. G., 146 Marecek, J., 163 Marin, B., 459 Marin, G., 459 Markey, C. N., 428 Markey, P. M., 428 Markus, H., 176 Marsh, S. C., 317 Marshall, J. P., 224 Marshall, S. K., 160 Marold, D. B., 160 Marson, S. M., 85-86 Martin, J. K., 335 Martinez, C. R., 369 Maruschak, L. M., 369-370 Marx, J. A., 472 Mascher, J., 145 Masse, L. N., 499 Masten, A. S., 32, 34, 79, 315-346 Mathur, S. R., 317, 320-321 Maton, K. I., 132, 277, 369 Matos, L., 87 Matsuda, S., 166, 192, 207 Maurice, A., 505 Maxwell, C., 170 Mayfield, J., 499, 502, 505 Mayseless, O., 132, 136, 189-190, 192-193, 531 Mayo-Wilson, E., 9, 303, 530 McAdams, D. P., 109 McClanahan, W. S., 46, 177, 203, 432, 439 McClellan, W., 88, 331 McClendon, S. C., 273, 284 McCord, J., 34, 235, 273 McCrary, S., 308 McDaniel, K., 150 McDonald, K. E., 92, 527 McDonald, S., 8, 21-22, 136, 150, 180, 207, 291-292, 306 McDougall, J., 86 McEwen, B., 121, 123-124 Mcguire, A., 500 McGuire, L., 326 McKenna, K. Y. A., 262 McIsaac, M. S., 449 McLaughlin, C. S., 160-161 McLellan, J. A., 295 McMaken, J., 6, 34, 178, 212-213, 234, 403, 487, 489, 506, 512 McManus, S. E., 471

546   HANDBOOK OF YOUTH MENTORING McMillen, J. C., 134, 150, 346-347, 352 McNichol, E., 385-386 Meade, N. L., 401 Meade, T., 262, 264 Mecham, C., 361-363, 519 Meehan, B. T., 36, 149, 307, 332, 461 Meeus, W., 161-162 Meissen, G. J., 135, 445, 457 Mekinda, Megan A., 489 Merriam, S. B., 259, 264 Messian, N., 147 Metz, E. C., 400 Metz, H., 473 Mickus, M., 166, 347 Middlebrook, J. S., 413, 416 Mihalic, S., 31, 38 Mikulincer, M., 399, 471 Milburn, M., 356 Millenky, M., 428-429, 434 Miller, A., 189, 192-193, 196, 457, 514 Miller, G., 121 Miller, H., 259, 261-262, 264-265, 267-268 Miller, J. B., 160 Miller, L., 159, 166 Miller, M., 499, 502, 505 Miller, S., 437 Miller-Johnson, S., 29 Milliken, Bill, 234 Minuchin, P., 371 Miyake, K., 192 Miyashita, T., 189, 192 Moffitt, T. E., 33 Moher, D., 495 Monaghan, D., 309 Moneta, G., 530 Montoya, R. M., 147 Moodie, M. L., 505 Moody, J., 130 Moon, A., 280 Moon, T. R., 461 Moore, K., 19, 526 Morgan, D., 130 Morland, Lyn, 364, 447 Morris, A. S., 99 Morris, P. A., 495 Morrow, K. V., 37, 47, 49, 66, 68-71, 74, 78, 109, 135, 164, 444, 447-448, 457, 460 Morrow-Howell, N., 308 Mortimer, J. T., 292, 294-295 Morton, T., 332 Mousseau, J., 196 Movelli, G., 192 Moya, J., 222-223, 225-228 Mrazek, P. J., 29-30 Muir, K., 153, 155, 361 Muller-Kohlenberg, H., 78, 195

Muller-Ravett, S., 428-429, 434 Mumola, C. J., 369 Mun, E. Y., 239, 242 Munakata, T., 193-194 Munson, M. R., 134, 150, 346-347, 352 Munoz, R. F., 29 Murphy, J. M., 336 Murphy, R., 328 Murphy, S. E., 149, 159, 266, 280, 284, 405, 443 Murray, C., 327 Murray, J., 369 Murray, L., 18 Mustillo, S., 325 Myrick, R. D., 247 Nakkula, Michael J., 9, 37, 46-50, 56, 59, 64, 75, 84, 89, 91, 110, 205, 212, 291-293, 306, 355, 430, 442, 444, 531 Neal, M., 130 Neely, M. A., 86 Negash, E., 355 Nelson, G., 517 Nelson, J. R., 333-334 Nelson, M. M., 325 Neville, H. A., 35, 47, 74, 76, 165, 178, 205, 308, 343, 348, 439, 444, 515, 517 Newgent, R. A., 32 Newman, K. S., 295 Newton, R., 369 Ng, T., 9, 303, 329, 394 Niec, L. N., 32, 135, 278-279, 331-332, 334 Niego, S., 165 Nissen, E., 120 Noam, Gil G., 35, 99, 102-103, 105, 109, 113, 133, 206, 459 Norcross, J. C., 48 Nordness, P. D., 333-334 Norem-Hebeisen, A., 234-235 Nores, M., 499, 502, 505 Norwood, C., 244 Notaro, P. C., 371 Nov, O., 264 Nutley, S., 10 Nye, B., 211 Nygaard, J., 372 Nyonyintono, R. M., 193-194 O’Brien, B., 500, 506 O’Brysan, E., 261 O’Connell, M. E., 38 O’Connor, R., 402 O’Grady, K. E., 32, 152, 279 Obradović, J., 32 O’Donnell, Cliff, 65, 91, 234 O’Hara, D., 244, 251 O’Neill, D. K., 261, 265, 268-269

Author Index  547 O’Regan, C., 189 Obasi, E. M., 166, 433 Offord, D. R., 30 Ogbu, J. A., 358 Ogle, N. T., 32 Ogrodniczuk, J. S., 471 Oishi, S., 399, 406 Okubo, Y., 244, 342, 361-363 Okun, M. A., 398 Olafsdottir, S., 335 Olds, D. L., 500 Olfson, M., 326 Oliver, B. B., 416, 419-420, 422-423 Olsho, L., 6, 34, 164, 178, 208, 469, 473, 486, 515 Olson, L., 412-414, 416 Omoto, A. M., 398-399, 402, 406 Onuoha, F. N., 193-194 Oosahwe, E. S. L., 149 Oppedal, B., 359 Orens, A., 499 Orenstein, L., 398 Ormrod, R., 412, 414, 417 Ornelas, B., 362-363 Ortega, M., 152 Ortiz, B. G., 398 Osborne, L. N., 71, 75, 85, 448 Osterling, K. L., 347 Ouellet-Morin, I., 119, 123 Pachan, M., 35, 210 Padilla, A. M., 361 Pahl, K., 147, 150, 163 Palmer, D. J., 48, 515 Parise, M. R., 109 Parke, R. G., 371-372 Parra, G. R., 35, 47, 74, 76, 165, 178, 205, 308, 343, 348, 439, 444, 515, 517 Patel, H., 500 Patel, S., 147 Paterson, J. E., 161 Patrick, H., 204 Patterson, G. R., 321, 446 Patterson, J., 412, 416 Payne, R., 183 Pearce, N., 294, 530 Pederson, P. J., 75 Peirson, L., 517 Pelham, W., 325 Peloso, E., 121 Penn, L. S., 472 Pennucci, A., 499, 502, 505 Pepper, S. K., 46, 203, 225 Peppler, K. A., 222-223, 225-228 Perez-Brena, N., 362-363 Perez-Vidal, A., 359 Peris, T., 325

Perkins, D. F., 19, 228 Perlick, D., 327 Perlmutter, F., 236, 530 Perry, B., 335 Perry, D. G., 163 Pescosolido, B. A., 335 Petersen, A. C., 160 Petersson, M., 120 Pettigrew, T., 181, 183 Pettit, G. S., 30 Pfaff, D., 120 Phelps, C., 500 Phelps, E., 17, 19-21, 500 Philip, K., 177 Phillips, J., 32, 270, 526 Phinney, J., 153, 362-363 Piacentini, J., 325 Piaget, J., 99-100,103, 105-106 Pianta, R. C., 204, 441, 446 Pickrel, S. G., 500 Pietromonaco, P. R., 162 Pike, K., 283 Pina, A., 325 Piotrowski, C., 151, 168, 320, 362-363 Piper, W. E., termination in psychotherapy and, 471 Pipkin, C., 264 Pistrang, N., 165 Pittman, K., 19-21 Planty, M., 400 Pleiss, M. K., 445-446 Podorefsky, D., 104 Poehlmann, J., 371-372, 473, 476 Pollak, S., 121-122 Ponterotto, J., 161 Porcellini, Lorraine, 135, 360, 427 Porche, M. V., 160 Portes, A., 356, 359 Portillo, N., 3, 32, 76-77, 83, 92, 100, 147, 163, 175, 190, 223, 245, 278, 291, 316, 330, 341, 356, 411, 429, 457, 492, 528 Postmes, T., 264 Pothong, P., 32, 279 Portwood, S. G., 70, 87, 204, 208-210, 273, 355, 447 Pott, M., 192 Potts, I., 294, 430 Poulin, F., 34, 235 Povinelli, N., 35, 74, 178, 205, 308, 348, 439, 517 Powell, B., 17, 109, 135, 242, 309, 462 Powell, S. R., 207, 238, 242, 246 Powers, L. E., 341 Prelow, H., 274 Presson, C. C., 30 Preuss, L., 335 Prilleltensky, I., 147, 517

548   HANDBOOK OF YOUTH MENTORING Pryce, Julia, 72-73, 75, 78, 134, 150-151, 169, 194, 212, 251, 433, 531 Probst, K., 246 Pryor, J., 161 Pugh-Lilly, A. O., 35, 47, 74, 76, 165, 178, 205, 308, 343, 348, 439, 444, 515, 517 Putnam, R. D., 177, 261 Puzzanchera, C., 315, 320 Quintana, S. M., 472 Quotah, E., 230 Ragan, J., 327 Raines, J. C., 430 Rajala, A. K., 88, 208, 450, 457, 463 Raley, B., 225 Ramsay, D., 121 Randall, B., 283 Randall, P., 327 Randolph, K. A., 303-304 Rao, B., 264 Rappaport, C. D., 6, 34, 164, 178, 208, 469, 486, 515 Rappee, R. M., 206, 327 Raskoff, S. A., 400 Rathunde, K., 441 Raver, C. C., 176 Realmuto, G. M., 333, 461, 464 Redding, R. E., 351 Reddy, R., 36, 46, 89-90, 135, 308, 317, 443 Regnier, M., 400 Reid, J. B., 371-372 Reinherz, H., 235, 530 Reiter, S., 32, 262-265, 267-270 Resch, N. L., 31, 34, 135, 209, 308, 430, 491 Resnick, L. B., 296 Rew, L., 189, 195 Reyes, M., 189 Reyes, O., 134, 207, 360, 457, 530 Rhodes, Jean E., 5, 8, 9, 17, 22, 31-32, 34-37, 39, 45-51, 67, 71, 76, 83, 85, 89-91, 96, 111, 118, 122-123, 126, 133, 135-136, 146-149, 161-163, 164-165, 169, 175, 177-178, 183, 185, 190, 192, 196, 204, 206, 208, 211-212, 216, 221, 223-224, 234, 237, 240, 245, 259, 261, 264, 267, 269, 273, 278, 285-286, 291, 295, 305-306, 308-309, 320, 322, 326-330, 333-335, 337, 341-343, 344-345, 347-349, 356, 360, 371, 380, 398, 400, 402-404, 411, 416, 427, 429-430, 434, 439-440, 443-445, 447-448, 452, 457, 469, 472-474, 486, 492-493, 495-496, 511-514-515, 517-518, 528-529, 531 Rice, M., 122 Richardson, L. P., 8, 346 Richman, J. M., 204

Richmond, L., 274, 318 Rickwood, D., 165 Rigol-Dahn, R. M., 531 Riketta, M., 400, 405 Rincon, C., 132, 512 Ringle, J., 276 Rio, A., 359 Ritter, G., 210, 307 Roach, A. T., 87, 91 Roaf, P. A., 402, 411-412, 414-417, 421 Roberts, H., 189 Roberts, K., 183 Roder, A. E., 442, 450 Roder, L. E., 273 Rodrigues, S., 123 Rodriguez-Planas, N., 307-308 Roe, D., 472 Roffmann, J. G., 22, 36, 46, 89-90, 317, 356, 443 Rogers, C. R., 67 Rogers, T., 412-414, 416 Rogoff, B., 296 Rohner, R. P., 342, 459 Rohrbach, L. A., 207 Rollin, S. A., 294, 430 Rosenberg, W. L., 262, 265, 267-269 Rosenblum, A., 244 Rosenfeld, L. B., 204 Rosenthal, R., 204-205 Rosopa, P. J., 240, 531 Rotenberg, L., 261 Roth, J. L., 18-20, 22 Rothbaum, F., 192 Rothman, A. J., 399, 406 Rothney, A., 196 Rowland, M. D., 322 Rowley, S. A. J., 147 Roy, M., 121, 205 Roy, R., 205 Roy-Carlson, L., 149, 404, 432 Roysamb, E., 359 Roysircar, G., 152 Rubin, R., 120 Russell, J. E. A., 471 Russell, L., 189, 500, 504 Rutherford, R. B., 347, 320-321 Rutter, M., 30, 343, 371 Ruttle, P., 122 Ryan, C. S., 398 Ryan, J. A. M., 18, 204 Ryan, R. M., 190, 196, 305, 401 Ryerson-Espino, S., 358 Sabatelli, R. M., 344 Sadler, R., 261 Sagiv, L., 191 Saito, R. N., 8, 259, 374

Author Index  549 Saiz, C. C., 51, 56 Sam, D. L., 359 Sameroff, A. J., 371 Sánchez, Bernadette, 78, 134, 138, 145-146, 149-151, 154-155, 159, 179-180, 183, 196, 207, 287, 309, 349, 360, 427, 432, 457, 459, 519, 530 Sanderson, J., 344, 372 Sandler, I. N., 161 Santisteban, D., 359 Santon-Salazar, R., 361 Santos, A. B., 333 Santos, K., 248, 251, 428, 430, 434 Saslow, L., 123 Sass, D., 163, 207 Savoy, H. B., 151, 154 Sawrikar, P., 153, 155, 361 Sazama, J., 165, 205, 517 Scales, P. C., 3, 6, 17, 511, 526 Scandura, T. A., 259 Scealy, M., 270 Scharf, Miri, 146, 189, 192 Scheithauer, H., 100 Schippers, M., 149-150, 166-167, 432 Schmiesing, R., 411-412, 417 Schneider, B. H., 485 Schneirla, T. C., 19 Schoenwald, S. K., 316, 322, 333 Schoeny, M., 9, 486 Schonert-Reichl, K. A., 160 Schrepf, S., 333-334 Schultheiss, O. C., 84 Sculpher, M., 506 Schultz, L. H., 69, 79, 103-104 Schulz, K. F., 495 Schwartz, S. E. O., 8, 36-37, 39, 135, 190-191, 193-194, 196-197, 208, 211-212, 215, 297, 429, 431, 473, 493, 515, 531 Schweinhart, L., 499, 502, 505 Schweitzer, L. J., 499 Scielzo, S. A., 240, 531 Scott, D., 88 Scott, J., 175 Scott, Jr., L. D., 346 Scott-Nakai, K., 246, 248 Sears, H. A., 160-162 Seay, A., 33 Seeman, T., 121, 123 Seidl, A., 499 Seifer, R., 371 Sekol, I., 369 Sellers, R. M., 147, 153 Selman, R. L., 69, 79, 103-104, 204, 206 Sendi, P., 506 Sepulveda, S., 121 Serido, J., 228

