Violence and Crime in the Family : Patterns, Causes, and Consequences [9] 9781785602634, 9781785602627

In virtually all societies, crime is an ever-present problem. Although families are often envisioned as a 'safe hav

128 77

English Pages xix, 442 [463] Year 2015

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Front Cover
Violence and Crime in the Family: Patterns, Causes, and Consequences
Copyright page
Contents
List of Contributors
Editorial Board
Foreword
Family of Origin and Peer Influences on Intimate Partner Violence during Emerging Adulthood
Introduction
Background
Intimate Partner Violence and the Intergenerational Transmission of Violence
Peer Influences
Current Study
Analytic Plan
Methods
Measures
Dependent Variable
Family Violence
Peers’ Behaviors and Attitudes
Relationship Characteristics
Family of Origin Characteristics
Sociodemographic and Background Characteristics
Results
Descriptive Analysis
Predicting Intimate Partner Violence
Discussion
Notes
Acknowledgments
References
A Latent Class Approach to Understanding the Intergenerational Transmission of Violence in Emerging Adult Relationships
Intergenerational Transmission of Violence
Gender
Approaches to Studying the Intergenerational Transmission of Violence
Focusing on One Form of Violence
Focusing on the Presence of Violence
Person-Oriented Approach to Studying the Intergenerational Transmission of Violence
Person-Oriented Theory
Person-Oriented Methods
The Current Study
Methods
Participants
Procedures
Measures
Maltreatment and Intimate Partner Violence
Childhood Family Environment
Emotional Reactivity
Analytical Approach
Results
Prevalence of Intimate Partner Violence
Model Selection
Description of Three-Class Solution
Differences in Childhood Family Environment as a Function of Class Membership
Differences in Emotional Reactivity as a Function of Class Membership
Discussion
Limitations and Future Directions
Implications
Conclusion
Notes
References
Gender, Domestic Violence, and Patterns of Conviction: Analysis of India’s Supreme Court Rulings
Gender, Family, Violence, and Law
Family, Dowry, and Domestic Violence in India
Family, Domestic Violence, and the Law
Domestic Violence Laws in India
Data and Methods
Analytic Strategy
Analysis: Gender, Domestic Violence, and Court Rulings
Overview of Cases
Patterns of Conviction
Thematic Analysis
Family Relationships and Violence
Homeless in Their Own Home
Discussion and Conclusion
Notes
Acknowledgment
References
Structural Basis of Gender Violence in Cross-Regional Marriages: Tale from Haryana, India
Introduction
Skewed Sex Ratio Leading to Marriage Squeeze
Socio-Economic Context of Cross-Regional Marriages
Political Economy of Haryana
Sex Ratio of Haryana
Media Reportage from Haryana
Methodology
Findings from the Field
Marriage in Haryana: Social Obligation or Personal Need
Caste Barriers Redefined: Altruism or Vanity
Marriage in West Bengal: Social Obligation or Offloading the Burden?
Fixing a Marriage Alliance: Commodification of Brides
Crossing the Regions: Challenges and Beyond
Adjusting to New Ways of Life
Adjusting to Violent Physical Behaviour
Differing Relations to Work
Identity Crises: Constrained Existence
Conclusion
Notes
References
Family Violence in Syria
Introduction
Objective of the Study
Forms of Violence in the Family
Physical Violence
Sexual Violence
Psychological, Mental, Emotional Violence
Economic Violence
Social Violence
Violence against Women in Arab, Asian, Non-Western, and Western Societies
Violence in the Family Is Underdiagnosed
Causes of Violence in the Family
Prevalence of Family Violence in Syria
Impact of Violence against Individuals, the Family and Society
Violence and Physical Health Consequences
Violence, Mental Illness, and Mental-Health Effects
Solutions and Alternatives to Reduce the Phenomenon of Domestic Violence and Its Negative Impact
Purpose
Method
Participants
Inclusion Criteria
Measures
Questionnaire on the Phenomenon of Violence and Its Effects
Data Analysis
Results
Socio-Demographic Aspects
Causes of Family Violence in This Study
Discussion
Conclusion
Limitations and Future Research Needs
Role of the Funding Source
Conflict of Interest
Acknowledgments
References
Violent Socialization and Youth Violence across Different Nations: International Variations in Familial and Contextual Factors
A Typology of Violent Socialization
Violent Socialization within the Family
National Norms of Violent Socialization
Methods
Sample
Variables
Youth Violence
Individual-Level Variables
Violent Family Socialization
Group-Level Variables
Contextual Violent Socialization
Analysis
Results
Youth Violence
Violent Socialization within the Family
Violent Socialization as a Contextual Effect
Violent Socialization and Youth Violence across Different Nations
Discussion
Conclusion
Notes
References
“He is Still My Son”: Aging and Living in the Shadow of an Abusive Adult Child with Mental Disorder
Elder Abuse as a Crime within the Family: Theoretical Framework
Elder Abuse by ACMD
Aging Parents Experience in Caring for an ACMD
Method
Participants and Sample
Data Collection
Procedure
Data Analysis and Trustworthiness
Findings
Constructing Parental Identity in a Shared Reality of Violence
“No-one Else Came but Me”: A Devoted Parenthood
“I’m Alive but I’m not Living”: A Parenthood of “Martyrdom”
“What Kind of a Life Have I Had?”: Bitter and Deprived Parenthood
“My Experience Gives Me Strength”: Parenthood as Bravery against All Odds
Social and Family Networks as a Resource in Coping with ACMD
Keeping a Daily Life Routine as an Anchor in a Vulnerable, Abusive Relationship
Discussion
Limitations and Recommendations for Further Study
Practical Implications
References
Parental Influence and Dating Violence among Students at Nigerian Universities
Introduction
Objectives of the Study
Justification of the Study
Method
Research Design
Participants
Results
Discussion
Limitations of the Study
Conclusion
References
Long-Term Effects of Neglect on Violence: Are They Spurious or Indirect?
Neglect and Academic Achievement
Neglect and Physical Abuse
Neglect and General Offending
Neglect and Parental Attachment
Neglect and Alcohol Abuse
Negative Emotionality
Being in School
Method
Data
Measures
Dependent variables
Independent Variable
Neglect
Control Variables
Demographic Characteristics
Income
Independent
Married
In School
Current Employment
Parent Education
Community Disorder
Physical Abuse
Alcohol Use
Drug Use
Parental Attachment
Academic Achievement
Negative Emotionality
Frequency of Nonviolent Offending
Analytic Approach
Findings
Testing Candidate Confounding Factors
Indirect Effects
Follow-Up Analyses
Conclusions
References
Situating the Experience of Maternal Incarceration: Childhood and Young Adult Context
Introduction
Background
Children’s Experiences of Maternal Incarceration
Existing Links between Paternal and Maternal Incarceration
Incarceration and Young Adulthood
Methods
Analytic Strategy
Measures
Family Structure
Socioeconomic Disadvantage
Exposure to Crime and Violence
Findings
Family Structure
Socioeconomic Disadvantage
Exposure to Violence, Crime, and Substance Use
Discussion
Conclusion
Notes
Acknowledgments
References
Delinquent Effects of Childhood Exposure to Violent Victimization: A Latent Longitudinal Class Analysis
Introduction
The Impact of Childhood Exposure to Violence
The Impact of Child Abuse and Exposure to Violence in the Family of Origin
The Impact of Adolescent Dating Violence Victimization
The Impact of Street and Peer Violence Exposure
The Present Study
Methodology
Data
Variables
Delinquency
Exposure and Experience of Violent Victimization
Analysis Plan
Results
Discussion
Key Findings and Policy Implications
Limitations and Future Research
Notes
Acknowledgments
References
Girls Offenders Pathways into the Spanish Juvenile Justice
Introduction
Literature Review
Research Design and Methodological Issues
Analysis and Interpretation of Data
Ethical Considerations of the Research
Girls’ Pathways into the Juvenile Justice System: Contexts and Process of Social Exclusion
Limitations and Difficulties Encountered
Discussion and Final Considerations
Notes
References
A Binary Model of Broken Home: Parental Death-Divorce Hypothesis of Male Juvenile Delinquency in Nigeria and Ghana
Introduction
Theoretical Background
Methodology
An Overview of Literature: Broken Homes
Parental Death and Parental Divorce
Critique of Evidence
Conclusion
Notes
Acknowledgements
References
Delinquency, Gender, and Social Control within the Familial Context
The Social Bond
Female Delinquency
Female Delinquency Typologies
Male/Female Social Control Studies
Methods
Definition of Variables
Social Bond Variables
Delinquency Variables
Individual Characteristics Variables
Hypotheses
School Problems
Substance Abuse
Running Away
Fights
Arrests
Fights
Arrests
Conclusions
References
The Self and the ‘Selfie’: Cyber-Bullying Theory and the Structure of Late Modernity
Introduction
Is Cyber-Bullying Just Another Type of Bullying?
The Self and the ‘Selfie’ in Late Modernity
Conclusion
Notes
References
Setting the Problem of Missing Children in the Republic of Ireland against Contemporary International Research Perspectives
Introduction
The United States of America
Summary of U.S. Literature
Great Britain
Summary
The European Union
Ireland: Case Study
Volunteer Case Study Profile 1: Homelessness
Volunteer Case Study Profile 2: Runaway
Volunteer Case Study Profile 3: Runaway
Concluding Discussion
Notes
Acknowledgements
References
About the Authors
Recommend Papers

Violence and Crime in the Family : Patterns, Causes, and Consequences [9]
 9781785602634, 9781785602627

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

VIOLENCE AND CRIME IN THE FAMILY: PATTERNS, CAUSES, AND CONSEQUENCES

CONTEMPORARY PERSPECTIVES IN FAMILY RESEARCH Series Editor: Sampson Lee Blair Recent Volumes: Volume 1:

Through the Eyes of the Child Re-Visioning Children as Active Agents of Family Life Edited by Michael Abrams, Johnson Matthey, B. A. Murrer, Felix M. Berardo and Constance L. Shehan, 2000

Volume 2:

Families, Crime and Criminal Justice Charting the Linkages Edited by Greer Litton Fox and Michael L. Benson, 2000

Volume 3:

Minding the Time in Family Experience Emerging Perspectives and Issues Edited by Kerry Daly, 2001

Volume 4:

Intergenerational Ambivalences New Perspectives on Parent-Child Relations in Later Life Edited by Karl A. Pillemer and Kurt K. Luscher, 2003

Volume 5:

Families in Eastern Europe Mihaela Robila, 2004

Volume 6:

Economic Stress and the Family Sampson Lee Blair, 2012

Volume 7:

Visions of the 21st Century Family: Transforming Structures and Identities Edited by Patricia Neff Claster and Sampson Lee Blair, 2013

Edited by Edited By

Volume 8A: Family Relationships and Familial Responses to Health Issues Edited by Jennifer Higgins McCormick and Sampson Lee Blair, 2014 Volume 8B: Family and Health: Evolving Needs, Responsibilities, and Experiences Edited by Sampson Lee Blair and Jennifer Higgins McCormick, 2014

CONTEMPORARY PERSPECTIVES IN FAMILY RESEARCH VOLUME 9

VIOLENCE AND CRIME IN THE FAMILY: PATTERNS, CAUSES, AND CONSEQUENCES EDITED BY

SHEILA ROYO MAXWELL Michigan State University, East Lansing, MI, USA

SAMPSON LEE BLAIR The State University of New York, Buffalo, NY, USA

United Kingdom North America India Malaysia China

Japan

Emerald Group Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2015 Copyright r 2015 Emerald Group Publishing Limited Reprints and permissions service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. Any opinions expressed in the chapters are those of the authors. Whilst Emerald makes every effort to ensure the quality and accuracy of its content, Emerald makes no representation implied or otherwise, as to the chapters’ suitability and application and disclaims any warranties, express or implied, to their use. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-78560-263-4 ISSN: 1530-3535 (Series)

ISOQAR certified Management System, awarded to Emerald for adherence to Environmental standard ISO 14001:2004. Certificate Number 1985 ISO 14001

CONTENTS LIST OF CONTRIBUTORS

ix

EDITORIAL BOARD

xiii

FOREWORD

xv

FAMILY OF ORIGIN AND PEER INFLUENCES ON INTIMATE PARTNER VIOLENCE DURING EMERGING ADULTHOOD Mallory D. Minter, Monica A. Longmore, Peggy C. Giordano and Wendy D. Manning

1

A LATENT CLASS APPROACH TO UNDERSTANDING THE INTERGENERATIONAL TRANSMISSION OF VIOLENCE IN EMERGING ADULT RELATIONSHIPS Peter M. Rivera and Frank D. Fincham

23

GENDER, DOMESTIC VIOLENCE, AND PATTERNS OF CONVICTION: ANALYSIS OF INDIA’S SUPREME COURT RULINGS Preethi Krishnan and Mangala Subramaniam

45

STRUCTURAL BASIS OF GENDER VIOLENCE IN CROSS-REGIONAL MARRIAGES: TALE FROM HARYANA, INDIA Sonali Mukherjee

73

FAMILY VIOLENCE IN SYRIA Muaweah Ahmad Alsaleh

99

v

vi

CONTENTS

VIOLENT SOCIALIZATION AND YOUTH VIOLENCE ACROSS DIFFERENT NATIONS: INTERNATIONAL VARIATIONS IN FAMILIAL AND CONTEXTUAL FACTORS Aime´e X. Delaney

129

“HE IS STILL MY SON”: AGING AND LIVING IN THE SHADOW OF AN ABUSIVE ADULT CHILD WITH MENTAL DISORDER Tova Band-Winterstein, Hila Avieli and Yael Smeloy

153

PARENTAL INFLUENCE AND DATING VIOLENCE AMONG STUDENTS AT NIGERIAN UNIVERSITIES Olufemi Adeniyi Fawole and Ebenezer Bayode Agboola

177

LONG-TERM EFFECTS OF NEGLECT ON VIOLENCE: ARE THEY SPURIOUS OR INDIRECT? Joanne Savage and Amanda Murray

189

SITUATING THE EXPERIENCE OF MATERNAL INCARCERATION: CHILDHOOD AND YOUNG ADULT CONTEXT Xing Zhang and Allison Dwyer Emory

219

DELINQUENT EFFECTS OF CHILDHOOD EXPOSURE TO VIOLENT VICTIMIZATION: A LATENT LONGITUDINAL CLASS ANALYSIS Henriikka Weir and Catherine Kaukinen

255

GIRLS OFFENDERS PATHWAYS INTO THE SPANISH JUVENILE JUSTICE Rosario Pozo Gordaliza

285

A BINARY MODEL OF BROKEN HOME: PARENTAL DEATH-DIVORCE HYPOTHESIS OF MALE JUVENILE DELINQUENCY IN NIGERIA AND GHANA Suleman Ibrahim

311

Contents

vii

DELINQUENCY, GENDER, AND SOCIAL CONTROL WITHIN THE FAMILIAL CONTEXT Melissa A. Menasco

341

THE SELF AND THE ‘SELFIE’: CYBER-BULLYING THEORY AND THE STRUCTURE OF LATE MODERNITY Shahid Alvi, Steven Downing and Carla Cesaroni

383

SETTING THE PROBLEM OF MISSING CHILDREN IN THE REPUBLIC OF IRELAND AGAINST CONTEMPORARY INTERNATIONAL RESEARCH PERSPECTIVES Michael Rush

407

ABOUT THE AUTHORS

435

This page intentionally left blank

LIST OF CONTRIBUTORS Ebenezer Bayode Agboola

College of Social and Management Sciences, Afe Babalola University, Ado Ekiti, Nigeria

Muaweah Ahmad Alsaleh

Department of Psychology, University of Aleppo, Syria

Shahid Alvi

Faculty of Social Science and Humanities, University of Ontario Institute of Technology, Canada

Hila Avieli

Faculty of Social Sciences, University of Ariel, Israel; Faculty of Welfare and Health Sciences, University of Haifa, Israel

Tova Band-Winterstein

Department of Gerontology, University of Haifa, Israel

Carla Cesaroni

Faculty of Social Science and Humanities, University of Ontario Institute of Technology, Canada

Aime´e X. Delaney

Worcester State University, MA, USA

Steven Downing

Faculty of Social Science and Humanities, University of Ontario Institute of Technology, Canada

Allison Dwyer Emory

Departments of Sociology and Policy Analysis and Management, Cornell University, NY, USA

Olufemi Adeniyi Fawole

Department of Sociology, University of Ilorin, Nigeria

Frank D. Fincham

College of Human Sciences, The Florida State University, FL, USA

ix

x

LIST OF CONTRIBUTORS

Peggy C. Giordano

Department of Sociology, Bowling Green State University, Bowling Green, OH, USA

Suleman Ibrahim

Centre for Doctoral Training in Cyber Security, Royal Holloway University of London, UK

Catherine Kaukinen

University of Central Florida, FL, USA

Preethi Krishnan

Department of Sociology, Purdue University, IN, USA

Monica A. Longmore

Department of Sociology, Bowling Green State University, Bowling Green, OH, USA

Wendy D. Manning

Department of Sociology, Bowling Green State University, Bowling Green, OH, USA

Melissa A. Menasco

Criminal Justice Department, The State University of New York, College at Buffalo, NY, USA

Mallory D. Minter

Department of Sociology, Bowling Green State University, Bowling Green, OH, USA

Sonali Mukherjee

Institute of Social Sciences, New Delhi, India

Amanda Murray

Department of Psychology, American University, DC, USA

Rosario Pozo Gordaliza University of Balearic Irelands, Spain Peter M. Rivera

College of Human Sciences, The Florida State University, FL, USA

Michael Rush

School of Social Policy, Social Work and Social Justice, University College Dublin, Ireland

Joanne Savage

Department of Justice, Law and Criminology, American University, DC, USA

Yael Smeloy

Department of Gerontology, University of Haifa, Israel

xi

List of Contributors

Mangala Subramaniam Department of Sociology, Purdue University, IN, USA Henriikka Weir

School of Public Affairs, University of Colorado at Colorado Springs, CO, USA

Xing Zhang

Department of Policy Analysis and Management, Cornell University, NY, USA

This page intentionally left blank

EDITORIAL BOARD Josip Obradovic´ University of Zagreb, Croatia

Clarence M. Batan University of Santo Tomas, Philippines

Gary W. Peterson Miami University, USA

Eli Buchbinder University of Haifa, Israel Yu-Hua Chen National Taiwan University, Taiwan

Matthias Pollman-Schult Social Science Research Center Berlin, Germany

Patricia Neff Claster Edinboro University, USA

Allison J. Pugh University of Virginia, USA

Teresa M. Cooney University of Colorado-Denver, USA

Ria Smit University of Johannesburg, South Africa

Rosalina Pisco Costa University of E´vora, Portugal

Helen M. Stallman University of South Australia, Australia

Alda Britto da Motta Federal University of Bahia, Brazil

Fleur Thome´se VU University, Amsterdam, The Netherlands

Giovanna Gianesini University of Bologna, Italy Cardell K. Jacobson Brigham Young University, USA

xiii

This page intentionally left blank

FOREWORD The focus on Crime, Violence, and the Family in this volume of CPFR provides an important avenue for the exchange of ideas and research on the many forms of violence that face families across many cultures. The family is universally considered the provider of elemental support and sustenance for societies, as made evident by nurturing its children, providing social and economic support for its members, and providing the haven that people need in order to be productive members of society. However, these nurturance and sustenance missions are all too often challenged by forces from outside and, sadly, also from inside the family unit. Violence or systemic aggression that are perpetrated by a family member against another, or violence experienced by the family from outside that disrupts its functions, are among the major challenges facing communities across the globe. In Western nations, and particularly the United States, researchers have made great strides over the past several decades to better understand the nature of intimate partner violence, as well as child-directed and child-witnessed violence. In addition, research on the impact of community and neighborhood violence on the family is developing, although this is still limited. These same advances cannot be said of many other nations, where information about familial violence is less common and less accessible. This volume brings together contributions by researchers from four continents, and thus provides an important avenue toward understanding the status, patterns, and consequences of intra- and extra-family violence across varied cultural and social settings. Among the most examined areas in family violence research is the question of violence perpetration by those who have previously witnessed parental violence, or those who have experienced direct abuse. Two papers in this volume looked at the perpetration of intimate partner violence (IPV) by emerging adults, each using a unique approach. In their chapter, “Family of Origin and Peer Influences on Intimate Partner Violence during Emerging Adulthood,” Mallory D. Minter, Monica A. Longmore, Peggy C. Giordano and Wendy D. Manning explore the social influences of intimate partner violence (IPV) perpetration among emerging adults (aged 22 29) using the Toledo Adolescent Relationships Study (TARS). xv

xvi

FOREWORD

The authors examine the competing effects of witnessing parental violence during adolescence, and peer involvement and attitudes toward IPV, on the subjects’ own IPV perpetration. Although both these factors have often been considered in IPV and delinquency research, seldom are these two factors examined together, or their competing effects assessed in explaining IPV perpetration. The authors find that witnessing parental violence and peers’ experiences and toward IPV have separate and statistically significant effects on IPV perpetration. This adds an important theoretical dimension on the social influences of IPV perpetration and expands the current knowledge base about IPV. In a similar vein, Peter M. Rivera and Frank D. Fincham also assess IPV perpetration among emerging adults in their chapter: “A Latent Class Approach to Understanding the Intergenerational Transmission of Violence in Emerging Adult Relationships.” Their study provides a unique approach to examining intergenerational transmission of violence with their use of latent class modeling. The latent class approach uses a person-oriented model rather than the more commonly used variable-focused model in explaining IPV perpetration. The authors identify three classes of individuals who perpetrate violence among those who experienced violence: the full transmission group, the psychological transmission group, and the no transmission group. This unique method and perspective calls for continued research into possible subclasses or groups of potential perpetrators among those who experienced violence. There is also a need to recognize the importance of understanding culture and context in the study of domestic and sexual violence. Several papers incorporate in their analyses the unique cultural nuances in the framing of domestic violence and sexual assault, and underscore how cultural nuances need to be more carefully considered in the definitions and understandings of domestic and sexual violence. In “Gender, Domestic Violence, and Patterns of Conviction: Analysis of India’s Supreme Court Rulings,” Preethi Krishnan and Mangala Subramaniam examine the circumstances of domestic violence in India through their study of Supreme Court rulings of domestic violence between 1995 and 2011. The unique framing of the family in Indian culture assists in better understanding the circumstances and contexts of violence. Sonali Mukherjee, in “Structural Basis of Gender Violence in Cross-Regional Marriages: Tale from Haryana, India,” offers a thought-provoking glimpse into the consequences of a structure that aims to masculinize society through male-birth preference, and the ensuing lack of marriageable women due to this partiality. This structure sets up the need for migrant brides and the resulting hardships brought about by cross-border marriages. This intensive qualitative

Foreword

xvii

study provides an in-depth picture of some of the hardships faced by Bangladeshi women who are brought into arranged marriages with Haryani men. Muaweah Ahmad Alsaleh similarly provides a glimpse of family violence in a society that few in the West are familiar with. In “Family Violence in Syria,” Alsaleh underscores the unique characteristics of Syrian society and the need for greater study, and also the extant difficulties of studying family violence in that nation. A large part of family violence research pertains to the effects of this violence on the children who experience this violence in their homes. Researchers have studied the intergenerational transmission (IGT) of violence, as well as general delinquency and antisocial behaviors exhibited by affected children. This volume includes several contributions across various cultures on the general delinquency and antisocial behaviors exhibited by children who experienced or witnessed IPV in their homes. In “Violent Socialization and Youth Violence Across Different Nations: International Variations in Familial and Contextual Factors,” Aime´e X. Delaney assesses variations in youth violence across countries and finds that community and familial violence in these countries have significant effects on youth violence. In ““He Is Still My Son”: Aging and Living in the Shadow of an Abusive Adult Child with Mental Disorder,” Tova Band-Winterstein, Hila Avieli, and Yael Smeloy assess a different dimension of child-aggression that is not well understood and not extensively examined: that of elder abuse particularly perpetrated by children who have mental disorders. This qualitative piece provides an important foundation for future research and theory development. Olufemi Adeniyi Fawole and Ebenezer Bayode Agboola’s study, “Parental Influence and Dating Violence among Students at Nigerian Universities,” extends our understanding of the effects of direct and witnessed parental violence on the perpetration of sexual violence by students in a Nigerian university. This volume also includes several quantitative pieces which examine the effects of family violence on the children who were exposed to this violence. In “Long-Term Effects of Neglect on Violence: Are They Spurious or Indirect?,” Joanne Savage and Amanda Murray use the National Longitudinal Study of Adolescent to Adult Health (Add Health) data to assess the direct and indirect effects of child neglect on youth violence, while controlling for physical abuse and other indicators of offending behavior. They find a direct and robust relationship between neglect and youth violence. Xing Zhang and Allison Dwyer Emory also used Add Health data in their analysis of the effects of maternal incarceration on children and young adults in their chapter “Situating the Experience of Maternal

xviii

FOREWORD

Incarceration: Childhood and Young Adult Context.” Although maternal incarceration is not, by itself, a form of violence, evidence in criminological research attests that children of incarcerated parents, particularly of an incarcerated mother, experience long-term hardships including displacement and violence as a result of this incarceration. This combination of factors contributes to the “cycle” of violence. Zhang and Emory substantiate this result using the Add Health longitudinal database. Henriikka Weir and Catherine Kaukinen, in “Delinquent Effects of Childhood Exposure to Violent Victimization: A Latent Longitudinal Class Analysis,” also use Add Health data and Longitudinal Latent Class modeling to assess possible classes of offending trajectories by children exposed to violence. Their findings of three distinct trajectories of offending and a narrowing gap between male and female delinquency over time underscore the need for more research inquiries, particularly those using longitudinal panel data, and lay the foundation for theory development. The chapters by Rosario Pozo Gordaliza, “Girls Offenders Pathways into the Spanish Juvenile Justice,” and Suleman Ibrahim, “A Binary Model of Broken Home: Parental Death-Divorce Hypothesis of Male Juvenile Delinquency in Nigeria and Ghana,” highlight the need to more actively examine male and female delinquency across cultures using varieties of appropriate methodologies and theories. Gordaliza reports on a qualitative study of 44 female delinquents from a female juvenile reform institution in Andalucia, Spain. Ibrahim reports on a review of relevant literature on Nigerian/Ghanaian research on delinquency/antisocial behaviors. Both papers present unique perspectives on how delinquency may be examined within Spanish and Nigerian/Ghanaian cultures, noting important characteristics unique to each society that must be considered when assessing outcome behaviors like delinquency. These chapters present important considerations for all researchers who conduct cross-cultural studies on the family and the impact of family dysfunctions. The family environment can also be seen as one of the contexts within which maladaptive behaviors develop. In “Delinquency, Gender, and Social Control within the Familial Context,” Melissa A. Menasco explores the social bonds which are supposed to hinder or limit adolescent delinquency. Interestingly, she finds that substantial differences are evident in the respective behaviors of girls and boys. Shahid Alvi, Steven Downing, and Carla Cesaroni address the increasing role of technology, as it pertains to maladaptive juvenile behaviors. In “The Self and the “Selfie”: CyberBullying Theory and the Structure of Late Modernity,” they demonstrate not only the growing problem of online intimidation, harassment, and

Foreword

xix

bullying, but also how contemporary criminological theories need to have a broader perspective in order to fully comprehend the very nature of cyberbulling. In many instances, abused and neglected children, particularly girls, run away from home. Running away can lead to circumstances that precipitate missing children occurrences and reports, yet, repeat runaways are generally missed by Missing Children hotlines. In “Setting the Problem of Missing Children in the Republic of Ireland against Contemporary International Research Perspectives,” Michael Rush examines just this problem and recommends a new approach to policies on missing children that should incorporate repeat runaways. This is particularly important for those children who repeatedly run away due to chronic and systemic violence in their homes and who then often fall prey to gangs and other criminal groups. Overall, the contributions to this volume’s focus on Crime, Violence, and the Family are methodologically and also geographically diverse. Contributions from the African, Asian, European, and the American continents provide an expansive view of the challenges and cultural idiosyncrasies within familial relationships that are important in understanding family violence and its outcomes. Although a large body of research has been generated about family violence in Western societies, and particularly in the United States, much more is needed to understand this dynamic in non-Western societies. Efforts to expand empirical research and theoretical framework by including diverse perspectives across cultures are indeed necessary. This volume provides a step toward that goal. We thank all the authors who provided important contributions to this volume, and all the anonymous reviewers who provided thoughtful and detailed reviews.

Sheila Royo Maxwell Sampson Lee Blair Editors

This page intentionally left blank

FAMILY OF ORIGIN AND PEER INFLUENCES ON INTIMATE PARTNER VIOLENCE DURING EMERGING ADULTHOOD Mallory D. Minter, Monica A. Longmore, Peggy C. Giordano and Wendy D. Manning ABSTRACT Purpose Prior researchers have documented significant effects of family violence on adult children’s own risk for intimate partner violence (IPV). Yet, few studies have examined whether exposure to family violence while growing up as well as emerging adults’ reports of their current peers’ behaviors and attitudes influenced self-reports of intimate partner violence perpetration. The current study based on interviews with a large, heterogeneous sample of men and women assessed the degree to which current peers’ attitudes and behaviors contributed to risk of intimate partner violence perpetration, net of family violence. Methodology/approach Using data from the Toledo Adolescent Relationships Study (TARS) (n = 928), we examined associations between family violence indicators, peers’ behaviors and attitudes, and

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 1 22 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009001

1

2

MALLORY D. MINTER ET AL.

self-reports of intimate violence perpetration among adults ages 22 29. We used ordinary least squares regression and controlled for other known correlates of IPV. Findings For men and women, we found a significant relationship between witnessing parental violence during adolescence and IPV perpetration in emerging adulthood, and a positive relationship between current peers’ IPV experiences and attitudes and respondents’ perpetration. We also found that for respondents who reported higher, compared with lower, peer involvement in partner violence, the effects of parental violence were stronger. Originality/value We provided a more comprehensive assessment of peers’ IPV to this body of research, which tends to focus on family violence. Studies have examined peers’ attitudes and behavior during adolescence, but we extended this work by examining both peer and familial influences into emerging adulthood. Keywords: Intimate partner violence; peers’ influence; family violence; emerging adulthood

INTRODUCTION Emerging adulthood is a distinct stage in the life course characterized by the development of intimate relationships. Although prevalence rates vary, many researchers have reported that rates of intimate partner physical violence peak during the life stage of emerging adulthood (e.g., Breiding, Black, & Ryan, 2008; Halpern, Spriggs, Martin, & Kupper, 2009; Johnson, Giordano, Manning, & Longmore, 2015; Thompson et al., 2006), and that much of this violence is mutual or reciprocal. Prior theorizing has suggested that the family of origin often provides a direct model for how to be violent toward loved ones. As such, researchers have examined intergenerational or family influences, such as witnessing parental violence (e.g., Franklin & Kercher, 2012; Turcotte-Seabury, 2010), experiencing harsh parenting (e.g., Herrenkohl et al., 2004) or both (e.g., Ehrensaft et al., 2003) to explain emerging adults’ self-reports of perpetrating violence toward their own intimate partners. In addition to the family of origin, peers may influence emerging adults’ acceptance and tolerance of intimate partner violence by their involvement in violent relationships (Arriaga & Foshee, 2004) and their expressed

Family and Peer Influences on IPV during Emerging Adulthood

3

support for violence against partners (Silverman & Williamson, 1997; Williamson & Silverman, 2001). Moreover, there may be linkages between the intergenerational transmission of violence and the influence of peers in that emerging adults who witnessed parental violence or experienced violence first hand while growing up may be more likely to associate with a deviant crowd as emerging adults (Silverman & Williamson, 1997). Yet, few studies have examined whether, both exposure to family violence while growing up as well as emerging adults’ reports of their current peers’ behaviors and attitudes affect self-reports of intimate partner violence perpetration. We analyzed data from a community sample, the Toledo Adolescent Relationships Study (TARS). This dataset focuses on individuals’ experiences with parents, peers, and intimate partners as they transitioned from adolescence to emerging adulthood. Drawing on experiences when respondents were ages 22 29, we examined two indicators of family violence (i.e., witnessing parental violence, and harsh parenting), and two indicators of current peer influence (i.e., intimate partner violence experiences and attitudes toward partner violence), as well as other known correlates of emerging adults’ intimate partner violence perpetration. We hypothesized that in addition to family violence indicators, peers’ experiences with and attitudes toward intimate partner violence would positively influence emerging adults’ self-reports of perpetrating relationship violence. Although we build on prior research, our unique contribution is examining adult peer, in addition to family of origin, influences. We expected that peer influences may be stronger than family violence indicators, and that women, compared with men, may be more susceptible to peer influences.

BACKGROUND Intimate Partner Violence and the Intergenerational Transmission of Violence Intimate partner violence is common among emerging adults. Estimates from the Longitudinal Study of Adolescent to Adult Health (Add Health) indicated that 25% of unmarried women and men, ages 18 24, experienced relationship violence (Halpern et al., 2009). That so many individuals report intimate partner aggression during this stage in the life course underscores the need to understand its precursors because a key developmental

4

MALLORY D. MINTER ET AL.

task during emerging adulthood is learning to get along with intimate partners. A large body of literature has theorized ways in which violence is learned (Bandura, 1977), and that witnessing parental violence as well as experiencing harsh parenting while growing up teaches individuals to be violent (e.g., Ehrensaft et al., 2003). Through these early experiences of violence, children and adolescents learn meanings, symbols, and definitions that can be called on and continually re-shaped when exposed to other situations with potential for violence (Burke, Stets, & Pirog-Good, 1988). In other words, violence is learned in the home by witnessing and experiencing violence first hand and may be re-created in subsequent intimate relationships. There is an extensive empirical literature that supports the notion of the intergenerational transmission of violence during emerging adulthood. Turcotte-Seabury’s (2010) multi-national study (n = 14,252), for example, concluded that witnessing parental violence limited individuals’ capabilities of managing anger and that this was significantly associated with emerging adults’ perpetration of intimate partner violence. Band-Winterstein (2014) found through in-depth interviews with men and women (n = 25) that witnessing parental violence is an experience that individuals carry with them as a constant echo of the past as they evaluate their own lives and relationships throughout adulthood. Goldblatt (2003) noted similar findings from interviews focused on adult respondents (n = 21) whose parents were still engaged in violence. Goldblatt found that respondents felt that they were forced to grow up earlier, and that they felt ambivalence toward, as well as isolation, from their parents. Respondents, nevertheless, noted some upsides of these negative experiences in that they felt that witnessing parental violence gave them a unique perspective on assessing future relationships to avoid potentially violent partners. In addition to witnessing violence, the experience of harsh parenting has consequences for emerging adults. Maas, Herrenkohl, and Sousa (2008) reviewed 12 recent studies on the effect of child maltreatment on teenagers and emerging adults including their involvement in intimate partner violence. They concluded that harsh parenting consistently predicted subsequent intimate partner violence. Jouriles, Mueller, Rosenfield, McDonald, and Dodson (2012) found that both witnessing parental violence and experiencing corporal punishment as a teenager predicted severe teen dating violence. Other studies (e.g., Ehrensaft et al., 2003; Herrenkohl et al., 2004) have also concluded that both parental violence and harsh parenting are associated with the intergenerational transmission of violence.

Family and Peer Influences on IPV during Emerging Adulthood

5

Peer Influences Even when behavior is learned at home, there are other factors that contribute to its actual imitation at later stages of the life course. Deviant peer associations may exacerbate aggressive behaviors witnessed and experienced at home. Scholars (e.g., Booth, Farrell, & Varano, 2008; Lonardo, Giordano, Longmore, & Manning, 2009; Olds, Thombs, & Tomasek, 2005; Piquero, Gover, MacDonald, & Piquero, 2005; Thornberry, Lizotte, Krohn, Farnworth, & Jang, 1994; Williams & Guerra, 2007) have theorized and empirically examined ways in which peers influence various anti-social and criminal activities. Recently, researchers have examined both parental violence and peer dating violence as direct models for teen dating violence. Evaluating a dating violence prevention program for adolescents, ages 12 17, Arriaga and Foshee (2004) found among the untreated portion of the sample (n = 526) that both peer dating violence and parental violence significantly increased teens’ victimization and perpetration of intimate partner violence, but that only peers’ dating violence was associated with respondents’ dating violence six months later. There are several limitations of this study, however, that we attempted to overcome in the current study, such as the use of a single item to examine peers’ relationship violence, and assessing intimate partner violence among a younger age group, although emerging adulthood is associated with higher rates of intimate partner violence (e.g., Breiding et al., 2008; Halpern et al., 2009; Thompson et al., 2006). Moreover, this study did not take into account characteristics of the relationship in which the violence occurred. In addition to peers’ behaviors, some prior studies have examined peers’ support and attitudes toward violence. Silverman and Williamson (1997), examining college men (n = 193), mean age 21, reported that controlling for witnessing violence between parents, male peers’ support for the use of violence had a direct effect on men’s perpetration of relationship violence. DeKeseredy and Kelly (1993) analyzed peers’ support for violence in their study of Canadian college men (n = 1,037) who were asked about advice received from male peers that encouraged sexual, physical, and psychological assaults on dating partners. They found a significant correlation between peer support and men’s views regarding abuse of female dating partners. Rosen, Kaminski, Parmley, Knudson, and Fancher (2003) found that among married male soldiers (n = 713) a normative climate that included conversations that degraded women significantly increased the frequency of intimate partner violence. Moreover, group recognition of respect and needs of spouses, importance of family

6

MALLORY D. MINTER ET AL.

life, and specific attitudes of fidelity and non-violence toward partners, significantly reduced the severity and frequency of perpetrating intimate partner violence. Taken as a whole, these studies highlight significant relationships between peers’ views and attitudes regarding the use of partner violence and their influences on individuals’ own intimate partner violence by accepting and normalizing this specific violent behavior. However, because many prior studies examined adolescent peer influences and the data sets either excluded women and adults, or utilized samples that are too specific for generalization, such as college or military samples, there is a gap in the empirical research. While building on these studies, we sought to overcome these limitations. In the current study, we accounted for the influence of family violence experiences, as well as current peer relationship violence on young adults’ intimate partner violence. Few prior studies have examined these issues among individuals who are older than age 21 so this study allows us to develop a more comprehensive portrait of the correlates of emerging adults’ intimate violence perpetration.

CURRENT STUDY The current study examined the extent to which exposure to family violence as well as current peers’ intimate partner violence and attitudes toward violence influenced emerging adults’ self-reports of intimate partner violence perpetration. This study extended prior literature in several ways. Prior research that has examined family violence and peers’ intimate partner violence behavior and attitudes is limited in using single questions or only examining female victimization. Recognizing that peers may use a range of violent tactics and approve of partner violence for different reasons, we assessed these influences with multiple questions among men and women. This study is based on a community sample of emerging adults, therefore making it more generalizable and relevant to the age group associated with high rates of intimate partner violence. In the multivariate models we included relationship, sociodemographic and family of origin characteristics, which have been found in prior studies to also influence intimate partner violence. Relationship characteristics including union status influence intimate partner violence with cohabiting individuals reporting higher odds compared with married and dating

Family and Peer Influences on IPV during Emerging Adulthood

7

individuals (Brown & Bulanda, 2008; Cui, Ueno, Gordon, & Fincham, 2013). Further, longer relationship duration was associated with increased odds of intimate partner violence (Kenney & McLanahan, 2006; Rickert, Wiemann, Harrykissoon, Berenson, & Kolb, 2002). We also controlled for the following sociodemographic and background characteristics: gender (Whitaker, Haileyesus, Swahn, & Saltzman, 2007), age (Foster, Hagan, & Brooks-Gunn, 2004; Thompson et al., 2006), race and ethnicity (Martin, Cui, Ueno, & Fincham, 2013), education (Tillyer & Wright, 2014), employment status (Alvira-Hammond, Longmore, Manning, & Giordano, 2014), and individual and peer general deviance (Capaldi, Dishion, Stoolmiller, & Yoerger, 2001; Johnson et al., 2015; Ramirez, Paik, Sanchargin, & Heimer, 2012). Family of origin factors included household structure (Brown & Bulanda, 2008; Fang & Corso, 2007), and mother’s education, a proxy for social class background (Alvira-Hammond et al., 2014). We expected in multivariate analyses that exposure to family violence and peers’ attitudes and behaviors would positively influence respondents’ self-reports of intimate partner violence perpetration net of the other variables. We then assessed whether the effects of the peer variables were stronger than the family variables and whether effects were stronger for women based on prior research (e.g., Foshee, Linder, MacDougall, & Bangdiwala, 2001), which found that peers had a greater influence for women than men.

Analytic Plan In Table 1, we presented the means/percentages and ranges for each variable, by total sample, and then for men and women separately, with significant differences indicated. To assess the association between family violence indicators, peers’ intimate partner violence indicators, and respondents’ perpetration of intimate partner violence, in Table 2, we examined the zero-order associations (model 1). We examined models that included the two indicators of family violence (model 2), and then the association between the two indicators of peers’ intimate partner violence behavior and attitudes (model 3) using ordinary least squares (OLS) regression. Each of these models included the control variables. Model 4, the full model, included all of the variables. We also tested interactions between gender and peers’ intimate partner violence and gender and peers’ attitudes towards intimate partner violence.

Table 1. Descriptive Statistics for Emerging Adults’ Intimate Partner Violence, Family Violence, Peers’ Intimate Partner Violence Behaviors and Attitudes, and Control Variables (n = 928). Mean/Percentages Total sample Dependent variable Perpetration frequency Independent variables Family violence Harsh parentinga Witnessing parental violence Peers’ behaviors and attitudes Peers’ partner violence Peers’ attitudes toward partner violence Relationship characteristics Union status (dating) Cohabiting Married Duration Sociodemographic/background characteristics Female Age Race/ethnicity (Anglo)a Black Hispanic Respondent education (high school) Less than high school Some college College or more Employment status (full-time) Part-time Unemployed Respondents’ prior intimate partner violence Respondents’ general deviance Peers’ general deviance Family of origin factorsb Family structure (two biological) Single parents Step parent Other Mother’s education (high school) Less than high school Some college College or more

Males

Ranges Females

0.61

0.45*

0.71

45.5% 31.3%

42.4%† 31.3%

48.0% 31.2%

0 12

1.47 2.26

1.46 2.08***

1.47 2.40

1 5 1 5

44.4% 32.3% 23.3% 3.43

46.2% 34.8% 18.9%** 3.19*

42.9% 30.2% 26.9% 3.62

0.5 14

54.6% 25.43 66.9% 21.3% 11.3% 18.4% 8.2% 38.7% 34.7% 55.9% 19.2% 24.9% 55.6% 0.28 0.74

25.47 66.6% 21.3% 11.4% 22.3%** 8.5% 37.4% 31.8%† 63.0%*** 15.2%** 21.8%* 61.1%** 0.38*** 0.97***

25.39 67.2% 21.3% 11.3% 15.2% 7.9% 39.7% 37.2% 50.0% 22.5% 27.5% 51.0% 0.20 0.54

53.5% 21.0% 13.6% 12.0% 32.4% 10.8% 33.4% 23.4%

58.1%* 19.9% 12.3% 9.7%† 32.5% 9.7% 33.4% 24.4%

49.6% 21.9% 14.6% 13.8% 32.4% 11.7% 33.4% 22.5%

22 29

0 3.4 0 5.6

Source: Toledo Adolescent Relationships Study (2011/2012). a Constructed from initial interview, respondent questionnaire. b Variables constructed from initial interview, parent questionnaire; remaining variables constructed from fifth interview of interviewing; reference categories in parentheses. †p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001.

Model 1 (Zero Order) b 0.12** 0.33***

(0.04) (0.04)

0.23*** 0.19***

(0.03) (0.02)

b 0.06 0.20***

Model 3 (SE)

b

Model 4 (SE)

(0.04) (0.04)

b

(SE)

0.06 0.16***

(0.04) (0.04)

0.11*** 0.07**

(0.03) (0.02)

0.10*** 0.05*

(0.03) (0.02)

0.10** −0.03 0.02**

(0.04) (0.04) (0.01)

0.06 −0.01 0.02**

(0.04) (0.06) (0.01)

0.07 0.01 0.02**

(0.04) (0.06) (0.01)

0.07 0.01 0.02**

(0.04) (0.06) (0.01)

0.11** −0.01

(0.04) (0.01)

0.17*** −0.01

(0.04) (0.01)

0.14*** −0.01

(0.04) (0.01)

0.14*** −0.01

(0.04) (0.01)

0.19*** 0.10†

(0.05) (0.06)

0.07 −0.01

(0.05) (0.06)

0.03 −0.00

(0.05) (0.06)

0.02 −0.02

(0.05) (0.06)

0.19** 0.04 −0.21***

(0.07) (0.04) (0.04)

−0.07 −0.07 −0.09

(0.07) (0.05) (0.06)

−0.09 −0.06 −0.07

(0.17) (0.05) (0.06)

−0.08 −0.05 −0.05

(0.07) (0.05) (0.06)

0.05 0.20*** 0.24***

(0.05) (0.04) (0.04)

0.04 0.12** 0.15***

(0.05) (0.04) (0.04)

0.03 0.10* 0.15***

(0.05) (0.04) (0.04)

0.03 0.10* 0.14***

(0.05) (0.04) (0.04)

9

Family violence Harsh parentinga Witnessing parental violence Peers’ behaviors and attitudes Peers’ partner violence Peers’ attitudes toward partner violence Relationship characteristics Romantic relationship (dating) Cohabiting Married Duration Sociodemographic/background characteristics Female Age Race/ethnicity (Anglo)a Black Hispanic Respondent education (high school) Less than high school Some college College or more Employment status (full-time) Part-time Unemployed Respondents’ prior intimate partner violence

(SE)

Model 2

Family and Peer Influences on IPV during Emerging Adulthood

Table 2. Coefficients and Standard Errors for the OLS Regression of Family Violence, and Peers’ Intimate Partner Violence Behaviors and Attitudes on Frequency of Perpetrating Intimate Partner Violence (n = 928).

Model 1 (Zero Order) b Respondents’ general deviance Peers’ general deviance Family of origin factorsb Family structure (two biological) Single parents Step parent Other Mother’s education (high school) Less than high school Some college College or more R2

(SE)

10

Table 2. (Continued ) Model 2 b 0.10* 0.11***

Model 3

Model 4

(SE)

b

(SE)

b

(SE)

(0.05) (0.03)

0.12** 0.08**

(0.05) (0.03)

0.11* 0.08**

(0.05) (0.03)

0.19*** 0.16***

(0.04) (0.02)

0.19*** 0.04 0.12*

(0.05) (0.05) (0.06)

0.07 −0.02 0.02

(0.05) (0.05) (0.06)

0.09† 0.01 0.05

(0.05) (0.05) (0.06)

0.07 −0.01 0.04

(0.05) (0.05) (0.06)

0.18*** 0.04 −0.16***

(0.06) (0.04) (0.04)

0.04 0.06 0.01 0.20

(0.06) (0.04) (0.05)

0.04 0.04 0.01 0.21

(0.06) (0.04) (0.05)

0.02 0.05 0.01 0.22

(0.06) (0.04) (0.05)

MALLORY D. MINTER ET AL.

Source: Toledo Adolescent Relationships Study (2011/2012). a Constructed from initial interview, respondent questionnaire. b Variables constructed from initial interview, parent questionnaire; remaining variables constructed from fifth interview of interviewing; reference categories in parentheses. †p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001.

Family and Peer Influences on IPV during Emerging Adulthood

11

METHODS The data, from the Toledo Adolescent Relationships Study (TARS), were based on a stratified random sample derived from enrollment records in the 2000 academic year from Lucas County, Ohio adolescents in 7th, 9th, and 11th grades. There were five interviews of data collected, beginning in 2001, and in 2002, 2004, and 2006, with the fifth interview recently collected in 2012. Additionally, in the 2001 interview we collected information directly from respondents’ parents (primarily mothers) or caretakers. The original sample (n = 1,321), devised by the National Opinion Research Center, encompassed 62 schools across seven different school districts. Students did not have to attend school to be included in the sample. Data from the 2010 U.S. Census indicated that Lucas County has similar socioeconomic and demographic characteristics compared to national averages with respect to marital status, income, educational levels, and racial distribution. The TARS sample included oversamples of Black and Hispanic respondents. Respondents at the time of the fifth interview ranged from age 22 to 29, with a mean age 25. The dataset is well suited for the study since it included multiple measures of respondents’ perceptions of peers’ involvement in intimate partner violence and attitudes toward violence and indicators of family violence. At the fifth interview there were 1,021 respondents or 77.3% of the initial interview. The analytic sample included all respondents who participated in the fifth interview, but excluded those who reported that they have never dated (n = 71). We excluded those who reported their race/ethnicity to be other than non-Hispanic Anglo, non-Hispanic Black and Hispanic due to small sample size (n = 22). The final analytic sample was composed of 928 respondents.

Measures Dependent Variable Intimate partner violence perpetration measured at the time of the fifth interview included likert responses to 12 items from the Revised Conflict Tactics Scale (CTS2) (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). These included how often the respondent had done the following: (1) “thrown something at”; (2) “twisted arm or hair”; (3) “used a knife or gun”; (4) “punched or hit with something that could hurt”; (5) “choked”; (6) “slammed against a wall”; (7) “beat up”; (8) “burned or scalded on

12

MALLORY D. MINTER ET AL.

purpose”; (9) “kicked”; (10) “pushed, shoved, or grabbed”; (11) “slapped in the face or head with an open hand”; and (12) “hit” (α = .99). Responses ranged from “never” to “very often.” We analyzed partner violence as a continuous variable, which was logged (due to skewness) summation of respondents’ perpetration acts; the non-logged measure ranged from 0 to 12. About 23.3% of respondents reported any physical violence, while 15.5% reported perpetrating violence. These percentages were consistent with many prior studies and reviews of the literature based on community samples (e.g., Capaldi & Langhinrichsen-Rohling, 2012), which have found that much violence among young adults is reciprocal or mutual; thus, individuals are both victims as well as perpetrators. In analyses in which we combined the frequency of perpetration and victimization to reflect mutual or reciprocal violence (available from authors), the multivariate findings were similar.

Family Violence We measured harsh parenting at the time of the first interview when respondents were adolescents. This was a dichotomous measure composed of responses to the following two questions: “When you and your parents disagree about things, how often do they do the following:” “call you names or insult you”; “push, slap, or hit you.” Affirmative responses to either of these questions were coded as 1 (46%) (α = .67). We measured witnessing parental violence (Straus et al., 1996) using retrospective responses, from the fifth interview, in which respondents were asked, “How often did either one of your parents:” “throw something at the other”; “push, shove, or grab the other”; “slap the other in the face or head with an open hand”; and “hit the other.” We dichotomized scores with affirmative responses coded as 1 (31%) (α = .95).1

Peers’ Behaviors and Attitudes Peers’ intimate partner violence referred to responses at the fifth interview when individuals were asked to think about their friends over the past 12 months and answer the following questions from the abbreviated version of the Revised Conflict Tactics Scale, “How often did your friends and their romantic partners:” “throw something at each other”; “push, shove, or grab each other”; “slap each other in the face or head with an open

Family and Peer Influences on IPV during Emerging Adulthood

13

hand”; and “hit each other.” Responses ranged from “never” to “very often.” We calculated the mean score with imputed responses for respondents missing on each variable (1.19%; n = 11), and results ranged from 1 to 5. Higher scores represented greater frequency of violence (α = .94). We measured peers’ attitudes toward intimate partner violence with the following statement: “Regardless of how you feel about it, how strongly would you say your friends agree or disagree with the following reasons why it might be understandable that someone could hit a partner.” The four reasons included the following: (1) “A partner hit them first”; (2) “They catch a partner cheating on them”; (3) “A partner embarrasses or belittles them in front of others”; and (4) “A partner continually nags them.” Responses ranged from “strongly disagree” to “strongly agree” with higher scores indicating greater agreement with reasons for participating in intimate partner violence. This scale is a mean score with imputed responses for respondents missing on each variable (0.32%; n = 3) and ranged from 1 to 5 (α = .85).

Relationship Characteristics Union status included dichotomous indicators for dating (44.4%), cohabiting (32.3%), and married (23.3%), with dating as the reference category in the multivariate analyses. Duration referred to how long respondents had been with their current or most recent partners, measured in years with mean imputation for missing data, and a mean of 3.43 and a range of 0.5 14.

Family of Origin Characteristics Family structure during adolescence, from the first interview asked, “During the past 12 months, who were you living with most of the time?” Respondents selected one of 25 categories, which we collapsed into four categories: two biological parents (reference group) (53.45%), single parent (21.01%), stepparents (13.58%), or “other family” including living with other family members or foster care (11.96%). Mother’s education from the parent’s questionnaire at the time of the first interview, was a proxy for socioeconomic background, and responses included less than high school (10.78%), high school (32.44%), some college (33.41%), and college graduate (23.38%), with high school as the reference in multivariate analyses.

14

MALLORY D. MINTER ET AL.

Sociodemographic and Background Characteristics Female, a dichotomous variable, indicated whether the respondent was female (54.53%). Age (mean = 25.42) was the difference between date of birth and the fifth interview date. Race/ethnicity consisted of three selfreported categories: Anglo (reference group) (66.92%), Black (21.34%), and Hispanic (11.31%). Respondent’s education included less than high school (8.19%), high school graduate (reference group) (18.43%), some college (38.69%), or college or more (34.70%). Employment status was measured as full-time (reference group) (55.93%), part-time (19.18%), and unemployed (24.89%). Respondents’ prior intimate partner violence, measured from all previous interviews assessed whether the respondent had any involvement in intimate partner violence. Each prior interview (1 4) included a short version of the CTS2, which asked how often the respondent and partner had done the following: “thrown something at”; “punched or hit with something that could hurt”; “pushed, shoved, or grabbed”; or “hit.” We dichotomized scores with affirmative responses coded as 1 (55.6%) to indicate any violence in previous relationships (α = .91). Respondents’ general deviance (Elliott & Ageton, 1980) measured at the fifth interview asked respondents, “In the last two years (or 24 months), how often have you: stolen (or tried to steal) things worth $5 or less; damaged or destroyed property on purpose; carried a hidden weapon other than a plain pocket knife; stolen (or tried to steal) something worth more than $50; attacked someone with the idea of seriously hurting him/her; sold drugs; been drunk in a public place; broken into a building or vehicle (or tried to break in) to steal something or just to look around; and used drugs to get high (not because you were sick).” We calculated a mean score with the mean imputed for missing responses (0.54%; n = 5). Scores ranged from 0 to 3.4 (α = .73). Peers’ general deviance asked respondents the same items as those in the respondent general deviance scale. We calculated a mean score with imputed responses for missing information on each variable (0.22%; n = 2). Scores ranged from 0 to 5.6 (α = .82).

RESULTS Descriptive Analysis The average level of intimate partner violence perpetration based on the CTS2 was 1.54 (range = 0 12), with a standard deviation of 3.85 (not

Family and Peer Influences on IPV during Emerging Adulthood

15

shown) indicating relatively low levels of partner violence. Moreover, 23.3% (n = 216) of respondents reported any relationship violence. About 15.5% (n = 144) of the total sample reported perpetrating violence, and 20.5% (n = 190) reported being victimized by their current or most recent partner. Consistent with prior literature (e.g., Alvira-Hammond et al., 2014; Jennings, Reingle, Staras, & Maldonado-Molina, 2012), there was much overlap as 54.6% (n = 118) reported perpetration and victimization (not shown), indicating that the majority of relationship violence reported was mutual. In Table 1, we examined descriptive statistics and whether there were significant gender differences. Women compared with men reported higher rates of violence perpetration (0.74 compared to 0.45, respectively), which is consistent with other studies based on community samples (e.g., Whitaker et al., 2007). Regarding family violence there was a marginally significant difference between men and women who experienced harsh parenting; women compared to men reported slightly higher levels (48.0% compared to 42.4%, respectively). In contrast, witnessing parental violence was reported at similar levels among men and women. With regard to peers’ influence, there were no significant gender differences between peers’ intimate partner violence behaviors (mean score of 1.47 on a five point scale). Women, on average, (women = 2.40, men = 2.08) had friends who were significantly more likely to report supportive attitudes toward violence. In terms of peers anti-social or deviant behavior, men, on average (women = 0.54; men = 0.97), reported higher mean levels.

Predicting Intimate Partner Violence In Table 2, at the zero order, witnessing parental violence, experiencing harsh parenting, peers’ intimate partner violence, and peers’ attitudes supportive of partner violence were positively associated with respondents’ greater frequency of intimate partner violence perpetration. Cohabiting and relationships of longer duration were associated with higher frequency of intimate partner violence perpetration. Black and Hispanic compared with Anglo respondents were significantly more likely to report intimate partner violence perpetration. Respondents’ lower educational level was positively associated, and higher educational levels were negatively associated with intimate partner violence. Respondents who were unemployed, reported prior intimate partner violence, as well as general deviance and peers’ general deviance were significantly more likely to report intimate partner violence perpetration. Growing up in single parent households and

16

MALLORY D. MINTER ET AL.

“other” households (e.g., foster care, relatives) were associated with perpetrating intimate partner violence. Mothers’ self-reports of having less than a high school education was positively associated, and having a college education was negatively associated with frequency of intimate partner violence perpetration. Model 2 examined the influence of family violence indicators on respondents’ intimate partner violence perpetration net of the other variables. Although harsh parenting was not significantly related to violence perpetration, witnessing parental violence was associated with respondents’ intimate partner violence perpetration. Model 3 included the influence of peers’ participation in intimate partner violence and peers’ attitudes about the use of partner violence under certain circumstances, net of the control variables. Peers’ participation in intimate partner violence was associated with respondents’ intimate partner violence perpetration. Additionally, peers’ attitudes accepting of the use of intimate partner violence were significantly related to respondents’ frequency of intimate partner violence perpetration. The full model (model 4) estimated the influence of all of the variables. Witnessing parental violence during adolescence was related to greater frequency of intimate partner violence perpetration. Peers’ intimate partner violence behaviors and attitudes continued to be significantly related to respondents’ frequency of intimate partner violence perpetration net of control variables. Regarding the control variables, relationships of longer duration were associated with greater frequency of intimate partner violence perpetration. Female respondents continued to be significantly related to violence perpetration. Unemployment, compared to full-time employment, was significantly related to violence perpetration. Respondents’ prior intimate partner violence, general deviance, and peers’ general deviance were associated with intimate partner violence perpetration. We examined a set of interaction models that included the moderating effects of gender with each key peer variable on respondents’ intimate partner violence (available from authors). We did not find a significant moderating relationship between gender and peers’ intimate partner violence, or gender and peers’ attitudes toward partner violence. Thus, the influences of peers’ intimate partner violence, and peers’ attitudes toward partner violence were similar for men and women. Additionally, we did not find a significant moderating relationship between gender and harsh parenting, and gender and witnessing parental violence. Therefore, the influences of harsh parenting and witnessing parental violence on intimate partner violence were similar for men and women. Supplemental analyses (available from authors) found that compared to those with low peers’ involvement in

Family and Peer Influences on IPV during Emerging Adulthood

17

intimate partner violence, respondents with high peer involvement, combined with witnessing parental violence or experiencing harsh parenting in adolescence was significantly related to respondents’ intimate partner violence.

DISCUSSION Many researchers have reported that rates of mutual intimate partner violence peak during the life stage of emerging adulthood. Much research has examined the influence of family and peers during the life stage of adolescence (e.g., Arriaga & Foshee, 2004; Haynie & Osgood, 2005). Yet, scholars have often theorized that peers’ influence eclipses parents’ influence as individuals’ transition to young adulthood. Although researchers have found that peers’ attitudes and behaviors significantly affect older teens’ and young adults’ behavior with respect to substance use including alcohol, cigarettes, and various drugs (Andrews, Tildesley, Hops, & Li, 2002; Gardner & Steinberg, 2005), intimate partner violence may function differently than general deviance and criminal activity (Moffitt, Krueger, Caspi, & Fagan, 2000). This current study established that peers’ intimate partner violence behaviors, and attitudes regarding those behaviors, were associated with individuals’ intimate partner violence in young adulthood. We found that these relationships were generally similar for men and women. In addition, we found that when individuals have prior histories of parental violence, whether harsh parenting or witnessing parental violence, having peers’ who were highly involved in relationship violence in emerging adulthood strongly influenced an individual’s own intimate partner violence perpetration. Combined, parental violence and peers’ involvement in relationship violence can be particularly detrimental in emerging adulthood. The association between current peers’ behaviors and attitudes regarding intimate partner violence, and emerging adults’ intimate partner violence remains significant. While family of origin factors are important, the learning process of this violence obviously does not cease. Although the influence of peers may be lessened in emerging adulthood with regard to general deviance and violent crimes, they can continue to create a normative climate of violence among intimate partners into young adulthood. Although this paper addresses an important issue regarding the continued influence of peers and prior family violence, there are a few limitations.

18

MALLORY D. MINTER ET AL.

First, it is cross-sectional and only captures one time period. Yet, peer characteristics during adolescence have also been shown to have a lasting effect on adult behaviors, particularly with regard to intimate partner violence (Capaldi et al., 2001). Our future work will take a more comprehensive longitudinal examination of these connections, as well as the continued influence of peers further into adulthood. Additionally, there may be selection effects in which individuals who perpetrate intimate partner violence seek out friends who behave similarly and who support the use of violence. Third, this study did not incorporate other peer factors that have been identified as influential on intimate partner violence such as peer group characteristics, gender composition and the number of peers involved in delinquent acts (e.g., Casey & Beadnell, 2010; Connolly, Furman, & Konarski, 2000; Ramirez et al., 2012). Fourth, although the distributions of race, gender and familial background characteristics are comparable to national samples, this study was not itself nationally representative. Our next research step is to examine how family violence and peers’ attitudes and behaviors regarding intimate partner violence influence other aspects of the individual’s life, such as pro-violent attitudes as well as experiencing depressive symptoms, both of which are likely correlates of intimate partner violence. Further examination of the influences of family and peer normative climates should consider the community level, such as community norms that support family violence and intimate partner violence. We would expect that such norms would contribute to a normalizing climate where individual intimate partner violence is more likely to occur (Raghavan, Rajah, Gentile, Collado, & Kavanagh, 2009). Additionally, our future investigations will take a more in-depth look at the processes behind these peer influences in emerging adulthood, including qualitative aspects of this relationship, particularly with regard to gender differences. As Arriaga and Foshee (2004) noted, it may be the witnessing of peers as solely perpetrators or victims of this violence that causes gender differentiations, which this study could not distinguish. Despite these limitations, this study showed that after controlling for various sociodemographic, familial background and relationship characteristics, parents and peers’ partner violence significantly influenced respondents’ conduct within their own intimate relationships. Although peers’ influence on general deviance has been shown to decrease as individuals grow older (e.g., Monahan, Steinberg, & Cauffman, 2009), this study demonstrated that this is not necessarily the case with regard to intimate partner violence and that peers’ current involvement in this specific violent behavior can influence respondent’s own involvement. These findings are potentially useful in designing programs

Family and Peer Influences on IPV during Emerging Adulthood

19

and policies in the prevention or cessation of intimate partner violence and the role that families and friends have in an individual’s life. For example, although a person cannot control their parents’ or friends’ violence, they can recognize it and their reasoning for the use of violence, but chose not to accept it. Individuals can then remain non-violent themselves, and become better models for their own peers.

NOTE 1. Additional analyses were done with Witnessing Parental Violence designed as an interval variable and results had similar outcomes.

ACKNOWLEDGMENTS This research received support from The Eunice Kennedy Shriver National Institute of Child Health and Human Development (HD036223 and HD044206), the Department of Health and Human Services (5APRP A006009), the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice (Award Nos. 2009-IJ-CX-0503 and 2010-MUMU-0031), and the Center for Family and Demographic Research, Bowling Green State University, which received core funding from The Eunice Kennedy Shriver National Institute of Child Health and Human Development (R24HD050959). The opinions, findings, conclusions, and recommendations expressed in this publication/program/exhibition are those of the authors and do not necessarily reflect the official views of the Department of Justice or National Institutes of Health.

REFERENCES Alvira-Hammond, M., Longmore, M. A., Manning, W. D., & Giordano, P. C. (2014). Gainful activity and intimate partner aggression in emerging adulthood. Emerging Adulthood, 2(2), 116 127. Andrews, J. A., Tildesley, E., Hops, H., & Li, F. (2002). The influence of peers on young adult substance use. Health Psychology, 21(4), 349 357. Arriaga, X. B., & Foshee, V. A. (2004). Adolescent dating violence: Do adolescents follow in their friends’, or their parents’, footsteps? Journal of Interpersonal Violence, 19(2), 162 184.

20

MALLORY D. MINTER ET AL.

Band-Winterstein, T. (2014). The impact of lifelong exposure to IPV on adult children and their aging parents. Journal of Family Issues, 35(4), 439 461. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall. Booth, J. A., Farrell, A., & Varano, S. P. (2008). Social control, serious delinquency, and risky behavior: A gendered analysis. Crime and Delinquency, 54(3), 423 456. Breiding, M. J., Black, M. C., & Ryan, G. W. (2008). Prevalence and risk factors of intimate partner violence in eighteen U.S. states/territories, 2005. American Journal of Preventative Medicine, 34(2), 112 118. Brown, S. L., & Bulanda, J. R. (2008). Relationship violence in young adulthood: A comparison of daters, cohabitors, and marrieds. Social Science Research, 37, 73 87. Burke, P., Stets, J., & Pirog-Good, M. (1988). Gender identity, self-esteem, and physical and sexual abuse in dating relationships. Social Psychology Quarterly, 51, 272 285. Capaldi, D. M., Dishion, T. J., Stoolmiller, M., & Yoerger, K. (2001). Aggression toward female partners by at-risk young men: The contribution of male adolescent friendships. Developmental Psychology, 37(1), 61 73. Capaldi, D. M., & Langhinrichsen-Rohling, J. (2012). Informing intimate partner violence prevention efforts: Dyadic, developmental, and contextual considerations. Prevention Science, 13(4), 323 328. Casey, E. A., & Beadnell, B. (2010). The structure of male adolescent peer networks and risk for intimate partner violence perpetration: Findings from a national sample. Journal of Youth and Adolescence, 39, 620 633. Connolly, J., Furman, W., & Konarski, R. (2000). The role of peers in the emergence of heterosexual romantic relationships in adolescence. Child Development, 71(5), 1395 1408. Cui, M., Ueno, K., Gordon, M., & Fincham, F. D. (2013). The continuation of intimate partner violence from adolescence to young adulthood. Journal of Marriage and Family, 75(2), 300 313. DeKeseredy, W. S., & Kelly, K. (1993). Woman abuse in university and college dating relationships: The contribution of the ideology of familial patriarchy. Journal of Human Justice, 4(2), 25 52. Ehrensaft, M. K., Cohen, P., Brown, J., Smailes, E., Chen, H., & Johnson, J. G. (2003). Intergenerational transmission of partner violence: A 20-year prospective study. Journal of Consulting and Clinical Psychology, 71(4), 741 753. Elliott, D. S., & Ageton, S. S. (1980). Reconciling race and class differences in self-reported and official estimates of delinquency. American Sociological Review, 45, 95 110. Fang, X., & Corso, P. S. (2007). Child maltreatment, youth violence, and intimate partner violence: Developmental relationships. American Journal of Preventive Medicine, 33(4), 281 290. Foshee, V. A., Linder, F., MacDougall, J. E., & Bangdiwala, S. (2001). Gender differences in the longitudinal predictors of adolescent dating violence. Preventive Medicine, 32, 128 141. Foster, H., Hagan, J., & Brooks-Gunn, J. (2004). Age, puberty, and exposure to intimate partner violence in adolescence. Annals of New York Academy of Sciences, 1036, 151 166. Franklin, C. A., & Kercher, G. A. (2012). The intergenerational transmission of intimate partner violence: Differentiating correlates in a random community sample. Journal of Family Violence, 27(3), 187 199.

Family and Peer Influences on IPV during Emerging Adulthood

21

Gardner, M., & Steinberg, L. (2005). Peer influence on risk taking, risk preference, and risky decision making in adolescence and adulthood: An experimental study. Developmental Psychology, 41(4), 625 635. Goldblatt, H. (2003). Strategies of coping among adolescents experiencing interparental violence. Journal of Interpersonal Violence, 18(5), 532 552. Halpern, C. T., Spriggs, A. L., Martin, S. L., & Kupper, L. (2009). Patterns of intimate partner violence victimization from adolescence to young adulthood in a nationally representative sample. Journal of Adolescent Health, 45(5), 508 516. Haynie, D. L., & Osgood, D. W. (2005). Reconsidering peers and delinquency: How do peers matter? Social Forces, 84(2), 1109 1130. Herrenkohl, T. I., Mason, W. A., Kosterman, R., Lengua, L. J., Hawkins, J. D., & Abbott, R. D. (2004). Pathways from physical childhood abuse to partner violence in young adulthood. Violence and Victims, 19(2), 123 136. Jennings, W. G., Reingle, J. M., Staras, S. A., & Maldonado-Molina, M. M. (2012). Substance use as a risk factor for intimate partner violence overlap: Generational differences among Hispanic young adults. International Criminal Justice Review, 22(2), 139 152. Johnson, W. L., Giordano, P. C., Manning, W. D., & Longmore, M. A. (2015). The age IPV curve: Changes in the perpetration of intimate partner violence during adolescence and young adulthood. Journal of Youth and Adolescence, 44(3), 708 725. Jouriles, E. N., Mueller, V., Rosenfield, D., McDonald, R., & Dodson, M. C. (2012). Teens’ experiences of harsh parenting and exposure to severe intimate partner violence: Adding insult to injury in predicting teen dating violence. Psychology of Violence, 2(2), 125 138. Kenney, C. T., & McLanahan, S. S. (2006). Why are cohabiting relationships more violent than marriages? Demography, 43(1), 127 140. Lonardo, R. A., Giordano, P. C., Longmore, M. A., & Manning, W. D. (2009). Parents, friends, and romantic partners: Enmeshment in deviant networks and adolescent delinquency involvement. Journal of Youth and Adolescence, 38, 367 383. Maas, C., Herrenkohl, T. I., & Sousa, C. (2008). Review of research on child maltreatment and violence in youth. Trauma, Violence, and Abuse, 9(1), 56 67. Martin, B., Cui, M., Ueno, K., & Fincham, F. D. (2013). Intimate partner violence in interracial and monoracial couples. Family Relations, 62, 202 211. Moffitt, T. E., Krueger, R. F., Caspi, A., & Fagan, J. (2000). Partner abuse and general crime: How are they the same? How are they different? Criminology, 38(1), 199 232. Monahan, K. C., Steinberg, L., & Cauffman, E. (2009). Affiliation with anti-social peers, susceptibility to peer influence, and anti-social behavior during the transition to adulthood. Developmental Psychology, 45(6), 1520 1530. Olds, S. R., Thombs, D. L., & Tomasek, J. R. (2005). Relations between normative beliefs and initiation intentions toward cigarette, alcohol and marijuana. Journal of Adolescent Health, 37, 75.e7 75.e13. Piquero, N. L., Gover, A. R., MacDonald, J. M., & Piquero, A. R. (2005). The influence of delinquent peers on delinquency: Does gender matter? Youth and Society, 36(3), 251 275. Raghavan, C., Rajah, V., Gentile, K., Collado, L., & Kavanagh, A. M. (2009). Community violence, social support networks, ethnic group differences, and male perpetration of intimate partner violence. Journal of Interpersonal Violence, 24(10), 1615 1632.

22

MALLORY D. MINTER ET AL.