Serumaga-Zake, P. A., 193-194 Sesma, A., 17, 526 Sethre-Hofstad, L., 122 Shadduck-Hernandez, J., 362-363 Shaffer, D., 328 Shakeshaft, C., 413 Shanahan, M., 8-9, 124 Shao, P., 262, 265, 267-269 Shapira, N., 262 Shaver, K. H., 445 Shaver, P. R., 161, 399, 428, 469, 471 Shehan, C. L., 344 Sheldon, J., 221, 225 Shelton, J. N., 147 Sheridan, M., 326 Sherk, J., 284-286 Sherman, R., 85-86 Sherman, S. J., 30 Sherrod, L., 19 Shin, S., 146, 189, 195 Shin, Y., 195 Shirk, S. R., 51, 56 Shirtcliff, Elizabeth A., 119-123 Shlafer, R. J., 371-372, 473, 476 Shoenwald, S. K., 500 Shollenberger, T., 373 Short, J. A., 264 Shotton, H. J., 149 Shpigelman, Carmit-Noa, 262-265, 267-270, 350, 449 Shucksmith, J., 189, 457 Shulha, L. M., 484 Shwiff, S. A., 499 Siegel, J., 150-151, 500, 504 Siegel, T. C., 326 Siever, L., 121 Silbert, E., 151, 168, 320, 362-363 Silverman, B., 216 Silverman, W. , 325, 328 Silverthorn, N., 3, 21-22, 30, 32, 76-78, 83, 100, 134, 137, 147, 150-151, 163, 175, 178, 190, 203, 207, 223, 245, 278, 291-292, 309, 316, 329-330, 341, 356, 371, 411, 427, 429, 443, 457, 472, 485, 489-492, 494, 527-528 Simon, B. D., 152 Simon, P., 239, 242 Simon, S. A., 471 Simons, J., 87 Sims, V. M., 177 Sinclair, M. F., 139, 332, 530 Singh, J., 360, 530 Single, P. B., 259, 264, 267 Single, R. M., 259, 264, 267 Sipe, C. L., 8, 46, 70, 83, 177, 203, 259, 273, 307, 311, 355, 374, 432, 439, 442, 447, 450, 457 Sipe, W. S., 48

550   HANDBOOK OF YOUTH MENTORING Sirey, J., 327 Slade, S., 101 Slicker, E. K., 48, 515 Smalling, S. E., 346 Smashnaya, S., 342 Smedley, A., 145-146 Smedley, B. D., 145-146 Smith, A., 33, 35-36, 38, 86 Smith, D. K., 333 Smith, H. T., 499-500 Smith, K. E., 531 Smith, L. H., 241, 248 Smith, M. A., 147, 411-414, 416-417, 419-423 Smith, T. B., 148 Smith, T. J., 370, 374 Smith-Jentsch, K. A., 240, 531 Smolen, A., 8-9, 124 Snarey, J., 100 Snibbe, A., 176 Snider, J. B., 8, 193, 292 Snyder, Mark, 398-400, 403, 406, 445 Soenens, B., 87 Solomon, F., 152, 155 Sosik, J. J., 320 Soto, D., 284 Soucy, N., 161, 304, 306-308, 459 Sovik-Johnston, A., 183, 283 Sparks, J., 437 Spencer, Renée, 35, 48, 64, 67, 84, 109, 133, 135, 138, 148-149, 151, 164-165, 169, 175, 177, 180-182, 185, 196, 205-206, 213, 217, 236, 259, 264, 342, 346, 352, 405, 412, 416, 429, 433, 439, 444-448, 460, 463, 466, 469-470, 473, 475, 476, 512-515, 517, 519, 527, 531 Spears, R., 264 Spina, S. U., 132, 134, 149 Spitz, Rene, 117 Spitznagel, E., 308 Spracklen, K. M., 321, 446 Spragins, W., 9 Springer, J. F., 32 Sribney, C., 440 Sroufe, L. A., 100, 371 St. Aubin, E., 109 St. Claire-Smith, S., 351 Stansbury, K., 122 Stanton-Salazar, R. D., 132, 134, 149 Staudinger, U. M., 19 Stavrinides, P., 161 Steckler, A., 485 Steele, C. M., 516 Steenbeek, D., 86-87 Steiman, A. H. K., 244 Stein, B. D., 356 Steinberg, L., 99

Stellar, E., 123 Steptoe, A., 121 Stern, D., 295 Stern, M. J., 499 Stevenson, R., 270 Stewart, A. J., 168, 520 Stewart, M., 506 Stice, E., 327 Stiver, I. P., 159-160 Stoddart, G. L., 500 Stolinski, A., 398 Stone, S., 430 Stoolmiller, M., 122 Streit, K., 526 Strom-Gottfried, K., 514 Stuhlman, M. W., 204 Stukas, Arthur A., 397, 399-402, 404, 436, 445, 485 Stürmer, S., 398, 403 Styles, M. B., 37, 47, 49, 66, 68-71, 74, 78, 109, 135, 164, 444, 447-448, 457, 460 Suarez-Orozco, C., 356, 447 Suarez-Orozco, M. M., 447 Sue, D. W., 177, 519 Sue, S., 148 Suhr, J. A., 264 Sukhodolsky, D., 275 Sullivan, J., 413 Sullivan, K., 160 Sullivan, P. J., 134, 230 Sundeen, R. A., 400 Sundell, K., 190 Surrey, J. L., 159 Sussman, S., 207 Suter, J. C., 333 Swank, P. R., 531 Swinton, J., 249 Syed, M., 182, 432 Szapocznik, J., 359 Takeuchi, S. A., 429 Tal, S. A., 265 Talge, N., 121-123 Tallon, J., 411-413 Tanis, M., 264 Tanti, C., 397, 399-402, 404 Tarabulsy, George M., 90, 132, 205, 305-306, 309, 311, 432 Taussig, H. N., 31-32, 149, 347, 349-350 Taylor, Andrea S., 48, 51-52, 70, 88, 109, 119, 121, 134-136, 166, 177, 208, 273, 355,360, 427, 447, 451, 462-464, 470 Taylor-Ritzler, T., 85, 92 Teague, G. B., 432 Tebes, J. K., 30, 137, 484

Author Index  551 Tennant, A., 86 Tennes, K., 121 Terry, K., 411-413 Tharp, R. G., 65 Theodorou, E., 177 Theokas, C., 17, 20-21, 526-527 Thomas, D., 245 Thomas, K. M., 149, 432 Thomas, R. E., 9 Thomason, Jessica D., 74, 139 Thomson, N., 48, 51-52, 136, 166 Thorndike, A., 183 Thurman, T. R., 189 Thurlow, M. L., 312, 333 Tierney, J. P., 34, 38, 51, 64, 71, 209, 234, 370, 402, 411-412, 414-417, 421, 462, 491, 495, 505 Timlin-Scalera, R. A., 161-162 Tobach, E., 19 Tobler, N. S., 211 Todis, B., 317, 319 Tolan, P., 9, 486 Tolman, D. L., 160 Topping, K., 235 Torrance, G. W., 500 Torres Campos, C., 362-363 Townsend, T. N.88, 208, 450, 457, 463 Toyokawa, T., 166, 192, 207, 520 Tracy, A. J.58, 165, 346 Tranter, D., 90 Trautman, P., 328 Travis, J., 370 Triandis, H. C., 191-192 Trickett, E. J., 322, 358-359 Trivette, C. M., 345 Tropp, L. R., 181, 183 Tsikalas, K., 262, 264 Tu, M., 119, 123 Tu, X., 37 Tupler, L., 325 Turnbull, B., 189, 193 Turner, S. G., 151, 168, 320, 362-363 Turner-Stokes, L., 86, 91-92, 96 Tyler, J. H., 303-304, 311-312 Tyrrell, C. L., 432 Uhl, J., 151, 433 Underwager, R. C., 235 Unger, D., 151 Unwin, L., 295 Urban, J. B., 17, 20-21 Urquiola, M., 295 Urso Spina, S., 361 USAID, 193, 195 Usher, B., 121

Usher, L., 346 Usita, P. M., 402 Utsey, S. O., 273, 282, 284 Uvnas-Moberg, K., 119-120 Uzoebo, V., 348 Vaisey, S., 8-9, 124 Valentine, J. C., 3, 32, 34, 50, 64, 76-77, 83, 100, 135, 147, 163, 166, 175, 190, 209, 223-224, 235-236, 245, 259, 278, 291, 306, 316, 330, 341, 374, 411, 429, 431, 439, 457, 481, 491-492, 494-495, 528 van Aken, M. A, 161 van Dam, C., 411, 414 van der Werf, M. P. C., 205 Van Gool, K., 506 van Nijnatten, C., 369, 372 Van Ryzin, M., 206, 276-278 Vang, Z., 273 Vansteenkiste, M., 87, 190 Varenhorst, B., 234, 246, 253, 530 Vazquez, D., 121, 123 Vazsonyi, A. T., 8, 193, 292 Vidourek, R. A., 88, 331 Viggers, D., 89 Villareal, J., 191-192 Villarruel, F. A., 19 Vincent, T. A., 322 Vinogradov, S., 274 Viswesvaran, C., 325 Vohs, K. D., 401, 469 von Eye, A., 170 Vrasidas, C., 449 Vygotsky, L. S., 105, 112 Wade, D. T., 85 Wahlsten, D., 118 Waintrup, M., 317, 319 Wakefield, W. D., 147 Walker, G., 6, 221, 223-225, 227-228, 404-405 Walker, K. E., 227-230, 294, 530 Wallace, B. C., 239 Waller, N. G., 399 Walsh-Samp, K., 161, 211, 320, 473 Walters, R. H., 10, 162 Wampold, B. E., 66-67 Wandersman, A., 406 Wang, Y., 151, 154 Wänke, M., 30 Ward, D. M., 500 Ward, R., 342 Waris, R., 87 Warner, K. E., 38 Wasley, P., 394 Wasserman, D., 85-86

552   HANDBOOK OF YOUTH MENTORING Wasserman, R., 150 Wasserman, S., 137 Waters, P. L., 160 Watkins, M., 275 Watkins, N. D., 134, 230 Watton, J., 351 Watts, A. M., 32, 279 Waul, M., 370 Way, N., 147, 150, 163, 295 Webster, B., 163, 386 Weeden, K. A., 177 Weer, C., 151, 433 Wei, G., 85-86 Weinberger, Susan G.230, 234, 390, 475, 485, 423, 464 Weinstein, M. C., 500, 504 Weinstein, T., 358 Weiss, P. L., 32, 262-265, 267-270 Weissberg, R. P., 35, 210, 440 Weissman, M., 369, 372 Weisz, J. R.161, 192, 332, 334 Weitlauf, J., 85 Wellman, B., 130 Wells, R., 121 Wenger, E., 296 Werner, E. E., 34, 101, 180, 371, 469 Wernli, M. A., 398 Wurtele, S., 414 Wertheim, C., 190, 193 West, H., 370 West, J., 145, 155, 175, 179 Westerlund, D., 242-243 Westhues, A., 351 Westman, M., 132 Wheeler, M. E., 6, 9-10, 34-35, 75, 135, 210, 261, 312, 429, 442, 449-450, 462, 489, 491, 496 Westermeyer, J. F., 109 Western, B., 369 Whalen, C., 123 White, D. R., 130 Whitesell, N. R., 160 Whiting, M. A., 308, 311 Wickrama, K. A. S., 327 Widstrom, A., 120 Wildeman, C., 369 Wiley, T., 514 Williams, H., 86 Williams, I., 65 Williams, L., 165 Williams, O., 273 Williams, R. A., 160 Willis, P., 245 Wilson, K., 65 Wilson, W. J., 132 Wilson-Ahlstrom, A., 483, 492-493 Wimer, C., 458

Wingfield, J., 121 Wise, D. L., 87 Wohlfarth, T., 177 Wolfaardt, U., 440 Wolfe, B. L., 501, 505 Wong, K., 3, 216, 221, 225 Wong, M., 356 Wong, V., 231 Woo, S. M., 328 Wood, S., 9, 303, 530 Woodhouse, S. S., 471-472 Woody, J. D., 460, 463, 466 Woolum, S., 75 Worth, K., 399 Wright, V., 86 Wrobel, T., 386 Wyman, P. A., 37, 75 Xiaoxia, H., 506 Xue, Y., 149 Yakushko, O. F., 151, 154 Yalom, I., 274 Yamauchi, L. A., 152 Yancey, A., 150-151, 343 Yarbrough, C. S., 240, 531 Yarbrough, D. B., 484 Yates, B. T., 30, 137, 484, 500 Yeh, C. J., 244, 342, 361-363 Yellin, E. M., 318 Yohalem, N., 483, 492-493 Young, C. H., 105 Younggren, J. N., 513 Youniss, J., 19, 295, 400 Yu, Q., 37 Zahn-Waxler, C., 121 Zak, P., 125-126 Zambrano, E., 237 Zand, D. H., 48, 51-52, 136, 166 Zarrett, N., 21 Zeki, S., 119 Zerbe, R. O., 500-502 Zhang, D., 36, 149, 307, 332, 461 Zheng, W., 264 Zhou, E., 121 Zhou, M., 359 Ziegler, T., 122 Zigler, E. F., 160 Zimmerman, H., 506 Zimmerman, M., 8, 149-150, 162, 165, 179-181, 207-208, 216, 223, 236, 315, 371, 459, 527-528 Zimmerman, S., 20 Zippay, A., 136 Zirkel, S., 162 Zurbriggen, E. L., 182, 432

SUBJECT INDEX

Abt Associates study, school-based mentoring (SBM) and, 208-209 Academically at-risk student (AARS) attachment theory and, 304 checklist for practitioners and, 310 determinants of participation in mentoring and, 306 effects and explanatory processes of mentoring and, 307-308 matching mentors and mentees, 311 mediating processes and, 308-309 mentoring relationship and, 311-312 mentoring sociomotivational model and, 305 mentor relationship quality (MRQ) and, 306-307 mentor training and supervision, 311 moderating processes and, 308 natural mentoring and, 309, 312 negative external influences and, 303 negative internal forces and, 303 overview of, 303 program promotion and mentor selection, 310-311 quasi-experimental studies and, 309 research and, 306-309 systemic model of youth mentoring and, 305 theory and, 304-306 youth mentoring model and, 305 Acculturation theory, immigrant youth and, 357-359 Across Ages program family involvement, 464 in-person training and, 450-451 Add Health survey mentoring assessment from, 292, 528 naturally occurring mentors and, 346 Adolescent connectedness theory, cross-age peer mentoring and, 237-238 Adrenocorticotropin-releasing hormone (ACTH), reactive Cortisol and, 121 Adult Relationships Scale, 51, 56 Advocacy, mentoring activities and, 65-66, 76-77