Ramirez, M., Paik, A., Sanchargin, K., & Heimer, K. (2012). Violent peers, network centrality, and intimate partner violence perpetration by young men. Journal of Adolescent Health, 51, 503 509. Rickert, V. I., Wiemann, C. M., Harrykissoon, S. D., Berenson, A. B., & Kolb, E. (2002). The relationship among demographics, reproductive characteristics, and intimate partner violence. American Journal of Obstetrics & Gynecology, 187(4), 1002 1007. Rosen, L. H., Kaminski, R. J., Parmley, A. M., Knudson, K. H., & Fancher, P. (2003). The effects of peer group climate on intimate partner violence among married male U.S. army soldiers. Violence against Women, 9(9), 1045 1071. Silverman, J. G., & Williamson, G. M. (1997). Social ecology and entitlements involved in battering by heterosexual college males: Contribution of family and peers. Violence and Victims, 12(2), 147 164. Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised conflict tactics scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17(3), 283 316. Thompson, R. S., Bonomi, A. E., Anderson, M., Reid, R. J., Dimer, J. A., Carrell, D., & Rivara, F. P. (2006). Intimate partner violence: Prevalence, types, and chronicity in adult women. American Journal of Preventative Medicine, 30(6), 447 457. Thornberry, T. P., Lizotte, A. J., Krohn, M. D., Farnworth, M., & Jang, S. J. (1994). Delinquent peers, beliefs, and delinquency behaviors: A longitudinal test of interactional theory. Criminology, 32(1), 47 83. Tillyer, M. S., & Wright, E. M. (2014). Intimate partner violence and the victim-offender overlap. Journal of Research in Crime and Delinquency, 51(1), 29 55. Turcotte-Seabury, C. (2010). Anger management and the process mediating the link between witnessing violence between parents and partner violence. Violence and Victims, 25, 306 318. Whitaker, D., Haileyesus, T., Swahn, M., & Saltzman, L. S. (2007). Differences in frequency of violence and reported injury between relationship with reciprocal and nonreciprocal intimate partner violence. American Journal of Public Health, 97(5), 941 947. Williams, K. R., & Guerra, N. G. (2007). Prevalence and predictors of internet bullying. Journal of Adolescent Health, 41, S14 S21. Williamson, G. M., & Silverman, J. G. (2001). Violence against female partners: Direct and interactive effects of family history, communal orientation, and peer-related variables. Journal of Social and Personal Relationships, 18(4), 535 549.

A LATENT CLASS APPROACH TO UNDERSTANDING THE INTERGENERATIONAL TRANSMISSION OF VIOLENCE IN EMERGING ADULT RELATIONSHIPS Peter M. Rivera and Frank D. Fincham ABSTRACT Purpose Research on the intergenerational transmission of violence has been limited by reliance on variable-oriented methodology that does not capture heterogeneity that exists within experiences of violent interpersonal conduct. The current study therefore examines the utility of a person-oriented statistical method in understanding patterns of maltreatment and intimate partner violence. Approach Guided by person-oriented theory, the current study utilizes latent class analysis, a person-oriented method used with cross-sectional data, to examine the heterogeneity within this transmission process in a sample of emerging adults (N = 150). This study also examined whether

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 23 44 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009002

23

24

PETER M. RIVERA AND FRANK D. FINCHAM

the classes identified differed on reported emotional reactivity and childhood family environment. Findings Three classes emerged from the latent class analysis, labeled full transmission, psychological transmission, and no transmission. Those comprising the full transmission subgroup reported the lowest levels of childhood family cohesion, accord, and closeness. The full transmission subgroup also reported significantly more emotional reactivity than the psychological transmission and no transmission subgroups. Implications To understand fully the etiology of intimate partner violence for maltreated offspring, a multidimensional view of violence is needed. The current study represents a step in this direction by demonstrating the utility of a person-oriented approach in understanding the IGT of violence. Keywords: Intergenerational transmission of violence; maltreatment; intimate partner violence; latent class analysis; person-oriented methods

Maltreated offspring have been shown to be at an increased risk for experiencing violence within their later intimate relationships (e.g., Rivera & Fincham, 2015); the intergenerational transmission (IGT) of violence. Weak-to-moderate support for the IGT of violence hypothesis (Stith et al., 2000) may reflect an overreliance on variable-oriented approaches that do not capture the heterogeneity that exists within experiences of violent interpersonal conduct (Bogat, Levendosky, & Von Eye, 2005; Herrenkohl & Herrenkohl, 2007; Swartout & Swartout, 2012). The failure to adopt a more multidimensional view has hindered our ability to generalize findings across studies, develop more sophisticated theories, and interrupt the transmission of violence across generations. It is therefore essential in advancing our understanding of the IGT of violence to begin utilizing analytical techniques that account for the heterogeneity that may exist within the transmission process, such as person-oriented methods (von Eye & Bogat, 2006). Researchers have yet to utilize these techniques in the study of the IGT of violence. The current study therefore aims to demonstrate the utility of a person-oriented method (i.e., latent class analysis) in understanding patterns of maltreatment and intimate partner violence, while considering the type and severity of reported maltreatment1 and intimate partner violence.

Intergenerational Transmission of Violence

25

INTERGENERATIONAL TRANSMISSION OF VIOLENCE Social Learning theory (SLT; Bandura, Ross, & Ross, 1962) is often utilized in explaining the IGT of violence, which postulates that offspring learn that intimate partner violence is acceptable behavior by witnessing parents model such behavior during conflicts (Mihalic & Elliott, 1997). Prospective and retrospective data provide support for links between varying forms of maltreatment and varying forms of intimate partner violence. Research on emerging adults using longitudinal designs shows that exposure to interparental violence is related to psychological and physical intimate partner violence perpetration and victimization (Cui, Durtschi, Donnellan, Lorenz, & Conger, 2010; Fergusson, Boden, & Horwood, 2006; Smith, Ireland, Park, Elwyn, & Thornberry, 2011). Further, there is evidence that experiences of physical maltreatment and neglect in childhood are related to reported injuries sustained through intimate partner violence (Ehrensaft et al., 2003). However, it should be noted that most research on the IGT of violence has utilized cross-sectional study designs and variable-oriented methods and has provided inconsistent support for relationships among varying forms of experienced maltreatment and intimate partner violence (e.g., Gover, Kaukinen, & Fox, 2008; Rivera & Fincham, 2015; Simons, Simons, Lei, Hancock, & Fincham, 2012).

Gender An element that may have contributed to inconsistencies within IGT of violence research is gender. There is evidence that the IGT of violence may function differently for males and females and is dependent on the gender of the perpetrator and victim of violence. Gender identification models of the IGT of violence have been proposed and empirically supported (e.g., Jankowski, Leitenberg, Henning, & Coffey, 1999) and suggest that the likelihood of violent interpersonal conduct being transmitted across generations is increased when the perpetrator of maltreatment is of the same sex as the offspring. Most recently, Milletich, Kelley, Doane, and Pearson’s (2010) study of 703 emerging adults supported a gender identification model of the IGT of violence. Their findings indicated that the likelihood for perpetrating physical intimate partner violence was

26

PETER M. RIVERA AND FRANK D. FINCHAM

strongest when male and female emerging adults reported a history of exposure to interparental violence perpetrated by the same sex parent. Despite these findings and previous evidence supporting a gender identification model of the IGT of violence, this model has not been consistently supported (e.g., Gover et al., 2008). As such, these findings are far from conclusive and further research on the IGT of violence that considers the role of gender is needed to better understand its impact on the transmission process.

APPROACHES TO STUDYING THE INTERGENERATIONAL TRANSMISSION OF VIOLENCE Focusing on One Form of Violence Despite evidence indicating that maltreatment and intimate partner violence are multidimensional constructs (Bogat et al., 2005; Herrenkohl & Herrenkohl, 2009; Swartout & Swartout, 2012), research on the IGT of violence has overwhelmingly adopted variable-oriented methods and has approached violent interpersonal conduct as a unidimensional phenomenon (e.g., Black, Sussman, & Unger, 2010; Cui et al., 2010; Karakurt, Keiley, & Posada, 2013; Simons et al., 2012). One such approach involves focusing on a particular type of maltreatment and/or intimate partner violence in isolation (e.g., Gay, Harding, Jackson, Burns, & Baker, 2013; Lee, Reese-Weber, & Kahn, 2014; Rosen, Bartle-haring, & Stith, 2001). Even though these efforts have contributed to our understanding of the IGT of violence, by focusing on one type of violence without controlling for others extant research fails to deal with the problem of comorbidity or co-occurring experiences of violence. Offspring who experience maltreatment often experience more than one type (Arata, LanghinrichsenRohling, Bowers, & O’Farrill-Swails, 2005; Berzenski & Yates, 2011; Teicher, Samson, Polcari, & McGreenery, 2006), and individuals who report experiencing intimate partner violence also rarely experience one form alone (Sullivan, McPartland, Armeli, Jaquier, & Tennen, 2012). It is probable that confounding single and co-occurring maltreatment experiences have contributed to current inconsistencies within IGT of violence research.

Intergenerational Transmission of Violence

27

Focusing on the Presence of Violence In addition to focusing on a single type of maltreatment, IGT of violence research has relied on classifying the presence of maltreatment (e.g., Gover et al., 2008; Millett, Kohl, Jonson-Reid, Drake, & Petra, 2013) or intimate partner violence (e.g., Gover et al., 2008; Millett et al., 2013; Whiting, Simmons, Havens, Smith, & Oka, 2009) when testing the IGT of violence hypothesis. Classifying the presence of violent interpersonal conduct overlooks evidence indicating that different types of maltreatment produce different effects on psychosocial health (Hahm, Lee, Ozonoff, & Van Wert, 2010; Teicher et al., 2006) and that co-occurring maltreatment types effects the well-being of children and adolescents differently (Hazen, Connelly, Roesch, Hough, & Landsverk, 2009; Villodas et al., 2012). For example, Berzenski and Yates (2011) identified subgroups of emerging adults based on multiple indicators of maltreatment, and found significant subgroup differences across a number of psychosocial outcomes. This approach of simply focusing on the presence of violence also neglects evidence suggesting that the likelihood of violent conduct being transmitted across generations will depend on the severity of violence experienced (Berzenski, Yates, & Egeland, 2014; Litrownik et al., 2005; Straus & Michel-Smith, 2014). In a study examining the effects of different dimensions of maltreatment (i.e., type, severity, chronicity, and age at first report) on child functioning, English et al. (2005) concluded that the type of maltreatment (indicated by the maximum severity rating for each type) was the most consistent predictor of child outcomes. These findings suggest that particular subgroups of maltreated offspring may be more vulnerable to the IGT of violence, and by considering and assessing all maltreatment types and severity concurrently, IGT of violence researchers may begin to identify shared and distinctive contributions of maltreatment types and severity to the risk of experiencing intimate partner violence.

PERSON-ORIENTED APPROACH TO STUDYING THE INTERGENERATIONAL TRANSMISSION OF VIOLENCE Variable-oriented approaches (e.g., multiple regression) assume that populations are homogenous and that variables operate the same way across

28

PETER M. RIVERA AND FRANK D. FINCHAM

populations. Conversely, person-oriented approaches assume that significant heterogeneity exists within populations and that shared patterns exist within subgroups of populations (Bergman & Magnusson, 1997; von Eye & Bogat, 2006). These assumptions have direct implications for how violent interpersonal conduct is operationalized within IGT of violence research. A person-oriented approach can be segmented into two parts, theory and methods (Bergman & Wa˚ngby, 2014).

Person-Oriented Theory A holistic-interactionistic view (Bergman & Magnusson, 1997) considers an individual to be “an organized whole with elements operating together to achieve a functioning system in a dynamic process with interactions between components” (Bergman & Wa˚ngby, 2014, p. 31). This view leads researchers to consider the many possible constellations of maltreatment and intimate partner violence types that one may have experienced when examining the IGT of violence. The principles of a person-oriented approach were developed upon this holistic-interactionistic perspective (Bergman, 2001; Sterba & Bauer, 2010; von Eye & Bergman, 2003). This perspective includes the view that patterns across variables offer insight into the development of behavior and that a finite number of patterns exist across individuals and populations, some with greater or lesser frequency. For a more detailed overview of the holistic-interactionistic perspective the reader is referred to Bergman and Magnusson (1997).

Person-Oriented Methods Three assumptions guide methodology within the person-oriented approach (von Eye & Bogat, 2006). The first presumes that samples derive from populations made-up of many subgroups, which can be specified or identified. This can be achieved a number of ways, such as through the use of a person-oriented method (e.g., latent class analysis) that decomposes a sample into homogenous subgroups based on observed variables. The second assumption pertains to external validity and assumes that subgroups will significantly vary on at least one variable not used to specify the subgroups. Finally, the third assumption suggests that subgroups must have substantive interpretation and theoretical meaningfulness (von Eye & Bogat, 2006).

Intergenerational Transmission of Violence

29

Latent class analysis is a person-oriented method that may be particularly promising for IGT of violence research that relies on cross-sectional study designs. This analytical method uses finite mixture modeling to empirically identify patterns across individuals based on observed variables that explain the unobserved heterogeneity within the phenomenon under investigation. In other words, latent class analysis uses response patterns of observed variables (e.g., maltreatment types) to assign individuals from a heterogeneous sample to homogenous subgroups (Roesch, Villodas, & Villodas, 2010; von Eye & Bogat, 2006); maximizing the homogeneity within group and heterogeneity between groups (for review see Roesch et al., 2010). Latent class analysis could offer insight into the IGT of violence by uncovering meaningful heterogeneity with the transmission process by not only modeling the heterogeneity that exists within experiences of maltreatment but also experiences of intimate partner violence. However, research has yet to extend latent class analysis to the IGT of violence.

THE CURRENT STUDY The current study extends research employing person-oriented methods to violent interpersonal conduct by applying a person-oriented method to the IGT of violence. Through the use of latent class analysis, the current study aims to identify meaningful subgroups of emerging adults based on histories of maltreatment and intimate partner violence, while considering the type and severity of violence reported. To establish the external validity of subgroups identified in the current study, mean differences in emotional reactivity and childhood family environment are examined. Emotional reactivity seems to be especially salient in understanding conflict within emerging adults’ intimate relationships (Wei, Vogel, Ku, & Zakalik, 2005) and has been demonstrated to be associated with one’s family environment (Gardner, Busby, & Brimhall, 2007) and childhood exposure to interparental violence (McKee & Payne, 2014). As such, emotional reactivity may be especially relevant to understanding increased risk for experiencing intimate partner violence among individuals with maltreatment histories. The following hypotheses were tested: H1. Latent class analysis will identify at least one distinct subgroup characterized by histories of co-occurring maltreatment types.

30

PETER M. RIVERA AND FRANK D. FINCHAM

H2. The class characterized with the highest likelihood of reporting exposure to interparental violence, maltreatment perpetrated by either parent, and intimate partner violence will report the lowest levels of childhood family cohesion, accord, and closeness. H3. The class characterized with the highest likelihood of reporting exposure to interparental violence and maltreatment perpetrated by either parent, and intimate partner violence will report the highest level of emotional reactivity.

METHODS Participants Participants were 150 emerging adults from a university in the southeastern United States. Participants were predominately female (94%), between the ages of 18 and 28 (M = 20.20, SD =1.47), with 73.3% identifying as White, 15.3% African American, 5.3% biracial, 3.4% did not report, 2% Asian, and .7% American Indian or Alaska Native. The majority of participants reported being in their third or fourth year of their respective programs; 3 were in their first year, 50 sophomore year, 69 junior year, and 27 reported being in their senior year. Participant’s also reported on their current GPA, which ranged from 2.00 to 4.00 (M = 3.22, SD = .42). Regarding their families, the majority of participants’ families were currently intact (64%), followed by separated/divorced (18%), stepfamily (10%), other (4.7%), and never married (3.3%). Participants were asked to estimate their family’s annual income; 34.7% reported between 50K and 100K, 31.3% reported above 100K, 19.3% between 30K and 50K, 8% below 30K, and 6.7% did not know their family annual income. Procedures After approval was obtained from the local Institutional Review Board, students from three undergraduate courses were offered several options to earn extra credit; one option was to complete the instruments used in the current study. The surveys were completed during a lab visit that lasted approximately 1 hour. All participants read a consent form prior to their lab visit explaining the voluntary nature of their participation.

Intergenerational Transmission of Violence

31

Measures Maltreatment and Intimate Partner Violence The Revised Conflict Tactics Scales (CTS2; Straus, Hamby, Boney-McCoy, & Sugarman, 1996) were used to assess lifetime experiences of maltreatment and most recently experienced intimate partner violence. Items from the parent-parent, parent-child, and partner-partner versions of the CTS2 were used. Responses were given on a 1 (“this has never happened”) to 7 (“more than 20 times”) scale. For the parent-parent and partner-partner versions of the CTS2, items from the severity subscales of the physical assault and psychological aggression scales were used yielding eight indicators each for interparental and intimate partner violence, respectively (2 [severity: severe vs. minor] × 2 [type: physical vs. psychological], and 2 [target]). For the parent-child version of the CTS2 items from the physical assault and psychological aggression scales were used yielding four indicators (2 [parent: mother vs. father] × 2 [type; physical vs. psychological]). Final subscale scores were recoded to “1” if one or more of a subscale’s acts occurred previously and “0” if none of the items had occurred. Table 1 presents a detailed list of the categorical indicators used in the following analyses. Childhood Family Environment The Childhood Family Environment Scale was used to assess reports of interpersonal relationships in the family during childhood (i.e., cohesion, accord, and closeness; King et al., 2003). Responses to the 15-items were recorded on a 1 (“almost none of the time”) to 5 (“almost all of the time”) scale, and summed to create a final score. Example items include “People in my family did things together” and “Family members avoided each other.” Higher scores reflected childhood family environment characterized by higher cohesion, accord, and closeness. The alpha coefficient for the current sample was .92. Emotional Reactivity The Emotion Reactivity Scale (Nock, Wedig, Holmberg, & Hooley, 2008) was used to assess participants’ emotional reactivity. Participants responded to 21-items on a 1 (“not at all like me”) to 4 (“completely like me”) scale. Example items include “I experience emotions very strongly” and “I get angry at people very easily.” Items were averaged to create final scores (α = .96 in the current sample), which ranged from 1 to 4 with higher scores reflecting higher levels of emotional reactivity.

32

Table 1.

PETER M. RIVERA AND FRANK D. FINCHAM

Categorical Indicators That Informed the Latent Class Analyses (Coded 1 = Yes, 0 = No). Number of Participants Endorsing a Particular Construct

Exposure to interparental violence Father-to-mother Physical assault minor Psychological aggression minor Physical assault severe Psychological aggression severe Mother-to-father Physical assault minor Psychological aggression minor Physical assault severe Psychological aggression severe Experienced Maltreatment Father perpetrated Physical abuse Psychological abuse Mother perpetrated Physical abuse Psychological abuse Intimate Partner Violence Perpetrated Physical assault minor Psychological aggression minor Physical assault severe Psychological aggression severe Victimization Physical assault minor Psychological aggression minor Physical assault severe Psychological aggression severe

Yes

No

26 96 8 27

122 52 140 121

23 100 8 19

148 48 140 129

34 90

115 59

75 131

75 19

46 112 11 26

104 38 139 124

33 102 6 13

117 48 144 137

Analytical Approach Latent class analysis was conducted with Mplus (Muthe´n & Muthe´n, 1998 2012) to examine the structure underling the set of 12-maltreatment (four parent-child; eight interparental) and eight-intimate partner violence constructs. This approach is particularly useful for the current study design

Intergenerational Transmission of Violence

33

because it models the heterogeneity that exists within response patterns to maltreatment and intimate partner violence experiences based on type and severity. In conducting latent class analysis, the current study used the Pseudo-class (PC) approach (Asparouhouv & Muthen, 2013). First, latent class analysis was conducted testing k classes against k − 1 classes, until the appropriate number of classes to characterize the data emerged. The following fit indices and inferential tests guided this processes: Akaike information criterion (AIC), Bayesian information Criterion (BIC), sample size adjusted BIC (A-BIC), Entropy, and the Lo-Mendell-Rubin Adjusted LRT (LMRT). Lower AIC, BIC, and A-BIC indicate better model data fit. Entropy ranges from 0 to 1, where closer to 1 indicates how well classes have been distinguished. Further, Bootstrapped likelihood ratio test (BLRT) provides a p-value that when significant indicates that a model with k classes significantly fits the data better than the model with k − 1 classes (Roesch et al., 2010). Probabilities of class membership assignment and the substantive interpretation and theoretical meaningfulness of each model also informed this process. Subsequent to identifying the appropriate number of classes that fit the data best, we examined the associations among class membership, childhood family environment, and emotional reactivity. The PC method uses posterior probability-based multiple imputation and pseudo-class Wald chi-square significance tests to determine differences in means and has been demonstrated to be most effective when entropy (i.e., class separation) is greater than .60 (Asparouhouv & Muthen, 2013).

RESULTS Prevalence of Intimate Partner Violence In the current study, over half of participants reported experiencing at least one form of intimate partner violence (n = 119). Moreover, approximately 76% of participants reported perpetration (n = 114) and 70% (n = 105) victimization of intimate partner violence. Our estimates align with past estimates of emerging adults who have experienced intimate partner violence in the United States, which have ranged from 25% to 75% (HalpernMeekin, Manning, Giordano, & Longmore, 2013; Renner & Whitney, 2012; Rennison & Welchans, 2000; Whitaker, Haileyesus, Swahn, & Saltzman, 2007).

34

PETER M. RIVERA AND FRANK D. FINCHAM

Model Selection Table 2 presents the fit indices and inferential test statistics for one-, two-, three-, four-, and five-class solutions. Entropy was particularly high for each model, suggesting that these models discriminated the classes well. AIC and A-BIC decreased for every solution over one-class, suggesting a five-class solution to be the best fit for the data. However, BIC increased when going from a three- to four-class solution and the LMRT test was not significant when comparing a three- to four-class solution, indicating a three-class solution as the best fitting and most parsimonious model. As such, a three-class solution was selected. Description of Three-Class Solution Each of the three subgroups corresponds to an underlying segment of emerging adults in our sample characterized by specific patterns of maltreatment and intimate partner violence. The conditional response probabilities depicted in Fig. 1 provides information for interpreting and labeling each class. Table 2. Fit Indices for Latent Class Models. Model

AIC

BIC

A-BIC

Entropy

LMRT

Classes: n, %

1 Class 2 Class

2804.82 2537.89

2865.04 2661.32

2801.74 2531.57

N/A .96

N/A p < .001

3 Class

2379.72

2566.38

2370.16

.96

p < .01

4 Class

2334.63

2585.51

2321.83

.95

ns

5 Class

2312.48

2625.58

2296.44

.95

p < .05

1. n = 150, 100% 1. n = 100, 66.67% 2. n = 50, 33.33% 1. n = 27, 18.00% 2. n = 71, 47.33% 3. n = 52, 34.67% 1. n = 64, 42.67% 2. n = 38, 25.33% 3. n = 34, 22.67% 4. n = 14, 9.33% 1. n = 15, 10.00% 2. n = 26, 17.33% 3. n = 58, 38.67% 4. n = 15, 10.00% 5. n = 36, 24.00%

Note: AIC = Akaike information criterion; BIC = Bayesian information criterion; A-BIC = sample size adjusted BIC; LMRT = Lo-Mendell-Rubin-adjusted LRT; N/A = not available; ns = not significant.

35

Intergenerational Transmission of Violence

Full Transmission

Psychological Transmission

No Transmission

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

Reported maltreatment

Fig. 1.

Reported intimate partner violence

Conditional Response Probabilities for a Three-Class Solution.

Individuals in the first latent class labeled full transmission are characterized by high probabilities of exposure to minor physical and psychological, and severe psychological interparental violence perpetrated by either parent. In addition, this subgroup had high probabilities of reporting psychological maltreatment perpetrated by either parent, and physical maltreatment perpetrated by mothers. Regarding intimate partner violence, class-one had high probabilities of reporting perpetration of minor physical, psychological, and severe psychological violence, in addition to reporting minor physical and psychological victimization. Individuals in the second class are characterized by high probabilities of reporting exposure to minor psychological interparental violence and psychological maltreatment perpetrated by either parent. This class also had high probabilities of reporting perpetration and victimization of minor psychological aggression; thus, this class was labeled psychological transmission. The third latent class comprised individuals who are likely to report mother perpetrated psychological maltreatment, and perpetration and victimization of minor psychological aggression, as such, this group was called no transmission.

36

PETER M. RIVERA AND FRANK D. FINCHAM

Table 3. Emerging Adult Descriptive Statistics and Mean Differences on Reports of Childhood Family Environments and Emotional Reactivity as a Function of Class Membership (N = 150). Overall Sample

Three-Class Solution Class 1 “Full Transmission” (n = 27)

Average age (SD) 20.20 (1.47) 20.26 (.90) Sex Female 141 (94%) 25 (93%) Male 9 (6%) 2 (7%) Relationship status Single 61 (41%) 14 (52%) Dating someone 16 (11%) 2 (7%) LT dating 62 (41%) 7 (26%) relationship Dating around 7 (5%) 2 (7%) Engaged 2 (1%) 1 (4%) Married 2 (1%) 1 (4%) Current family form Intact 96 (64%) 12 (44%) Separated/divorced 27 (18%) 10 (37%) Stepfamily 15 (10%) 5 (19%) Never married 5 (3%) − (0%) Other 7 (5%) − (0%) Annual family income Below 30k 12 (8%) 4 (15%) 30k 50k 29 (19%) 8 (29%) 50k 100k 52 (35%) 7 (26%) Above 100k 47 (31%) 7 (26%) Do not know 10 (7%) 1 (4%) Parents currently violent towards each other? Yes 3 (2%) 2 (7%) No 147 (98%) 25 (93%) Distal outcome M (SD) Family-of-origin 58.46 51.52 cohesion, accord, (10.59) (13.78)1,2 and closeness Emotional reactivity 2.08 (.72) 2.48 (.82)1,2

Class 2 “Psychological Transmission” (n = 71)

Class 3 “No Transmission” (n = 52)

20.13 (1.80)

20.27 (1.25)

66 (93%) 5 (7%)

50 (96%) 2 (4%)

26 (37%) 7 (10%) 33 (47%)

21 (40%) 7 (14%) 22 (42%)

3 (4%) 1 (1%) 1 (1%)

2 (4%) − (0%) − (0%)

48 (68%) 13 (18%) 7 (10%) 2 (3%) 1 (1%)

36 (69%) 4 (8%) 3 (6%) 3 (6%) 6 (11%)

3 (4%) 10 (14%) 29 (41%) 24 (34%) 5 (7%)

5 (9%) 11 (21%) 16 (31%) 16 (31%) 4 (8%)

1 (1%) 70 (99%)

− (0%) 52 (100%)

56.99 (9.00)1,3

64.08 (7.66)2,3

1.97 (.63)1

2.01 (.72)2

Note: Matching superscripts denote significant post hoc differences; M = mean; SD = standard deviation; LT = long-term.

Intergenerational Transmission of Violence

37

Differences in Childhood Family Environment as a Function of Class Membership To validate the differences between subgroups, mean scores for family environment during childhood were examined. Results from Wald chi-square significance tests indicated that childhood family environment varied across the three classes (see Table 3). Individuals comprising the full transmission subgroup reported the lowest levels of childhood family cohesion, accord, and closeness (M = 51.52), followed by individuals in the psychological transmission subgroup (M = 56.99), with individuals in the no transmission subgroup reporting the highest levels of childhood family cohesion, accord, and closeness (M = 64.08).

Differences in Emotional Reactivity as a Function of Class Membership Results associated with Wald chi-square significance tests that determined whether subgroup differences on emotional reactivity were statistically significant are presented in Table 3. Individuals comprising the full transmission subgroup reported significantly more emotional reactivity (M = 2.48) than those in the psychological transmission (M = 1.97) and no transmission (M = 2.01) subgroups. However, there were no significant differences in emotional reactivity between the individuals in the psychological transmission and no transmission subgroups.

DISCUSSION The purpose of this study was to demonstrate the utility of a personoriented approach to understanding the IGT of violence. This was achieved through latent class analysis using multiple indicators of maltreatment and intimate partner violence that accounted for the multiple types and severity of violent conduct. A three-class solution emerged as most parsimonious, and the classes were labeled full transmission, psychological transmission, and no transmission, reflecting the IGT of violence that most accurately characterized each subgroup. Thus, subgroups of emerging adults were identified across multiple indicators of maltreatment and intimate partner violence. These subgroups were validated by examining reports of emotional reactivity and childhood family environment. Consistent with our

38

PETER M. RIVERA AND FRANK D. FINCHAM

second and third hypotheses, the full transmission subgroup was characterized by the lowest reported levels of childhood family cohesion, accord, and closeness. The full transmission subgroup also reported significantly more emotional reactivity than the psychological transmission and no transmission subgroups. In line with our first hypothesis, distinct subgroups of emerging adults were found based on multiple indicators of maltreatment and intimate partner violence. Notably, the full transmission subgroup was characterized by multiple forms of maltreatment and intimate partner violence that varied in severity. The psychological transmission subgroup was characterized by minor psychological maltreatment and intimate partner violence. Lastly, the no transmission subgroup did not report any forms of maltreatment, yet did have a likelihood of reporting minor psychological aggression within their intimate relationships. This finding is consistent with past research indicating psychological aggression to be a common characteristic within emerging adult intimate relationships (Black et al., 2010). This finding also offer support for a SLT explanation of the IGT of violence and aligns with person-oriented principles by illustrating several possible constellations of maltreatment and intimate partner violence types within the IGT of violence. Validation of the identified subgroups indicated that each subgroup significantly varied from the others on self-reported childhood family environment. In line with our second hypothesis, the full transmission subgroup’s reported the lowest levels of childhood family cohesion, accord, and closeness, followed by the psychological transmission and no transmission subgroups. These findings suggest that a multidimensional view of the IGT of violence should consider family environment processes beyond conflict when examining the IGT of violence hypothesis. Research will be needed that takes a more comprehensive multidimensional view of maltreated offspring’s family-of-origin experiences, which may be achieved by considering family social and environmental characteristics and child maltreatment experiences concurrently; such efforts may uncover overlooked subgroups of maltreated offspring who are vulnerable to the IGT of violence. Consistent with our third hypothesis, the full transmission group reported significantly more emotional reactivity than the psychological transmission and no transmission subgroups. This finding is consistent with the role accorded emotion regulation or self-control in the genesis of violent conduct. Even though the psychological transmission subgroup reported more emotional reactivity than the no transmission subgroup, this difference was not significant. These findings align with past research

Intergenerational Transmission of Violence

39

indicating that patterns of co-occurring maltreatment may have an accumulative effect leading to maladjustment (Arata et al., 2005; Berzenski & Yates, 2011) and underscores the need for IGT of violence research to consider the presence of multiple forms of maltreatment when testing the IGT of violence hypothesis. This can be achieved through adopting a personoriented approach. These findings also align with existing research by suggesting that emotional reactivity is especially salient in understanding conflict within emerging adults’ intimate relationships when a history of maltreatment is present (Rivera & Fincham, 2015; Rosen et al., 2001).

Limitations and Future Directions The current study demonstrates the utility of a person-oriented approach in examining the IGT of violence; however, due to the nature of the data several limitations should be considered when interpreting our findings. Due to the cross-sectional nature of this study, longitudinal research employing person-oriented methods to the IGT of violence is still needed. Another limitation was the sample used in the current study, which is restricted by the disproportionate number of female participants; it will be necessary for research to replicate the current findings with samples more representative of the emerging adult population. Lastly, the sample size in the current study limited our ability to consider all the dimensions of violence in our analyses. There is a need for person-oriented research testing the IGT of violence hypothesis that accounts for the type, severity, chronicity, duration, and frequency of maltreatment perpetrated and later intimate partner violence.

Implications To better address violence transmitted across generations, researchers should consider using person-oriented methods to account for and examine the heterogeneity that exists within the experiences of maltreatment and intimate partner violence. There is a growing consensus that research involving maltreatment or intimate partner violence must move away from examining isolated effects of violence and begin to approach maltreatment and intimate partner violence as multidimensional constructs (Bogat et al., 2005; Herrenkohl & Herrenkohl, 2009; Swartout & Swartout, 2012). Person-oriented approaches offer a range of analytical techniques that will

40

PETER M. RIVERA AND FRANK D. FINCHAM

allow IGT of violence researchers to achieve this, and ultimately will assist in developing a better understanding of this phenomenon.

CONCLUSION To understand fully the etiology of intimate partner violence for maltreated offspring, a multidimensional view of violence is needed. The current study represents a step in this direction by demonstrating the utility of a personoriented approach in understanding the IGT of violence. Our results indicated that heterogeneity within reports of maltreatment and intimate partner violence can be modeled through a person-oriented approach by considering the type and severity of violence perpetrated. The current findings also support family-of-origin processes and offspring emotional reactivity as possible factors that may assist in explaining the IGT of violence.

NOTE 1. We define maltreatment as experiences of neglect, physical, emotional, and/or sexual abuse directed towards an offspring, including exposure to interparental violence.

REFERENCES Arata, C. M., Langhinrichsen-Rohling, J., Bowers, D., & O’Farrill-Swails, L. (2005). Single versus multi-type maltreatment: An examination of the long-term effects of child abuse. Journal of Aggression, Maltreatment & Trauma, 11, 29 52. doi:10.1300/j146v11n04_02 Asparouhouv, T., & Muthen, B. (2013). Auxiliary variables in mixture modeling: A 3-step approach using Mplus. Mplus web notes: No. 15. Retrieved from http://www.statmodel. com/examples/webnotes/webnote15.pdf Bandura, A. J., Ross, R., & Ross, S. (1962). Transmission of aggression through imitationof aggressive models. Journal of Abnormal and Social Psychology, 63, 575 582. doi:10.1037/h0045925 Bergman, L. R. (2001). A person approach in research on adolescence: Some methodological challenges. Journal of Adolescent Research, 16, 28 53. doi:10.1177/0743558401161004 Bergman, L. R., & Magnusson, D. (1997). A person-oriented approach in research on developmental psychopathology. Development and psychopathology, 9, 291 319. doi:10.1017/ S095457949700206X

Intergenerational Transmission of Violence

41

Bergman, L. R., & Wa˚ngby, M. (2014). The person-oriented approach: A short theoretical and practical guide. Estonian Journal of Education, 2, 29 49. doi:10.12697/ eha.2014.2.1.02b Berzenski, S. R., & Yates, T. M. (2011). Classes and consequences of multiple maltreatment a person-oriented analysis. Child Maltreatment, 16, 250 261. doi:10.1177/1077559511 428353 Berzenski, S. R., Yates, T. M., & Egeland, B. (2014). A multidimensional view of continuity in intergenerational transmission of child maltreatment, In Handbook of child maltreatment (pp. 115 129). The Netherlands: Springer. doi:10.1007/978-94-007-7208-3_6 Black, D. S., Sussman, S., & Unger, J. B. (2010). A further look at the intergenerational transmission of violence: Witnessing interparental violence in emerging adulthood. Journal of Interpersonal Violence, 25, 1022 1042. doi:10.1177/0886260509340539 Bogat, G. A., Levendosky, A. A., & Von Eye, A. (2005). The future of research on intimate partner violence: Person-oriented and variable-oriented perspectives. American Journal of Community Psychology, 36, 49 70. doi:10.1007/s10464-005-6232-7 Cui, M., Durtschi, J. A., Donnellan, M. B., Lorenz, F. O., & Conger, R. D. (2010). Intergenerational transmission of relationship aggression: A prospective longitudinal study. Journal of Family Psychology, 24, 687 688. doi:10.1037/a0021675 Ehrensaft, M. K., Cohen, P., Brown, J., Smailes, E., Chen, H., & Johnson, J. G. (2003). Intergenerational transmission of partner violence: A 20-year prospective study. Journal of Consulting and Clinical Psychology, 71, 741 753. English, D. J., Upadhyaya, M. P., Litrownik, A. J., Marshall, J. M., Runyan, D. K., Graham, J. C., & Dubowitz, H. (2005). Maltreatment’s wake: The relationship of maltreatment dimensions to child outcomes. Child Abuse & Neglect, 29, 597 619. doi:10.1016/j. chiabu.2004.12.008 Fergusson, D. M., Boden, J. M., & Horwood, L. J. (2006). Examining the intergenerational transmission of violence in a New Zealand birth cohort. Child Abuse & Neglect, 30, 89 108. Gardner, B. C., Busby, D. M., & Brimhall, A. S. (2007). Putting emotional reactivity in its place? Exploring family-of-origin influences on emotional reactivity, conflict, and satisfaction in premarital couples. Contemporary Family Therapy, 29, 113 127. doi:10.1007/ s10591-007-9039-x Gay, L. E., Harding, H. G., Jackson, J. L., Burns, E. E., & Baker, B. D. (2013). Attachment style and early maladaptive schemas as mediators of the relationship between childhood emotional abuse and intimate partner violence. Journal of Aggression, Maltreatment & Trauma, 22, 408 424. doi:10.1080/10926771.2013.775982 Gover, A. R., Kaukinen, C., & Fox, K. A. (2008). The relationship between violence in the family of origin and dating violence among college students. Journal of Interpersonal Violence, 23, 1667 1693. doi:10.1177/0886260508314330 Hahm, H. C., Lee, Y., Ozonoff, A., & Van Wert, M. J. (2010). The impact of multiple types of child maltreatment on subsequent risk behaviors among women during the transition from adolescence to young adulthood. Journal of Youth and Adolescence, 39, 528 540. doi:10.1007/s10964-009-9490-0 Halpern-Meekin, S., Manning, W. D., Giordano, P. C., & Longmore, M. A. (2013). Relationship churning, physical violence, and verbal abuse in young adult relationships. Journal of Marriage and Family, 75, 2 12. doi:10.1111/j.1741-3737. 2012.01029.x

42

PETER M. RIVERA AND FRANK D. FINCHAM

Hazen, A. L., Connelly, C. D., Roesch, S. C., Hough, R. L., & Landsverk, J. A. (2009). Child maltreatment profiles and adjustment problems in high-risk adolescents. Journal of Interpersonal Violence, 24, 361 378. doi:10.1177/0886260508316476 Herrenkohl, R. C., & Herrenkohl, T. I. (2009). Assessing a child’s experience of multiple maltreatment types: Some unfinished business. Journal of Family Violence, 24, 485 496. doi:10.1007/s10896-009-9247-2 Herrenkohl, T. I., & Herrenkohl, R. C. (2007). Examining the overlap and prediction of multiple forms of child maltreatment, stressors, and socioeconomic status: A longitudinal analysis of youth outcomes. Journal of Family Violence, 22, 553 562. doi:10.1007/ s10896-007-9107-x Jankowski, M. K., Leitenberg, H., Henning, K., & Coffey, P. (1999). Intergenerational transmission of dating aggression as a function of witnessing only same sex parents vs. opposite sex parents vs. both parents as perpetrators of domestic violence. Journal of Family Violence, 14, 267 279. doi:10.1023/A:1022814416666 Karakurt, G., Keiley, M., & Posada, G. (2013). Intimate relationship aggression in college couples: Family-of-origin violence, egalitarian attitude, attachment security. Journal of Family Violence, 28, 561 575. doi:10.1007/s10896-013-9526-9 King, D. W., King, L. A., & Vogt, D. S. (2003). Manual for the deployment risk and resilience inventory (DRRI): A collection of measures for studying deployment-related experiences of military veterans. Boston, MA: National Center for PTSD. Lee, M., Reese-Weber, M., & Kahn, J. H. (2014). Exposure to family violence and attachment styles as predictors of dating violence perpetration among men and women: A mediational model. Journal of Interpersonal Violence, 29, 20 43. doi:10.1177/0886260513 504644 Litrownik, A. J., Lau, A., English, D. J., Briggs, E., Newton, R. R., Romney, S., & Dubowitz, H. (2005). Measuring the severity of child maltreatment. Child Abuse & Neglect, 29, 553 573. doi:10.1016/j.chiabu.2003.08.010 McKee, J. R., & Payne, B. K. (2014). Witnessing domestic violence as a child and adulthood emotionality: Do adults “feel” the consequences of exposure to partner abuse later in the life course? Journal of Aggression, Maltreatment & Trauma, 23, 318 331. doi:10.1080/10926771.2014.881947 Mihalic, S. W., & Elliott, D. (1997). A social learning theory model of marital violence. Journal of Family Violence, 12, 21 47. Milletich, R. J., Kelley, M. L., Doane, A. N., & Pearson, M. R. (2010). Exposure to interparental violence and childhood physical and emotional abuse as related to physical aggression in undergraduate dating relationships. Journal of Family Violence, 25, 627 637. doi:10.1007/s10896-010-9319-3 Millett, L. S., Kohl, P. L., Jonson-Reid, M., Drake, B., & Petra, M. (2013). Child maltreatment victimization and subsequent perpetration of young adult intimate partner violence an exploration of mediating factors. Child Maltreatment, 18, 71 84. doi:10.1177/ 1077559513484821 Muthe´n, L. K., & Muthe´n, B. (1998 2012). Mplus user’s guide. Los Angeles, CA: Muthe´n & Muthe´n. Nock, M. K., Wedig, M. M., Holmberg, E. B., & Hooley, J. M. (2008). The emotion reactivity scale: Development, evaluation, and relation to self-injurious thoughts and behaviors. Behavior Therapy, 39, 107 116. doi:10.1016/j.beth.2007.05.005

Intergenerational Transmission of Violence

43

Renner, L. M., & Whitney, S. D. (2012). Risk factors for unidirectional and bidirectional intimate partner violence among young adults. Child Abuse & Neglect, 36, 40 52. doi:10.1016/j.chiabu.2011.07.007 Rennison, C. M., & Welchans, S. (2000). Intimate partner violence. Washington, DC: U.S. Department of Justice. Retrieved from http://www.bjs.gov/content/pub/pdf/ipv.pdf Rivera, P. M., & Fincham, F. (2015). Forgiveness as a mediator of the intergenerational transmission of violence. Journal of Interpersonal Violence, 30, 895 910. doi:10.1177/088626 0514539765 Roesch, S. C., Villodas, M., & Villodas, F. (2010). Latent class/profile analysis in maltreatment research: A commentary on Nooner et al., Pears et al., and looking beyond. Child Abuse & Neglect, 34, 155 160. doi:10.1016/j.chiabu.2010.01.003 Rosen, k. H., Bartle-haring, S., & Stith, S. M. (2001). Using Bowen theory to enhance understanding of the intergenerational transmission of dating violence. Journal of Family Issues, 22, 124 142. doi:10.1177/019251301022001006 Simons, L. G., Simons, R. L., Lei, M. K., Hancock, D. L., & Fincham, F. D. (2012). Parental warmth amplifies the negative effect of parental hostility on dating violence. Journal of Interpersonal Violence, 27, 2603 2626. doi:10.1177/0886260512436387 Smith, C. A., Ireland, T. O., Park, A., Elwyn, L., & Thornberry, T. P. (2011). Intergenerational continuities and discontinuities in intimate partner violence: A twogenerational prospective study. Journal of Interpersonal Violence, 26, 3720 3752. Sterba, S. K., & Bauer, D. J. (2010). Matching method with theory in person-oriented developmental psychopathology research. Development and Psychopathology, 22, 239 254. doi:10.1017/S0954579410000015 Stith, S. M., Rosen, K. H., Middleton, K. A., Busch, A. L., Lundeberg, K., & Carlton, R. P. (2000). The intergenerational transmission of spouse abuse: A meta-analysis. Journal of Marriage and Family, 62, 640 654. doi:10.1111/j.1741-3737.2000.00640.x Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283 316. doi:10.1177/019251396017003001 Straus, M. A., & Michel-Smith, Y. (2014). Mutuality, severity, and chronicity of violence by father-only, mother-only, and mutually violent parents as reported by university students in 15 nations. Child Abuse & Neglect, 38, 664 676. doi:10.1016/j.chiabu.2013.10.004 Sullivan, T. P., McPartland, T. S., Armeli, S., Jaquier, V., & Tennen, H. (2012). Is it the exception or the rule? Daily co-occurrence of physical, sexual, and psychological partner violence in a 90-day study of substance-using community women. Psychology of Violence, 2, 154 164. doi:10.1037/a0027106 Swartout, K. M., & Swartout, A. G. (2012). Shifted perspectives: Applying personoriented analyses to violence research. Psychology of Violence, 2, 309 312. doi:10.1037/ a0029910 Teicher, M., Samson, J., Polcari, A., & McGreenery, C. (2006). Sticks, stones, and hurtful words: relative effects of various forms of childhood maltreatment. American Journal of Psychiatry, 163, 993 1000. doi:10.1176/appi.ajp.163.6.993 Villodas, M. T., Litrownik, A. J., Thompson, R., Roesch, S. C., English, D. J., Dubowitz, H., & Runyan, D. K. (2012). Changes in youth’s experiences of child maltreatment across developmental periods in the LONGSCAN consortium. Psychology of Violence, 2, 325 338. doi:10.1037/a0029829

44

PETER M. RIVERA AND FRANK D. FINCHAM

von Eye, A., & Bergman, L. R. (2003). Research strategies in developmental psychopathology: Dimensional identity and the person-oriented approach. Development and Psychopathology, 15, 553 580. doi:10.1017/S0954579403000294 von Eye, A., & Bogat, G. A. (2006). Person-oriented and variable-oriented research: Concepts, results, and development. Merrill-Palmer Quarterly, 52, 390 420. doi:10.1353/mpq. 006.0032 Wei, M., Vogel, D. L., Ku, T. Y., & Zakalik, R. A. (2005). Adult attachment, affect regulation, negative mood, and interpersonal problems: The mediating roles of emotional reactivity and emotional cutoff. Journal of Counseling Psychology, 52, 14 24. doi:10.1037/0022-0167.52.1.14 Whitaker, D. J., Haileyesus, T., Swahn, M., & Saltzman, L. S. (2007). Differences in frequency of violence and reported injury between relationships with reciprocal and nonreciprocal intimate partner violence. American Journal of Public Health, 97, 941 947. doi:10.2105/ ajph.2005.079020 Whiting, J. B., Simmons, L. A., Havens, J. R., Smith, D. B., & Oka, M. (2009). Intergenerational transmission of violence: The influence of self-appraisals, mental disorders and substance abuse. Journal of Family Violence, 24, 639 648. doi:10.1007/ s10896-009-9262-3

GENDER, DOMESTIC VIOLENCE, AND PATTERNS OF CONVICTION: ANALYSIS OF INDIA’S SUPREME COURT RULINGS$ Preethi Krishnan and Mangala Subramaniam ABSTRACT Purpose The practices and arrangements within a family can create grounds for violence. Although we agree that family processes are important, we think that these explanations downplay the structure of families (nuclear, extended) and thereby the ways in which gender relations are organized. In this paper, domestic violence is explored as an intra-family dynamic that extends beyond the intimate partner relationship and which seeps into court rulings of cases of such violence. Methodology/approach Using archival data from 164 Supreme Court case decisions on domestic violence in India for the period 1995 2011, we examine both the patterns of conviction and the complexities of

$

An earlier version of this paper has been accepted for presentation at the 2015 ASA annual meetings.

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 45 72 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009003

45

46

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

gender relations within the family by systematically coding the Court’s rulings. Findings Analysis of court rulings show that mothers-in-law were convicted in 14% cases and the husband was convicted in 41% cases. We call attention to the collective nature of the domestic violence crime in India where mothers-in-law were seldom convicted alone (3% of cases) but were more likely to be convicted along with other members of the family. Two dominant themes we discuss are the gendered nature of familial relations beyond the intimate partner relationship and the pervasiveness of such gendered relationships from the natal home to the marital family making victims of domestic violence isolated and “homeless.” Research limitations/implications Future research may benefit from using data in addition to the judgments to consider caste and class differences in the rulings. An intersectionality perspective may add to the understanding of the interpretation of the laws by the courts. Social implications Insights from this paper have important policy implications. As discussed in the paper, the unintended support for violence from the natal family is an indication of their powerlessness and therefore further victimization through the law will not help. It is critical that natal families re-frame their powerlessness which is often derived from their status as families with daughters. Considering that most women in India turn to their natal families first for support when they face violence in their marriages, policy must enable such families to act and utilize the law. Originality/value By examining court rulings on cases of domestic violence in India we focus on the power exerted by some women particularly within extended families which is central to understanding gender relations within institutions. These relations are legitimized by the courts in the ways they interpret the law and rule on cases. Keywords: Gender; family; domestic violence; court judgments; India

Violence against women takes diverse forms across the world. Yet, most women experience violence not from strangers but from known persons.1 A multi-country study by the World Health Organization estimates the

Gender, Domestic Violence, and Patterns of Conviction

47

lifetime prevalence of domestic violence against women to vary from 15% to 71% in 15 sites across 10 countries including the United States and India (Garcı´ a-Moreno, Jansen, Ellsberg, Heise, & Watts, 2005). Intimate partner violence (IPV), and in particular wife abuse by husbands, is a human rights and public health concern throughout the world. Intimate partner violence (IPV) includes physical violence, sexual violence, stalking, and psychological aggression (including coercive tactics) by a current or former intimate partner (Breiding, Basile, Smith, Black, & Mahendra, in press). An intimate partner is a person with whom one has a close personal relationship that may be characterized by the partners’ emotional connectedness, regular contact, ongoing physical contact and sexual behavior, identity as a couple, and familiarity and knowledge about each other’s lives; however, the relationship need not involve all of these dimensions (Breiding et al., in press). In the western context, domestic violence is often seen as intimate partner violence with little consideration of the structure of the family (nuclear versus extended). Domestic violence in South Asia has received special attention because of both its prevalence and severity. More than 60% of Indian women reported three or more episodes of physical or psychological violence in their lifetime (Ahuja et al., 2000). Indian studies indicate that 33% 40% of domestic violence cases include the wider household, including female relatives of the husband as perpetrators or instigators of violence (Dave & Solanki, 2000; Panchanadeswaran & Koverola, 2005). Even where it does not culminate in death, many young Indian married women often suffer violence at the hands of their mothers-in-law. Some evidence exists of mothers-in-law being arrested and convicted in cases of dowry-related deaths (Cherukuri, Britton, & Subramaniam, 2009). Other family members of the husband too are involved in perpetrating violence. In fact, the husband and his relatives are included in India’s laws as potential perpetrators of domestic violence. In 2011, more than 40% of crimes against women in India were accounted under “cruelty by husband and relatives” (National Crime Records Bureau, 2011). It includes women being convicted for inflicting violence on other women within the extended family. Such violence has implications for gender theory and for the interpretation of the law by the court in case rulings. First, our analysis has implications for gender theory. We note that intra-family dynamics involves not only men exerting power, it also includes women wielding some degree of control particularly in extended (or joint) families. Men and women are often held accountable to stereotypes that define their sex category and who engage in gendered practices

48

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

for social approval (Martin, 2003; West & Zimmerman, 1987). In some cultural contexts, such as India, some women are afforded formal and informal power within the family. Therefore, analyzing the involvement of women in exerting power adds to insights about how gender permeates social institutions such as the extended family. We examine how practices and arrangements of the family are associated with violence and have implications for who is charged by the court as well as how this seeps into court rulings particularly in convictions. Second, many countries/territories worldwide, including India, have adopted laws, or revised existing legislation and amendments to their constitutions, to address violence against women more effectively (Krishnan & Subramaniam, unpublished; Ortiz-Barreda & Vives-Cases, 2013). Lawmaking in itself does not guarantee enforcement of the law. The court’s interpretation of the law determines who is charged and convicted in cases of domestic violence. So the text of the court’s judgments can provide insights into whether and how power exerted by some women in extended families is legitimized (or not). While much scholarly work on domestic violence in the Indian context focuses on the family and dowry including the law (cf. Agnes, 1992, 2011; Cherukuri et al., 2009; Ghosh & Choudhuri, 2011; Subramaniam, Remedios, & Mitra, 2009; Visaria, 2000), there has been no systematic analysis of the rulings of India’s Supreme Court in terms of who (if any) is convicted in the cases and whether, and how, the underlying basis of the judgments is about the gender relations in the family.2 We undertake such an analysis in this paper. Using archival data from 164 Supreme Court decisions on domestic violence cases in India for the period 1995 2011, we examine both the patterns of charges and of conviction that is who from the family was implicated and convicted. We also analyze the complexities of gender relations within the family by systematically coding the textual data of court rulings. Our analysis show that India’s Supreme Court has convicted women as perpetrators of the violence with the mother-in-law being implicated in 61% of the cases in contrast to the husband being implicated in more than 90% of the cases. The trends in conviction show that the court is not hesitant in implicating women who are typically viewed as passive and docile in the Indian cultural context. In addition to examining patterns in conviction across members of the family, we discuss two dominant themes in the rulings: the gendered nature of familial relations beyond the intimate partner relationship and the pervasiveness of such gendered relationships from the natal home to the marital family making victims of domestic violence isolated and “homeless.”

Gender, Domestic Violence, and Patterns of Conviction

49

GENDER, FAMILY, VIOLENCE, AND LAW Gender scholars have contributed extensively to examining and understanding domestic violence across countries (cf. Abraham, 2005; Agnes, 1992; Cherukuri et al., 2009; Dobash & Dobash, 1979; Narayan, 2013; Visaria, 2000). One of the biggest challenges to criminalizing domestic violence lies in the private nature of the crime itself. Violence in the private space of the family has been documented as a critical location for violence in the form of physical, mental, and sexual abuse against women and children (Abraham, 2005; Cherukuri et al., 2009; Crenshaw, 1991; Collins, 1998; Dobash & Dobash, 1979; Narayan, 2013; Visaria, 2000). Women’s subordination has been legitimized through the family structure and is further reinforced by other social, political and economic institutions (Dobash & Dobash, 1979). In this paper, we emphasize an analysis of the state’s gender regime or the “state of play of gender relations in a given institution” (Connell, 1987, p. 120) specifically the family. Research on violence and family during the past decade has focused on the types of violence and its interconnections with larger structures of power and inequality (cf. Anderson, 2010; Loseke, 2005). Particularly, applying structural approaches and considering intersections of gender, race, and class has extended understandings of the connections between power and family violence. In her review essay on feminism and family research, Ferree (1990) states that “feminists agree that male dominance within families is part of a wider system of male power, is neither natural nor inevitable, and occurs at women’s cost” (p. 866). Violence grows out of inequality within marriage and reinforces men’s dominance and women’s subordination within the home and outside of it. The conceptualization of violence as coercive control is important for considering it as a tactic of entitlement and power that is deeply gendered (Yllo¨, 2005). The practices and arrangements within a family can create grounds for violence. Family processes such as marital relations including marital instability can lead to violence (Hoffman, Demo, & Edwards, 1994). Although we agree that family processes are important, we think that these explanations downplay the structure of families nuclear versus extended and therefore the ways in which gender relations are organized. Our interest in this paper is to examine the role of extended family members in cases of domestic violence adjudicated by the court, a major state institution. State institutions are central to the critical analyses work of gender scholars (cf. Brown, 1995; Brush, 2003; Ferguson & Gupta, 2002; Sunder

50

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

Rajan, 2003). Among them are those who emphasize the positive role of women legislators, judges, and bureaucrats active inside the state referred to as “femocracts” who can change the governance of gender (Eisenstein, 1996; Mazur, 2001; Stetson, 2001). This group includes feminists who find promise in India’s laws that pertain to dowry and domestic violence (Kapur, 2005; Menon, 2004). They, however, point out that merely getting laws enacted will not alter the structural subordination of women who are viewed as passive and dependent. Yet, in the Indian context some women within the family exert power and so the court must recognize the structure of gender relations within families. The court cannot afford to just mechanically follow the rules laid down by the legislature; they must interpret the rules so as to reconcile them with the wider objectives of justice. Because gender relations within families, and particularly extended families, varies across cultures and contexts, we begin with a detailed discussion of family, dowry, and domestic violence in India.

FAMILY, DOWRY, AND DOMESTIC VIOLENCE IN INDIA In the Indian context, the notion of the family is complex because it often includes the extended family members and is referred to as the “joint” (or extended) family.3 Structural forces such as economic development, increased education and literacy, and advanced communication and information systems impact cultural values and in turn family structures and relationships. India is a part of such change but it is also a country of contrasts and complexities. In India, arranged marriages are generally the norm. In fact, marriage is viewed as an alliance between two families rather than two individuals. While there have been some shifts in trends in marriage, Netting’s (2010) study in a small city in India confirms the continuing role of the family in arranging marriages. Marriages based on familial introductions are seen as generally more stable because they guarantee similarity of religion and social class. Women generally marry into patrilineal, patrilocal families where authority is based on the relations among adult males rather than on the conjugal bonds between spouses (Madan, 1993). Below we discuss the interrelatedness of two strands of literature that are important for our analysis of India’s Supreme Court’s rulings on cases of violence: intra-family dynamics and the state and law.

Gender, Domestic Violence, and Patterns of Conviction

51

Family, Domestic Violence, and the Law While there are large variations, intra-family dynamics in the Indian context are based on gender relations. A woman’s place is, in large part, defined by her relationship to the family, where her power and involvement in decision making can vary based on her age and whether she has a son. Gender differences are also the basis of basic needs. Studies provide accounts of practices that differentially allocate resources such as, food and medical care based on gender. Within households, particularly poor households, there is a strong preference for supporting sons over daughters and men over women (Griffiths, Matthews, & Hinde, 2002; Gupta, 1987; Messer, 1997; Sen, 1993). Generally, older women have more power than younger women, especially new brides, in the family. However, the birth of a son bestows considerable status on the woman (the mother). The family structure is a paradox in that it is constraining as well as supportive of women. Women are often viewed as liabilities in their natal families and are expected to “adjust” in their marriages even in the face of harassment. Yet, because the family (particularly the extended family) is supportive in matters such as child care (Karlekar, 1998), women seldom choose to break away from their families even when they face violence. Social relations in the domestic sphere are further complicated by the fact that many women enter their husband’s families at very young ages as a result of arranged marriages. A desire for very early marriages results in urgency among parents who are willing to pay any “price” (dowry) to “buy” a groom. Dowry refers to the payment of money, goods, and property to the groom’s family by the bride’s family. Dowry, in its most basic form, has been interpreted as a form of pre-mortem inheritance for women who otherwise lack inheritance rights equal to those of men. Over the decades, the dowry demands of the groom and his family have grown and continue through the period of marriage. Not meeting the demands leads to harassment, violence, and sometimes death. Dowryrelated harassment, sometimes resulting in so-called “dowry death” or “bride burning” (Butalia, 2002; Leslie, 1998) became an increasing burden facing women. The pervasiveness of dowry-related violence is evident in the increasing number of cases of dowry deaths. According to the Indian government, there were 8,083 dowry deaths in 2013 (National Crime Records Bureau, 2013).4 Prior to the 1980s, violence against women in India received limited attention from either women’s organizations, political parties, or the media.

52

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

While state initiated actions put issues involving women’s economic welfare on the agenda, it was women’s groups outside of the government that placed body politics on the agenda (Katzenstein, 1989, 1991). The framing of dowry-related deaths as private and as an issue internal to a family were challenged by women’s groups who showed that many official “suicides” were “murders.” Some scholars call for thinking about whether it is the dowry system itself or more specifically the abuse linked to dowry that has to be rallied against (cf. Kishwar, 1988). For Kishwar, the potential of dowry to provide women with some form of inheritance that could improve their position within their marital homes is to suggest that they make even greater martyrs of themselves than society makes of them (2005). Others such as Palriwala (2009) and Lakshmi (1989), reject Kishwar’s position and argue that dowry itself is inherently problematic for women because it is primarily transacted between men and is premised on the inferiority of the bride’s future contribution to her husband and his family who must therefore be financially compensated for. Although there has been a tendency to focus on the violent aspects of dowry-giving and legal solutions (maybe rightly so), it is important to recognize the underlying structural gender inequalities in examining the practice of dowry. It is necessary to distinguish between cultural practices and patriarchal practices. Where cultural practices reflect and perpetuate gender discrimination, the law must be ever more cognizant of the underlying causes of violence against women. We underscore two important points about domestic violence in India. The first is that domestic violence may be related to or unrelated to dowry demands and second, it need not always be perpetuated by men alone; the husband and his family may also be involved in the act of violence. Scholars provide a variety of understandings of domestic violence in India. While some conceptualize it as primarily violence by men against women (Kapur & Cossman, 1996; Mehra, 1998), others include mothers-in-law and others from the husband’s family and explain it as a proxy of men’s acts of violence (Fernandez, 1997). These understandings are not incorporated in analyses of legislative debates in India (Gangoli & Rew, 2011). Less is known about whether state institutions such as the courts implicate (or not) the mother-in-law or other extended family members and on what basis. Feminist legal studies in India and elsewhere examine how law is embedded in patriarchal power. There are some who see no possibility of achieving justice through legal intervention (cf. Menon, 1998). While

Gender, Domestic Violence, and Patterns of Conviction

53

feminist politics rarely makes a dent into the state agenda, feminist involvement with the law provides the state a possibility to enter realms which were closed to it, such as the family. Moreover, feminist intervention also legitimizes state power. Some scholars, such as Menon (2004), therefore call for discarding the legal arena as a fruitful realm for change. Others, such as the NGO, Lawyers Collective of India and the state institution, National Commission for Women differ in their opinion and have been actively collaborating to pressure the state (cf. Fernandez, 1997; Kandiyoti, 1988; Subramaniam, Krishnan, & Bunka, 2014). The differing notions of working with the state in itself pose a challenge to gendered policy-making. Yet making of policy or a law is only one dimension to addressing violence; the interpretation of the law in court rulings is a critical piece that needs systematic scrutiny. An overview of domestic violence related laws in India will be useful for understanding the analysis.

Domestic Violence Laws in India In this paper we analyze cases that refer to the four domestic violence laws: The Dowry Prohibition Act 1961, Section 498A, Section 304B and Protection of Women from Domestic Violence Act 2005. The first legislation to ban dowry (not the related domestic violence per se) is the Dowry Prohibition Act, 1961. This 1961 law defines dowry as “property or valuable security given or agreed to be given either directly or indirectly … at or before or any time after the marriage in connection with the marriage of the said parties.” The Act explicitly covers both the giver and taker of dowry (parents or other relatives or guardian of a bride or bridegroom) as being punishable. This definition of dowry, as we discuss below, is key to adjudication of cases of dowry deaths. The 1961 law saw minor amendments in the 1980s. The 1980s also saw two major additions to the Indian Penal Code (IPC): Section 498A and Section 304B. The Indian Penal Code, comprising more than 500 sections, is a comprehensive code that covers all substantive aspects of criminal law. Section 498A was inserted into the IPC through an amendment in 1983 to address dowry harassment, suicide, and included mental cruelty. This section made cruelty to the wife by husband or his relatives as a cognizable, non-bailable offense. Section 304B was inserted into the IPC through an amendment in 1986. It specifically deals with dowry deaths. Acknowledging the absence of circumstantial evidence in most cases of murders that happen within the privacy of the family,

54

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

Section 304B deals with dowry deaths and transfers the burden of proof to the husband or his family. In 2005, the Protection of Women from Domestic Violence (PWDVA) Act was enacted to fill the gaps of existing domestic violence laws in India particularly the three laws (noted above) which focused on dowry-related violence. The PWDVA Act 2005 is the first of its kind to provide a comprehensive definition to domestic violence by including physical, emotional, sexual, verbal, and economic abuse as violence. While the earlier three domestic violence laws were criminal laws, the PWDVA is a civil law aimed at protecting women from future violence. The PWDVA ensures that women have access to their marital home, claim to maintenance and right to custody over children, rather than merely punishing the husband (Suneetha & Nagaraj, 2005). An innovative aspect of the PWDVA is the inclusion of a woman’s “right to shared household.” The husband and his relatives often have the wife leave the marital home and so in order to protect women from such situations, the law provides them the right to the marital home, even if the property may not be in her or her husband’s name. By analyzing the rulings of the highest court, Supreme Court, in India, we provide insights into how extended family members may be charged and convicted (or not) as perpetrators. As state institutions often legitimize gender inequality and tend to discourage any challenge to this order (cf. Dobash & Dobash, 1979), we expect the Supreme Court of India to also resist challenges to the existing social order. Consequently, while adjudicating domestic violence cases, we expect that the Supreme Court will operate on gendered assumptions by prioritizing the family over women’s rights. The court’s rulings, we propose, will fail to consider notions of gender relations within Indian families and will therefore be less likely to convict women from the husband’s family.

DATA AND METHODS The data for this paper are case decisions of India’s Supreme Court for the period 1995 2011. During this period, a total of 164 cases refer to one or more of the four domestic violence laws in the Indian legal system Dowry Prohibition Act 1961, Section 304B in the Indian Penal Code, Section 498A in the Indian Penal Code and Protection of Women from Domestic Violence Act 2005. Of the 164 cases, more than 35 cases referred

Gender, Domestic Violence, and Patterns of Conviction

55

to the Dowry Prohibition Act, 1961, about 90 cases referred to Section 498A, and about 50 cases utilized Section 304B. Only six of the 164 cases referred to the 2005 PWDVA. Most judgments use a combination of laws which include but are not limited to the four domestic violence laws. The 164 case decisions comprising 2,025 pages are publicly available at http://judis.nic.in/supremecourt/chejudis.asp. Using the search engine available on this website, we searched for and downloaded cases that refer to the above mentioned laws for the period from 1995 to 2011. These data are part of a larger project examining judgments pertaining to domestic violence. We are therefore well acquainted with the data. Each judgment typically contains: the year the decision was delivered, the name of the appellant, name/s of the respondent/s, the name of the panel of judge/s (one or two), and a record of the text of the decision. The text of the decision typically outlines the case background such as demands (as dowry) made, the incident of violence or death by fire or other means, and the sentence, if any. However, not all judgments contain all the information.

Analytic Strategy Keeping in mind, the above referred key laws that address domestic violence, dowry-related or not, we utilized qualitative methods to analyze the court decisions. We used a combination of directed content analysis and conventional content analysis (Berg, 2009). “Directed content analysis involves the use of more analytic codes and categories derived from existing theories and explanations relevant to the research focus. In this case, the investigator will immerse himself or herself in the raw data, using these themes and those that may emerge from the data itself” (Berg, 2009, p. 341). Conventional content analysis uses a grounded theory approach whereby codes are derived inductively from the data. Combining a deductive and inductive approach, we open-coded the text of the decisions using NVivo. The analytic strategy was limited to the language of the decisions and value judgments about specific rulings were avoided. A gendered analysis requires the researcher to look for “disparate treatment” and “disparate impact” (England, 1998). In this study, we examine the disparate treatment and impact of domestic violence laws and their interpretation (cf. Krishnan and Subramaniam, unpublished). Disparate treatment suggests that the policies by themselves are discriminatory towards women but they become complicated in extended family settings as some women exert power over

56

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

others, such as the mother-in-law over a daughter-in-law. We therefore examine the Supreme Court’s rulings for patterns of conviction, if women were implicated, and for broader themes of family dynamics. Several cases included the mother-in-law or sister-in-law as a perpetrator (discussed below). Combining a grounded theory approach for an initial set of themes with systematic coding, we looked for gendered practices engaged in by women which resulted in other women’s subordination. In the first step, the court rulings were coded as (a) favorable, (b) unfavorable, and (c) not applicable. A verdict was coded “favorable,” when either the husband or relative was convicted of the crime or the husband or his relatives were denied bail or the woman was offered maintenance if that was her claim. There were cases for which rulings were partially favorable or partially unfavorable. For example, the woman would have died and the husband or his relative would have been convicted only for cruelty but acquitted for charges concerning her death. Such partial verdicts were coded “favorable” in order to be conservative in our analysis. Cases in which the woman herself did not want to pursue the criminal case and requested for compromise were coded as “not applicable.” Such compromises have often been allowed by the Supreme Court. Two cases in which the wife had filed a case against her husband’s girlfriend were coded as “not applicable.” In both these cases, the Court ruled that the girlfriend cannot be a relative and therefore cannot be charged under these laws. While this verdict is indeed against the wife, the defendant as a woman may be in a position similar to that of the wife. In contrast, cases in which the husband or a relative was acquitted of domestic violence charges were coded “unfavorable” even if they were charged under other sections such as polygamy. In the second step, patterns in terms of who was implicated (charged) and who was convicted were coded. The third step entailed discerning key themes of family dynamics invoked in the court rulings. In this analysis, we are indirectly engaging with women who had faced violence by reading through narrations in case decisions written by judges. Although we are Indian by ethnicity, our upbringing was predominantly urban with access to education and job opportunities. So, while we are in a unique position to interpret cultural practices, we felt some instances of extreme violence as “foreign.” While our familiarity with cultural norms does provide some insights into analyzing the narratives in these case decisions, our personal experiences with Indian families and state institutions, alongside the theoretical framework of gender provide a unique lens in this analysis.

57

Gender, Domestic Violence, and Patterns of Conviction

ANALYSIS: GENDER, DOMESTIC VIOLENCE, AND COURT RULINGS Our analysis of the 164 Supreme Court rulings comprises two parts. In the first part, we present patterns in outcomes for the victim of violence (the woman who is the wife) followed by the patterns in charges and convictions for domestic violence crimes. In the second part, we describe two main themes that emerged from the systematic qualitative analysis of the textual data. The first theme relationships of power within the family demonstrates the complexities of the Indian family structure and the hegemony of the gender structure beyond the intimate partner relationship. The second theme homeless in their own homes demonstrates how gender is embedded within families across the bride’s natal home and her marital home. These analyses draw attention to variations in the structure of families across cultural contexts which in turn is important for understanding how gender permeates the social institution of the family affording some women power. This makes victims of domestic violence isolated and “homeless.”

Overview of Cases As discussed above, our analytic strategy involved the coding of the court verdict as favorable, unfavorable, or not applicable (Table 1). For the period 1995 2011, about 58% of the cases led to favorable verdicts, about 36% were unfavorable, and 6% were not applicable. The outcomes also varied depending on who was the petitioner, the family ties of the petitioner (of the wife’s family or husband’s family), who was the judge, and whether the woman (victim) of violence was alive or not (Table 2). In about 73% (N = 164) of cases, the primary petitioners were men. About 20% of primary petitioners were women and about 7% of cases were taken forward by the state. There was only a small difference in the Table 1.

Outcome of Rulings.

Outcomes

Favorable

Favorable Unfavorable Not applicable Total

95 (57.9%) 59 (36.1%) 10 (6.1%) 164 (100%)

58

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

favorable outcome (as favorable to the woman) when the primary petitioner was a woman versus a man. While about 61% (n = 33) of outcomes were favorable to the wife when a woman was the petitioner, 56% (n = 119) of outcomes were favorable when a man was the petitioner. However, the petitioner whether man or woman could be from the wife’s family or from the husband’s family. Thus, the gender of the petitioner may not have a direct influence on the outcome of the case. Table 2. Outcomes

Outcome by Petitioner and Victim Alive or Not. Favorable

Petitioner Man 67 (56.3%) Woman 20 (60.6%) State 7 (58.3%) Total Petitioner’s ties Husband’s family 70 (53.8%) Wife’s family 17 (77.3%) State 7 (58.3%) Total Victim of violence (when case filed) Dead 71 (64.0%) Alive 23 (43.4%) Total

Unfavorable

Not Applicable

Total

47 (39.5%) 10 (30.3%) 4 (33.3%)

5 (4.2%) 3 (9.1%) 1 (8.3%)

119 (72.6%) 33 (20.1%) 12 (7.3%) 164 (100%)

53 (40.8%) 4 (18.2%) 4 (33.3%)

7 (5.4%) 1 (4.5%) 1 (8.3%)

130 (79%) 22 (13%) 12 (7%) 164 (100%)

37 (33.3%) 24 (45.3%)

3 (2.7%) 6 (11.3%)

111 (68%) 53 (32%) 164 (100%)

The primary petitioner was from the husband’s family in a majority of the cases (79%) and only 54% of these outcomes were favorable to women. In contrast, when the petitioner was from the wife’s family the outcome was favorable to the wife and/or her family in 77% cases (n = 22). A distressing statistic as seen in Table 2 is that the woman had died in the majority of cases (68%; N = 164) reaching the Supreme Court (when the case was filed). Her death could be a suicide or homicide. Thus, the language in many of the decisions were rather sympathetic toward the woman and included comments about dowry as a social evil.5

Patterns of Conviction A total of 107 (65%) families in these cases are extended (joint) families; 34 (21%) are nuclear families, and the family form in the remaining 23 (14%) is unclear. Patterns of conviction by the Supreme Court demonstrate the

59

Gender, Domestic Violence, and Patterns of Conviction

influence of the extended family on the occurrence of domestic violence. In India, even legislations addressing domestic violence include the husband and his relatives as possible perpetrators of domestic violence. For instance Section 498A of the Indian Penal Code made cruelty to the wife by the husband or his relatives as a cognizable, non-bailable offense. Similarly, laws that prohibit dowry also consider the act of demanding dowry as a crime that may be committed by the husband and/or his relatives. We coded the judgments to analyze whether the husband, father-in-law or mother-in-law was implicated, convicted, or acquitted. Out of 164 cases, the husband was convicted in 68 cases while he was implicated of the crime in more than 147 cases. Mothers-in-law were convicted in 23 cases while they were implicated of the crime in 100 cases. Thus, the husband was implicated in 90% of the cases and the mother-in-law was implicated in 61% of the cases. In many cases, several members of the family were implicated simultaneously. The trend in the conviction pattern points to the court recognizing the involvement of the extended family in cases of domestic violence which is as articulated in the laws (Table 3).

Table 3.

Charges and Convictions.

Family Member/s Mother-in-law and husband and& father-in-law Mother-in-law and husband Mother-in-law and father-in-law Mother-in-law alone Mother-in-law, husband, father-in-law and others Total cases mother-in-law was implicated Husband alone Husband and father-in-law Father-in-law alone Not clear/not applicable Total cases mother-in-law was not implicated or charge was unclear/was not convicted TOTAL a

Implicated

Convicted

68 (42%) 21 (13%) 4 (2%) 5 (3.0%) 2 (1.2%) 100 (61%) 51 (32%) 7 (4%) 0 5 (3%)a 64 (39%)

7 (4.3%) 7 (4.3%) 5 (3%) 4 (2.4%) 0 23 (14%) 49 (30%) 5 (3%) 1 (1%) 86 (52.4%)b 141 (86%)

164 (100%)

164 (100%)

In some cases, it was not clear whether other members of the family were included. There were three cases where it was not clear if the mother-in-law was implicated along with the husband. There were two cases where it was not clear if the father-in-law was implicated along with the husband. b No one was convicted because of procedural impasse. That is in many cases the accused had died before the judgment and therefore the court does not proclaim a verdict.

60

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

The conviction of the mother-in-law is often accompanied by conviction of the husband and/or the father-in-law. As noted in Table 3, the mother-in-law alone was implicated in 5 (3%) of the cases while she is implicated with the husband and father-in-law in 68 (42%) of the cases. The mother-in-law was implicated alone or with other members of the family in 100 (61%) of the cases but was convicted in 23 (14%) of cases. In contrast, the husband alone was charged in 51 (32%) of the cases and convicted in 49 (30%) cases. The husband was therefore charged and implicated in more number of cases than the mother-in-law. Although it is clear that the associations of the mother-in-law with men in the family are a key aspect of domestic violence in India, the court rulings are less likely to implicate and charge the mother-in-law. We elaborate on the quality of these associations in our discussion of the two main themes in the next section. Although the law includes the husband and his relatives as potential perpetrators of violence, we find that husbands are most likely to be implicated and convicted of the crime, especially when the victim (the woman) had died. The collective nature of domestic violence crimes results in contrasting judgments regarding conviction or acquittal of the husband. In fact, in some cases, the husband, by virtue of his closeness to the victim was expected to explain the sequence of events which led to the death of the victim. There were cases where the husband was convicted if he was unable to explain the unnatural death of the victim. For instance, in the case of State of West Bengal versus Halder (2009), the decision says: No reason has been assigned as to why accused Dipak Halder did not explain in his examination under Section 313 Cr.P.C. as to how fire occurred. No reason has been explained as to what led the deceased to come out of the house and seek shelter at the varandah of others. There is not a single evidence that accused Dipak Halder made any attempt to put off the fire of his wife. Even in the examination under Section 313, Cr.P.C. he does not say that he tried to put off the fire. On the contrary there is sufficient circumstantial evidence that accused Dipak Halder resisted the deceased from going out of the house to save her life. (State of West Bengal versus Halder, 2009)

In this case the burden of proof is placed on the husband to explain the occurrence of the fire that caused the death of the woman. Circumstantial evidence is critical in the case of domestic violence crimes particularly because of the private nature of the crime. However, there were cases where absence of evidence was used to acquit the husband. This is demonstrated in the case of State of Karnataka versus Manjunathegowda and Anr (2003) in which the decision says:

Gender, Domestic Violence, and Patterns of Conviction

61

… the entire prosecution story with regard to the offence of murder under Section 302 IPC, rests on the circumstantial evidence. It is now well-established principle of law that in the case of circumstantial evidence the chain must unerringly link to the guilt of the accused. There is no evidence on record to show that there was a trail of blood from the house to the place where the dead body was found in the well. There was no evidence of dragging the body. The recovery of blood stained shirt and lungi and MO 11 (axe), said to have been stained with blood, seized at the instance of the accused was disbelieved by the High Court. We are also of the view that dragging of body of the deceased from the house to the well, where the dead body was found, could not be the handy work of an individual. The accused must have been assisted by the other inmates of the house in as much as there were seven inmates in the house on the fateful day. The High Court on reappreciation of the evidence, with regard to the offence under Section 302 IPC, came to the conclusion that the prosecution had failed to establish complete and conclusive chain of circumstances to bring the guilt of A-1 beyond reasonable doubt.

In this case the court was convinced that a homicide had occurred. Yet, the husband was acquitted of the crime because the court believed that the murder could not have been the “handy work of an individual” and the state could not provide adequate evidence to suggest that the husband was responsible. This points to the ambiguity with which the Indian legal system approaches domestic violence. While the law recognizes that domestic violence can be a collective crime, the criminal justice system is designed to adjudicate on the individual’s role in the criminal action. Due to the private nature of the crime, accounting for individual culpability in a collective crime is sometimes a challenge. However as shown above, in most cases the husband is ascribed the largest responsibility for the crime by virtue of his relationship with the victim. Of all the laws discussed here, section 498A has seen support and opposition particularly because the offense is non-bailable. A spate of reports about the misuse of the section by means of false/exaggerated allegations and implication of several relatives of the husband were received by the courts. Keeping in view the representations received from various quarters and observations made by the Supreme Court and the High Courts, India’s Home Secretary through his letter in 2009, requested the Law Commission of India to consider suggesting amendments, if any, to Section 498A or other measures to check the alleged misuse of the said provision. The conviction rate in respect of the cases under 498A is quite low it is about 20% and this includes cases in courts at the state and national level (Government of India, 2012). Interestingly, there is no reliable data to reveal the extent of abuse or misuse of 498A (Government of India, 2012). However, it is possible that

62

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

the judgments which are merely appeals for quashing a case at a lower court may pertain to section 498A. Quashing a case in the lower court suggests that from the available evidence, the Supreme Court did not see the case worthy of indictment. A total of 26 cases (about 16%) pertain to quashing cases as not being in the jurisdiction of the lower court which is a little lower than the 20% reported by the Law Commission for all cases across courts lower courts, state High Courts, and national level Supreme Court. The most recent (2012) report about Section 498A by India’s Law Commission notes that “… the point to be noted is that the value to be attached to the rights of women are no less than the value to be attached to the family as a unit and vice-versa. The challenge before the community is to ensure the promotion of both values. The emphasis should therefore be on wise moderations without overlooking the need and relevance of the retention of penal sanctions necessary to protect and promote women’s rights and interests” (p. 15). However, the safeguard against arbitrary and unwarranted arrests lies in strictly observing the letter and spirit of the conditions laid down in the law as relating to power of arrest and sensitizing the police on the modalities to be observed in cases of this nature. The purported misuse of 498A also provides some insights into the inadequacy of the criminal law in dealing with various types of violence ranging from verbal abuse to even murder. The criminalization of verbal abuse by extended family members is often interpreted as misuse of 498A. However it may be useful to examine the content of the judgments. We turn now to the discussion of the two predominant themes in our qualitative analysis.

Thematic Analysis Our analysis of the textual data yielded two main themes. The first theme examines the gendered nature of familial relations beyond the intimate partner relationship that comes to bear on domestic violence against the wife. The second theme we discuss is the cultural pervasiveness of the gendered relationships from the natal home to the marital family making victims of domestic violence isolated and “homeless.” Family Relationships and Violence The Supreme Court’s rulings clearly note that domestic violence in the Indian context is somewhat of a collective crime occurring within the

Gender, Domestic Violence, and Patterns of Conviction

63

extended family or the joint family. In 107 of the 164 cases we analyzed, the victim lived in a joint family. In this section, we examine the social relations of power within the family; not merely that between intimate partners. As described above, age and gender play an important role in how power relations are organized in the Indian family. Young brides who live with their extended family are expected to be submissive and obedient to men and to older women in the family. Thus, a “good” bride is expected to conform to gender norms. A newly arrived daughter-in-law is expected to be adept in all household chores but the situation changes as women grow older. Women gain formal and informal power in the family through their position as the mother of a son or as the wife of the man who is the oldest son (in term of birth order). As discussed above, dowry is negotiated at the time of marriage but demands may continue later in marriage as well. Husbands and their family members often demand dowry as an entitlement. This notion of entitlement is embedded in the gendered structure of the family in that men are privileged and women’s families are therefore expected to meet the demands of the groom and his family. In order to maintain this gendered structure, women from the husband’s family, specifically the mother-in-law, may perpetrate violence against the daughter-inlaw. Consider the case of Satya Narayan Tiwari versus State of Uttar Pradesh (2010) in which Satya Narayan’s wife was brutally murdered by her husband and mother-in-law for dowry. In this case, a car was being demanded from the bride’s family. There is the testimony of demand of Maruti Car [Maruti is the brand name] being pressed by the two accused persons after about six months of the marriage of the deceased (which took place about three years before the incident) and of her being pestered, nagged, tortured and maltreated on non-fulfillment of the said demand which was conveyed by her to her parents from time to time on her visits to her parental home and on telephone … Seemingly, violence had first been applied to her inside the bedroom by the accused and offering resistance she had somehow run out to the Chajja [balcony] adjoining the room and the blood dropped there …. To incapacitate her of any meaningful resistance, the accused persons interfered with her breathing process with the compression of the windpipe of her neck before burning her. Strangulating her half way to overpower her and to render her incapable of offering any meaningful resistance, the two accused then poured kerosene over her and burnt her. (Satya Narayan Tiwari versus State of Uttar Pradesh 2010)

In this case the mother-in-law and her son engaged in violence against the young wife or the bride, when her natal family (the wife’s) was unable

64

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

to meet the demand for a car. This case ruling by the court demonstrates how the mother-in-law exerts power through her son. In the case of Satya Narayan Tiwari versus State of Uttar Pradesh 2010, the father of the bride had met with her father-in-law to seek his assurance of his daughter’s safety. The father-in-law named Ghanshyam had assured the father of the bride that his daughter would be safe. However, later in the judgment the Supreme Court decision states: It appears that Ghanshyam could not control the disposition of his wife and son (the two accused) and they continued to pursue their greed by tormenting and maltreating the young lady (deceased) to get a Maruti Car in dowry from her parents. (Satya Narayan Tiwari versus State of Uttar Pradesh 2010)

Thus, both the father of the bride as well as the court assume that the father-in-law is in a position to “control” the actions of his wife (husband’s mother), the mother-in-law of the victim. Such assumptions suggest that the mothers-in-law may achieve power in the family, but still as perceived by the court, may occupy positions that are secondary to the men in the family. In another case of Rajendran versus State Assistant Commissioner (2008), the husband, mother-in-law and the father-in-law engaged in violence against the bride because she gave birth to a girl child. Moreover, because there was a death in the family after the birth of the girl child, the family members believed that the birth of the girl child had brought bad luck to the family. As a result, the other family members, including the mother-in-law began to torture the young woman. The daughter-in-law of the family is considered to be the carrier of good luck and prosperity for the family and her perceived inability to do so, provided the justification for inflicting violence on her. Therefore, protecting the family was a legitimate reason to engage in violence against the young wife. In both cases, discussed above, the legitimacy for the mother-in-law to engage in violence against the young bride is derived from her position as the mother of the son. If analyzed in isolation, the actions of the motherin-law do not seem to conform to gender norms. Usually, men and women are held accountable to their sex category and therefore are expected to behave accordingly (West & Zimmerman, 1987). However, as evident in these cases, some women engage in atypical behavior within the family a sense of power in complicity with men typically the son that subordinates women. Thus women are able to reject gender norms when it contributes to domination of men and submission of women and consequently enable maintaining gender relations within the family.

Gender, Domestic Violence, and Patterns of Conviction

65

Homeless in Their Own Home Typically, young brides entering their marital home experience familial gender relations that do not entirely differ from their natal home. Under these circumstances, women victims of domestic violence lack support even from their natal families. When the young wife recognizes that her life is in danger in the marital home, she often seeks assistance from her own parents and/or siblings but she seldom receives support. In most discussions of domestic violence, the focus is on the husband and his family. In the analysis that follows, we draw attention to the prevalence of the structure of gendered relations not only in the marital families but also in the natal families of young women. The focus on the natal family is important in the study of domestic violence in India because women often turn to their birth families, rather than law-enforcement institutions, for support. We argue that gender relations are embedded in the natal family as well as the marital home which renders the young woman “homeless” when she faces violence. By focusing on the natal family we also complicate the notion of the young woman (the wife) as being a “passive” victim of domestic violence particularly in India. This theme recurred in about 11% of the cases. In most of the domestic violence cases reaching the Supreme Court, the young woman had already died (in 68% of the total cases). The Court views young women in Indian families as passive, helpless victims. This is evident in the case of K. Prema Rao and Anr versus Yadla Srinivasa Rao and others (2002) where the judgment says, “Frailty thy name is woman,” that is how in one of his plays Shakespeare described one of the female characters in his play. This description is more and more in evidence particularly in rural Indian society where married women, who are unable to muster courage to fight against cruelty and harassment, meted out to them by their spouses and family members find no escape other than ending their own life. (K. Prema Rao and Anr versus Yadla Srinivasa Rao and others, 2002)

However, our analysis suggests that victims of domestic violence were neither passive nor afraid to fight violence. In most cases, as women faced violence in their marital homes, they usually informed their natal family and sought support. Unfortunately, more often than not, the natal family responded by negotiating with the husband and his relatives. Moreover, parents and other members of the natal family convince the victim to “adjust” to the environment of her marital family inspite of knowing that her (daughter) life was in danger. This is evident in the case of Deen Dayal

66

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

versus State of Uttar Pradesh (2009) where the judgment quotes the father of the young wife as follows: Two months before death of Asha Devi, Deen Dayal father-in-law of Asha Devi had come to our house for taking her. Deen Dayal had demanded chain of gold for his son and rupees ten thousand and he had asked to send Asha Devi. My daughter was not prepared to go. But we made her to comprehend and then she was sent. Deen Dayal took Asha Devi with him in anger.

Later the judge corroborates the above testimony of the young wife’s father. Their inability to meet his demand had caused him annoyance and anger. Asha Devi was naturally apprehensive and was very reluctant to go with him. But they somehow prevailed upon her and made her depart with him. There is thus positive evidence of her being subjected to harassment.

As evident in this case, Asha Devi the young wife, is not a passive victim of domestic violence. Rather she had informed her natal family about her experience of violence in her marital relationship and even declined to return to her marital home. Even in the face of violence, the victim’s family convinced her to join her husband. This case shows that young women’s lives, before and after marriage, are entangled in familial gender relations. Similar instances are evident in several of the 164 cases we analyze. For instance, in Jagjit Singh versus State of Punjab (2009) the judgment by the court describes a similar situation: Jasprit Kaur had also informed her father that the accused had made her life miserable at their house and made living there impossible. The complainant Avtar Singh and his wife consoled Jasprit Kaur and sent her back to the house of the accused by telling her that they will inform her about the demand in a day or two.

In this case too, the parents of the young wife had her return to her marital home even when she informed them of the danger awaiting her. Cases such as that of Asha Devi and Jasprit Kaur show that power or lack of it is transmitted to other members of the family as the gender structure is deeply embedded in the Indian family system. Moreover, the father of the victim has relatively less power in relation to the groom’s family and in contrast to the power achieved by the mother-in-law through her son. Rather than question the violence inflicted on his daughter, he tries to meet the demands of the groom’s family, however unreasonable it may be. Thus, gender inequality is reinforced through family relations (both marital and natal) leaving the victim isolated when she faces violence.

Gender, Domestic Violence, and Patterns of Conviction

67

These cases also bring to the forefront another important issue and that is the victim not approaching law-enforcement agencies in the absence of support from either the natal or the birth family. One plausible reason, as evident from the rulings, is the fear of losing a marital life that is deeply valued in society. An example is the case of Satya Narayan Tiwari versus State of Uttar Pradesh (2010). In this case, the natal family chooses to seek a compromise and the victim too does not choose to take legal recourse for fear of losing her marital life. The judgment says: On her visits to her parental house, she (deceased) used to narrate to him (this witness) her torture and maltreatment. She had also informed him in this behalf on telephone. Both the accused were there at their home at Farrukhabad and repeated the demand of Maruti car. On his expressing inability to meet this demand, he and Jaideo Awasthi were insulted and humiliated and turned out of the house. Both the accused told them not to visit their house again without meeting their demand of a Maruti Car. Surya Kant Dixit PW1 then went to Geeta’s father-in-law at the place of his employment- State Bank because he was a gentleman. He apprised him of the conduct of his wife and son (accused) pressing the demand of Maruti Car. He, however, offered consolation. Geeta, daughter of Surya Kant Dixit DW 1, also advised him not to take any action and he went away. The victim might have thought that making of FIR by her father at that juncture would ruin her matrimonial life and so she advised him not to take any legal step at that time. (Satya Narayan Tiwari versus State of Uttar Pradesh, 2010) [our italics]

In this case, the victim, Geeta, informed her natal family about the violence but did not choose to take legal recourse. The priority accorded to the institution of marriage even in the face of violence is yet another part of the gendered structure of the Indian family structure. Not only do the groom’s family feel entitled to make demands of the bride’s family, but the bride’s family too do not desire to break the relationship even if it means the daughter is facing violence.

DISCUSSION AND CONCLUSION In this paper we discuss domestic violence as an intra-family dynamic that extends beyond intimate partner relationships. While the law recognizes that domestic violence can be a collective crime, the criminal justice system is designed to adjudicate on the individual’s role in the criminal action. The Supreme Court’s rulings clearly note that domestic violence in the Indian context is somewhat of a collective crime occurring within the extended family or the joint family. Due to the private nature of the crime, accounting for individual culpability in a collective crime is sometimes a challenge.

68

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

We call attention to the collective nature of the domestic violence crime in India where mothers-in-law were seldom convicted alone (3% of cases) but were more likely to be convicted along with other members of the family. First, this paper demonstrates that in an extended family, women in positions of power engage in violence against other women in the family but this does not lead to an overall change in gender relations within the institution of the family. The judgments analyzed in this paper highlight the power of the mothers-in-law as they engage in violence against the newly married daughters-in-law. Yet, such display of power does not necessarily change gender relations within the family but rather exacerbates it. The mother-in-law, inspite of her power, continues to occupy a secondary position in the family as she derives power through her association with men in the family. It is only by upholding the higher status of the men in the family that she is able to subordinate the young wife. Second, our analysis demonstrates how atypical gender behavior is accepted when the outcome is the continued maintenance of the gender structure. Gender theory suggests that people are held accountable to their sex category (West & Zimmerman, 1987). As perpetrators of violence, mothers-in-law do not conform to gendered behavior but they are not held accountable within their own families for the non-conformance (although they are held accountable by the court). This suggests that nonconformance to gender roles is accepted when it aids maintaining the gender structure within the institution of the family. Transgressing gender norms is considered an important act of resistance and therefore one that contributes to social change (Deutsch, 2007; Risman, 2004). However, our analysis show that transgressions by mothers-in-law enable preserving gender inequality while assisting individual women exert power even if it may be secondary to the men in the family. Finally, our analysis brings attention to the pervasiveness of the gender structure beyond the intimate partner relationship. Most policies which address domestic violence focus on the intimate partner and the marital family (in the case of Indian law). The role of the natal family in being somewhat complicit in this gendered violence is made invisible most of the time.6 From when the unmarried daughter is considered a burden, and until when those in the natal family advise their daughter to “adjust” to the marital family, the natal family enables the system of violence. However, unlike the marital family, where violence is about power, the natal family’s unintended support of violence is a demonstration of their powerlessness in the larger gender structure. By drawing attention to this “unintended” violence, we argue that any policy that seeks to address domestic violence has to

Gender, Domestic Violence, and Patterns of Conviction

69

consider the structure of gender relations in both the marital and natal home. However, punitive action against the natal family cannot be the way forward. As we saw in some cases discussed above, the unintended support for violence from the natal family is an indication of their powerlessness and therefore further victimization through the law will not help. However, insights from this paper have important policy implications. It is critical that natal families re-frame their powerlessness which is often derived from their status as families with daughters. Considering that most women in India turn to their natal families first for support when they face violence in their marriages, policy must enable such families to act and utilize the law.

NOTES 1. http://www.un.org/en/women/endviolence/pdf/pressmaterials/unite_the_ situation_en.pdf 2. In India, the Supreme Court is the highest appellate court which considers appeals against judgments rendered by state level High Courts. 3. We use the terms “joint” and “extended” interchangeably. 4. http://ncrb.gov.in/CD-CII2013/figure%20at%20a%20glance.pdf. Accessed on May 3, 2015. 5. Although mothers-in-law and sisters-in-law may be involved in the legal proceedings, when we mention “woman,” we mean the wife on whom violence has been inflicted. 6. A daughter is seen as a liability by most parents because she has to be married and she “belongs” in the marital family.

ACKNOWLEDGMENT We thank the anonymous reviewers and the editors for their insightful comments and suggestions.

REFERENCES Abraham, M. (2005). Fighting back: Abused South Asian women’s strategies of resistance. In N. J. Sokoloff & C. Pratt (Eds.), Domestic violence at the margins: Readings on race, class, gender, and culture (pp. 253 271). New Brunswick, NJ: Rutgers University Press. Agnes, F. (1992). Protecting women against violence-review of a decade of legislation, 1980 89. Economic and Political Weekly, 27(17), WS19 WS21 + WS24 WS33.

70

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

Agnes, F. (2011). The concubine and notions of constitutional justice. Economic and Political Weekly, 46(24), 31 33. Ahuja, R., Bangdiwala, S., Bhambal, S., Jain, D., Jeyaseelan, L., & Kumar, S. (2000). Domestic violence in India − A summary report of a multi-site household survey. Washington, DC: International Center for Research on Women. Anderson, K. L. (2010). Conflict, power, and violence in families. Journal of Marriage and Family, 72(3), 726 742. Berg, B. L. (2009). Qualitative research methods for the social sciences (7th ed.). Boston, MA: Pearson. Breiding, M. J., Basile, K. C., Smith, S. G., Black, M. C., & Mahendra, R. R. (in press). Intimate partner violence surveillance: Uniform definitions and recommended data elements: Version 2.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control Atlanta. Brown, W. (1995). States of injury: Power and freedom in late modernity. Princeton, NJ: Princeton University Press. Brush, L. D. (2003). Gender and governance. Walnut Creek, CA: Altamira Press. Butalia, S. (2002). The gift of a daughter: Encounters with victims of dowry. New Delhi: Penguin Books. Cherukuri, S., Britton, D. M., & Subramaniam, M. (2009). Between life and death women in an Indian state prison. Feminist Criminology, 4(3), 252 274. Collins, P. H. (1998). It’s all in the family: Intersections of gender, race, and nation. Hypatia, 13(3), 62 82. Connell, R. W. (1987). Gender and power: Society, the person and sexual politics. Stanford, CA: Stanford University Press. Crenshaw, K. (1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43(6), 1241 1299. Dave, A., & Solanki, G. (2000). Domestic violence in India. A summary report of four research studies. Washington, DC: International Centre for Research on Women. Deen Dayal versus State of Uttar Pradesh. (2009). Retrieved from http://judis.nic.in/supremecourt/imgs1.aspx?filename=33460. Accessed on July 26, 2015. Deutsch, F. M. (2007). Undoing gender. Gender & Society, 21(1), 106 127. Dobash, R. E., & Dobash, R. (1979). Violence against wives: A case against patriarchy. New York, NY: Free Press. Eisenstein, H. (1996). Inside agitators: Australian femocrats and the state. Philadelphia, PA: Temple University Press. England, P. (1998). What do we mean when we say something is gendered? In Organizations, and Work, A section of the American Sociological Association, Fall 1998 Newsletter. Ferguson, J., & Gupta, A. (2002). Spatializing states: Toward an ethnography of neoliberal governmentality. American Ethnologist, 29(4), 981 1002. Fernandez, M. (1997). Domestic violence by extended family members in India: Interplay of gender and generation. Journal of Interpersonal Violence, 12(3), 433 455. Ferree, M. M. (1990). Beyond separate spheres: Feminism and family research. Journal of Marriage and Family, 52(4), 866 884. Gangoli, G., & Rew, M. (2011). Mothers-in-law against daughters-in-law: Domestic violence and legal discourses around mother-in-law violence against daughters-in-law in India. Women’s Studies International Forum, 34(5), 420 429.

Gender, Domestic Violence, and Patterns of Conviction

71

Garcı´ a-Moreno, C., Jansen, H. A., Ellsberg, M., Heise, L., & Watts, C. (2005). WHO multicountry study on women’s health and domestic violence against women: Initial results on prevalence, health outcomes and women’s responses. Geneva: World Health Organization. Ghosh, B., & Choudhuri, T. (2011). Legal protection against domestic violence in India: Scope and limitations. Journal of Family Violence, 26(4), 319 330. Government of India. (2012). Law commission of India. Report No. 243 on Section 498A IPC. Retrieved from http://lawcommissionofindia.nic.in/reports/report243.pdf. Accessed on March 24, 2015. Griffiths, P., Matthews, Z., & Hinde, A. (2002). Gender, family, and the nutritional status of children in three culturally contrasting states of India. Social Science & Medicine, 55(5), 775 790. Gupta, M. D. (1987). Selective discrimination against female children in rural Punjab, India. Population and Development Review, 13(1), 77 100. Hoffman, K. L., Demo, D. H., & Edwards, J. N. (1994). Physical wife abuse in a non-Western society: An integrated theoretical approach. Journal of Marriage and Family, 56(1), 131 146. Jagjit Singh versus State of Punjab. (2009). Retrieved from http://judis.nic.in/supremecourt/ imgs1.aspx?filename=34031. Accessed on July 26, 2015. Kandiyoti, D. (1988). Bargaining with patriarchy. Gender & Society, 2(3), 274 290. Kapur, R. (2005). Erotic justice: Law and the new politics of postcolonialism. London: Glass House Press. Kapur, R., & Cossman, B. (1996). Subversive sites: Feminist engagements with law in India. New Delhi: Sage. Karlekar, M. (1998). Domestic violence. Economic and Political Weekly, 33(27), 1741 1751. Katzenstein, M. F. (1989). Organizing against violence: Strategies of the Indian women’s movement. Pacific Affairs, 62(1), 53 71. Katzenstein, M. F. (1991). Getting women’s issues onto the public agenda: Body politics in India. Samya Shakti, 6, 1 16. Kishwar, M. (1988). Rethinking dowry boycott. Manushi, 48, 10 13. Krishnan, P., & Subramaniam, M. Dowry, domestic violence and gender: Legal interpretations by India’s Supreme Court. Unpublished paper. Lakshmi, C. S. (1989). On kidneys and dowry. Economic and Political Weekly, 24(4), 189 190. Leslie, J. (1998). Dowry, dowry deaths, and violence against women: A journey of discovery. In W. Menski(Ed.), South Asians and the dowry problem (pp. 21 35). London: Trentham Books Limited. Loseke, D. R. (2005). Through a sociological lens: The complexities of family violence. In D. R. Loseke, R. J. Gelles, & M. M. Cavanaugh (Eds.), Current controversies on family violence (pp. 35 48). Thousand Oaks, CA: Sage. Madan, T. N. (1993). The kinship map of India. In P. Uberoi (Ed.), Family, kinship and marriage in India (pp. 416 434). New Delhi: Oxford University Press. Martin, P. Y. (2003). “Said and Done” versus “Saying and Doing” gendering practices, practicing gender at work. Gender & Society, 17(3), 342 366. Mazur, A. (2001). State feminism women’s movements and job training: Making democracies work in the global economy. New York, NY: Routledge. Mehra, M. (1998). Exploring the boundaries of law, gender and social reform. Feminist Legal Studies, 6(1), 59 83. Menon, N. (1998). State gender community-citizenship in contemporary India. Economic and Political Weekly, 33(5), PE3 PE10.

72

PREETHI KRISHNAN AND MANGALA SUBRAMANIAM

Menon, N. (2004). Recovering subversion: Feminist politics beyond the law. Champaign, IL: University of Illinois Press. Messer, E. (1997). Intra-household allocation of food and health care: Current findings and understandings Introduction. Social Science & Medicine, 44(11), 1675 1684. Narayan, U. (2013). Dislocating cultures: Identities, traditions, and third world feminism. New York, NY: Routledge. National Crime Records Bureau. (2013). Crime in India 2013. Ministry of Home Affairs. Retrieved from http://ncrb.gov.in/CD-CII2013/home.asp. Accessed on May 3, 2015. National Crime Records Bureau, Ministry of Home Affairs. (2011). Crime in India 2011 compendium. Retrieved from http://ncrb.nic.in/CD-CII2011/compendium2011.pdf. Accessed on December 15, 2014. Netting, N. S. (2010). Marital ideoscapes in 21st-century India: Creative combinations of love and responsibility. Journal of Family Issues, 31(6), 707 726. Ortiz-Barreda, G., & Vives-Cases, C. (2013). Legislation on violence against women: Overview of key components. Revista Panamericana de Salud Pu´blica, 33(1), 61 72. Palriwala, R. (2009). The spider’s web: Seeing dowry, fighting dowry. In T. Bradley, E. Tomalin & M. Subramaniam (Eds.), Dowry: Bridging the gap between theory and practice (pp. 1 28). New Delhi: Women Unlimited. Panchanadeswaran, S., & Koverola, C. (2005). The voices of battered women in India. Violence against Women, 11(6), 736 758. Rajendran versus State Assistant Commissioner. (2008). Retrieved from http://judis.nic.in/ supremecourt/imgs1.aspx?filename=33199. Accessed on July 26, 2015. Rao, K. P. and Anr versus Yadla Srinivasa Rao and others. (2002). Retrieved from http:// judis.nic.in/supremecourt/imgs1.aspx?filename=20660. Accessed on July 26, 2015. Risman, B. J. (2004). Gender as a social structure theory wrestling with activism. Gender & Society, 18(4), 429 450. Satya Narayan Tiwari versus State of Uttar Pradesh. (2010). Retrieved from http://judis.nic. in/supremecourt/imgs1.aspx?filename=37055. Accessed on July 26, 2015. Sen, A. (1993). The economics of life and death. Scientific American, 268, 40 47. State of West Bengal versus Halder. (2009). Retrieved from http://judis.nic.in/supremecourt/ imgs1.aspx?filename=34662. Accessed on July 26, 2015. Stetson, D. M. (2001). Abortion politics, women’s movements, and the democratic state: A comparative study of state feminism: A comparative study of state feminism. Oxford: Oxford University Press. Subramaniam, M., Krishnan, P., & Bunka, C. (2014). Women’s movement groups in state policy formulation: Addressing violence against women. Indian Anthropologist, 44(1), 37 52. Subramaniam, M., Remedios, K., Mitra, D. (2009). Dowry and transnational activism. In T. Bradley, E. Tomalin, & M. Subramaniam (Eds.), Dowry: Bridging the gap between theory and practice (pp. 197 225). London: Zed, Women Unlimited. Sunder Rajan, R. (2003). The scandal of the state: Women, law, and citizenship in postcolonial India. Durham, NC: Duke University Press. Suneetha, A., & Nagaraj, V. (2005). Adjudicating (Un)domestic battles. Economic and Political Weekly, 40(38), 4101 4103. Visaria, L. (2000). Violence against women. Economic and Political Weekly, 35(20), 1742 1751. West, C., & Zimmerman, D. H. (1987). Doing gender. Gender & Society, 1(2), 125 151. Yllo¨, K. A. (2005). Through a feminist lens: Gender, diversity, and violence: Extending the feminist framework. In D. Loseke, R. J. Gelles, & M. Cavanaugh (Eds.), Current controversies on family violence (pp. 19 34). Thousand Oaks, CA: Sage.

STRUCTURAL BASIS OF GENDER VIOLENCE IN CROSS-REGIONAL MARRIAGES: TALE FROM HARYANA, INDIA Sonali Mukherjee ABSTRACT Purpose Primary objective of the research is to study the various forms of violence existent in a cross-regional marriage. This form of marriage is the outcome of gendered marriage squeeze as there is discordance between supply and demand of brides in the marriage market of Haryana. Hence, brides are imported into Haryana from far-off poorer regions of the country like West Bengal, Tripura, Assam, Chhattisgarh etc. This implies a big change or adjustment in traditional marriage behaviour for both the host as well as the feeder society. Methodology/approach Fieldwork was conducted in Haryana and West Bengal to get a first-hand understanding of the research problem. Focussed Group Discussions and In-Depth interviews were conducted to get a nuanced understanding of the issues in hand. An attempt for the first time has been made to study one of the regions, that is West Bengal,

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 73 97 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009004

73

74

SONALI MUKHERJEE

which is providing the brides. It is thought provoking to understand the reasons for occurrence of such marriages in West Bengal. Findings Prevalent socio-economic conditions of Haryana and West Bengal have been analysed to derive the causal factors for such a phenomenon. Haryana is a patriarchal, patrilineal and patrilocal society. Interplay of caste and class has significant repercussions on these brides who are married to the Haryana men. These brides go through a marriage which may not always be violent on the outside but definitely so in the inside. These brides face contestations in their everyday existence. Originality/value It would be of significance to the researchers in the field of sociology, anthropology, development studies, gender studies and demography. It would also interest human rights activists. Keywords: Daughter devaluation; cross-regional marriage; migrant bride; adverse sex ratio; violence; vulnerability

INTRODUCTION Discrimination against the girl child is rampant in India. This gendered violence from birth is due to son preference in all the regions (Agnihotri, 1995; Miller, 1997; Visaria & Ramachandran, 2007). This has created an adversity in the sex ratio of the country as evidenced by the census figures1 (2011) which show the presence of only 940 females per 1,000 males in the population. Though it is seven points higher than 2001 census yet it is not reflective of any change in the situation as further devaluation of the girl child is revealed in the child sex ratio (0 6 years). Child sex ratio (CSR) which was 927 in 2001 census has declined to 914 in the 2011 census. So son preference leads to lesser number of daughters being born. In other words, daughters are prevented from being born by sex selective abortion. If some of those daughters who were preferred to be dead happened to be born then they are mercilessly killed in the first few hours of their existence.2 If that is also not possible then they are neglected to the point of extinction.3 Gendered birth preference to annihilation of the girl child is indicative of a premeditated murder which is executed effortlessly by many. Here it becomes important to mention that northern, north western and central regions of India take the lead in achieving this demographic aberration. Statewise changes in the Child Sex Ratio in the 2011 census over the

Structural Basis of Gender Violence in Cross-Regional Marriages

75

2001 census are given in Fig. 1. It also gives the location of different States in the country which would enable readers to situate the study.

Skewed Sex Ratio Leading to Marriage Squeeze Lesser number of females in the population has resulted in a phenomenon which probably was never contemplated by those who were responsible for the same. It created a huge dent in the marriage market as the number of marriageable males was much more than females and hence it created gendered marriage squeeze. Incidences of complexities, anxieties, difficulties and failure to find a spouse in the marriage market for both males and females resulting from the imbalance in the number of marriageable males and marriageable females in the marriage cohort (culture specific age groups within which marriages are fixed) has been referred to as marriage squeeze. ‘Marriage squeeze reflects the relationship between supply and demand in the marriage market within a certain period of time and space. Once a person enters his or her marriageable age, he or she is involuntarily drawn into the marriage market, is exposed to the system of supply and demand of the spouses, and then selects the spouse or is selected, compared or matched by others’ (Kuangshi, 2014, p. 1643). Discordance between supply and demand of brides in marriage market implies a big change or adjustment in traditional marriage behaviour. To overcome this problem another phenomenon emerged which was that of cross-regional marriages (CRM). These cases have been primarily reported from Haryana, Punjab, Rajasthan and Uttar Pradesh. Shortage of marriageable women in these regions is resulting in the ‘import’ of brides from poorer regions of India like West Bengal, Assam, Chattishgarh, Tripura, Jharkhand, etc. (Chaudhry & Mohan, 2011; Das Gupta & Li, 1999; Kaur, 2004; Mukherjee, 2013). This paper is the outcome of an empirical research based on cross-regional marriages in Haryana, a northern State and West Bengal, an eastern State wherein Haryana is the bride taker and West Bengal is the bride provider. Point to ponder is whether the societies which are not kind to their own girls would be tolerant to ‘outsiders’ or would simply accept them for their reproductive and productive labour. Again there are reports on the variability of these situations, that is some of the women may be staying within the premise of the house but many are kept in a small hutment on the fringes of their agricultural field (Chowdhry, 2005), definitely away from their main dwelling unit. In some cases women did not seem to be in a comfortable situation and again in

76

Child Sex Ratio in Age Group 0 6 Census 2001 and 2011.

SONALI MUKHERJEE

Fig. 1.

Structural Basis of Gender Violence in Cross-Regional Marriages

77

some, women reported to be better-off than their parental homes, at least economically. This kind of arrangement does raise many a questions which would be answered subsequently in the course of the paper. Haryana is a patrilinear patrilocal society therefore; migrant brides4 (Mukherjee, 2013) reside with their grooms in Haryana only. Contestations within these marriages can only be studied where they are located. Hence, the paper primarily revolves around Haryana. West Bengal has been included to uncover the most plausible explanation for these migrant brides to venture so far away from their parental homes to an unknown land. Migration for marriage could well be a strategy for upward mobility. This mobility has been observed to be primarily economic in nature. In the context of transnational marriages, Constable (2004) proposes the concept of ‘paradoxical hypergamy’. Though there could an improvement in the economic resources of the migrants after moving to a richer country yet their social stature remains low in the husbands’ family as well as in the society, in fact it is lower than their parent society. A lot of literature (Agnihotri, 2000; Arokiasamy, 2004; Bhatnagar et al., 2005; Mazumdar & Krishnaji, 2001; Nayar, 2006; Patel, 2007) exists on the quantitative and qualitative aspects of the missing girl child but few scholars have ventured to study the phenomenon of marriage squeeze (Das Gupta et al, 2003; Kuangshi, 2014) and even fewer (Chaudhry & Mohan, 2011; Kaur, 2004, 2009) so in India. These few studies have analysed the phenomenon at the location of marital residence only. They have not covered any state which provides the brides. This study for the first time attempts to do so with a view to uncover the layers of gendered equations and structural inequities in these marriages. This paper begins with an analysis of the socio-economic context within which these CRM are situated. It also outlines the impact of the extended families on the gendered relations existent in the families. This would help in understanding the dilemmas and contestations of a migrant bride vis-a`vis the rest of the family and also vis-a`-vis the larger society. Oppressive structures have been further discussed in this section to understand the undercurrent of violence prevalent in these marriages. Political economy of Haryana sets the backdrop for understanding the structural roots of oppression in Haryana. This discussion is followed by a note on the sex ratio of Haryana/Rewari where these marriages are set. Next section includes some stories from the print media. This is followed by the section on the methodology of the research. Severity of the problem of cross-regional marital alliances, contestations and adjustments are delved in the subsequent section. Implications of the findings have also been discussed at length in this section. Concluding section gives a synthesis of the primary data.

78

SONALI MUKHERJEE

SOCIO-ECONOMIC CONTEXT OF CROSSREGIONAL MARRIAGES This paper is set in the rural backdrop of Haryana; therefore a review of rural India follows. Indian economy is primarily dependent on agriculture. Agriculture employs over 70% of the total rural workforce whether on their own lands or on the lands of the others (Haque, 2012). According to National Sample Survey Organisation (2006, Government of India), about 70% of the rural households have marginal landholdings of average size of less than 0.40 hectares and 10% households are landless. Of the total farmers, 41.8% are women and 58.2% are men. Among the females, 83.2% are agricultural workers and among the males, only 66.5% are agricultural workers. Still women do not have land rights as land records are mostly never in the name of the women. Most of the landholdings are jointly held by extended families with the patta5 in the name of the male patriarch. This is a practice prevalent everywhere in the country. Though women work as much as or even harder than their male counterparts yet they do not have ownership of land (Agarwal, 2002). Also the economic returns from selling the dairy products or the food crops are not shared with women (Saxena, 2011). This necessitates their dependence on men. Reliance on men for economic reasons renders women helpless and gives the men a sense of power. Situation is no different in Haryana. Literature has proved beyond doubt that violence has a significant correlation with notions of power. Powerlessness has a connotation of weakness and weak have perennially been suppressed. Further ‘oppressive social systems and processes facilitate conditions within which violence is more likely to occur (Hollomotz, 2012). This paper reflects on this very nature of the perpetuating social system. Most of the rural households are comprised of extended families. This prevents the land from being disaggregated into small plots and also control of the property remains with the head of the household. There exists a vertical ladder where power rests with the patriarch who is at the vertex and decreases as one comes down the ladder. Of course, this structure is gendered as men are more powerful than the women. Sons who are economically capable have a better stake than even senior female members of the household. But, among the women, hierarchy is based on age and also on social position. For example, mother-in-law would subjugate all daughter-in-laws; wife of the eldest son will yield more power than the wife of the younger son. If the younger son/s is/are incapable of earning a livelihood then he/they along with his/their families would

Structural Basis of Gender Violence in Cross-Regional Marriages

79

be excluded from all important decisions of the family. In any case, economic decisions are mostly taken by men. Decisions regarding marriage alliances, relationship maintenance, day to day household affair etc include women. Indian rural men enjoy a sense of power as the women in their household are economically dependent on them and this creates space for domestic violence or violence in intimate relations. These abusive equations could be between husband and wife; father and daughter; father-in-law and daughter-in-law; and even between mother-in-law and daughter-in-law; any two individuals where one is dependent on the other. Such is the case of the Haryana women residing in Haryana. They do not have much say in anything, what to talk of women coming from outside the region. There may not always be a volatile expression on the front but a surreal calmness which conveys a lot. As most feminist scholars would agree that gender violence in the context of India is rooted primarily in the patriarchal systems that pervade every sphere of the society (Sahoo & Raju, 2007). Variable access to resources further creates barriers among people (Agarwal, 1994; Mohanty, 2004). Systemic differences based on gender and class is further highlighted by caste.6 Hierarchical orders based on caste, methodically defines the inequalities that add to the gendered equations in the society (Dube, 2001). Violence in any form impedes the quality of life of any individual who is in an oppressive situation. Also, the migrant women are from impoverished rural families. Many of them are married off in lieu of money. One key question in these marriages is the extent of agency shown by the migrant bride. Following discussion on political economy further outlines the structural roots of supremacy of men especially of the upper caste. This in turn allows one to visualise the constricting domains of existence of the women in Haryana society.

POLITICAL ECONOMY OF HARYANA7 Haryana was created out of Punjab in 1966 to accommodate the demands of the non-Sikh and non-Punjabi speaking population of Punjab. Haryana and Punjab flourished during the Green Revolution in the 1980s. Currently, Haryana scores high on economic indicators such as per capita income, per capita investment, per capita plan expenditure and resource mobilisation. Jats, an upper caste group constitute the single largest caste

80

SONALI MUKHERJEE

group with 24% population. Dalits are the next largest group at around 20%. The districts of Bhiwandi, Jind, Sonepat, Rohtak, Jhajjar and Hisar constitute the Jat ‘heartland’ where the populations of the Jats are more. The social superiority of the Jats rests on their ownership and control of the major portion of land in the state, which, in turn, has allowed them to control the labour of those who are subordinate to them in the caste hierarchy. Jats are today at the pinnacle of practically every formal and informal institution in Haryana.

Sex Ratio of Haryana Haryana has a skewed sex ratio in favour of men for the past one century at least (Mukherjee, 2013). While the earlier years of this adversity could be because of migration patterns or enumeration problems but later years are surely ascribed to ‘son preference’. Studies have revealed a declining trend in fertility since the 1990s. Pressures of planning a family has led to fertility decline which has inadvertently tilted towards the decline in female population (Kaur, 2007). These inferences do imply that they are battling the issue of lesser females and thus more bachelors in their community for quite some time. The sex ratio at birth (SRB) is still very low as evidenced by the 2008 2010 SRS data in all the states of India irrespective of caste, class or religion. Since this study is based in Haryana, census figures of Haryana are being quoted in this text. Sex ratio at birth (SRB) is 848 in 2008 2010.8 SRB is an indicator of preference for the birth of a particular gender. India has shockingly continued its travail towards eliminating female foetuses shirking all legal bindings. Medical and para-medical practitioners are abetting female genocide shamelessly without any moral underpinnings. Child sex ratio (CSR for 0 6 years of age) in Haryana is no better at 819 in 2001 and 830 in 2011. Though it might have increased by a few points in the overall figures but Rewari, the district which has been studied for this paper show a further decline (see Table 1). Table 1. Comparative Child Sex Ratio of Rewari District 2011.

2001 and

Districts

2001

2011

Rewari

811

784

Source: Registrar General of India, Census Division.

Structural Basis of Gender Violence in Cross-Regional Marriages

81

Media Reportage from Haryana These instances have received wide coverage from the print media from time to time. Some of the stories which were carried by popular newspaper are being quoted here.

Story 1: Onu (aged 20 years) from Assam, was sold to Ram Phal and Kirpal Singh, two brothers in Haryana for a sum of Rs 20,000 by her parents in Assam. On paper, she is only married to Ram Phal, an ageing widower. But, the younger brother, Kripal Singh, 45, has conjugal rights over Onu. ‘Staying here’, she says, ‘is perhaps better than being a burden on my parents’. Both brothers, who are poor vegetable farmers, cannot afford to buy two brides. So they decided to share one. After years of searching for a bride in their own state, the brothers say they were compelled to hire a tout who led them outside Haryana to Assam to broker a ‘deal’ with Onu’s poor family.9

Story 2: Fourteen-year-old Tripala from Jharkhand, was sold by her povertystricken mother in Ranchi (Jharkhand) to Ajmer Singh of Dahola village, Jind district, Haryana for $300. Ajmer Singh claimed that he was ‘buying’ her as a bride for his brother. After having been kept captive in shanties in Delhi for some time, she was brought to Dahola and kept locked for days in Singh’s farmhouse. And when she refused to sleep with his brother who couldn’t find a wife, she was beheaded with a machete.10

Story 3: Seema’s parents sold her to a man almost twice her age because that was the only way out of poverty. Seema is 21 years of age and she hails from West Bengal. The man, Mahavir Singh, is a farmer from

82

SONALI MUKHERJEE

Haryana. ‘I am happy here’, Seema says, because there is enough to eat. Life in Bengal was wretched. She would barely get a bowl of rice a day to eat back home. Here, with curd, milk abundantly available, she has become healthier and nods to being asked if she is treated well.11

Story 4: Sumari was widowed young. She is presently 40 years of age. She was brought from her native village to northern India and sold to a man who abused her sexually and physically. He imprisoned her and her daughter in his house. ‘It went on for three or four months, until he sold us off to another man’, says Mumtaz, Sumari’s daughter, who is now about 20. Sumari is one of the luckier women, having eventually found a good husband after being sold repeatedly in a thriving human trade in northern India that is blamed on local customs and a shortage of women. ‘There aren’t enough girls here. Locals won’t give their girls to widowers, ageing and handicapped men’, says Fatima, Sumari’s neighbour in Mewat, a district of Haryana state where there are 820 girls for every 1,000 boys under the age of six.12

Story 5: Hameedan (aged approximately 30 years) from Assam, was cheated by an acquaintance who laced her food with a sedative and smuggled her out of her native village. She woke up in Mewat, Haryana, about 16 years ago, and has been here since then. ‘I did not know anyone here. I did not know the language. It took me 10 years to establish contact with my family’, she says. She was a teenager at the time while the man she was forced to marry was in his mid-50s. He is violent and beats her up routinely. ‘But a woman cannot hit back at her children’s father, can she’? she asks.13

Structural Basis of Gender Violence in Cross-Regional Marriages

83

METHODOLOGY Present research began with the idea of understanding the phenomenon of cross-regional marriages that was being regularly reported in the print media. Stories (few of them mentioned in the last section) reflected suffering as well as hope. This amalgamation led to curiosity regarding these marriages and thus emerged the idea of doing an in-depth analysis in Haryana where the brides are sent and in West Bengal, one of the states from where the brides are procured. It must be mentioned here that this paper is the outcome of a study done for a much bigger research project on ‘Structural Basis of Gender Violence: Analysis of implications of adverse sex ratio in Haryana and West Bengal’, sponsored by Indian Council of Social Science Research (ICSSR). Three districts from each state have been chosen for the full analysis. Rewari, Jind, Ambala from Haryana and the districts of South 24-Parganas, Murshidabad and Cooch Behar from West Bengal have been chosen. For the purpose of this paper, data from Rewari and Cooch Behar has been presented in this paper. This paper relies on primary source of data. Standard social science techniques such as observation, in-depth interview, household survey and FGDs were used for collection of data. Districts chosen for the larger study are spatially distributed in the state. Also Rewari has one of the lowest CSR (child sex ratio) in the state, Jind is reported to have cases of fraternal polyandry and Ambala has one of the highest female literacy rates in the district (data would help us in drawing a link between ‘literacy of the girl’ and ‘her voice in marital decisions, both pre marriage and post-marriage). Murshidabad is a trafficking prone district where adolescent girls under the guise of marriage are being sexually exploited. 79% of the girls in the district are married when they are still in their adolescence, in most cases below the age of 16 years. So is trafficking rampant in South 24-Parganas. Cooch Behar is known to be an area where women and girls both from Bangladesh and Bhutan are trafficked to India through this district. In all the three districts, migration for the purpose of marriage and for the purpose of trade is taking place simultaneously, almost in an overlapping fashion. The topic of research is extremely sensitive, therefore it necessitated that enough precaution was taken in the field to prevent being ousted from there without collecting enough information and also to save oneself from physical injury. Author is a Bengali Brahmin (upper caste) residing in Delhi. She is well versed in both Bengali and Hindi languages which are used in West

84

SONALI MUKHERJEE

Bengal and Haryana respectively. In both the States, informants were women especially the migrant brides in case of Haryana and migrant bride’s mother or sister in case of Bengal. Other than these, ten men from both the districts were interviewed using interview guides. Since Rewari is only 93 km away, therefore it was possible to commute from Delhi everyday to collect data. As per the initial research plan, a village with approximately 200 households was being sought with the presumptuous idea that one could locate 20 30 such cases and one would be able to draw a rough estimate of the presence of cross-regional marriages14 and also be able to correlate between the two types of marriages, that is cross-regional marriages and the marriages that are taking place within Haryana. Initial investigations were carried out to identify the villages where such cases were present. Local shopkeepers and others who are present at the market places were queried first. It was realised from these interactions that such cases are there all over Rewari but not every village would have them. Again, it was informed that in a village of 200 300 households, one could possibly locate 3 4 cases only. They directed towards some villages which were visited. In every village, Pradhan15 of the gram panchayat was met. After failed attempts in two villages, third village, Ghurkawas acknowledged the presence of such cases. Initially it was informed that this village had approximately 200 households, however after the survey was completed; it was found that only 144 households with only three cases of CRM were present there. Other villages where CRM cases were picked included Nayagaon, Gokalgarh, Lishana, Gangaicha Ahir, Bikaner and Kakodia. All these villages were located within a periphery of 10 15 kms. in the Rewari block only. In total, 34 case studies were collected over a period of 15 days in January February 2014. Next challenge was to enter the CRM household and talk to the migrant bride. In some cases it was not just the physical space of the narrow lanes and small houses that was constricting but also the familial atmosphere. Possibly the upper caste look16 of the author helped her in gaining entry into their households but the initial information seemed guarded and sketchy. Even after she explained her purpose of visit, they were in doubt. She spent some time discussing her study with them and also affirmed that it meant no harm to them. Each household took a lot of time. Despite that, it was difficult to gain trust of some of the families. In these households, talking to the migrant bride was a little constraining than those cases where people were very open about their decision to import a bride. Also, 22 of the migrant brides spoke Bengali as they were from the States of West

Structural Basis of Gender Violence in Cross-Regional Marriages

85

Bengal, Tripura and Assam. Bengali is one of the languages spoken by the author as well, therefore it was easier for her to get the information without much interference from the rest of the family. Fieldwork in the district of Cooch Behar, West Bengal had different challenges. With the help of a local NGO, Anwesha and civic police, cases were identified in Toofanganj block. Without the help of these people, it would have been impossible to identify these cases as locals did not seem much aware of such marriages. Some of the traders did mention the existence of such cases. Villages that were visited were Santoshpur, Krishnopur, Debogram, Devcharai and Jhaljhali. Cases had to be literally handpicked from these villages. In total, 17 such cases could be identified and family members could be interview where girls have married into Haryana, Rajasthan, Uttar Pradesh, Kashmir and Himachal Pradesh. After reaching the households, getting information on these marriages was not difficult as they seemed forthcoming. These interviews took place in March 2015.

FINDINGS FROM THE FIELD Marriage in Haryana: Social Obligation or Personal Need Before studying the phenomenon of cross-regional marriages, it was thought imperative to gather some local explanation that necessitates marriage in Haryana so much so that they have to import brides from other States. ‘Marriage is absolutely essential for everyone in the Haryana society’ proclaimed an informant of Rewari district. Procreation was cited as one of the important reasons for marriage as it creates descendants to take the family name forward and also generates extra hands for labour. It also denotes power for some as more members (sons) in a family adds to the strength of the family. There is also an understanding that children (especially sons) could take care in their old age. Apart from these rationales, loneliness is another motive for marriage. It was felt by many of the older women that if their sons do not find a bride, there would be no one to feed their sons after them. Men also feel the need for companionship as one of them said ‘friends and family are there for some time but a wife is always there with you whether in joy or in grief’. These local explanations do indicate an acute need for marriage for Haryana men/boys. Both unmarried girls and boys are looked down upon by the society after they

86

SONALI MUKHERJEE

have crossed their marriageable age. In Haryana, girls are married between the age of 18 and 22 years and boys are married between 22 and 26 years. If anyone crosses this ripe age of marriage, they are pitied as they slowly start fading from the local marriage market. Men17 are referred to as ‘bechara’ (Kaur, 2008). It is assumed that they would either have to remain unmarried all their lives or they would have to bring brides from other regions. As already mentioned marriage entails patrilocal residence which means the married couple has to stay near or at the residence of the man’s father. As noted in the field, except eight women, all the others were married to men who were more than 20 years older to them. These women themselves got married when they had crossed 20 years of age which was considered late in their native places. Therefore, these women are married to men who are either old (19 of them) or physically challenged (two of them) or youngest of the brothers (five of them). So, she is situated at the bottom rung of the family. She is constantly reminded of the fact that she has been brought from a poor family where she was hardly fed. In a way, her marital family always portrays an image of obligation towards her. Since she is economically dependent on the man, it helps the family to dominate her completely. This entails a form of violence which is structurally rooted in the patriarchal Haryana society.

Caste Barriers Redefined: Altruism or Vanity In Haryana, these marriages are quite rampant and identity of those families who have procured brides from outside is well known. Caste endogamy is important for Haryana marriages. Much of the contestations regarding these marriages reverberate around the caste identity which the upper caste and class men strongly hold against these marriages. As one elderly upper caste Haryanvi man contemptuously said, ‘who knows where these women come from? What is their caste? Do they even have a birth certificate’? This sentiment does reflect on the fact that out of the 34 migrant brides interviewed, 31 strictly claimed that their birth families belonged to the same caste as their husbands. Closer examination revealed that only 19 men have married into their own caste and rest others have either married lower or higher caste girls. In some sense, it is heartening to know that the rigid caste and sub-caste boundaries are ruptured through these marriages but at what cost? It is not by choice but by compulsion of out-marriage and hence subverting the entire argument through the

Structural Basis of Gender Violence in Cross-Regional Marriages

87

casteless identity of women. Then the biggest question remains about them being marginalised by the society at large. Here one could argue as to what is a bigger cause for concern; marginalisation through caste or through poverty? Overarching dimension of impoverishment is probably the single largest reason for these brides to come to Haryana. Barring three cases, none of the brides had any complaints regarding their marriage but they did not seem happy. Three of the brides from Cooch Behar and two from Assam were never married. They were brought into Haryana promising work but were sold off. Two of them were bought (rescued as they mentioned) by their present partner (an old man aged 84 years with two wives) with whom they have children. Blanchet’s study documented similar stories of distrust where they were sold to men by dalals (middle-men).

Marriage in West Bengal: Social Obligation or Offloading the Burden? ‘Kanyadaan is mahadaan’ (giving away a daughter is the greatest offering a man can ever make) and ‘pitridai theke mukti’ (to be released from the obligation of a father) are some of the oft quoted and oft repeated answers. Both these convictions are associated with another one which states that the father would be able to go to heaven after his death if in his lifetime, he has fulfilled these obligations. Almost 100% of the informants used either of these or both these reasons to get their daughters married. Deeper probing does reveal that societal and familial pressures of marriage are huge. Often comparisons are made between girls who could got married with the ‘unfortunate’ ones who could not. Also it is believed that there is a fixed time for every life cycle ritual and if they are not fulfilled in time, deviant behaviour could be expected. ‘Girls could go astray’ as one old lady claimed. Also ‘boys start stalking the girls’, another one said. In any case, it is unthinkable for them to keep ‘grown up18’ girls at home. Average age of marriage for girls was found to be 16 years. Out of 17 cases, only three respondents said that they are willing to let their daughters study if they themselves wish to. Dowry demanded at the time of marriage is very high. An unemployed youth would also demand a minimum amount of a lakh of rupees other than motorbike, etc. Also, the number of girls in a family is usually more than one in such cases. Poverty and also stories of other girls successfully getting married outside the region seemed propelling reasons for the 17 girls to venture out. In five of these cases, respondents narrated that since

88

SONALI MUKHERJEE

these girls could not be married here, therefore they went out of Bengal and got married in Bihar. Also, a number of families have reportedly migrated to bigger cities in search of work. Two sisters reportedly got married in Haryana when they accompanied their parents to Haryana who worked as wage labourers. It has been narrated by many that once a girl or a family goes out, they pull the others. Similarly for marriage, these girls start acting as go-betweens for other nuptial arrangements.

Fixing a Marriage Alliance: Commodification of Brides Fixing these alliances is not much difficult as in 25 of the cases, groom’s family claimed that the girl’s sister is married in a nearby village; therefore it was easy for them to find such a match. Deeper probing does reveal the existence of a go-between who is herself a migrant bride. Northern states of Haryana and Punjab have the reputation of having abundance of food and prosperity due to their agricultural prosperity. This imagery is used by most go-betweens to cultivate newer client base and get more brides into this region. They even justify their act by saying that they are doing social service or serving the cause of humanity by getting these poor fellows brides and these poor girls a better deal in marriage by getting them fed and clothed. Interestingly all these go-betweens seemed to reside near the respective groom’s maternal grandfather’s house.19 This does point out that an attempt is made to find an alliance quietly and incorporate the bride20 into the family with least fanfare. As far as the bride’s opinion is concerned, 25 migrant brides said that they were aware that they were being married to Haryana men. This awareness certainly does not always include their choice. This is not a surprise as women are not able to exercise their choice in most parts of patriarchal India. But there are distinct regional differences based on rural-urban divide; access to education; access to better employment opportunities; etc. But in this case, what unites all these women is their poverty. Inability to meet the dowry demands of the local grooms makes the local marriage market inaccessible to them. At the same time, parents of these migrant brides are lured into settling their daughters into households which are economically better-off than theirs. Hence the marriage puzzle is solved by each group fulfilling the demands of the other group. So, this spatially hypergamous union is sanctified longevity by its sheer need based approach.

Structural Basis of Gender Violence in Cross-Regional Marriages

89

CROSSING THE REGIONS: CHALLENGES AND BEYOND Adjusting to New Ways of Life Most of these brides had to learn a new way of life starting from learning a new language, new forms of cooking, different household and farming chores to newer forms of religious practices more so if the bride was originally a Muslim. Different food habits probably hit the hardest as Haryana is typically wheat eating area and Bengal, Assam, and Tripura are rice eating areas. Also Bengalis consume fish on a regular basis which is completely missing from the diet of Haryana. Concept of the purdah was another difficult issue to handle for all these Bengali brides. To negotiate between all kinds of restrictive rules of behaviour was very constricting. Everyday rituals and customs are somewhat elaborate among the Bengali Hindus in their birth place which they miss in Haryana. Life cycle rituals are also different and they have to follow the customs of their marital home. But when they go back to their parents place, they follow the rituals prevalent there. They have a similar arrangement for most of their religious festivals. But the Muslim brides are worse affected in this as their religious identity is completely hidden and they have to follow Hindu pantheon of belief which is nothing less than a catastrophe as evidenced in Blanchet’s study. Religious conversion is unacceptable and hence the returnee brides or disassociated brides (ones who ran away from their marriage) had a tough time back in Bangladesh. Adjusting to Violent Physical Behaviour Based on the above discussion it is quite clear that the pattern of transition from one’s own (birth) cultural beliefs to prevailing (Haryana) cultural idioms was not a smooth progression but an adaptation of a newer cultural state. Apart from these adjustments, these women had other reasons to worry. Intolerance of men towards their women, even for petty reasons leads to domestic violence. Was it a cause for culture shock for these migrant brides? Local women are more adapted towards unreasonable behaviour of their men but migrant brides from the east of India are not used to so much physical violence. As stated by a local woman, ‘as such in our society, there is no concept of ruthna-manana.21 If the wife gets upset over something, she still has to carry on as if nothing has happened. But if quarrels go out of control then our birth family members intervene to

90

SONALI MUKHERJEE

make peace for our sake’. With these functional mechanisms present, these local women are able to sustain themselves better. Fifteen migrant brides complained of domestic violence which is physical in nature. Though most of them have got used to being beaten once in a while yet they feel they can handle verbal abuse better. Two important reasons cited for violence in the initial days of marriage, included, (i) not able to speak Haryanvi language properly, (ii) not able to follow the customary rule of purdah as unable to distinguish between different kins and identify the jeth22 properly. While discussing the issue of violence, they did not seem much agitated; on the contrary, they seemed complacent as they had resigned to their fate. It is a distinct possibility that these women have no common sharing platform to share their life struggles as they have very few friends with whom they can share their woes and also they lack the community support from which they can derive some solace.

Differing Relations to Work In different societies, there are differing men’s and women’s relationship to agriculture. In Haryana, women have an important role in agriculture and allied activities. Especially among the lower caste and class, women play an active role in the agriculture field which is not so among the higher caste and class. These migrant brides who married into poorer households (18 of the informants) found this to be a daunting task in the beginning. Many of them had never made cow-dung cakes prior to their marriage. Postmarriage, they had to not only make them but also carry them on their heads over a distance. All these works along with other household chores of cooking, cleaning etc. was a handful for these women. They have struggled their way into learning these activities. These activities renders Haryana women of lower caste and class, more mobile unlike the higher caste women who are like bounded entities and can move only within a restricted space and that too with the permission of the male members of their families. Also many feminist scholars in other spheres have documented that chance of women’s mobility increases post-marriage as honour of a woman is safely handed over to her husband. But these migrant brides might be facing restrictions in mobility due to other reasons like social ostracisation. Though it was not clearly stated during the course of the fieldwork yet it was felt that these women are mostly confined to their homes. They go out only to perform some of the household chores like making cow-dung cakes or getting fodder for the animals, etc.

Structural Basis of Gender Violence in Cross-Regional Marriages

91

This mobility is restricted to the periphery of the house and the agricultural field. They did confirm their participation in local festivities along with local Haryana women but local women themselves did not seem comfortable about mixing with these migrant brides. Identity Crises: Constrained Existence Social affiliations through marital relations and network of friends are deep rooted in any society, especially when a couple takes up virilocal residence post-marriage. But in the case of these marriages, movement between consanguinal kins of the husband and the couples themselves is not much. If other kins have also got migrant brides, then there is some chance of movement but otherwise there is hardly any social interaction. In that case, at times of sudden crises, whether economic or any other (say, inter or intra family disputes), these couples cannot seek help from their relatives. This kind of social exclusion emerging from restrictive behaviour of the various individuals stands to corner these couples to the fringes of the society. Further, children of these couples would find it extremely hard to get social recognition in Haryana society. Couples who were interviewed for another earlier study were either childless or having young children. Actual state of their social standing could not be gathered during this study. Remarkably, landed farmers do not face such acute social ostracisation. Two such couples who had land (5 acres and above) are getting along fine with Haryana society. In one case, there are three children (two daughters and one son) out of whom, one daughter and one son have studied engineering and the other daughter has done graduation in science and also got a degree in bachelor of education. Both daughters are presently married to Haryana boys. It was pointed out that the boy could marry a girl of his choice from anywhere (not necessarily from Haryana) not because of shortage of girls in Haryana but because youth these days prefer to marry equally qualified partners. Similar logic cannot be applied to sons of poor couples. A senior man from landed Gujjar Caste queried, ‘who will give them daughters? They will also get brides from other regions’.

CONCLUSION There are two sets of phenomenon which have parallel existence in Haryana. One is the continuous in-migration of brides into Haryana due to

92

SONALI MUKHERJEE

a compelling need for reproductive and productive labour for all those Haryana men who are unable to secure a bride from within. Other phenomenon is the denigration of such couples especially the migrant brides. Overwhelming patriarchal Haryana society dominated by the landed upper caste would like to treat them as aberrations but there is no denying the fact that under this facade lies the reality wherein it is a common strategy to build these marital alliances with the so-called ‘outsiders’. These alliances have existed since the last four decades (Blanchet, 2003) and can be safely foreseen that they would exist for another few decades. Due to acute shortage of brides, Haryana men are not averse to marrying women of other caste, sub-caste, religion, region, etc. Territorial limitations are not seen. Caste and gotra exogamy is overlooked, religious biases are swept away. The tightly bound marital norms are violated with convenience. Existence of these marriages for a long time implies that the prevalent social practices have been altered or are in the process of being altered. Closer examination does reveal that the web of social relations might have changed for some families but the hegemony of the dominant landed class or upper caste pervades every sphere of social activity in such a way that the host culture has subsumed within itself all forms of variants which have come through in the form of brides from other cultures. It is the migrant brides who have transformed their identities to suit the Haryana men. As seen earlier, very little space is available to them to negotiate and seek changes in their husbands. These hypergamous23 unions are lopsided. Power dynamics are already stated as economic transactions are weighed heavily against these women. Earlier literature and present study leaves little room for doubt that it is the women who have to bear with uncertainties and insecurities whereas the men who are getting these brides, either have a nonchalant attitude or they take out their angst on these hapless women. Genocide of female population has not consciously escaped the law makers and other power brokers of the society but they have been deliberately put aside and somewhere there is a pretence that the problem existed because of ‘others’ or issue of unavailability is leading to alliances in ‘others’ families. Men of landholding caste somehow want to problematise the issue through the channel of lower caste and class. Dilemmas and contentions in these marriages which the migrant brides face in these marriage alliances seem to go unrecognised by the larger sections of the Haryana society. There are three primary reasons for this social exclusion. Firstly, Haryana women do not share an equivalent platform with their men. If these women with their backing of local social network are not able to legitimise their claims of equity, then it is impossible for the migrant brides

Structural Basis of Gender Violence in Cross-Regional Marriages

93

to do so. Secondly, there is a palpable anxiety of the upper class and caste men over the stereotypical images of migrant brides and their children as ‘inferior others’. Hence their needs and aspirations are sidelined. Thirdly and most importantly, Haryana is ruled by a parallel set of governance, the Khap Panchayats. These panchayats comprise of men of higher caste and class who carry on with their sycophancies. They have meticulously managed to create a reign of terror by castrating or killing those men or women or couples who have dared to challenge them. Khap panchayats have devised a technique by way of social sanction which ensures the marriage of the limited pool of Haryana women to their own men (Kaur, 2010). Therefore their kith and kin; and their inner circle of friends and allies are able to sustain the Haryana gene pool by this mechanism. Now the question is how larger section of Haryana has sustained its cultural matrix despite so much of inter-mixing of Haryana gene pool. One would imagine changes in the dress pattern, language, rituals, festivities etc. of Haryana but it is amazing to see that these migrant brides have transformed themselves completely. They talk, walk, eat, behave or perform any function like any other Haryana woman in similar socio-economic condition. There are possibly two reasons for them to adapt themselves to Haryana, one is the social pressure of acceptance and the other which is linked to the former, psychological pressure of abandonment. Acceptance is also at two levels, one at the level of the family where she needs to be fully incorporated and also at the level of the community where she and the family in which she is married needs to be accommodated. If she is unable to make her way into the family then it is likely that she would have to bear the brunt of violent behaviour and also desertion. In that situation she would have very little chances of survival. She probably cannot go back to her parents; if she could then she would never have married so far in an alien land with a different culture. Since there are other women willing to come to Haryana in marriage, her husband could get another woman. Therefore, she is always a victim of circumstances. These migrant brides are coming from impoverished families and travelling long distances to get married into a different culture. Also the marriage proceedings are bare minimal and they take place at the birth place of the bride. All these factors along with the initial problem of communication gaps within husbands’ households tend to weaken their status. Further, they are getting married to men of low status and therefore their status is further low in society. Women in Haryana do not have any property rights despite the presence of Hindu Succession (amended) Act 2005. Migrant brides are further away from ascertaining their rights as they do not have

94

SONALI MUKHERJEE

the support of their kin members to support their claim. As documented by Chowdhry (2012) women in Haryana, like their counterparts in any other patriarchal setting, face violence within the households. Dominance of the men is reinforced through violence. There seems to be an existence of social sanction towards such behaviour. Nonetheless Chowdhry has differentiated between public and private sphere where there is a great level of resistance in private sphere towards patriarchal oppression. It is altogether a different thing that men would never accept such resistance publicly as he would be shunned by the society for acquiescing to his wife. In such a situation, getting a share in the property by the women is not only difficult but also inappropriate as per the customary social sanction. This study would like to conclude by stating that these migrant brides are not gaining any social stature by marrying into the Haryana society. Some of them have access to better economic resources and hence are happier in host society, yet the social standing of these women is no better than what it was prior to migration. Therefore, there is an urgent need for a policy recommendation towards the protection of these migrant women and their children and allocating resources for improving their political, economic and social life in Haryana society. Absolutely fundamental is the wider cognisance of these marriages. To achieve this some form of law should be enforced upon the community to approve these forms of marriages. Otherwise, under no circumstance the power brokers of the Haryana society can be made to accept these brides. Compulsorily enforce the marriage registration act of Haryana so as to bring security to the brides and their children, in case of desertion; trafficking; reselling; and any other form of abuse. Wider campaigns are required to improve their accessibility to community resources and services and also to garner a humane support from the local people who tend to denigrate them on the basis of their skin colour; and the way they have been incorporated into the Haryana society. Policy intervention is required to initiate active participation of these brides in some economic activity which would not only alleviate their poverty but also instill confidence in them.

NOTES 1. Census of India (2011). 2. Scores of media reports exist on this phenomenon as more often than not body of an infant girl child is recovered from a drain or the backyard of a private hospital, etc.