Africa, protégés and, 193 African American youth attitudes toward mentoring and, 345 “bridge” programs for, 361 CPY program and, 376 cultural competence and, 152 cultural similarity or dissimilarity and, 148-149 Friends of the Children (FOTC) program, 377 girls preference for mentors and, 166 high school dropout rates and, 304 incarcerated parents and, 370 juvenile offenders and, 318 mentoring roles and, 151 students’ trust beliefs about adult mentors and, 433 work-based mentoring programs and, 136 African Immigrant Mentoring (AIM) program, 363 African Youth Advisory Council, 363 After-school programs adult-youth interactions and, 230 characteristics of mentoring relationships and, 222-223 characteristics of mentoring relationships and, 225-226 community resources and, 230 determinants of mentoring behavior and, 223-224, 227-228 environmental determinants of mentoring behavior and, 223-224 individual characteristics of, 223 mentoring arrangement and, 222-223 mentoring behaviors and, 223 outcomes and, 228 overview of, 221 program experience and outcomes, 224 recommended future directions and, 228-229 revised framework for investigating mentoring relationships and, 222 staff and, 229-230 theory and, 221-224 Aggressive children, mentoring programs for, 36, 108, 332, 461-462 553

554   HANDBOOK OF YOUTH MENTORING AIDS, skills and abilities of mentors and, 193-194, 398, 464 Alliance for Youth, 525 Allo-parenting, definition of, 3 American Association of Fundraising Counsel and Giving Institute, program funding, 391 American Psychological Association (APA) assigning to racial categories and, 145 ethical issues and, 512 Ethics Code and, 516 mentor training and, 448 ongoing self-evaluation regarding values and culture, 519 America’s Promise, Colin Powell and, 525 Aos study, costs and benefits of mentoring programs, 505 Asian American youth, cultural similarity or dissimilarity and, 148-149 Asset Builder’s Guide to Training Peer Helpers, 246 Attachment theory aggressive children and, 108 children of incarcerated parents and, 371, 376 corrective emotional experiences and, 429, 433 first few years of life and, 105 foster care and, 348, 351 gender and, 161-162 mentoring pairings and, 428 mentoring research and, 304 nature and influence of close relationships and, 343-344 overview of, 100-101 prior experiences and, 104 prevention literature view and, 111 same- versus cross-matched mentoring and, 159 youth matching and, 428 youth mentoring literature and, 427 Attention-deficit/hyperactivity disorder (ADHD), 325 Australian National Youth Affairs Research Scheme, 155 Australian Youth Mentoring Network, 8 Authorship as collaboration and, 75-76 mentoring activities and, 79-80 Avoidant behaviors anxious-ambivalent and, 101 anxious-avoidant and, 101 Balu und Du (Baloo and You), 78 Banny, Adrienne, 439 Basal cortisol, 120-121 Beacon Centers, 221, 226-228 Bernard, Claude, functioning in the face of environmental changes and, 119

Big Brothers Big Sisters of America (BBBSA) Boys & Girls Club sites and, 230 child abuse and, 348, 413 children of incarcerated parents, 370, 373 cross-age peer mentoring and, 238 economic evaluation of, 505 effect of longer relationships and, 515 “enhanced” model of SBM and, 217 female-led households and, 211 formal mentoring and, 307, 310 Group Assessment of Interpersonal Traits and, 402 high-profile evaluations of, 496 High School Bigs Demonstration Program, 240, 249 match support and, 449 meeting and match closure practices and, 252 mentor-mentee relationship and, 181 mentor recruitment and, 402-403 mentor relationship quality (MRQ) and, 306-307 mentor screening and, 415 mentor training and, 442, 450 mentor training and supervision, 311 “natural mentoring” and, 530 peer mentoring programs and, 247 power of volunteerism and, 234 program evaluation and, 487, 490 Public/Private Ventures impact study and, 208-209 randomized controlled trial (RCT), 329 ratings of relationship closeness and, 166 study of relationship formation in sex-matched pairs and, 165 trial evaluation of, 489 using teen mentors and, 235 web-based mentor training programs and, 451-452, 451 youth matching and, 430 Big Brothers Big Sisters International (BBBSI), 8, 10 Big Brothers Big Sisters of Canada (BBBSC), 91, 94-95 child sexual abuse and, 415-416 cross-age peer mentoring and, 238 High School Teen Mentoring Activity Book and, 247-248 meeting and match closure practices and, 252 teen mentoring programs and, 235 Big Brothers Big Sisters of Northern Minnesota, 183 Big Buddies Project, 89 Bill & Melinda Gates Foundation, 10-11 Biological perspective allostatic load and, 123

Subject Index  555 applying biology to youth mentoring and, 125 attunement research and, 122 basic biological processes and, 119 Cortisol and, 118, 120-123 “dandelion children” and, 122 hormonal processes and, 119-120 Oxytocin and, 118, 120, 122-123, 126 positive adult-youth relationships and, 117-126 practice and, 124-126 psychobiological approach and, 118-119 psychobiological benefits of youth mentoring and, 124 reactive Cortisol and, 121 research and, 124 terms and definitions for, 118 theory and, 118-124 Blocks, Arthur, program funding and, 391 Blue Ribbon Mentor-Advocates (BRMA) program, 139 Boy Scouts of America, child sexual abuse and, 413 Boys & Girls Clubs of America after-school programs and, 221, 224-229 group mentoring, 277 statistics for, 5 Boys to Men Mentoring Network, 168 Bridging Refugee Youth and Children’s Services (BRYCS), 363-364 Brothers Project, 168 Building the Foundation, web-based mentor training course and, 452 Bullied children, 20-21, 32, 538, 540 Bush, George W., importance of faith and volunteerism, 370 California Psychological Inventory, mentor screening and, 415 Canada, high school dropout rates and, 303 Canadian Big Brother Big Sister, 193 Center for Evidence-Based Mentoring, 7, 10, 12, 545 Center for Interdisciplinary Mentoring Research, 7, 9, 12, 538, 542 Centers for Medicare and Medicaid Services, 29, 40 Chafee Foster Care Independence Act of 1999, 342 ChalleNGe mentoring program, 431 Chamber of Commerce, program funding and, 390 Chapel Hill/Carrboro City Schools’ Blue Ribbon Mentor-Advocate program, 155 Check and Connect Program, 139, 312, 332-333 Child abuse Big Brothers Big Sisters of America (BBBSA) study, 348 foster care and, 341-342 statistics and, 341-342

Child Abuse Potential Inventory (CAP), 414-415 Children with Adolescent Mentors, 451 Child Safety Audit, child sexual abuse cases and, 415 Child sexual abuse Child Abuse Potential Inventory (CAP), 414-415 definition of, 412 Diana Screen and, 415 sex offender registries and, 422 statistics and, 412-413, 417 China, cross-cultural study and, 194 Clinical Analysis Questionnaire, mentor screening and, 415 Closure of mentoring relationships careful monitoring of matches and, 476 checklist for practitioners and, 475 developing clear policies regarding when relationships should end and, 476 exit interviews and, 476 guidelines for post-program contact and, 477 MENTOR “termination ritual” and, 475 overview of, 469 phase models of mentoring and, 470-471 positive closure and, 477 preparing volunteer mentors and, 476 providing closure activities and, 477 psychotherapy and, 471-472 research and, 472-474 start ending from the beginning and, 475-476 supporting participants throughout the closure process and, 477 theory and, 470-472 Clover Model, 100, 102, 105-107, 110-113 Club Amigas, 362 Club Pocahontas, 225 Cognitive Behavioral Principles Within Group Mentoring, 278 Cognitive-behavioral therapy (CBT) group psychotherapy and, 275 internalizing disorders and, 327-328 major depressive disorder (MDD), 325 mental health needs and, 331-332, 337 Collaboration, mentoring activities and, 65, 79 Committed Partners for Youth–Big Brothers Big Sisters of Lane County (CPY), 374-377 Communities in Schools (CIS), 73, 209-210 Community-based mentoring (CBM) activities and, 89-90, 217 overview of, 203-204 youth characteristics and, 211 Companionship Therapy: Studies in Structured Intimacy (Goodman), 64 Competence positive youth development (PYD) and, 19, 23 risk related and, 48

556   HANDBOOK OF YOUTH MENTORING Computer Clubhouses, 225-227 Computer-mediated communication (CMC) development of, 259-262 e-mentoring and, 265, 268 personalization processes and, 264-265 supportive behaviors and, 264 Computer Supported Cooperative Work (CSCW) system, 265 Confidence, positive youth development (PYD) and, 19, 24 Connecticut Department of Children, foster care and, 348 Contemplation, youth mentoring and, 84 Cool Girls, Inc., 74, 78 Corporation for National and Community Service, 9, 292, 295, 299, 397 Corticotropin-releasing hormone (CRH), reactive Cortisol and, 121 Cortisol, 118, 120-123 Craig, Ashley, 439 Crandall, Floyd, 117 Cross-Age Mentoring Program (CAMP) 3-2-1 activities and, 254 active listening and, 253 adolescent connectedness theory and, 237-238 BBBS High School Bigs program and, 240 checklist for practitioners and, 249-250 closure of mentoring relationships, 477 cognitive developmental theory and, 109 conflict resolution and, 253 deviancy training and, 236-237 family involvement and, 462, 464 helper therapy principle and, 236, 243 high school mentors and creating curriculum and, 253 implementation of, 248 inclusion of diverse populations and, 243-244 matching and, 251 meeting and match closure practices and, 252 mentor and mentee interactions and, 241-243 mentor roles and, 253 mentor training and, 246, 451 mentors with higher scores on social interest and attitudes toward youth and, 251 monitoring program implementation and, 251-252 monitoring status of relationship and, 253-254 negotiation strategies and, 253 origins of, 233-234 overview of, 233 parent and teacher involvement in, 252 peer mentoring interactions and, 79 practice closures and, 252 program evaluation and, 239-243

recruitment and screening of prospective mentors and, 251 relationship life cycle and, 252-253 research and, 238-245 skills-based training activities and, 246 theory and, 236-238 training mentors and, 252 video presentations and, 253 youth development and self-psychology, 237 Cross cultural youth mentoring characteristics of the relationship and, 193-194 checklist for practitioners and, 197 goals and success of, 194-195 implications for, 191-192 overview of, 189-190 regulatory focus theory and, 190-191, 193-197 self-determination theory and, 189-191, 193-197 Cross-sex mentoring, impact on youth and, 168-169 Czech Republic, mentoring programs and, 193 Daura, Sharon, 439 Diana Screen, reducing sexual risk to children and, 415 Digital Heroes Campaign (DHC), e-mentoring and, 260, 266 DO-IT, e-mentoring, 260, 262 Early Risers program family involvement and, 464 mental health needs and, 333 skill building for families and, 461 Economic evaluation benefit-cost analysis and, 501-502 Big Brothers Big Sisters (BBBS), 505 cost-effectiveness acceptability curve (CEAC) and, 502-503 discounting and, 501-502 incremental cost-effectiveness ratio (ICER) and, 502-504 literature on the economics of mentoring and, 505-506 market failure and, 501 overview of, 499 questions to consider and, 507 return on investment and, 502 use of public funds and, 500-501 Electronic Emissary Project, e-mentoring and, 260 Electronic Socio-Emotional Support (ESES), 263, 264 conceptual framework for, 262 e-mentoring and, 270 importance of ESES framework component and, 266 importance of key moderators in, 268

Subject Index  557 Israeli e-mentoring program and, 267 personalization processes and, 264-265 Elements of Effective Practice for Mentoring (EEPM) group mentoring and, 273-274 mentor screening and, 9 mentor training and, 450, 452 Elements of Effective Practice, match closure and, 474 E-mentoring checklist for practitioners and, 269 CMC supportive behaviors and, 264 Computer Supported Cooperative Work (CSCW) system, 265 conceptual framework for, 262 definition of, 259 developmental e-mentoring and, 266-267 development of, 259-260 Digital Heroes Campaign (DHC) and, 266 DO-IT program, 262 face-to-face (FtF) mentoring programs, 267 focus of, 261-262 GED Exit Option program, 267 instrumental programs and, 265-266 intrinsic sources of motivation and, 262, 264 key challenges of, 268-269 MentorNet, 267 personalization processes and, 264-265 research and, 265-268 roots of, 259 theory and, 261-265 typologies of, 261 Empowerment, mentoring activities and, 76-77 Ethical issues acting with integrity and, 515-516 beneficence and, 513 establishing clear expectations and program guidelines, 519 ethical approach to mentoring and, 512 guiding questions for practitioners and, 518 inappropriate boundaries and, 513-514 misuse of power and, 513 notion of fidelity and, 514-515 ongoing self-evaluation regarding values and culture, 519 overview of, 511 promoting justice and, 516 respecting people’s rights and dignity, 516-517 respecting the rights and dignity of youth and, 519-520 screening and training of mentors, 517-518 sensitizing mentors and, 518 supporting mentor behaviors and, 519 theory and, 512-517

Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD), 439 European Union, high school dropout rates and, 303 Evaluation standards framework for, 484-485 institutional review board (IRB), 484 Joint Committee on Standards for Educational Evaluation (JCSEE), 484 Evidence-based prevention tools, 33-40 Face-to-face (FtF) mentoring programs e-mentoring and, 261-262, 267 overview of, 259 Families and Mentors Involved in Learning With Youth (F.A.M.I.L.Y) project, 464 Families One-on-One Mentoring program, foster care and, 348 Family Hardiness Index (FHI), 460 Family involvement Across Ages program, 464 checklist for practitioners and, 465 Cross-Age Mentoring Program (CAMP), 462, 464 Early Risers, 464 Families and Mentors Involved in Learning With Youth (F.A.M.I.L.Y) project, 464 Family Hardiness Index (FHI), 460 family systems theory and, 459 Lunch Buddy program, 461-462, 464 Mentors in Across Ages, 464 overview of, 457 parental acceptance theory and, 459 Parent-Child Relationship Inventory (PCRI), 460 PrimeTime program, 461-462, 464 research and, 460-463 resources for mentoring and, 465 skill building for families and, 461 strategic approaches for, 458 theory and, 458-459 youth-only mentoring and, 461-463 Federal Mentoring Council, 7, 9, 12 Federal Office of Refugee Resettlement, refugee youth and, 363 FLEX, skill building for families and, 461 Focus, mentoring activities and, 68-69, 78 Formal mentors, foster care and, 347-348 Foster care attachment theory and, 351-352 Chafee Foster Care Independence Act of 1999, 342 checklist for practitioners and, 351

558   HANDBOOK OF YOUTH MENTORING effect of mentoring relationships on mentors and, 349 formal mentors and, 347-348 Fostering Connections to Success and Increasing Adoption Act of 2008, 342 human attachment theory and, 343-344 Keller’s contextual mentoring model and, 344 naturally occurring mentors and, 346-347 natural mentors and, 352 Orphan Foundation of America (OFA), 350 parental acceptance-rejection (PAR) theory, 344 research and, 345-349 Rhodes model and, 342-343 self-esteem and, 343 social exchange theory, 344 statistics and, 346 tailoring mentoring practices and, 350 theory and, 342-345 training mentors and, 350-351 unrelated adult mentors and, 150 VISIONS program, 348 youth attitudes toward mentoring and, 345 youth transitioned out of foster care and, 342 FosterClub contract, social networks and, 138 Fostering Connections to Success and Increasing Adoption Act of 2008, 342 Friday Night Live Mentoring Program, 248-249 Friends for Youth, 79, 414 Friends of the Children (FOTC) program, 377-379 GED Exit Option program, e-mentoring, 260, 267 Gender and identity in context, 162-163 attachment theory and, 161-162 checklist for practitioners and, 167-168 differences in mentoring and other identity characteristics, 169 help-seeking behaviors and, 160-161 mentoring practices and, 159-170 relational-cultural theory (RCT) and, 160 relationship development and, 159-160 research and, 163-167 similarity and attraction theory, 162 theory and, 159-163 youth mentoring practice and, 167-169 Germany protégés and, 195 research on apprenticeship and, 293 GirlPOWER!, 74, 78, 168, 216-217, 488 Girl Scout Research Institute, 150 Girls on the Run, 168 Goal attainment scaling (GAS) application of, 86-87 historical background of, 86 original formula for, 86

overview of, 85-86 pilot evaluation of a GAS approach and, 90 process of implementing and, 93-96 recommendations for future research and, 91 used along with standardized measures and, 92 GOAL program, 88 Goal setting Big Brothers Big Sister of Canada and, 94 checklist for practitioners and, 92-93 community-based mentoring and, 89-90 goal attainment scaling and, 85-86 overviw of, 83-84 recommendations for future research and, 90-91 role of, 84-85 school-based mentoring programs and, 85-90 SMART goals and, 94 Go-Betweener Mentoring Program, 363 Great Britain e-mentoring and, 266 mentoring relationships and, 194 “modern apprenticeships” and, 295 National Health Service and, 499 paid professional mentors and, 193 recidivism and, 319 Greeson, Johanna, 439 Group mentoring checklist for practitioners and, 285 Cognitive Behavioral Principles Within Group Mentoring, 278 cognitive-behavioral therapy (CBT), 275 cultural competence and ethnic identity, 286 developmental theory and, 275-276 formalized settings and, 277 insights from group psychotherapy and, 274-275 matching and, 286 mutual help and support group interventions and, 275 opportunities for interaction within, 275 overview of, 273 overview of findings and, 278 programs and, 278-283 recommendations for future research and, 283-284 recruitment and training of mentees, 286 recruitment and training of mentors, 285-286 research and, 276-278, 283-284 School Transition Environment Project (STEP), 277 supervision, reflection, and planning, 286 theoretical model of, 277 theory and, 273-276 Young Women’s Leadership Project, 283 Youth Development Program, 278, 286

Subject Index  559 Health care statistics and, 29 Healthy Kids Mentoring Program, 88, 331 Help-intended communications, 64-65 High School Bigs Demonstration Program, 240-241, 245, 249 Hillside Work-Scholarship Connection program, 297 Hogan Personality Inventory, mentor screening and, 415 Hong Kong, mentoring relationships and, 194 Human attachment theory, foster care and, 343-344 Hypothalamic-pituitary-adrenal (HPA) axis, individual’s stress response system and, 120-122 I Could Be, e-mentoring and, 260, 265 iMentor Interactive (iMi) e-mentoring and, 260 youth matching and, 435 Immigrant youth, see also Refugee youth acculturation and resettlement process, 364 acculturation theory and, 357-359 African Immigrant Mentoring (AIM) program, 363 bicultural adjustment and, 359 Bridging Refugee Youth and Children’s Services (BRYCS), 364 checklist for practitioners and, 365 conceptual model of application of acculturation theory and, 357 ecological theory and, 358 “family mentoring” and, 362, 364-365 family acculturation gaps and, 359 formal mentoring programs and, 361-362 Go-Betweener Mentoring Program, 363 “immigrant paradox” and, 356 mentoring goals and, 359 mentoring structures and, 360 overview of, 355 posttraumatic stress disorder (PTSD) and, 355 research and, 360-363 school-based programs and, 361-362 theory and, 356-360 Incarcerated parents Big Brothers Big Sisters (BBBS), 373 Friends of the Children (FOTC) program, 377-379 Mentoring Children of Prisoners (MCP) program, 370, 372, 376 population characteristics and, 369-370 professional mentoring and, 377-379 recommendations for practitioners and, 375 research and, 372-373

statistics and, 369-370 theory and, 371-372 transactional model of mentoring relationships and, 372 volunteer mentoring and, 376-377 Indiana Through Mentoring (AIM) program, 316-320 Individualized education plan (IEP), mental health needs, 333 Initiation, youth mentoring and, 84 Institutional review board (IRB), evaluation standards, 484 Intergroup contact theory (ICT) differences in social class within mentoring relationships and, 181 four “essential” conditions of, 179 initial phase of mentoring and, 178 reducing social class prejudice and, 185 International Telementoring Project, e-mentoring and, 260 International youth mentoring characteristics of mentors and protégés, 192-193 characteristics of the relationship and, 193-194 goals and success of, 194-195 overview of, 189-190 research and, 192-195 Israel e-mentoring program and, 267 Perach project and, 193 John Templeton Foundation, 17 Joint Committee on Standards for Educational Evaluation (JCSEE), evaluation standards and, 484 Joint productive activity (JPA), 65 Juvenile offenders checklist for practitioners and, 321 evaluation reports and, 318 guidelines for future research and, 320 Indiana Through Mentoring (AIM) and, 316-320 literature review summary and, 319-320 meta-analysis of youth mentoring programs, 319 overview of, 315 peer-reviewed studies and, 316-317 reentry programs and, 317 research and, 316-321 review of literature on, 316-319 theory and, 315-316 transitional coordinators (TCs) and, 317 Juvenile Rehabilitation Administration (JRA), recidivism and, 318 King Odysseus, 68

560   HANDBOOK OF YOUTH MENTORING Latino youth Club Amigas and, 362 CPY program and, 374 cultural competence and, 152 cultural similarity or dissimilarity and, 148-149 family involvement and, 460 Friends of the Children (FOTC) program, 377 gender study and, 164 high school dropout rates and, 304 incarcerated parents and, 370 mentoring roles and, 151 natural mentoring and, 309 studies of sources of support around academic success and, 360-361 Lunch Buddy program, 78, 80, 111, 212, 239-240, 242, 307, 308, 332, 461-462, 464 Major depressive disorder (MDD), 325, 327 Māori youth, cultural competence and, 152 Match Characteristics Questionnaire (MCQ), 56-57 Maxwell School of Citizenship and Public Affairs, program funding and, 391 Mental health needs attention-deficit/hyperactivity disorder (ADHD), 325 behavioral support specialist (BSS) and, 334 Check and Connect Program, 332-333 checklist for practitioners and, 335-336 cognitive-behavioral therapy (CBT), 331-332, 337 coping and, 327-328 destigmatization and, 327 Early Risers program, 333 existing treatments for, 325-326 externalizing problems and, 328 individualized education plan (IEP), 333 internalizing problems and, 327-328 Lunch Buddy program, 332 major depressive disorder (MDD), 325, 327 multidimensional treatment foster care (MTFC), 333-334 multisystemic therapy (MST), 333 oppositional defiant disorder (ODD), 325, 333 PrimeTime program, 332 research and, 329-335 studies of mentoring in nonclinical samples and, 329-330 suicidal thoughts and, 328-329 theory and, 326-329 treatment initiation and, 327 wraparound services and, 333 Youth-Nominated Support Team (YST) and, 330-331

Mentee Support-Seeking scale, mentoring activities and, 75 MENTOR definition of mentoring and, 99 effective mentoring programs for youth and, 91 Elements of Effective Practice and, 153 Elements of Effective Practice Toolkit and, 155 formation of Mentoring Research and Policy Council of, 7 “mentoring gap” and, 9, 34 program evaluation and, 485 Research in Action Series of, 7, 10, 13, 157, 380, 497 “termination ritual” and, 475 web-based mentor training programs and, 451-452 Mentoring, see also youth mentoring as context and, 37-38 attachment perspective on development and, 100-101 attachment processes and, 108-109 attachment theory and, 111, 113 bullied children and, 20-21, 32, 538, 540 Clover Model and, 105-107 cognitive development theories and, 109 definitions of, 292-293 effects and explanatory processes of, 307-308 effects of network relationships on, 132-135, 137-139 evidence-based prevention tools and, 33-40 formal mentoring and, 307-308, 347-348, 361-362 functionalist perspective on development and, 101-102, 113 funding a mentoring program and, 388 global popularity of, 189 inconsistent mentors and, 108 juvenile offenders and, 315-321 “natural mentoring” and, 309, 312, 352, 530 naturally occurring mentors and, 346-347 NRI-conflict scores and, 56 phase models of mentoring and, 470-471 Piaget paradigm and, 105 prevention science and, 29-40 professional mentoring and, 378-379 recommendations for practitioners and, 113 social-cognitive developmental abilities and, 110-111 social-cognitive developmental theory and, 112-113 social-cognitive perspective on development and, 102-104 social information processing (SIP) theory and, 104 social networks and, 129-140

Subject Index  561 statistics and, 34 structuring the mentoring meeting time developmentally and, 109-110 theory and, 100-107 training mentors and, 216, 350-351 trust and, 102, 109 volunteer mentoring and, 376-377 youth with mental health needs and, 326-337, 326 Mentoring activities advocacy and, 65-66, 76-77 authorship and, 69-70, 79-80 authorship as collaboration and, 75-76 checklist for practitioners and, 79-80 client-centered approach and, 67-68 collaboration in activity settings and, 65, 79 Companionship Therapy Study and, 76 developmental vs. instrumental mentoring styles and, 70-71 empowerment and, 76-77 focus and, 68-69, 78 goal-directed matches in schools and, 73 help-intended communications and, 64-65 hierarchy of interventions and, 66-67, 80 importance of balance in interaction and, 71-72 joint productive activity (JPA) and, 65 Mentee Support-Seeking scale and, 75 mutuality and, 77 overview of, 63-64 playful doing vs. serious talking and, 74-75 Problem-Behavior theory and, 68 psychotherapeutic parallels and, 67 purpose and, 78-79 purpose of an interaction and, 69 reciprocity and, 77 relational, goal directed, or balanced and, 73-74 Relationship Quality Scale and, 75 research and, 70-77 Theoretically Evolving Activities in Mentoring (TEAM), 68 Mentoring Alliance Scale (MAS), 51, 56 Mentoring Central, web-based mentor training and, 452 Mentoring Children of Prisoners (MCP) program, 7-8, 370, 372, 374, 376, 381, 447, 539, 546 Mentoring Enhancement Demonstration Program, 8, 10 Mentoring relationship quality (MRQ), 45 administering surveys and, 60 Adult Relationships Scale and, 51 authenticity and, 48 authorship and, 50 basic relational competence and, 47-48 closeness and, 48 connectedness and, 48

Durability/Resilience and, 49 external match quality and, 46, 50 focus and, 50 guidelines and, 58-60 instrumental compatibility and, 49 instrumental match quality and, 49 internal match quality and, 46 interpreting and using results, 60 intimacy and, 48 logistics and, 50 longevity and, 48 match duration and, 48-49, 51, 57 match quality and, 48-49 match structure and, 46, 49-50 meeting frequency and, 48 Mentoring Alliance Scale (MAS) and, 51, 56 Mentor-Youth Alliance Scale (MYAS) and, 51 Network of Relationships Inventory (NRI) and, 56 objective indicators and, 48-50 overview of, 45, 47 parents or guardians and, 50 received support and, 49 recommendations for future research and, 58 relational bond and, 51 Relational Health Indices (RHI) and, 58 relational match quality and, 46-49 relational satisfaction and, 48 research and, 51, 56-58 risk-related relational competence and, 47-48 Strength of Relationship Measure (SoR) and, 57 subjective indicators and, 46-50 theory and, 45-50 trust and, 48 working alliances and, 48 Youth-Mentor Relationship Questionnaire (YMRQ) and, 51 Mentoring Works!, Facilitator’s Guide and, 248, 251 MentorNet, e-mentoring and, 260, 267 Mentor recruitment attachment styles and, 399 checklist for practitioners and, 406 motivation of mentors and, 402 motivations for volunteering and, 399-400 overview of, 397 recommendations for communities and, 406 recommendations for organizations and, 405-406 recommendations for potential mentors and, 404-405 research and, 401-404 retention and, 401-404 skills and abilities of mentors and, 398-399, 401-402

562   HANDBOOK OF YOUTH MENTORING statistics and, 400, 402-403 strategies for, 400-401 theory and, 397-401 Mentor relationship quality (MRQ), academically at-risk student (AARS) and, 306-307 Mentor screening application process and, 421-422 checklist for practitioners and, 423 Child Abuse Potential Inventory (CAP), 414-415 child sexual abuse and, 412-413, 417 general principles for, 419-420 high turnover rates and, 420 maintaining an applicant database and, 420 mentee training and, 421 mentor recruitment and, 421 National Center for Missing and Exploited Children, 416 National Child Protection Act/ Volunteers for Children Act, 411, 417, 422 postmatch screening and, 422-423 program design and, 420-421 recommendations by manuals, 417, 419 research and, 414-416 specific pecommended practices and, 420 theory and, 412-414 Mentor training building a positive mentoring relationship and, 447-448 Building the Foundation web-based mentor training and, 446 Children with Adolescent Mentors, 451 closing the mentoring relationship and, 448 Cross-Age Mentoring Program (CAMP), 451 Elements of Effective Practice for Mentoring (EEPM), 450, 452 ethical issues and, 448 expectations for mentoring relationship and, 445 first meeting with mentee and, 445 guiding questions for practitioners and, 453 having fun and, 447 in-person training and, 450-451 mentor-mentee relationship quality and, 440-441 motivations of mentors and, 444-445 need for mentor training programs and, 440 overview of, 439 population-specific content and, 447 relationship length and, 442-443 relationship quality and, 442-443 research and, 442-450 role clarity and, 445-446 successful mentoring relationships and, 446-447

theory and, 440-442 training content and, 444 training format and, 448-449 training length and, 443 training manuals and, 451 training prevalence and, 442 training timing and, 443-444 web-based mentor training programs and, 451-452 Mentor-Youth Alliance Scale (MYAS), 51 Mexican American youth mentor behaviors regarding the promotion of education and career, 166 nonfamily adult mentors and, 361 Mexican youth, high rates of violence against and, 355 Meyer, Graig, 439 Minnesota Multi-Phase Inventory, mentor screening and, 415 MIRES program mentor training and supervision, 311 Multidimensional treatment foster care (MTFC), 333-334, 333 Multisystemic therapy (MST), mental health needs and, 333 National 4-H Council, 17, 20-22, 25, 293 National Agenda for Action, 7, 42 National Association of Social Workers (NASW), mentor training and, 448 National Center for Missing and Exploited Children, mentor screening and, 416, 514 National Child Protection Act/ Volunteers for Children Act, 417, 422 National Guard Youth Challenge program, 297 National Institute of Mental Health, 30, 86 National Institutes of Health, 439 National Longitudinal Study of Adolescent Health families with more resources study, 291-292 mentoring assessment from, 528 naturally occurring mentors and, 346 National Longitudinal Survey of Youth, linking truancy to long-term earnings and, 505 National Mentoring Center, program evaluation and, 485 National Mentoring Partnership Elements of Effective Practice for Mentoring and, 182-183 mentor recruitment and, 402 National Mentoring Summit, 7, 9, 13 National Science Foundation, 17, 271 National Service-Learning Clearinghouse, 293 Native American youth CPY program and, 376 juvenile offenders and, 318