Structural Basis of Gender Violence in Cross-Regional Marriages

95

3. Many such incidents are reported in John et al. (2008). 4. Women who are coming from outside the region and getting married in Haryana are referred so. 5. It is the land record maintained in the land revenue offices. Though computerisation of land records have started since 1992 1993, it has been achieved in very few states. 6. Castes are occupationally based endogamous groups which stratifies the Indian society. It plays a crucial role in perpetuating violence, especially among the dalits (earlier referred to as untouchables, currently classified under scheduled castes) in rural India. 7. This entire section is taken from Caves (2014). Speak! The Truth Is Still Alive: Land, Caste and Sexual Violence against Dalit Girls & Women in Haryana. 8. Data provided by Sample Registration System, Office of Registrar General of India. 9. http://www.wunrn.com/news/2006/09_11_06/091706_india_bride.htm 10. Ibid. 11. Ibid. 12. http://afp.google.com/article/ALeqM5jP_j2BMuoejUW0bTK77IA8jX85oQ 13. Ibid. 14. No such data is available in the secondary source also which could possibly estimate the percentage of such marriages. Anecdotal evidence and some of the research studies have thrown light on the presence of such marriages only. 15. Chief of the elected village council. 16. It cannot be said with great conviction whether any such look really exists. Long years back a reasonably renowned feminist writer told the author that she looks like an upper caste Hindu Brahmin and it might cause hindrances to her collecting information in a muslim dominated area. It of course, proved false. Looking back it is realised that one might look like anything; it is the interpersonal skill and ability to build rapport with the informants which helps in getting information. Probably, one’s looks help in as much as the entry into the household is concerned; much depends on one’s expertise as an ethnographer. 17. Conditions in which women remain unmarried are not being discussed in this paper as it is not within the purview of this paper. 18. Girls who have attained puberty were referred so in the field. 19. It must be mentioned here that all these cross-regional marriages in Haryana are first generation marriages in their respective families. So the groom’s parents were married as per Haryana marriage rules and groom’s mother comes from a distant village. 20. These marriages are usually performed at the bride’s birth place. Some of them are also registered in court. 21. Ruthna implies when wife gets annoyed, husband somehow manages to make up which is referred as manana. 22. Literally means brothers elder to her husband, in this context, all kin types in the category of jeth. 23. Brides are coming from a poorer region to an agriculturally rich, hence economically prosperous region.

96

SONALI MUKHERJEE

REFERENCES Agarwal, B. (1994). A field of one’s own: Gender and land rights in South Asia. Cambridge: Cambridge University Press. Agarwal, B. (2002). Are we not peasants too? Land rights and women’s claim in India, Population Council Report. Agnihotri, S. B. (1995). Missing females: A disaggregated analysis. Economic and Political Weekly, 30(19), 2074 2084. Agnihotri, S. B. (2000). Sex ratio patterns in the Indian population: A fresh exploration. New Delhi: Sage. Arokiasamy, P. (2004). Regional patterns of sex bias and excess female child mortality in India. Population (English Edition, 2002), 59(6), 833 863. Bhatnagar, R. D., Dube, R., & Dube, R. (2005). Female infanticide in India: A feminist cultural history. New York, NY: State University of New York Press. Blanchet, T. (2003). Bangladeshi girls sold as wives. Study submitted to the Academy for Educational Development (AED, Dhaka). With the collaboration of Anisa Zaman, Hannan Biswas, Monzur Hasan Dabu and Masuda Aktar Lucky. Caves, J. L. (2014). Speak! The truth is still alive: Land, caste and sexual violence against dalit girls & women in Haryana. A Report by Women Against Sexual Violence and State Repression, Progressive Printers, Delhi. Census of India. (2001). Retrieved from http://www.censusindia.gov.in/ Census of India. (2011). Retrieved from http://www.censusindia.gov.in/ Chaudhry, S., & Mohan, T. D. (2011). Of marriage and migration: Bengali and Bihari brides in a U.P. village. Indian Journal of Gender Studies, 18(3), 311 340. Chowdhry, P. (2005). Crisis of masculinity in Haryana: The unmarried, the unemployed and the aged. Economic and Political Weekly, 40(49), 5189 5198. Chowdhry, P. (2012). Infliction, acceptance and resistance: Containing violence on women in rural Haryana. Economic and Political Weekly, 47(37), 47 59. Constable, N. (Ed.) (2004). Cross-border marriages: Gender and mobility in transnational Asia. USA: University of Pennsylvania Press. Das Gupta, M., Jiang, Z., Li, B., Xie, Z., Chung, W., & Hwa-Ok, B. (2003). Why is son preference so persistent in East and South Asia? A cross-country study of China, Indian and the republic of Korea. Journal of Development Studies, 40(2), 153 187. Das Gupta, M., & Li, S. (1999). Gender bias in China, South Korea and India 1920 1990: Effects of war, famine and fertility decline. Development and Change, Special Issue in Gender, Poverty and Well-being, 30(3), 619 652. Dube, L. (2001). Anthropological explorations in gender: Intersecting fields. New Delhi: Sage. Haque, T. (2012). Land policies for inclusive growth. New Delhi: Concept Publishing Company. Hollomotz, A. (2012). Disability, oppression and violence: Towards a sociological explanation. Sociology, 47(3), 477 493. John, M. E., Kaur, R., Palriwala, R., Raju, S., & Sagar, A., (2008). Planning families, planning gender: The adverse child sex ratio in selected districts of Madhya Pradesh, Rajasthan, Himachal Pradesh, Haryana and Punjab, ActionAid and IDRC. Kaur, R. (2004). Across-region marriages: Poverty, female migration and the sex ratio. Economic and Political Weekly, 39(25), 2595 2603.

Structural Basis of Gender Violence in Cross-Regional Marriages

97

Kaur, R. (2007). Declining juvenile sex ratios: Economy, society and technology explanations from field evidence. Margin: The Journal of Applied Economic Research, 1(2), 231 245. Kaur, R. (2008). Dispensable daughters and bachelor sons: Sex discrimination in North India. Economic and Political Weekly, 43(30), 109 114. Kaur, R. (2009). Strangers as spouses: Marriage implications of India’s skewed sex ratio. Paper presented in SASNET Seminar in collaboration with The Department of Economic History, Lund University. Kaur, R. (2010). Khap panchayats, sex ratio and female agency. Economic & Political Weekly, 45(23), 14 16. Kuangshi, H. (2014). Marriage squeeze in China: The past, present and future. Journal of Family Issues, 35(12), 1642 1661. Mazumdar, V., & Krishnaji, N. (Eds.) (2001). Enduring conundrum: India’s sex ratio. Delhi: Rainbow Publishers Limited. Miller, B. (1997). The endangered sex: Neglect of female children in rural India. Delhi: Oxford University Press. Mohanty, M. (2004). Class, caste and gender. UK: Sage Publications. Mukherjee, S. (2013). Skewed sex ratio and migrant brides in Haryana: Reflections from the field. Social Change, 43(1), 37 52. National Sample Survey Organisation, 62nd Round. (2006). Household consumer expenditure. Ministry of Statistics and Programme Implementation, Government of India. Nayar, U. (2006). The unborn daughters of Delhi. Delhi: The Women Press. Patel, T. (Ed.). (2007). Sex-selective abortion in India: Gender, society and new reproductive technologies. New Delhi: Sage. Saxena, N. C. (2011). Women: Land and agriculture in India. UN Women Report. Sahoo, H., & Raju, S. (2007). Domestic violence in India: Evidences and implications for working women. Social Change, 37(4), 131 152. Visaria, L., & Ramachandran, V. (Eds.). (2007). Abortion in India: Ground realities. New Delhi: Routledge.

This page intentionally left blank

FAMILY VIOLENCE IN SYRIA Muaweah Ahmad Alsaleh ABSTRACT Purpose Family violence is a universal problem which is beginning to grow to a significant scale in Syria. Although it has existed for a long time, the actual characteristics of this scourge in our country are not known. The main aim of this study was to evaluate the family violence in Syrian society. Design/Methodology/Approach This work consisted of an epidemiological approach to domestic violence in Syria during the year 2010. A questionnaire had been developed which is used for the study of the sociodemographic profile of families and the study of violence in the family. This study has been conducted on a survey of 365 women. Findings The analysis of the results reveals the following characteristics: 16% of the women in the sample were victims of physical violence. The youth is a risk factor for these women, the age range most affected by violence (45%) is that of women aged between 20 and 40 years. Violence affects all social, economic, and cultural classes; anger is an aggravating factor of domestic violence; in fact, 27.3% of spouses who assaulted their wives were in an angry state. Originality/Value The violence in the family is a very sensitive issue and very common, but the exact prevalence of violence in the family is

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 99 127 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009005

99

100

MUAWEAH AHMAD ALSALEH

not known. Therefore, the violence in the family is underdiagnosed. An urgent response plan is needed to reduce the spread of this scourge and its consequences. Keywords: Family violence; women; negative impact; prevalence; underdiagnostic

INTRODUCTION Family can be one of the most protective forces in the lives of children and youth (Maalouf & Campello, 2014). But family violence is a serious public health problem, the level of this violence is seriously increasing in Syria and in the world. Although it has existed for a long time, the policy and research ignore the real characteristics as well as the consequences of this plague in our country. In 2008, 535,000 people died by homicide in worldwide and 95% of these deaths occurred in low- and middle-income countries (LMIC) (Maalouf & Campello, 2014; Murray, Cerqueira, & Kahn, 2013). Acts of violence account for an estimated 1.43 million deaths worldwide annually (Siever, 2008; WHO, 2007). The violence against women in developing countries is emerging as a growing concern for public health practitioners because it is well known that women are vulnerable to many forms of violence, with domestic violence being the most common form (Heise, Pitanguy, & Germaine, 1994). Domestic violence, or Intimate Partner Violence (IPV), is defined by the American Medical Association as a pattern of physical, sexual, and/or psychological abuse by a person with whom the victim has had an intimate relationship (Flitcraft et al., 1992). The World Health Organization (WHO) defines domestic violence as any act of “gender-based violence that results in, or is likely to result in, physical, sexual, or mental harm or suffering to women, including threats of such acts, coercion, or arbitrary deprivation of liberty, whether occurring in public or in private life.” (WHO Consultation, 1996). Domestic violence is a significant cause of morbidity and mortality for women in every country where these associations have been studied (WHO Consultation, 1996). In a literature review of population-based studies, Krug et al. found that 10 69% of women reported that they had experienced physical abuse from a male partner (Krug et al., 2002). This paper is a study of epidemiological data on violence in the world and its causes, and consequences, in particular, in Syria.

101

Family Violence in Syria

Objective of the Study The main aim is to study the causes of violence in Syria and its impact on women.

Forms of Violence in the Family Worldwide, there are different types of violence against women. International reports agree that violence can be physical, verbal, sexual, and psychological (mental). Any other forms, for example, economic or social violence, have emerged as a result of the differences in the definitions and measurement methods of violence (Cohen et al., 2000; Haj-Yahia, 2000b; Heise et al., 1994; Michau & Naker, 2003). Physical Violence The use of physical force, which involves evident effects, causes physical injuries and may lead to chronic conditions (Campbell, 2002; Karol, Micka, & Kuskowski, 1992; Nduna & Goodyear, 1997). Physical violence is the most visible and common form of violence. It includes such acts as beating, pulling, biting, kicking, and inflicting bone fractures and burns. This form also includes denial of basic needs, such as food, water, sleep, and shelter, the lack of which causes physical harm (Abu Ghazaleh, 2008). Sexual Violence Coercion into sexual contact; encouragement of or intimidation into prostitution; or coercion into viewing sexual acts. It includes rape, sexual harassment, unaccepted sexual remarks, and resort to sexual practices that contravene religious and ethical rules of sexual contact (Abu Ghazaleh, 2008). Psychological, Mental, Emotional Violence Any act that may result in direct or indirect harm threatening a woman’s emotions, feeling of self-esteem and ability to control her life. It includes threats, insults, scorn, hurtful words, deprivation, reproach, and throwing doubts on her capacities, which may lead to various degrees of feelings of lowliness, despair, and depression. Emotional abuse can have severe longterm effects (Bustamante- Gavino, 1998).

102

MUAWEAH AHMAD ALSALEH

Economic Violence Depriving a woman of the right to dispose of her economic resources or take part in making financial decisions that concern her, impact her future, and make her totally dependent on others. Economic violence includes denial of dispensation of properties and spending on her basic needs, disinheritance, denial of owning properties, lower wages for women, and exposure to economic exploitation (Abu Ghazaleh, 2008). Social Violence Any act or behavior that deprives a woman of her social rights, such as interference in her social relations, isolating her from society, or severing her from her social contacts (Abu Ghazaleh, 2008).

Violence against Women in Arab, Asian, Non-Western, and Western Societies Data from industrialized and developing countries as well as from transitional countries (see Table 1) provide an overview of the global problem (Khan, 2000; WHO, 1999). Table 1.

Domestic Violence against Women (Khan, 2000; WHO, 1999).

Industrialized countries Canada Japan New Zealand Switzerland United Kingdom The United States France

Germany

17 29% of women reported being physically abused by their partner (Cohen et al., 2005). 59% of women reported being physically abused by their partner. 20% of women reported being physically abused by their partner. 20% of women reported being physically abused by their partner. 25% of women reported being physically abused by their partner. 28% of women reported being physically abused by their partner. 1.8% declared to have been victims of emotional blackmailing, 37% were victims of psychological pressure, 2.5% of the women were victims of physical aggression, and 0.9% endured rape and other enforced sexual acts (Country report, 2012a; Hagemann-White, 2010; Jaspard et al., 2000). 23% of women have experienced physical violence from their intimate partners and 7% sexual violence from their intimate partners. The majority (99%) of perpetrators of intimate partner violence are men (Country report, 2012b; Federal Ministry for Family Affairs, Senior Citizens, Women and Youth, 2003).

Family Violence in Syria

103

Asia and the Pacific Cambodia India Korea Thailand Karachi

16% of women reported being physically abused by their partner; 8% report being injured. 45% of women reported being physically abused by their partner. 38% of women reported being physically abused by their partner. 20% of women reported being physically abused by their partner. 34 49.4% of the women had been physically abused by their husbands. 15% had been abused even while pregnant. 72% of physically abused women were anxious/depressed (Fikree & Bhatti, 1999; Fikree, Razzak, & Durocher, 2005).

Middle East Egypt Syria

Jordan Africa Kenya Uganda Zimbabwe

35% of women reported being physically abused by their partner (Diop-Sidibe´, Campbell, & Becker, 2006). 26% of women reported being victims of psychological violence, 18% physical violence, and 4% sexual violence (Alkesh & Hassan, 2010; Brik & Alkesh, 2011). The size of the violence problem ranges from 7.7% to approximately 78%, regardless of types and forms. 42% of women reported being physically abused by their partner. 41% of women reported being physically abused by their partner. 32% of women reported being physically abused by their partner.

Latin America and the Caribbean Chile Colombia Mexico Nicaragua Haiti

26% of women reported being physically abused by their partner. 19% of women reported being physically abused by their partner. 30% of women reported being physically abused by their partner. 52% of women reported being physically abused by their partner. 16.4% of women reported being physically abused by their partner (Gage, 2005).

Central and Eastern Europe/CIS/Baltic States Estonia Poland Russia Tajikistan

29% of women aged 18 24 fear domestic violence, affecting 52% of women 65 or older. 60% of women reported being physically abused by their partner. 25% of girls (and 11% of boys) reported unwanted sexual contact. 23% of women reported being physically abused by their partner.

104

MUAWEAH AHMAD ALSALEH

In summary, studies on violence in the family against women in several non-Western patriarchal societies (such as surveys of several countries in Asia, Rani & Bonu, 2009; Armenia; Kazakhstan; Turkey; Nepal; Bangladesh; etc.) have revealed perspectives that are similar to those prevailing in Arab societies (Maier et al., 1999).

Violence in the Family Is Underdiagnosed Rates of violence reported in victimization surveys are higher than “officially reported rates” because the political and personal filters are removed from the reporting (Craig & Rowell Huesmann, 2003). In general, the prevalence of violence in the family is very common and very high in the world, but the exact prevalence of violence in the family is not known. Therefore, the violence in the family is underdiagnosed.

Causes of Violence in the Family The reported reasons for abuse included talking to strangers and jealousy. Abuse has also been found to be associated with low income (Cohen et al., 2005; Diop-Sidibe´ et al., 2006; Gage, 2005; Harwell, Moore, & Spence, 2003) and there are other causes (see Table 2) (Heise et al., 1994). Measuring societal rates of violence is fraught with difficulties because politics affects definitions and reporting of violence and/or crime (Craig & Rowell Huesmann, 2003).

Prevalence of Family Violence in Syria Domestic violence has long been an issue in Syria. In 2006, its first countrywide domestic violence survey, sponsored by the United Nations Development Fund for Women, found that between 10% and 25% of Syrian women had experienced physical violence (Leigh, 2014; Roumani, 2006). In 2004, the Syrian Ministry of Labor & Social Affairs conducted a research experiment about gender in its establishments. The research showed that 25.5% of the workers at the Ministry consider there to be discrimination between males and females in the

Family Violence in Syria

Table 2. Cultural

105

Factors that Perpetuate Domestic Violence (Heise et al., 1994).

1. Gender-specific socialization. 2. Cultural definitions of appropriate sex roles. 3. Expectations of roles within relationships. 4. Belief in the inherent superiority of males. 5. Values that give men proprietary rights over women and girls. 6. Notion of the family as the private sphere and under male control. 7. Customs of marriage (bride price/dowry). 8. Acceptability of violence as a means to resolve conflict.

Economic 1. Women’s economic dependence on men. 2. Limited access to cash and credit. 3. Discriminatory laws regarding inheritance, property rights, use of communal lands, and maintenance after divorce or widowhood. 4. Limited access to employment in formal and informal sectors. 5. Limited access to education and training for women. Legal

1. Lesser legal status of women either by written law and/or by practice. 2. Laws regarding divorce, child custody, maintenance, and inheritance. 3. Legal definitions of rape and domestic abuse. 4. Low levels of legal literacy among women. 5. Insensitive treatment of women and girls by police and judiciary.

Political

1. Underrepresentation of women in power, politics, the media, and in the legal and medical professions. 2. Domestic violence not taken in account seriously. 3. Notions of family being private and beyond control of the state. 4. Risk of challenge to status quo/religious laws. 5. Limited organization of women as a political force. 6. Limited participation of women in organized political system.

nominations for high administrative positions, and that 51.7% of the men share in the domestic chores against 19.7% and sometimes 28% who of men do not share. In answer to the following question posed to both males and females “Do you believe that the woman is capable of management and planning?” about 40% answered in the negative (Arabi, 2006). Syria is a state party to the CEDAW since 2003. In addition, Syria remains internationally committed to investigate, prevent, prosecute, and punish authors of violence against women because it expressed reservations on some articles (Nasar, 2013). Although

106

MUAWEAH AHMAD ALSALEH

politicians and institutions have recognized the problem in the Syrian society, they do not intervene except in severe cases causing access to hospital and/or in case of death because the violence remained within the family as an internal affair. According to the results of the study (Alkesh & Hassan, 2010; Brik & Alkesh, 2011), the majority of women (45.1%) reported that they had experienced verbal abuse (conflict) and physical violence from their husbands. The most widespread methods of physical violence are slapping, hitting, or punching. With regard to physical violence, 18% reported receiving beatings by their husbands. The majority of women reported feeling stressed by the conflict with their husbands (26% of the women) and sexual violence (4% of the women). The study showed that the most widespread methods of psychological violence are first, screaming and reprimanding; second, criticism, ridicule, and defamation; and third, cursing and spitting. The results of the study showed of several forms of sexual harassment, such as the follow-up messages and harassment by telephone (4.3%), considering prosecution (7.6%), harassment to speak (5.8%), touching (4.7%), coercion into sexual behavior (4.9%), attempted rape (3.0%), and rape (2.4%) (Alkesh & Hassan, 2010; Brik & Alkesh, 2011). The violence in the Syrian family increased after the Syrian conflict. Experts working there say that the displacement has triggered a shift in traditional family dynamics, leading to a fast escalation in domestic violence among the nearly 3 million-strong refugee population. This includes physical, emotional, and sexual abuse, along with cases of neglect (Leigh, 2014). A majority of victims refuse to speak about violence they have been exposed to, in particular, when the offense has a sexual dimension, because they fear social stigma attached to these abuses. Syrian women exposed to sexual abuses subsequently found themselves victimized not only by the crime itself, but also by enduring the silence that surrounds the crime and the social pressure related to it. Uncovering instances of violence such as rape and sexual violence is therefore one of the most necessary, though challenging tasks in the Syrian context (Nasar, 2013). This displacement in which men who were gainfully employed at home find themselves without jobs, apartments, cars, or a normal life has also led to shifts in traditional family dynamics. Fathers who were once breadwinners find themselves unable to provide for their wives and children, leading to anger and frustration that often turns violent (Leigh, 2014). The rate of violence against women in Syria and other countries is high (see Table 3).

Author Series (1999)

Kristof and WuDunn (2009) The Economist (2011)

Country

Violence against Women. Rate of Violence against Women

Barbados Canada Egypt New Zealand Switzerland The United States Philippines and Paraguay India

30% of women are victims of domestic violence (Series, 1999). 29% 34% 35% 21% 33% Less than 10%

Turkey

42% of women over 15 have suffered physical or sexual violence (The Economist, 2011). 37% were victims of psychological pressure and 2.5% of the women were victims of physical aggression (Country report, 2012a; Hagemann-White, 2010; Jaspard et al., 2000). 37% of women in Germany have experienced some form of violence in their lifetime (Country report, 2012b; Federal Ministry for Family Affairs, Senior Citizens, Women and Youth, 2003).

France

Federal Ministry for Family Affairs, Senior Citizens, Women and Youth (2003) Haj-Yahia et al. (2012) and Khawaja and Barazi (2005) El-Zanaty, Hussein, Shawky, Way, and Kishor (1996) and HajYahia et al. (2012) Alkesh and Hassan (2010) and Brik and Alkesh (2011)

Germany

Jordan

Egypt

Syria

70% of women are victims of domestic violence (Kristof & WuDunn, 2009).

44.7% of the women have suffered spousal physical violence during lifetime, 17.4% have suffered in the past year, and 14.2% have suffered during pregnancy (Haj-Yahia et al., 2012; Khawaja & Barazi, 2005). One out of three women has been subjected to marital physical violence at least once in their lifetime. 45% and 17% have suffered physical violence at least once and at least three times in the past year, respectively (Haj-Yahia et al., 2012; El-Zanaty et al., 1996). 26% women have been subjected to psychological violence and 18% physical violence and 4% sexual violence (Alkesh & Hassan, 2010; Brik & Alkesh, 2011).

107

Jaspard et al. (2000)

Family Violence in Syria

Table 3.

Author

Country

Haj-Yahia (1999, 2000a, 2000b, 2012)

Palestine

Douki, Nacef, and Belhaj (2003) Al-Badayneh (2012)

Egypt, Palestine, and Tunisia Arab Society

Domestic violence and pregnancy Bacchus, Mezey, Bewley, United Kingdom and Haworth (2004) Huth-Bocks, Levendosky, The United States and Bogat (2002) and Torres, Campbell, and Campbell et al. (2000) Ireland

108

Table 3. (Continued ) Rate of Violence against Women 91%, 53%, and 39% of the women reported experiencing psychological abuse, physical violence, and sexual abuse at least once in the past year, respectively (Haj-Yahia, 1999; Haj-Yahia, 2000a, 2000b; Haj-Yahia et al., 2012). Surveys show that at least one out of three women is beaten by her husband (Douki et al., 2003). Rates of physical violence against Arab women ranging from (26%) to (87%) (AlBadayneh, 2012). 3.4% of women are victims of domestic violence during pregnancy (Bacchus et al., 2004). 3.2 33.7% of women are victims of domestic violence during pregnancy (HuthBocks et al., 2002; Torres et al., 2000).

MUAWEAH AHMAD ALSALEH

12.5% of women are victims of domestic violence during pregnancy (O’Donnell, Fitzpatrick, & McKenna, 2000).

Family Violence in Syria

109

Impact of Violence against Individuals, the Family and Society Family violence includes any attack against the physical or mental integrity of a person. The violence against the family unit is a multifaceted problem; it is equally a social, health, economic, and cultural problem. It is also one of the most significant causes of family disorganization, whose consequences affect all family members, especially children. The WHO has declared violence as a health problem in view of the physical injuries and disabilities it inflicts. Furthermore, violence causes mental disorders, such as anxiety, tension, and insecurity (Abu Ghazaleh, 2008). The consequences of violence do not stop at physical injury and negative mental effects. The violence has economic effects, depicted in the costs of providing health and psychological care to both victims and perpetrators. The public sector’s health and social institutions invariably bear these costs, which are spent on treating the physical injuries and mental disorders resulting from violence and which lead to negative impacts on production and investment. Undoubtedly, the impact of violence is harder on the economies of poor countries (Abu Ghazaleh, 2008). There are variables associated with violence such as behavioral problems, stress management, prosocial behaviors, family aggression, and conflict (Maalouf & Campello, 2014). In addition, violence against family has consequences and implications on divorce, children (see Table 4) (Abdul Wahab, 1976), the academic achievement of pupils, children’s mental health, diseases or mental disorders, underachievement, failing, academic failure, drug abuse, moral deviation, behavioral deviation, etc.

Violence and Physical Health Consequences Increased health problems such as injury, chronic pain, gastrointestinal, and gynecological signs including sexually transmitted diseases, depression, and posttraumatic stress disorder are well documented by controlled research in abused women in various settings. Physical health effects can manifest as poor health status, poor quality of life, and high use of health services. Battering is a significant direct sign and indirect risk factor for various physical health problems frequently seen in healthcare settings. Intimate partner violence is one of the most common causes of injury in women. Battered women were more likely to have been injured in the head, face, neck, thorax, breasts, and abdomen than women injured in other ways (Campbell, 2002).

110

Table 4.

MUAWEAH AHMAD ALSALEH

Impact of Family Violence on Children and Its Negative Effects on Academic Achievement (Abdul Wahab, 1976).

From the physical point of view

From the social point of view

From the psychological point of view

When the child pretends to be tired and exhausted and complains that his health does not allow him to go to school, or sometimes he tries to harm himself, either to have compassion from others or because of the desire to revenge of them. 1. Inability to socially adjust and has a negative attitude toward the environment around him or her. 2. Tendency to exercise violence on others and harm them. 3. Often inclined to practice bad habits such as theft and smoking. 4. Difficulty with social adjustment tends to be isolated and introverted. 1. Lack of self-confidence, the feelings of anxiety and frustration. 2. Negative attitude toward family and school. 3. Occurrence of neurological disorders and loss of confidence and feeling comfortable. 4. Hatred and tendency to be aggressive and to vandalize. 5. Inferiority and instability. 6. Negative and severe emotions.

Violence, Mental Illness, and Mental-Health Effects Mental disorders are neither necessary, nor sufficient causes of violence. The major determinants of violence continue to be socio-demographic and socioeconomic factors such as being young, male, and of lower socioeconomic status. Members of the public undoubtedly exaggerate both the strength of the relationship between major mental disorders and violence, as well as their own personal risk from the severely mentally ill. It is far more likely that people with a serious mental illness will be the victim of violence. Substance abuse appears to be a major determinant of violence and this is true whether it occurs in the context of a concurrent mental illness or not. Those with substance disorders are major contributors to community violence, perhaps accounting for as much as a third of self-reported violent acts, and 7 out of every 10 crimes of violence among mentally disordered offenders (Stuart, 2003). Depression and posttraumatic stress disorder, which have substantial comorbidity, are the most prevalent mental-health sequelae of intimate partner violence. Depression in battered women has also been associated with other life stressors that often accompany domestic violence, such as

111

Family Violence in Syria

childhood abuse, daily stressors, many children, changes in residence, forced sex with an intimate partner, marital separations, negative life events, and child behavior problems (Campbell, 2002).

Solutions and Alternatives to Reduce the Phenomenon of Domestic Violence and Its Negative Impact According to the landmark World Report on Violence and Health (WHO, 2002), the first steps needed toward violence prevention are: (i) the collection of as much basic knowledge as possible on the magnitude and nature of the problem and (ii) research identifying causes and correlates of violence which might be modified through interventions. Given its economic, cultural, and social context, it is possible that risk factors for crime identified in other contexts have different effects in each state, and this needs empirical testing (Murray et al., 2013; Rutter, 1999). Family skill interventions have been found to be effective in encouraging safe and nurturing relationships between parents (or caregivers) and children in their early years and as such preventing many problem behaviors including violence (Maalouf & Campello, 2014). There are solutions and alternatives to reduce the phenomenon of domestic violence and its negative impact (see Table 5) (Issawi, 2003). In summary, the most literature on violence against women is concentrated on the Western societies but very little information is known about violence against women in the Arab society (Al-Badayneh, 2012), especially violence against women in Syria. This study contributes to the understanding the causes of violence against women in a random sample of women in Syria.

Table 5.

Preventive Solutions (Issawi, 2003). Therapeutic Solutions

Holding seminars and conferences aimed Create family organizations concerned with the at increasing consciousness of family. treatment of parents who abuse to deal with Increase the level of per capita income and their children, etc. improve the standard of living, etc.

112

MUAWEAH AHMAD ALSALEH

Purpose This research aim is to explain the phenomenon of violence against women in Syria, disclosure of its causes, knowledge of the underlying factors behind of its occurrence, and its negative impact and find appropriate solutions.

METHOD Participants This is a cross-sectional epidemiological study, having covered a sample of 365 women and students. Data was collected in 2010. The questionnaire was completed after an explanation of the purpose of the investigation and after obtaining informed consent. Self-administered questionnaires were distributed to 365 female students and women. Responses were collected on the same day. A random sample of women was chosen between the ages of 19 and 45 years. This study relied on the university and the villages to prepare for the research in the local community. Participation was voluntary and responses were anonymous.

INCLUSION CRITERIA Women and female students between 19 and 45 years old were included in the study. All women and female students were Syrian.

MEASURES Questionnaire on the Phenomenon of Violence and Its Effects In this study, the questionnaire used was in the form of questions directed at women and female students on the phenomenon of violence, its causes and its effects. The questionnaire was developed to serve the current search, based on literature review and the local community. This tool consisted of 10 items that measured some variables. The anonymous questionnaire assessed educational level; economic level; causes of violence; and its

113

Family Violence in Syria

negative impact on the child’s personality; ask for help and resort to in case she was subjected to violence; and divorce. The questions could be scored with a dichotomous yes/no fashion. In this study, the dichotomous (yes/no) scoring approach was used for answer the questions.

DATA ANALYSIS In this study, descriptive analysis was conducted to determine ranked proportions and frequencies for each of the causes of family violence (see Table 6) and to analyze each item included in the tool.

RESULTS Socio-Demographic Aspects Some variables measured in this study (number of Sample n = 365) were, respectively: age; sex; educational level; economic level; causes of violence and its negative impact on the child’s personality; ask for help and resort to in case she was subjected to violence; and divorce. The average age of women was 26 years old with a minimum of 19 years and a maximum of 45 years. There is a predominance of age range between 20 and 28 years (45%). The original rural women represented 76% of study participants. Twenty percent of women had completed secondary education. More than half (56.2%) were housewives.

Causes of Family Violence in This Study The prevalence of battered women was 18%. Women in the range of ages between 20 and 40 years most affected by the violence (45%), while 34% of battered women were aged between 23 and 38 years. Most battered women (65%) were housewives. The most widespread cause of violence in this study was the conflict with the in-laws (46.3%). In 10.38% of cases, women reported violence caused by claiming money to their spouse and 9.82% reported that unemployment was the cause for which they were beaten. A large majority of women (79%) agreed that husbands are not completely responsible for the

114

MUAWEAH AHMAD ALSALEH

violence because the in-laws and the husbands’ daily life circumstances play a major role in causing violence (see Table 6). Forty-one percent of women reported that family violence is one of the social problems which are affected by the conditions prevailing in the society. Table 6.

Causes of Violence Reported by Women.

Causes

Percent

In-laws problem (relatives families) Asking husband for money Unemployment Husband demanding money Working problem Other Total

46.3 10.38 9.82 4 3.2 26.3 100

In the overall sample, 46.3% of women say that the primary cause for man’s violence against woman are the interference of the family (in-laws, relatives’ families). Problems with in-laws are consistently implicated in violence against their sons’ wives. These problems are complicated by the normative nature of patrilocality where the majority of the sons stay within their parents’ home even after marriage, while married women join their husbands in their in-laws household. Thus, the in-laws often have a vested interest in perpetuating practices of control and power over their sons’ wives (daughters-in-law), because they have finally obtained a relative position of power in the family. This represents a socioculturally specific form of patriarchal bargaining that has significant implications regarding addressing violence against the woman endemic in the Arab society and in Syrian society, with equally distinct psychological implications. Thus, these problems with in-laws mean violence against women remains under diagnosed, because the sons are afraid of being forced to leave the house if they did not do so under the authority of the Father. Furthermore, the husbands believe that wives are responsible for the violence against them in this case. In addition, the second cause is economic (26.4%). Finally, the other causes such as difference in opinion because men’s desire to subject women to his sovereignty are present in family. After the results of the study, women at risk are: • Women aged less than 40 years; • Housewives with no fixed income and dependent on their spouse; • Women with low socioeconomic status;

115

Family Violence in Syria

• Women who were victims of violence in their childhood and experienced family violence while living with their parents. • These women are likely to have psychological problems including depression, anxiety, stress, etc.

DISCUSSION The violence affects the lives of millions of women worldwide, of all socioeconomic and educational classes. However, the studies conducted reported that women with low educational status, unemployed women, and newly married women were more frequently abused. The current studies support those of previous studies done in Egypt (34%) (Diop-Sidibe´ et al., 2006), Canada (17%) (Cohen et al., 2005), Haiti (16.4%) (Gage, 2005), Bangladesh (42%) (Bates et al., 2004), India (30%), and some squatter settlements of Pakistan (34%) (Fikree & Bhatti, 1999; Fikree et al., 2005; Harwell et al., 2003; Shaikh, 2000), but the figures are much higher. In France, 16% experienced a climate of violence between their parents and 28% of women who suffered abuse in childhood have been abused in the past 12 months against 6% of those who experienced no difficulty in their childhood (Jaspard, 2011). In Arab societies, family violence against women and/or children is not normally considered as a crime because of the prevailing sociopolitical structure and cultural context in these societies (Haj-Yahia et al., 2012). In Arab and Islamic countries, domestic violence is not yet considered a major concern despite its increasing frequency and serious consequences (Douki et al., 2003). The study of domestic violence in Arab and Islamic countries carries the distinction of being a source of disagreement regarding the concept of this violence. The lines between traditional, pathological, and religious are not yet subject to a societal consensus. In the Arab and Islamic countries, domestic violence is still not considered a major problem, although its frequency is high enough. Investigations in these countries showed that the proportion of women who were exposed to violence by their husbands is at least one in three women (Douki et al., 2003). Studies have reported prevalence rates of physical abuse among Arab women in the Arab countries ranging from (26%) to (87%) (Al-Badayneh, 2012). But others Studies have reported rates of physical violence against Arab women between 30% and 53% (Haj-Yahia et al., 2012).

116

MUAWEAH AHMAD ALSALEH

The Jordanian social culture accepts the use of violence with children or women as a kind of discipline, and this acceptance is supported by cultural and social norms (Al-Badayneh, 2012). Furthermore, The Palestinian social culture accepts the use of violence with women (56% of the men and 41% of the women have reported that wife beating is acceptable in Palestinian society) (Maier et al., 1999). The social culture in Syrian society accepts also the use of violence with children and/or women as a kind of discipline. Because of the prevailing customs and traditions in these societies, the use of violence against children and/or women is very frequent. Although the use of violence is not accepted in Islam, these societies use violence as a tool for control and discipline against women and/or children. In answer to the question of who the woman would resort to in case she was subjected to violence, more than 66% said they would resort to their families. This shows that the issues related to violence are still being solved inside the family or are not publicized. Twenty percent of the sample said they would resort to no one, and this is a large percentage. It’s rarely said they resort to the police (0.4%) and/or to the physician (0.2%). At the various educational and economic levels, women facing violence would resort to their folks, friends, or relatives but woman would not resort the physician and the police, this is related to the prevailing customs and traditions in the society. The issues related to violence are still being solved within the family because the patriarchy, the prevailing customs and traditions in the society, and cultural view see that violence against the woman is a personal and family problem rather than a social problem. Furthermore, this violence deals within the family without using interventions from strangers and people tend not to wish to discuss it, especially with outsiders. Similarly, studies conducted among Palestinians and Jordanians reported that there is very low support for the punishment of violent husbands and for the use of social services and/or police intervention. In addition, the mistrust of the social and/or law services in general causes the low tendency to seek assistance from these services that fight violence against women in society (Haj-Yahia et al., 2012). In the question related to the economic level in this sample, this study found out that the economic causes of domestic violence represent 26.4% of the causes of violence in the family (asking husband for money 10.38%; unemployment 9.82%; husband demanding money 4%; working problem 3.2%); and that this violence is linked to the profession whereby violence decreases in the upper professions in terms of culprit and victim, and increases in the lower professions and within the marginalized segment of the society.

Family Violence in Syria

117

In answer to the question of why battered women do not seek divorce? More than 77% said it is because they are keen on their children’s interests. This was the main reason for all women. Furthermore, this was the main reason in each country, in particular, in Arab and Islamic countries. In general, the woman does not seek divorce because there is no place to resort to with her children. This is directly related to the number of children, and inversely to the woman’s age. Generally, the divorcee finds it difficult to return to her folks because the society holds the wife responsible for the failure of the marriage and asks her to sacrifice more than the man for the sake of the continuity of the marriage. In addition of holding the wife responsible for the failure of the marriage, the society considers the divorced women a different point of view that can probably carry and lead the inferiority look to these women because of the prevailing customs and traditions in this society. In answer to the question of domestic violence magnified of low the educational level? 62% of women reported that weakness of level of education among the couple leads to the emergence of differences between them. In the overall sample, the most frequent violence against the woman was the imposition of the husband’s viewpoint on the wife. The highest rate among the housewives was 65%, because their status is linked to their low educational level. But among the female employees, the rate was 6%, because of their higher educational level. In the question related to women’s educational level and violence in the sample, this study concluded in general that the increase in the educational level of the woman and her family enables her to attain her rights and to take the best decisions about her life with regard to the choice of knowing the fiance before marriage (by 50% of the illiterate women, 54% at the primary education level, and 72% at the university level). The women educational level is proportionately affected by the rise of the educational level of the father or mother. In answer to the question of wives represent the majority of domestic violence victims? 56% of women reported that violence is directed by the husband to his wife. The violence has a negative effect on their personality, their self-esteem, their image in front of their children. And in answer to the question of violence and conflict between parents has a negative effect on the child’s personality? 48% of women reported that the conflicts between parents will weaken the child’s personality and threaten his social adaptation and psychological stability. In this study of 365 women, the prevalence of battered women during their lifetime is 18%. The health studies of women and domestic violence conducted by WHO in 71 countries showed that the prevalence of physical

118

MUAWEAH AHMAD ALSALEH

violence by an intimate partner during the life of a woman is between 13% and 61%. The prevalence varies between 23% and 49% for most countries (OMS, 2002). These estimates of the prevalence of violence are based on: (1) Sample population; (2) Diagnostic criteria used. Various factors influence the quality and comparability of data on domestic violence, including: • • • •

Definitions of the variables of family violence; Variations in the selection criteria for study participants; Differences resulting from data sources; The willingness of respondents to speak out of an experience from the violence. The estimates in samples vary widely because of differences in:

(1) (2) (3) (4) (5)

The definitions of violence; Scales; The samples studied; Different study methodologies; Criteria for diagnosis incompatible and inconsistent.

In addition, domestic violence in the culture is a very sensitive issue, is a private matter, and people tend not to wish to discuss it, especially with strangers. Because of these factors, the violence in the family is underdiagnosed and underdetected. This study shows that there are motives and reasons behind violence against women: subjective from within the person, economic, educational, psychological and social motives emanating from prevailing customs and traditions in Syrian society. All of these are related to the prevailing societal culture among the people. These motives vary from one country to another one because of the social and cultural context, a patriarchal ideology and of the prevailing customs and traditions in each society. This study concludes that the prevalence of domestic violence among such women is high and this violence puts the women under considerable stress. The main reported reasons for the violence are public health issues which should be addressed by health professionals. There is therefore a need for women’s empowerment through increased schooling and improved financial capacity so that they are able to participate in the decision-making

119

Family Violence in Syria

process of a household. The media could play a leading role in persuading society to be more supportive of women and their role in society. But there are barriers to women empowerment through schooling and to the media in Syrian society: women do not know their rights and these rights do not reach them, thus they are exposed to violence in their life. Education is a means of obtaining work and independence in the decision and equality with man who takes the economic decision in the family. Although the government is taking all measures to ensure education for all, especially for girls and women. In addition, Syrian law provides free and compulsory education up to the age of 15, but many fathers (neo-patriarchal or patricentric) force their daughters to leave education. This could be to give them away to an early marriage where an education would not be needed, to send them into employment, or for other reasons. Because of the prevailing customs and traditions in Syrian society, the most of the women do not inherit or do not get any education. Nevertheless, the father (neopatriarchal or patricentric) and the university graduates know that the woman has the right to inherit the land and get an education. Islam, through the teaching of the Koran, encourages the education and passing on of inheritance to women and girls. However, women are prevented from learning and/or deprived of inheritance. So, awareness is the first step toward a more supportive and tolerant society because the woman’s awareness is linked to her education. Hundred percent of the university graduates know the woman’s right to be nominated for elections and inherit, compared to 80% among the housewives. Twenty percent of the illiterate women do not know the woman’s right to inherit land and to be nominated for elections because they are not in contact with the regulations and laws.

CONCLUSION The findings of this study were based on a relatively small sample. Results may not be generalized to other women in Syrian society because the sample size is too small. Further work with larger sample size is ongoing. This study cannot assume that sample is representative female population in Syria in general. However, it can be assumed to represent low-income women and low education level women in Syria and especially in countryside because this is the case in the most of the Syrian women.

120

MUAWEAH AHMAD ALSALEH

Data was collected from 365 women between ages 19 and 45. The violence experienced during lifetime of the respondents was measured. Sixteen percent of women in Syria have lived some act of physical violence/abuse since the age of 19. The types of violence committed are: women do not take part in politics, are not elected, are deprived of inheritance, and are deprived of education. All of them are linked to the fact that the woman does not know her rights and these rights do not reach her, thus she is exposed to violence in her life. On the one hand, the permanent relations between in-laws (family, father and mother, etc.) and their son and on the other hand, with his wife are probably the primary cause for man’s violence against the woman, according to the sample of this study. Similar to previous research findings (Al-Badayneh, 2012), violence is a private matter and a very sensitive issue in the Syrian society. The violence against women may be understood in light of social and cultural context of a patriarchal ideology. In addition, this violence is one way of maintaining dominance of husband over wife. Further studies using nonclinical and/or clinical samples are needed to replicate the results. In a study concerning women in the Syrian society, violence against women was related with their educational level and mental distress (Maziak & Asfar, 2003). Similarly, in this study of Syrian women, violence against women was associated with their educational level and with their economic level. It has been reported that one-quarter of all men and approximately one-half as many women report acts of physical aggression after age 18 (Robins & Regier, 1991). Thirty-seven percent of women in Germany have experienced any act of physical violence/abuse since the age of 16, while 13% of women have disclosed experiencing any act of sexual violence/ abuse since the age of 16 and 40% have experienced one of the forms or both. Sexual harassment has been experienced by 58% of women in Germany and 42% of women experienced psychological violence/abuse (Country report, 2012b; Federal Ministry for Family Affairs, Senior Citizens, Women and Youth, 2003). Domestic violence is a major public health problem that exists in all countries and in all social, economic, religious, and cultural backgrounds. Family violence, even if it appears less important than other forms of widespread violence, poses a serious threat to the individual and society because the consequences caused by the imbalance of power within the family cause an imbalance of values and behaviors, especially among children. This is causing a regeneration of violence both in the family and in society.

Family Violence in Syria

121

The violence and crime in the family present in all healthcare settings with many different physical and mental-health problems and demographic characteristics. Women’s multiple roles in the society put them at greater risk for developing common mental disorders such as anxiety and depression, and many argue that social factors play an important role for the preponderance of common mental problems seen in women (Maier et al., 1999; Maziak et al., 2002; World Health Organization, 2000). The prevalence of psychiatric distress in the Syrian sample was 55.6% (Maziak et al., 2002). In addition, this study shows that the violence in the family is underdiagnosed and underdetected. In addition, the violence in the family weakens the child’s personality and threatens his social adaptation and psychological stability and makes the child had psychiatric diseases and aggression against others, etc. The intertwined relationships between abuse and physical and mental health outcomes should be of interest to healthcare practitioners as well as researchers. The violence is a risk factor for many healthcare problems and crimes in the family, but the causes and extent of such a risk are only beginning to be understood. In addition, the violence is only ever partially resolved through legal processes (Das, 2000; Hastrup, 2003; Rew, Gangoli, & Gill, 2013). Despite political and institutional recognition of the problem, violence remained within the family as an internal affair, particularly in the Arab countries, including Syria. Hence, the difficulty in attaining information in this study. The underreporting and delayed reported of violence against women and its types as sexual, physical, psychological, mental, emotional, economic, and social violence are endemic, making an assessment of the magnitude difficulty. Despite this, violence has received little attention from academics and policy makers. Because of the prevailing customs, traditions and masculine domination, and little attention from academics and policy makers, women have been subjected to psychological, physical, sexual, economic, social, symbolic violence.

Limitations and Future Research Needs A key conclusion of this study is that the results of research and the prevalence of the violence are still unavailable in Syria and other countries. The violence is still underdiagnosed. There is great need for more systematic

122

MUAWEAH AHMAD ALSALEH

data collection on violence in Syria and in the world. In particular, police recording of offenses and offenders needs to be conducted systematically across the country, and records nationally/worldwide compiled to monitor numbers of officially recorded crimes and offenders, and characteristics of offenders worldwide. There is also an urgent need to mount regular national victimization surveys in Syria to provide reliable estimates of violence frequency and time trends that can be compared with official statistics and with other countries. Given that reducing violence is one of the main challenges facing Syria today, it is imperative to continue collecting this basic information to monitor progress made. Longitudinal studies should be conducted to examine risk factors for violence from childhood onwards. Even so, the relationship between violent experiences and mental-health problems needs further study. According to Al-Badayneh (2012), diagnostic tools such as scales or questionnaires to measure violence against women are needed to be developed to reflect the components of Jordanian culture and social system. A qualitative research is needed too as a methodology especially to study victims of violence against women (Al-Badayneh, 2012). This issue applies to the Syrian society. In addition, social workers must be trained in each governorate and society. The violence against women has largely focused on violence committed by men because there is abundant evidence that this is the most endemic form of gender-based violence. Studies in the United Kingdom show that 92% of family violence perpetrators are men, and 91% of victims are women (Rew et al., 2013). Some women perpetrate violence against other women erodes many cross-cultural ideals of solidarity between women across generations and social classes (Rew et al., 2013). Western, Easterly and Oriental feminist debates on violence against women have largely focused on violence committed by men against women. But, another type of violence needs to be more researched into Syria and in the world. This type of violence is violence perpetrated by women against other women. The violence against women and their societal suffering is associated with social deprivation. This requires continuous researches and greater efforts so that women may fulfill their role in making future in Syrian society. In addition, future research is needed in other segments of Syrian society such as urban, rural, and Bedouin areas. The policy and research need to develop a deeper understanding of the continuum of violence and abuse involved, and the causes behind it, if these abuses are to be addressed.

123

Family Violence in Syria

ROLE OF THE FUNDING SOURCE There is no funding source or financial interest in this research.

CONFLICT OF INTEREST There is no conflict of interest in this research.

ACKNOWLEDGMENTS The author would like to thank Mr. Ahmad and Mrs. Sabrie´ Alsaleh Dr. Aude Ventura, Mrs. Maggie Hughes, and Mrs. Re´gine Fabien for their participation in giving the counseling.

REFERENCES Abdul Wahab, L. (1976). Family violence. Damascus : Damascus University Publication. Abu Ghazaleh, H. (2008). Violence Against Women: Joint vision for effecting change (No. 2, Periodic Newsletter). Jordan: National Council for Family Affair. Al-Badayneh, D. M. (2012). Violence against women in Jordan. Journal of Family Violence, 27, 369 379. Alkesh, M. A., & Hassan, N. K. (2010). Family violence against women in Syria. Syrian Commission for Family Affairs. Retrieved from http://wonews.net/ar/index.php?act= post&id=2992 Arabi, B. (2006). The women’s rights and family violence in Syria. Department of Sociology, Damascus University. Bacchus, L., Mezey, G., Bewley, S., & Haworth, A. (2004). Prevalence of domestic violence when midwives routinely enquire in pregnancy. BJOG, 111(5), 441 445. Bates, L. M., Schuler, S. R., Islam, F., & Islam, M. K. (2004). Socioeconomic factors and processes associated with domestic violence in rural Bangladesh. International Family Planning Perspectives, 30, 190 199. Brik, J., & Alkesh, M. A. (2011). Family violence against women in Syria. Communication Center and Strategic Research. Arab media website for the Geneva Institute for Human Rights. Bustamante- Gavino, M. I. (1998). The effects of the Nurse program on the self-esteem of Filipino abused women. Doctoral thesis, University of the Philippines, Manila. Campbell, J. C. (2002). Health consequences of intimate partner violence. Lancet, 359, 1331 1336.

124

MUAWEAH AHMAD ALSALEH

Cohen, M., Deamant, C., Barkan, S., Richardson, J., Young, M., Holman, S., … Melnick, S. (2000). Domestic violence and childhood sexual abuse in HIV-infected women and women at risk for HIV. American Journal of Public Health, 90(4), 560 565. Cohen, M., Forte, T., Du Mont, J., Hyman, I., & Romans, S. (2005). Intimate partner violence among Canadian women with activity limitations. Journal of Epidemiol Community Health, 59(10), 834 839. Country report. (2012a). Violence against women and migrant and minority women (pp. 94 101), France. Country report. (2012b). Violence against women and migrant and minority women (pp. 108 113), Germany. Craig, A. A., & Rowell Huesmann, L. (2003). Human aggression: A social-cognitive view. In M. A. Hogg & J. Cooper (Eds.), The Sage handbook of social psychology. Thousand Oaks, CA: Sage. Das, V. (2000). Violence and the work of time. In A. P. Cohen (Ed.), Signifying identities: Anthropological perspectives on boundaries and contested values. London: Routledge. Diop-Sidibe´, N., Campbell, J. C., & Becker, S. (2006). Domestic violence against women in Egypt wife beating and health outcomes. Social Science & Medicine, 62(5), 1260 1277. Douki, S., Nacef, F., Belhadj, A., Bouasker, A., & Ghachem, R. (2003). Violence against women in Arab and Islamic countries. Archives of Women’s Mental Health, 6(3), 165 171. El-Zanaty, F., Hussein, E., Shawky, G., Way, A., & Kishor, S. (1996). Egypt demographic and health survey 1995. Calverton, MD: Macro International. Fikree, F. F., & Bhatti, L. I. (1999). Domestic violence and health of Pakistani women. International Journal of Gynecology & Obstetrics, 65(2), 195 201. Fikree, F. F., Razzak, J. A., & Durocher, J. (2005). Attitudes of Pakistani men to domestic violence: A study from Karachi, Pakistan. Journal of Men’s Health and gender, 2(1), 40 58. Flitcraft, A. H., Hadley, S. M., Hendricks-Matthews, M. K., McLeer, S. V., & Warshaw, C. (1992). Diagnostic and treatment guidelines on domestic violence. American Medical Association. Gage, A. J. (2005). Women’s experience of intimate partner violence in Haiti. Social Science & Medicine, 61(2), 343 364. Hagemann-White C. (2010). Protecting women against violence: Analytical Study of the Results of the Third Round of Monitoring the Implementation of Recommendation Rec (2002) 5 on the Protection of Women against Violence in Council of Europe Member States. Directorate General of Human rights and Legal Affairs, Strasbourg. Haj-Yahia, M. M. (1999). Wife abuse and its psychological consequences as revealed by the first Palestinian national survey on violence against women. Journal of Family Psychology, 13, 642 662. Haj-Yahia, M. M. (2000a). Implications of wife abuse and battering for self-esteem, depression, and anxiety as revealed by the Second Palestinian national survey on violence against women. Journal of Family Issues, 21, 435 463. Haj-Yahia, M. M. (2000b). The incidence of wife abuse and battering and some sociodemographic correlates as revealed by two national surveys in Palestinian society. Journal of Family Violence, 15(4), 347 374.

Family Violence in Syria

125

Haj-Yahia, M. M., Wilson, R. M., & Naqvi, S. A. M. (2012). Justification, perception of severity and harm, and criminalization of wife abuse in the palestinian society. Journal of Interpersonal Violence, 27(10), 1932 1958. Harwell, T. S., Moore, K. R., & Spence, M. R. (2003). Physical violence, intimate partner violence, and emotional abuse among adult American Indian men and women in Montana. Preventive Medicine, 37(4), 297 303. Hastrup, K. (2003). Violence, suffering and human rights: Anthropological reflections. Anthropological Theory, 3, 309. Health, Well-Being and the Personal Safety of Women in Germany: A Representative Study of Violence against Women in Germany. (2003). Federal Ministry for Family Affairs, Senior Citizens, Women and Youth. Heise, L. L., Pitanguy, J., & Germaine, A. (1994). Violence against Women. The hidden health burden. Discussion Paper No. 225. The World Bank, Washington, DC, p. 46. Huth-Bocks, A. C., Levendosky, A. A., & Bogat, G. A. (2002). The effects of domestic violence during pregnancy on maternal and infant health. Violence & Victims, 17(2), 169 185. Issawi, A.-R. (2003). The family and its role in solving the problems of the child. Jordan: Osama House Publishing. Jaspard, M. (2011). Les violences contre les femmes (2nd ed., pp. 58 61). Paris: La De´couverte «Repe`res ». Karol, R. L., Micka, R. G., & Kuskowski, M. (1992). Physical, emotional, and sexual abuse among pain patients and health care providers: Implications for psychologists in multidisciplinary pain treatment centers. Professional Psychology: Research and Practice, 23(6), 480 485. Khan, M. (2000). Domestic violence against women and girls. UNICEF, Innocenti Research Centre. Innocenti Digest no. 6. Khawaja, M., & Barazi, R. (2005). Prevalence of wife beating in Jordanian refugee camps: Reports by men and women. Journal of Epidemiology and Community Health, 59, 840 841. Kristof, N. D., & WuDunn, S. (2009). The women’s crusade. The New York Times Magazine. Krug, E. G., Dahlberg, L. L., Mercy, J. A., Zwi, A. B., & Lozano, R. (2002). World report on violence and health (pp. 147 182). Geneva: World Health Organization. Leigh, K. (2014). Domestic violence on the rise among syrian refugees. Retrieved from http:// kristof.blogs.nytimes.com/2014/08/29/domestic-violence-on-the-rise-among-syrian-refu gees/?_r=1 Maalouf, W., & Campello, G. (2014). The influence of family skills programmes on violence indicators: Experience from a multi-site project of the United Nations Office on drugs and crime in low and middle income countries. Aggression and Violent Behavior, 19(6), 616 624. Maier, W., Gansicke, M., Gater, R., Rezaki, M., Tiemens, B., & Urzua, R. F. (1999). Gender differences in the prevalence of depression: A survey in primary care. Journal of Affect Disorder, 53(3), 241 252. Maryse, J., Brown, E., Condon, S., Firdion, J.-M., Fougeyrollas-Schwebel, D., Houel, A., … Schiltz, M.-A. (2000). Enqueˆte nationale sur les violences envers les femmes en France (Enveff).

126

MUAWEAH AHMAD ALSALEH

Maziak, W., & Asfar, T. (2003). Physical abuse in low-income women in Aleppo, Syria. Health Care for Women International, 24(4), 313 326. Michau, L., & Naker, D. (2003). Mobilising communities to prevent domestic violence: A resource guide for organisations in East and Southern Africa. Kampala, Uganda: Raising Voices. Murray, J., Cerqueira, D. R. C., & Kahn, T. (2013). Crime and violence in Brazil: Systematic review of time trends, prevalence rates and risk factors. Aggression and Violent Behavior, 18(5), 471 483. Nasar, S. (2013). Violence against Women, Bleeding Wound in the Syrian Conflict. EuroMediterranean Human Rights Network. Copenhagen K, Denmark. Nduna, S., & Goodyear, L. (1997). Pain too deep for tears: Assessing the prevalence of sexual and gender violence among burundian refugees in Tanzania. New York, NY: International Rescue Committee. O’Donnell, S., Fitzpatrick, M., & McKenna, P. (2000). Abuse in pregnancy The experience of women. Irish Medical Journal, 93(8), 229 230. OMS. (2002). Rapport mondial sur la violence et la sante´ (pp. 100 101), Gene`ve. Rani, M., & Bonu, S. (2009). Attitudes toward wife beating: A cross-country study in Asia. Journal of Interpersonal Violence, 24, 1371 1397. Rew, M., Gangoli, G., & Gill, A. K. (2013). Violence between female in-laws in India. Journal of International Women’s Studies, 14(1), 147 160. Robins, L. N., & Regier, D. A. (1991). Psychiatric disorders in America. New York, NY: Free Press. Roumani, R. (2006). Study reveals domestic abuse is widespread in Syria. The Christian Science Monitor CSMonitor.com. Retrieved from http://www.csmonitor.com/2006/ 0425/p04s01-wome.html Rutter, M. (1999). Social context: Meanings, measures and mechanisms. European Review, 7(1), 139 149. Series, L. (1999). Ending violence against women Population reports. Center for Health and Gender Equity (CHANGE). Shaikh, M. A. (2000). Domestic violence against women Perspective from Pakistan. Journal of Pakistan Medical Association, 50(9), 312 314. Siever, L. J. (2008). Neurobiology of aggression and violence. American Journal of Psychiatry, 165(4), 429 442. Stuart, H. (2003). Violence and mental illness: An overview. World Psychiatry, 2(2), 121 124. The Economist. (2011). Women in Turkey: Behind the veil. Istanbul. Torres, S., Campbell, J., Campbell, D. W., Ryan, J., King, C., Price, P., … Laude, M. (2000). Abuse during and before pregnancy: Prevalence and cultural correlates. Violence & Victims, 15(3), 303 321. Wasim, M., Asfar, T., Mzayek, F., Fouad, F. M., & Kilzieh, N. (2002). Socio-demographic correlates of psychiatric morbidity among low-income women in Aleppo, Syria. Social Science & Medicine, 54, 1419 1427. WHO. (1999). Violence Against Women, Women in Transition. Regional Monitoring Report, UNICEF, and a study by Domestic Violence Research Centre, Japan. WHO (World Health Organization). (2002). World report on violence and health. Geneva, Switzerland.

Family Violence in Syria

127

WHO (World Health Organization). (2007). Third milestones of a global campaign for violence prevention report 2007: Scaling up. Geneva, Switzerland. WHO Consultation. (1996). Violence against women. Geneva: World Health Organization, 5 7 February, 1996 (document FRH/WHD/96.27). Retrieved from http://whqlibdoc. who.int/hq/1996/FRH_WHD_96.27.pdf. Accessed on March 18, 2005. World Health Organization. (2000). Women’s mental health: An evidence based review. WHO.

This page intentionally left blank

VIOLENT SOCIALIZATION AND YOUTH VIOLENCE ACROSS DIFFERENT NATIONS: INTERNATIONAL VARIATIONS IN FAMILIAL AND CONTEXTUAL FACTORS Aime´e X. Delaney ABSTRACT Purpose The present study sought to determine whether or not there is an association between contextual effects of violent socialization on violent youth behavior across different nations. Design/Methodology/Approach The data in this study derive from the International Dating Violence Study, a dataset of over 17,000 college students collected in 32 different nations. Variables consist of various scales from the Personal Relationship Profile that focuses on experiences and behaviors occurring prior to age 15 years, as well as national indicators of violence. Multilevel modeling analysis was used to analyze the data.

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 129 151 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009006

129

130

AIME´E X. DELANEY

Findings The results indicate two important findings. First, violent socialization significantly varies across different national contexts and this contextual effect accounts for a significant proportion of variation in youth violence. Second, violent socialization, both within individual families and as a contextual effect within different nations, is significantly associated with increases in violent youth behavior across the nations. Research Limitations/Implications The International Dating Violence Study is a cross-sectional convenience sample of college students, which is not representative of specific nations nor college students. Further, the present study classifies nations as a “community” rather than explore microlevel communities within a nation. Future research should focus on examining variation of area specific norms for subsets of communities within nations with representative samples of a general population. Originality/Value The present study appears to be one of the first published studies offering empirical evidence for international research on the theoretical argument of the contextual effects of violence within a nation, and begins to increase knowledge among criminologists of such contextual factors being associated with youth violent behavior across different nations. Keywords: Youth violence; violent socialization; delinquency; family violence; community violence; contextual effects

Ineffective socialization is one of the best predictors of behavioral issues among youth (e.g., see Hao & Matsueda, 2006; Loeber & Dishion, 1983; Sousa et al., 2011), such that ineffective socialization can contribute to youth aggression (Cooper, Masi, & Vick, 2009). The prospect for violent behavior may be especially salient for youth exposed to multiple risk factors (Loeber & Stouthamer-Loeber, 1986; Rutter, Quinton, & Hill, 1990), including exposure to different types of violence. This exposure can occur concurrently within the family (i.e., at the individual level) and within a context outside of the family, such as norms within a nation, posing as cumulative risk for violent behavior among youth. For example, disciplinary techniques could be unique behavior within each individual family; at the same time, socialization may also reflect a context in which norms within a

Violent Socialization and Youth Violence across Different Nations

131

nation shape the family’s socialization of youth. One context may be patterns in violence within nations (Eriksson, 2010) and the disciplinary techniques used to control youth behavior (Forjuoh & Zwi, 1998). Such norms for socialization can be classified as a contextual effect, that is, socially approved methods of socialization consistent within the same nation. Violent socialization occurring within multiple contexts, such as the family and norms within a nation, whether intentionally or not, contribute to youth’s development of proviolence identification (cf. Mead, 1934) and the inculcation of values and expected behaviors (Kendall, 2010) consistent with proviolence. Violent socialization may include violent discipline, exposure to violence, and proviolent messages. Socialization embedded with violence may place youth at risk for engaging in violent behavior and positively contribute to youth’s violent behavior. Yet, there is little published research that has examined the effect of violent socialization being positively associated with youth violent behavior among international studies. While laws vary nation to nation, one common definition of violent behavior across different nations is assault or threat of assault Shaw, van Dijk, and Rhomberg (2003) Youth violent behavior then is defined as threatening or assaulting other people. Despite a large body of research on family violence and violence within the community, there appears to be no published studies exploring international variation in violent socialization. The association between such variation and violent youth behavior across different nations seems to be an understudied area of criminology. This includes empirical evidence that either supports or refutes arguments of contextual effects of violent socialization on violent youth behavior. What, then, might be the association between violent socialization and youth behavior across different nations? The present study addresses this limitation by exploring a context of violence, violence that occurs both within the family (i.e., individual level) and within different nations (i.e., group level), which may be associated with youth violent behavior. When there are differences in violent socialization across nations, might such variation predict differences in youth violent behavior within each nation?

A TYPOLOGY OF VIOLENT SOCIALIZATION Much research has demonstrated that effective familial socialization prevents youth’s criminal behavior (e.g., Beaver, Wright, & Delisi, 2007; De

132

AIME´E X. DELANEY

Li, 2004; Gibbs, Giever, & Martin, 1998; Johnson, Giordano, Manning, & Longmore, 2011; Piquero & Bouffard, 2007). Although methods to control youth criminal behavior have been conceptualized as interconnected dimensions (Vold, Bernard, and Snipes, 1998), and criminologists have argued that explanations for youth’s criminal behavior require the study of interdependent relationships among multiple social factors (Baker, 2010), less is known about the degree to which violence occurring within the socialization of youth may contribute to youth perpetrating violence. Yet, the current literature seems to indicate the importance of studying multiple facets of violence within the socialization of youth without fully defining violent socialization. Socialization that includes a combination of different types of violence originating from multiple sources could be classified as violent socialization. Best defined as a typology, violent socialization involves a combination of methods of socialization embedded in violence: physical discipline, exposure to violence, and proviolent messages. Excessive physical discipline, which is frequent and severe physical punishment occurring during both childhood and adolescence, is one mode of violent socialization. Exposure to violence involves youth witnessing violence by others, which may suggest to youth a norm, both within the family and occurring within a national context. Parents and other community members may advise youth to respond to violence with their own violence, which manifests, over time, as proviolent messages. Violent socialization may be interwoven to a degree that different types of violent socialization cannot be separated into singular dimensions. Instead, violent socialization needs to be examined as a typology initially to explore the association between violent socialization and youth violent behavior across different nations.

Violent Socialization within the Family The current literature suggests a general relationship between abuse and youth criminal behavior (Goddard, Goff, Melancon, & Huebner, 2000). For example, physical abuse during childhood tends to predict youth criminal behavior later during adolescence (e.g., Rebellon & Van Gundy, 2005; Robertson, Baird-Thomas, & Stein, 2008; Teague, Mazerolle, Legosz & Sanderson, 2008) and adolescents engaging in the most serious types of crimes tend to report a history of experiencing physical abuse (e.g., Goddard et al., 2000). And research specifically supports that physical discipline of youth positively contributes to youth violence. Ga`mez-Guadix, Straus, and

Violent Socialization and Youth Violence across Different Nations

133

Hershberger (2012) found that the degree to which youth experience violence, including experiences with physical discipline by the family, is related to youth engaging in violent behaviors toward a dating partner later in life. Many family violence researchers and professionals have argued that physical discipline and physical abuse are conceptually equivalent, as both represent violence against children (e.g., Bauman & Friedman, 1998; Crosson-Tower, 2008; Straus & Donnelly, 1994; Taylor, Hamvas, & Paris, 2011; United Nations, 2006; Zolotor, Theodore, Chang, Berkoff, & Runyan, 2008). Experiences with physical discipline/ abuse may unintentionally instill youth with orientations toward violence as a means to resolve conflict (cf. Straus & Donnelly, 1994), hence socializing youth toward the use of violence and positively contribute to youth violent behavior. Yet, family violence encompasses more than just physical discipline/abuse. Examining the effects of physical discipline alone does not fully capture violent socialization nor fully explain youth violence. Violence within the family often manifests co-currently and in multiple different ways (Crosson-Tower, 2008). Current research has shown that youth who are exposed to violence within their family later act out violently toward others (e.g., Forbes & Adams-Curtis, 2001; Margolin & Gordis, 2000). Further, research indicates a direct link between exposure to family violence and youth criminal behavior (Hamby, Finkelhor, Turner, & Ormrod, 2011; Herrenkohl, Sousa, Tajima, Herrenkohl, & Moylan 2008; Smith & Thornberry, 1995). Many of these studies examined exposure to violence within the family (i.e., witnessing family violence/abuse) without examining other manifestations of violence. In a recent study, Sousa et al. (2011) looked at the combined effects of physical abuse and exposure to family violence: abuse alone does not fully predict youth violence, and exposure to family violence significantly increase the likelihood of violent behavior by youth but not necessarily youth violence. Yet, the combined effects of abuse and exposure to family violence significantly increases the likelihood of criminal behavior and violent behavior by youth (Sousa et al., 2011). Violence within the family may be socializing youth toward proviolent orientations for future behavior, but violent socialization manifests within the family as more than physical discipline/abuse and exposure to violence. The literature on violent socialization within the family seems to have focused predominantly on examining the associations between singular dimensions of socialization, such as physical abuse or witnessing family violence and youth behavior, rather than explore a typology of violent

134

AIME´E X. DELANEY

socialization. As part of socialization, parents may advise youth respond to violence with their own violence. For example, a youth may be told to hit another youth if the first youth hits or insults them. In general, such advice tends to be a deviant belief within contemporary mainstream society (Ohene, Ireland, McNeely, & Borowsky, 2006). When advised to use violence to resolve conflict, youth are receiving proviolent messages and these proviolent messages may instill a sense that violence is an acceptable means in which to resolve conflict (Straus & Donnelly, 1994). Therefore, youth who receive proviolent messages, even in the form of “advice,” may be more likely to engage in violence. Singular family level variables alone do not capture the full impact of socialization on youth criminal behavior (Morrongiello, Corbett, & Bellissimo, 2008). Socialization by the family may coincide within a context of socialization by other agents, including national norms. And violent socialization could vary across different nations, as familial socialization, in general, varies across different nations (e.g., Keller, Otto, Lamm, Yovsi & Kartner, 2008; Urquiza & Goodlin-Jones, 1994). Therefore, associations between violent socialization and youth violence need to be explored through a multilevel perspective.

National Norms of Violent Socialization Contextual factors can influence the socialization of youth above and beyond individual families. While the current international literature is limited in this area, there are some ancillary examples of research supporting socialization occurring within the context of a nation and the association between national contextual factors and youth behavior. Specifically, current research has already identified that the unique social structure of certain nations influences socialization (e.g., see Greenfield, Flores, Davis & Salimkhan, 2008; Keller et al., 2008; Runyan et al., 2010). For example, Shaw (2010) argues that the social change currently occurring in some Asian and Pacific nations impacts the ability of that nation to enforce control over youth behavior, including youth criminal behavior. It has been argued elsewhere that structural influences, such as wide-spread, positive national views on the use of physical discipline, contribute to the socialization of youth within that nation (Forjuoh & Zwi, 1998), including violent social influences, and contribute to different types of criminal behavior (e.g., DeGue, DiLillo, & Scalora, 2010). As a whole, these theoretical arguments about socialization being associated with youth behavior seem to

Violent Socialization and Youth Violence across Different Nations

135

suggest patterns of macrolevel socialization that are reflective of contextual factors within a nation, such that common patterns of socialization may be within nations. Common patterns can become norms, over time, among those living in the same nation; members of the same nation may embrace similar methods of socialization, whether knowingly or unintentionally. For example, Knowles (1996) argues that physical discipline is one method of socialization in which there is no discerned consensus across nations, but rather is part of the shared values within the same nation. When members of a nation collectively engage in the physical discipline of their youth, allow youth to be exposed to violence, and send youth proviolent messages, a context of violent socialization is created within the nation. Thus, violent socialization is not necessarily isolated to the socialization of an individual family but instead may be a function of a contextual effect within a nation. When this context becomes a norm, violent socialization becomes a norm within a nation. Socialization occurring within a context of violence may be associated with youth violent behavior. The concept of violent socialization occurring within a context across different nations seems to remain mostly theoretical. There is some limited international research showing a contextual effect in the use of corporal discipline across different nations. In one example of a published study, Runyan and colleagues (2008) attributed their findings of international differences in the use of harsh physical discipline on youth directly to variation in socioeconomic status of the nation, with families living in low- and middle-income nations engaging in more harsh physical discipline than families living in high-income nations. Other studies have found that economic disadvantage within a nation may positively influence criminal behavior (Farrington, 2003; Lipsey & Derzon, 1998). However, there may be other indicators of violence within a nation that may also contribute to a context of violence, such as the legally sanctioned use of the death penalty and violent crime rates. Yet published research examining the varying effects of violent socialization, as a whole, while controlling for other macrolevel supports of violence, across different nations remains sparse. The present study explores the association between violent socialization and youth violent behavior, hypothesizing that, while violent socialization will vary nation to nation, contextual violent socialization will be associated with more violent youth behavior across different nations compared to family violent socialization (alone) such that higher levels of violent socialization will significantly increase youth’s violent behavior across different nations.

AIME´E X. DELANEY

136

METHODS Data from the International Dating Violence Study was used to test the hypothesis. Conducted between 2001 and 2006, the International Dating Violence Study (IDVS) dataset is a cross-sectional, international convenience sample of college and university students in 32 different nations. Table 1 lists each of the nations included in the dataset and the number of Table 1. Country Australia Belgium Brazil Canada China England Germany Greece Guatemala Hungary India Iran Israel Japan Lithuania Malta Mexico Netherlands New Zealand Portugal Romania Russia Scotland Singapore South Africa South Korea Sweden Switzerland Taiwan Tanzania United States Venezuela

Number of Respondents from Each Nation in the International Dating Violence Study. n 210 574 275 1199 2502 218 471 286 249 176 190 99 153 207 448 112 235 410 137 424 271 450 213 231 124 243 433 161 258 208 4236 249

137

Violent Socialization and Youth Violence across Different Nations

respondents within each of these nations. IRB approval was granted at each site. Demographic information for each of these nations has been published elsewhere (see Straus, 2009).1 Although there are several limitations to using this dataset (discussed in conclusion), the IDVS appears to be the only accessible international dataset including measures of both violent socialization and youth violence.

Sample The original dataset of 17,404 cases was filtered to remove any cases due to non-response (i.e., missing data), as deleting cases tend to produce the least biased parameter estimates and standard errors compared to imputation (Allison, 2002). To control for the possibility of measurement error in responses to violent behavior during adolescence, due to potential issues with memory recall, respondents older than age 25 years were filtered from the dataset. The final sample size for the present study was 15,652 respondents. Table 2 includes demographic data for respondents.

Table 2. Level-1 (n = 15,652) Family violence Family violence (c) Sex Age Age (c) Peer delinquency Ses (c) Social desirability Social desirability (c) Level-2 (n = 32) Contextual violence Contextual violence (c) GNP (c) vcrate vcrate (c) Corporal punishment law Death penalty law Youth violence scale, logged

Descriptive Statistics.

Minimum

Maximum

Mean

Standard Deviation

5.00 −3.38 0 18.00 −3.11 0 −3.24 13.00 −15.36

20.00 11.62 1.00 27.00 5.89 1.00 3.56 44.00 15.64

8.38 0 0.30 21.11 0 0.54 0 28.36 0

2.68 2.68 0.46 2.50 0.98 0.50 0.98 3.93 3.93

3.00 −3.23 −0.56 0 −5.43 0 0 4.00

12.00 5.77 0.12 49.56 44.13 1 1 16.00

6.23 0 0 5.43 0 0.63 0.41 5.97

1.90 1.90 0.14 7.922 7.922 0.49 2.03

AIME´E X. DELANEY

138

Variables The IDVS questionnaire was translated from a member of the Consortium at each institution, then back-translated to ensure conceptual equivalence (Ga`mez-Guadix et al., 2012) across each nation. Independent and dependent variables derive from the Personal Relationship Profile scales (PRP),2 included within the IDVS. Numerous previous studies have tested these variables (see http://pubpages.unh.edu/~mas2/ID-Sample-4-04.htm for a sample published articles), including examining the validity and reliability of the variables (e.g., Straus & Mouradian, 1999; Straus, 2004; Straus, Hamby, Boney-McCoy, & Sugarman, 1996). Control variables, both included within and independent of the IDVS, were included to rule out spurious relationships between the violent socialization and violent behavior. Table 2 lists descriptive statistics for variables. Youth Violence Youth violence, the dependent variable, was measured using two items from the early onset subscale of the criminal history scale (Straus, Hamby, Boney-McCoy, & Sugarman, 2010) measuring self-reported violent crime (i.e., assault) occurring before age 15 years. Items include “Before age 15, I physically attacked someone with the idea of seriously hurting them” and “Before age 15, I hit or threatened to hit my parents.” Scores for these measures are based on the respondent’s level of agreement that they had engaged in violent crime during adolescence, ranging from strongly disagree (1) to strongly agree (4). Higher scores represent more violent behavior. Due to positive skew, this scale was transformed by the natural log which produced a normal distribution. Individual-Level Variables Violent Family Socialization. The individual-level main predictor (violentfamily) is from the family subscale for the violent socialization scale (Straus et al., 2010). This subscale has five items, measuring the degree to which the respondent agreed that they experienced direct or vicarious victimization within the family, as well as being counseled by family to engage in violence to resolve conflict. Example items include being hit frequently by parents, witnessing violence by family members, and being advised by a parent to use violence to resolve conflict. Higher scores represent more experiences with violent socialization within individual families. Other studies have used similar measures, including excessive corporal discipline

Violent Socialization and Youth Violence across Different Nations

139

(Teague et al., 2008) and parental physical abuse (Rebellon & Van Gundy, 2005). Gender. The sex of the respondent was used as a control variable, coded to males (1) and females (0). Age. Age of each respondent was another control variable. In the present study, the ages of the respondents ranged from 18 years old to 25 years old. Peer involvement in delinquency. Some criminologists have argued that peers provide more support and group cohesion than the family (cf. differential association theory; Sutherland & Cressey, 1978). To control for this possible effect of peer group influence on delinquency, this variable (peers) measured whether or not the respondent associates with peers who engage in delinquency. Two variables, “I have friends who committed crimes” and “I spend time with criminal friends,” were dichotomized from four response categories (i.e., strongly disagree, disagree, agree, strongly agree) to two response categories (i.e., disagree or agree). This new variable represents the respondent’s agreement that their friends had criminal behavior or they associated with friends that engaged in criminal behavior. Socioeconomic status. Socioeconomic status (SES), another control variable, is an original variable from the IDVS dataset, consisting of a scale combining both the father’s and mother’s completed education with the family’s yearly income. Scores range from low SES to high SES. Social desirability. It is possible that some of the respondents may have answered questions in a manner that might make them look favorable rather than answering truthfully. Research has shown that questionnaires inquiring about both personal (Reynolds, 1982) and sensitive information (Lee, 1993) are subject to social desirability bias (Crowne & Marlowe, 1960). Thus, another control variable (socdesire) included the limited disclosure scale from the PRP to control for socially desirable responses. This scale was based on an adaptation of Reynolds’ short form of the Crowne Marlow scale (Straus et al., 2010). The limited disclosure scale included 13 items that measure for the possibility of respondents providing false answers about sensitive information for a more favorable personal image, such as underreporting violent behavior. Higher scores represent more socially desirable responses. Group-Level Variables Contextual Violent Socialization. To create the level-2 variable (conviolence), the three items in the non-family subscale of the violent socialization scale from the PRP were used. The 32 cases were created by averaging the

140

AIME´E X. DELANEY

responses for all respondents within each nation to create a national-level mean for each separate nation, then centered on the grand-mean of all cases within that same nation for multilevel modeling (MLM) analysis. This variable represents an approximation of the degree to which each nation has its own unique contextual factors in the violent socialization of youth (i.e., a contextual effect), with higher scores indicating the presence of more violent socialization. Variance Inflation Factor test ruled out collinearity between violent family socialization and contextual violent socialization (results provided upon request). Gross domestic product. A level-2 variable to control for economic inequality across nations, gross domestic product (GNP), is an original variable in the IDVS dataset measuring the percent of national monies spent on the military compared to national monies spent on education and human services (Straus & Ramirez, 2000). Violent crime rate. Legal definitions of crime vary nation to nation. Homicide data tends to be the most valid crime measure across different nations (LaFree, 2005) and, according to INTERPOL, defined consistently across nations as the intended illegal killing of an individual by another individual (United Nations, 2012). This control variable was created using data from the United Nations Office on Drugs and Crime (http://www. unodc.org/unodc/en/data-and-analysis/homicide.html). Violent crime rate (vcrate) is an average of the rate of homicides, per 100,000 in the population, for each nation in the present study for 10 years (1996 2006). Legal status of physical discipline. This variable is the first of two variables used to control national-level support of the use of violence against others, through national laws. The Global Initiative to End All Corporal Punishment of Children (2012) provides information regarding the legal status of corporal discipline across 172 different nations. This variable (discipline) measures whether or not each nation has laws prohibiting the use of physical discipline on children, coded as the nations with any law (within schools or by families) prohibiting physical discipline (1), and nations without such laws (0). Legal status of the death penalty. This is the second variable used to control for national-level support of the use of violence against others. The Death Penalty Information Center (2012) lists whether or not the death penalty is a legal form of criminal punishment across 198 different nations. This variable (deathlaw) was coded as nations that have legally abolished the death penalty (1), and nations still having laws allowing for the use of the death penalty (0). Data was coded missing for one nation in the current study (South Korea).

Violent Socialization and Youth Violence across Different Nations

141

Analysis MLM regression was used to test the contextual effect of violent socialization on youth violence. MLM regression techniques were designed to separate individual- and group-level effects on the dependent variable (Bryk & Raudenbush, 1992; Goldstein, 1988). Characteristics that occur at a higher level (i.e., group-level attributes) can influence individual-level characteristics (Luke, 2004). For example, respondents within the same nation may be exposed to similar standards of socialization (e.g., see Keller et al., 2008), a context in which these respondents are more similar to each other than to respondents from different nations, which could produce more homogeneous experiences within any given nation. In other words, because socialization varies across different nations, both in beliefs and practices (Urquiza & Goodlin-Jones, 1994), youth within the same nation may have comparable socialization experiences compared to youth from different nations, which creates a contextual effect in which the respondents from the same nation would cluster together. Using ordinary least squares (OLS) regression would be problematic. OLS regression assumes cases are statistically independent (Bryk & Raudenbush, 1992) and measurement error for each predictor variable in the model is independent and unrelated (Hamilton, 1992). Because the data in the IDVS was collected as groups of students within the same college/ university then aggregated by nation (Straus, 2009) and compiled together into a single dataset, the data are considered hierarchical data. Assumptions of independence are violated because of the presence of hierarchical data. This violation, then, could increase the probability of rejecting a null hypothesis (Hamilton, 1992) because error terms for similar cases may be correlated (Luke, 2004). As such, OLS regression is not appropriate for this particular study; individual cases within each of the nations are not fully independent. All continuous variables were centered on the grand-means of cases in order to provide meaningful interpretation of true values for the dependent variable (Fuwa, 2004; Singer & Willett, 2003); the intercept from the level-1 analysis becomes an outcome in the level-2 analysis of MLM regression analysis (Luke, 2004). The estimated intraclass correlation coefficient (ICC) was used as a parameter estimate to determine the proportion of total variance that accounts for the clustering (Bickel, 2007; Hayes, 2006) of violent socialization for each nation. The ICC indicates that contextual violent socialization explains about 5% of the total variation in youth violence.

AIME´E X. DELANEY

142

RESULTS Exploratory multilevel models were used to determine the degree to which youth violence varied among the nations in the IDVS (i.e., the “unconditional model”), the association between violent family socialization with youth violence (i.e., “random coefficient model”), and the association of contextual violent socialization on youth violence (i.e., “means as outcome model”). The results from each of these preliminary models are shown in Table 3.

Youth Violence The unconditional model, Model 0 in Table 3, contains no predictors (Wang, 1998), in order to determine the amount of variation in youth violence within and between nations. This model shows the average national score for youth violence across all nations, a mean score of about 5.8 (800 = 5.827, p > 0.001) on average. In other words, there tends to be, on average, low levels of youth violence across the nations. The unconditional model also shows that there is significant variation in youth violence scores both within each nation (σ2 = 4.922, p < .001) and between the different nations (Τ00 = 0.257, p < .001) with over 15 times more variation in youth violence scores within each nation than between the different nations.

Violent Socialization within the Family The random coefficient model (model 1, Table 3) adds respondents’ experiences of violent socialization within individual families to the unconditional model. This model examines the degree to which youth violence varies as a result of violent family socialization. The results are relatively similar: the mean youth violence score for respondents with average levels of violent socialization within their family is about 5.9 (800 = 5.875, p < .001), on average. However, higher levels of violent socialization within individual families tend to significantly increase youth violence scores across nations. For every point violent family socialization scores increase, mean youth violence scores increase by about 0.39 points (810 = 0.389, p < 0.001), calculated as 6.264 points on average.

143

Violent Socialization and Youth Violence across Different Nations

Table 3. Results from Fitting the Taxonomy of Multilevel Models to Selfreported Youth Violence Regressed on Level-1 and Level-2 Predictors (n of College Students = 16,652; n of Institutions = 32). Fixed effects Intercept

800

violentfamily

810

peer

820

gender

830

age

840

ses

850

socdesire

860

conviolence

801

GNP

802

vcrate

803

discipline

804

deathlaw

805

Model 0

Model 1

Model 2

Model 3

Model 4

5.827*** (0.093)

5.8750*** (0.077) 0.389*** (0.006)

2.718*** (0.090)

3.926*** (0.094) 0.278*** (0.007)

0.504*** (0.009)

0.314*** (0.009)

3.954*** (0.172) 0.254*** (0.007) 0.270*** (0.032) 0.970*** (0.034) 0.001 (0.007) −0.089*** (0.015) −0.097*** (0.004) 0.216*** (0.009) −0.838** (0.251) 0.003 (0.006) −0.182 (0.178) 0.416* (0.154)

3.915*** (0.044) 0.172** (0.047)

4.050*** (0.041) 0.152* (0.042)

3.658*** (0.041) 0.155** (0.042)

3.304*** (0.038) 0.132** (0.037)

20.5 33.1

17.7 40.9

25.7 39.7

32.9 48.6

Variance components (random effects) Level-1 σ2 4.922*** (0.056) Level-2 Τ00 0.257*** (0.070) Pseudo-R2 Within nations Between nations *** p < 0.001; ** p < 0.01; * p < 0.05

The pseudo-R2 indicates that violent socialization within the family explains about 20.5% of the variation in youth violence scores within the same nation and about 33% of the variation in youth violence scores across different nations.

144

AIME´E X. DELANEY

Violent Socialization as a Contextual Effect The means as outcome model (model 2, Table 3) examine the contextual effect of contextual violent socialization on youth violence. Nations with mean levels of violent socialization average youth violence score of about 2.7 (800 = 2.718, p > 0.001), on average. Higher levels of contextual violent socialization tend to significantly increase youth violence scores. For each point that contextual violent socialization increases, mean youth violence scores tend to increase by about 0.5 points (801 = 0.504, p < .001), calculated as 3.222 points on average. Once accounting for this contextual effect, there are significant differences in youth violence scores within nations (σ2 = 4.050, p < .001), explaining about 18% of the variation in youth violence within the same nation. Youth violence scores also significantly vary across different nations (Τ00 = 0.152, p < .001), and about 41% of the variation in youth violence between different nations.

Violent Socialization and Youth Violence across Different Nations Once violent family socialization, contextual violent socialization, and controls are included in the model (model 4, Table 3), mean youth violence scores across nations reduce by about 32% to about 4 points on average (800 = 3.954, p > 0.001). While this indicates low levels of youth violence across the nations, even when accounting for the effects of violent socialization, experiencing higher levels of contextual violent socialization still significantly increases youth violence scores. Youth violence scores increase by about 0.22 points (801 = 0.216, p > 0.001) for each increase in contextual violent socialization scores, with a calculated score of 4.170 points on average. The results are similar for violent family socialization, such that for each increase in violent family socialization scores, youth violence scores increase by about 0.25 points (810 = 0.254, p > 0.001) with a calculated score of 4.208 points on average. Both violent family socialization (model 1) and contextual violent socialization (model 2) are independently significantly associated with youth violence. However, still significantly associated with youth violence, the contextual violent socialization coefficient (801 = 0.314 in model 3) reduces by about 35% once accounting for violent family socialization, and remains reduced, by over 55%, net of all controls (810 = 0.216 in model 4). There is significant variation in youth violence across the different nations once accounting for the effects of violent socialization and controls

Violent Socialization and Youth Violence across Different Nations

145

(Τ00 = 0.132, p > 0.001). However, there is over 25% more variation in youth violence scores within each nation (σ2 = 3.304, p > 0.001) compared to between the nations. And violent socialization, both within individual families and within the community, accounts for about 33% of the variation in youth violence within nations and 47% of the variation across the different nations. It would appear that while violent socialization within individual families may provide slightly better explanation for youth violent behavior within each individual nation and violent socialization as a contextual effect has more explanatory strength for youth violent behavior between different nations, violent socialization occurring within multiple contexts (e.g., family, national norms, laws) best explains the variance in youth violent behavior overall. These findings support the research question and hypothesis: variation in violent socialization across nations is associated with differences in youth violence such that experiencing higher levels of violent socialization, at both the micro- and macrolevels, significantly increase violent behavior across different nations. However, the effects of both family (individual-level) and contextual (group-level) violent socialization on youth violent behavior are about the same rather than group-level violent socialization having a greater influence on youth violence behavior.

DISCUSSION The findings from the present study indicate that violent behavior tends to increase if both families and norms within a nation favor violence as part of the socialization of youth. Violence may be so inculcated within socialization that families and citizens within a nation may not be aware of engaging in or promoting violence. Such obscure socialization occurring across different families and nations may reinforce a context in which the violent socialization of youth is supported. Violent socialization then becomes a contextual effect. Given the argument that the discipline of criminology has failed to construct universal explanations for the causes of crime across different nations (Johnson & Barak-Glantz, 1983), this finding was expected. This study preliminarily demonstrates that, at least among this sample of 32 different nations, youth violence does vary as a function of a contextual effect of violent socialization. This could be interpreted as meaning that significant variation in the effects of violent socialization across different nations is associated with differences in violent behavior but in different ways.

AIME´E X. DELANEY

146

This is important to note because there is consistency across the 32 nations for the definitions of youth violence: assault or threat of assault (as outlined in the section “Methods”). With such consistency in conceptualization of the dependent variable, the focus is to explore a potential correlate of youth violence: violent socialization within each nation. With about 40%, the “explainable” variance in youth violence between the 32 different nations attributed to violent socialization as a national context effect alone3 and the effects of violent socialization significantly varying nation to nation, this project presents empirical data, at least preliminarily, demonstrating an association between violent socialization and youth violence that differ for each nation. While research has already shown that parents lacking the skills to effectively socialize youth increases likelihood these youth will have criminal behavior (Church, Wharton, & Taylor, 2009), this study begins to show that violent socialization occurring within context, in this instance as a norm within any given nation, also contributes to youth violence. The current research project supports the contention among criminologists that a combination of influences tends to best explain criminal behavior (Dillon, Pantin, Robbins, & Szapocznik, 2008), including youth violence. These findings provide a reference point for future comparative international research on norms within and across different nations being associated with youth’s violent behavior. Further, the present study begins to offer support that variation in violent socialization, both within the family and as a context within a nation, can differentiate youth’s violent behavior across different nations.

CONCLUSION Recently argued by Sellers, Ku¨bler, Walter-Rogg, and Walks (2012), the social sciences cannot advance understanding of subnational characteristics unless more internationally based research is conducted. More specifically, they discussed the need to examine social phenomenon through national context effects that identify similarities within the same nation, while also recognizing variation across different nations (Sellers et al., 2012). The present study offers preliminary evidence in this call for international research on contextual effects, by exploring a context of violent socialization within a nation and the association with youth violent behavior across different nations.

Violent Socialization and Youth Violence across Different Nations

147

The findings from the present study are particularly important because this study appears to be one of the first to demonstrate support favoring theoretical arguments for context effects being associated with crime. Failure to account for the multiple dimensions of violent socialization provides misleading and incomplete research in the study of violent behavior. While the present study opens pathways for increasing international understanding of familial and contextual factors associated with violent behavior, the results should be interpreted cautiously as the study is exploratory in nature and causal effects cannot be established. The results are based on data from a cross-sectional convenience sample of college students. It has been argued elsewhere that valid conclusions can be drawn from crossnational research despite the use of unrepresentative convenience samples (Straus, 2009). Further, although not necessarily representative of the nations as a whole, nor college students within these nations, valid empirical conclusions have been drawn using data from the IDVS (see Straus et al., 2010). The current criminological literature explaining correlates of youth violence would benefit from additional international research. The present study classifies the nation as a “community” recognizing that there may be several microlevel communities within a nation. Future research should focus on examining variation of area specific norms for subsets of communities within nations. Future international research should also attempt to capture samples of a more general population across different nations, and preferably representative of that nation, rather than use convenience samples of college students, in order to provide more generalizable results. The present study can, at least, provide initial understanding in the degree to which violent socialization is positively associated with violent behavior among youth. And more importantly, the present study helps to move forward family-based research that seeks to examine the potential consequences of socially approved violence, violence occurring within both families and communities, and the extent to which this violence may become part of the context in which youth are socialized.

NOTES 1. Specific details regarding each of these nations, including demographic data for each nation and the colleges/universities from each nation included in the IDVS, can also be located at http://pubpages.unh.edu/~mas2 2. Contact author for information on the psychometric properties of each scale.

AIME´E X. DELANEY

148

3. The pseudo-R2 for between nations in Model 1, the means as outcome model which tests the effects of the level-2 variable violent socialization as a national context effect only, is 59.3.

REFERENCES Allison, P. D. (2002). Missing data. Thousand Oaks, CA: Sage. Baker, J. O. (2010). The expression of low self-control as problematic drinking in adolescents: An integrated control perspective. Journal of Criminal Justice, 38, 237 244. doi:10.1016/j.jcrimjus.2010.02.011 Bauman, L. J., & Friedman, S. B. (1998). Corporal punishment. Pediatric Clinics of North America, 45, 403 414. doi:10.1016/S0031-3955(05)70015-8 Beaver, K. M., Wright, J. P., & Delisi, M. (2007). Self-control as an executive function: Reformulating Gottfredson and Hirschi’s parental socialization thesis. Criminal Justice and Behavior, 34, 1345 1361. doi:10.1177/0093854807302049 Bickel, R. (2007). Multilevel analysis for applied research. New York, NY: Guilford. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Application and data analysis methods. Newbury Park, CA: Sage. Church, W. T., Wharton, T., & Taylor, J. K. (2009). An examination of differential association and social control theory: Family systems and delinquency. Youth Violence and Juvenile Justice, 7, 3 15. doi:10.1177/1541204008324910 Cooper, J. L., Masi, R., & Vick, J. (2009). Social and emotional development in early childhood: What every policymaker should know. New York, NY: National Center for Children in Poverty, Columbia University. Retrieved from http://www.nccp.org/publications/pdf/ text_882.pdf Crosson-Tower, C. (2008). Understanding child abuse and neglect (7th ed.). Boston, MA: Allyn and Bacon. Crowne, D. P., & Marlowe, F. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349 354. De Li, S. (2004). The impacts of self-control and social bonds on juvenile delinquency in a national sample of midadolescents. Deviant Behavior, 25, 351 373. doi:10.1080/ 01639620490441236 DeGue, S., DiLillo, D., & Scalora, M. (2010). Are all perpetrators alike? Comparing risk factors for sexual coercion and aggression. Sexual Abuse: A Journal of Research and Treatment, 22, 402 426. doi:10.1177/1079063210372140 Dillon, F. R., Pantin, H., Robbins, M. S., & Szapocznik, J. (2008). Exploring the role of parental monitoring of peers on the relationship between family functioning and delinquency in the lives of African American and Hispanic adolescents. Crime & Delinquency, 54, 65 94. doi:10.1177/0011128707305744 Eriksson, M. (2010). Children who “witness” violence as crime victims and changing family law in Sweden. Journal of Child Custody, 17, 93 116. doi:10.1080/15379418.2010. 508258 Farrington, D. P. (2003). Developmental and life-course criminology: Key theoretical and empirical issues-the 2002 Sutherland award address. Criminology, 41, 221 225. doi:10.1111/j.1745-9125.2003.tb00987.x

Violent Socialization and Youth Violence across Different Nations

149

Forbes, G. B., & Adams-Curtis, L. E. (2001). Experiences with sexual coercion in college males and females: Role of family conflict, sexist attitudes, acceptance of rape myths, self-esteem, and the big-five personality factors. Journal of Interpersonal Violence, 16, 865 889. doi:10.1177/088626001016009002 Forjuoh, S. N., & Zwi, A. B. (1998). Violence among children and adolescents: International perspectives. Pediatric Clinics of North America, 45, 2 426. Fuwa, M. (2004). Macro-level gender inequality and the division of household labor in 22 countries. American Sociological Review, 69, 751 767. Ga`mez-Guadix, M., Straus, M. A., & Hershberger, S. (2012). Childhood and adolescent victimization and perpetration of sexual coercion by male and female university students. Deviant Behavior, 32, 712 742. doi:10.1080/01639625.2010.514213 Gibbs, J. J., Giever, D., & Martin, J. S. (1998). Parental management and self-control: An empirical test of Gottfredson and Hirschi’s general theory. Journal of Research in Crime and Delinquency, 35, 40 70. Goddard, H. W., Goff, B. G., Melancon, M. V., & Huebner, A. J. (2000). Profiles of delinquency: A comparison of delinquent behavioral groups. Journal of Social Behavior and Personality, 15, 19 50. Goldstein, H. (1988). Multilevel statistical models. New York, NY: Wiley. Greenfield, P. M., Flores, A., Davis, H., & Salimkhan, G. (2008). What happens when parents and nannies come from different cultures? Comparing the caregiving belief systems of nannies and their employers. Journal of Applied Developmental Psychology, 29, 326 336. doi:10.1016/j.appdev.2008.04.002 Hamby, S., Finkelhor, D., Turner, H., & Ormrod, R. (2011). Children’s exposure to intimate partner violence and other family violence. (OJJDP Publication No. 2005 JL FX 0048). Washington, DC: U.S. Government Printing Office. Hamilton, L. C. (1992). Regression with graphics. Pacific Grove, CA: Brooks/Cole. Hao, L., & Matsueda, R. L. (2006). Family dynamics through childhood: A sibling model of behavior problems. Social Science Research, 35(2), 500 524. doi:10.1016/j.ssresearch. 2004.10.003 Hayes, A. F. (2006). A primer on multilevel modeling. Human Communication Research, 32, 385 410. doi:10.1111/j.1468-2958.2006.00281.x Herrenkohl, T. I., Sousa, C., Tajima, E. A., Herrenkohl, R. C., & Moylan, C. A. (2008). Intersection of child abuse and children’s exposure to domestic violence. Trauma, Violence, & Abuse, 9, 84 89. doi:10.1177/1524838008314797 Johnson, E., & Barak-Glantz, I. (1983). Comparative criminology. Beverly Hills, CA: Sage. Johnson, W. L., Giordano, P. C., Manning, W. D., & Longmore, M. A. (2011). Parent-child relations and offending during young adulthood. Journal of Youth and Adolescence, 40, 786 799. doi:10.1007/s10964-010-9591-9 Keller, H., Otto, H., Lamm, B., Yovsi, R. D., & Kartner, J. (2008). The timing of verbal/vocal communications between mothers and their infants: A longitudinal cross-cultural comparison. Infant Behavior & Development, 31, 217 226. doi:10.1016/j.infbeh.2007.10.001 Kendall, D. (2010). Sociology in our times: Essentials (7th ed.). Belmont, CA: Wadsworth Cengage Learning. Knowles, C. (1996). Family boundaries: The invention of normality and dangerousness. Orchard Park, NY: Broadview Press. LaFree, G. (2005). Evidence for elite convergence in cross-national homicide victimization trends, 1956 to 2000. The Sociological Quarterly, 46, 191 211. doi:10.1111/j.1533-8525.2005.00009.x

150

AIME´E X. DELANEY

Lee, R. M. (1993). Doing research on sensitive topics. Thousand Oaks, CA: Sage. 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. P. Farrington (Eds.), Serious & violent juvenile offenders: Risk factors and successful interventions (pp. 86 105). Thousand Oaks, CA: Sage Publications, Inc. Loeber, R., & Dishion, T. (1983). Early predictors of male delinquency: A review. Psychological Bulletin, 94, 68 99. Loeber, R., & Stouthamer-Loeber, M. (1986). Family factors as correlates and predictors of juvenile conduct problems and delinquency. In M. Tonry & N. Morris (Eds.), Crime and justice: A review of research (pp. 29 149). Chicago, IL: University of Chicago Press. Luke, D. A. (2004). Multilevel modeling. Thousand Oaks, CA: Sage. Margolin, G., & Gordis, E. B. (2000). The effect of family and community violence on children. Annual Review of Psychology, 51, 445 479. Mead, G. H. (1934). Mind, self, and society. Chicago, IL: University of Chicago Press. Morrongiello, B. A., Corbett, M., & Bellissimo, A. (2008). Do as I say, not as I do: Family influences on children’s safety and risk behaviors. Health Psychology, 27, 498 503. doi:10.1037/0278-6133.27.4.498 Ohene, S. A., Ireland, M., McNeely, C., & Borowsky, I. W. (2006). Parental expectations, physical punishment, and violence among adolescents who score positive on a psychosocial screening test in primary care. Pediatrics,117, 441 447. doi:10.1542/peds.2005-0421 Piquero, A. R., & Bouffard, J. A. (2007). Something old, something new: A preliminary investigation of Hirschi’s redefined self-control. Justice Quarterly, 24, 1 27. doi:10.1080/ 07418820701200935 Rebellon, C. J., & Van Gundy, K. (2005). Can control theory explain the link between parental physical abuse and delinquency? A longitudinal analysis. Journal of Research in Crime and Delinquency, 42, 247 274. doi:10.1177/0022427804271926 Reynolds, W. M. (1982). Development of reliable and valid short forms of the MarloweCrowne Social Desirability Scale. Journal of Clinical Psychology, 38, 119 125. Robertson, A. A., Baird-Thomas, C., & Stein, J. A. (2008). Child victimization and parental monitoring as mediators of youth problem behaviors. Criminal Justice and Behavior, 35, 755 772. doi:10.1177/0093854808316096 Runyan, D. K., Shankar, V., Hassan, F., Hunter, W. M., Jain, D., Paula, C. S., … Bordin, I. A. (2010). International variations in harsh child discipline. Pediatrics, 126, E701 E711. doi:10.1542/peds.2008-2374 Rutter, M., Quinton, D., & Hill, J. (1990). Adult outcomes of institution-reared children. Males and females compared. In L. N. Robins & M. Rutter (Eds.), Straight and devious pathways from childhood to adulthood (pp. 135 157). Cambridge: Cambridge University Press. Sellers, J., Ku¨bler, D., Walter-Rogg, M., & Walks, A. (2012). The political ecology of the metropolis. Essex: ECPR Press. Shaw, M., van Dijk, J., & Rhomberg, W. (2003). Determining trends in global crime and justice: An overview of results from the united nations surveys of crime trends and operations of criminal justice systems. Forum on Crime and Society, 3(1&2), 35 63. Shaw, V. N. (2010). Crime and social control in Asia and the Pacific: An examination of larger social institutions and processes. Criminal Justice Studies, 23, 303 323. doi:10.1080/ 1478601X.2010.516526

Violent Socialization and Youth Violence across Different Nations

151

Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and event occurrence. New York, NY: Oxford University Press. Smith, C. A., & Thornberry, T. P. (1995). The relationship between childhood maltreatment and adolescent involvement in delinquency. Criminology, 33, 451 481. Sousa, C., Herrenkohl, T. I., Moylan, C. A., Tajima, E. A., Klika, J. B., Herrenkohl, R. C., & Russo, M. J. (2011). Longitudinal study on the effects of child abuse and children’s exposure to domestic violence, parent child attachments, and antisocial behavior in adolescence. Journal of Interpersonal Violence, 26, 111 136. doi:10.1177/0886260510362883 Straus, M. A. (2004). Cross-cultural reliability and validity of the revised conflict tactics scales: A study of university student dating couples in 17 nations. Cross-Cultural Research, 38, 407 432. Straus, M. A. (2009). The national context effect: An empirical test of the validity of crossnational research using unrepresentative samples. Cross-Cultural Research, 43, 183 205. doi:10.1177/1069397109335770 Straus, M. A., & Donnelly, D. A. (1994). Beating the devil out of them: Corporal punishment in American families. New York, NY: Lexington Books. Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. (1996). The revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283 316. Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. (2010). Manual for the Personal and Relationships Profile (PRP). Retrieved from http://pubpages.unh.edu/~mas2/ Straus, M. A., & Mouradian, V. E. (1999). Preliminary psychometric data for the Personal and Relationships Profile (PRP): A multi-scale tool for clinical screening and research on partner violence. Retrieved from http://pubpages.unh.edu/~mas2/ Straus, M. A., & Ramirez, I. L. (2000). Detailed research design for the international dating violence study. Retrieved from http://pubpages.unh.edu/~mas2/detailed-research-design.htm Sutherland, E. H., & Cressey, D. R. (1978). A sociological theory of delinquency. Criminology. Philadelphia, PA: J.B. Lippincott Co. Taylor, C. A., Hamvas, L., & Paris, R. (2011). Perceived instrumentality and normativeness of corporal punishment use among black mothers. Family Relations, 60, 60 72. doi:10.1111/j.1741-3729.2010.00633.x Teague, R., Mazerolle, P., Legosz, M., & Sanderson, J. (2008). Linking childhood exposure to physical abuse and adult offending: Examining mediating factors and gendered relationships. Justice Quarterly, 25, 313 348. doi:10.1080/07418820802024689 United Nations. (2006). World report on violence against children. Retrieved from http://www. unicef.org/violencestudy/reports.html United Nations. (2012). United nations office on drugs and crime homicide statistics. Retrieved from http://www.unodc.org/unodc/en/data-and-analysis/homicide.html Urquiza, A. J., & Goodlin-Jones, B. L. (1994). Child sexual abuse and adult revictimization with women of color. Violence and Victims, 9, 223 232. Vold, G. B., Bernard, T. J., & Snipes, J. B. (1998). Theoretical criminology (4th ed.). New York, NY: Oxford University Press. Wang, J. (1998). Opportunity to learn: The impacts and policy implications. Educational Evaluation and Policy Analysis, 20, 137 156. Zolotor, A. J., Theodore, A. D., Chang, J. J., Berkoff, M. C., & Runyan, D. (2008). Speak softly-and forget the stick Corporal punishment and child physical abuse. American Journal of Preventive Medicine, 35, 364 369. doi:10.1016/j.amepre.2008.06.031

This page intentionally left blank

“HE IS STILL MY SON”: AGING AND LIVING IN THE SHADOW OF AN ABUSIVE ADULT CHILD WITH MENTAL DISORDER Tova Band-Winterstein, Hila Avieli and Yael Smeloy ABSTRACT Purpose In face of global deinstitutionalization policy, some aging parents find themselves confronting violence and crime in the family due to abusive behavior from their adult child with mental disorder. The aim of this paper is to explore and understand the meaning given by aging parents to this deviant behavior and the different ways in which they cope with a lifetime in the shadow of violence. Design/Methodology/Approach Data collection was performed through in-depth semi-structured interviews with 16 parents, followed by content analysis. Findings Three themes that expressed the meaning attributed to life with ACMD in the shadow of violence: (1) constructing parental identity in a shared reality of violence, (2) social and family networks as a

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 153 176 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009008

153

154

TOVA BAND-WINTERSTEIN ET AL.

resource in coping with ACMD, and (3) keeping a daily life routine as an anchor in a vulnerable, abusive relationship Practical Implications Intervention with such families should focus on the life review process as a therapeutic tool. Interventions should also provide a “safety belt,” including health services, public social networks, and knowledge regarding their right for self-protection. Originality/Value Old age becomes an arena for redefined relationships combining increased vulnerability, needs of both sides, and its impact on the well-being of the ageing parents. This calls for better insights and deeper understanding in regard to intervention with such families. Keywords: Aging; elder abuse; qualitative research; violence and crime

The elder abuse phenomenon, with its varying prevalence and scope, has given rise to various definitions, referring to the numerous dimensions of the abuse and the abusive relationship (Dixon et al., 2010; Perel-Levin, 2008). The World Health Organization (2001) defines elder abuse as: “a single or repeated act or lack of appropriate action occurring within any relationship where there is an expectation of trust, which causes harm or distress to an older person.” Various forms of elder abuse have been observed, such as physical, psychological/emotional, sexual, and financial abuse (Comijs, Smit, Pot, Bouter, & Jonker, 1999; Lachs & Pillemer, 2004; Muehlbaur & Crane, 2006; O’Keeffe et al., 2007; Pillemer & Finkelhor, 1988). Intentional or unintentional neglect can also be included in the phenomenon (Fulmer et al., 2005; Reay & Browne, 2001). More recently, abuse and neglect have been combined and defined as “mistreatment,” which implies both an omission and a commission dimension (Muehlbaur & Crane, 2006). This definition has served as a guide for practitioners involved with elder abuse intervention units. In recent decades, social and medical changes have influenced the lives of people with mental disorders. These changes have included deinstitutionalization, which has contributed to the integration of the mentally ill into the general community, new drug treatments, and greater access to social resources (Cook, Cohler, Pickett, & Beeler, 1997). Consequently, people with mental disorders who previously lived in various forms of institutionalized

Living with an Abusive Adult Child with Mental Disorder

155

arrangements are now residing in the community, often with their parents (Saunders, 2003). Studies on elder abuse have focused on older parents’ daily experience of physical and emotional harm at the hands of their children (Anetzberger, Korbin, & Austin, 1994; Nahmiash, 2004; Pillemer & Finkelhor, 1988). The literature suggests a wide range of characteristics of abusive adult children and includes a unique group with mental disorder (Lachs & Pillemer, 2004). Although families are often envisioned as a “safe haven,” aging parents face challenges when living with an abusive adult child with mental disorder. Despite the growing awareness of elder abuse, research into the family’s perspective of the phenomenon is limited as most of these studies focus on the nature and extant of the violence (Schiamberg & Gans, 2000; Vaddadi, Gilleard, & Fryer, 2002; Vaddadi, Soosai, Gilleard, & Adlard, 1997) while the subjective experience of the abused parents has received little attention (Band-Winterstein, 2015). Moreover, the particular population of adult children with mental disorder (ACMD) that are diagnosed with serious mental disorders such as schizophrenia and drug or alcohol abuse are a significantly relevant for research as they are profiled with high-risk abusive behaviors (Schiamberg & Gans, 2000). The aim of this paper is to explore and understand the meaning given by aging parents to this deviant behavior and the different ways in which they cope with a lifetime in the shadow of violence.

ELDER ABUSE AS A CRIME WITHIN THE FAMILY: THEORETICAL FRAMEWORK The concept of elder abuse as a crime within the family attempts to explore it in the context of crime and victimology and thus understand the different roles of the “actors” involved (Lachs et al., 2006; Payne, 2005; Payne & Gainey, 2005). Studies have shown that certain characteristics are typical of adult children who abuse their parents, such as alcohol addiction (Homer & Gilleard, 1990) and mental disorder (Greenberg, McKibben, & Raymond, 1990; Lowenstein, Eisikovits, Band-Winterstein, & Enosh, 2009). These children usually reside with their parents (Lowenstein et al., 2009; Marmolejo, 2008) and are prone to life crises (Lowenstein et al., 2009). Several theories have been offered to explain why these adult children abuse their older parents, such as the dependence theory (Wolf, 1996), caregiver burnout (Brandel et al., 2007), transgenerational violence theory (Korbin, Anetzberger, & Austin, 1995), and ageism (Brandel et al., 2007;

156

TOVA BAND-WINTERSTEIN ET AL.

Walsh et al., 2007). Another explanation for elder abuse by an adult child with mental disorder (ACMD) is that inner psychological and mental processes cause abnormal and deviant behavior (Payne, 2005). Such processes can occur when the abuser has a mental disorder (Payne, 2005; Ramsey-Klawsnik, 2000). Schiamberg and Gans (2000) suggest an applied ecological bifocal intergenerational model as a basis for framing and understanding elder abuse in the family context. These authors view parents and ACMD as part of a dyed that evolves over several ecological levels, thus the child’s mental illness plays a significant role by their risk behaviors that affect older parents.

ELDER ABUSE BY ACMD Mental disorder is defined as a behavioral or psychological syndrome which is clinically significant and which is associated with distress, disability, or risk of suffering or losing freedom (American Psychiatric Association [APA], 1994). The ICD (WHO, 2001) describes the term disorder as a clinically manifesting set of behavior or symptoms with distress or disturbance of personal functions. Thus, most operational definitions of mental illnesses are syndromal, based on a convergence of signs and symptoms usually linked with some form of disability (Lauronen et al., 2004). The increase in life expectancy in general population, together with process of deinstitutionalization, new drug treatments, and greater access to social resources (Cook et al., 1997), contributes to growing numbers of adults with mental disorders living within the community (Niimura et al., 2011). It is estimated that between 35% and 75% of adults with mental illness reside with family members, especially parents (Kaufman, Scogin, MacNeil, Leeper, & Wimberly, 2010). As a consequence, both the parents and the ACMD share a long-term reality of abusive dynamics (BandWinterstein, Smeloy, & Avieli, 2014). Thus, the responsibility of living with and caring for ACMD is falling on an increasing number of aging parents (Cohler, Pickett, & Cook, 1991; Lefley, 1987). Some of this growing population of aging parents are living in high-risk situations as a result of the unpredictable and often aggressive behavior from their ACMD (BandWinterstein et al., 2014; Cooper, Selwood, & Livingston, 2008). Such behavior might include shouting, swearing, threats, and serious injuries (Vaddadi et al., 2002). Studies have revealed that 32 40% of family members had been assaulted by a relative with a mental disorder (Vaddadi et al., 2002, 1997). Other studies point on an increase of cases of elder

Living with an Abusive Adult Child with Mental Disorder

157

abuse by ACMD in the past decade from 30.1% in 1990 to 36.7% in 1996 and 55% in 2007 (Cooper et al., 2008). These data indicates on severity of the phenomenon and its potential effect on a growing number of people. Thus, the influence of mental health on family violence should be paid attention to both in research and clinical settings. Researchers have suggested various explanations for this violent behavior among persons with mental disorders, such as the person’s aggressive temperament, a lack of self-control, psychopathic personality, and feelings of anger and even rage (Monahan & Steadman, 1994). Other researchers emphasized the relationship dynamics as a major factor in the etiology of the violent behavior. These researchers used the theory of codependency to explain the relationships between abused parents and their abusive children (Lachs, Williams, O’Brien, Hurst, & Horwitz, 1997; Wolf, 1996). Codependency refers to a relationship in which a person is controlled or manipulated by another who is affected by a pathological condition (Davis, 2009). In this context, the mentally ill child is strongly dependent on the abused parents and vice versa (Pillemer, 1985; Sadock, Kaplan, & Sadock, 2007). The few studies that attempted to address this topic focused on the timing of the violence following hospitalization (Straznickas, McNiel, & Binder, 1993), family members who are involved in the violence (McNiel & Binder, 1994), the types of violent acts that were used (Swan & Lavitt, 1988), and the causes of the violent incidents (Monahan & Steadman, 1994). They did not focus on the parents’ subjective experience and the meaning they gave to this relationship through the life course.

AGING PARENTS EXPERIENCE IN CARING FOR AN ACMD The aging of parents of ACMD involves a dual element of stress, regarding the ongoing care for their child alongside the need to cope their own agingrelated changes in health and functioning, such as physical disability and greater vulnerability to illness (Lefley, 2003). Sharing a household with ACMD requires the parents to resume the role of active caretakers, regardless of their age and their physical and mental condition (McGarry & Arthur, 2001, p. 182). In studies of older parents of ACMD, it has been suggested that these parents experience greater levels of caregiver burden (Cook, Lefley, Pickett, & Cohler, 1994; Greenberg, Greenley, & Benedict, 1994; Provencher & Mueser, 1997), poorer health (Ghosh & Greenberg, 2009; Seltzer, Greenberg, Floyd, Pettee, & Hong, 2001), higher rates of

158

TOVA BAND-WINTERSTEIN ET AL.

chronic health conditions (Magan˜a, Greenberg, & Seltzer, 2004), neglect of health, higher risk of mortality (Addington, McCleery, & Addington, 2005), and elevated levels of general medical symptoms (Seltzer et al., 2001) than other parents. The relationship dynamics between older parents and their ACMD have been poorly documented. A possible reason for this is that up to the past decade, examining the association of violence and mental disorders was generally restricted to inpatient settings (Solomon, Cavanaugh, & Gelles, 2005). The body of knowledge about elder abuse and families with ACMD has evolved as two separate topics. The few studies that have attempted to address them as a unified theme did not focus on the subjective aging experience of the parents living with ACMD and its impact on their aging process. The knowledge gap on this topic emphasizes the importance of the current study. The main research questions of the present study were as follows: How do parents experience and perceive their parenthood in the context of an abusive relationship? How do they find resources to deal with this situation? How do they frame and position themselves within the abusive child − parent relationship?

METHOD As the phenomenon under study was so complex, one of the significant challenges in the present study was selecting the research method. We chose a phenomenological qualitative approach because it provides a better understanding of the participants’ experience through their feelings, thoughts, insights, and understandings. Through this perspective, we were able to focus on the “lived experiences” (Denzin, 1983) of parents coping with abuse by their ACMD and on the meanings they ascribed to these experiences (Moustakas, 1994). Another challenge in this study was to reach a sample that can best address the phenomenon in depth, in-line with the qualitative tradition that does not depend on random sampling techniques (Patton, 2002).

Participants and Sample The participants were purposefully selected (Patton, 2002) by criterion sampling, to obtain the widest possible variation of respondents among parents

Living with an Abusive Adult Child with Mental Disorder

159

coping with abuse by their ACMD. Sixteen parents (11 mothers and 5 fathers), aged 58 − 94, participated in the study. Participants were recognized by the welfare services as individuals exposed to abuse and mistreatment by their ACMD including seven adult children with schizophrenia, one with chronic depression, and eight adult children with co-morbid syndromes of schizophrenia and drug/alcohol abuse, epilepsy, and personality disorder (Table 1). The sampling criteria were that the participants be Hebrew-speakers, with no cognitive deterioration, good verbal capabilities, and residing with ACMD. This criteria evaluation was performed during ongoing interactions with welfare professionals at the local welfare agencies. Most of the participants who were initially approached agreed to take part in the study. One woman refused to participate, but gave no explanation why, and one man was not able to attend the interview due to ill health on the appointed date. All participants had medium- to low-income rates with basic education level, and deteriorating health status. We used theoretical saturation to determine the final sample size (16 participants). In other words, saturation was reached when no new relevant information was added (Bowen, 2008; Green & Thorogood, 2006; Morse, 2000).

Data Collection Data collection was performed via in-depth semi-structured interviews using an interview guide. It included several content categories such as the nuclear family: parent − child relations (example: what can you tell me about your family members?); abusive relationships over the years (example: can you describe your son’s deviant behavior over the years?); abusive relationships in parents’ later years (example: how do you perceive your parenthood to this child at this point of your life?), and ways of coping (example: how do you manage to address your own needs in your current life context?).

Procedure The data presented in this paper were drawn from a wide-scale study conducted by a team of researchers from the University of Haifa. The research team comprised two head researchers (with PhDs in gerontology and criminology) and a graduate gerontology student who conducted the interviews.

Table 1. Name

Age

Family Status

Characteristics of the Participants and Their Adult Children.

Country of Origin

Number of Children

Years of Residing with the Child

Type of Abuse

Years of Being Abused by the Child

Psychological Psychological Psychological, financial, physical Psychological, physical Psychological, physical Psychological

25 30 13

Joushoa Abraham Gadi

85 79 75

Widower Married Widower

Romania Romania Libya

3 3 7

55 50 13

Haim Moses Mimi

89 90 75

Widower Widower Married

Poland Soviet union Morocco

3 1 3

55 56 40

Gila

75

Widow

Iraq

9

33

Lea

79

Divorced

Kurdistan

3

48

Bella

80

Married

Soviet union

2

53

Nili

75

Widow

Soviet union

3

50

Miri

68

Widow

Israel

2

25

Lili

58

Divorced

Israel

4

25

Orna

65

Widow

Israel

3

Funy Suzanne

78 72

Widow Widow

Tunisia Morocco

Firha

70

Widow

Morocco

Note: Names of the participants were changed.

Psychological, financial, physical Psychological, financial, physical Psychological, physical

Child’s Mental Illness

9

Schizophrenia Chronic depression Schizophrenia and drug abuse Schizophrenia Schizophrenia Schizophrenia and drug abuse Schizophrenia

48

Schizophrenia

53 50

Schizophrenia and personality disorder Schizophrenia

7

Schizophrenia

25

Schizophrenia and alcohol abuse Schizophrenia and drug abuse Schizophrenia and Epilepsy Schizophrenia and drug abuse Schizophrenia and epilepsy

27 28 25

47

Psychological, financial, physical Psychological, financial, physical Psychological, financial, physical Psychological, financial

25

8 9

39 40

Psychological Psychological, neglect

39 2

8

30

Psychological, neglect

30

Living with an Abusive Adult Child with Mental Disorder

161

Participants were recruited through welfare agencies. The researcher who conducted the interviews is also a social worker engaged closely with the elder population, and is well trained in forming a therapeutic alliance with abused older parents. Nevertheless, the recruitment of participants lasted a year because of the difficulty in locating candidates who met the sample criteria. Based on existing knowledge about performing sensitive research (Lee & Renzetti, 1993), we followed the following procedure: The researcher explained the aim of the study to social workers in elder abuse prevention units and that they were seeking older parents who are abused by their ACMD. The social workers provided a list of potential participants, who had applied to the elder abuse units at this stage of life following incidences of hurt and distress at the hands of their children. Subsequently, the researcher contacted them directly by telephone, introducing herself and seeking their consent to be interviewed. This introductory telephone conversation was necessary to inform the participants of the topic of inquiry and to establish an initial rapport and trust. Face-to-face interviews took place at the participants’ location of choice in most cases, in the participant’s home, or in public locales, such as parks and cafe´s. The research aim and topic were presented with appropriate sensitivity, taking into consideration the stereotypes that often accompany mental illness. Therefore, the aim of the study was presented as an exploration of the conflictual relationship between older parents residing with adult children who have dealt with mental problems. Thus, in-line with the interview guide, the issue of elder abuse was directly exposed early on in the interview. All participants signed an informed consent form. Each interview lasted approximately an hour and a half, was tape-recorded, and later transcribed verbatim. When the parents’ sense of intimidation and insecurity during an interview at home was apparent because of an unexpected interruption by the ACMD, we stopped the interview, reported the incident to the social workers treating the family, and continued the interview when the incident had passed.

Data Analysis and Trustworthiness The qualitative researcher is a central figure, actively involved in all stages of the research, who constructs interpretations, rather than reporting “facts” or “truth” (Ben-Ari & Enosh, 2011; Hertz, 1997). Consequently, the researcher, the participants, and their relationships all contribute to

162

TOVA BAND-WINTERSTEIN ET AL.

the research and meanings are discussed in a specific social context. In other words, the qualitative research is a shared product of the researcher − participant relationship, and also of the meanings attributed by both sides to the phenomenon under study (Finlay, 2011; King & Horrocks, 2010). Therefore, different researchers might gain different insights regarding the same data. The subjective nature of the qualitative research and the involvement of the researcher require specific attention throughout the research process. One of the ways to deal with this and to achieve trustworthiness is the use of reflexivity (Band-Winterstein, Doron, & Naim, 2012; Finlay, 2011). Data analysis was performed according to the phenomenological method. First, to establish “bracketing,” the researchers reflected on their experiences, biases, and prejudices regarding elder abuse and mental illness (Gearing, 2004; Tufford & Newman, 2012). The research team was involved in analyzing the data. Each researcher, separately, then conducted thematic content analysis of the data prior to a joint comparative examination of the individual analyses. The analysis was performed using a horizontal coding process by pinpointing statements about the participants’ experiences with the phenomenon, for example, identifying a wide range of parental feelings about the ACMD. The next step was grouping the statements into units of meaning, including quotes to describe the participants’ experiences and perceptions, for example, gathering quotes relating to the parents’ role in the child’s life. The following step involved identifying the emerging themes by shifting from the descriptive to the interpretive levels of analysis, for example, the conceptualization of the parental identity (Lincoln & Guba, 1985; Smith, Flowers, & Larkin, 2009). Data were organized based on agreed-upon themes identified in participants’ narratives (King & Horrocks, 2010). Adherence to this procedure enhanced the study’s credibility (Lincoln & Guba, 1985).

FINDINGS The findings revealed three themes that expressed the meaning attributed to life with ACMD in the shadow of violence: (1) constructing parental identity in a shared reality of violence, (2) social and family networks as a resource in coping with ACMD, and (3) keeping a daily life routine as an anchor in a vulnerable, abusive relationship.

Living with an Abusive Adult Child with Mental Disorder

163

CONSTRUCTING PARENTAL IDENTITY IN A SHARED REALITY OF VIOLENCE From the perspective of old age, parents were able to reflect on their parental identity that emerged as a result of the unique life experience of living with ACMD. Manifestations of suffering and victimhood were common to all parents and served as a “flag” symbolizing the type of parental identity. Some viewed their parenthood as devoted, others constructed it into “martyrdom,” another group of parents viewed it as a bitter and deprived parenthood, and others constructed their parental identity as bravery against all odds. These four categories were mutually exclusive and represented by a somewhat similar amounts of parents in each category. These constructions enabled the parents in this stage of life to cope with the harsh life reality and to give it meaning and purpose.

“No-one Else Came but Me”: A Devoted Parenthood Devoted parenthood (represented by three parents) is described as characterized by ongoing care and a strong sense of responsibility as demonstrated in the following quote: … He was hospitalized … I’d visit him every day; no-one else came but me. It was hard for me. I would go out to work at five in the morning, finish at four in the afternoon and go to see him. It was like that for two and a half years. He has been living here with me for 25 years since then. I make sure he gets up to work every morning … Some people my age go to nursing homes but I have to suffer from his brutal attacks … he is like a lion, like a lion in the jungle … After everything he does to me, I still take care of him, he is still my son. (Haim, 89)

Haim constructs his devoted parental identity over two phases: the first relates to his son’s years of hospitalization and the second to living with and caring for his child. He describes his loyalty and consistency over the years, which construct his identity as a devoted parent despite the child’s severely violent behavior. Haim acknowledges that, at this point in his life, he should have been free from everyday parental obligations, which stands in sharp contrast to his current situation. His selfconstruction as a devoted father serves as a resource that keeps him going in this Sisyphean role.

164

TOVA BAND-WINTERSTEIN ET AL.

“I’m Alive but I’m not Living”: A Parenthood of “Martyrdom” Other parents emphasized the agony accompanying life with an ACMD and described an exhausting daily struggle. This type of parenthood emphasizes total self-deprecation and giving up their own personal fulfillment. This parenthood of “martyrdom” was represented by four parents and is illustrated by Bella in the following quote: … she [the daughter] came along and got sick. I took care of everything, what could I do? … Throw her out? She hits me, she yells at me, she turns the house upside down. These are my troubles. I’m alive but I’m not living … my life is gone … for good or for bad; that’s how it’s been for years. My life is worth nothing …. (Bella, 80)

Bella perceives herself as a total victim due to her self-sacrifice and the loss of her entire life. The essence of this self-sacrifice narrative is conveyed through the statement: “I’m alive but I’m not living.” This means that she was overwhelmed by the ongoing hardships of residing with her daughter. She acknowledges the fact that she gave up her life and continues to fulfill her parental obligation to her daughter. Constructing a parental identity of martyrdom strengthens Bella’s ability to self-sacrifice, apparently enabling her to accept her destiny and keep living in this harsh reality.

“What Kind of a Life Have I Had?”: Bitter and Deprived Parenthood Parents who constructed a bitter and deprived parental identity express frustrations and anger toward their surroundings including family members, social policy makers, and authorities. This type of parenting differs from the former categories by the intensity of negative feelings that “color” their reality as represented by four parents and illustrated in the following quote: … He has been living with me now for 24 years … This government is useless, they don’t care … there is nothing I can do … What kind of a life have I had? What have I done? I didn’t see anything in my life. I just had babies … Now the married ones come for an hour or two and after that I’m left alone again with this one [the abusive son] … other people get some help from their children or their husband. My husband died young … I have nothing … let them all go to hell … I feel so bad for my son … he is 33 years old … I must keep quiet and accept the good and the bad. Sometimes, he curses me, but he has a good heart …. (Gila, 75)

Living with an Abusive Adult Child with Mental Disorder

165

Gila describes a lifetime of suffering and hardship. Her bitterness is expressed by a strong sense of missed opportunity, which colors her entire life and parental identity. Although she managed to cope with her son alone during the earlier years, at this stage of her life, she feels deprived of help and support from her family and from the authorities. This construction of parental identity seems to preserve her “fighting spirit” and motivation, allowing her to withstand and survive her son’s behavior.

“My Experience Gives Me Strength”: Parenthood as Bravery against All Odds Parents who presented a parental identity of bravery against all odds (represented by five parents) tended to construct an empowered view of life that is rooted in their vast experience in several life arenas. These experiences provide resources for coping with the ACMD from a position of strength and confidence, as illustrated in the following quote: I got pregnant when I was 18 … So I kept on going to work and when I gave birth, I thought, I’m so young and without a husband, how am I going to cope with everything? … and I also bought an apartment in a public housing plan that helps single mothers … Now I have the tools to take care of her and to stand her violent attitude towards me. My experience gives me strength, so it’s much easier now … so I’m here to help her out … and I think things will get better. (Lily, 58)

Lily presents a narrative of a strong, capable woman, who prevailed against all odds. Her intense life and the challenges with which she coped successfully ultimately led to her construction of a strong parental identity that contains elements of self-growth. She emphasized her current position as empowered by life’s experience that provides her with control over her relationship with her daughter. Old age is apparently perceived as a source of strength rather than of weakness in her parental experience. In summary, it seems that all parents are deeply involved in the relationship with the ACMD. They recognize their deviant behavior and are all preoccupied in defining their parental identity in relation to the child’s behavior and condition. Each parental type represents differences in the array of feelings, attitudes and coping strategy toward life with ACMD.

166

TOVA BAND-WINTERSTEIN ET AL.

SOCIAL AND FAMILY NETWORKS AS A RESOURCE IN COPING WITH ACMD The existential experience of the vulnerability of the aging parent emphasizes his need of a social and familial network. This network provides support, a sense of a “safe haven,” and assists in gathering strength. The importance of friendship for the aging parent is illustrated in the following quote: I don’t have enough tears in my eyes anymore. … my friends call me and rescue me: “Are you coming to the mall with us? Let’s go and get ice-cream.” They don’t want me to stay at home. They know what I’m going through. I have fantastic friends, I shower and dress up; I go out and come back at night. My friends don’t like TV. We talk about life, smoke a cigarette, share a piece of cake. Sometimes, we have a cup of coffee at the beach. (Miri, 86)

Miri emphasizes the importance of friendships at her present stage of life. These friendships provide her with feelings of intimacy and empathy. For her, her friends are a high-quality resource of support that extracts her from the whirlpool of the relationship with her son, providing her with space and a refuge from the daily abusive experience. In addition, these friendships give meaning to a life with constant suffering and introduce sparks of belonging, satisfaction, and well-being into her life. Another way of giving meaning to life in the shadow of violence is through volunteering outside the home: I dont want any money. I’ve been volunteering with blind people for 30 years. It feels good. Now I want start volunteering in a cancer organization … they love me, I have a good heart …. (Lea, 79)

Lea uses her volunteer activity to construct a positive self that is so important to her in light of her daily difficulties with her son. These activities give her a sense of devotion, of giving, and of love, and give meaningful justification for being out of the house. Another way of creating meaningful social networks is through participation in a senior citizens day center: I come here and it makes me feel good. Sometimes, a woman comes over. She plays the piano and we sing along with her. She always looks for me: “Where is Funy?” She knows I love singing and dancing; she knows I’m cheerful. She asks me to sing songs from my homeland, so I get up and dance around with her … I love this; it helps me forget about my daughter for a little while …. (Funy, 78)

Funy channels her difficulties into participation in social activities. Her outings to the senior citizens center, where she is appreciated for her

Living with an Abusive Adult Child with Mental Disorder

167

positive personality and her innate joy of life, are a source of belonging and a sense of value. Some of the participants’ social networks were based on family relationships. Most of the participants reported estrangement, alienation, and distancing from other family members due to the stigmatizing presence of mental illness and the deviant abusive behavior of the ACMD. Thus, cooperation and support from family members were not taken for granted: My youngest son is a good man. Every weekend, he is here with his kids. Never said a bad word to me. Never. He helps out; he pays the mortgage. Once I could manage on my own. My other sons ran away from home. They didn’t care at all. He was my only help. (Nili, 73)

Nili presents a complex familial relationship, in which only some of the family members share her problems and help out, while others distance themselves from the situation. Over the years, Nili has coped with raising an abusive ACMD on her own. It seems that, in this stage of her life, she needs help on a daily basis, both financially and physically, and her younger son provides this help, leaving her grateful for his presence in her life. This creates a sort of balance between her experience as an abused parent and a valued mother.

KEEPING A DAILY LIFE ROUTINE AS AN ANCHOR IN A VULNERABLE, ABUSIVE RELATIONSHIP With the parents’ vulnerability due to aging, they talk about the need for continuity and to cling to a daily routine. The involvement in the daily routine serves as an anchor in a complex reality that enables the construction of the meaning of life around the feeling of being needed and competent. The participants describe the positive aspects of maintaining a routine and the ways keeping it in context of abusive relationship is an ongoing struggle. I don’t want to live with him but I have no choice; he can’t live on his own. Without me, he will be out on the street. There is no-one to take care of him except me. There is no peace around the house. I’m never at peace, but I try to also take care of myself: watch some TV, take my medication, eat, drink, take a shower …. (Mimi, 75)

Mimi presents a life with a dead end. She has to take care simultaneously of her son’s needs and her own changing needs that come with aging. In this chaotic, shared reality, she struggles to maintain a daily routine that addresses her needs. This struggle may be a difficult one in light of the child’s deviant and unpredictable behavior.

168

TOVA BAND-WINTERSTEIN ET AL.

The need for tending the aging needs of the parent and permanently meeting them is illustrated in the following quote: She gets up in the middle of the night and goes into the shower. I get up and want to go to the bathroom, and I can’t because she is already there! And she doesn’t let me in. I’m taking Fusid [a diuretic] and I can’t hold it in … when I come to the bathroom, she tells me to go to the neighbor’s bathroom. Well, how can I go knock on their door in the middle of the night? And this Fusid, it has no mercy, I just have to go, excuse me for saying this, but I finally pee inside a bottle. I’m embarrassed to tell you this …. (Lea, 79)

Although Lea is used to her daughter’s deviant behavior, at this point in her life, she has difficulty integrating her new health-related needs into the relationship. The daughter seems to be oblivious to her mother’s situation. She continues to put her own needs first while subjecting her mother to health-threatening conditions. Lea is aware of her own needs and of the changes in her physical condition due to medication but has to struggle to meet her modest needs. Another aspect of gaining a sense of competence from a daily routine is by clinging to everyday chores: I have a lot of laundry to do. From Friday, I’m cleaning, washing the dishes, doing the laundry. I’m not a person who can sit around all day, and I already said my prayers today. The children come over every Saturday. I make them dinner, it makes me feel good, I feel that I can give. I cook and bake and get everything ready, and when I’m doing this, I don’t think about her [the abusive daughter] (Suzzane, 72).

Suzzane’s day is packed with instrumental housekeeping tasks and caring for her daughter’s needs. She elaborates on her daily chores, giving them a significant meaning, which provides relief from the constant involvement and confrontation with her daughter. Maintaining a daily routine helps her perceive herself as competent and needed, and this ensures that she remains a significant figure in the family context.

DISCUSSION This paper proposes a significant perspective on violence and crime in aging families, enabling exploration of the different ways in which parents try to cope in such an extreme familial context. One way of coping is to construct a meaningful parental identity that helps them to commit to living with the violent child until the end of life. Another way is to seek external social resources to balance the physical and emotional

Living with an Abusive Adult Child with Mental Disorder

169

distress and intense friction caused by the abusive relationship. A third way is to use the daily routine as an anchor for maintaining feelings of self-worth in a vulnerable abusive relationship. In normative families, adult children have the ability and the willingness, as well as the moral dictum, to care for their parents (Allen, Blieszner, & Roberto, 2000). In violent families, they are aware of this expectation but have no wish to act accordingly (Pillemer, 1985; Suitor & Pillemer, 1988; Suitor, Pillemer, Keeton, & Robison, 1995). On the contrary, it seems that the parental obligation toward the children is timeless and never-ending even now, in old age. The constant need to take care of the child leads the parents to become very involved in the child’s life, at which point their own needs become secondary. These parents may be described as “eternal parents” (Kelly & Kropf, 1995) as they continue to attend to their child’s needs even though the child has reached adulthood. The findings show that the parental identity varies among parents (parental identities of devotion, martyrdom, bitterness, and bravery against all odds), while eternal parenting is common to all these types of identity. From the point of view of all the participants, victimhood was found to be a central motif, which characterized the formation of reciprocal relationships in these families. Victimhood is known to be a product of oppression and situations of power. People in such a state experience a loss of self-control (Rifai & Ames, 1977). An additional aspect of victimhood is old age, which increases the older person’s vulnerability to various types of victimhood, including physical, functional, and social losses (Reiman, 1975; Yin, 1985). By placing themselves in the role of victim, the parents created a retrospective narrative, which served as a cornerstone in the construction of parental identity. The role of victim seems to challenge parents to maintain the obligation of endless caring for their children. The construction of a personal and unique parental identity might explain these parents’ stamina and determination as they approach their own challenges of old age. Another possible explanation to the parent’s resiliency is the development of mechanisms of self-defense or survival that enables them to maintain their parental role in such circumstances. This may contribute to the construction of “survivor” parental identity. Another aspect of the findings reveals the existential suffering that accompanies life with ACMD (Frankl, 1985). It is evident from the findings that even though the lives of parents with ACMD might be paved with hardship and loss, at this point in their lives, they find a way to balance this grim picture and create compensating mechanisms such as seeking the support of social networks and maintaining a daily life

170

TOVA BAND-WINTERSTEIN ET AL.

routine. It seems that the suffering that is well embedded into these parents’ lives is now intensified by old age on the one hand, and becomes bearable by the stock of life’s knowledge and experience, on the other (Turner, 1982). In other words, in this stage of their lives, parents are aware of their own needs and vulnerability and of the upcoming future that is bound to be difficult. Thus, the suffering becomes an achievement (Frankl, 1985) as they creatively gather social sources of support and maintain an organized daily routine that helps them cope with life’s reality. These resources cannot be taken for granted in old age and serve as a compensating mechanism of survival (Baltes & Baltes, 1990). Relying on rewarding social ties and keeping a comfortable daily routine echoes the concept of “socio-emotional selectivity” (Carstensen, 1991). This concept emphasizes the individual’s awareness of “time left,” which is associated with the pursuit of emotionally rewarding states. Activities that are pleasant or meaningful are more compelling under conditions in which time is perceived as limited (Carstensen, Isaacowitz, & Charles, 1999). In accordance with this perspective, it seems that participants in the current study chose to emphasize positive experiences such as seeking activities outside of the house, enjoying the company of friends and family members, and preserving a self-care routine. Another perspective that may be relevant to our findings is the framework of enduring (Morse, 2010), which is conceptualized as an adaptive form of suffering as it enables functioning, prevents disintegration of self, and includes enduring to survive, to live, and to die (O¨hman, So¨derberg, & Lundman, 2003). The participants in our study, who suffer while living in a reality of violence, can also be viewed as people who endured, because they have coped and adapted to the reality of living with ACMD. The findings regarding maintaining a daily routine is a good example of this adaptive process of enduring. To stress the importance of the process of enduring, Foss and Na˚den (2009) asserted that in order to be prepared and strong enough to deal with emotional suffering, one must have the ability to recognize what has been endured. Finally, the findings show the parent’s ability to create a continuity and routine within the harsh abusive reality. In his perception of normal aging, Atchley (1989) claims that older adults attempt to preserve and maintain existing internal and external structures, and they prefer to accomplish this objective by using strategies tied to their past experiences of themselves and their social world. Continuity is perceived as an adaptive strategy that is promoted by both individual preference and social approval. For the participants in the current study, continuity was perceived as a way of balancing

Living with an Abusive Adult Child with Mental Disorder

171

between the challenges of old age and the difficulties of caring for an ACMD. By so doing, parents preserve self-worth, feelings of competence, and of being needed.

LIMITATIONS AND RECOMMENDATIONS FOR FURTHER STUDY This study focused on the parent’s perspective. To broaden the understanding of the phenomenon, triangulation can be used by participatory observation, interviews with the child with the mental disorder, or with other family members. In addition, the participants had all been labeled by the welfare authorities as victims of violence, and as such, were defined as a clinical sample. Future research including a sample of self-declared victims of elder abuse might enrich the findings of this study. Finally, the research did not cover a wide range of mental disorders and focused mainly on schizophrenia and drug and alcohol abuse. Therefore, we suggest including other mental disorders such as borderline personality disorder, antisocial personality disorder, and others.

PRACTICAL IMPLICATIONS The present study revealed parents’ perspectives regarding their life with ACMD and the way they cope with this reality. We recommend that intervention with such families focuses on the life review process as a therapeutic tool (Butler, 1964), which will enable parents to achieve closure to their life story, emphasizing enduring and viewing suffering as an achievement rather than defeat. Together with this, interventions should provide a “safety belt,” including health services, public social networks, and knowledge regarding their right for self-protection. Since parents far along in their aging process, relevant social policies that provide sheltered housing for both parents and ACMD disorder, which will address the needs of both population, should be considered. The knowledge regarding the finiteness of life can be utilized to emphasize the inevitable separation process and thus provide future oriented scripts for these children. These programs should include a rehabilitation plan that will enable both parents and children to undergo this process.

172

TOVA BAND-WINTERSTEIN ET AL.

REFERENCES Addington, J., McCleery, A., & Addington, D. (2005). Three-year outcome of family work in an early psychosis program. Schizophrenia Research, 79(1), 107 116. Allen, K. R., Blieszner, R., & Roberto, K. A. (2000). Families in the middle and later years: A review and critique of research in the 1990s. Journal of Marriage and the Family, 62, 911 926. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders. Diagnostic and Statistical Manual of Mental Disorders. Anetzberger, G. J., Korbin, J. E., & Austin, C. (1994). Alcoholism and elder abuse. Journal of Interpersonal Violence, 9(2), 184 193. Atchley, R. C. (1989). A continuity theory of normal aging. The Gerontologist, 29(2), 183 190. Baltes, P. B., & Baltes, M. M. (1990). Psychological perspectives on successful aging: The model of selective optimization with compensation. In P. B. Baltes & M. M. Baltes (Eds.), Successful aging: Perspectives from the behavioral sciences (pp. 1 34). Cambridge: Cambridge University Press. Band-Winterstein, T. (2015). Whose suffering is this? Narratives of adult children and parents in long-term abusive relationships. Journal of Family Violence, 30, 123 133. doi:10.1007/s10896-014-9660-z Band-Winterstein, T., Doron, I. I., & Naim, S. (2012). Elder self-neglect: A geriatric syndrome or a life course story? Journal of Aging Studies, 26(2), 109 118. Band-Winterstein, T., Smeloy, Y., & Avieli H. (2014). Shared reality of the abusive and the vulnerable: The experience of aging for parents living with abusive adult children coping with mental disorder. International Psychogeriatrics, 26, 1917 1927. doi:10.1017/ S1041610214001495 Ben-Ari, A., & Enosh, G. (2011). Processes of reflectivity knowledge construction in qualitative research. Qualitative Social Work, 10(2), 152 171. Bowen, G. A. (2008). Naturalistic inquiry and the saturation concept: A research note. Qualitative Research, 8(1), 137 152. Brandel, B., Dyer, C. B., Heiseler, C. J., Otto, J. M., Stiegel, L. A., & Thomas, R. W. (2007). Elder abuse detection and intervention: A collaborative approach. New York, NY: Springer. Butler, R. N. (1964). The life review: An interpretation of reminiscence in the aged. In R. Kastenbaum (Ed.), New thoughts on old age (pp. 265 280). Berlin Heidelberg: Springer. Carstensen , L. L. (1991). Socioemotional selectivity theory: Social activity in life-span context. Annual Review of Gerontology and Geriatrics, 11, 195 217. Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. (1999). Taking time seriously: A theory of socioemotional selectivity. American Psychologist, 54(3), 165. Cohler, B. J., Pickett, S. A., & Cook, J. A. (1991). The psychiatric patient grows older: Issues in family care. In E. Light, S. A. Pickett & J. A. Cook (Eds.), The elderly with chronic mental illness (pp. 82 110). New York, NY: Springer. Comijs, H. C., Smit, J. H., Pot, A. M., Bouter, L. M., & Jonker, C. (1999). Risk indicators of elder mistreatment in the community. Journal of Elder Abuse & Neglect, 9(4), 67 76.