Subject Index  563 Network of Relationships Inventory (NRI), 56 New Zealand Youth Mentoring Network, 8 No Child Left Behind legislation, 35 Nonprofits, program funding and, 386-387, 393 Obama, Barack budget cuts and, 40 cross-age peer mentoring programs and, 234 Office of Juvenile Justice and Delinquency Prevention (OJJDP), xii, 8, 10, 315, 318, 323, 401, 403, 416, 535, 542, 545, 547 Oppositional defiant disorder (ODD), 325, 333 Orphan Foundation of America (OFA), 350 Oxytocin, 118, 120, 122-123, 126 Palo Alto School District Peer Counseling Program, 234-235 Parental acceptance-rejection (PAR) theory, 344 Parent-Child Relationship Inventory (PCRI), 460 Pasifika youth, cultural competence and, 152 Peer Buddy Program, 246-247, 251 Peer Connection Program, 246 Peer Group Connection, 238, 242 Phoenix Youth At-Risk, 183 Poland, high school dropout rates and, 303 Portugal, high school dropout rates and, 303 Positive youth development (PYD) absence of an accepted vocabulary for, 18 checklist for practitioners and, 23-24 deficit model of youth and, 18 definition of, 17 Five Cs of, 19-21 implications for future PYD-Mentoring research and, 23 necessity of, 17-19 promoting PYD-Mentoring linkage and, 21-23 “sixth C” and, 20, 24 strength-based vision for America’s young people and, 19 youth-context alignment and, 20-21 Powell, Colin, America’s Promise and, 234, 525 Prevention science core tenets of, 31-33 dynamic developmental processes and, 32 efficacy trials and, 31 equifinality and, 33 evidence-based prevention tools and, 33-40 fixed markers and, 33 “generic” risk and protective factors and, 33 key terms and concepts in, 30-31 knowledge of fundamental causal processes and, 31-32 mentoring relationships and, 35-36 moderator studies and, 31

multifinality and, 30, 33 overview of, 29-30 population at risk and, 32 practice of youth mentoring and, 38-40 pre-intervention research and, 30-31 prevention research cycle and, 30-31 prevention trials and, 33 risk factors and, 30, 32-33 Type 1 translational research, 38 Type 2 translational research, 38 PrimeTime program, 307-308, 461-462, 464 Princeton Center for Leadership Training, Peer Connection Program and, 246 Probability matching theory, youth matching and, 429 Program evaluation adherence and, 486 CONSORT and, 495 data analysis and, 493-494 dosage and, 489 evaluation standards and, 484 exposure and quality of delivery and, 486-487 linking process and outcome components of program evaluation, 494-495 measurement and, 492-493 outcome evaluation and, 489-490 overview of, 481 process evaluation and, 485-486 program differentiation and, 488-489 recommendations for good practice in, 482-483 responsiveness and, 488 study design and sampling, 490-491 study design and sampling, 490-492 two-staged least squares (TSLS) and propensity score analysis, 495 utilization of evaluation findings and, 495 Program funding accepting rejection and, 394 American Association of Fundraising Counsel and Giving Institute, 391 board of directors and, 386-387 Chamber of Commerce and, 390 checklist for practitioners and, 394 corporate giving and, 389-390 creating a fundraising plan and, 387-388 direct mail and, 392-393 evidence of program quality and, 393 fees for program and product services, 393 government funding and, 389 implementing fundraising plans and, 385-388 individual giving and, 391-392 key sources of, 388-394 local initiatives and, 393 matching gift programs and, 390 nonprofit status and, 386-387, 393

564   HANDBOOK OF YOUTH MENTORING overview of, 385 planned giving and, 392 public and private foundations, 390-391 thanking donors and, 393-394 third-party events and, 392 United Way of America, 393 Project START, 461 PROTECT ACT Criminal Background Check Demonstration Project, 9, 514 Psychotherapy, closure of mentoring relationships and, 471-472 PsycINFO, 148, 163 Public/Private Ventures (P/PV) study BBBS SBM program and, 208-209 child abuse and, 348 children of incarcerated parents and, 370 evaluation of BBBSA school-based mentoring program and, 493-495 impact study and, 34-35, 38, 70, 209-213, 490-495, 541 program evaluation and, 490 Public Relations Society of America (PRSA), e-mentoring and, 260-261 Quantum Opportunity Program (QOP), 307, 310 Race cultural competence and, 148, 151-152, 155 cultural similarity or dissimilarity and, 146-149, 152-154 ethnic identity and, 150-151, 155 incarcerated parents and, 370 limitation of studies of, 146 oppression and, 147, 150, 154-155 racial and ethnic identity, 147-148 research and, 148-153 Randomized controlled trial (RCT) Big Brothers Big Sisters (BBBS), 329 integrative and meta-analytic views of, 210-211 multisystemic therapy (MST), 333 SBM program model and, 208 size of SBM program impacts and, 213 Redefinition, youth mentoring and, 84 Refugee youth, see also Immigrant youth acculturation and resettlement process, 364 acculturation theory and, 357-359 bicultural adjustment and, 359 Bridging Refugee Youth and Children’s Services (BRYCS), 363-364 checklist for practitioners and, 365 conceptual model of application of acculturation theory and, 357 ecological theory and, 358

“family mentoring” and, 364-365 family acculturation gaps and, 359 family-focused mentoring relationship and, 362 Federal Office of Refugee Resettlement and, 363 formal mentoring programs and, 361-362 Go-Betweener Mentoring Program, 363 mentoring goals and, 359 mentoring structures and, 360 overview of, 355 research and, 360-363 statistics and, 355 theory and, 356-360 Regulatory focus theory, 190-191, 193-197 Relational-cultural theory (RCT), 160 Relational Health Indices (RHI), 58 Relationship Quality Scale, mentoring activities and, 75 Research academically at-risk student (AARS) and, 306-309 after-school programs and, 224-229 attunement research and, 122 closure of mentoring relationships and, 472-474 cross-age peer mentoring and, 238-245 developmental perspective and, 107-111 e-mentoring and, 265-268 Erikson’s functionalist approach and, 109 family involvement and, 460-463 foster care and, 345-349 group mentoring and, 276-278, 283-284 incarcerated parents and, 372-373 international youth mentoring and, 192-195 juvenile offenders and, 316-321 mental health needs and, 329-335 mentoring activities and, 70-77 mentoring relationship quality (MRQ) and, 51, 56-58 mentor recruitment and, 401-404 mentor screening and, 414-416 mentor training and, 442-450 psychobiology and youth mentoring, 124 race and, 148-153 refugee youth and, 360-363 role of social class in mentoring relationships and, 180-182 school-based mentoring (SBM) and, 87-88, 207-213 service-learning and, 294-297 social-cognitive developmental abilities and, 110-111 social networks and, 134-137 youth matching and, 429-433

Subject Index  565 Rochester Resilience Project, 75 Rwanda, effectiveness of a mentoring intervention and, 194 SafetyNET program, 7, 9, 12, 416-417, 419, 425 San Antonio Mentoring Program and, 209-210 School Transition Environment Project (STEP), group mentoring and, 277 Second Life, e-mentoring and, 260 Self-determination theory, 189-191, 193-197, 305 Service-learning, see also Work balance of power and, 294-295 cognitive transformation and, 295 connections across settings and, 295 Current Population Survey and, 292 goal-directed activity and, 294 high-intensity work and, 295 learner engagement and, 295 overview of, 291 personal and social competencies, 293-294 practice questions and recommendations for, 297-298 project-based learning and, 296 research summary and recommendations, 296 supporting mentoring and, 297 teaching and mentoring at work, 295-296 theory and, 293-294 working at challenging jobs and, 294 SimTeach, e-mentoring and, 260 SMILE study, youth characteristics and, 211 Social change “oughts” for, 18-19 talking about positive attributes of young people and, 18 Social class cultural definition of, 176-177 cultural differences in models of agency and, 176 environmental definition of, 176 intergroup contact theory (ICT) and, 178 material definition of, 176 measurement of, 177-178 mentor-mentee relationship and, 181-182 model of youth mentoring and, 178-179 outcomes and, 180-181 overview of, 175 perspectives on, 175-176 research and, 180-182 theory and, 175-180 youth identity and, 177 Social control theory, delinquency and, 34 Social exchange theory, foster care and, 344 Social information processing (SIP) theory, 104 Social networks checklist for practitioners and, 137

concepts and principles of, 131 effects of network relationships on mentoring, 132-136, 138-139 effects of network structure on mentoring and, 131-134, 137-139 future directions of, 136-137 hypothetical mentor-mentee social networks and, 130 Latino youth and, 134 research and, 134-137 size of youth networks and, 137 theory and, 129-134 Social Support Behavior Code, FtF interactions and, 264 Society for Prevention Research, Standards of Evidence and, 495 Socioeconomic status (SES) cultural aspects of class and, 182 development of mutual understanding and respect for, 185 effects of mentoring and, 178 lower SES backgrounds and, 177 measurement of, 180 mentoring programs and, 175 South Africa, mentoring as an intervention for youth and, 194 STAR (Stop and take a deep breath), 88 Statistics Boys & Girls Clubs of America and, 5 child abuse and, 341-342 child sexual abuse and, 412-413, 417 college enrollment and, 292 Current Population Survey and, 292 effect sizes of after-school programs and, 210-211 focus on relational activities and, 74 foster care and, 341, 346 government program funding and, 389 graduation rate and, 242 health care in the United States and, 29 high school dropout rates and, 303 incarcerated parents and, 369-370 Indiana Through Mentoring (AIM) program, 319 international affiliates of BBBSI and, 10 juvenile murder arrest rates and, 315 mental health needs of youth and, 325-326 “mentoring gap” and, 6 mentor recruitment and, 400-403 mentor training and, 442, 450 P/PV study of the BBBS SBM program and, 213 program funding and, 386, 391 “proven” effects of youth mentoring and, 34 refugee youth and, 355

566   HANDBOOK OF YOUTH MENTORING school-based mentoring (SBM) and, 203 teachers as attachment figures and, 277 West Indian and Haitian adolescents, 147 working with youth and, 273 youth matching and, 430 youth mentoring and, 526-527 STEM (science, technology, engineering, and math) program, 182 Strategic Enhancement to Mentoring Programs, 8, 10 Strength of Relationship Measure (SoR), 57 Structured program activities with older peer mentors, connecting youth to employment and, 136 Student Mentoring Program initiative, 5, 7-8, 34-35, 208, 489 Student-Teacher On-line Mentoring Project (S.T.O.M.P.), 261 Study of Mentoring in the Learning Environment (SMILE), 7, 73, 209-211, 402-403, 535 Suicidal thoughts, mental health needs and, 328-329 Summer Institute on Youth Mentoring, ix, 7, 10, 13, 538, 542-543 Sweden, mentoring programs and, 193 Syracuse University, program funding and, 391

Thrive Foundation for Youth, 17 Trojan War, 68 Tutoring school-based mentoring (SBM) and, 210 Type 1 translational research and, 38

Take Stock in Children, 78 Teen Mentoring Initiative, 251 Teen Trendsetters, peer mentoring and, 238 Theoretically Evolving Activities in Mentoring (TEAM), 68, 70-72, 76, 78 Theory academically at-risk student (AARS) and, 304-306 after-school programs and, 221-224 biological perspective and, 118-124 closure of mentoring relationships and, 470-472 cross-age peer mentoring and, 236-238 e-mentoring and, 261-265 family involvement and, 458-459 foster care and, 342-345 group mentoring and, 273-276 incarcerated parents and, 371-372 mental health needs and, 326-329 mentor recruitment and, 397-401 mentor screening and, 412-414 mentor training and, 440-442 school-based mentoring (SBM) and, 204-207 service-learning and, 293-294 social-cognitive developmental theory and, 206-207 social expectancy theory and, 205-206 social networks and, 129-134 youth matching and, 427-429

VISIONS program, foster care and, 348 “Voice and choice” framework, youth matching and, 428-429

Uganda, mentoring as an intervention for youth and, 194 United Way of America, program funding and, 393 University of Washington, DO-IT program, 262 U.S. Department of Education child sexual abuse in school settings and, 413 funding mentoring programs and, 10 Student Mentoring Program initiative and, 5, 7-8, 34, 208-209, 489 U.S. Department of Education’s Office of Safe and Drug-Free Schools, mentor recruitment and, 405 U.S. Department of Education’s Student Mentoring Program, formal mentoring programs and, 489 U.S. Department of Health and Human Services (DHHS), children with incarcerated parents and, 10, 370 U.S. Office of Juvenile Justice and Delinquency Prevention, compensating mentors and, 401

Wallace Foundation, 221 Washington State Institute for Public Policy, recidivism and, 318, 320 White youth attitudes toward mentoring and, 345 cultural similarity or dissimilarity and, 149 FLEX program and, 461 Friends of the Children (FOTC) program, 377 incarcerated parents and, 370 William T. Grant Foundation, 221 Work, see also Service-learning balance of power and, 294-295 cognitive transformation and, 295 connections across settings and, 295 Current Population Survey and, 292 goal-directed activity and, 294 high-intensity work and, 295 learner engagement and, 295 overview of, 291 personal and social competencies, 293-294 practice questions and recommendations for, 297-298 project-based learning and, 296 research summary and recommendations, 296