Living with an Abusive Adult Child with Mental Disorder

173

Cook, J. A., Cohler, B. J., Pickett, S. A., & Beeler, J. A. (1997). Life-course and severe mental illness: Implications for caregiving within the family of later life. Family Relations, 46, 427 436. Retrieved from http://www.jstor.org/stable/585102 Cook, J. A., Lefley, H. P., Pickett, S. A., & Cohler, B. J. (1994). Age and family burden among parents of offspring with severe mental illness. American Journal of Orthopsychiatry, 64, 435 447. Cooper, C., Selwood, A., & Livingston, G. (2008). The prevalence of elder abuse and neglect: A systematic review. Age and ageing, 37, 151 160. doi:10.1093/ageing/afm194 Davis, L. J. (2009). Obsession: A history. Chicago, IL: University of Chicago Press. Denzin, N. K. (1983). A note on emotionality, self, and interaction. American Journal of Sociology, 89, 402 409. Dixon, J., Manthorpe, J., Biggs, S., Mowlam, A., Tennant, R., Tinker, A., & McCreadie, C. (2010). Defining elder mistreatment: Reflections on the United Kingdom study of abuse and neglect of older people. Ageing and Society, 30, 403. Finlay, L. (2011). Phenomenology for therapists: Researching the lived world. New York, NY: Wiley. Foss, B., & Na˚den, D. (2009). Janice Morse’theory of suffering A discussion in a caring science perspective. Nordic Journal of Nursing Research & Clinical Studies/Va˚rd I Norden, 29(1), 14 18. Frankl, V. (1985). Man’s search for Meaning. New York, NY: Pocket Books. Fulmer, T., Paveza, G., VandeWeerd, C., Fairchild, S., Guadagno, L., Bolton-Blatt, M., & Norman, R. (2005). Dyadic vulnerability and risk profiling for elder neglect. The Gerontologist, 45(4), 525 534. Gearing, R. E. (2004). Bracketing in research: A typology. Qualitative Health Research, 14, 1429 1452. Ghosh, S., & Greenberg, J. (2009). Aging fathers as caregivers to an adult son or daughter with schizophrenia: Implications for practice. Psychiatric Services, 60, 982 984. Green, J., & Thorogood, N. (2006). Qualitative methods for health research. London: Sage. Greenberg, J. R., McKibben, M., & Raymond, J. A. (1990). Dependent adult children and elder abuse. Journal of Elder Abuse and Neglect, 2(1 2), 73 86. doi:10.1300/ J084v02n01_05 Greenberg, J. S., Greenley, J. R., & Benedict, P. (1994). Contributions of persons with serious mental illness to their families. Psychiatric Services, 45, 475 480. Hertz R. (1997). Reflexivity and voice. Thousand Oaks, CA: Sage. Homer, A. C., & Gilleard, C. (1990). Abuse of elderly people by their carers. British Medical Journal, 301, 1359 1362. doi:10.1136/bmj.301.6765.1359 Kaufman, A. V., Scogin, F., MacNeil, G., Leeper, J., & Wimberly, J. (2010). Helping aging parents of adult children with serious mental illness. Journal of Social Service Research, 36, 445 459. Kelly, T. B., & Kropf, N. P. (1995). Stigmatized and perpetual parents: Older parents caring for adult children with life-long disabilities. Journal of Gerontological Social Work, 24(1 2), 3 16. King, N., & Horrocks, C. (2010). Interviews in qualitative research. London, England: Sage. Korbin, J. E., Anetzberger, G. J., & Austin, C. (1995). The intergenerational cycle of violence in child and elder abuse. Journal of Elder Abuse & Neglect, 7(1), 1 15. doi:10.1300/ J084v07n01_01

174

TOVA BAND-WINTERSTEIN ET AL.

Lachs, M., Bachman, R., Williams, C. S., Kossack, A., Bove, C., & O’Leary, J. R. (2006). Violent crime victimization increases the risk of nursing home placement in older adults. The Gerontologist, 46, 583 589. doi:10.1093/geront/46.5.583 Lachs, M. S., & Pillemer, K. (2004). Elder abuse. Lancet, 364, 1263 1272. Lachs, M. S., Williams, C., O’Brien, S., Hurst, L., & Horwitz, R. (1997). Risk factors for reported elder abuse and neglect: A nine-year observational cohort study. The Gerontologist, 37, 469 474. Lauronen, E., Veijola, J., Isohanni, I., Jones, P. B., Nieminen, P., & Isohanni, M. (2004). Links between creativity and mental disorder. Psychiatry, 67(1), 81 98. Lee R. M., & Renzetti C. M. (1993). The problems of researching sensitive topics: An overview and introduction. In C. M. Renzetti & R. M. Lee (Eds.), Researching sensitive topics (pp. 3 13). Newbury Park, CA: Sage. Lefley, H. P. (1987). Impact of mental illness in families of mental health professionals. The Journal of Nervous and Mental Disease, 175, 613 619. Lefley, H. P. (2003). Changing caregiving needs as persons with schizophrenia grow older. In C. I. Cohen (Ed.), Schizophrenia into later life: Treatment, research, and policy (pp. 251 268). Arlington, VA: American Psychiatric Publishing. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage. Lowenstein, A., Eisikovits, Z., Band-Winterstein, T., & Enosh, G. (2009). Is elder abuse and neglect a social phenomenon? Data from the first national prevalence survey in israel. Journal of Elder Abuse & Neglect, 21(3), 253 277. doi:10.1080/08946560902997629 Magan˜a, S. M., Greenberg, J. S., & Seltzer, M. M. (2004). The health and well-being of black mothers who care for their adult children with schizophrenia. Psychiatric Services, 55, 711 713. Marmolejo, I. I. (2008). Elder abuse in the family in Spain. Queen Sofia Center, Valencia, Spain. Retrieved from http://www.inpea.net/images/Spain_Report_2008_Elder.pdf McGarry, J., & Arthur, A. (2001). Informal caring in late life: A qualitative study of the experiences of older carers. Journal of Advanced Nursing, 33(2), 182 189. McNiel, D. E., & Binder, R. L. (1994). The relationship between acute psychiatric symptoms, diagnosis, and short-term risk of violence. Psychiatric Services, 45(2), 133 137. Monahan, J., & Steadman, H. J. (Eds.). (1994). Violence and mental disorder: Developments in risk assessment. Chicago, IL: University of Chicago Press. Morse, J. M. (2000). Determining sample size. Qualitative Health Research, 10, 3 5. doi:10.1177/104973200129118183 Morse, J. M. (2010). The praxis theory of suffering. In J. B. Butts & K. L. Rich (Eds.), Philosophies and theories for advanced nursing practice. Sudbury, MA: Jones & Bartlett. Moustakas, C. (1994). Phenomenological research methods. Thousand Oaks, CA: Sage. Muehlbaur, M., & Crane, A. (2006). Elder abuse and neglect. Journal of Psychological Nursing, 44(11), 43 48. Nahmiash, D. (2004). Powerlessness and abuse and neglect of older adults. Journal of Elder Abuse & Neglect, 14(1), 21 47. Niimura, H., Nemoto, T., Yamazawa, R., Kobayashi, H., Ryu, Y., Sakuma, K., & Mizuno, M. (2011). Successful aging in individuals with schizophrenia dwelling in the community: A study on attitudes toward aging and preparing behavior for old age. Psychiatry and Clinical Neurosciences, 65, 459 467. O¨hman, M., So¨derberg, S., & Lundman, B. (2003). Hovering between suffering and enduring: The meaning of living with serious chronic illness. Qualitative Health Research, 13, 528 542.

Living with an Abusive Adult Child with Mental Disorder

175

O’Keeffe, M., Hills, A., Doyle, M., McCreadie, C., Scholes, S., Constantine, R., … Erens, B. (2007). UK study of abuse and neglect of older people: Prevalence survey report. London: National Centre for Social Research. Patton, M. Q. (2002). Qualitative research and evaluation methods (3rd ed.). Thousand Oaks, CA: Sage. Payne, B. K. (2005). Crime and elder abuse: An integrated perspective. Springfield, IL: Charles C Thomas. Payne, B. K., & Gainey, R. (2005). Family violence and criminal justice: A life course approach (2nd ed.). Cincinnati, OH: Anderson. Perel-Levin, S. (2008). Discussing screening for elder abuse at primary health care level. Geneva: World Health Organization. Pillemer, K. (1985). The dangers of dependency: New findings on domestic violence against the elderly. Social Problems, 33(2), 146 158. Retrieved from http://www.jstor.org/ stable/pdfplus/800558.pdf Pillemer, K., & Finkelhor, D. (1988). The prevalence of elder abuse: A random sample survey. The Gerontologist, 28, 51 57. doi:10.1093/geront/28.1.51 Provencher, H. L., & Mueser, K. T. (1997). Positive and negative symptom behaviors and caregiver burden in the relatives of persons with schizophrenia. Schizophrenia Research, 26 (1), 71 80. Ramsey-Klawsnik, H. (2000). Elder abuse offender: A typology. Generations, 24(2), 17 22. Reay, A. C., & Browne, K. D. (2001). Risk factor characteristics in carers who physically abuse or neglect their elderly dependants. Aging & Mental Health, 5(1), 56 62. Reiman, J. H. (1975). Aging as victimization: Reflections on the American way of (ending) life. In J. Goldsmith & S. S. Goldsmith (Eds.), Crime and the elderly (pp.77 83). Lexington, MA: Lexington. Rifai, M. A. Y., & Ames, S. A. (1977). Social victimization of older people: A process of social exchange. In M. A. Y. Rifai (Ed.), Justice and older Americans (pp. 47 62). Lexington, MA: Lexington. Sadock, B. J., Kaplan, H. I., & Sadock, V. A. (2007). Kaplan & Sadock’s synopsis of psychiatry: Behavioral sciences/clinical psychiatry. Philadelphia, PA: Lippincott Williams & Wilkins. Saunders, J. C. (2003). Families living with severe mental illness: A literature review. Issues in Mental Health Nursing, 24(2), 175 198. Schiamberg, L. B., & Gans, D. (2000). Elder abuse by adult children: An applied ecological framework for understanding contextual risk factors and the intergenerational character of quality of life. International Journal of Aging and Human Development, 51, 329 360. Seltzer, M. M., Greenberg, J. S., Floyd, F. J., Pettee, Y., & Hong, J. (2001). Life course impacts of parenting a child with a disability. American Journal of Mental Retardation, 106, 265 286. Smith, J. A., Flowers, P., & Larkin, M. (2009). Interpretative phenomenological analysis: Theory, method and research. Thousand Oaks, CA: Sage. Solomon, P. L., Cavanaugh, M. M., & Gelles, R. J. (2005). Family violence among adults with severe mental illness: A neglected area of research. Trauma, Violence, & Abuse, 6, 40 54. Straznickas, K. A., McNiel, D. E., & Binder, R. L. (1993). Violence toward family caregivers by mentally ill relatives. Psychiatric Services, 44(4), 385 387. Suitor, J. J., & Pillemer, K. (1988). Explaining intergenerational conflict when adult children and elderly parents live together. Journal of Marriage and Family, 50, 1037 1047. Retrieved from http://www.jstor.org/stable/352113

176

TOVA BAND-WINTERSTEIN ET AL.

Suitor, J. J., Pillemer, K., Keeton, S., & Robison, J. (1995). Aged parents and aging children quality. In R. Blieszner & V. H. Bedford (Eds.), Handbook of aging and family (pp. 223 241). Westport, CT: Greenwood Press. Swan, R. W., & Lavitt, M. (1988). Patterns of adjustment to violence in families of the mentally ill. Journal of Interpersonal Violence, 3(1), 42 54. Tufford, L., & Newman, P. (2012). Bracketing in qualitative research. Qualitative Social Work, 11, 80 96. Turner, J. H. (1982). Alternative approach to interactionist theorizing: Phenomenology and ethnomethodology. In J. H. Turner (Ed.), The structure of sociological theory (pp. 389 399). Homewood, IL: The Dorsey Press. Vaddadi, K. S., Gilleard, C., & Fryer, H. (2002). Abuse of carers by relatives with severe mental illness. International Journal of Social Psychiatry, 48, 149 155. doi:10.1177/ 002076402128783208 Vaddadi, K. S., Soosai, E., Gilleard, C. J., & Adlard, S. (1997). Mental illness, physical abuse and burden of care on relatives: A study of acute psychiatric admission patients. Acta Psychiatrica Scandinavica, 95, 313 317. Walsh, C. A., Ploeg, J., Lohfeld, L., Horne, J., MacMillan, H., & Lai, D. (2007). Violence across the lifespan: Interconnections among forms of abuse as described by marginalized Canadian elders and their care-givers. British Journal of Social Work, 37, 491 514. doi:10.1093/bjsw/bcm022 Wolf, R. S. (1996). Elder abuse and family violence: Testimony presented before the U.S. senate special committee on aging. Journal of Elder Abuse & Neglect, 8(1), 81 96. doi:10.1300/J084v08n01_05 World Health Organization (WHO) & International Network for the Prevention of Elder Abuse. (2001). Missing voices: The views of older persons on elder abuse. Geneva: WHO. Yin, P. (1985). Victimization and the aged. Springfield, IL: Charles C. Thomas Publisher.

PARENTAL INFLUENCE AND DATING VIOLENCE AMONG STUDENTS AT NIGERIAN UNIVERSITIES Olufemi Adeniyi Fawole and Ebenezer Bayode Agboola ABSTRACT Purpose Dating violence has, in recent times, been a social problem that has been creating different levels of concern especially among parents, and those in the academia, in Nigeria. Studies have shown causes to be largely due to personality types, but little relate it with violence between the parents of the perpetrator. This study examines the influence of violence between parents and the effect on dating violence among students in Nigerian Universities. Design/Methodology/Approach Questionnaires were administered to 460 students who had experienced violence in their dating relationship. The study had 55.7% of the respondents being females. Findings All of the respondents had experienced dating violence at one point or the other in their relationship. About 36.7% of the

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 177 187 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009009

177

178 OLUFEMI ADENIYI FAWOLE AND EBENEZER BAYODE AGBOOLA

respondents reported to having been in dating relationship with a partner who had witnessed violence in the home. Data analyzed using Pearson Product Moment Correlation Co-efficient indicate that the variables of parental conflict and dating violence were significantly positively correlated among the students. Originality/Value The study was limited because it focuses on only one university, and research in the area of dating violence in Nigeria has not been extensively reported. The study therefore emphasizes the impact of socialization process on dating behavior of young adults in Nigeria as well as the need to have further studies on these dating patterns. This study will serve as addition to the gradually increasing literature on dating behavior of young adults in the Nigerian society. Keywords: Dating violence; Nigerian Universities; violent behavior; parental conflict; intimacy

INTRODUCTION Dating violence has been conceptualized as involving range of incidents from physical force, emotional, and verbal abuse to murder (Hickman, Jaycox, & Aronoff, 2004). It can also be defined as a form of relationship which involves insult or assault. According to Iconis (2013), dating violence is a significant problem on college campuses. Studies have shown the prevalence of this incident. For instance, Ajuwon, Funmilayo, and Osungbade (2011) found that students in Nigerian Universities are both victims and perpetrators of violent behaviors in dating. Boladale et al. (2004) reported that about a third of students in Nigerian Universities have experienced dating violence. Lewis and Fremouw (2001) reported that as many as one in three college couples will be involved in at least one incident of violence during their dating relationship. In Nigeria, as in many other African countries, beating of wives and children is widely sanctioned as a form of discipline (UNICEF, 2001). By this act, parents believe they are instilling discipline in their children, husbands on their wives. This kind of violence functions therefore as a means of enforcing conformity with the role of a woman. Violent punishments such as corporal punishment have been found to lead to delinquency in

Parental Influence and Dating Violence among Students

179

adolescents and later to violent behaviors or crimes. Violence in adolescence may make the adolescent angry and confused and may direct the anger toward the parents or other children. This could account for such behaviors as violent behaviors toward dating partners in the university. While some injuries may not be physical, they could be emotional. Victims often have low self-esteem, deep feelings of anger and stress, which could lead to depression. An abused child may go into relationships with an aggressive mode (CDC, 2006). Dating violence is a common occurrence and while it exists with either males or females being victims, violence in dating relationship is mutual, though the nature of violence can be different. Molidor, Tolman, and Kober (2000) found that girls were most likely to have severe forms such as being punched or forced to engage in sexual activities; boys were more likely to be pinched, slapped, or scratched. Dating violence may be part of a continuum of violence beginning in dating relationships and continuing through marriage (Boladale, Adesanmi, & Olutayo, 2013). The theory of intergenerational transmission of violence is one of the theories that attempts to explain contributing factors to dating violence. Based on Bandura’s (1977) social learning theory, the intergenerational transmission hypothesis posits that because violence is a learned behavior, exposure to violence in the family of origin may lead to violence in later intimate relationships. Studies that have assessed family interactions more generally have found that family conflicts and negative interactions in the family predict dating violence perpetration and victimization (Andrews, Foster, Capaldi, & Hops, 2000; Linder & Collins, 2005). Also, poorly managed parental conflicts place adolescents at the risk of experiencing dating violence in their own intimate relationships. For instance, parents tend to be restrictive on their views of what is age-appropriate behavior, while adolescents tend to be more accepting than adults of actions that violate inter-personal and social expectations of age-appropriate behavior. Also, parents are more experienced, while adolescents, generally, are adventurous and learn faster from the wealth of parents’ experience, and sometimes without the parents knowing it. Furthermore, the frequency of inter-parental conflicts and the negative behavior during such conflicts as well as the conflict-resolution strategies employed by parents usually affect and influence the adolescents’ ideal picture of a romantic or dating relationship. In addition, inter-parental violence might influence adolescents’ own conflict management skills. These dimensions of parental conflict have deleterious effects on the adjustment

180 OLUFEMI ADENIYI FAWOLE AND EBENEZER BAYODE AGBOOLA

and behavior of children and adolescents (Cummings & Davies, 2002; Grych & Fincham, 1990; Zimet & Jacob, 2001). For instance, more frequent conflict between parents is related to higher levels of emotional distress and behavioral problems among children and adolescents (Kerig, 1996; Tschann et al., 2002). Negative parental behaviors during conflicts, such as anger and withdrawal, are related to children’s externalizing behavior problems (Katz & Gottman, 1993). In addition, poor conflict resolution is related to emotional distress among both children and adolescents (Cummings, Ballard, El-Sheik, & Lake, 1991; Tschann et al., 2002); while adolescent who have poor management skills may be at an increased risk of disagreements with date partners escalating into violence which in turn affects their relational temperament, in all. A study in Nigeria by Izugbara, Emmanuel, and Peter (2008) brought to light few studies on dating violence in Nigeria. They found that women’s views regarding their abuse reinforced the cultural belief that men are naturally violent and that women are sometimes to be blamed. Other studies (Mapayi et al., 2011; Odujirin, 1993) on spousal violence in Nigeria showed that with the theory of intergenerational transmission of violence in play, the courtship or dating violence becomes an important matter of concern in Nigeria. Exposure to violence during childhood is also one of the most commonly studied correlates of dating violence. Dating violence could be linked to the perpetuation of spousal violence which also creates an environment for continued dating violence as children experience and witness violence at home and thus exposing the children to tendency to develop violence in their future relationships (Boladale et al., 2013). Dating violence is a growing concern in Nigeria (Boladale et al., 2013) and appears to be directly related to witnessing of spousal violence which in turn creates a conducive environment for continued dating violence as children experience and witness violence in the home. This thus enables them to develop the propensity to exhibit violence in their own future relationships.

Objectives of the Study The aim of this study is to determine the prevalence of and present coherent dimensions of dating violence among students of University of Ilorin, Nigeria. Specific objectives include:

Parental Influence and Dating Violence among Students

181

1. To examine the prevalence violence among parents of the students of University of Ilorin. 2. To highlight the prevalence of dating violence among students of University of Ilorin. 3. To determine the relationship between violence witnessed at home and dating violence.

Justification of the Study Strong parent child relationships as well as positive parental modeling are necessary for proper development of the child. The study therefore is intended to broaden knowledge in understanding the effects of developing intimate and cordial relationships between parents and with their children. A study of this sort can help to provide objective information that can help parents on how to effectively bring up their children, counselors, and teachers to make more positive and effective impacts on the adolescents and to help guide them in making responsible decisions. It is also intended to enable counselors, sociologists, parents, and teachers to know the significance of effective parenting, keeping conflicts, particularly involving violence away from the home and the impacts on the adolescents’ behavior. Research into dating violence in Nigeria, with particular emphasis on parental violent conflicts, is limited. This study intends to fill the gaps created by earlier studies by emphasizing the functions of parental cordial relationships as well as their influence on the development process of the child.

METHOD Research Design The descriptive survey method was chosen and used for this research. The descriptive survey establishes the incidence, the distribution, and interpretations among sociological and psychological variables in a population. Furthermore, it focuses on people and their beliefs, opinions, perceptions, motivations, and behaviors. The total population included students of the University of Ilorin, Kwara State, Nigeria.

182 OLUFEMI ADENIYI FAWOLE AND EBENEZER BAYODE AGBOOLA

PARTICIPANTS A total of 600 respondents in the third year of their programs in the University of Ilorin were randomly selected to participate in this study, out of which 460 had experienced some form of violence in their dating relationship. Questionnaires were distributed and correctly filled out. Interviews were also conducted on a total of 79 respondents. For the purpose of interpreting all gathered data logically, frequency count and simple percentage were used and Pearson Product Moment Correlation Coefficient was used to analyze the data.

RESULTS A total of 600 questionnaires were distributed out of which 460 had experienced or witnessed violence in their home while growing up and 460 were used in this study. From the total of 460 that had witnessed violence in the home while growing up, a total of 256 of the respondents were females (55.7%), while 235 (44.3%) were males (Table 1). Also from the study, respondents were measured by their perception of the major factor causing the violence in the relationship they were into. A total of 124 (36.7%) of respondent believed that the partner, having witnessed violence in the home, most likely was conditioned into believing that violence was the normal reaction to conflicts in relationships. Another 117 (27.8%) of the respondents were of the opinion that jealousy was the predominant factor influencing violence by their partners, while 106 (13.9%) of respondents confirmed that uncontrollable anger was the cause of dating violence and 113 (21.5%) were of the view that the predominant cause was their denying the partner of sexual intercourse (Table 2).

Table 1.

Distribution of Respondents by Gender.

Gender

Frequent

Percentage (%)

Male Female

235 256

44.3 55.7

Total

460

100.0

183

Parental Influence and Dating Violence among Students

Participants also gave information on who they shared their experience of dating violence with. Majority of the respondents claimed they discussed with their friends more about their dating violence (49.4%), while 24.1% kept it to themselves. Another 16.5% shared it with their parents; some 6.3% with lecturers on campus and 3.8% had discussions with professional counselors whenever such incidences occurred (Table 3). Hypothesis for this study states that there is no significant relationship between parental violence and dating violence of male students in intimate relationships in the University of Ilorin. This was tested using the Pearson correlation coefficient presented in Table 4. The table presents data on the relationship between parental conflicts of students in intimate relationships and dating violence. The results from the table indicate that the variables of parental conflict and dating violence were significantly positively correlated among the students. This is because the calculated r-value of 0.90 is greater than the critical r-value. This invariably implies that violence between parents influences dating violence tendencies of students in intimate relationships in University of Ilorin.

Table 2.

Distribution of Respondents by Perceived Predominant Cause of Dating Violence.

Option

Frequent

Percentage (%)

Partner’s witness of violence in the home Jealousy Uncontrollable anger Denial of sexual intercourse

124 117 106 113

27 25 23 25

Total

460

100.0

Table 3. Distribution of Respondents by Who They Share Their Dating Violence Experience with. Opinion

Frequent

Percentage (%)

Parents Friends Lecturers Keep silence A professional counselor

76 226 28 110 17

16.5 49.4 6.3 24.1 3.8

Total

460

100.0

184 OLUFEMI ADENIYI FAWOLE AND EBENEZER BAYODE AGBOOLA

Table 4. Correlation Coefficient of Parental Violence and Dating Violence of Male Students in Intimate Relationships. Variable Parental conflict Dating violence

N

Mean

SD

df

Calculated r-Value

Critical r-Value

235 235

53.76 63.86

5.62 8.06

233

0.90*

0.14

* Significant p < 0.05.

DISCUSSION The findings of the hypothesis for this study showed that there is a significant relationship between parental violence and dating violence of male students in intimate relationships in University of Ilorin. This goes in-line with the studies of Mapayi et al. (2011), and Odujirin (1993) on spousal violence in Nigeria which showed that with the theory of intergenerational transmission of violence in play, courtship or dating violence becomes an important matter of concern in Nigeria. Exposure to violence during childhood is also one of the most commonly studied correlates of dating violence. Boladale et al. (2013) also found that dating violence could be linked to the perpetuation of spousal violence which also creates an environment for continued dating violence as children experience and witness violence at home and thus exposing the children to tendency to develop violence in their future relationships. In their own study, Kinsfogel and Grych (2004) found that parental conflicts, reflecting a combination of verbal and physical aggression, have been found to be related to dating aggression among young men.

LIMITATIONS OF THE STUDY 1. It was conducted in only one University in Nigeria, out of over 50 universities in the country. 2. The sample was limited to only 460 participants out of over 20,000 students in total population. 3. The study attempted to measure variables based only from the personal experiences of the respondents, and not involving the parents. 4. As the measures were all self-reported, there were bound to be elements of bias in the responses.

Parental Influence and Dating Violence among Students

185

5. The study was restricted to measuring only the variable of parental conflict, and leasing aside other possible contributing factors. However, it is expected that a study of this sort will provoke further research studies of this nature using the similar or other approaches in the fields of dating violence, adolescents and the family.

CONCLUSION From all investigation with regard to this study, dating violence is rather a pertinent issue. Particularly with 460 out of 600 respondents randomly selected having experienced violence from their dating partners who themselves have witnessed some sort of violence in their homes. A study of this sort also brings to the limelight the necessity to create opportunities for victims of dating violence to have places of refuge or solitude and proper counseling sessions. Also, parents’ relationship thus has impact on the manner their wards conduct their own relationship, whether it is little or much influence. Moreover, it is interesting to know that adolescence appeared to confide more in friends, themselves, than they do their parents. Given the prevalence of violent behavior, particularly, dating among university students, the study offers a number of recommendations. The school authority should create special centers/shelters or hotlines where student’s can report incidence of violence, particularly, relational violence. In addition, parents ought to create an ambience of positive learning, as they are models to their wards, to do otherwise could be inimical as adolescence can become “so fitted by been fit in an unfit fitness,” which will tell on their worldview, particularly, on relational issues.

REFERENCES Ajuwon, A. J., Funmilayo, O., & Osungbade, K. O. (2011). Experience and penetration of violent behaviors among secondary school students in Ibadan, Nigeria. Sierra Leone Journal of Biomedical Research, 3(1), 27 35. Andrews, J. A., Foster, S. L., Capaldi, D., & Hops, H. (2000). Adolescent and family predictors of physical aggression, communication, and satisfaction in young adult couples: A prospective analysis. Journal of Consulting and Clinical Psychology, 68, 195 208.

186 OLUFEMI ADENIYI FAWOLE AND EBENEZER BAYODE AGBOOLA Bandura, A. (1977). A social learning analysis. Englewood Cliffs, NJ: Prentice Hall. Boladale, M., Adesanmi, A., & Olutayo, A. (2013). Experience of childhood violence and help-seeking behavior of students exposed to dating violence at Obafemi Awolowo university, Ile-Ife. Gender and Behavior, 11(2), 5762 5774. Boladale, M., Yetunde, O., Adesanmi, A., Olutayo, A., & Olanrewaju, I. (2004). Personality profiles and psychopathology among students exposed to dating violence at the Obafemi Awolowo university, Ile-Ife. Journal of Interpersonal Violence, 30(1), 168 190. CDC. (2006). Intimate partner violence-fact sheet. Retrieved from www.cdc.gov/injury. Accessed on November 8, 2014. Cummings, E. M., & Davies, P. T. (2002). Effects of marital conflict on children: Recent advances and emerging themes in process-oriented research. Journal of Child Psychology and Psychiatry, 43, 31 63. Cummings, E. M., Ballard, M., El-Sheik, M., & Lake, M. (1991). Resolution and children’s responses to interadult anger. Developmental Psychology, 27, 462 470. Grych, J. H., & Fincham, F. D. (1990). Marital conflict and children’s adjustment: A cognitive contextual framework. Psychological Bulletin, 108, 267 290. Hickman, L. J., Jaycox, L. H., & Aronoff, J. (2004). Dating violence among adolescents: Prevalence, gender distribution and prevention program effectiveness. Trauma Violence Abuse, 5(2), 123 142. Iconis, R. (2013). Dating violence among college students. Contemporary Issues in Education Research, 6(1), 111 114. Izugbara, C. O., Emmanuel, J. C., & Peter, O. D. (2008). Women and male partner dating violence in Nigeria. Indian Journal of Gender Studies, 15(3), 461 484. Katz, L. F., & Gottman, J. M. (1993). Patterns of marital conflict predict children’s internalizing and externalizing behaviors. Development Psychology, 29, 940 950. Kerig, K. (1996). Assessing the links between interparental conflict and child adjustment: The conflicts and problem Solving scales. Journal of Family Psychology, 10, 454 473. Kinsfogel, K. M., & Grych, J. H. (2004). Interparental conflict and adolescent dating relationships: Integrating cognitive, emotional, and peer influence. Journal of Family Psychology, 18, 505 515. Lewis, S. F., & Fremouw, W. (2001). Dating violence: A critical review of the literature. Clinical Psychology Review, 21(1), 105 127. Linder, J. R., & Collins, W. A. (2005). Parent and peer predictors and physical aggression and conflict management in romantic relationships in early adulthood. Journal of Family Psychology, 19, 252 262. Mapayi, B., Makanjuola, R., Fatusi, A. O., & Afolabi, O. T. (2011). Socio-demographic factors associated with intimate partner violence. Journal of Gender and Behavior, 9(1), 346 364. Molidor, C., Tolman, R. M., & Kober, J. (2000). Gender and contextual factors in adolescent dating violence. Prevention Research, 7(1), 1 4. Odujirin, O. (1993). Wife battery in Nigeria. International Journal of Gynecology and Obstetrics, 41, 159 164. O’ Leary, K. D., & Slep, A. M. S. (2003). A dyadic longitudinal model of adolescent dating aggression. Journal of Clinical Child and Adolescent Psychology, 32, 314 327.

Parental Influence and Dating Violence among Students

187

Tschann, J. M., Flores, E., Marin, B. Y., Pasch, L. A., Baisch, E. M., & Wibblesman, C. J. (2002). Interparental conflict and risk behaviors among Mexican American adolescents: A cognitive-emotional model. Journal of Abnormal Child Psychology, 30, 373 385. UNICEF. (2001). In A. Hodge (Ed.), Children and women’s rights in Nigeria: A wake-up call situation assessment and analysis. National Population Commission and UNICEF. Zimet, D. M., & Jacob, T. (2001). Influences of marital conflict on child adjustment: Review of theory and research. Clinical Child and Family Psychology Review, 4, 319 335.

This page intentionally left blank

LONG-TERM EFFECTS OF NEGLECT ON VIOLENCE: ARE THEY SPURIOUS OR INDIRECT? Joanne Savage and Amanda Murray ABSTRACT Purpose In the present paper we explore the long-term influence of childhood neglect on violent behavior in the transition to adulthood. In particular, we test whether neglect is spuriously related to violence due to their common association with academic achievement, physical abuse, and general offending. We then ask whether neglect has an indirect effect on violence through its impact on parental attachment, alcohol use, emotional negativity, academic achievement, or staying in school. Methodology/approach We use two waves of data from the National Longitudinal Survey of Adolescent Health (Add Health) and employ both regression models and INDIRECT, a syntax file that allows for the testing of indirect effects using SPSS (Preacher & Hayes, 2008). Findings We find that the long-term association between childhood neglect and violence in the transition to adulthood is robust in models controlling for GPA, physical abuse, and other forms of offending. Neglect did not have an indirect effect on violence through attachment,

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 189 218 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009010

189

190

JOANNE SAVAGE AND AMANDA MURRAY

negative emotionality, or academic achievement but did have indirect effects on violence through its association with staying in school and with alcohol use. Research implications This set of analyses was exploratory in nature. Further research on neglect should be undertaken, using finely tuned measures and research questions. In addition, our findings imply that the association between neglect and later violent behavior may be intertwined with certain dynamics of physical abuse and alcohol use, which should be further studied. Keywords: Violence; neglect; abuse; transition to adulthood

In recent years, a significant amount of attention across disciplines has shifted to the transition to adulthood. This stage of life has many distinct circumstances. For those interested in violence and other forms of antisocial activity, this stage is particularly important because most adolescents desist from delinquent behavior, and the persistence of criminal activity into adulthood signals a more serious problem. Thus, understanding violence in the transition to adulthood would be an advance in developmental and criminological science. One candidate predictor of violence in the transition to adulthood is child neglect. Research on maltreatment has consistently demonstrated that abused children are at a greater risk of violent behavior in childhood and adolescence. Some of the findings about abuse in the etiology of crime indicate that the magnitude of its effect is not trivial. In one study, participants who had been abused or neglected were 38% more likely than controls to have been arrested for a violent crime (Lansford et al., 2007). These participants were also 53% more likely to have been arrested as a juvenile (Lansford et al., 2007). Less is known about the lasting effects of abuse. Many studies treat various forms of abuse together as “maltreatment.” In studies where the authors disaggregate the effects by type of abuse, we find that the association between physical abuse and adult antisociality is consistent, but the evidence on other forms of abuse is more mixed. We make the case here that neglect is important and understudied. Neglect is the most common form of child abuse and accounted for 78.3% of “substantiated maltreatment victims in the United States” in 2012 (Snyder & Merritt, 2014).

Long-Term Effects of Neglect on Violence

191

Additionally, an estimated 70% of deaths related to maltreatment were a result of child neglect (Snyder & Merritt, 2014). Thus, the problem is neither minor nor rare. There are many studies where neglect is combined with other forms of abuse, most notably physical abuse, so associations in those studies cannot be assumed to hold for neglect alone. Among studies that partial out effects of neglect, the vast majority suggest a significant association between neglect and general delinquency (e.g., Grogan-Kaylor & Otis, 2003; Haapasalo & Moilanen, 2004; Maxfield & Widom, 1996; Zingraff, Leiter, Johnsen, & Myers, 1994). Others have reported correlations between neglect and violence (Carter, Stacey, & Shupe, 1988; Chapple, Tyler, & Bersani, 2005; Fang & Corso, 2007; Haapasalo & Moilanen, 2004; Maxfield & Widom, 1996; Mersky & Reynolds, 2007; Straus & Savage, 2005). Importantly, some studies have not found an association between neglect and violence (e.g., Cusick, Havlicek, & Courtney, 2012; Haapasalo & Moilanen, 2004). In a Finnish sample, neglect was not correlated with violence (Haapasalo & Moilanen, 2004). Cusick et al. (2012) used a very conservative model and found that neglect was not associated with violent arrests, but it is unclear which of the many other variables in the model mediated the effect (or if there was no simple correlation at all). Savage and Wozniak (forthcoming) recently reviewed the evidence and report that the body of studies on neglect do not yet show that it has a special relationship with violence, above and beyond its effect on nonviolent offending. In addition, in studies where an effect has been reported, few models include all of the control variables necessary to confirm that the reported association is not spurious. Thus, more research is needed to understand the paths from neglect to violence and the conditions under which the effect is manifested. Further, the fact that most studies of neglect and violence do not control for potential confounds tells us that additional attention, with careful analysis would make an important contribution to the literature. While the intergenerational transmission of aggression thesis would predict that physical abuse might have a special relationship with violence (Savage, Palmer, & Martin, 2014), theories about neglect and research on its effects on children also suggest that neglect might also be specially associated with violence. In this paper we will discuss a series of potential mediating factors and then provide an analysis of Add Health data to test whether childhood neglect is associated with violence in the long term. We have chosen to look at two issues. First we will explore whether neglect is spuriously related to violence due to their common association with “third” factors. We will emphasize three candidates for sources of spuriousness

192

JOANNE SAVAGE AND AMANDA MURRAY

that have not been adequately addressed in the literature thus far: academic achievement, physical abuse, and general offending. We will then explore whether neglect has an indirect effect on violence. Five candidates for nodes in this indirect path are parental attachment, alcohol use, emotional negativity, academic achievement, and staying in school.

NEGLECT AND ACADEMIC ACHIEVEMENT One of the most disturbing findings in the literature on neglect is the significant association between neglect and low IQ (e.g., Jaffee & MaikovichFong, 2011; Kent, 1976; Yates, Dodds, Sroufe, & Egeland, 2003). The potential for maltreatment to have a “profound impact” due to rapid neuronal growth is recognized by many authors (e.g., Lee & Hoaken, 2007). It is now understood that the neuronal system is plastic, with a sensitive period early in life. Research on brain plasticity suggests that development is partly subtractive, starting with an overproduction of synapses and proceeding through “selective preservation” of synapses that are used and the loss of synapses that are not used (Lerner, 1984). Because neglect may result in lack of sensory and intellectual stimulation for children, it may cause enduring effects on intelligence and related cognitive functions such as problem solving. Although physical abuse is usually the object of concern when we think about brain damage, research suggests that, in addition to physical brain injury, uncontrollable stress can impair function in the pre-frontal cortex (Lee & Hoaken, 2007). Maltreated children appear to have problems regulating cortisol (the stress hormone) and studies of abused children have also found decreased metabolism in the orbital frontal gyrus, the amygdala and the pre-frontal cortex, responsible for cognitive and social functioning. In one study, the corpus callosum in abused children was smaller than that in comparison children (Lee & Hoaken, 2007). Although these studies tend to combine victims of various forms of abuse, their findings are consistent with a hypothesis that lack of stimulation in neglected children, or perhaps the stress they endure, can engender permanent changes in brain structure. An important implication is that neglected children have a greater chance of academic problems than other children, a phenomenon commonly seen in the literature. Low academic achievement is a very consistent predictor of violence (e.g., Bellair & McNulty, 2005; Ellickson & McGuigan, 2000; Salts, Lindholm, Goddard, & Duncan, 1995). Academic

Long-Term Effects of Neglect on Violence

193

problems can be not only due to low intelligence, cognitive dysfunction, or learning problems, but also due to social forces related to neglect such as missing school and failure to form bonds to school. Eckenrode, Laird, and Doris (1993) compared neglected, physically abused, and sexually abused children to controls and found that neglected children had the lowest scores on standardized tests and grades. In a study of children in nine Georgia counties, neglected children had very low scores in language (mean = 29) and math (mean = 28) compared to the control group (mean = 67 and 78 respectively) (Wodarski, Kurtz, Gaudin, & Howing, 1990). In that study, a shocking 60% of neglected children had repeated a grade. Because academic achievement has such a strong association with violence, it may be an important source of spuriousness and indirect effects. In this paper we test two hypotheses about the dynamics of neglect, academic achievement, and violence: that the association between neglect and violence is spuriously associated, due to their common association with academic achievement, and that the effect of neglect on violence is indirect, via its impact on academic achievement.

NEGLECT AND PHYSICAL ABUSE Sadly, victims of one type of abuse are frequently victims of other forms of abuse as well. For example, abused children are more likely to be exposed to domestic violence between their parents (e.g., Gelles, 1998; Haskett, Portwood, & Lewis, 2010). A high co-incidence of neglect and physical abuse has been reported in numerous studies as well. For example, in one study of emotionally neglected children, 32% reported being physically abused (Dong et al., 2004). While current research indicates that the causes and risk factors contributing to physical abuse may differ from factors contributing to neglect, Hartley (2002) maintains that, “it is clear that domestic violence does co-occur with child neglect” and these families experience “multiple vulnerabilities” (p. 357). Dong et al. (2004) argue that by ignoring this co-occurrence, it becomes difficult to understand or measure the cumulative influence of multiple forms of abuse and maltreatment. This could lead to the misattribution of some outcomes to the wrong type of abuse. Because physical abuse has been consistently associated with violent behavior in victims (e.g., Barnow, Lucht, & Freyberger, 2001; Benda & Corwyn, 2002; Herrenkohl et al., 2004; Lansford et al., 2007; Rebellon &

194

JOANNE SAVAGE AND AMANDA MURRAY

Van Gundy, 2005; Widom, 1989), it is certainly possible that associations between neglect and violent behavior are due to their common association with physical abuse. Savage and Wozniak (forthcoming) have recently reported that physical abuse has a robust association with violence, even in studies where nonviolent offending is accounted for. Thus controlling for physical abuse, and checking whether neglect affects violence in the absence of physical abuse, may be important for fully understanding the phenomenon.

NEGLECT AND GENERAL OFFENDING The association between neglect and general offending is no longer controversial. Savage et al. (2014) argue that authors who wish to be certain that their findings apply to violence specifically must take several precautions in their analyses to reduce the chances that associations are merely due to general delinquency. Because nonviolent offending and violent offending are highly correlated, it is possible that associations we have seen in the literature between neglect and violent behavior are due to their common association with nonviolent offending behavior.

NEGLECT AND PARENTAL ATTACHMENT Neglected children have displayed difficulties forming selective attachment to caregivers in humans and rhesus monkeys (Sanchez & Pollak, 2009), and insecure bonds to parents have been associated with problem behavior in childhood and adolescence. Unfortunately, there are not enough studies of this issue to form any firm conclusions. Some studies of general maltreatment and physical abuse suggest that their associations with violence survive the imposition of controls for parent death or desertion (Heck & Walsh, 2000), family attachment (Fagan, Hansen, & Jang, 1983), and maternal warmth (Stacks, Oshio, Gerard, & Roe, 2009) for example. Neglect during childhood may also impact the ways in which a child develops attachments with others. Attachment theory postulates that the type and quality of the attachment that a child develops with the parent is dependent on the degree to which a parent is willing to provide a “secure base” for the child (e.g., Finzi, Ram, Har-Even, Shnit, & Weizman, 2001). This requires emotional availability and responsiveness to the needs of the

Long-Term Effects of Neglect on Violence

195

child. When these needs are not met, the child may develop an insecure attachment style. For this reason, it is clear that neglected children are at a heightened risk of attachment problems. Research has demonstrated that early attachment security tends to be stable across childhood (e.g., Ammaniti, Speranza, & Fedele, 2005) and, in the past, many authors emphasized (and came to exaggerated conclusions about) the enduring effects of infant attachment on later relationships in life. Research does suggest modest correlations between attachment to parents and sibling relationships and friendships in childhood (e.g., Berlin, Cassidy, & Appleyard, 2008) but associations between attachment and distal relationships later in life are likely to be attenuated (Thompson, 2008) and perhaps indirect (e.g., Berlin et al., 2008). Recent writings by Belsky (2005) and others are insistent that the stability of attachment relations over time depends on childrearing circumstances, such as maternal sensitivity, which is likely to remain fairly consistent over time (e.g., Thompson, 2008). This means that neglected children remain at an increased risk for poor attachment and low bonds to parents over time. Savage (2014) reviews the reasoning for the association between attachment and physically aggressive or violent behavior, emphasizing empathy development, attributional style, physiological underarousal, peer rejection, negative emotionality, and emotion regulation. According to Finzi et al. (2001) avoidant behavior develops in insecurely attached children in order to decrease feelings of fear, anxiety, and anger that the child feels due to continuous maternal rejection and unresolved needs. Avoidant behavior eventually may lead to “later emotional insulation, lack of empathy, and antisocial and aggressive behavior” (Finzi et al., 2001, pp. 771 772). These behaviors could become the root of violent thinking and behavior. An individual who feels emotionally isolated and lacks empathy has fewer barriers preventing violence. Thus, if neglect disrupts the attachment process, it could cause violent behavior through this indirect path. Relatedly, attachment to parents is important for informal social controls. The bonds a child feels toward his or her family, friends, or community serve as primary sources of social control (Snyder & Merritt, 2014). We might speculate that neglect can cause these bonds to deteriorate at rates determined by the severity of abuse. Without having both short-term and long-term consequences for unacceptable behavior, children are unable to learn which behaviors should be reinforced. This may lead to both nonviolent and violent antisocial behavior (Snyder & Merritt, 2014). It is for these reasons that we test whether neglect has an indirect effect on violence through attachment.

196

JOANNE SAVAGE AND AMANDA MURRAY

NEGLECT AND ALCOHOL ABUSE Neglect may also operate on violence through its influence on alcohol abuse. Adults who report experiences of neglect as children, compared to those who do not, “also report considerably higher rates of virtually every type of psychopathy including depression, anxiety, drug and alcohol disorders, and generalized distress” (Horwitz, Widom, McLaughlin, & White, 2001). Alcohol abuse is a consistent predictor of violence (e.g., Boles & Miotto, 2003; MacCoun, Kilmer, & Reuter, 2003; Parker & Auerhahn, 1998). Few studies have tested the mediation hypothesis, and these have not looked specifically at neglect. Widom and her colleagues found mixed evidence in their sample (White & Widom, 2003; Widom, Schuck, & White, 2006). Horwitz et al. (2001) found that abused and neglected females, but not males, were more likely to have higher rates of alcohol symptoms or diagnoses when compared to controls (Horwitz et al., 2001). We believe that more attention should be paid to this very compelling hypothesis. Our findings show that the dynamics of neglect, physical abuse, and alcohol may be complex; see below.

NEGATIVE EMOTIONALITY Another indirect route between neglect and violence might be through negative emotionality. Neglect is likely to affect emotional systems and skills in children. Sanchez and Pollak (2009) point to the plasticity of the central nervous system and alterations to brain maturation due to abuse and the likelihood that these may have long-term effects on emotion processing. In socially deprived monkeys, researchers have seen decreased white matter volume, and alterations in the development of receptors related to fearful and anxious behavior (Sanchez & Pollak, 2009). Researchers have also seen differences with non-abused children in how abused children perceive emotion in faces, attend to emotional cues and learn intuitions about outcomes from positive situations (Sanchez & Pollak, 2009). Studies suggest that abuse engenders negative emotionality as manifested in tantrums, chronic crying, enuresis and hypervigilance, emotional lability, depression, and anxiety (e.g., Conaway & Hansen, 1989; Egeland, Sroufe, & Erickson, 1983; E´thier, Lemelin, & Lacharite´, 2004; Kent, 1976; Loos & Alexander, 1997; Maschi, Bradley, & Morgen, 2008). Abuse has also been associated with the child’s anger (Brezina, 1998;

Long-Term Effects of Neglect on Violence

197

Hollist, Hughes, & Schaible, 2009; Kent, 1976) and Dutton (1999) draws a link between traumatic stress in early childhood and later “intimate rage.” Anger has been an important component in theories of physical aggression and violence. In the field of criminology, anger is central to general strain theory as well (Agnew, 1992). Most of these studies, however, have not distinguished between types of abuse so the association between neglect and negative emotionality is not yet clear and no firm conclusions can currently be drawn about the effect of neglect on anger. There is also no consensus regarding the mediating role of negative emotions in the path between abuse and later violence.

BEING IN SCHOOL In the transition to adulthood, there are many lifestyle characteristics that take on new significance for their impact on antisocial behavior. One of these is remaining in school. Although there is not a great deal of research on this issue, apart from overall academic attainment, theory suggests that being a student would be associated with lower violent behavior and existing research is somewhat consistent with this expectation. Findings on overall academic attainment are mixed in statistical significance but consistently suggest that greater academic attainment is associated with reduced violent behavior. Some studies have reported statistically significant associations between overall education and violence (e.g., Jang & Franzen, 2013; Lansford et al., 2007) and dropping out of high school and violence (e.g., Doherty & Ensminger, 2013; Pollock, Mullings, & Crouch, 2006). While rates of general delinquency have a tendency to decrease as individuals age and become young adults, one study suggests that this transition takes place more slowly for individuals who had dropped out of school (Henry, Caspi, Moffitt, Harrington, & Silva, 1999). Additional studies have demonstrated that the relationship between dropping out and eventual delinquent behavior was strengthened when disliking school prompted the drop out (Henry et al., 1999). For young adults, not being in school may have a dual effect on delinquency (Henry et al., 1999). Some authors have found that those who left school displayed reduced delinquency, and speculated that this was due to being separated from a criminogenic context. Most, however, expect that leaving school is likely to be correlated with increased delinquency. Leaving school may mean early entry into the work force and its attendant stresses.

198

JOANNE SAVAGE AND AMANDA MURRAY

There is research to suggest that employed adolescents engage in more delinquent behavior than their unemployed counterparts (Weerman, 2010). This pattern has a number of potential explanations, including economic freedom from parents, as well as a potential increase in exposure to delinquent peers in the work environment. There may also exist a correlation between a lack of motivation or interest in education and the propensity to seek employment (Weerman, 2010). School may be protective against antisocial behavior due to positive associations with other students and optimism for the future. Most of the reasons that being in school might be associated with reduced violence are likely due to a spurious association with factors such as intelligence, poverty, and previous academic problems. Here, however, we imagine that neglect has many far-reaching effects on academic life and if neglect in childhood decreases the chances that young adults are in school, it may have indirect effects on violence through this path.

METHOD Data We analyzed data from the National Longitudinal Study of Adolescent Health (Add Health). Collection of Add Health data began with subjects in grades 7 through 12 in the United States during the 1994 1995 school year (Wave 1) (Harris & Udry, 1994 2000). The Add Health cohort has been followed over time using in-home interviews as well as contextual data on families, neighborhoods, schools, communities, and peers. In the present analysis, we relied on data from Wave 3 self-report and Wave 1 self-report and parent-report surveys. In Wave 3, most subjects were between 19- and 25-years-old, thus this is the best wave to use for outcomes related to the “transition to adulthood.” One advantage of using Add Health data is that the sample comes from the general population, so the findings are not vulnerable to problems found in studies of clinical populations such as small sample sizes, selection effects, and the influence of interventions. Measures Dependent variables The Add Health data set has been used to study violent and delinquent behavior in many published studies and we model our measures of violent

199

Long-Term Effects of Neglect on Violence

and nonviolent offending after them. Subjects in the Add Health sample were asked to report how often they had committed a series of violent acts within the past 12 months (see Table 1). Because the responses were categorized as “never, once, more than once,” or “never, 1 2 times, 3 4 times, 5 + times” the scale does not estimate the total number of acts, rather the resulting additive scale reflects both variety and frequency. Frequency of violent offending was computed by summing responses to items indicative of violent behavior (see Table 1). The indicators for this scale are “formative” and therefore the additive computation was used (see, e.g., Brown, 2006). For formative scales, where the causal flow is from the indicators to the construct, no Cronbach’s alpha is reported. Independent Variable Neglect. In Wave 3 of the Add Health survey, retrospective questions about potentially abusive experiences before the subjects were in sixth grade were included in the questionnaire. To measure neglect we combined two items that indicate caregiver neglect (see Table 2). The response categories were coded: 0 = “This has never happened,” 1 = one time, 2 = two times, 3 = three to four times, 4 = six to ten times, 5 = more than 10 times, so the sum truncates the actual frequency but a high number still reflects more frequent abuse. Because each item suggests a different type of neglect, it is appropriate to add them together, as higher values indicate greater levels of neglect. Most subjects do not report any hitting, neglect, or sexual abuse, however approximately 27% of subjects overall (males and females) report being hit at least once, and more than 40% report some neglect. Table 1.

Violent and Nonviolent Offending Items.

Nonviolent Offending Damage property Steal something worth > $50 Go into a house or building to steal Steal something worth < $50 Buy, sell, or hold stolen property Use someone else’s credit card, bank card, or ATM card without permission

Violent Offending Use or threaten to use weapon to get something from someone Have a physical fight with a group of friends Use a weapon in a fight How many fights were you injured needed doctor/nurse care How often did you hurt someone badly enough they required doctor/nurse care?

200

JOANNE SAVAGE AND AMANDA MURRAY

Table 2.

Items Used to Measure Child Abuse Victimization in Add Health, Wave 3.

By the time you started 6th grade, Neglect How often had your parents or other adult caregivers left you home alone when an adult should have been with you? How often had your parents or other adult caregivers not taken care of your basic needs, such as keeping you clean or providing food or clothing? Physical discipline How often had your parents or other caregivers slapped, hit, or kicked you? Sexual abuse How often had one of your parents or other adult caregivers touched you in a sexual way, forced you to touch him or her in a sexual way, or forced you to have sexual relations?

Control Variables Demographic Characteristics. Demographic characteristics include age and gender, coded 1 = male, 2 = female (as originally done by the Add Health investigators). To code for race/ethnicity, we examined the associations between measures of violence and being Hispanic, Black, American Indian, or Asian, and found that a dummy code indicating whether or not the subject was from a disadvantaged minority (Hispanic, Black, or American Indian) performed best. Income. Income was measured as self-reported total household income before taxes reported in Wave 3. We use the natural log to correct for heteroscedasticity in this variable. In Wave 1, when the subjects were all much younger and living with their parents, family income was reported by parents in dollar amounts in the Add Health survey. Independent. The subjects in this age group lived in a variety of residential arrangements ranging from home with their parents to living with spouses and their own children. We were most concerned that the effectiveness of the income control variable would be attenuated by the fact that some subjects lived at home with parents and some lived on their own. Therefore, we added a control variable to distinguish these groups. We created a dummy code that indicates if the subject was living independently (i.e., in his or her “own place” or a group situation away from parents; coded 1) or with parents (coded 0). Married. The variable is coded 1 if the participant was married and living together with his or her spouse in Wave 3, and 0 otherwise.

Long-Term Effects of Neglect on Violence

201

In School. We also controlled for whether or not the participant was in school in Wave 3, coded 1 if so, and 0 otherwise. Current Employment. Due to its centrality in life-course theories of criminal behavior, and in its relevance for the economic life and routine activities of individuals in the transition to adulthood, we included a control variable for current employment. In Wave 3 of the Add Health survey, subjects were asked if they had worked for pay, and whether they had worked continuously since the beginning of the year. We created a dummy code indicating whether or not the subject reporting working continuously since the beginning of the year or not. Parent Education. Parent education has been consistently associated with violence in children in many studies. In Wave 1, parent respondents reported their own education level (highest grade completed). Community Disorder. Community disorder has been associated with violence in many studies. Here we include it as a control variable because of its likely association with neglect, and other constructs of interest such as staying in school (see nuances on this relationship by Harding, 2009). The Add Health data set had several items related to community disorder, most of which were rated by parents in Wave 1 on a scale of 1 = “No problem,” 2 = “Small problem,” and 3 = “Big problem,” and two items rated only as “Yes or No.” We dummy coded each item such that “1” reflected at least a small problem, and 0 otherwise. The items indicated whether or not the parent respondent knew the neighbors (reverse coded), whether or not the neighborhood was safe (reverse coded), and if trash or drugs were a problem in the neighborhood. We added these together to create a perceived community disorder scale (0 = no problems to 4 = all four problems); a higher score indicates greater community disorder. Physical Abuse. To measure physical abuse, we used the following item: “By the time you started 6th grade, how often had your parents or other care-givers slapped, hit, or kicked you?” The response categories were coded: 0 = “This has never happened,” −1 = one time, 2 = two times, 3 = three to four times, 4 = six to ten times, 5 = more than 10 times, so the variable truncates the actual frequency but a high number still reflects more frequent abuse.

202

JOANNE SAVAGE AND AMANDA MURRAY

Alcohol Use. Subjects were asked in Waves 1 and 3, “During the past 12 months, on how many days did you drink alcohol?” The responses varied on a 6-point scale from “never” (0) to every day/almost every day (6). Drug Use. In Wave 3 of Add Health, the authors ask about drug use. We used a dummy code to indicate whether the participant had used marijuana, cocaine, or other illegal drugs in the past year. Parental Attachment. We created this scale by adding together adolescent respondents’ answers to two items for each parent in Wave 1: “How close do you feel to your Mother/Father?” and “How much do you think s/he cares about you?” These items are highly correlated (r = .84). These were rated on a 5-point scale so the maximum score was 20 points. In Wave 3, participants were not asked about how much their parents cared about them. In Wave 3, the question about closeness was asked about the current residential, previous residential, and biological mother and then father. We used only one score per parent, prioritizing them in the order given such that if there was no score for a current residential mother, we used previous residential mother, and so on. The score for closeness to the mother was added to the score for closeness to the father, with a maximum total of 10 points. Academic Achievement. Academic achievement may be a source of spuriousness and may also be an indirect route between neglect and violence. Here we use self-reported grades in school from Wave 1 when the subjects were aged 12 17. At that time, subjects reported their most recent grades in math, science, English, and social studies and we computed an estimate of their grade point average (GPA); a higher number reflects better grades. Negative Emotionality. There were many items in the Add Health questionnaire that could be interpreted as indicators of negative emotionality. In order to reduce the chances of redundancy with other constructs (such as delinquency or attachment), we used theory to drive our choice of items. The final scale that we used was taken in Wave 1, and included six items asking the participant the extent to which s/he has felt moody, hopeful about the future, depressed, happy, that people dislike him or her and how often he or she has enjoyed life (note: moodiness is measured over 12 months, the others just the past week). The scale for moodiness ranges from 0 to 4, never to every day, and the scale for the other items ranges from 0 to 3, never to most of the time. Of course, the various moods are

Long-Term Effects of Neglect on Violence

203

not generally highly correlated with each other, but it is reasoned that individuals who experience less happiness, more depression and moodiness have greater “negative emotionality” than those who report high frequency of just one of these indicators. Thus, an additive scale was used, the happy/ optimistic items were reverse scored and the responses were then summed so that a higher score is indicative of more “negative emotionality.” Frequency of Nonviolent Offending. This was computed by summing selfreported frequency ratings for a series of items related to damaging property, stealing, selling drugs, etc. (see Table 1). As with the violence measure, this indicator dually reflects variety and frequency of nonviolent offending.

ANALYTIC APPROACH We created an analytic sample by selecting cases that had no missing values for the main variables of interest and variables that could not be substituted (frequency of violence, neglect, minority status, and sex). There were 4,442 such cases. Two sets of analysis were originally planned. First, we simply test the robustness of the association between neglect and violence using a multivariate regression model. Because the violence variable is highly skewed, we converted it to its natural log to improve normality (anchoring its value at 1, rather than 0, as is necessary and seen as best practices, Osborne, 2013, p. 174). However, because interpreting coefficients related to a natural log is not easy for the reader, and the conclusions (if not the coefficient sizes) are essentially identical when the untransformed variable is used, we present the coefficients for the analyses using the untransformed variable in our tables, as others have done. We developed a base model, including all control variables first. We eliminated variables that detracted from statistical power. We began with controls for age, minority status, sex, childhood income (Wave 1), community disorder, intact family (Wave 1), (Wave 3) income (natural log), parent education, drug use (Wave 3), and whether the subject was in school, married and living together with spouse, currently employed (working more than 10 hours per week), or living independently. In this process, the variables which explain the least amount of squared error are removed one at a time until all variables in the model have an absolute t value of 1 or more (this is the level at which statistical power for the remaining partial coefficient tests is improved). It is important to emphasize that we used this method only to

204

JOANNE SAVAGE AND AMANDA MURRAY

develop the base control model, not to test our coefficients. The following variables were removed from our set of control variables because their presence in the model detracted from statistical power: living independently, intact family (Wave 1), community disorder (Wave 1), and family income (Wave 1). See the first column in Table 4 for the base model. Then we ran regressions adding potential confounds one by one. We allowed for mean substitution for missing cases. Then we tested indirect effects using INDIRECT, a publicly available syntax file that allows for the testing of indirect effects using SPSS (Preacher & Hayes, 2008a, 2008b). The module uses bootstrap confidence intervals for inference testing (set to 5,000 resamples) (Hayes, 2013). In these indirect models, parallel or serial multiple mediation models are estimated and control variables are accounted for as in Ordinary Least Squares regression, but an estimate is made of the indirect paths between the independent variable and the dependent variable through the mediators. Hayes (2013) has shown that this procedure results in very similar estimates to those achieved using Structural Equations Modeling (SEM) software and argues that the findings are essentially identical. We wish to comment on language. The term “mediate” has been used in two different ways in the literature. First, it has been used in traditional regression to describe the influence of a third factor on the association between x and y. If z fully “mediates” the effect of x on y, this means that x is no longer significant in a regression model where z is included and this is taken as a sign of spuriousness (x is only related to y because of their common association with z). More recently, the term “mediation” has been used to reflect the operation of a third factor whereby the independent variable of interest, x, affects y in part because it affects z (which affects y). In Hayes’ indirect models (Hayes, 2013), the analyst selects an independent variable, a dependent variable, and “mediators” to test indirect effects of x on y through the proposed mediators. In this paper, we will attempt to avoid ambiguity by using the term “confounding factors” rather than “mediators” for the former case.

FINDINGS Table 3 displays the descriptive data for the variables in our analysis. Correlations between primary measures of interest are displayed in Table 4. Self-report neglect was significantly, positively correlated with frequency of

205

Long-Term Effects of Neglect on Violence

Table 3. Descriptive Information. Variable Dependent variable Violence, Wave 3 Independent variable Neglect Mediators Attachment (Wave 1) Attachment (Wave 3) Alcohol use (Wave 1) Alcohol use (Wave 3) Negative emotionality GPA (Wave 1) Control variables, continuous Age (Wave 3) Income (Wave 3) Parent education Nonviolent offending (Wave 3) Physical abuse

Mean

Median

Standard Deviation

Minimum

Maximum

.267

.000

.938

.00

13.0

1.29

.00

2.02

.00

10.0

15.2 7.88 1.08 2.21 5.04 2.81

18.0 8.0 .00 2.00 5.00 2.75

5.08 2.20 1.47 1.75 3.16 .772

.00 .00 .00 .00 .00 1.00

20.0 10.0 6.0 6.0 19.0 4.00

21.8 $12,627 5.55 .353 73

22.0 $10,000 6.0 .00 0

1.81 $15,702 2.35 1.17 1.44

18 $0 1.00 0 0

28 $250,000 9.00 18.0 5

Binary Variables Minority Sex In school (Wave 3) Current job (Wave 3) Drug use (Wave 3) Married, living together (Wave 3)

Frequencies n = 1,605 n = 2,144 38.2% 69% 33.8% 15.7%

Yes Male Yes Yes Yes Yes

n = 2,837 n = 2,297 61.7% 31% 65.3% 84.3%

No Female No No No No

violence in Wave 3 (r = .083). The correlation was not significant for violent behavior in Wave 2 or Wave 1 (see Table 4). This is surprising because, in general, more proximate measures of related factors usually have stronger correlations than more distal ones. Thus, this finding begs the question: is there a delayed or indirect effect of neglect on later violence, or is the correlation between neglect and violence due to the retrospective method (subjects reported abuse experiences in Wave 3 and their current relationship with their parents might have influenced their ratings of abuse in the distant past)? Another possibility is that child and adolescent reporting of parental warmth and attachment to parents is not as candid as the reports we get once the participants are in the transition to adulthood. We will return to this problem below. Using OLS regression we then estimated the association between neglect and violence in Wave 3, controlling for the base model variables and each

206

Table 4.

JOANNE SAVAGE AND AMANDA MURRAY

Correlations between Neglect, Violence, and Possible Mediators of the Neglect Violence Relationship. Neglect

Neglect Age Minority Community disorder (Wave 1) Sex (1 = Male, 2 = Female) Income (Wave 1) Income (Wave 3) Parent education (Wave 1) In school (Wave 3) Married and living with spouse (Wave 3) Currently employed (Wave 3) GPA (Wave 1) GPA Wave 2) Physical abuse Nonviolent offending (Wave 1) Nonviolent offending (Wave 3) Parental attachment (Wave 1) Parental attachment (Wave 3) Alcohol use (Wave 1) Alcohol use (Wave 3) Substance use (Wave 3) Negative emotionality (Wave 1)

Wave 1 Violence

Wave 2 Violence

Wave 3 Violence

.034* .041** .013 .041** .012 .008 .002 .045** .003

.010 .004 .035* .085** .008 .070** .008 .126** .046** .021

.004 .001 .005 .006 .002 .007 .027 + .019 .002 .027

.082** .076** .089** .006 .199** .020 .040** .017 .064** .079**

.025 + .006 .002 .296** .008 .100** .008 .173** .006 .035* .051** .014

.001 .260** .243** .004 .454** .017 .153** .025 + .232** .021 .013 .192**

.002 .007 .023 .006 .019 .003 .007 .017 .020 .004 .023 .041**

.053** .012 007 073** .029* .325** .022 .018 .017 .101** .137** .018

+p ≤ .10; *p ≤ .05; **p ≤ .01.

of the potential confounds, entered one at a time. The findings are displayed in Table 5. Neglect remained positively and significantly associated with violence in a multivariate analysis, including standard controls for age, minority status, and sex and in a model including other potentially important controls (parent education, past year drug use, and family income) and situational variables for those in the transition to adulthood: in school and whether the subject was currently employed, or married and living together with spouse (see Table 5). Testing Candidate Confounding Factors Table 5 also displays the findings for the tests of spuriousness for academic achievement, physical abuse, and nonviolent offending. In our

207

Long-Term Effects of Neglect on Violence

Table 5. The Influence of Potential Confounds on the Association between Neglect and Later Violent Behavior (Standardized Beta Coefficients Are Displayed). Base Model Age Disadvantaged minority Sex Ln(Income) Parent education Past year drug use In school (Wave 3) Currently employed Married and living with spouse Neglect GPA (Wave 1) Physical abuse Nonviolent offending (Wave 3) N R2

Neglect

GPA (Wave 1)

Physical Abuse

Nonviolent Offending (Wave 3)

−.087** −.084** .088** .085**

−.084** .085**

−.086** .084**

−.067** .077**

−.188** −.024 + −.019 .107** −.085** −.050** −.035*

−.186** −.024 + −.019 .104** −.082** −.052** −.036*

−.186** −.024 + −.017 .104** −.082** −.052** −.036**

−.186** −.026 + −.019 .100* −.084** −.054** −.036**

−.151** −.026 + −.015 .052** −.086** −.039** −.029*

.061**

.061** −.011

.047**

.038**

.048** .275** 4,442 .078

4,442 .082

4,442 .082

4,442 .084

4,442 .152

+p ≤ .10; *p ≤ .05; **p ≤ .01.

multivariate model, GPA was not associated with violent behavior and neglect remained significantly associated with Wave 3 violence when GPA was controlled. Physical abuse was significantly associated with violence, and though its inclusion attenuated the effect of neglect somewhat, the effect of neglect on violence was still statistically significant. Unexpectedly, nonviolent offending in Wave 1 was not associated with violence in Wave 3 (not displayed), but nonviolent offending in Wave 3 was significantly associated with violent behavior in Wave 3, and it partially mediated the association between neglect and violence (which was still statistically significant). Our findings suggest that the long-term association between neglect and violence is robust with respect to sources of potential spuriousness tested here. An early reviewer raised concerns about mean substitution. Some variables in the Add Health data set have many fewer cases than others. We re-ran all the regression models displayed in Table 5, without mean substitution. Neglect was still significantly associated with violent behavior in Wave 3, except in the final model where we controlled for frequency of

208

JOANNE SAVAGE AND AMANDA MURRAY

nonviolence. In that model, the association was still positive, but barely failed to achieve statistical significance (p = .057). In the analysis using the natural log of violent behavior in Wave 3 as the dependent variable, the association between neglect and violence in this model was statistically significant (p = .007), likely due to the improvement in statistical power when dependent variables with smaller variance (such as the natural log transformation) are used.

Indirect Effects We tested the indirect effect of neglect on violence in Wave 3 through attachment to parents (Wave 1 and Wave 3), alcohol use (Wave 1 and Wave 3), negative emotionality, grades and staying in school in the transition to adulthood. The findings are displayed in Table 6. Parental attachment in Wave 1 was not a mediator of the association between neglect and violence. While neglect was negatively associated with the participants’ assessments of attachment to parents in Wave 3, that Table 6.

Summary of Indirect Effects of Neglect on Wave 3 Violence. Path Coefficients Are Displayed.

Neglect → Parental attachment (Wave 1) → Violence Neglect → Parental attachment (Wave 3) → Violence Neglect → Alcohol use (Wave 1) → Violence Neglect → Alcohol use (Wave 3) → Violence Neglect → Negative emotionality (Wave 1) → Violence Neglect → GPA (Wave 1) → Violence Neglect → Staying in school (Wave 3) → Violence

Neglect to Mediator

Mediator to DV

Neglect to DV

Total Effect of Neglect

R2

−.016

.002

.019*

.019*

.084**

−.194**

.004

.021**

.020**

.084**

−.001

.007

.019*

.019*

.084**

.005

.037**

.020**

.020**

.087**

.020

−.001

.020**

.020**

.082**

−.001

−.011

.020**

.020**

.083**

−.001**

−.167**

.020**

.022**

.083**

Note: All models control for age, minority status, sex, Wave 3 income, parent education, Wave 3 drug use, current employment, and married and living together. All models except for the indirect effect model through “in school” Wave 3 also include “in school” as a control. *p ≤ .05; **p ≤ .01.

Long-Term Effects of Neglect on Violence

209

attachment was not associated with violence in Wave 3 so that the total effect of neglect was equal to its direct effect on violence in this model. Wave 1 alcohol use was not a mediator of the association between neglect and violence, either. Many of the subjects in Wave 1 were still quite young to drink alcohol. Alcohol use in Wave 3, however, partially mediated the effect of neglect on violence. Childhood neglect had a marginal long-term association with alcohol use in Wave 3 and both were significantly associated with violence in Wave 3. The total effect of neglect was only slightly larger than the direct effect after adding in this indirect path. Negative emotionality was not a mediator of the association between neglect and violence and neither was academic achievement. Staying in school, however, was negatively associated with neglect and also negatively associated with violence. Thus part of the association between neglect and violence was mediated through the association between neglect in childhood and staying in school in young adulthood. This effect was not large, however, and the direct effect of neglect on violence remained robust and comprises the majority of its effect on violence.

Follow-Up Analyses We had two important remaining questions. First, we wish to return to the surprising finding that neglect was associated with (less proximate) Wave 3 violence and not more proximate Wave 1 and Wave 2 violence. Such a finding could be due to a methodological problem that because subjects are reporting abuse retrospectively, the association between violence and ratings of abuse is inflated due to current relations with their parents. We cannot adjudicate this issue completely, but we explored the data to try to elucidate the problem. Reasoning that if Wave 3 relationships with parents could influence retrospective reports of abuse, we looked at the correlation between Wave 3 ratings of warmth and attachment and reports of neglect. We found that these were correlated: participants who currently reported high attachment to parents were less likely to report neglect in their childhood. This is consistent with two explanations: either the current attitude influenced the report of neglect, or that childhood neglect influenced current attachment to parents. In order for Wave 3 attitudes about parents to confound the association between the retrospective measure of neglect and the current indicator of violence, those attitudes must also be associated with Wave 3 violence. They were not. The correlation between Wave 3

210

JOANNE SAVAGE AND AMANDA MURRAY

attachment and Wave 3 violent behavior was only r = −.018 and was not statistically significant. Although this cannot eliminate the possibility that the retrospective methodology inflated the estimate, it does not support that hypothesis. In addition, we came across an interesting pattern in our analysis of the indirect effects of neglect through alcohol use. The coefficients suggest that childhood neglect and Wave 3 alcohol use are both associated with violence, but neglect has only a marginal effect on alcohol use. When we added physical abuse to the indirect effect model as a control variable, wondering whether the indirect effect was somehow confounded by the mutual associations with physical abuse and alcohol use, we discovered that the association between neglect and alcohol use was attenuated dramatically and the direct effect of neglect on violence was only marginally significant (see Fig. 1). We then ran the model installing physical abuse as the independent variable, alcohol use as the mediator, and neglect as a control variable (see Fig. 2). In this model, physical abuse is strongly associated with alcohol use in Wave 3, which is strongly associated with violence, and physical abuse maintains a significant direct effect on violence. Neglect as a control variable in this model maintains a marginally significant association

Control Variables

Alcohol Use Wave 3 0.00 0.038* 0.014 Neglect

Violence

0.021 Physical Abuse

*p ≤ .05

Fig. 1.

Follow-Up Analysis of the Indirect Effect of Neglect on Violence through Alcohol Use in Wave 3, Controlling for Physical Abuse.

211

Long-Term Effects of Neglect on Violence

Control Variables

0.032+

Alcohol Use Wave 3 0.038** 0.021+

Physical Abuse

Violence

0.014+ Neglect +p ≤ .10; **p ≤ .01

Fig. 2.

Follow-Up Analysis Comparing the Indirect Effect of Physical Abuse on Violence through Alcohol Use in Wave 3, Controlling for Neglect.

with violence. This suggests that perhaps the initial estimates of the indirect effect of neglect on violence, via alcohol use, may be influenced by the dynamics of physical abuse, alcohol use, and violence, though we have introduced this final analysis merely as a follow-up in an already exploratory paper. A more defined test should be conducted before firm conclusions are drawn.

CONCLUSIONS The adverse effects of child neglect on society are myriad, but in this paper we have explored some of the intricacies of the long-term association between neglect in childhood and violence in the transition to adulthood. We found that the association between neglect and violence was robust and withstood the test of numerous important potential confounding factors. In these data, the association between neglect and violence is not due to academic achievement, their common association with physical abuse, or their common association with general offending. We also expected that we would find significant indirect effects of neglect on violence through

212

JOANNE SAVAGE AND AMANDA MURRAY

parental attachment, alcohol use, negative emotionality, academic achievement, and staying in school. We did find small indirect effects of neglect through alcohol use and we also found a significant indirect effect of neglect through staying in school. In all of these analyses, the direct association between neglect and violence remained statistically significant. In a follow-up analysis, we also found that the effect of neglect on violence is entangled in a common association with physical abuse, which exerts a strong effect on alcohol consumption, which, in turn, was significantly associated with violence. In that complex model, neglect was only marginally associated with violence. Because that analysis was run as a follow-up, we do not place too much emphasis on it and we recommend that future researchers explore this possibility in detail. The findings imply that the association between neglect and violence deserves more attention than it has previously received. Although some authors have disaggregated their analysis of abuse effects by type of abuse, almost none have made an attempt to theorize about specific associations between neglect and violence (over nonviolent antisocial behavior). Our findings of a robust effect of neglect on violence suggest that a better elucidation of theory and more empirical tests are warranted. Although we discovered two routes through which neglect affects violent behavior (alcohol use and staying in school), the direct effects of neglect on violence remained important, demanding more propositions about mechanisms through which neglect wields its influence, which are not patently obvious. We might expect that effects on intelligence, executive functioning, problem solving and the like might be good prospects, and the data set used here did not allow us to measure these constructs. In addition, it is possible that certain protective factors play a strong role in the association. Testing interactions in future research would be useful for learning the conditions under which neglect will result in violence rather than the many other potential outcomes. An early reviewer of this manuscript pointed out that the effect sizes and model fit coefficients in our regression analyses are quite small. Although many of the comparisons we report are “statistically significant,” their impact is not large. This may be due, in part, to the somewhat vague operational definition of neglect, which did not distinguish in fine detail levels of neglect including serious abuse. We speculate that better measures of neglect would not only result in larger coefficients, but would certainly be better for application to policy or intervention. We acknowledge limitations to the generalizability of these findings. First, a non-clinical sample has a strong advantage: it eases concerns about

Long-Term Effects of Neglect on Violence

213

sampling bias that are present in smaller studies of abused children compared to non-abused children. However, the types of violence reported by participants in a national probability sample are likely to be less severe than those in a clinical sample, so the findings may not be generalizable to severe offending. Second, retrospective measures of neglect and physical abuse have weaknesses. Many studies suggest that the inherent lapse of time from the traumatic event to the time of recall in cases of retrospective memories can cause difficulties for the individual when trying to maintain accuracy. Additionally, many questions regarding neglect or abuse are sensitive in their nature and “variability in responses may occur due to the … subjects’ knowledge of the ‘social taboos’ of responding to such questions (Dube, Williamson, Thompson, Felitti, & Anda, 2004, p. 730). Unpleasant childhood experiences can also cause severe stressful or anxious feelings that have the potential to impair one’s ability to properly acquire and assimilate information (Dube et al., 2004). Particularly in cases involving impaired psychological functioning as a contributing factor for violent behavior, when individuals are asked to retrospectively recall cases of abuse or neglect, it may be difficult to preserve accuracy and reliability (Lansford et al., 2007). This psychological impairment may exacerbate already apparent memory and recall unreliability (Lansford et al., 2007). Although we tried to assuage our concerns about the measure by controlling for other parent ratings made in Wave 3, it may still be the case that our coefficients are biased due to correlations between retrospective ratings of abuse, current violent behavior, and current animosity toward parents. Our measure of “negative emotionality” did not have the level of face validity we would have liked. We were limited to the Add Health survey items, which were not intended for the purpose of measuring negative emotionality pursuant to abuse. If we were to test this in the future, we would introduce items related to anger and guilt and shame, in particular. The final scale that we used included items about moodiness, depression, hopefulness, enjoying life, and whether the participant believes that others dislike him or her. It is possible that we missed an important indirect effect by inadequately capturing the type of emotional states that link previous neglect with violent behavior. We employed a conservative set of covariates in our models but neglected children are highly likely to be exposed to multiple risk factors (e.g., Widom, 1997). We did not control for cumulative risk, which is also a strong predictor, on its own, of serious antisocial behavior. For example, neighborhoods that are more likely to have families where children are

214

JOANNE SAVAGE AND AMANDA MURRAY

neglected are also more likely to have visible violence that the children may be exposed to on a daily basis (Snyder & Merritt, 2014). Neglecting families are also more likely to endure problems such as poverty, unemployment, parental alcohol and drug problems, parental incarceration, and inadequate family functioning (Horwitz et al., 2001). Additionally, abused or neglected children are more likely to be taken out of the home or placed in foster care, which alone may have implications for future violence. Children who are more severely or chronically abused or neglected are the most likely to be taken out of the home, which may cause correlations to appear stronger than they are in reality (Lansford et al., 2007). As with all studies, research methodology may also impact the reliability of the results obtained.

REFERENCES Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30, 47 87. Ammaniti, M., Speranza, A. M., & Fedele, S. (2005). Attachment in infancy and in early and late childhood: A longitudinal study. In K. A. Kerns & R. A. Richardson (Eds.), Attachment in middle childhood (pp. 115 136). New York, NY: Guilford Press. Barnow, S., Lucht, M., & Freyberger, H.-J. (2001). Influence of punishment, emotional rejection, child abuse, and broken home on aggression in adolescence: An examination of aggressive adolescents in Germany. Psychopathology, 34(4), 167 173. Bellair, P. E., & McNulty, T. L. (2005). Beyond the bell curve: Community disadvantage and the explanation of black-white differences in adolescent violence. Criminology, 43(4), 1135 1168. Belsky, J. (2005). Attachment theory and research in ecological perspective: Insights from the Pennsylvania infant and family development project and the NICHD Study of early child care. In K. E. Grossmann, K. Grossmann, & E. Waters (Eds.), Attachment from infancy to adulthood: The major longitudinal studies (pp. 71 97). New York, NY: Guilford. Benda, B. B., & Corwyn, R. F. (2002). The effect of abuse in childhood and in adolescence on violence among adolescents. Youth and Society, 33(3), 339 365. Berlin, L. J., Cassidy, J., & Appleyard, K. (2008). The influence of early attachments on other relationships. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 333 347). New York, NY: Guilford Press. Boles, S. M., & Miotto, K. (2003). Substance abuse and violence: A review of the literature. Aggression and Violent Behavior, 8, 155 174. Brezina, T. (1998). Adolescent maltreatment and delinquency: The question of intervening processes. Journal of Research in Crime and Delinquency, 35(1), 71 99. Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York: Guilford.

Long-Term Effects of Neglect on Violence

215

Carter, J., Stacey, W. A., & Shupe, A. W. (1988). Male violence against women: Assessment of the generational transfer hypothesis. Deviant Behavior, 9(3), 259 273. Chapple, C. L., Tyler, K. A., & Bersani, B. E. (2005). Child neglect and adolescent violence: Examining the effects of self-control and peer rejection. Violence and Victims, 20(1), 39 53. Conaway, L. P., & Hansen, D. J. (1989). Social behavior of physically abused and neglected children: A critical review. Clinical Psychology Review, 9, 627 652. Cusick, G. R., Havlicek, J. R., & Courtney, M. E. (2012). Risk for arrest: The role of social bonds in protecting foster youth making the transition to adulthood. American Journal of Orthopsychiatry, 82, 19 31. Doherty, E., & Ensminger, M. E. (2013). Marriage and offending among a cohort of disadvantaged African Americans. Journal of Research in Crime and Delinquency, 50(1), 104 131. Dong, M., Anda, R. F., Felitti, V., Dube, S. R., Williamson, D. F., Thompson, T. J., …. Giles, W. H. (2004). The interrelatedness of multiple forms of childhood abuse, neglect, and household dysfunction. Child Abuse & Neglect, 28(7), 771 784. Dube, S. R., Williamson, D. F., Thompson, T., Felitti, V. J., & Anda, R. F. (2004). Assessing the reliability of retrospective reports of adverse childhood experiences among adult HMO members attending a primary care clinic. Child Abuse & Neglect, 28(7), 729 737. Dutton, D. G. (1999). Traumatic origins of intimate rage. Aggression and Violent Behavior, 4(4), 431 447. Eckenrode, J., Laird, M., & Doris, J. (1993). School performance and disciplinary problems among abused and neglected children. Developmental Psychology, 29, 53 62. Egeland, B., Sroufe, L. A., & Erickson, M. F. (1983). The developmental consequences of different patterns of maltreatment. Child Abuse and Neglect, 7, 459 469. Ellickson, P. L., McGuigan, K. A. (2000). Early predictors of adolescent violence. American Journal of Public Health, 90(4), 566 572. E´thier, L. S., Lemelin, J., & Lacharite´, C. (2004). A longitudinal study of the effects of chronic maltreatment on children’s behavioral and emotional problems. Child Abuse & Neglect, 28(12), 1265 1278. Fagan, J., Hansen, K. V., & Jang, M. (1983). Profiles of chronically violent juvenile offenders: An empirical test of an integrated theory of violent delinquency. In J. R. Kluegel (Ed.), Evaluating juvenile justice (pp. 91 119). Beverly Hills, CA: Sage. Fang, X., & Corso, P. S. (2007). Child maltreatment, youth violence, and intimate partner violence: Developmental relationships. American Journal of Preventive Medicine, 33(4), 281 290. Finzi, R., Ram, A., Har-Even, D., Shnit, D., & Weizman, A. (2001). Attachment styles and aggression in physically abused and neglected children. Journal of Youth & Adolescence, 30(6), 769. Gelles, R. (1998). Family violence. In M. Tonry (Ed.), The handbook of crime and punishment (pp. 178 206). New York, NY: Oxford University Press. Grogan-Kaylor, A., & Otis, M. D. (2003). The effect of childhood maltreatment on adult criminality: A Tobit regression analysis. Child Maltreatment, 8(2), 129 137. Haapasalo, J., & Moilanen, J. (2004). Official and self-reported childhood abuse and adult crime of young offenders. Criminal Justice and Behavior, 31(2), 127 149. Harding, D. J. (2009). Collateral consequences of violence in disadvantaged neighborhoods. Social Forces, 88(2), 757 784.

216

JOANNE SAVAGE AND AMANDA MURRAY

Harris, K. M., & Udry, J. R. (1994 2002). National longitudinal study of adolescent health (Add Health), 1994 2002 [Computer file]. ICPSR21600-v2. Ann Arbor, MI: Interuniversity Consortium for Political and Social Research [distributor], 2009-03-11. Hartley, C. C. (2002). The co-occurrence of child maltreatment and domestic violence: Examining both neglect and child physical abuse. Child Maltreatment, 7(4), 349 358. Haskett, M. E., Portwood, S. G., & Lewis, K. M. (2010). Physical child abuse. In C. J. Ferguson (Ed.), Violent crime: Clinical and social implications (pp. 207 228). Los Angeles, CA: Sage. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford. Heck, C., & Walsh, A. (2000). The effects of maltreatment and family structure on minor and serious delinquency. International Journal of Offender Therapy and Comparative Criminology, 44(2), 178 193. Henry, B., Caspi, A., Moffitt, T. E., Harrington, H., & Silva, P. A. (1999). Staying in school protects boys with poor self-regulation in childhood from later crime: A longitudinal study. International Journal of Behavioral Development, 23(4), 1049 1073. Herrenkohl, T. I., Mason, W., Kosterman, R., Lengua, L. J., Hawkins, J., & Abbott, R. D. (2004). Pathways from physical childhood abuse to partner violence in young adulthood. Violence and Victims, 19(2), 123 136. Hollist, D. R., Hughes, L. A., & Schaible, L. M. (2009). Adolescent maltreatment, negative emotion, and delinquency: An assessment of general strain theory and family-based strain. Journal of Criminal Justice, 37(4), 379 387. Horwitz, A. V., Widom, C. S., McLaughlin, J., & White, H. R. (2001). The impact of childhood abuse and neglect on adult mental health: A prospective study. Journal of Health and Social Behavior, 42(2), 184 201. Jaffee, S. R., & Maikovich-Fong, A. K. (2011). Effects of chronic maltreatment and maltreatment timing on children’s behavior and cognitive abilities. Journal of Child Psychology and Psychiatry, 52, 184 194. Jang, S. J., & Franzen, A. B. (2013). Is being “spiritual” enough without being religious? A study of violent and property crimes among emerging adults. Criminology, 51(3), 595 628. Kent, J. T. (1976). A follow-up study of abused children. Journal of Pediatric Psychology, 2, 25 31. Lansford, J. E., Miller-Johnson, S., Berlin, L. J., Dodge, K. A., Bates, J. E., & Pettit, G. S. (2007). Early physical abuse and later violent delinquency: A prospective longitudinal study. Child Maltreatment, 12(3), 233 245. Lee, V., & Hoaken, P. N. (2007). Cognition, emotion, and neurobiological development: Mediating the relation between maltreatment and aggression. Child Maltreatment, 12(3), 281 298. Lerner, R. M. (1984). On the nature of human plasticity. New York, NY: Cambridge University Press. Loos, M. E., & Alexander, P. C. (1997). Differential effects associated with self-reported histories of abuse and neglect in a college sample. Journal of Interpersonal Violence, 12(3), 340 360. MacCoun, R., Kilmer, B., & Reuter, P. (2003). Research on drugs-crime linkages: The next generation. In U.S. Department of Justice (Ed.), Toward a drugs and crime research agenda for the 21st century (pp. 65 96). Washington, DC: Office of Justice Programs.

Long-Term Effects of Neglect on Violence

217

Maschi, T., Bradley, C. A., & Morgen, K. (2008). Unraveling the link between trauma and delinquency: The mediating role of negative affect and delinquent peer exposure. Youth Violence and Juvenile Justice, 6(2), 136 157. Maxfield, M. G., & Widom, C. S. (1996). The cycle of violence revisited 6 years later. Archives of Pediatrics and Adolescent Medicine, 150, 390 395. Mersky, J. P., & Reynolds, A. J. (2007). Child maltreatment and violent delinquency: Disentangling main effects and subgroup effects. Child Maltreatment, 12(3), 246 258. Osborne, J. W. (2013). Best practices in data cleaning. Los Angeles, CA: Sage. Parker, R. N., & Auerhahn, K. (1998). Alcohol, drugs, and violence. Annual Review of Sociology, 24, 291 311. Pollock, J. M., Mullings, J. L., & Crouch, B. M. (2006). Violent women: Findings from the Texas women Inmates Study. Journal of Interpersonal Violence, 21(4), 485 502. Preacher, K. J., & Hayes, A. F. (2008a). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879 891. Preacher, K. J., & Hayes, A. F. (2008b). SPSS macro for multiple mediation: INDIRECT. Written by Andrew F. Hayes, The Ohio State University. Retrieved from http://www. afhayes.com. Accessed on May 1, 2014. Rebellon, C. J., & Van Gundy, K. (2005). Can control theory explain the link between parental physical abuse and delinquency? A longitudinal analysis. Journal of Research in Crime and Delinquency, 42(3), 247 274. Salts, C. J., Lindholm, B. W., Goddard, H. W., & Duncan, S. (1995). Predictive variables of violent behavior in adolescent males. Youth and Society, 26, 377 399. Sanchez, M. M., & Pollak, S. D. (2009). Socioemotional development following early abuse and neglect: Challenges and insights from translational research. In M. De Haan & M. R. Gunnar (Eds.), Handbook of developmental neuroscience (pp. 497 520). New York, NY: Guilford. Savage, J. (2014). The association between attachment, parental bonds and physically aggressive and violent behavior: A comprehensive review. Aggression and Violent Behavior, 19(2), 164 178. Savage, J., Palmer, J. E., & Martin, A. B. (2014). Intergenerational transmission: Physical abuse and violent vs. nonviolent criminal outcomes. Journal of Family Violence, 29(7), 739 748. Savage, J., & Wozniak, K. (forthcoming). Thugs and thieves: The differential etiology of violence. New York, NY: Oxford University Press. Snyder, S. M., & Merritt, D. H. (2014). Do childhood experiences of neglect affect delinquency among child welfare involved-youth? Children & Youth Services Review, 46, 64 71. Stacks, A. M., Oshio, T., Gerard, J., & Roe, J. (2009). The moderating effect of parental warmth on the association between spanking and child aggression: A longitudinal approach. Infant and Child Development, 18, 178 194. Straus, M. A., & Savage, S. A. (2005). Neglectful behavior by parents in the life history of university students in 17 countries and its relation to violence against dating partners. Child Maltreatment, 10(2), 124 135. 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, NY: Guilford Press.

218

JOANNE SAVAGE AND AMANDA MURRAY

Weerman, F. (2010). Delinquency after secondary school: Exploring the consequences of schooling, working and dropout. European Journal of Criminology, 7(5), 339 355. White, H. R., & Widom, C. S. (2003). Intimate partner violence among abused and neglected children in young adulthood: The mediating effects of early aggression, antisocial personality, hostility and alcohol problems. Aggressive Behavior, 29(4), 332 345. Widom, C. S. (1989). The cycle of violence. Science, 244(4901), 160 244. Widom, C. S. (1997). Child abuse, neglect, and witnessing violence. In D. M. Stoff, J. Breiling, & J. D. Maser (Eds.), Handbook of antisocial behavior (pp. 159 170). New York, NY: Wiley. Widom, C. S., Schuck, A. M., & White, H. R. (2006). An examination of pathways from childhood victimization to violence: The role of early aggression and problematic alcohol use. Violence and Victims, 21(6), 675 690. Wodarski, J. S., Kurtz, P., Gaudin, J. M., & Howing, P. T. (1990). Maltreatment and the school-age child: Major academic, socioemotional, and adaptive outcomes. Social Work, 35(6), 506 513. Yates, T. M., Dodds, M. F., Sroufe, L., & Egeland, B. (2003). Exposure to partner violence and child behavior problems: A prospective study controlling for child physical abuse and neglect, child cognitive ability, socioeconomic status, and life stress. Development and Psychopathology, 15(1), 199 218. Zingraff, M. T., Leiter, J., Johnsen, M. C., & Myers, K. A. (1994). The mediating effect of good school performance on the maltreatment-delinquency relationship. Journal of Research in Crime and Delinquency, 31(1), 62 91.

SITUATING THE EXPERIENCE OF MATERNAL INCARCERATION: CHILDHOOD AND YOUNG ADULT CONTEXT Xing Zhang and Allison Dwyer Emory ABSTRACT Purpose We descriptively examined measures of family structure, socioeconomic disadvantage, and exposure to crime, violence, and substance use in young adulthood and childhood for those who experienced maternal incarceration as children. Methodology/Approach We used data from waves I and IV of the National Longitudinal Study of Adolescent to Adult Health. We compared these individuals to two groups: those who did not experience maternal incarceration and those who experienced paternal incarceration. We generated weighted means and conducted F-tests using bivariate regressions to determine where these groups significantly differed. Findings We found that individuals whose mothers were incarcerated during their childhoods experienced greater hardships in both childhood and young adulthood than those whose mothers were not incarcerated.

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 219 254 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009011

219

220

XING ZHANG AND ALLISON DWYER EMORY

Individuals who experienced maternal incarceration reported similar levels of socioeconomic disadvantage and exposure to crime and violence as those who experienced paternal incarceration. One notable exception was family structure, where maternal incarceration was associated with significantly fewer respondents reporting living with their mother or either biological parent. Social Implications With the exception of family structure, the childhood and transition to adulthood were comparable for individuals experiencing any form of parental incarceration. These children were significantly more disadvantaged and exposed to more risk factors than those whose parents were never incarcerated. Additional support and resources are necessary for families who have incarcerated parents, with special outreach made to families without a biological mother in the household. Originality/Value of Paper There has been no overarching, descriptive study comparing child and young adult outcomes of those with an incarcerated mother using a nationally representative, longitudinal dataset in the United States. Keywords: Maternal incarceration; socioeconomic disadvantage; childhood

INTRODUCTION Parental incarceration is a defining life experience for many children in the United States and affected nearly 2.7 million in 2010 (Pew Charitable Trust, 2010). A small but growing proportion of these children experienced maternal incarceration. While the number of parents in state and federal prisons grew by 79% between 1991 and 2007 (Glaze & Maruschak, 2008), maternal incarceration grew by 122% over this same period. Despite the increase in incarcerated mothers, relatively little research has examined maternal incarceration from their children’s perspective. Most of what is known about maternal incarceration comes from studies of incarcerated women, which provide estimates of the prevalence and the context of the women but cannot speak to the experiences of their children. Incarcerated mothers reported more involvement in the care and support of their children prior to arrest than incarcerated fathers (Glaze & Maruschak, 2008). This family engagement suggests that a mother’s incarceration may be particularly disruptive of family life (Arditti, 2012a; Glaze & Maruschak, 2008; Johnson & Waldfogel, 2004).

Situating the Experience of Maternal Incarceration

221

Due in part to the larger number of incarcerated men, much more is known about the implications of paternal incarceration for children. Paternal incarceration is associated with family instability (Braman, 2004; Turney, 2015; Turney & Wildeman, 2013; Western et al., 2004), economic hardship in the form of lost income and direct costs (Arditti, 2012a; Comfort, 2008; Pager, 2003; Pettit & Western, 2004; Western et al., 2001), and increased risk of exposure to crime and violence (Foster & Hagan, 2007; Johnston, 1995). Paternal incarceration also has long-term implications for intergenerational disadvantage (Foster & Hagan, 2007, 2009; Hagan & Foster, 2012; Roettger & Swisher, 2011). Without more information on the experiences of children with incarcerated mothers, it is difficult to determine whether existing research on paternal incarceration can be applied to this population or build a similar body of causal research. To address this gap in the literature and build the needed descriptive foundation, this paper addresses three key research questions. First, what is the childhood context of individuals who experienced maternal incarceration before the age of 18? Second, how does this childhood context conform to or differ from the context of children whose fathers were incarcerated? Third, to what extent are the patterns of differences in childhood continued into young adulthood? This paper used the National Longitudinal Study of Adolescent to Adult Health (Add Health), which contained data on both adolescents and young adults with a sufficient sample size to identify the small population of individuals who experienced maternal incarceration. Childhood context and young adult experiences were studied in three broad areas: family structure, socioeconomic factors, and exposure to crime and violence. Comparisons were made between individuals whose mothers were incarcerated during their childhood and those whose mothers were not, as well as between individuals with incarcerated mothers or fathers. We found that children exposed to maternal incarceration had different family structures, faced socioeconomic disadvantages, and had greater exposure to abuse and violence during childhood. In comparison to those who experienced paternal incarceration, the maternal incarceration sample reported more diverse family structures, disproportionate representation of non-Hispanic blacks, and higher rates of maternal alcoholism. Outside of these areas, the experiences of maternal and paternal incarceration were similar. The lingering differences found in young adulthood suggest that maternal incarceration may play a role in the intergenerational transmission of disadvantage similar to that of paternal incarceration.

222

XING ZHANG AND ALLISON DWYER EMORY

BACKGROUND Incarcerated mothers make up a small but growing proportion of incarcerated parents in the United States. Due to the small number of women who experience incarceration, data on these women and their families has been limited and research on their children has found inconsistent results. Data on incarcerated women mostly come from government statistics on state and federal prisoners and qualitative studies of incarcerated women. A snapshot of state and federal prisoners in 2007 found that the majority of imprisoned women were mothers, and around 40% had multiple children (Glaze & Maruschak, 2008). Mothers in this prison sample were most likely to be non-Hispanic white (50%), followed by non-Hispanic black (28%) and Hispanic (15%). These mothers also reported many risk factors prior to their incarceration that may be relevant for their children such as relatively high levels of homelessness (9%), mental health problems (73%), substance abuse (64%), and physical or sexual abuse (60%). Incarcerated mothers not living with their children were even more likely to report these hardships. While these figures only account for a subset of all everincarcerated women, the snapshot provides a valuable starting point for understanding the experiences of their children. Qualitative work has followed mothers during and after their incarceration, describing factors that influence how the family handled the incarceration (Arditti, 2012a, 2012b; Enos, 2001; Hanlon et al., 2005; Poehlmann, 2005; Siegel, 2011). These studies have found that incarcerated mothers and their families faced stressors like poverty, crime within the family, unstable family structures, substance abuse, and maternal unemployment (Arditti, 2012a; Enos, 2001; Poehlmann, 2005; Siegel, 2011). These qualitative findings were confirmed by research using the Fragile Families and Child Wellbeing study, which found that mothers with a history of incarceration were more likely to have low incomes, a history of substance abuse and domestic violence, and unstable housing (Wildeman & Turney, 2014).

Children’s Experiences of Maternal Incarceration While there are a variety of sources examining the characteristics and experiences of incarcerated mothers, fewer have addressed their children. There has been an ongoing discussion as to whether incarceration has a positive, negative, or null effect on children. Several studies have found positive or null effects of incarceration due to difficulties mothers faced

Situating the Experience of Maternal Incarceration

223

prior to their incarceration like addiction, disruptive or criminal behavior, or other risk factors (Siegel, 2011; Turanovic et al., 2012). On the other hand, maternal incarceration may harm children through the trauma of separation or the direct stress of incarceration (Arditti, 2012a; Siegel, 2011). This perspective was supported by studies that have found increased stress, decreased educational attainment, and family instability among children with incarcerated mothers (Arditti, 2012a; Hagan & Foster, 2012; Johnson & Waldfogel, 2004). The incarceration may be associated with additional instability if children were separated from siblings or if changes in neighborhood and community ties occurred (Harm & Thompson, 1995; LaVigne et al., 2008; Phillips et al., 2006). Wildeman and Turney (2014) provided the most comprehensive quantitative study of the effects of maternal incarceration on children. They found that most of the 21 child outcomes examined were not associated with maternal incarceration after controlling for selection factors. The two exceptions were that teacher reported self-control problems increased in response to maternal incarceration, while rule-breaking behavior decreased (Wildeman & Turney, 2014). Why this experience has such ambiguous effects despite what appears to be a large disruption in children’s lives is less well understood. Establishing the context of children who experience maternal incarceration, and how it differs from other populations, is critical for developing more sophisticated causal models to address this debate. The existing literature suggests that maternal incarceration is most likely to disrupt the family structure of children. In a study comparing the implications of maternal and paternal incarceration on children’s living arrangements, Johnson and Waldfogel (2004) found that children were more likely to live with grandparents or other relatives, or in another arrangement not involving parents, during their mother’s incarceration. This was especially likely for children whose mothers reported more risk factors prior to their incarceration. Children with incarcerated fathers overwhelmingly stayed with their mothers, while only 17% of the children of incarcerated mothers lived with fathers (Johnson & Waldfogel, 2004). The Bureau of Justice Statistics study of parents in prison found similar results in later years. Nearly 65% of women in state prisons reported living with their children prior to their arrest or incarceration, and over 40% reported being a single mother in the month before arrest (Glaze & Maruschak, 2008). Incarcerated mothers reported their child’s caregiver as a grandparent (45%), the child’s father (37%), and other relatives or foster care (42%), with many mothers who sent siblings to different caregivers (Glaze & Maruschak,

224

XING ZHANG AND ALLISON DWYER EMORY

2008). It is unclear from the existing literature how temporary these living arrangements are, how engaged mothers are in these families, or what implications these differences in family structure may have for children’s relationships and resources. Maternal incarceration can disrupt the resources available to children directly or indirectly. Prior to their incarceration, 52% of incarcerated mothers reported providing the primary financial support for their children (Glaze & Maruschak, 2008). In addition to lost income, incarceration can impose direct costs on families through visits, phone calls, and funds sent to the incarcerated person (Arditti et al., 2003; Braman, 2004; Comfort, 2008). Wives, girlfriends, and mothers typically paid these costs for incarcerated men (Braman, 2004; Comfort, 2008), though whether incarcerated women receive the same level of support is unclear. Studies of custodial grandparents, the most common arrangement for children with incarcerated mothers, have shown that these families were at particularly high risk for financial distress (Bachman & Chase-Lansdale, 2005; Brandon, 2005; Dunifon, 2013). Finally, children who experience maternal incarceration may be at higher risk for behavioral problems, involvement in crime, delinquency, and exposure to violence. Mothers who were incarcerated reported high levels of exposure to violence within the home throughout their lives (Arditti, 2012a; Glaze & Maruschak, 2008), which given the high rates of coresidence may also expose their children to domestic violence. These children may also have higher exposure to illegal drug use through their mother’s addictions and relationships with men engaged in the drug market (Arditti, 2012a). During a mother’s incarceration, children may have less supervision to enforce rules or check antisocial behavior (Arditti et al., 2003; LaVigne et al., 2008; Siegel, 2011).

Existing Links between Paternal and Maternal Incarceration A larger body of research has focused on paternal incarceration due in part to data availability. While some of this literature may apply more broadly to parental incarceration, there may also be differences between the implications of maternal and paternal incarceration for children. Perhaps the largest known difference between paternal incarceration and maternal incarceration is family structure. Data on parents in state and federal prison suggests that relative to incarcerated mothers, incarcerated fathers were less likely to have been living with their children before the

Situating the Experience of Maternal Incarceration

225

incarceration and less likely to have been the primary source of daily care (Glaze & Maruschak, 2008). Children also overwhelmingly lived with their mothers during their father’s incarceration, while those with incarcerated mothers were more likely to be living with relatives other than their fathers (Glaze & Maruschak, 2008; Johnson & Waldfogel, 2004). Paternal incarceration has been associated with economic distress for families ranging from the short-term loss of income and direct costs of incarceration for families (Arditti et al., 2003; Pager, 2003; Pettit & Western, 2004; Western, 2001) to disadvantages in areas like school readiness that carry long-term implications for children’s own economic future (Haskins, 2014). Imprisoned fathers and mothers were equally likely to report being their children’s primary source of financial support, though mothers who had lived with their children were most likely to be that primary source (Glaze & Maruschak, 2008). Previous research has suggested that experiencing a father’s incarceration can increase children’s exposure to violence and crime. Within the household, recent research has suggested that a father’s incarceration increased mothers’ physical aggression toward children (Turney, 2014). The experience of incarceration can also encourage men to participate in violent behavior or adopt tougher identities that make reentering into families more difficult (Arditti, 2012a; Nurse, 2002). Housing instability induced by paternal incarceration (Geller & Franklin, 2014) may force families to move to neighborhoods with more crime in response to changed financial circumstances (Braman, 2004). Whether maternal incarceration similarly elevates the risk of exposure to violence is unknown, and the differences in family structure between maternal and paternal incarceration may even exacerbate this risk. Paternal incarceration has also been linked to children’s own participation in crime, violence, and risky behaviors. Children and adolescents whose fathers were incarcerated reported elevated aggression, delinquent and reactive behavior, drug use, and early sexual activity (Arditti et al., 2003; Braman, 2004; Geller et al., 2012; Hanlon et al., 2005; Miller, 2006; Murray et al., 2012; Poehlmann, 2005; Snyder et al., 2001; Turney et al., 2012; Wildeman, 2010). Studies with strong research designs have consistently associated paternal incarceration with more aggression in young children (Cragie, 2011; Geller et al., 2012; Wakefield & Wildeman, 2011; Wildeman, 2010) and higher levels of delinquency in young adulthood (Murray et al., 2012). The effect of a father’s incarceration was both distinct from other causes of father absence (Geller et al., 2012) and heightened if the father was coresident prior to his incarceration (Geller et al.,

226

XING ZHANG AND ALLISON DWYER EMORY

2012; Wildeman, 2010). As female prisoners were more likely to report living with their children prior to incarceration than male prisoners were (Glaze & Maruschak, 2008), the stronger effects of coresidence may be particularly salient for maternal incarceration. In a comparison between children who had experienced maternal incarceration and paternal incarceration using the FFCW study, however, Wildeman and Turney concluded that paternal incarceration was more strongly associated with children’s behavior outcomes in their sample (2014).

Incarceration and Young Adulthood The extent to which the implications of parental incarceration continue into young adulthood is less well understood. Overall, research on parental incarceration and young adults has largely focused on outcomes related to individual behaviors like criminal activity and substance use. Less is known about how family instability experienced during childhood, especially for children with incarcerated mothers, may be reproduced in young adulthood. Previous studies of incarcerated women found that 75 95% experienced intimate partner violence (Browne et al., 1999; Zust, 2008), and 50% experienced sexual and physical abuse (Glaze & Maruschak, 2008). These findings suggest that domestic violence and negative relationship patterns may be modeled in households that also experience maternal incarceration. Children who experience abuse directly are significantly more likely to be arrested as juveniles and adults (Widom & Ames, 1994) or be involved in intimate partner violence themselves (Browne et al., 1999). Due to the prevalence of domestic violence among women who have been incarcerated, young adults who experienced maternal incarceration may be at greater risk for domestic violence. The implied link between childhood experiences of parental incarceration and adult outcomes is born out in the few studies looking at adult outcomes. Though primarily a study of adolescents, Murray et al. (2012) found that parental incarceration was associated with increased theft among young men. Among young adults in the Add Health data, paternal incarceration was associated with more delinquent behavior (Foster & Hagan, 2007; Roettger & Swisher, 2011) and arrests (Roettger & Swisher, 2011). Using the National Longitudinal Survey of Youth 1979, Huebner and Gustafson (2007) found a significant and positive association between conviction and probation in adulthood and experiences of maternal incarceration. Results have been more mixed regarding the effects of parental

Situating the Experience of Maternal Incarceration

227

incarceration on substance use of adult offspring. Murray et al. (2012) found no effects of parental incarceration on marijuana use on a sample of young men, while others found increased likelihood of marijuana and other illegal drug use in young adults whose fathers were incarcerated (Roettger et al., 2011). This paper builds on the existing literature by addressing three foundational research questions: (1) what is the childhood context of individuals who experienced maternal incarceration at some point during their childhood, (2) how does this childhood context conform to or differ from the context of children whose fathers were incarcerated, and (3) to what extent are the patterns of differences in childhood continued into young adulthood? These questions are addressed by comparing individuals who reported maternal incarceration during their childhood to both those who did not and those who experienced paternal incarceration. We examined three key areas to develop a descriptive picture of childhood and young adulthood: family structure, socioeconomic disadvantage, and exposure to violence and crime. We used data from the National Longitudinal Study of Adolescent to Adult Health (Add Health) to provide insight into the question of maternal incarceration. One key advantage of this dataset was the sampling frame of adolescents in selected schools, with interviews conducted with both adolescents and their primary caregivers. This frame may help mitigate the selective attrition on mother’s coresidence present in household or birth cohort studies. There was not a sufficient N of children experiencing maternal incarceration, however, to directly address whether children’s experiences varied by their mother’s coresidence during or after incarceration. The tradeoff for this gain in childhood context and representativeness, however, was that detailed data on the biological mother was limited in the many households where the mothers were not residing with their children at the time of the survey. Due to this limitation, this dataset lacked sufficient data on the biological mother’s selection factors to build a causal model. This research, however, provides a necessary foundation for future models of the implications of maternal incarceration for children.

METHODS This paper used the National Longitudinal Study of Adolescent to Adult Health (Add Health), a nationally representative school-based sample of

228

XING ZHANG AND ALLISON DWYER EMORY

adolescents from 1994 to 1995, with additional surveys gathering information from the respondent’s primary caregivers. Data came from wave I, when respondents were between the 7th and 12th grades, and wave IV, conducted in 2008 when respondents ranged in age from 24 to 34 years old. Wave IV had a sample size of 15,701 follow-up interviews from wave I (wave I response rate 79%; wave IV response rate 80.3%). In wave IV, interviews were conducted with individuals who were incarcerated at the time of the survey, which mitigated attrition based on outcomes of interest. Data on childhood experiences came from the wave I parent questionnaire and self-report, while young adult experiences and retrospective reports of incarceration came from wave IV. All respondents in wave IV with non-missing information on maternal incarceration history were included in the analytic sample. In the analytic sample, there were 643 respondents whose mothers had ever been or were currently incarcerated, but only 317 (49%) of those were known to have experienced maternal incarceration before the respondent’s 18th birthday. Just over 2% of the full analytic sample experienced maternal incarceration during their childhood. Approximately, 10% of the analytic sample reported paternal incarceration (N = 1,477). Wave IV interview questions related to intimate relationships, drug use, criminal offending, and adult mistreatment were conducted on laptops using computer-assisted selfinterview software (Harris et al., 2009). Missing data was not imputed, so sample size varied between measures due to item specific non-response.

Analytic Strategy The three research questions of this paper were considered in three broad areas: family structure, socioeconomic disadvantage, and exposure to crime and violence. Within each of these areas, several variables were considered. The first research question was addressed by comparing those who experienced maternal incarceration during their childhood to those who did not. The comparison group who did not experience maternal incarceration included those who experienced paternal incarceration and neither parent’s incarceration. In the rest of this paper, the group that experienced maternal incarceration is referred to as maternal, while the group that did not experience maternal incarceration is referred to as non-maternal. The second research question was addressed by comparing four mutually exclusive categories of respondents who experienced maternal incarceration, paternal incarceration, the incarceration of both parents, and no parental

Situating the Experience of Maternal Incarceration

229

incarceration during childhood. To avoid confusion, these groups are referenced throughout the paper as maternal-only, paternal-only, both, or neither. Finally, the third research question was addressed by including both childhood and young adulthood measures in the analysis. All of the analyses presented in this paper used the survey weights provided by Add Health with jackknife estimates of standard errors. The survey weights accounted for attrition between waves I and IV and adjusted for the initial survey design. This made the sample nationally representative of individuals who were in grades 7 12 in 1994. Weighted means were calculated for each parental incarceration group, and bivariate regressions were used as F-tests to determine where groups significantly differed using bivariate regressions. Bivariate regressions took the form of logistic regressions for dichotomous measures, OLS regressions for continuous variables, and multinomial logistic regressions for categorical variables. The first comparison used F-tests to determine whether the sample that experienced maternal incarceration differed from the sample that did not. The second comparison used F-tests to determine whether there were differences among the four parental incarceration experiences. Where the F-test of parental incarceration was significant, additional t-tests were used to compare the subcategories of parental incarceration. To protect against erroneous significances due to multiple comparisons, a Bonferroni correction was used to adjust all p-values by a factor of seven. Comparisons within parental incarceration type are represented by subscripts in tables. All tables report weighted means or percentages rather than coefficients for straightforward interpretation.

MEASURES Maternal incarceration during childhood was a dichotomous measure of whether the respondent, at wave IV, reported that his or her parent had ever been in jail or prison before the respondent’s 18th birthday. This measure was a combination of two variables. First, respondents were asked whether their biological mother ever spent time in jail or prison. If yes, respondents reported how old they were when their mother was first incarcerated and when she was most recently released. Respondents were counted as having experienced maternal incarceration during their childhood if their mother was first incarcerated between their birth and 18th birthday, or if she was released during that period. Due to data constraints, this is a conservative measure that does not capture mothers who were

230

XING ZHANG AND ALLISON DWYER EMORY

incarcerated for either the respondent’s entire childhood or both before the child’s birth and after their 18th birthday. This construction also relied on the child’s retrospective report of their age at the time their biological mother was incarcerated. A consistent measure of paternal incarceration during childhood was constructed using the parallel set of questions asked about the biological father. Type of parental incarceration was measured with mutually exclusive categories for children experiencing maternal incarceration during childhood, paternal incarceration during childhood, both parents’ incarceration during childhood, or neither parent’s incarceration during childhood.

Family Structure The first measure of family structure for children was the parent interviewed during the first wave of the survey. This primary caregiver provided the majority of the information about the household, the child’s background, and the child’s behavior in wave I. Add Health favored female family members in the following order: biological mother, stepmother, other female guardian, father, stepfather, and other male guardian. Household structure was measured using a snapshot of parents in the household at wave I. While this snapshot occurred during the window in which maternal incarceration was measured, it provides a conservative measure of family structure. Two additional measures were used to identify the primary parent figures during childhood. In wave IV, respondents retrospectively reported whom they considered the woman and man who raised them. This measure provided a retrospective report of childhood mother and father figures. Responses were collapsed into five categories: the biological mother/father, stepmother/stepfather, kin, nonkin, and no mother/father figure. Kin included a grandmother/grandfather, sister/brother, aunt/uncle, or other family member. The category of nonkin included a foster mother/father, adopted mother/father, and other unrelated caregiver. Finally, the age of first sexual intercourse was included as a measure of risk for new family formation during adolescence based on a report from wave IV. Family structure in young adulthood was measured through the lens of family formation. Young adults reported in wave IV how many sexual partners they had, and whether they had ever been married. Whether respondents had children of their own was measured through the number of live births reported.

Situating the Experience of Maternal Incarceration

231

Socioeconomic Disadvantage Socioeconomic disadvantage was conceptualized in this paper as having insufficient resources or social disadvantage. Respondents’ race and ethnicity were categorized as non-Hispanic white, non-Hispanic black, Hispanic, and other race. The responding parent at wave I reported the financial resources of the household in which the child was living at the time of the survey. Childhood measures included whether the parent reported having enough money to pay the bills, the household income in 1994, and whether the family was receiving public assistance at wave I in the form of welfare, AFDC, food stamps, or public housing. Several measures of disadvantage in young adulthood were included from the wave IV survey. Respondents were asked six questions about their financial situation in the past 12 months: whether the respondent lived in a household without phone service because there wasn’t enough money, couldn’t pay rent, was evicted for not paying rent, couldn’t pay bills, had the utilities turned off, or was worried about food running out. These items were collapsed into a scale with good internal consistency (α = .74) ranging from zero, indicating no economic hardship, to one, indicating affirmative responses to each economic hardship question. To capture earning potential, measures were included for whether the respondent reported working at least 10 hours per week and their highest level of education at the time of the wave IV survey.

Exposure to Crime and Violence Children’s exposure to crime and violence was separated into two primary areas: exposure to violence and crime, and exposure to substance use. Violence within the household was measured as whether the respondent reported in wave IV that a parent or caregiver had hit, kicked, or physically harmed them before they turned 18. Exposure to violence at wave I was measured using the eight-item Exposure to Physical Violence Scale in Add Health. Respondents reported whether they saw someone shoot or stab another person, got into a physical fight, got jumped, pulled a knife or gun on someone, shot or stabbed someone, had someone pull a knife or gun on him/her, were shot, or were cut or stabbed in the previous year. The exposure to violence scale had good internal consistency (α = .72) and ranged from zero, indicating the respondent reported none of these

232

XING ZHANG AND ALLISON DWYER EMORY

incidents, to one, indicating the respondent reported experiencing all of these incidents. Adolescents reported their own participation in delinquent acts on the Add Health Delinquency Scale. The scale included 15 items measuring the frequency of behaviors with good internal consistency (α = .84). These actions included painting graffiti in a public place, damaging property, lying to parents/guardians about where they had been, stealing from a store, getting into a physical fight, hurting someone so badly they needed bandages, running away from home, driving a car without permission, stealing something worth more than and less than $50, going into a home to steal something, using or threatening to use a weapon, selling marijuana, fighting where a group of friends was against another group, and being loud, rowdy, and unruly in a public place. The childhood delinquency scale ranged from zero, indicating no delinquent behavior, to three, indicating participation in all of the behaviors five or more times in the past year. Children’s exposure to substance use included three variables: mother alcoholism, father alcoholism, and the respondent’s own substance use. The responding caregiver reported whether each of the biological parents had a problem with alcoholism. This measure was not an objective measure of alcohol consumption, but was the only measure of substance use asked specifically about biological rather than resident parents. Whether the respondent used any substances themselves was a dichotomous measure of whether they reported binge drinking in the last year or using illegal substances in the last month. Exposure to violence and crime in young adulthood was measured as arrests, delinquent behaviors, and intimate partner violence reported at wave IV. Respondents reported how many times they had been arrested since their 18th birthday. As arrests are not necessarily synonymous with criminal behavior, the adult delinquency scale addressed behavior rather than criminal justice involvement. Young adults reported how often in the last 12 months they did each the following: damage property, go into a house or building to steal, sell drugs, steal something worth less than $50, participate in a fight with a group of friends against another group, get into a serious physical fight, or steal something worth more than $50. These items were collapsed into an adult delinquency scale with satisfactory internal consistency (α = .66) ranging from zero to three, where zero indicated no delinquent behavior and three indicated the respondent participated in all of these activities five or more times. Two measures of intimate partner violence in young adulthood were included to distinguish between relationship victimization and relationship

Situating the Experience of Maternal Incarceration

233

aggression. Respondents were asked whether four types of violence had occurred in their current or most recent relationship in the past year or since the beginning of the relationship. Approximately 3% of the analytic sample reported not having a relationship and were dropped from the sample for these variables.1 The questions were items from the Sexual Coercion Scale and Injury Scale of the Revised Conflict Tactics Scale (CTS2) (Straus et al., 1996). The relationship abuse scale (α = .76) included how often their partner had threatened violence, struck or kicked the respondent, injured the respondent, or forced unwanted sexual contact with the respondent. The relationship aggression scale (α = .71) was based on how often the respondent threatened violence, struck or kicked, injured, or forced unwanted sexual relations with his or her partner. Responses for both scales ranged from never (0) to more than 20 times (7). Substance use in young adulthood was a dichotomous measure of whether the respondent reported any binge drinking, marijuana use, or other illicit drug use in the past 12 months. Binge drinking was defined as four drinks in a row for women or five drinks in a row for men. The illicit drug referred to the most frequently used of the following list: sedatives, tranquilizers, stimulants, painkillers, steroids, or cocaine. A second measure of whether the respondent had used any illegal substances in the last 30 days was also reported.

FINDINGS As shown in Fig. 1, in this nationally representative sample parental incarceration was not a common experience during childhood. Approximately, 12% of respondents reported experiencing either maternal or paternal incarceration. Of those who did experience a parent’s incarceration, for most (10% of the full sample) the parent was their father. Just over 1% experienced a mother’s incarceration only, and less than 1% of respondents reported experiencing both parents’ incarceration during their childhood. The experiences of these individuals were examined in three key areas: family structure, socioeconomic disadvantage, and exposure to violence and crime. In each of these areas, outcomes were measured in both childhood and young adulthood, and compared first by their experiences of maternal incarceration and second by parental incarceration type. Comparisons between those who experienced maternal incarceration and those who did not are referred to as maternal and non-maternal. For the

234

XING ZHANG AND ALLISON DWYER EMORY 1.14%, N = 215

9.61%, N = 1,448 0.73%, N = 129 Mother only Father only Both parents Neither parent

88.24%. N = 12,747

Fig. 1.

Weighted Proportion Experiencing Parental Incarceration.

four-point comparison between those who experienced the incarceration of their mother, father, both parents, and neither parent, we use the terms maternal-only, paternal-only, both, and neither to prevent confusion.

Family Structure Table 1 documents family structure in childhood and young adulthood. The caregiver interviewed during the first wave provides insight into who was in the household at the time of the survey. Which caregiver was interviewed varied dramatically by maternal incarceration status. For the nonmaternal group, about 80% of caregivers interviewed were biological mothers, compared to about 53% of those in the maternal group. After biological mothers, the most frequently interviewed caregivers for the maternal group were grandmothers (11%), aunts (5%), and other caregivers including men (24%). These differences persist when comparing those who experienced only maternal incarceration to those who experienced only paternal incarceration. Only 46% of biological mothers for those in the maternal-only group were interviewed, compared to 80% of biological mothers for those in the paternal-only group. Apart from biological mothers, a significantly greater percentage of stepmothers, adoptive mothers, grandmothers, and other caregivers were interviewed for those who had a mother incarcerated compared to those who had a father incarcerated. The paternal-only group was

Non-maternal

Maternal

Family Structure. F-Test of Difference

Parental Incarceration during Childhood Maternal only

Paternal only

Both parents

Neither parent

45.67 7.08a 3.29a 2.08a 8.27a 6.79a 26.83a

79.68 1.29b 0.31b 0.08b 3.12b 0.95b 14.57b

61.14 0.00c 2.08ab 0.00c 13.05a 1.34ab 22.84ab

80.74 1.66b 0.28b 0.07b 1.27c 0.33b 15.64b

10.65 21.83a 10.12a 21.66a 6.83a 28.91a

29.79 22.62a 1.76b 33.69a 4.43b 7.72b

17.51 15.66a 4.09ab 39.51a 5.59abc 17.64a

61.74 11.15b 2.48b 17.01b 2.80c 4.82c

59.12

54.00

88.08

67.25

93.96

2.82 3.50 33.44 1.12

5.00a 4.98a 35.90a 0.12a

F-test of difference

Family structure in childhood Caregiver interviewed at wave I (%) Biological mother1 80.39 Stepmother 1.55 Adoptive mother 0.46 Foster mother 0.07 Grandmother 1.70 Aunt 0.43 Other caregiver 15.39 (including males) Family structure at wave I (%) Two biological parents1 Mom/stepdad Dad/stepmom Single mom Single dad Other

*** 52.63 4.37 2.70 1.17 10.72 4.51 23.91

***

*** 56.01 13.37 2.31 19.74 2.88 5.68

***

***

***

0.74b 0.95b 9.34b 0.88a

0.00c 0.30b 30.65a 1.79a

1.09b 0.58b 3.90c 0.47a

235

Retrospective report of mother figure Biological mother 92.92 (baseline) (%)1 Stepmother (%) 1.05 Nonkin (%) 0.76 Kin (%) 4.77 None (%) 0.50

12.16 19.40 7.28 29.43 5.61 26.12

Situating the Experience of Maternal Incarceration

Table 1.

Non-maternal

F-Test of Difference

Parental Incarceration during Childhood Maternal only

Paternal only

Both parents

Neither parent

47.21 10.73ab 11.33a 12.98ab 17.75a 15.09ab

54.71 18.38b 2.85b 10.59a 13.47ab 15.72a

42.98 15.45b 1.97abc 22.19b 17.41ab 14.72b

82.07 7.41ac 1.42c 3.78c 5.32c 16.81c

***

***

F-test of difference ***

39.56 14.56 7.24 18.70 19.94 14.95

***

12.61

17.04

*

17.00a

14.88a

17.95ab

12.35b

**

49.99 1.58

43.42 1.96

**

43.09a 1.91ab

46.32a 1.71b

41.10a 2.03ab

50.49a 1.55a

* **

Family characteristics in young adulthood Lifetime number of sexual partners Ever married (%) Number of live births

Figures reported use mother or father sample weights. Significance tested via binary OLS regression for continuous variables, logistic regression for dichotomous variables, and multinomial logistic regression for categorical variables. Means in the same row that share a subscript are not significantly different from one another, while means in the same row that do not share a subscript differ at p < .05 after adjustment for multiple comparisons. * p < .05; ** p < .01; *** p < .001. 1 This is the reference category in the multinomial logistic regression used to generate the F-statistic. No comparisons were made between incarceration types within the reference category.

XING ZHANG AND ALLISON DWYER EMORY

Retrospective report of father figure Biological father1 75.91 Stepfather 9.82 Nonkin 2.00 Kin 5.10 None 7.17 Age at first intercourse 16.67

Maternal

236

Table 1. (Continued )

Situating the Experience of Maternal Incarceration

237

statistically indistinguishable from the neither parent group for all caregiver groups except grandmothers. Those who experienced maternal incarceration had significantly different family structures in childhood than either the non-maternal group or those who experienced paternal incarceration. While in the non-maternal sample, the majority of adolescents lived in two-parent homes (56% with biological parents and 16% in another two-parent arrangement) at wave I, the majority of children who experienced maternal incarceration were living with a single parent or in other arrangements (61%). About a third of respondents in the maternal group reported living with a single mother compared to 20% for those in the non-maternal group. In addition, about 26% of respondents who experienced maternal incarceration had a family arrangement that did not include either biological parent, compared to 6% for those who did not experience maternal incarceration. The maternal-only group was also distinct from both the paternal-only and neither categories. Those in the paternal-only group were almost three times more likely to be living with both biological parents at the time of interview relative to those in the maternal-only group (30% vs. 11%), though both groups differed sharply from the 62% of the neither group with both biological parents. The maternal-only group was more likely to report living with a dad/stepmom, single dad, or other arrangement than those who experienced paternal incarceration. There were not significant differences, however, between the percentages with a single mother or mom/stepdad family structure. The most striking difference in family structure between those who experienced maternal and paternal incarceration was that 29% of those in the maternal-only group reported a family arrangement that did not involve either biological parent, compared to 8% of those in the paternal-only group. Unlike in the previous measure, however, the neither parent group was significantly different from the either parent incarceration group for family structure across most categories. Respondents’ choices of mother and father figures in wave IV echoed the pattern of differences in family structure by parental incarceration experiences. Young adults in the non-maternal group overwhelmingly reported their biological mother as the woman who raised them (about 93%), while those in the maternal group were more likely to report a different mother figure. Slightly over a third of the maternal group reported another relative as the primary mother figure, compared to 5% of those in the non-maternal group. A closer look at the patterns between the maternal-only and the paternal-only groups reveals that mother figures were significantly less

238

XING ZHANG AND ALLISON DWYER EMORY

likely to be biological mothers for respondents who experienced maternal incarceration, and more likely to be stepmothers, unrelated caregivers, or relatives. The paternal-only group, for the most part, was more similar to the neither group than the maternal-only group. Those in the paternal-only group were also more likely to report their biological mother as a primary mother figure, with almost 88% reporting their biological mother compared to 54% for those in the maternal-only group. About 9% of those in the paternal-only group reported a kin mother figure compared to 36% for those in the maternal-only group. There were also striking differences in the father figure young adults reported, suggesting that children who experienced maternal incarceration were also less likely to have their biological father playing a key role in their upbringing. There were more differences between the neither group, where 82% reported their biological father, and any parental incarceration category than between parental incarceration categories. The majority of children who experienced maternal incarceration, including those in the maternal-only and both groups, did not report their biological father as their main father figure. About 20% of those in the maternal group reported having no father figure growing up, compared to about 7% of those in the non-maternal group. Approximately 47% of those in the maternal-only group reported their biological father as their primary father figure, and 55% of those in the paternal-only group. Those in the maternalonly group were more likely to report nonkin father figures than the paternal-only group. This contrasts with the majority of children in the non-maternal group (76%) or those in the neither group (82%), whose biological fathers were their main father figure. Adolescents exposed to maternal incarceration had sexual intercourse about a year and a half earlier on average than those whose mothers were not incarcerated just under 15 compared to almost 17. This younger age was not significantly different from youth whose fathers were incarcerated. Both experiences of parental incarceration were significantly younger than those who had neither parent incarcerated. In young adulthood, differences in family characteristics persisted, with the largest differences found between those who experienced maternal incarceration and those who did not. Both the lifetime number of sexual partners and the number of live births were significantly higher for the maternal group in comparison to the non-maternal group. A smaller proportion of those in the maternal group were ever married compared to those in the non-maternal group, but the differences were not significant between the two groups.

Situating the Experience of Maternal Incarceration

239

Family characteristics in young adulthood were not significantly different within parental incarceration categories. Comparing the maternal-only to the paternal-only group, there were no significant differences in the number of sexual partners or the number of live births. Similarly, differences in the proportion ever married were not significant after correcting p-values. The main differences in family structure and stability in young adulthood were between those who experienced either parent’s incarceration and those who experienced neither.

Socioeconomic Disadvantage Table 2 shows socioeconomic disadvantages across the life course for each type of parental incarceration. One concern about incarceration in both the policy and academic world is the question of selection or disproportionate impact. The racial and ethnic makeup of respondents suggests that these concerns are relevant to our population. While maternal incarceration was observed in every racial category, the population who experienced maternal incarceration was majority non-white. Children who experienced maternal incarceration were significantly more likely to be non-Hispanic black than either children whose mothers were not incarcerated or children whose fathers were incarcerated. The population of children in the maternal group was 36% non-Hispanic black and 12% Hispanic. The population in the non-maternal group, on the other hand, was only 15% non-Hispanic black and 12% Hispanic. Almost 40% of those in the maternal-only group were non-Hispanic black, a proportion significantly higher than those in the paternal-only group. Paternal incarceration, in contrast, appeared more proportionate to the population who did not experience incarceration about 21% of respondents who experienced paternal incarceration were non-Hispanic black, a significantly larger proportion than the neither category but significantly smaller than the maternal-only group. Those who experienced maternal incarceration also experienced economic hardship during childhood. Almost a third of respondents in the maternal group lived in families that did not have enough money to pay the bills, compared to less than 20% of those in the non-maternal group. The mean income of the household they were living in at wave I was also significantly lower than those who did not experience any maternal incarceration $27,000 compared to $46,000 in the non-maternal group. Those who had a mother incarcerated were more likely to receive public assistance than those who did not have a mother incarcerated. Over half of those in

Nonmaternal

Socioeconomic Disadvantage.

Maternal

F-Test of Difference

240

Table 2.

Parental Incarceration during Childhood Maternal only

Paternal only

Both parents

Neither parent

9.68a 38.55a 46.42 5.35a

14.17a 21.08b 61.31 3.44a

13.95a 31.50ab 52.37 2.18a

11.54a 14.36c 69.03 5.07a

F-test of difference

Socioeconomic disadvantage in childhood Respondent race/ethnicity (%) Hispanic Non-Hispanic black Non-Hispanic white1 Other

***

11.76 36.29 48.28 3.67

82.95 46.07

70.29 26.62

*** ***

71.04a 30.52 ab

74.27a 33.17 a

66.74a 20.67b

84.45b 48.52c

*** ***

14.81

50.38

***

45.57ab

32.15a

58.34b

11.96c

***

0.08 77.67

0.22 73.77

***

0.18a 74.96

0.14a 74.14

0.23a 73.06

0.08b 78.24

***

8.85 17.80 9.34 33.52 30.50

23.04 23.60 13.04 29.22 11.09

19.87a 24.30a 13.13 a 30.90 11.80a

15.15a 20.92a 12.29 a 38.37 13.27a

20.76a 29.66a 10.65 a 27.21 11.72ab

7.87b 17.05b 9.08a 32.77 33.23b

Socioeconomic disadvantage in young adulthood Economic hardship scale2 Currently working at least 10 hours/ week (%) Highest level of education achieved Some high school or less High School or GED Vocational/some vocational Some college1 College or more

***

***

Figures reported use mother or father sample weights. Significance tested via binary OLS regression for continuous variables, logistic regression for dichotomous variables, and multinomial logistic regression for categorical variables. Means in the same row that share a subscript are not significantly different from one another, while means in the same row that do not share a subscript differ at p < .05 after adjustment for multiple comparisons. * p < .05; ** p < .01; *** p < .001. 1 This is the reference category in the multinomial logistic regression used to generate the F-statistic. No comparisons were made between incarceration types within the reference category. 2 Scale of six items ranging from 0 to 1, where 1 indicates the most hardship while 0 indicates the least.

XING ZHANG AND ALLISON DWYER EMORY

Household has money to pay bills (%) Household income at wave 1 ($ thousands) Family receiving public assistance at wave 1 (%)

*** 12.07 15.29 67.79 4.84

Situating the Experience of Maternal Incarceration

241

the maternal group received public assistance, compared to 15% of those in the non-maternal group. Breaking down further by parental incarceration type, approximately 46% of those in the maternal-only group received public assistance compared to 32% of those in the paternal-only group, but these differences were not significant. While families in the maternal-only group reported greater financial hardship in terms of not having enough money to pay the bills and lower household incomes than those in the paternal-only group, these differences were not significant. All differences between any form of parental incarceration and neither were significant, however, and children who experienced both parents’ incarceration were significantly more disadvantaged than those who only experienced paternal incarceration. For those who experienced maternal incarceration, economic hardship persisted in the transition to young adulthood. Scores on the economic hardship scale, which measured items such as being unable to pay rent and bills and range from zero (least hardship) to one (most hardship), were significantly higher for those who had an incarcerated mother than those who did not. Those who had mothers incarcerated reported higher scores on the economic hardship scale than those whose fathers were incarcerated. Similar to the childhood measure, however, this difference was not significant. None of the groups differed on current employment. Educational attainment was much lower for those in the maternal group with 47% not going beyond high school, compared to 27% of those in the non-maternal group. A greater percentage of those who had a mother incarcerated did not complete high school or complete any college education relative to those who had a father incarcerated, but these differences were not significant. The group that had neither parent incarcerated was significantly more likely to have completed college or more and less likely to have stopped at high school than respondents who experienced any parental incarceration.

Exposure to Violence, Crime, and Substance Use Table 3 shows exposure to violence, crime, and substance use in childhood and young adulthood. Overall, the largest differences were found between those who experienced maternal incarceration and those who did not. In general, the maternal-only and the paternal-only groups were not significantly different from each other, with the exception being a significant difference in maternal alcoholism. A large proportion of respondents whose

242

Table 3. Nonmaternal

Exposure to Violence, Crime, and Substance Use.

Maternal

F-Test of Difference

Parental Incarceration during Childhood Maternal only

Paternal only

Both parents

Neither parent

F-test of difference

Exposure to violence and crime in childhood 18.48

38.04

***

35.16a

31.16a

40.10a

16.56b

***

0.10

0.18

***

0.17a

0.14a

0.16a

0.09b

***

0.28

0.40

**

0.40a

0.36a

0.38a

0.27b

**

28.85 34.65 38.76

*** *** *

31.15a 27.77a 38.00ab

4.37b 40.89a 38.65b

27.67a 46.59a 39.66ab

2.07c 11.65b 31.11a

*** *** **

2.19 0.11 0.30 0.55

** *** * *

2.06ab 0.10ab 0.25 ab 0.42 ab

1.31 b 0.07a 0.28 a 0.40 b

2.11ab 0.14a 0.39 ab 0.71 b

0.62a 0.04b 0.14b 0.27a

*** *** *** ***

Exposure to substance abuse in childhood Mother alcoholism (%) Father alcoholism (%) Any substance use (%)

2.45 15.13 31.88

Violence and crime in young adulthood Adult arrests Adult delinquency scale3 Relationship aggression4 Relationship abuse4

0.70 0.05 0.16 0.29

XING ZHANG AND ALLISON DWYER EMORY

Experienced physical abuse (%) Exposure to violence scale1 Delinquency scale2

Substance use last 12 months Illegal drug use last 30 days

57.47

62.76

19.70

34.67

***

59.67a

61.89a

70.34a

56.98a

*

32.28a

29.40a

37.29a

18.39b

***

Figures reported use mother or father sample weights. Significance tested via binary OLS regression for continuous variables, logistic regression for dichotomous variables, and multinomial logistic regression for categorical variables. Means in the same row that share a subscript are not significantly different from one another, while means in the same row that do not share a subscript differ at p < .05 after adjustment for multiple comparisons. * p < .05; ** p < .01; *** p < .001. 1 Scale of eight items ranging from 0 (no exposure to violence) to 1 (exposure to all types of violence included in scale). 2 Scale of whether the respondent participated in the delinquent behaviors, ranging from 0 (no delinquent acts) to 1 (participated in all 15 delinquent behaviors). 3 Delinquent behavior is a scale measure ranging from 0 (no delinquent acts in the last year) to 3 (multiple and frequent delinquent acts). 4 Scale measure of relationship abuse/aggression ranging from 0 (no abuse/aggression) to 7 (frequent or multiple forms of abuse/ aggression).

Situating the Experience of Maternal Incarceration

Substance use in young adulthood

243

244

XING ZHANG AND ALLISON DWYER EMORY

mothers were incarcerated reported experiencing physical abuse as a child. Just over 38% of those who experienced maternal incarceration reported experiencing violence as a child. Of those who did not experience maternal incarceration, about 18% reported such abuse. Over a third of those who had either a mother or a father incarcerated experienced physical abuse in childhood, but the percentage of respondents experiencing childhood abuse was not significantly different between these groups. Respondents who had a mother incarcerated reported almost twice as much exposure to violence as those whose mothers were not incarcerated. Differences between maternal and paternal incarceration were not significant, however. The exposure to violence scale measured whether the respondent saw and participated in violent behaviors like shooting, stabbing, and physical fights. These adolescents also participated in significantly more delinquent behaviors relative to those without a mother incarcerated, but not significantly more than those whose fathers were incarcerated. In childhood, respondents in the maternal group were more likely to have a parent report alcoholism and use substances themselves. Significantly higher proportions of those in the maternal group reported having a mother with alcoholism (29%) or father with alcoholism (35%), compared to those in the non-maternal group (about 3% and about 15%, respectively). Comparing those who experienced maternal to paternal incarceration, the most striking difference was in maternal alcoholism over a third of the incarcerated mothers were reported to have alcoholism during the respondent’s childhood, compared to 4% of mothers for those in the paternal-only group. While a higher proportion of incarcerated fathers were reported to have alcoholism than fathers in the maternal-only category, that difference was not significant. In addition, early substance use was significantly higher for those in the maternal group relative to those in the non-maternal group, with about 39% who reported using any illegal substances for the maternal group. However, early substance use was not significantly different for those in the maternal-only group compared to those in the paternal-only group, with about 38 39% reporting any substance use. The only significant difference with regard to early substance use was between paternal-only and neither incarceration. In young adulthood, higher exposure to violence and crime persisted. Comparing the maternal to the non-maternal group, the maternal group reported 2.2 adult arrests, which was significantly higher than the nonmaternal group who reported 0.7 arrests. A closer comparison of maternalonly and paternal-only showed that those with an incarcerated mother

Situating the Experience of Maternal Incarceration

245

experienced about 0.70 additional arrests, but that this difference was not significant. The adult delinquency scale was much higher for those in the maternal than the non-maternal group, though both groups fell low on the scale. The adult delinquency scale did not significantly differ by parental incarceration. An additional measure of violence in young adulthood was intimate partner violence, divided into relationship aggression and relationship abuse scales. Experiences of intimate partner violence, including relationship abuse and relationship aggression, were higher for the maternal group relative to the non-maternal group. While the difference appeared small, it is the equivalent of experiencing one more form of violence on the scale or increasing the frequency of an abusive behavior by one unit. Relationship abuse and relationship aggression were not significantly different between the maternal-only and paternal-only groups. The only significant differences among the groups with regard to relationship violence were between the paternal-only and neither groups. Substance use in young adulthood was measured by substance use in the last 12 months and illegal drug use in the last 30 days. Substance use in the last 12 months was not significantly different between maternal and nonmaternal groups or between parental incarceration categories. However, illegal drug use in the last month was significantly higher for the maternal group, with about 35% reporting use of an illegal drug relative to about 20% who in the non-maternal group. There were no significant differences in illegal drug use between those in the maternal-only and paternal-only groups, with about a third of both groups reporting illegal drug use in the past month. Illegal drug use in the past 30 days was significantly higher for those in the maternal-only, paternal-only and both groups compared to the neither group.

DISCUSSION These findings provide a descriptive picture of the lives of individuals whose mothers were incarcerated during their childhood. A minority of children in the United States experienced a parent’s incarceration, and of those who did, about 16% experienced maternal incarceration. The analysis provided in this paper centered around three key research questions: (1) what is the childhood context of individuals who experienced maternal incarceration at some point during their childhood, (2) how does this childhood context conform to or differ from the context of children whose

246

XING ZHANG AND ALLISON DWYER EMORY

fathers were incarcerated, and (3) to what extent do the patterns of differences in childhood continue into young adulthood? Addressing the first question, this paper found that individuals whose mothers were incarcerated during their childhood had significantly more disadvantaged childhood experiences than those whose mothers were not. This difference was perhaps most notable in the area of family structure. While the majority of the adolescents in the sample lived with their biological mother at the time of the survey, around half of the children whose mothers had been incarcerated reported a different arrangement. Other family members played a much larger role for this latter group, and both fathers and unrelated adults emerged as primary caregivers more frequently for adolescents with incarcerated mothers. These findings are consistent with previous research suggesting that children live with other relatives or go into the foster system when their mother is incarcerated (Glaze & Maruschak, 2008; Johnson & Waldfogel, 2004), but also extend these differences in structure beyond the window where mothers are in jail or prison. While some of the mothers reported in our sample may have been incarcerated during nearly the entire childhood of the respondent, most were not. These findings suggest that differences in family structure may be more than a short-term coping mechanism for current incarceration. The retrospective reports of the person who raised the respondent also suggest that living arrangements observed at the time of the survey were often more than just temporary caregiving relationships. Respondents whose mothers were incarcerated were less likely to give their biological parents that status, while other family members were more likely to be named as parent figures. This data cannot address, however, whether differences in family structure predated the incarceration or were continuations of arrangements established during the incarceration. While these different family structures are not necessarily harmful for children, caregivers other than the biological mother may require additional support and information about assistance programs. Prior research on fathers who step into caregiving roles suggests that even among biological parents, there can be confusion about access to benefits and programs (Waller, 2009). In addition to differences in family structure, those who experienced maternal incarceration were more likely than those who did not to live in disadvantaged households, be exposed to crime and violence, and have substance users in the family. The households of respondents whose mothers were incarcerated reported about half the income of the comparison sample. In addition, over half received public assistance, and though the

Situating the Experience of Maternal Incarceration

247

majority still reported having enough money to pay the bills, a larger proportion struggled to do so. Children in these households were also more likely to have parents with alcoholism and were more likely to have used alcohol or drugs themselves. Within their households, almost 40% of respondents reported some sort of physical abuse or harsh discipline. Within their communities and schools, they were nearly twice as likely to have reported observing or participating in violence and delinquent behavior. For context, these children observed or participated in approximately one additional item on the violence scale and participated in two additional delinquent behaviors than those whose mothers were not incarcerated. These findings suggest that children whose mothers were incarcerated faced many co-occurring hardships within their families and childhood, and faced heightened exposure to violence in their homes and communities. The second research question sought to explain how the experiences of individuals whose mothers were incarcerated during their childhood compared to those whose fathers were incarcerated during that window. Our findings suggest that the socioeconomic disadvantage and exposure to crime and violence were largely consistent between parental incarceration experiences despite large differences in family structure. These findings affirm that children whose fathers were incarcerated were likely to remain living with their mother (Glaze & Maruschak, 2008; Johnson & Waldfogel, 2004), which was in sharp contrast to children whose mothers were incarcerated. Fathers with a history of incarceration were less likely to be living with their children than those in families with no parental incarceration, but overall these households were more similar than paternal incarceration and maternal incarceration families. This highlights that while fathers or other family members can step up to a parenting role when a mother is incarcerated, most children remain with mothers who were already the primary and relatively stable parent if their fathers are incarcerated. Adolescents who experienced maternal or paternal incarceration reported similar levels of socioeconomic disadvantage and exposure to crime and violence. Across measures of economic hardship, there were no differences between children whose mothers were incarcerated and those whose fathers were incarcerated. The only significant difference with regard to exposure to substance use was the rate of maternal alcoholism, which was very low for those who experienced their father’s incarceration. One notable exception to this similarity, however, was the racial and ethnic composition of these populations. While paternal incarceration affected a much larger proportion of children and has been discussed as a source of racial stratification (Wildeman, 2009), the population experiencing

248

XING ZHANG AND ALLISON DWYER EMORY

maternal incarceration was even more disproportionately non-white. Over a third of young adults who experienced maternal incarceration during their childhood were non-Hispanic black, in contrast to less than a quarter of those who only experienced paternal incarceration. The estimates of Glaze and Maruschak (2008) suggest that the children of women in state and federal prisons may be even more disproportionately non-white than our findings. For the most part, these findings suggest that parental incarceration was more salient for increasing exposure to abuse, violence, and delinquency during childhood than having neither parent incarcerated. Finally, our descriptive approach found that most of the observed childhood patterns continued into young adulthood. This suggests that experiences of maternal and paternal incarceration during childhood may at least serve as a flag for understanding and addressing problems in young adulthood. Family structure in childhood was the clearest distinguishing feature of the maternal incarceration experience, but these differences softened in young adulthood. While those who experienced maternal incarceration during their childhood reported earlier sexual debut and more sexual partners, they were not significantly different from those who experienced paternal incarceration. The young adults who experienced maternal incarceration were equally likely to have been married in young adulthood, and the difference in the number of births was significant but relatively small. The socioeconomic disadvantage and exposure to crime and violence of young adults also varied by whether either parent was incarcerated than between parental incarceration experiences. Young adults whose mothers were incarcerated reported more economic hardship and less educational attainment than those whose mothers were not, but these disadvantages were comparable to those reported by the paternal incarceration sample. Similarly, while young adults whose parents were incarcerated had higher levels of reported delinquency, arrests, relationship violence, and substance use, there were not significant differences by which parent was incarcerated. These findings suggest that the two experiences of incarceration were comparable with respect to the exposure to socioeconomic hardship, substance use, domestic violence, and criminal activity in young adulthood. While this study answered many of the lingering questions about correlates of maternal incarceration for children, there were several limitations to this approach. Using the Add Health data sacrificed the ability to control for selection factors like the mother’s own substance use, delinquent behavior, impulsivity, or mental health. This limitation precluded a causal approach to linking maternal incarceration and child outcomes. This data provided a much richer set of contextual variables for both children and

Situating the Experience of Maternal Incarceration

249

young adults independent of their relationship with the incarcerated mother. Given these limitations, our findings presented the correlates for children and young adults of maternal incarceration but do not address whether these associations were caused by maternal incarceration, predisposed mothers to criminal behavior, or merely correlated with other selection factors for maternal incarceration like poverty. A second limitation was the conflation of jails and prisons within the data, institutions that may have different implications for families given differences in distance and sentence length (Arditti, 2012a). This data also lacked a sufficient sample size and question specificity to differentiate mothers by the duration or frequency of incarceration. Our findings should be interpreted conservatively as there may be heterogeneity by incarceration type masked by current measures. Despite these limitations, the present study provides a descriptive background of the co-occurring hardships faced by the children of incarcerated women that provides groundwork for future causal work on the impact of incarceration.