Subject Index  567 supporting mentoring and, 297 teaching and mentoring at work, 295-296 theory and, 293-294 working at challenging jobs and, 294 Yavapai Big Brothers Big Sisters, 45 YMCA, child sexual abuse and, 413 Young Women Leaders Program, 183 Young Women’s Leadership Project, group mentoring and, 283 Youth Advocate Programs, Inc., 77 Youth-context alignment, 20-21 Youth Development Program, group mentoring and, 278, 286 YouthFriends, 87 Youth Helping Youth (Myrick & Erney), 247 Youth Initiated Mentoring, 297 YouthLaunch ABCs of Teen Mentoring and, 247 Teen Mentoring Initiative, 251 Youth matching attachment theory and, 428 ChalleNGe mentoring program, 431 closing the research-practice gap and, 436 guiding questions for practitioners and, 435 meta-analyses of mentoring programs and, 431-432 overview of, 427 probability matching theory and, 429 recommendations for practitioners and, 434 research and, 429-433 single-study findings and individual similarities among matches, 432-433 theory and, 427-429 “voice and choice” framework and, 428-429 Youth mentoring, see also mentoring assessing mentoring relationship quality and, 52-55 Big Brothers Big Sisters of America (BBBSA) and, 6 checklist for practitioners and, 39, 59 community-based mentoring and, 89-90 connecting research and practice of, 9-10 contemplation and, 84 contemporary trends in, 10-11 current status of the quantity and quality of, 525-529 decline and dissolution, 84 defining and, 35-36 funding of, 10 goal attainment scaling (GAS) and, 85-86 goal setting and, 83-96

growth and maintenance, 84 inequities in access and, 526-527 initiation and, 84 international progress and, 10 “mentoring gap” and, 6 meta-analysis of program effectiveness and, 90 multilevel conceptualization of, 4-6 overview of, 525 point estimates and, 525-526 practice and policy of, 9, 38-40 public perceptions of, 6 quality and effectiveness of, 527-529 quality of youth mentoring relationships and, 84 recent developments in, 6, 8-10 recent developments in, 7-8 redefinition and, 84 research and, 6, 8-9, 38, 87-91 role of goal setting and, 84-85 role of race, ethnicity, and culture in, 145-156 strategies to improve quantity and quality of, 529-531 trends over time and, 527, 529 variability and, 529 Youth mentoring practice academically at-risk student (AARS) and, 309-312 after-school programs and, 229-230 closure of mentoring relationships and, 474-477 cross-age peer mentoring and, 245-254 family involvement and, 463-464 foster care and, 349-353 group mentoring and, 284-286 immigrant and refugee youth, 363-366 incarcerated parents and, 374-380 international youth mentoring and, 195-197 juvenile offenders and, 321 key challenges of e-mentoring and, 268-269 mental health needs and, 335-337 mentor recruitment and, 404-406 mentor screening and, 416-423 mentor training and, 450-453 school-based mentoring (SBM) and, 213-217 social class and, 182-185 work and service-learning, 296-297 youth matching and, 434-436 Youth Mentoring Survey (YMS), 56-57 Youth-Mentor Relationship Questionnaire (YMRQ), 51 Youth-Nominated Support Team (YST), mental health needs and, 330-331 Youth Research Center, 235

ABOUT THE EDITORS

David L. DuBois, PhD, is a professor in community health sciences within the School of Public Health at the University of Illinois at Chicago. He received his doctorate in clinical-community psychology from the University of Illinois at Urbana-Champaign. His research examines the contribution of protective factors, particularly self-esteem and mentoring relationships, to resilience and holistic positive development and on translating knowledge in this area to the design of effective youth programs. He has authored numerous peer-reviewed studies on these topics, including two widely cited meta-analytic reviews of the literature on the effectiveness of youth mentoring programs. Dr. DuBois was lead coeditor of the first edition of the Handbook of Youth Mentoring (Sage, 2005) and is coauthor of After-School Centers and Youth Development: Case Studies of Success and Failure (Cambridge University Press, 2012), each of which received Social Policy book awards from the Society of Research on Adolescence. His research has received funding from several federal sources, including the National Institutes of Health, Office of Juvenile Justice and Delinquency Prevention (OJJDP), Office of Minority Health, and the Institute of Education Sciences. He is a fellow of the American Psychological Association and Society for Community Research and Action and is a past distinguished fellow of the William T. Grant Foundation. He consults widely to mentoring programs nationally and internationally, is a member of the Research and Policy Council of MENTOR/ National Mentoring Partnership and the Research Advisory Council of Big Brothers Big Sisters of America (BBBSA), and is the founder and moderator of a listserv on youth mentoring research and practice that currently has over 600 members. Dr. DuBois also has been a mentor himself in a Big Brothers Big Sisters program. His current research includes an OJJDP-funded randomized controlled evaluation of youth outcomes associated with infusion

of the Step-It-Up-2-Thrive model of positive youth development into the BBBSA community-based mentoring program. Michael J. Karcher, EdD, PhD, is a professor of counseling at the University of Texas at San Antonio, where he coordinates the School Counselor Training Program. He received doctorates in human development and psychology from Harvard University (1997) and in counseling psychology from the University of Texas at Austin (1999). He conducts research on school-based and cross-age peer mentoring as well as on adolescent connectedness and pair counseling. He authored the Cross-Age Mentoring Program (CAMP), which includes a set of materials (manual, peer mentor handbook, activity curriculum, training activities, and evaluation guide) that are currently the subject of two effectiveness tests. Based on his research on cross-age peer mentoring, Dr. Karcher created an online training for teen mentors (for Big Brothers Big Sisters), available at www.highschool bigs.org. Dr. Karcher conducted one of the first largescale school-based mentoring studies, the Study of Mentoring in the Learning Environment (SMILE, 2003–2006), funded by the William T. Grant Foundation. With support from the Office of Juvenile Justice and Delinquency Prevention, he is currently investigating the role of advocacy in youth mentoring with delinquent youth, and is testing the TEAM framework of mentoring interactions presented in Play, Talk, Learn: Promising Practices in Youth Mentoring (Wiley, 2010), which he coedited with Michael Nakkula. With David L. DuBois, Professor Karcher coedited the Handbook of Youth Mentoring (Sage, 2005), which received a Social Policy book award from the Society of Research on Adolescence. He is on the editorial board for several national journals and the research and advisory boards of Big Brothers Big Sisters of America and MENTOR/ National Mentoring Partnership.

569

ABOUT THE CONTRIBUTORS

Kym R. Ahrens, MD, MPH, is an assistant professor and associate fellowship director in adolescent medicine. She holds a joint appointment in the Department of Pediatrics at the University of Washington and Seattle Children’s Hospital and Research Institute. Her research focuses on the development of resilience-oriented strategies to improve adult health and other outcomes for youth who have been exposed to early adversity. She has conducted both qualitative and quantitative research on the topic of natural mentoring relationships for youth in foster care. Amber L. Allison, MS, is a doctoral candidate at the University of New Orleans in the biopsychology program. She works with Dr. Elizabeth Shirtcliff in the Stress Physiology Laboratory. Her primary research interests include the physiological correlates of bonding, social support, affiliation, and romantic relationships. She is particularly interested in how our biology facilitates and maintains our connections with other people, and how those connections then shape our physiological functioning. She has utilized indices of the autonomic nervous system (heart rate, respiration) and the endocrine system (cortisol, oxytocin) as windows into how relationships get under the skin. After receiving her PhD, she hopes to continue on an academic trajectory and to continue researching the complexities of environment-biology interactions. Fabricio E. Balcazar, PhD, is a professor in the Department of Disability and Human Development at the University of Illinois at Chicago. Dr. Balcazar’s primary research interest is in developing effective strategies for enhancing consumer empowerment and personal effectiveness among individuals with disabilities. Dr. Balcazar is currently the director of the Center on Capacity Building for Minorities with Disabilities Research. As such, he has led the development of a cultural competence conceptual framework, training curriculum, and assessment instrument, conducting multiple workshops on this

topic and providing technical assistance to multiple organizations. Dr. Balcazar has published over 70 peer-reviewed journal articles and recently coedited a book entitled Race, Culture and Disability: Issues in Rehabilitation Research and Practice. Dr. Balcazar is a fellow of the American Psychological Association and the Midwestern Psychological Association. Luciano Berardi, PhD, is the associate director of the McNair Scholars Program and Research Coordination at the Center for Access and Attainment at DePaul University in Chicago. He received a degree in clinical psychology from Universidad de Belgrano, in Argentina. He worked on the development, implementation, and evaluation of communitybased mentoring programs at the District of Columbia Mental Health Department in Washington, D.C. Dr. Berardi has worked on several research projects at DePaul University and at the Center for Capacity Building for Minorities With Disability Research at the University of Illinois at Chicago. He earned his PhD in community psychology from DePaul University. His research interests focus on the role of natural mentors in youths’ and young adults’ academic attainment, social mobility, and well-being. Overall, his work is centered on fostering academic success and improving the educational environments of underrepresented students. Joseph Bergen is program director for the Portland (Oregon) chapter of the Friends of the Children youth mentoring program. Mr. Bergen has been an employee with Friends for 9 years and served as a professional mentor to eight boys for 5 years. His research interests focus on the identification of elements of effective mentoring relationships for highest-risk youth so that such youth have the support necessary to thrive during adolescence and adulthood. Dina Birman, PhD, is an associate professor of psychology at the University of Illinois at Chicago. She received her doctorate in clinical/community 571

572   HANDBOOK OF YOUTH MENTORING psychology from the University of Maryland. She has conducted studies of acculturation and adaptation of refugees and immigrants from the former Soviet Union, Vietnam, Central America, and Somalia. Her research has contributed to conceptualization and measurement of acculturation, family acculturation gaps, and understanding multiple cultural, ethnic, and religious identities of refugees in resettlement. She has studied mental health interventions for immigrants and refugees, and the process of implementation and adaptation of evidence-based practices within such programs. She is a fellow of the Society for Community Research and Action and the American Psychological Association (APA), and in 2010 was appointed to serve on the APA Presidential Task Force on Immigration. Jennifer E. Blakeslee, PhD, received her doctorate in social work and social research from Portland State University, where she also worked at the Center for Interdisciplinary Mentoring Research with Tom Keller. Her research has focused on aspects of youth mentoring and social support networks, particularly as these relate to the transition from foster care and into young adulthood. Recent publications include an article analyzing the youth mentoring literature and the network of coauthors publishing in the field of youth mentoring as well as a conceptual article on applying the social network perspective to expand the scope of inquiry into the formal and informal social support available to youth as they transition from foster care. G. Anne Bogat, PhD, is professor of psychology at Michigan State University. She received her doctorate in 1982 in clinical/community psychology from DePaul University in Chicago. Her research interests include mentoring, social support, and family violence. In 2002, she coedited, with Jean Rhodes, a special issue of American Journal of Community Psychology on youth mentoring. Her recent work focuses on the longitudinal assessment of risk and resilience factors for women and children living in households with intimate partner violence. Her work has appeared in journals related to community psychology, intimate partner violence, and child development. Michelle J. Boyd, PhD, completed her doctoral work at Tufts University and is currently an SRCD congressional fellow in Washington, D.C. Her research focuses on media literacy and active and engaged citizenship among youth. Preston A. Britner, PhD, is a professor of human development and family studies at the University of Connecticut. He is editor emeritus of the Journal of

Primary Prevention and holds several editorial appointments and administrative positions. His research covers the areas of youth mentoring (in particular among special populations), attachmentcaregiving relationships, child welfare and child maltreatment prevention, law and social policy, and family-focused prevention programs. Dr. Britner is a fellow of the American Psychological Association (APA) and served on APA’s Committee on Children, Youth, and Families. His latest book (coauthored with Martin Bloom) presents new models of clientcentered evaluation for the helping professions. Kristina S. Callina, MA, is in her final year of doctoral studies at the Eliot-Pearson Department of Child Development at Tufts University. Ms. Callina is currently a doctoral research assistant at the Institute for Applied Research in Youth Development, and her research focuses on the role of hope about the future in promoting positive youth development. Timothy A. Cavell, PhD, is professor and director of clinical training in the Department of Psychology at the University of Arkansas. Cavell’s work has focused on parent- and mentor-based interventions for children who are highly aggressive or chronically bullied and thus at risk for later delinquency, substance abuse, or psychopathology. His research has been funded by the National Institute on Drug Abuse, the Health Resources & Services Administration, the Verizon Foundation, and Big Brothers Big Sisters of Canada. He is author of over 50 journal articles and chapters as well as two books: Working With Parents of Aggressive Children: A Practitioner’s Guide and Anger, Aggression, and Interventions for Interpersonal Violence. His recent work has focused on the integration of youth mentoring and prevention science and, more specifically, on short-term, lunchtime mentoring as a school-based intervention for chronically bullied children. Jennifer Cearley, PhD, is a research associate at the Oregon Social Learning Center in Eugene. She is a National Institutes of Health–sponsored postdoctoral fellow with the Child Study, a randomized controlled trial of the Friends of the Children mentoring program. Dr. Cearley has worked as a researcher with the Parent Child Study, which evaluated impacts of a parent management training program delivered inside the Oregon state prison system, and with the Linking the Interests of Families and Teachers (LIFT) study, which evaluated the long-term impact of a school-based preventive intervention program. Her other research interests include the adjustment and social contexts of antisocial youth, particularly girls. She has

About the Contributors   573 worked as a practitioner within the juvenile justice system in Oregon for the past decade. Wing Yi Chan, PhD, is an assistant professor of psychology at Georgia State University. She received her doctorate in community psychology from the University of Illinois at Chicago in 2010. She then served as a National Institute of Mental Health postdoctoral fellow at the University of Minnesota from 2010 to 2011. Her research focuses on promoting positive youth development among immigrant and refugee youth. Her most recent research examines how civic and political participation can prevent problem behaviors and promote successful transition to adulthood for this population of youth. For example, her recent projects investigate the impact of service-learning programs on academic adjustment of immigrant college students; and the role of religion in promoting civic engagement among Korean immigrants. Her work also addresses the development, implementation, and evaluation of after-school programs designed to promote successful adaptation for immigrant children and adolescents. E. Gil Clary, PhD, is assistant vice provost for assessment at Kutztown University of Pennsylvania. He is the author of many empirical journal articles and book chapters about helping behavior, including many focused on the functional approach to volunteerism, which he originated with Mark Snyder. One product of this collaboration, a scale investigating volunteers’ motivations called the Volunteer Functions Inventory, has been used extensively by organizations and researchers. With Dr. Jean E. Rhodes, Dr. Clary edited the book Mobilizing Adults for Positive Youth Development: Strategies for Closing the Gap between Beliefs and Behaviors. He has been actively involved with the Association for Research on Nonprofit Organizations and Voluntary Action and belongs to numerous other scholarly organizations. Yarí Colón-Torres, PhD, is a staff psychologist at the VA Caribbean Healthcare system in San Juan, Puerto Rico. She received her doctorate in clinical/ community psychology from DePaul University in Chicago. Currently, her clinical and research interests include culturally informed evidenced-based practices, diversity, and behavioral health. She codeveloped a postdoctoral fellowship focusing on women veterans. She mentors postdoctoral fellows and interns and teaches a seminar on diversity issues. Becky Cooper has bachelor’s and master’s degrees in psychology and education from Stanford University. She serves as executive director of

Friends for Youth, Inc., which has been cited as a program of excellence by the John W. Gardner Center at Stanford and the U.S. Department of Education. Ms. Cooper gave the keynote address on mentoring at Singapore’s National Youth Council Conference, served on the California Governor’s Mentoring Partnership Quality Assurance Standards Committee, was a guest at a White House Mentoring Ceremony, was inducted into the San Mateo County Women’s Hall of Fame, and served on the national Mentoring Children of Prisoners Support Center Advisory Board. She coauthored Running a Safe and Effective Mentoring Program and SAFE (Screening Applicants for Effectiveness): Guidelines to Prevent Child Molestation in Mentoring and Youth-Serving Organizations. She also participated in the Kettering Institute’s research on out-ofschool learning and contributed to the Institute’s book Community Educators. Nancy L. Deutsch, PhD, is associate professor at the University of Virginia’s Curry School of Education. She received her doctorate in human development and social policy from Northwestern University. Her research examines after-school settings as contexts for positive youth development, focusing on adult-youth relationships. Her book Pride in the Projects: Teens Building Identities in Urban Contexts reports on 4 years of fieldwork at an urban after-school program. Her recent book, After-School Centers and Youth Development: Case Studies of Success and Failure, with Drs. Barton Hirsch and David L. DuBois (Cambridge University Press, 2011; winner of a Society for Research on Adolescence book award), examines how and why youth thrive (or not) in three centers of differing quality. Dr. Deutsch won the American Education Association’s Out-of-School Time SIG’s emerging scholar award in 2009. She is the program director for methodology at Youth-NEX, the UVA Center to Promote Effective Youth Development. Nicole Duffy, MA, is a research assistant and doctoral student in counseling psychology at Boston College. In addition to studying and promoting youth mentoring relationships, her interests also include innovative community-based mental health interventions and resilience and coping in individuals and communities. J. Mark Eddy, PhD, is director of research at Partners for Our Children in the School of Social Work at the University of Washington in Seattle. A key part of his work is developing and testing evidence-informed interventions designed to prevent aggression and related problems in children and

574   HANDBOOK OF YOUTH MENTORING families. He is the principal investigator on a multisite randomized controlled trial (RCT) of the Friends of the Children youth mentoring program as well as RCTs of several other psychosocial preventive intervention programs for children and families. For the past decade, Dr. Eddy has worked closely with the Oregon Department of Corrections and nonprofit service delivery agencies on the design and testing of a multisystemic program for incarcerated parents and their children and families. He serves on a number of commissions and boards relevant to children and families, including the Oregon Criminal Justice Commission. L. Christian Elledge, PhD, is a postdoctoral fellow in the Clinical Child Psychology Program at the University of Kansas. He is the recipient of the Ruth L. Kirschstein National Research Service award through the National Institute of Child Health and Human Development. Dr. Elledge has recently accepted a position as an assistant professor in the Department of Psychology at the University of Tennessee. His program of research draws from the science of risk and resilience to develop and evaluate effective interventions for children at risk for social-emotional dysfunction. He has particular interest in studying children whose contributions to or experience in antagonistic interpersonal relationships confer significant developmental risk. His recent research has focused on childhood aggression, with a particular emphasis on developing and evaluating school-based interventions for aggressive and bullied youth. Currently, he is conducting a randomized controlled trial testing the efficacy of a school-based mentoring intervention for aggressive children. Rachel Feuer, MA, is a doctoral candidate in clinical/community psychology at DePaul University in Chicago. She completed her undergraduate work at Northwestern University, where she founded a small mentoring program in which college students served as mentors to youth at a social service agency. Ms. Feuer has also consulted with a Chicagobased nonprofit organization to help them improve their mentoring programming. Her dissertation research explores the role of natural mentoring relationships and life stressors in the coping and academic outcomes of urban, low-income Latino youth. E. Michael Foster, PhD, is professor of health care organization and policy and of biostatistics in the School of Public Health at the University of Alabama at Birmingham. His research focuses on the causes and consequences of emotional and behavioral problems among children and youth.

Limor Goldner, PhD, teaches at the Graduate School of Creative Arts in the University of Haifa and at Oranim College in Israel. She is interested in adult-child close relationships, including mentoring relationships and in children’s drawings as a manifestation of their adjustment. She is also the manager of training at the Israeli nationwide mentoring and tutoring project (Perach). Sarah R. Guidone, is a project consultant within the area of organizational development for the Chicago Transit Authority. Ms. Guidone has been working with multisystem partnerships on program, organization, and workforce development for over 7 years. She has partnered with government agencies, nonprofit organizations, education institutions, and federal funders on collaborative work to impact services for young people and communities. Her focus in the youth work field includes juvenile justice, youth program assessment and evaluation, community-based mentoring, and municipal government programming. Ms. Guidone holds a bachelor’s degree in social work from Loyola University Chicago. Mary Agnes Hamilton, PhD, is a senior research associate and director of the Cornell Youth in Society Program in the Bronfenbrenner Center for Translational Research at Cornell University. She received her doctorate in human service studies at Cornell University. Her research goals are focused on understanding and promoting social inventions that improve institutional supports enabling lowincome and minority youth to become workers, citizens, and family members as adults. Her ethnographic research focuses on the quality of these opportunities, especially nonrelated adult mentoring relationships, and how quality affects the technical, social, and personal competence of youth. A recent action research project, Abriendo Caminos: Jóvenes en América Latina, explored ways in which four programs in Argentina, Mexico, and Colombia are creating new institutions that support the transition to adulthood. Stephen F. Hamilton, EdD, is professor of human development at Cornell University and associate director for youth development of the Bronfen­ brenner Center for Translational Research. His research on adolescent development and education investigates the interaction of school, community, and work during the transition to adulthood. As a Fulbright senior research fellow, he spent a year studying Germany’s apprenticeship system and then wrote Apprenticeship for Adulthood: Preparing Youth for the Future, which helped guide the

About the Contributors   575 School-to-Work Opportunities Act of 1994. With Mary Agnes Hamilton he is coeditor of The Youth Development Handbook: Coming of Age in American Communities. Their most recent project involved action research with four programs that support the transition to adulthood for vulnerable youth in Latin America. His master’s and doctoral degrees in education are from Harvard University. He taught social studies and English for three years in a Washington, D.C., vocational high school. Keoki Hansen, MA, worked as the director of research and evaluation with Big Brothers Big Sisters of America (BBBSA) for 14 years, focusing on bridging research and practice. One of her key accomplishments was developing a school-based mentoring program in collaboration with practi­ tioners and mentoring researchers to increase mentee program retention by 50%. Ms. Hansen currently has her own consulting business, Hansen Evaluation Services, providing comprehensive program outcome evaluations. In addition to work with BBBSA, Ms. Hansen has taught research methods and statistics at Boston College and Pine Manor College. She also has been a consultant for the U.S. Army and the Center for Cognitive Neuroscience at the University of Pennsylvania. Ms. Hansen earned her bachelor’s degree in psychology from Castleton State College, graduating Summa Cum Laude with honors, and has her master’s degree in cognitive psychology from Boston College. John T. Harris, EdM, founded Applied Research Consulting (ARC) in 2000 to help bridge the gap between research and practice in youth development programming. Through ARC, Mr. Harris strives to make high-quality evaluation and consultation accessible to programs that might otherwise not be able to afford it. Mr. Harris specializes in action research, instrument development, and the integration of quantitative and qualitative methods. His primary research focuses on youth mentoring with an emphasis on understanding how individual descriptors, programmatic practices, and match characteristics interact to promote youth outcomes. Mr. Harris is the lead author of the Match Characteristics Questionnaire (MCQ) and the Youth Mentoring Survey (YMS). His publications have focused on mentoring relationship quality and include peer-reviewed journal articles, toolkit contributions, and the chapter “Assessing Mentoring Relationships” in the first edition of Handbook of Youth Mentoring. Angela K. Henneberger, PhD, is a postdoctoral research fellow at Youth-Nex, the University of

Virginia Center to Promote Effective Youth Development. She received her doctorate in educational psychology/applied developmental science from the University of Virginia. Her research focuses on peer influence during adolescence, with a specific focus on the ways in which peers positively influence adolescents. David Henry, PhD, is professor of health policy and administration at the University of Illinois at Chicago, a fellow of the Institute for Health Research and Policy, and a President’s Professor at the University of Alaska–Fairbanks. A community psychologist by training, Dr. Henry’s primary research interest is in the ecological factors that affect the lives of children and adolescents. His work includes over 100 peer-reviewed publications on research methods, child development, peer relationships, psychopathology, and prevention. With funding from the National Institutes of Health, the Centers for Disease Control and Prevention, and the William T. Grant Foundation, he has conducted or collaborated on prevention studies related to violence, delinquency, substance abuse, suicide, and sexual risk with urban minority and rural indigenous populations. Carla Herrera, PhD, is an independent consultant, most recently a senior research fellow for Public/ Private Ventures (P/PV). During her 14 years with P/PV, she conducted several P/PV mentoring evaluations including studies on school-based, group, and community-based models. She recently led P/PV’s Big Brothers Big Sisters (BBBS) School-Based Mentoring Impact Study and has published several articles based on that work, exploring questions including which volunteers yield (and which youth experience) the strongest benefits and what programmatic practices may be key in supporting those benefits. Herrera is a member of both the Research Advisory Council of Big Brothers Big Sisters of America and the Research and Policy Council of MENTOR: The National Mentoring Partnership. She received her doctorate degree in developmental psychology from the University of Michigan. Barton J. Hirsch, PhD, is professor of human development and social policy at Northwestern University. He is the author of A Place to Call Home: After-School Programs for Urban Youth and first author of After-School Centers and Youth Development: Case Studies of Success and Failure. Each of these books won the Social Policy Award for Best Authored Book from the Society for Research on Adolescence. Dr. Hirsch is a fellow of the American Psychological Association and the

576   HANDBOOK OF YOUTH MENTORING Society for Community Research and Action, and general editor of the book series Adolescent Lives in Context (NYU Press). His current research addresses how to promote successful school-towork transitions among low-income youth, with a special focus on teaching job interview skills. Thomas E. Keller, PhD, is the Duncan and Cindy Campbell Professor for Children, Youth, and Families with an emphasis on mentoring in the School of Social Work at Portland State University (PSU). He also is director of the PSU Summer Institute on Youth Mentoring and the Center for Interdisciplinary Mentoring Research. His research interests include the development and influence of mentoring relationships, social networks in mentoring interventions, implementation of program innovations, and the professional development of mentoring program staff. His studies have been funded by the National Institute of Mental Health, Office of Juvenile Justice and Delinquency Prevention, and the Spencer Foundation. He serves on boards or research advisory committees for several mentoring organizations, including MENTOR: The National Mentoring Partnership, Big Brothers Big Sisters of America, Friends of the Children, and Oregon Mentors. Prior to earning his PhD at the University of Washington, he worked for several years with a Big Brothers Big Sisters chapter in Seattle as a caseworker, supervisor, and program director. Michael S. Kelly PhD, LCSW, is associate professor and MSW program director at Loyola University Chicago’s School of Social Work. Prior to coming to Loyola in fall 2006, he was a school social worker, family therapist, and youth minister in the Chicago area for 14 years. He has authored over 40 books, journal articles, and book chapters on school social work, evidence-based practice, and positive youth development. His most recent book is Christianity and Social Work: Readings in the Integration of Faith and Social Work Practice, 4th ed. Dr. Kelly serves on the editorial boards of School Mental Health, School Social Work Journal, and Children & Schools, and has recently brought his work on school social work and evidence-based practice to researchers and practitioners in Rhode Island, Wyoming, Chile, and Japan. David C. R. Kerr, PhD, is an assistant professor at the Oregon State University School of Psychological Science, a research scientist at the Oregon Social Learning Center, and an Oregon licensed psychologist. He completed his doctorate in clinical psychology at the University of Michigan in 2004. He has authored 35 journal articles and chapters, most of

which concern developmental psychopathology and prevention science. Dr. Kerr has a special interest in using long-term longitudinal studies to inform the prevention of depression, conduct problems, and suicide among adolescents and young adults. Christopher B. Keys, PhD, is associate dean for research at the College of Science and Health and professor of psychology at DePaul University in Chicago. He has served as chair of the psychology departments at the University of Illinois at Chicago and DePaul University and as president of the Society for Community Research and Action. He is a fellow of the American Psychological Association and a recipient of the Distinguished Contribution to Theory and Research Award of the Society for Community Research and Action. Dr. Keys has been studying and developing interventions concerning the empowerment of people with disabilities and their families for the past two decades. The empirically supported methods he, Fabricio Balcazar, and their colleagues have developed and studied include goal-setting interventions and adaptations of goal-attainment scaling. Cheryl A. King, PhD, ABPP, is a professor in the departments of psychiatry and psychology at the University of Michigan, where she also serves as director of the Institute for Human Adjustment. She received her doctorate in clinical psychology from Indiana University, specializing in clinical child and adolescent psychology. Dr. King has a long-standing program of programmatic research and published studies focused on improving youth suicide risk recognition, assessment, and prevention strategies, including strategies based on adult support and mentorship models. This research has been funded by the National Institute of Mental Health, the Centers for Disease Control and Prevention, and numerous private foundations. Dr. King is a fellow of the American Psychological Association and past president of the Society for Clinical Child and Adolescent Psychology, the American Association of Suicidology, and the Association of Psychologists in Academic Health Centers. Sarah E. Kremer, MA, is program director for Friends for Youth’s Mentoring Institute. She has a master’s degree in art therapy from the School of the Art Institute of Chicago. She has been invited to present at conferences across the United States and in the Caribbean, Canada, and Australia for Friends for Youth and as a consultant for the Center for Applied Research Solutions, the National Mentoring Center, and MENTOR: The National Mentoring Partnership. Ms. Kremer has been invited to attend the Summer

About the Contributors   577 Institute on Youth Mentoring at Portland State University as a key technical assistance provider partner since its inception in 2007. Her resource, the Mentoring Journal, in its third printing, provides programs with a creative tool for structuring and documenting the life of a match. She has worked with First Exposures, a photography-based mentoring program in San Francisco, since 1995 and her essay, Mentoring Through Photography, appeared in the program’s first book in 2006. Gabriel P. Kuperminc, PhD, is a professor of psychology at Georgia State University. He completed a William T. Grant Foundation Faculty Scholars award, examining ecological factors in the development of Latino adolescents from immigrant families. He has evaluated several youth programs, including Teen Outreach, Cool Girls, Inc., and After School All-Stars. He consults with the Centers for Disease Control and Prevention on youth development approaches to reducing adolescent health risk behaviors, is a panelist on Social and Behavioral Contexts for Academic Learning for the Institute of Education Sciences, and serves on editorial boards of several journals, including Child Development, and Journal of Early Adolescence. Janis B. Kupersmidt, PhD, is president and senior research scientist at innovation Research & Training. She received her doctorate in clinical child psychology at Duke University, internship at Yale University, and was a tenured professor at the University of North Carolina at Chapel Hill. She co-led a team at MENTOR: The National Mentoring Partnership that wrote the Elements of Effective Practice for Mentoring (EEPM), 3rd ed. Based on EEPM, she created EQUIP, the Elements Quality Improvement Process, that includes a web-based self-assessment tool for mentoring programs and resources for technical assistance providers. She is the principal investigator on a grant from the Eunice Kennedy Shriver National Institute on Child Health and Human Development to create and evaluate the effectiveness of a prematch mentor training program, www .mentoringcentral.net. She has created customized mentor training for Big Brothers Big Sisters of America and a corporate mentor training program, Mentor in a Box, with Viacom and MENTOR: The National Mentoring Partnership. Simon Larose, PhD, is a professor in educational psychology at Laval University in Québec City, Canada. He obtained his doctorate in developmental psychology from the same university. His main fields of research and teaching are adolescent and young adult development, attachment and family

relationships, mentoring, school transitions, and academic adjustment. Since 2011, he has been the principal investigator of the Mentoring by High School Teachers (MHST) project. The MHST project is aimed to teach high school teachers a series of best practices in mentoring; to supervise their mentoring intervention with academically at-risk students; and to assess the impact of training, supervision, and mentoring on mentored students’ adjustment during their transition from primary to high school. Edith C. Lawrence, PhD, is a professor of clinical and school psychology at the University of Virginia. She cofounded and directs the Young Women Leaders Program, a research-based mentoring initiative that pairs college women to middle school girls considered at risk, to enhance the leadership skills of both groups. Her research has focused on the ways in which combining group with one-on-one mentoring serves both mentees and mentors. She also is a licensed clinical psychologist, coauthor of Competence, Courage and Change: An Approach to Family Therapy and developer of the Family Inventory of Resources and Stressors, a competence-based family assessment tool for professionals who work with families facing multiple problems. Richard M. Lerner, PhD, is the Bergstrom Chair in Applied Developmental Science and the director of the Institute for Applied Research in Youth Development at Tufts University. Having received his PhD from the City University of New York in in 1971, Dr. Lerner has more than 550 scholarly publications, including more than 70 authored or edited books. He was the founding editor of the Journal of Research on Adolescence and of Applied Developmental Science, which he continues to edit. Lerner is known for his theory of relations between lifespan human development and social change, and for his research about the relations between adolescents and their peers, families, schools, and communities. His work integrates the study of public policies and community-based programs with the promotion of positive youth development and youth contributions to civil society. Belle Liang, PhD, is a clinical/community psychologist and associate professor at the Lynch School of Education in counseling, developmental, and educational psychology at Boston College. Her research focuses on relational interventions, especially youth mentoring and innovative social media–based interventions for diverse youth. She has published widely on youth mentoring; serves as a member of several boards and committees, including the Research and

578   HANDBOOK OF YOUTH MENTORING Policy Council of MENTOR: The National Mentoring Partnership; and has received honors such as the inaugural Distinguished Asian/Pacific American Alumni Award from Indiana University, the Boston College Teaching With New Media Award, the Many Faces of Counseling Psychology Award for innovations in social justice intervention and research, and recognition by Youth Service America as a Leading ServiceLearning Researcher. Tina Malti, PhD, is an assistant professor of developmental and clinical child psychology at the University of Toronto. She also holds an affiliate scientist position at the Jacobs Center for Productive Youth Development. She received her doctorate in developmental psychology in 2003 from the Max Planck Institute for Human Development and the Free University of Berlin. Her research interests reside in children’s and adolescents’ emotions in the context of morality, the development of prosocial and antisocial behavior, and the ways in which socialemotional and moral development can help to promote mental health and well-being and reduce aggression and antisocial behavior in school and out-of-school settings. Dr. Malti has published over 85 publications, and she is a consulting editor of Child Development. She received the New Investigator Award from the Canadian Institutes of Health Research in 2012 and the Young Investigator Award from the Society for Research in Adolescence in 2010. Megan A. Mekinda, MA, is a doctoral candidate in human development and social policy at Northwestern University. She studies the design and evaluation of youth programs, with special interest in those to prepare urban minority youth for the postsecondary transition. Her dissertation research is an in-depth qualitative evaluation of After School Matters, an apprenticeship-based after-school program for Chicago public school students and the largest singlecity after-school program for high school youth in the country. Lyn Morland, MSW, MA, is director of Bridging Refugee Youth and Children’s Services (BRYCS), an initiative of Migration and Refugee Services, United States Conference of Catholic Bishops, and national technical assistance provider for the federal Office of Refugee Resettlement. She holds an MSW and an MA in anthropology and medical behavioral sciences and is currently affiliated with Howard University’s School of Social Work. Her 30-year career in cross-cultural social work includes developing programs for out-of-school youth in the Philippines, a community-based program for Sierra

Leonean refugee youth, research on ethnic identity in children, and participatory action research with Central American immigrants. As BRYCS director, she promotes the key importance of a range of mentoring approaches for the positive development of refugee and immigrant youth. Current memberships include the board of directors of Heritage Multicultural Youth and Family Programs, Inc., serving West Africans in the Washington, D.C., metropolitan area and in Sierra Leone. Megan K. Mueller, MA, is in her final year of doctoral studies at the Eliot-Pearson Department of Child Development at Tufts University. She is currently a doctoral research assistant at the Institute for Applied Research in Youth Development, and her research focuses on the role of human-animal interaction as a context for promoting positive youth development. Michael J. Nakkula, EdD, is a practice professor and chair of the Division of Applied Psychology and Human Development at the University of Pennsylvania’s Graduate School of Education. He teaches courses on adolescent development and the intersection of counseling, mentoring, and education within urban public schools. Among his publications, Professor Nakkula is the lead author of Understanding Youth: Adolescent Development for Educators and Building Healthy Communities for Positive Youth Development. He is the coeditor, with Michael J. Karcher, of a special issue of New Directions in Youth Development (2010) on the ways in which youth mentoring relationships are organized, assessed, and understood to promote best practices within different contexts. Christopher M. Napolitano, MA, is in his final year of doctoral studies at the Eliot-Pearson Department of Child Development at Tufts University. He is currently a doctoral research assistant at the Institute for Applied Research in Youth Development, and his research focuses on serendipity as an instance of the person-context relations involved in positive youth development. Gil G. Noam, PhD, EdD, is the founder and director of the Program in Education, Afterschool & Resiliency (PEAR) and an associate professor at Harvard Medical School and McLean Hospital. Trained as a clinical and developmental psychologist and psychoanalyst in both Europe and the United States, Dr. Noam has a strong interest in supporting resilience in youth, especially in educational settings. He served as the director of the Risk and Prevention program, and is the founder of the RALLY Prevention

About the Contributors   579 Program, a Boston-based intervention that bridges social and academic support in school, after-school, and community settings. Dr. Noam has also followed a large group of high-risk children into adulthood in a longitudinal study that explores clinical, educational, and occupational outcomes. Julia Pryce, PhD, LCSW, is an assistant professor at Loyola University Chicago’s School of Social Work. Following her work as a clinician within the child welfare system, she received her doctorate in social service administration from the University of Chicago, and has spent the last 10 years studying the process of mentoring relationship development, including those that take place in schools and in the broader community. Her work has been recognized by the International Mentoring Association and has been funded by the Office of Juvenile Justice and Delinquency Prevention and the U.S. Department of Health and Human Services. Dr. Pryce has particular interest in the intersection between research and practice. Projects to date have focused on the development, implementation, and evaluation of high-quality mentoring programs for at-risk youth, and on the nature and development of supportive nonparental relationships among youth who are system involved. Kellie G. Randall, MA, is a doctoral student in the Department of Human Development and Family Studies at the University of Connecticut. She received her BA in psychology and economics from Williams College and her master’s degree in human development and family studies from the University of Connecticut. Her current research interests include positive youth development and evaluation of community-based programs that serve children and their families. Jean E. Rhodes, PhD, is the MENTOR/National Mentoring Partnership Professor of psychology and the director of the Center for Evidence-Based Mentoring at the University of Massachusetts, Boston. A clinical/community psychologist, Dr. Rhodes has devoted her career to understanding the role of intergenerational relationships in the social, educational, and career development of disadvantaged youth. She is a fellow of the American Psychological Association and the Society for Research and Community Action, and was a distinguished fellow of the William T. Grant Foundation. Dr. Rhodes is a member of the John D. and Catherine T. MacArthur Foundation Research Network on Connected Learning. She serves as chair of the Research and Policy Council of the National Mentoring Partnership, is an overseer at the Boston Institute of Contemporary Art, and sits on the board of

directors and research advisory boards of over a dozen mentoring and policy organizations. Katrina E. Roundfield, MA, is a doctoral candidate in clinical/community psychology at DePaul University in Chicago. Her research focuses on the interaction between academic achievement and mental health among low-income, ethnic minority youth. She has also conducted research on ethnic identity and mentoring relationships of youth. Ms. Roundfield is a recipient of the 2009 Ford Foundation Predoctoral and 2012 Dissertation Fellowship administered by the National Research Council. Bernadette Sánchez, PhD, is an associate professor of psychology at DePaul University in Chicago. She earned her PhD in community psychology from the University of Illinois at Chicago. Her research focuses on youth mentoring and positive youth development. She has specifically examined mentoring relationships; education; and race, ethnicity, and cultural issues in urban, low-income youth using quantitative and qualitative methodologies. She has also worked with multiple community-based organizations in helping them to develop, enhance, and evaluate their youth mentoring programs. Miri Scharf, PhD, is a senior lecturer of developmental psychology in the Department of Counseling and Human Development at the University of Haifa in Israel. She is interested in attachment relationships across the lifespan, parenting, and developmental trajectories of resilience and risk. Elizabeth A. (Birdie) Shirtcliff, PhD, is an early research professor at the University of New Orleans and director of the Stress Physiology in Teens (SPIT) laboratory. Dr. Shirtcliff received her doctorate in biobehavioral health from Pennsylvania State University in 2003 and completed postdoctoral training at the University of WisconsinMadison in affective neuroscience and the biological and behavioral approaches to typical and atypical development in children. Dr. Shirtcliff uses a variety of noninvasive tools to investigate the interplay of biological and behavioral factors unfolding across children’s lives, especially in adolescence. Dr. Shirtcliff’s focus is on hormones because the endocrine system is stress responsive, often mirroring a child’s social environment. This interdisciplinary research examines both short-term stress responses, as well as biological changes that can consistently or even permanently change an individual’s biology. Many of these are putatively positive contextual forces, such as warmth, sensitive caregiving, and support during stress.

580   HANDBOOK OF YOUTH MENTORING Carmit-Noa Shpigelman, PhD, is an assistant professor at the Department of Community Mental Health, University of Haifa, Israel. She conducted her postdoctoral research at the Department of Disability and Human Development, University of Illinois at Chicago. She is an internationally leading expert on computer-mediated support in the context of disability. Her research focuses primarily on health promotion of people with a wide range of disabilities through online support, e-mentoring, and social media. Dr. Shpigelman identifies as a woman with a physical disability and has been involved in disability advocacy in Israel. Mark Snyder, PhD, is professor of psychology at the University of Minnesota, where he holds the McKnight Presidential Chair in Psychology and is the director of the Center for the Study of the Individual and Society. His research interests include the motivational foundations of individual and collective social action. He is the author of the book Public Appearances/Private Realities: The Psychology of SelfMonitoring and coeditor of the volumes Cooperation in Modern Society: Promoting the Welfare of Communities, States, and Organizations; Cooperation: The Political Psychology of Effective Human Interaction; The Psychology of Prosocial Behavior: Group Processes, Intergroup Relations, and Helping; and The Oxford Handbook of Personality and Social Psychology. Renée Spencer, EdD, is an associate professor at the Boston University School of Social Work. She received her master’s degree in social work from the University of Texas at Austin and her doctorate in human development and psychology from the Harvard Graduate School of Education. Her research focuses on the relational processes at work in youth mentoring and the determinants of more and less successful relationships. She has written numerous articles and book chapters on both formal and informal youth mentoring relationships and serves as a member of several boards and committees, including the Research and Policy Council of MENTOR: The National Mentoring Partnership and the Research Advisory Council of Big Brothers Big Sisters of America. Jenny Stern-Carusone, MSW, is a technical assistance specialist for the Center for the Advancement of Mentoring in Newton, Massachusetts. Since 1996, she has worked as an advocate for youth in a number of capacities, including designing a defense detention alternative reform program, launching a rural peer court program, and helping launch a community-based family intervention program for

Latino youth involved in juvenile justice. Prior to her current position, she served as program director for Committed Partners for Youth (CPY)–Big Brothers Big Sisters of Lane County (Oregon). She served as manager of the CPY Mentoring Children of Prisoners grant from 2003 to 2011, and was one of the first members of the Mentoring Children of Prisoners National Advisory Board, founded in 2009. Ms. Stern-Carusone extends her advocacy for youth involved in the justice system through participation on several community boards. Arthur A. Stukas, PhD, is a senior lecturer in the School of Psychological Science at La Trobe University in Melbourne, Australia. Together with Drs. Gil Clary and Mark Snyder, he has written about and researched the topic of volunteerism for nearly two decades. Dr. Stukas is a long-standing member of the Association for Psychological Science and several other professional organizations (e.g., Society for the Psychological Study of Social Issues, Society for Personality and Social Psychology). In collaboration with his students and colleagues, he has written numerous scholarly articles in scientific journals and chapters for relevant books, such as Mobilizing Adults for Positive Youth Development: Strategies for Closing the Gap Between Beliefs and Behaviors and The Handbook of Consumer Psychology. Dr. Stukas is also an executive editor for the Journal of Social Psychology. George M. Tarabulsy, PhD, is professor of psychology at Laval University in Québec City, Canada. Trained at McGill and Laval Universities and the University of Western Ontario, Dr. Tarabulsy has focused on the development of attachment relationships and social-emotional development in high-risk contexts, as well as on the elaboration of intervention strategies with mother-infant dyads that involve a strong mentoring component. He has also been involved in the training (and mentoring) of different social practitioners involved in primary prevention efforts with vulnerable parents and children. Jessica D. Thomason, MA, is a doctoral candidate in the community psychology program at Georgia State University. Her research focuses on positive youth development, specifically community initiatives to promote adolescent well-being and sexual health. She is the project coordinator for the evaluation of Cool Girls, Inc., and is currently evaluating the Healthy Teen Atlanta program. Her dissertation is a mixed-method study of the delivery of sexuality education in a youth development context.

About the Contributors   581 Susan G. Weinberger, EdD, is the founder and president of the Mentor Consulting Group based in Norwalk, Connecticut. She received her doctorate from the College of Business and Public Management at the University of Bridgeport. She was a pioneer in developing school-based mentoring in the early 1980s. Her model has been replicated in the United States, Canada, and internationally. She lectures on both youth mentoring and internal coaching and mentoring for corporations.

She consults to the Office of Juvenile Justice and Delinquency Prevention programs, Departments of Housing and Urban Development and Education, the National Congress of American Indians, and the Governor’s Prevention Partnership in Connec­ ticut. Her publications include the My Mentor & Me series, Guidebook to Mentoring, Preparing My Mentor for Me, Manual for Mentors, and Strengthening Native Community Commitment Through Mentoring.

The essential online tool for researchers from the world’s leading methods publisher

More content and new features added this year!

Find exactly what you are looking for, from basic explanations to advanced discussion

Discover Methods Lists— methods readings suggested by other users

“I have never really seen anything like this product before, and I think it is really valuable.” John Creswell, University of Nebraska–Lincoln

Watch video interviews with leading methodologists

Search a custom-designed taxonomy with more than 1,400 qualitative, quantitative, and mixed methods terms

Explore the Methods Map to discover links between methods Uncover more than 120,000 pages of book, journal, and reference content to support your learning

Find out more at www.sageresearchmethods.com