CONCLUSION While there has been a substantial body of research built around paternal incarceration, and a smaller body of literature describing maternal incarceration from the mother’s perspective, this paper addressed the need for foundational research on children whose mothers were incarcerated. This population was particularly distinctive in both family structure and disproportionate minority representation. Children who experienced maternal incarceration were much more likely than other groups to live with someone other than their biological mother, and often lived with grandparents, other relatives, fathers, or unrelated adults instead. These diverse family types may require both assistance accessing resources available to families with children and special consideration when constructing policies aimed at the families of incarcerated women. This study found that the experiences of maternal and paternal incarceration were similar with regard to economic disadvantage and exposure to crime and violence. These similarities suggest that outside the area of family structure and the direct implications of family, there is room to read the existing literature on paternal incarceration more broadly as parental incarceration. Children and young adults who have had a parent incarcerated are at elevated risk for economic hardship, exposure to violence in

250

XING ZHANG AND ALLISON DWYER EMORY

their communities and relationships, and participation in delinquent behavior. At the very least, parental incarceration of any type can identify families that are more likely to be struggling and need additional support. The persistence of these hardships into young adulthood may also signal the relevance of parental incarceration for the intergenerational transmission of disadvantage. Isolating the experience of maternal incarceration from the co-occurring disadvantages faced by children with this experience remains a data challenge. These findings suggest that many of the known confounding variables and co-occurring hardships regularly used when modeling paternal incarceration can also apply to future causal models of maternal incarceration. The differences in family life experienced by these children suggest that maternal incarceration was associated with profound challenges, if not changes, for their children and their children’s caregivers. Family structure should be a component of a rigorous causal model of maternal incarceration’s implications for children. Research into the relevant mechanisms is needed to shed light on the question of how to support children with incarcerated mothers with resources that may be otherwise lacking in their lives.

NOTE 1. Of those not reporting a relationship, only 1.48% reported maternal incarceration during their childhood (n = 7), so excluding these people does not strongly impact the reported mean.

ACKNOWLEDGMENTS The authors would like to thank Rachel Dunifon, Kelly Musick, Kimberly Turner, and Christopher Wildeman for feedback on earlier versions of this paper. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant No. DGE1144153 and the Cornell Population Center. Earlier versions of this paper were presented at the 2014 meetings of the Add Health Users Conference, the American Sociological Association, and the Association for Public Policy Analysis and Management. This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan

Situating the Experience of Maternal Incarceration

251

Harris at the University of North Carolina at Chapel Hill, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health website (http://www.cpc.unc.edu/addhealth). No direct support was received from grant P01-HD31921 for this analysis.

REFERENCES Arditti, J. A. (2012a). Parental incarceration and the family: Psychological and social effects of imprisonment on children, parents, and caregivers. New York, NY: New York University Press. Arditti, J. A. (2012b). Child trauma within the context of parental incarceration: A family process perspective. Journal of Family Theory & Review, 4(3), 181 219. doi:10.1111/j.17562589.2012.00128.x Arditti, J. A., Lambert-Shute, J., & Joest, K. (2003). Saturday morning at the jail: Implications of incarceration for families and children. Family Relations, 52(3), 195 204. doi:10.1111/j.1741-3729.2003.00195.x Bachman, H., & Chase-Lansdale, P. (2005). Custodial grandmothers’ physical, mental, and economic well-being: Comparisons of primary caregivers from low-income neighborhoods. Family Relations, 54(4), 475 487. doi:10.1111/j.1741-3729.2005.00334.x Braman, D. (2004). Doing time on the outside: Incarceration and family life in urban America. Ann Arbor, MI: University of Michigan Press. Brandon, P. (2005). Welfare receipt among children living with grandparents. Population Research and Policy Review, 24(5), 411 429. doi:10.1007/s11113-005-4437-y Browne, A., Miller, B., & Maguin, E. (1999). Prevalence and severity of lifetime physical and sexual victimization among incarcerated women. International Journal of Law and Psychiatry, 22(3 4), 301 322. doi:10.1016/S0160-2527(99)00011-4 Comfort, M. (2008). Doing time together: Love and family in the shadow of the prison. Chicago, IL: University of Chicago Press. Cragie, T. (2011). The effect of paternal incarceration on early child behavioral problems: A racial comparison. Journal of Ethnicity in Criminal Justice, 9(3), 179 199. doi:10.1080/ 15377938.2011.594349 Dunifon, R. (2013). The influence of grandparents on the lives of children and adolescents. Child Development Perspectives, 7(1), 55 60. doi:10.1111/cdep.12016 Enos, S. (2001). Mothering from the inside: Parenting in a Women’s prison. New York, NY: State University of New York Press. Foster, H., & Hagan, J. (2007). Incarceration and intergenerational social exclusion. Social Problems, 54(4), 399 433. doi:10.1525/sp.2007.54.4.399 Foster, H., & Hagan, J. (2009). The mass incarceration of parents in america: Issues of race/ ethnicity, collateral damage to children, and prisoner reentry. Annals of the American Academy of Political and Social Science, 623, 179 194. doi:10.1177/0002716208331123

252

XING ZHANG AND ALLISON DWYER EMORY

Geller, A., Cooper, C. E., Garfinkel, I., Schwartz-Soicher, O., & Mincy, R. B. (2012). Beyond absenteeism: Father incarceration and child development. Demography, 49(1), 49 76. doi:10.1007/s13524-011-0081-9 Geller, A., & Franklin, A. W. (2014). Paternal incarceration and the housing security of urban mothers. Journal of Marriage and Family, 76(2), 411 427. doi:10.1111/jomf.12098 Glaze, L. E., & Maruschak, L. M. (2008). Parents in prison and their minor children. Bureau of Justice Statistics Special Report. U.S. Department of Justice, Washington, DC. Retrieved from http://www.bjs.gov/content/pub/pdf/pptmc.pdf Hagan, J., & Foster, H. (2012). Intergenerational educational effects of mass imprisonment in America. Sociology of Education, 85(3), 259 286. doi:10.1177/0038040711431587 Hanlon, T. E., Blatchley, R. J., Bennett-Sears, T., O’Grady, K. E., Rose, M., & Callaman, J. M. (2005). Vulnerability of children of incarcerated addict mothers: Implications for preventive intervention. Children and Youth Services Review, 27(1), 67 84. doi:10.1016/ j.childyouth.2004.07.004 Hanlon, T. E., O’Grady, K. E., Bennett-Sears, T., & Callaman, J. M. (2005). Incarcerated drug-abusing mothers: Their characteristics and vulnerability. American Journal of Drug and Alcohol Abuse, 31(1), 59 77. doi:10.1081/ADA-200037564 Harm, N., & Thompson, P. J. (1995). Needs assessment: Children of incarcerated women and their caregivers. Little Rock, AR: Centers for Youth & Families. Harris, K. M., Halpern, C. T., Whitsel, E., Hussey, J., Tabor, J., Entzel, P., & Udry, J. R. (2009). The national longitudinal study of adolescent to adult health: Research design. Retrieved from http://www.cpc.unc.edu/projects/addhealth/design Haskins, A. R. (2014). Unintended consequences: Effects of paternal incarceration on child school readiness and later special education placement. Sociological Science, 1, 141 157. doi:10.15195/v1.a11 Huebner, B. M., & Gustafson, R. (2007). The effect of maternal incarceration on adult offspring involvement in the criminal justice system. Journal of Criminal Justice, 35(3), 283 296. doi:10.1016/j.jcrimjus.2007.03.005 Johnson, E. I., & Waldfogel, J. (2004). Children of incarcerated parents: Multiple risks and Children’s living arrangements. In M. Patillo, D. Weiman & B. Western (Eds.), Imprisoning America: The social effects of mass incarceration. New York, NY: Russell Sage Foundation. Johnston, D. (1995). Effects of parental incarceration. In K. Gabel & D. Johnston (Eds.), Children of incarcerated parents (pp. 59 88). New York, NY: Lexington Books. LaVigne, N. G., Davies, E., & Brazzell, D. (2008). Broken bonds: Understanding and addressing the needs of children with incarcerated parents. Retrieved from http://www.urban. org/UploadedPDF/411616_incarcerated_parents.pdf Miller, K. M. (2006). The impact of parental incarceration on children: An emerging need for effective interventions. Child and Adolescent Social Work Journal, 23(4), 472 486. Murray, J., Loeber, R., & Pardini, D. (2012). Parental involvement in the criminal justice system and the development of youth theft, marijuana use, depression, and poor academic performance. Criminology, 50(1), 255 302. doi:10.1111/j.1745-9125.2011.00257.x Nurse, A. (2002). Fatherhood arrested: Parenting from within the juvenile justice system. Nashville, TN: Vanderbilt University Press. Pager, D. (2003). The mark of a criminal record. American Journal of Sociology, 108(5), 937 975. doi:10.1086/374403

Situating the Experience of Maternal Incarceration

253

Pettit, B., & Western, B. (2004). Mass imprisonment and the life course: Race and class inequality in US incarceration. American Sociological Review, 69(2), 151 169. Pew Charitable Trust. (2010). Collateral costs: Incarceration’s effect on economic mobility. Retrieved from http://www.pewtrusts.org/~/media/legacy/uploadedfiles/pcs_assets/2010/ CollateralCosts1pdf.pdf Phillips, S. D., Erkanli, A., Keeler, G. P., Costello, E. J., & Angold, A. (2006). Disentangling the risks: Parent criminal justice involvement and children’s exposure to family risks. Criminology & Public Policy, 5(4), 677 702. doi:10.1111/j.1745-9133.2006.00404.x Poehlmann, J. (2005). Children’s family environments and intellectual outcomes during maternal incarceration. Journal of Marriage and Family, 67(5), 1275 1285. doi:10.1111/ j.1741-3737.2005.00216.x Roettger, M. E., & Swisher, R. R. (2011). Associations of fathers’ history of incarceration with sons’ delinquency and arrest among black, white, and hispanic males in the United States. Criminology, 49(4), 1109 1147. doi:10.1111/j.1745-9125.2011.00253.x Roettger, M. E., Swisher, R. R., Kuhl, D. C., & Chavez, J. (2011). Paternal incarceration and trajectories of marijuana and other illegal drug use from adolescence into young adulthood: Evidence from longitudinal panels of males and females in the United States. Addiction, 106(1), 121 132. doi:10.1111/j.1360-0443.2010.03110.x Siegel, J. (2011). Disrupted childhoods: Children of women in prison. New Brunswick, NJ: Rutgers University Press. Snyder, Z. K., Carlo, T. A., & Coats Mullins, M. M. (2001). Parenting from prison: An examination of a children’s visitation program at a women’s correctional facility. Marriage and Family Review, 32(3 4), 33 61. Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised conflict tactics scales (CTS2) Development and preliminary psychometric data. Journal of Family Issues, 17(3), 283 316. doi:10.1177/019251396017003001 Turanovic, J. J., Rodriguez, N., & Pratt, T. C. (2012). The collateral consequences of incarceration revisited: A qualitative analysis of the effects on caregivers of children of incarcerated parents. Criminology, 50(4), 913 959. doi:10.1111/j.1745-9125.2012.00283.x Turney, K. (2014). The consequences of paternal incarceration for maternal neglect and harsh parenting. Social Forces, 92(4), 1607 1636. doi:10.1093/sf/sot160 Turney, K. (2015). Hopelessly devoted? Relationship quality during and after incarceration. Journal of Marriage and Family, 77(2), 480 495. doi:10.1111/jomf.12174 Turney, K., Schnittker, J., & Wildeman, C. (2012). Those they leave behind: Paternal incarceration and maternal instrumental support. Journal of Marriage and Family, 74(5), 1149 1165. doi:10.1111/j.1741-3737.2012.00998.x Turney, K., & Wildeman, C. (2013). Redefining relationships: Explaining the countervailing consequences of paternal incarceration for parenting. American Sociological Review, 78(6), 949 979. doi:10.1177/0003122413505589 Wakefield, S., & Wildeman, C. (2011). Mass imprisonment and racial disparities in childhood behavioral problems. Criminology & Public Policy, 10(3), 793 817. doi:10.1111/j.17459133.2011.00740.x Waller, M. R. (2009). Family man in the other america: New opportunities, motivations, and supports for paternal caregiving. Annals of the American Academy of Political and Social Science, 624, 156 176. doi:10.1177/0002716209334372 Western, B. (2001). Incarceration, unemployment and inequality. Focus, 21, 32 36.

254

XING ZHANG AND ALLISON DWYER EMORY

Western, B., Kling, J. R., & Weiman, D. (2001). The labor market consequences of incarceration. Crime and Delinquency, 47, 410 427. doi:10.1177/0011128701047003007 Western, B., Lopoo, L., & McLanahan, S. L. (2004). Incarceration and the bonds between parents in fragile families. M. Patillo, D. Weiman & B. Western (Eds.), Imprisoning America: The social effects of mass incarceration. New York, NY: Russell Sage Foundation. Widom, C. S., & Ames, M. A. (1994). Criminal consequences of childhood sexual victimization. Child Abuse & Neglect, 18(4), 303 318. doi:10.1016/0145-2134(94)90033-7 Wildeman, C. (2009). Parental imprisonment, the prison boom, and the concentration of childhood disadvantage. Demography, 46(2), 265 280. doi:10.1353/dem.0.0052 Wildeman, C. (2010). Paternal incarceration and children’s physically aggressive behaviors: Evidence from the fragile families and child wellbeing study. Social Forces, 89(1), 285 309. doi:10.1353/sof.2010.0055 Wildeman, C., & Turney, K. (2014). Positive, negative, or null? the effects of maternal incarceration on children’s behavioral problems. Demography, 51(3), 1041 1068. doi:10.1007/s13524-014-0291-z Zust, B. L. (2008). Assessing and addressing domestic violence experienced by incarcerated women. Creative Nursing, 14(2), 70 72. doi:10.1891/1078-4535.14.2.70

DELINQUENT EFFECTS OF CHILDHOOD EXPOSURE TO VIOLENT VICTIMIZATION: A LATENT LONGITUDINAL CLASS ANALYSIS Henriikka Weir and Catherine Kaukinen ABSTRACT Purpose The present study uses data from the National Longitudinal Study of Adolescent Health (Ad Health) to evaluate the effects of exposure to violent victimization in childhood on adolescent delinquency and subsequent adult criminality. Methodology/approach Using Longitudinal Latent Class Analysis (LLCA), the present study investigates whether there are distinct and diverse longitudinal delinquency trajectories among those exposed to violence in childhood. Findings Findings from the current study indicate that there are three distinct trajectories of delinquency and offending from age 14 to 27 for both males and females exposed to violence in childhood. Further, it

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 255 284 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009012

255

256

HENRIIKKA WEIR AND CATHERINE KAUKINEN

appears that violent victimization in childhood bridges the gender gap in delinquency between males and females. Thus, childhood violent victimization, and the fact that females are victimized by parents/caregivers and romantic partners at higher rates than males, might be partially responsible in explaining the narrowing of the gender gap between male and female offending in the recent decades. At the same time, childhood violent victimization also seems to impact males and females in somewhat different ways. Practically, all female victims stop offending by their late 20s, whereas a fairly large proportion of males exposed to violent victimization in childhood steadily continue offending. Research limitations/implications Although this study was able to identify the diverse impacts of violence exposure on engagement in subsequent delinquency, it did not examine the unique contributions of each type of violence on adolescent outcomes or the chronicity of exposure to each of these types of violent victimization. We were also not able to measure all types of violence experiences in childhood, such as exposure to parents’ or caregivers’ intimate partner violence. Social implications While early prevention would be the most desirable option for both genders for the most optimal outcome, the retrospective intervention and treatment programs should be gender-specific. For males, they should heavily focus on providing alternative ways to cope with anger, impulse control and frustration, as well as teach empathy, cognitive problem solving skills, verbal communication skills, and tangible life and job skills. For females, most successful intervention and treatment programs may focus on helping the girls through a transition from adolescence to adulthood while providing mental health, medical, and family support services. Originality/value The paper uses a unique methodological approach to identify distinct and diverse longitudinal delinquency trajectories. The findings demonstrate how more resilient individuals (in terms of externalizing behaviors) can bring down the mean scores of delinquency even though many other individuals can be severely affected by violence exposure in childhood. Keywords: Childhood exposure to violence; externalizing behaviors; Add Health; Longitudinal Latent Class Analysis; delinquency; violence

Delinquent Effects of Childhood Exposure to Violent Victimization

257

INTRODUCTION Childhood exposure to violence is an important public health concern and is one of the most commonly studied risk factors for involvement in a host of adolescent antisocial behaviors including substance use and abuse, delinquency, adult offending and criminal justice system involvement, and later relationship violence (Allwood & Widom, 2013; Widom, 1989; Widom, White, Czaja, & Marmorstein, 2007). Estimates of children’s exposure to diverse forms of violent victimization including community violence, street and peer victimization, child abuse, dating violence, and witnessing intimate partner violence suggest that many children are exposed to some type of violence in their daily lives (Finkelhor, Turner, Ormrod, Hamby, & Kracke, 2009). While official reports from the Child Welfare Information Gateway (2014) estimate an official response to over 3.4 million referrals involving the alleged maltreatment of approximately 6.3 million children, this figure represents a fraction of the problem of children exposed to violence. Empirical work by Finkelhor (2008) suggests that children are at greater risk of exposure to violence as compared to adults. Estimates suggest that as many as 60% of children were at risk of experiencing and/or witnessing violence within a 12 month period including community, family, and school violence (Finkelhor et al., 2009). Finkelhor, Turner, Shattuck, and Hamby’s (2013) most recent scholarship using data from the National Survey of Children’s Exposure to Violence finds that approximately twofifths of children and youth have experienced a physical assault within 12 months of data collection, resulting in one in ten children experiencing an assault-related injury. Work by Aizer (2008) suggests that one in four children have been exposed to violence within their neighborhoods and that these experiences have implications for childhood and adolescent development, a host of negative academic outcomes, and later violence outcomes (Aizer, 2008). Children react to exposure to violence in diverse ways, and while many children demonstrate remarkable resilience, while others experience a host of short- and long-term impacts. Research suggests that children are particularly vulnerable to a whole host of personal, social, academic, and behavioral consequences of exposure to violence (Davis, Ammons, Dahl, & Kliewer, 2015; Jouriles & McDonald, 2015; Kliewer & Lepore, 2015; Monahan, King, Shulman, Cauffman, & Chassin, 2015; Turner, Ormrod, Hamby, & Kracke, 2009). Importantly, a child’s exposure to one type of violence increases the likelihood that the child will be

258

HENRIIKKA WEIR AND CATHERINE KAUKINEN

exposed to and experience other types of violence and exposed multiple times (Department of Justice, 2009).

The Impact of Childhood Exposure to Violence Overall, the impacts of violence on children include a heightened risk for a host of risk of externalizing behaviors including subsequent involvement in crime and delinquency, dating violence, further victimization and child welfare involvement, and subsequent conflict with the law and involvement with the juvenile justice systems (Allwood & Widom, 2013; Finkelhor et al., 2009). Some common externalizing behaviors observed in the literature vary depending on the age group of those children and adolescents under study but have often included, aggression, bullying, lying, truancy, teen-pregnancy, risky sexual behaviors, substance abuse, violence, and varying degrees of delinquency/later criminal offending (Allwood & Widom, 2013; Crittenden, Wright, & Fagan, 2011). For example, work by Monahan et al. (2015) have explored how exposure to violence in childhood and adolescence shapes the development of both impulse control and future orientation for both adolescence and on into adulthood. It is important to note that while some empirical research finds substantial consequences of exposure to violence and experiences of violent victimization, the literature here is somewhat conflicting on the exact impact (Crittenden et al., 2011; Emery, 2006). The literature is complicated by an absence in some work to fully control for personal, family, and community variables that interact and perhaps confound the impact of exposure to violence. In examining the relationship between childhood exposure to violence and longitudinal outcomes, there is no one leading theory that satisfactorily explains this relationship. Emery (2006) suggests that there are three major relevant classes of theories that can explain the relationship between the violence exposure and consequent manifestation of externalizing symptoms: Theories of duress, theories of development, and theories of deviance. Criminologists almost exclusively study, test, and develop different theories of deviance. As the name implies, these theories attempt to explain the causal reasons why an individual or a group of individuals become offenders or exhibit other behaviors considered deviant in any given culture. While many criminological perspectives such as strain, general theory of crime, social learning, and life-course theories could explain why traumatic repeated events such as exposure to violence during

Delinquent Effects of Childhood Exposure to Violent Victimization

259

childhood could lead to increases in externalizing behaviors, few criminological theories have fully explicitly emphasized it as a potential significant causal variable (Cullen & Agnew, 2010). In outlining the literature below we highlight the work by Widom (1989) and demonstrate how the social learning framework, in particular, the cycle of violence has dominated the discussion on how violent victimization in childhood shapes later involvement in criminal and other externalizing behaviors. We draw on the work by Widom (1989) and Widom and Maxfield (2001) that notes that while it is a common belief that “violence via begets violence that today’s abused children become tomorrow’s violent offenders,” the majority of abused and neglected children do not become delinquent, criminal, or violent. Below, we provide an overview of the literature specific to three types of violent victimization experienced by children and adolescents and the impacts of these types of violence exposure child maltreatment, dating violence, and street/pee violence. We also identify the gendered consequences of these types of victimization for young boys’ and girls’ involvement in adolescent offending.

The Impact of Child Abuse and Exposure to Violence in the Family of Origin It is difficult to fully estimate the extent to which children are exposed to family violence given that the majority of experiences of violence within the home, including exposure to inter-parental intimate partner violence and child abuse and neglect do not come to the attention of law enforcement. Within the United States, Sullivan, Egan, and Gooch (2004) and Rossman, Hughes, and Rosenberg (2000) have estimated that approximately 10 million children each year are exposed to or witness intimate partner violence between their parents and/or caregivers. Data from the Finkelhor, Ormrod, and Turner (2007) estimates that between 35 and 50% of children live in a household characterized by some form of intimate partner violence and/or child abuse and neglect. The work specific to children’s experience of child abuse has produced clear and definitive findings around the negative impacts. There are both short- and longterm impacts leading to psychological and functional damage, substance abuse, delinquency, and aggression and violence (Allwood & Widom, 2013; Maas, Herrenkohl, & Sousa, 2008; Stouthamer-Loeber, Loeber, Homish, & Wei, 2001; Widom, 1989; Widom & Maxfield, 2001; Widom et al., 2007).

260

HENRIIKKA WEIR AND CATHERINE KAUKINEN

The work by Cathy Spatz Widom has served to guide much of the literature on explorations of the consequences of childhood experiences of child abuse and neglect on myriad of adolescent and adult outcomes. The work by Widom (1989) led her to outline the Cycle of Violence, in which she demonstrated that a history of family violence is associated with an increased likelihood of becoming an abuser via experience of familial abuse during childhood. In her work she has argued that for some victims of child abuse and neglect, they may accept and internalize abusive messages, which then in turn become a self-fulfilling prophecy. Hence, the idea that “violence begets violence.” These early victimization experiences alter life trajectories and heighten the risk of offending over the lifecourse. Her work in this area also includes an examination of a diverse set of consequential outcomes on how children and adolescents are impacted by child abuse and neglect. This includes running away, problematic drinking, illicit substance use and abuse, negative mental health, delinquency, and other antisocial outcomes, including official measures of criminal justice involvement (Allwood & Widom, 2013; Horwitz, Widom, McLaughlin, & White, 2001; Widom, 1989; Widom & Maxfield, 2001; Widom et al., 2007). In particular, Allwood and Widom (2013) have identified a diverse set of negative consequences including school failure and disengagement, impacts on employment opportunities, in addition to the increased likelihood of involvement in the criminal justice system as an adolescent and adult. Work by Silverman, Reinherz, and Giaconia (1996) has added to the literature laid out by Widom in demonstrating the impact of child abuse on mental health, showing that children who experienced abuse were more likely to develop psychiatric disorders and behavioral problems than those who experienced no abuse. Widom’s long-term follow up (Allwood & Widom, 2013) to her pivotal and landmark study has examined how child abuse and neglect is associated with long-term adult outcomes including depression, anxiety, and substance abuse. In particular, in this work she demonstrates that social support serves to mediate and moderate these outcomes. Other work in this area has explored how experiences of child abuse may also have differential consequences by gender. One explanation is that given girls and young women are more often the victims of these types of violence, it subsequently increases the risk of later adolescent offending. An alternative explanation is that girls and women’s experience with violence is qualitatively different than that of boys and young men and that this shapes the etiology and pathways of women’s offending (Belknap, Holsinger, & Dunn, 1997; Chesney-Lind, 1997). For example, some

Delinquent Effects of Childhood Exposure to Violent Victimization

261

researchers have suggested that girls and women’s entrance into crime and criminality is rooted in their unique experiences of sexual and physical victimization (Belknap et al., 1997; Chesney-Lind, 1997; Chesney-Lind & Rodriguez, 1983; Daly, 1992). Work by Asscher, Van der Put, and Stams (2015) have used Juvenile Court data to examine the connections between early experiences of childhood physical and sexual abuse and later criminal involvement. Among female adolescent offenders they found these girls more often had a history of abuse as compared to male offenders. Drawing on data and conclusions from a meta-analysis, Hubbard and Pratt (2002) similarly find that in addition to school and family relationships offending among female adolescents is highly predicted by experiences of physical and sexual victimizations. Child abuse in particular has been shown to have implications for subsequent involvement in violent dating relationships (Milletich, Kelley, Doane, & Pearson, 2010; Stith et al., 2000). At the same time this literature has demonstrated important gender differences in the consequence of child abuse. Some of the literature in this area suggests that the relationship between child abuse and later dating violence perpetration is stronger for boys versus girls (Doumas, Margolin, & John, 1994; Stith et al., 2000). For example, work Marshall and Rose (1988) has shown that males who have both witnessed domestic violence and experienced child abuse were at heightened risk of dating violence aggression. In contrast other work has shown that girls experiencing child abuse are at a heightened risk of both dating violence victimization and offending as compared to male child abuse victims (Gover, Kaukinen, & Fox, 2008; Kaukinen, Buchanan, & Gover, 2015; Milletich et al., 2010; Straus, 2004).

The Impact of Adolescent Dating Violence Victimization Approximately one in five high school-aged adolescents is the victim of dating violence each year (Wolfe et al., 2009). Adolescent dating violence often results in negative physical, psychological, health, and behavioral outcomes (Foshee, Linder, MacDougall, & Bangdiwala, 2001). This includes adverse mental health issues and other related problems, including depression, anxiety, eating disorders, school disengagement, and posttraumatic stress (Connolly, Friedlander, Pepler, Craig, & Laporte, 2010). The research literature has clearly demonstrated the relationship between dating violence experiences and later consequential outcomes for adolescents and young adults (Kaura & Lohman, 2007; Rutter, Weatherill,

262

HENRIIKKA WEIR AND CATHERINE KAUKINEN

Taft, & Orazem, 2012; Straight, Harper, & Arias, 2003; Zurbriggen, 2009). For example, work by Swahn, Bossarte, Palmier, Yao, and Van Dulmen (2013) has shown the impact of dating violence on a variety of later outcomes including poor mental health, substance use and abuse, and the risk of suicide and suicide attempts. There is also work that has explored the role of anger and aggression within dating violence etiology and consequences. Empirical work by Rutter et al. (2012) has shown that feelings of anger are important outcomes associated with dating violence victimization. More importantly, their work demonstrates important gender differences in finding that physical and psychological victimization is more strongly associated with different forms of anger for men as compared to women. Within the dating violence literature, most often scholars have examined the impact of child abuse and witnessing inter-parental violence on subsequent involvement in adolescent intimate partner violence. Less often have scholars explored the connection between dating violence and other types of violent perpetration and victimization over the adolescent life-course. Longitudinal research in particular, examining the connection between dating violence experiences and later involvement in other types of aggression and delinquency is limited. An exception is recent work by Orpinas, Nahapetyan, Song, McNicholas, and Reeves (2012) who have examined the association between dating violence trajectories and peer aggression. In their work they explore four distinct developmental dating violence victimization and perpetration trajectories, a low victimization/ low perpetration group, increasing victimization/increasing perpetration, high victimization/high perpetration, and increasing victimization/low perpetration. For all groups, they find that engagement in peer aggression decreased over time. At the same time they found that those reporting the highest rates of peer aggression were among the high victimization/ high perpetration group. At the same time the students in the lowest peer aggression group were more often among the low victimization/low perpetration group. They conclude that there exists a strong reciprocal relationship between those engage in dating violence and adolescents most often participating in violent aggression towards their peers. Similarly, research by Herrenkohl, Kosterman, Mason, and Hawkins (2007) has examined the connection between young people’s violence trajectories and the developmental course of intimate partner violence. In their work, they find that two patterns of youth violence, “chronic and continuing” and what they refer to as a “late-increaser pattern” of youth violence. Both patterns are related to a higher likelihood of later intimate partner

Delinquent Effects of Childhood Exposure to Violent Victimization

263

violence perpetration. Finally with respect to gender differences, work by Ozer, Tschann, Pasch, and Flores (2004) has explored the co-occurrence of different forms of violence including, peer violence, sexual aggression, and dating violence. They found that among adolescent boys, but not girls, those who perpetrate violence in one domain are more likely to perpetrate violence in other domains.

The Impact of Street and Peer Violence Exposure Children are often exposed to the risk of violent victimization among their peer networks and within their communities. Research by Finkelhor et al. (2009) estimate that approximately half of American youth have experienced an assault and as many as 13% of children and youth have been the victims of physically assaultive bullying. Research (Wolke, Copeland, Angold, & Costello, 2013) has shown that physically and emotionally bullied children are at risk for a number of negative physical, school, and mental health impacts. In particular bullying is more likely to lead to both social and behavioral outcomes that persist into adulthood including depression and anxiety, increased feelings of sadness and loneliness, changes in sleep and eating patterns, and loss of interest in activities they used to enjoy. Work specific to the consequences of street and peer violence by Grogger (1998) suggests that children who have experienced school violence victimizations have higher levels of disengagement from school including reduced attendance, reduced ability to participate in-school activities, compromised academic achievement, and consequently reduced graduation rates. Street and peer violence experiences also have been shown to be associated with mental health impacts for youth. For example, Kelly, Anderson, Hall, Peden, and Cerel (2012) have identified how youth exposure to gang violence are impacted in terms of their mental health and well-being of male youth. One way to understand the connection between adolescent experiences of street and peer violence and subsequent antisocial behavior is an exploration of retaliatory attitudes. Work by Copeland-Linder, Johnson, Haynie, Chung, and Cheng (2012) recruited a group of assault-injured youth. The adolescents were recruited from an emergency room setting and were interviewed at baseline, 6 months, and 18 months to estimate involvement in fighting behavior, retaliatory attitudes, weapon carrying, and injury. They concluded that assault experiences heighted the risk of retaliatory attitudes and a cycle of violence and aggression.

264

HENRIIKKA WEIR AND CATHERINE KAUKINEN

The Present Study While there is a large literature on the effects of violence exposure on children, the validity of the results from many of these studies is more problematic (Emery, 2006). Several studies are cross sectional, lack proper scales to measure externalizing behaviors, use unsuitable data sets, or utilize inappropriate statistical modeling practices. Even when the studies use longitudinal data, they often fail to employ appropriate research designs and statistical techniques to deal with confounding variables, and thus will not allow any causal inferences to be made from the results (Emery, 2006). Furthermore, the longitudinal studies investigating the developmental consequences of violence exposure on externalizing behaviors frequently fail to take into account that children can be differentially impacted by exposure to violence (Holt, Buckley, & Whelan, 2008). The described lack of methodological rigor and theoretical foundation can be a serious problem in the research on childhood exposure to violence exposure. The present work attempts to address this gap in the literature by using the National Longitudinal Study of Adolescent Health (Add Health) data with scales measuring a wide variety of self-reported externalizing behaviors such as delinquency, violence, substance abuse, and subsequent adult criminal offending. Using a nationally representative sample of males and females the present study utilizes an application of a groupbased trajectory modeling to identify the delinquent effects of childhood experiences of violent victimization. In particular, we examine the delinquent outcomes associated with childhood maltreatment, dating violence victimization, and street/peer violence. We also examine the gendered impacts of these three types of violence.

METHODOLOGY Data This study utilizes data from the non-restricted portion of the National Longitudinal Study of Adolescent Health (Add Health), which is a nationally representative multi-wave cohort study of adolescent transition into adulthood (Udry, 2003). To date, Add Health has collected four waves of data, with first wave gathering information from a sample of U.S. adolescents who were in grades 7 12 in 1994 1995. Using a stratified sampling

Delinquent Effects of Childhood Exposure to Violent Victimization

265

design, more than 90,000 students from 132 middle and high schools were selected to fill out in-school questionnaires during the first wave of data collection on topics such as parental education and occupation, social and demographic characteristics, as well as risks and social behaviors. Even more in depth in-home interview was then conducted on a subsample of 20,745 youth. Since the first wave, the original subsample of in-home participants has been interviewed three more times (1996, 2001 2002, and 2007 2008). Throughout all four data collection waves, information was collected from multiple sources, including self-, peer-, teacher-, parent-, sibling-, and romantic partner-reports. Information gathered comprises topics such as social and demographic characteristics, family and peer relationships, physical and mental health, school and later occupational achievement, as well as criminal offending of the participants. Add Health has also harvested extensive information on participants’ experiences with interpersonal violence victimization in terms of childhood maltreatment, intimate partner violence, and street violence. As indicated, the current study employs the publicly available data from all four waves of Add Health, restricting participants to those who were 14 or 15 years old at wave 1 (mean ages across the four waves were 14.12, 15.03, 20.45, and 27.02).1 The resulting subsample consisted of 2,096 participants. Of these, 62% (n = 1,299) were Caucasian and 57% (n = 1,194) were female. At wave four, 45% (n = 943) of the participants were married and majority (57%) of the sample (n = 1,195) had attended college or completed a vocational/technical training after high school.

Variables Delinquency The dependent variable for this study was delinquency, which was measured at each of the four waves. During each data collection point, the participants filled out self-reports regarding their engagement in various delinquent acts during the past 12 months, ranging from minor vandalism to robbery and aggravated assault.2 The present study utilized 12 closely corresponding items involving counts of illegal acts at each four waves. Each waves’ individual delinquency items were converted to dichotomous variables (i.e., no involvement in a specific delinquency = 0, at least one incident of a specific delinquency = 1). The 12 resulting dichotomous variables were then added together at each wave to create a variety scale of delinquency, possible scores ranging from 0 to 12 (x = 1.53 and s.d. = 1.98

266

HENRIIKKA WEIR AND CATHERINE KAUKINEN

at wave 1; x = 1.20 and s.d. = 1.85 at wave 2; x = 0.64 and s.d. = 1.33 at wave 3; x = .0.43 and s.d. = 0.96 at wave 4). Using a variety scale of delinquency (vs. frequency scale) follows the convention of several other studies utilizing delinquency measures from different Add Health waves (see Haynie, 2001; Porter & Vogel, 2014; Rees & Pogarsky, 2011).3,4 The internal consistencies (Cronbach’s alpha) for the four resulting delinquency scales (waves 1 through 4) were .75, .77, .73, and .73, respectively. Exposure and Experience of Violent Victimization5 Exposure and the experience of interpersonal violent victimization in childhood were constructed from three main sources. First, Child Maltreatment was constructed from items measuring wave 4 retrospective self-reports of emotional, physical, and/or sexual abuse prior to age 18 were added and transformed into a binary variable (0 = no incidents of childhood maltreatment; 1 = at least one incident of childhood maltreatment). As Table 1 shows, 39% of females (n = 375) and 33% of males (n = 241) retrospectively reported experiences of childhood maltreatment. Second, Dating Violence came from items where respondents were asked about their experiences with dating violence victimization during the past 12 months at all four waves. Since the purpose of the present study was to investigate childhood exposure to violence, only measures including waves 1 and 2 (i.e., prior to age 18) self-reports of dating violence were included and constructed into a binary variable (0 = no incidents of dating violence victimization; 1 = at least one incident of dating violence victimization). As Table 1.

Exposure to Interpersonal Violence in Childhood by Type and Gender.

Type Childhood maltreatment Males Females Dating violence Males Females Street violence victimization Males Females Overall exposure to violence Males Females

Frequency

Percentage

241 375

33 39

43 60

6 6

215 193

29 20

370 494

50 51

Delinquent Effects of Childhood Exposure to Violent Victimization

267

Table 1 shows, 6% of females (n = 60) and 6% of males (n = 43) reported dating violence victimization at wave 1 or 2. Third, respondents were asked about their experiences with Street and Peer Violent Victimization, this was a measure of any violent victimization regardless of the perpetrator during the past 12 months at all four waves. This measure mostly encompassed street and peer types of victimization (i.e., items such as “getting stabbed/ shot/beaten”). Again, since the purpose of the present study was to investigate childhood exposure to violent victimization, only measures including waves 1 and 2 self-reports of violent street and peer victimization were included and constructed into a binary variable (0 = no incidents of street/ peer violent victimization; 1 = at least one incident of street/peer violent victimization). As Table 1 shows, 20% of females (n = 193) and 29% of males (n = 215) reported street/peer violent victimization at wave 1 or 2. Finally, an overall dichotomous variable intended to capture exposure to interpersonal violence in childhood was constructed based o participants’ reported experiences with any of the above three types of victimizations (0 = no violent victimization; 1 = yes exposure to at least one type of violent victimization).6 As Table 1 indicates, 51% of females (n = 494) and 50% of males (n = 370) were exposed to violence in childhood.

Analysis Plan The purpose of this study is to assess whether the selected subsample of Add Health participants exhibit distinct developmental trajectories of delinquency from age 14 to 27 based on their violent victimization status in childhood. Group-based trajectory modeling (GBT), frequently utilized in longitudinal research investigating the development of behaviors and lifecourse transitions (Nagin & Piquero, 2010), lends itself to this type of investigations. This technique was originally formulated by Nagin and Land (1993) to identify meaningful developmental paths over time or age as well as factors that might predict or modify these developmental pathways (Nagin, 2005). GBT modeling has several advantages over traditional ways of analyzing longitudinal data. For example, whereas standard growth curve models only deliver the mean probability trajectory for the study sample and attempt to isolate reasons why some individuals deviate from the population mean, GBT modeling highlights that there might be distinct clusters of development, characterized by similar patterns (Nagin, 2005). Indeed, GBT modeling is data-driven and cases are appointed to distinct latent classes based on probabilities (Nagin, 2005). Factors predicting

268

HENRIIKKA WEIR AND CATHERINE KAUKINEN

group membership, the slope and/or the intercept among each given class can also be identified (Nagin, 2005; Nagin & Piquero, 2010). While there are several different types of group-based trajectory modeling techniques, longitudinal latent class analyses (LLCA) was utilized in the present study. For the purposes of this study, LLCA was deemed beneficial over some other GBT techniques. For example, LLCA permits complex models to be estimated without violating model assumptions because it does not adopt any specific time function (Feldman, Masyn, & Conger, 2009). In other words, rather than having to estimate intercepts and slopes, the change over a period of time can be explained by events impacting participants’ development (Feldman et al., 2009). As an example regarding the present study, participants’ levels of delinquency could have been diversely influenced by several experiences from 14 to 27 (such as biological and peer-influences, family and neighborhood environment, contact with criminal justice system, and the focus of our study violent victimization experiences prior to age 18). Indeed, covariates can also be used to predict the group membership that is, the probability of belonging to a specific latent class (Feldman et al., 2009).

RESULTS While it would be an interesting question to further explore the delinquent trajectories across separate and unique forms of violence exposure, it was beyond the scope of this paper. The examine the overlap current study explored the bivariate relationships among the violence exposure measures used in current paper to examine the overlap in violence exposure (i.e., child maltreatment exposure and dating violence exposure; dating violence exposure and street/peer violence exposure; as well as child maltreatment exposure and street/peer violence exposure). These analyses showed that there were substantial overlap in experiences of childhood exposure to violence. In particular, among the Add Health data, there was a strong overlap in dating violence experiences and both childhood maltreatment and street/ peer violence exposure. Among those youth experiencing dating violence, 40% were the victims of street/peer violence and 37% were the victims of child maltreatment. Children exposed to street violence were also more likely to have experienced child maltreatment. Fully, 38% of children who were victims of violence in their communities were the victims of child maltreatment.

Delinquent Effects of Childhood Exposure to Violent Victimization

269

Fig. 1 presents the mean levels of delinquency at each of the four waves of data collection across sex and exposure to violence in childhood. There are statistically significant differences in the mean levels of delinquency between the “non-exposed” and “exposed” groups for both females (t = −8.32; t = −7.22; t = −2.04; t = −2.62) and males (t = 9.11; t = −9.03; t = −3.93; t = −7.17) at all four waves. However, simply looking at mean scores of delinquency across any given time period, even if modeled by more sophisticated methods such as traditional growth curves that include covariates, can be misleading. Because different individuals respond to violence exposure in childhood in different ways, more resilient (in terms of externalizing behaviors) individuals can bring down the mean scores of delinquency even though many other individuals can be severely affected by violence exposure in childhood. Thus, our analysis will estimate the LLCA models to determine whether developmental trajectories of delinquency differ across sex and interpersonal violence exposure in childhood. The first step in LLCA modeling involves the selection of the best-fitting model. This is typically performed by estimating various unconditional trajectory models without covariates (see Nylund & Masyn, 2008). Model selection phase seeks to determine the number of classes that should be retained and the decision is typically rooted in the assessment of comparative fit indices, visual examination of graphs, as well as parsimony. In the present study, LLCA was deemed as the most appropriate approach for

3.00 2.50 Exposed Males 2.00 Not Exposed Males 1.50

Exposed Females Not Exposed Females

1.00 0.50 0.00 Wave 1

Fig. 1.

Wave 2

Wave 3

Wave 4

Mean Levels of Delinquency for All Four Subsamples (Not Trajectories) across Four Waves.

270

HENRIIKKA WEIR AND CATHERINE KAUKINEN

the longitudinal count outcome (i.e., delinquency across the four waves) following negative binomial distribution. Bayesian Information Criterion (BIC), posterior probabilities, class proportions, parsimony, and visualization plots7 were evaluated, and are presented in Tables 2 5, in order to decide an appropriate number of latent classes to retain for each subsample. Table 2 shows fit statistics for the 2-group, 3-group, 4-group, and 5group LLCA results for males who were exposed to interpersonal violence in childhood. Tables 3 5 present the same statistics for males with “no exposure to violence,” for females with “exposure to violence” and for females with “no exposure to violence” in childhood, respectively. Bayesian Information Criteria (BIC)8 for all subsamples implied that the five-class solutions would be the most appropriate in depicting the development of delinquency for all four subsamples. However, three-class models were selected over five-class solutions because of parsimony, larger class proportions, overall posterior probabilities, and visual inspections of the graphs for all other groups except for males not exposed to violence in Table 2.

BIC

Latent Longitudinal Class Analysis Results for Males Exposed to Interpersonal Violence. 2 Group

3 Group

4 Group

5 Group

3845.058

3766.134

3727.174

3674.817

Prob. Posterior probabilities and class proportions for 2 group model Class 1 0.982 Class 2 0.936 Posterior probabilities and class proportions for 3 group model Class 1 0.970 Class 2 0.940 Class 3 0.957 Posterior probabilities and class proportions for 4 group model Class 1 0.985 Class 2 0.957 Class 3 0.917 Class 4 0.955 Posterior probabilities and class proportions for 5 group model Class 1 0.990 Class 2 0.923 Class 3 0.915 Class 4 0.911 Class 5 0.953

%

#

83.5 16.5

198 39

77.6 11.8 10.6

184 28 25

70.5 15.2 10.5 3.8

167 36 25 9

64.1 3.0 13.1 16.0 3.8

152 7 31 38 9

Table 3.

BIC

Latent Longitudinal Class Analysis Results for Males Not Exposed to Interpersonal Violence. 2 Group

3 Group

4 Group

5 Group

3097.426

2941.079

2854.174

2775.532

Prob. Posterior probabilities and class proportions for 2 group model Class 1 0.999 Class 2 0.967 Posterior probabilities and class proportions for 3 group model Class 1 0.990 Class 2 0.999 Class 3 0.971 Posterior probabilities and class proportions for 4 group model Class 1 0.998 Class 2 1.000 Class 3 1.000 Class 4 0.942 Posterior probabilities and class proportions for 5 group model Class 1 0.984 Class 2 0.998 Class 3 0.953 Class 4 1.000 Class 5 1.000

Table 4.

BIC

%

#

92.9 7.1

235 18

86.6 5.9 7.5

219 15 19

80.2 7.1 5.9 6.7

203 18 15 17

76.7 11.5 8.7 2.0 1.2

194 29 22 5 3

Latent Longitudinal Class Analysis Results for Females Exposed to Interpersonal Violence. 2 Group

3 Group

4 Group

5 Group

3997.089

3836.706

3760.461

3701.534

Prob. Posterior probabilities and class proportions for 2 group model Class 1 0.995 Class 2 0.947 Posterior probabilities and class proportions for 3 group model Class 1 0.975 Class 2 0.996 Class 3 0.967 Posterior probabilities and class proportions for 4 group model Class 1 0.960 Class 2 0.926 Class 3 0.993 Class 4 0.996 Posterior probabilities and class proportions for 5 group model Class 1 0.990 Class 2 0.923 Class 3 0.915 Class 4 0.911 Class 5 0.953

%

#

91.5 8.5

300 28

10.7 83.2 6.1

35 273 20

9.5 6.1 80.5 4.0

31 20 264 13

75.6 3.4 14.6 6.1 0.3

248 11 48 20 1

272

Table 5.

BIC

HENRIIKKA WEIR AND CATHERINE KAUKINEN

Latent Longitudinal Class Analysis Results for Females Not Exposed to Interpersonal Violence. 2 Group

3 Group

4 Group

5 Group

2816.621

2637.368

2463.516

2443.838

Prob. Posterior probabilities and class proportions for 2 group model Class 1 1.000 Class 2 1.000 Posterior probabilities and class proportions for 3 group model Class 1 0.997 Class 2 1.000 Class 3 0.962 Posterior probabilities and class proportions for 4 group model Class 1 0.993 Class 2 1.000 Class 3 0.971 Class 4 1.000 Posterior probabilities and class proportions for 5 group model Class 1 0.985 Class 2 0.946 Class 3 0.999 Class 4 1.000 Class 5 0.903

%

#

97.1 2.9

306 9

83.8 2.9 13.3

264 9 42

82.5 1.6 13.0 2.9

260 5 41 9

4.4 5.1 78.1 2.9 9.5

14 16 246 9 30

childhood. For this group, the four-class solution was selected as the bestfitting model. Indeed, BIC alone should not be used in the model section making process because BIC often improves until the class counts get very small due to fact that LLCA is aimed to detect discrete homogeneous clusters among data. Further, when distinct latent classes become so small as to be included as few individuals as three (1.2%) as they did with “no exposure to violence” group with male five-class solution, it becomes impossible to investigate any potential impact of covariates within or between latent classes. Although it is certainly imperative to carefully consider fit statistics when conducting the model selection phase of LLCA, the ultimate decision regarding the final best-fitting model should also be theoretically driven and consider the preponderance of evidence. As mentioned, three-class solutions were selected as the best-fitting models for all other subsamples, except for males who had not been exposed to violence in childhood. Fig. 2 depicts the results from these analyses graphically (i.e., the final selected models). Regarding males who had been

273

Delinquent Effects of Childhood Exposure to Violent Victimization Development of Delinquency for Males Exposed to Interpersonal Violence (Ages 13 – 29, n = 237)

4

Delinquency

2

4 0

0

2

Delinquency

6

6

8

8

Development of Delinquency for Males not Exposed to Interpersonal Violence (Ages 13 – 29, n = 253)

1

2

3

1

4

2 Abstainers: 80.2% Mid-Level Persisters: 5.9%

Life-Course Persisters: 11.8%

4

Low-Level Persisters: 7.1% Adolescence-Limiteds: 6.7%

Development of Delinquency for Females not Exposed to Interpersonal Violence (Ages 13 – 29, n = 315)

4 0

0

2

2

4

Delinquency

6

6

8

8

Development of Delinquency for Females Exposed to Interpersonal Violence (Ages 13 – 29, n = 328)

Delinquency

3 Wave

Wave Abstainers: 77.6% Adolescence-Limiteds: 10.6%

1

2

3

4

Wave Early-Adulthood Limiteds: 10.7% Adolescence-Limiteds: 6.1%

Fig. 2.

1

2

3

4

Wave Abtainers: 83.2%

Abstainers: 83.8% Adolescence-Limiteds: 13.3%

Adult-Onsetters: 2.9%

Final Selected Longitudinal Latent Class Analysis (LLCA) Models for All Four Subsamples.

exposed to violence in childhood, 77.6% were assigned to a latent class that exhibit low levels of delinquency from age 14 to 27. This group was named “Abstainers.” The second latent class, termed “AdolescenceLimiteds” was made up of 10.6% of males exposed to violence in childhood. Their initial levels of delinquency were roughly three times higher than the delinquent behavior exhibited by “Abstainers.” However, by age 27, “Adolescent-Limiteds” were practically indistinguishable from “Abstainers.” Finally, there was an identifiable “Life-Course Persistent” (11.8%) group among the males exposed to violence in childhood. As Fig. 2 indicates, while members in this group started their offending paths at lower levels than “Adolescence-limited” group, and actually had much lower levels of offending at mean age of 15, by wave four, this group was offending at nearly four times higher rates than either “Abstainers” or “Adolescent-Limiteds.” Similarly to males exposed to violence in childhood, males who had not been exposed to violence in childhood were compromised of “Abstainers”

274

HENRIIKKA WEIR AND CATHERINE KAUKINEN

(80.2%) and “Adolescence-Limiteds” (6.7%), although both of these groups offended at substantially lower levels at all four waves than comparable groups among males who were exposed to violence. Notably, the “Abstainer” group was also larger among males not exposed to violence and the “Adolescent-Limited” group was smaller among males not exposed to violence when compared to males who were exposed to violence in childhood. As Fig. 2 shows, two additional small latent classes exhibiting fairly low levels of steady offending were identified among males not exposed to violence. These groups were named as “Low-Level Persisters” (7.1%) and “Mid-Level Persisters” (5.9%). Three distinct latent classes were identified (see Fig. 2) for females exposed to violence in childhood. Similarly to their male counter parts who had been exposed to violence in childhood, victimized females were compromised of “Abstainers” (83.2%) and “Adolescent-Limiteds” (6.1%). While both male and female “Adolescent-Limiteds” exhibited about the same rates of delinquency at wave 1, males had noticeable higher levels of delinquency at waves 2 and 3. However, by wave 4 both male and female “Adolescent-Limiteds” among those exposed to violence in childhood had practically seized offending. Among females exposed to violence, there was a third latent class, termed “Early-Adulthood Limiteds” (10.7%). This group steadily offended, but at fairly low levels, throughout the first two waves. Interestingly, this group’s offending peaked at wave 3 (in early adulthood), way surpassing the offending levels of “Abstainers” and “Adolescent-Limiteds.” However, even this group virtually stopped offending by mean age of 27 (i.e., wave 4). Finally, three distinct latent classes were discovered among females not exposed to violence in childhood. Similar to all other subsamples, recognizable groups of “Abstainers” (83.8%) and “Adolescent-Limiteds” (13.3%) could be identified. However, both of these groups exhibited considerably lower levels of offending in all four waves than rest of the subsamples. However, incredibly interestingly, a third very distinct latent class was discovered among this subsample called “Adult-Onsetters” (2.9%). While this group offended only at slightly higher levels throughout the first three waves than “Abstainers,” it displayed a sharp rise in its’ offending at wave 4. When contrasted overall (i.e., males vs. females; violence exposure in childhood vs. no violence exposure in childhood), the males exposed to violence in childhood had the highest levels of delinquency of the entire sample across all latent classes and throughout all four waves. Females who had been exposed to violence in childhood had the second highest levels of

Delinquent Effects of Childhood Exposure to Violent Victimization

275

delinquency across all latent classes and throughout all four waves. Indeed, females and males exposed to violence in childhood resembled each other more in delinquent trajectories than did males who had and had not been exposed to violence in childhood. However, practically all females exposed to violence in childhood seized all offending by age 27 while both male subsamples exhibited a latent class of individuals that continued offending at mean age of 27 (males exposed to violence offended at higher levels). Perhaps not surprisingly, females not exposed to violence in childhood exhibited by far the lowest overall levels of delinquency across all latent classes with one interesting exception. The “Adult-Onsetters” offended almost at same rates at wave 4 among females not exposed to violence than did “Life-Course Persisters” among males exposed to violence at wave 4. However, the “Adult-Onsetter” group among female non-victims was much smaller than the “Life-Course Persistent” group among male victims.

DISCUSSION Key Findings and Policy Implications There has been a large body of research examining the personal, social, and behavioral impacts of children’s exposure to and experience of violent victimization. Much of this research has focused on the relationship between child maltreatment and later delinquent, criminal, and violent outcomes (see Horwitz et al., 2001; Widom, 1989; Widom et al., 2007). Findings reported here are largely consistent with previous scholarly work. Not surprisingly, the present study discovered that the males exposed to violence in childhood had the highest levels of delinquency across all different trajectories of offending over time. Moreover, a much larger proportion of these males (almost 12%) continued offending at steady levels into their late 20s (i.e., assumed to be life-course persistent offenders) when compared to other males who had not been exposed to violence in childhood or to all females. This is also a lot larger percentage of life-course persistent offenders than empirical literature generally reports in unstratified samples (i.e., when looking at the population in general). Previous longitudinal examinations have found that life-course persistent offenders in general population make up a very small percentage, most likely less than 5% (see Moffitt, 1993, 2005). Thus, the current study highlights that while not all males exposed to violence in childhood continue persistent offending,

276

HENRIIKKA WEIR AND CATHERINE KAUKINEN

violent victimization in childhood certainly appears to be a risk factor for life-course persistent offending among males. Criminological literature has consistently reported that females engage in less delinquent and adult criminal behavior than males. Researchers have contributed this gender gap to lower rates of aggression and violence among females, better impulse control, and quicker neuropsychological maturation (Lauritsen, Heimer, & Lynch, 2009; Moffitt, 1993; Steffensmeier & Allan, 1996). However, the present study discovered that the females who had experienced violent victimization in childhood exhibited higher or similar levels of offending across all three offending trajectories when compared to males who had not been victimized in childhood. Indeed, when the adolescent-limited groups were compared, the victimized females offended at lot higher rates than the non-victimized males. This finding is totally contradictory to a vast array of literature on gender and crime and suggests that childhood violent victimization is a major risk factor for the increased adolescent-limited offending for females. Not surprisingly, and in agreement with a large body of empirical literature, females who had not experienced violent victimization in childhood had by far the lowest levels of offending across all trajectories over time with one notable exception. Approximately 3% of non-victimized females, who had practically not engaged in any previous offending, displayed almost as high rates of offending in their late 20s as did the life-course persistent male offenders who had been violently victimized in childhood. Explanations for this extremely surprising finding can only be speculated at this point. Perhaps this is a group of females that has been slow to mature and/or has lived shelter lives and are so to speak “late bloomers” in rebellious behaviors typically seen during teenage years. In summary, key findings from the present study suggest that violent victimization in childhood bridges the gender gap in delinquency between males and females. Thus, childhood violent victimization, and the fact that females are victimized by parents/caregivers and romantic partners at higher rates than males, might be partially responsible in explaining the narrowing of the gender gap between male and female offending in the recent decades. Yet, childhood violent victimization also seems to impact males and females somewhat different ways. Practically all female victims stop offending by their late 20s whereas a fairly large proportion of males exposed to violent victimization in childhood steadily continue offending. Of course, the female victims could be suffering from multitude of other negative life outcomes such as depression, anxiety, poor relationship quality, and so forth. However, in terms of externalizing behaviors, male

Delinquent Effects of Childhood Exposure to Violent Victimization

277

victims appear to be a lot higher risk for long-term negative outcomes. Thus, while the early prevention would be the most desirable option for both genders for the most optimal outcome, the retrospective intervention and treatment programs should be gender-specific. For males, they should heavily focus on providing alternative ways to cope with anger, impulse control and frustration, as well as teach empathy, cognitive problem solving skills, verbal communication skills, and tangible life and job skills (Belknap & Holsinger, 2006; Boney-McCoy & Finkelhor, 1995). For females, most successful intervention and treatment programs may focus on helping the girls through a transition from adolescence to adulthood while providing mental health, medical, and family support services (Belknap & Holsinger, 2006; Boney-McCoy & Finkelhor, 1995). Of course, for victims of both gender, many areas of intervention and treatment should also overlap to include family integration, potential substance abuse treatment, counseling services, medication therapy as needed, and spiritual services (Belknap & Holsinger, 2006; Boney-McCoy & Finkelhor, 1995).

Limitations and Future Research Findings of the present study identify childhood exposure to violent victimization as an important risk factor for delinquency and subsequent adult criminality. At the same time, it will be important for researchers to continue to explore the impact of unique types of violent victimization on both male and female adolescent and early adulthood development. It may be that experiences of child maltreatment and dating violence victimization may be particularly impactful for females and that experiences with these types of victimization likely shape relationship decision-making and other experiences that place them at risk for later involvement in violence as both victims and perpetrators. Researchers will also need to continue to explore how young men’s experience with peer and street violence shapes selfesteem, anger and aggression, and conflict styles within their later involvement in delinquency and violent offending. While scholars have begun to develop a solid literature on how childhood experiences shape later life-course outcomes, this empirical work and the present analyses are not without limitations. Although findings from the current study add to the literature on the connection between exposure and experience to violence and later delinquency and adult criminal outcomes, data limitations temper the conclusions. First, although this study was able to identify the diverse impacts of violence exposure on

278

HENRIIKKA WEIR AND CATHERINE KAUKINEN

engagement in subsequent delinquency, it did not examine the unique contributions of each type of violence on adolescent outcomes. It also did not estimate the chronicity of exposure to each of these types of violent victimization. Further, the present study was not able to measure all types of violence experiences in childhood, such as exposure to parents’ or caregivers’ intimate partner violence. Finally, exposure to victimization (other than childhood maltreatment) was really measured during adolescence versus childhood since the participants were already 14 at wave 1 and were asked about their experiences during the past 12 months. The experiences of childhood exposure to violent victimization have important implications for adolescent development and the transition to adulthood. To better understand these implications, future research should investigate how different types of childhood victimizations may or may not differently impact the longitudinal delinquent and subsequent adult criminal outcomes. Future research should also explore how different types of victimization experiences may differently impact males and females regarding delinquency and subsequent adult criminality. Further, future research should examine what different factors may protect or, alternatively, place childhood victims of violence at even higher risk for delinquency. Finally, the extremely interesting finding regarding adult-onset female offenders among non-victimized females should be further explored. The above suggestions could be achieved via making all groups to be compared statistically equivalent to each other via propensity score matching prior to conducting trajectory analyses and/or by investigating which covariates predict the probability of an individual in belonging to any given trajectory group (i.e., high- or low-delinquency). While advances in statistical techniques and increasingly rigorous research designs have certainly advanced the scholars’ understanding regarding childhood violent victimization and its impact on delinquency and crime, we still have a long way to go in order to fully detangle the complex connections between childhood violent victimization, delinquency, and many other potential interacting factors.

NOTES 1. The age range was limited due to well-established relationship between age and crime (i.e., the age-crime curve), which is one of the most robust findings in the criminological literature (Blumstein, Cohen, & Farrington, 1988a, 1988b; Farrington, 1986; Moffitt, 1993; Piquero, Farrington, & Blumstein, 2003). 2. The overall delinquency measure included the following individual items, all pertaining the last 12 months: “How often did you: Damage property that did not

Delinquent Effects of Childhood Exposure to Violent Victimization

279

belong to you?/Steal something worth less than $50?/Steal something worth more than $50?/Go into a house or a building to steal something?/Use or threatened to use weapon to get something from somebody?/Sell drugs?/Take a part in a physical fight where a group of your friends were against another group?/Get into a serious physical fight?/Hurt somebody badly enough that she/he needed medical care?/ Stabbed or shot someone?/Carried weapon or pulled a weapon on someone?” Additionally, a miscellaneous 12th delinquency item varying from wave to wave was included in the overall measure. 3. The number of items used to create variety scales of delinquency in each of these studies varied. 4. Transforming original frequency scales of each individual delinquency item to dichotomous measures limited the total range of delinquency/adult offending scale from 0 to 12 versus from 0 to extremely high number of offense counts (often same kind). Using a variety scale of delinquency versus frequency scale when including multiple items of offending across several waves of data collection attempts to encounter the issues brought on by extremely high zero counts, outliers, and skewed distribution. 5. Street and Peer Violent Victimization as well as Dating Violence were measured at all four waves in Add Health. During each interview, participants were asked about their experiences with these types of victimizations during the past 12 months. However, experiences with Childhood Maltreatment were only included at wave four, at which point the participants were asked about their experiences PRIOR to age 18 rather than during the last 12 months as they now were adults. 6. Exposure to violence in its many different forms overlap among children (See Cuevas, Finkelhor, Shattuck, Turner, & Hamby, 2013 for a full discussion). Finkelhor et al. (2009), using data from the National Survey of Children’s Exposure to Violence (NatSCEV), have clearly demonstrated that children who were exposed to one type of violence, were at far greater risk of experiencing other types of violence. Among children who noted exposure to one type of violence, fully 65% noted exposure to another type of violence. There are examples in other literatures on the impact of violence among children. For example, children exposed to child maltreatment are also at much higher risk for later dating violence (See Gover et al., 2008; Kendra, Bell, & Guimond, 2012; Leisring, 2013; Milletich et al., 2010). Other work (Duncan, 1999) has shown that there are large overlaps in the experience of child maltreatment and abuse by peers. This work found that those who were victims of peer violence had experienced higher rates of both emotional and physical maltreatment by parents two to three times higher than those who were not childhood victims of peer abuse. 7. Not included in the present paper but available upon request from corresponding author. 8. Smaller BICs typically indicate the better fitting model. However, with LLCA, BIC usually increases as the latent classes get higher. Thus, it is vital to consider many other factors in model decision-making.

280

HENRIIKKA WEIR AND CATHERINE KAUKINEN

ACKNOWLEDGMENTS This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health website (http://www.cpc.unc.edu/addhealth). No direct support was received from grant P01-HD31921 for this analysis.

REFERENCES Aizer, A. (2008). Neighborhood violence and urban youth. National Bureau of Economic Research Working Paper No. 13773. Allwood, M. A., & Widom, C. S. (2013). Child abuse and neglect, developmental role attainment, and adult arrests. Journal of Research in Crime and Delinquency, 50(4), 551 578. Asscher, J. J., Van der Put, C. E., & Stams, G. J. J. (2015). Gender differences in the impact of abuse and neglect victimization on adolescent offending behavior. Journal of Family Violence, 30(2), 215 225. Belknap, J., & Holsinger, K. (2006). The gendered nature of risk factors for delinquency. Feminist Criminology, 1(1), 48 71. Belknap, J., Holsinger, K., & Dunn, M. (1997). Understanding incarcerated girls: The results of a focus group study. Prison Journal, 77, 381 404. Blumstein, A., Cohen, J., & Farrington, D. P. (1988a). Criminal career research: Its value for criminology. Criminology, 26(1), 1 35. Blumstein, A., Cohen, J., & Farrington, D. P. (1988b). Longitudinal and criminal career research: Further clarifications. Criminology, 26(1), 57 74. Boney-McCoy, S., & Finkelhor, D. (1995). Psychosocial sequelae of violent victimization in a national youth sample. Journal of Consulting and Clinical Psychology, 63(5), 726 736. Chesney-Lind, M. (1997). The female offender: Girls, women, and crime. Thousand Oaks, CA: Sage. Chesney-Lind, M., & Rodriguez, N. (1983). Women under lock and key. Prison Journal, 63, 47 65. Child Welfare Information Gateway. (2014). Child maltreatment 2012: Summary of key findings. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau. Connolly, J., Friedlander, L., Pepler, D., Craig, W., & Laporte, L. (2010). The ecology of adolescent dating aggression: Attitudes, relationships, media use, and socio-demographic risk factors. Journal of Aggression, Maltreatment & Trauma, 19(5), 469 491.

Delinquent Effects of Childhood Exposure to Violent Victimization

281

Copeland-Linder, N., Johnson, S. B., Haynie, D. L., Chung, S. E., & Cheng, T. L. (2012). Retaliatory attitudes and violent behaviors among assault-injured youth. Journal of Adolescent Health, 50(3), 215 220. Crittenden, C. A., Wright, E. M., & Fagan, A. A. (2011). The effects of exposure to intimate partner violence on children’s development. Family and Intimate Partner Violence Quarterly, 4(2), 113 123. Cuevas, C. A., Finkelhor, D., Shattuck, A., Turner, H., & Hamby, S. (2013). Children’s exposure to violence and the intersection between delinquency and victimization. U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. NCJ 240555. Cullen, F. T., & Agnew, R. (2010). Criminological theory (4th ed.). Oxford: Oxford University Press. Daly, K. (1992). A women’s pathway to felony court. Review of Law and Women’s Studies, 2, 11 52. Davis, T., Ammons, C., Dahl, A., & Kliewer, W. (2015). Community violence exposure and callous unemotional traits in adolescents: Testing parental support as a promotive versus protective factor. Personality and Individual Differences, 77, 7 12. Department of Justice. (2009). Retrieved from http://www.justice.gov/defendingchildhood/ facts-about-children-and-violence Doumas, D., Margolin, G., & John, R. S. (1994). The intergenerational transmission of aggression across three generations. Journal of Family Violence, 9(2), 157 175. Duncan, R. D. (1999). Maltreatment by parents and peers: The relationship between child abuse, bully victimization, and psychological distress. Child Maltreatment, 4(1), 45 55. Emery, C. R. (2006). Consequences of childhood exposure to intimate partner violence. Final Technical Report. NCJ 215347, Washington, DC: United States Department of Justice. Farrington, D. P. (1986). Age and crime. Crime and Justice, 7, 189 250. Feldman, B. J., Masyn, K. E., & Conger, R. D. (2009). New approaches to studying problem behaviors: A comparison of methods for modeling longitudinal, categorical adolescent drinking data. Developmental Psychology, 45, 652 676. Finkelhor, D. (2008). Childhood victimization: Violence, crime, and abuse in the lives of young people. New York, NY: Oxford University Press. Finkelhor, D., Ormrod, R. K., & Turner, H. A. (2007). Revictimization patterns in a national longitudinal sample of children and youth. Child Abuse & Neglect, 31(5), 479 502. Finkelhor, D., Turner, H., Ormrod, R., Hamby, S., & Kracke, K. (2009). Children’s exposure to violence: A comprehensive National Survey. U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. NCJ 227744. Finkelhor, D., Turner, H., Ormrod, R., Hamby, S., & Kracke, K. (2009). National survey of children’s exposure to violence. Juvenile Justice Bulletin, 1 11. Finkelhor, D., Turner, H. A., Shattuck, A., & Hamby, S. L. (2013). Violence, crime, and abuse exposure in a national sample of children and youth: An update. JAMA Pediatrics, 167(7), 614 621. Foshee, V., Linder, F., MacDougall, J., & Bangdiwala, S. (2001). Gender differences in the longitudinal predictors of adolescent dating violence. Preventive Medicine, 32, 128 141. Gover, A. R., Kaukinen, C., & Fox, K. A. (2008). The relationship between violence in the family of origin and dating violence among college students. Journal of Interpersonal Violence, 23, 1667 1693.

282

HENRIIKKA WEIR AND CATHERINE KAUKINEN

Grogger, J. (1998). Local violence and educational attainment. Journal of Human Resources, 32, 659 682. Haynie, D. L. (2001). Delinquent peers revisited: Does network structure matter? American Journal of Sociology, 106, 1013 1057. Herrenkohl, T. I., Kosterman, R., Mason, W. A., & Hawkins, J. D. (2007). Youth violence trajectories and proximal characteristics of intimate partner violence. Violence and Victims, 22(3), 259 274. Holt, S., Buckley, H., & Whelan, S. (2008). The impact of exposure to domestic violence on children and young people: A review of the literature. Child Abuse & Neglect, 32(8), 797 810. Horwitz, A. V., Widom, C. S., McLaughlin, J., & White, H. R. (2001). The impact of childhood abuse and neglect on adult mental health: A prospective study. Journal of Health and Social Behavior, 184 201. Hubbard, D. J., & Pratt, T. C. (2002). A meta-analysis of the predictors of delinquency among girls. Journal of Offender Rehabilitation, 34(3), 1 13. Jouriles, E. N., & McDonald, R. (2015). Intimate partner violence, coercive control, and child adjustment problems. Journal of Interpersonal Violence, 30(3), 459 474. Kaukinen, C., Buchanan, L., & Gover, A. R. (2015). Child abuse and the experience of violence in college dating relationships: Examining the moderating effect of gender and race. Journal of Family Violence, (May), 1 14. Kaura, S. A., & Lohman, B. J. (2007). Dating violence victimization, relationship satisfaction, mental health problems, and acceptability of violence: A comparison of men and women. Journal of Family Violence, 22(6), 367 381. Kelly, S., Anderson, D., Hall, L., Peden, A., & Cerel, J. (2012). The effects of exposure to gang violence on adolescent boys’ mental health. Issues in Mental Health Nursing, 33(2), 80 88. Kendra, R., Bell, K. M., & Guimond, J. M. (2012). The impact of child abuse history, PTSD symptoms, and anger arousal on dating violence perpetration among college women. Journal of Family Violence, 27, 165 175. Kliewer, W., & Lepore, S. J. (2015). Exposure to violence, social cognitive processing, and sleep problems in urban adolescents. Journal of Youth and Adolescence, 44(2), 507 517. Lauritsen, J. L., Heimer, K., & Lynch, J. P. (2009). Trends in the gender gap in violent offending: New evidence from the national crime victimization survey. Criminology, 47(2), 361 399. Leisring, P. A. (2013). Physical and emotional abuse in romantic relationships motivation for perpetration among college women. Journal of Interpersonal Violence, 28, 1437 1454. Maas, C., Herrenkohl, T., & Sousa, C. (2008). Review of research on child maltreatment and violence in youth. Trauma, Violence, & Abuse, 9, 56 67. Marshall, L. L., & Rose, P. (1988). Family of origin violence and courtship abuse. Journal of Counseling & Development, 66(9), 414 418. Milletich, R. J., Kelley, M. L., Doane, A. N., & Pearson, M. R. (2010). Exposure to interparental violence and childhood physical and emotional abuse as related to physical aggression in undergraduate dating relationships. Journal of Family Violence, 25, 627 637. Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674 701.

Delinquent Effects of Childhood Exposure to Violent Victimization

283

Moffitt, T. E. (2005). The new look of behavioral genetics in developmental psychopathology: Gene-environment interplay in antisocial behaviors. Psychological Bulletin, 131(4), 533. Monahan, K. C., King, K. M., Shulman, E. P., Cauffman, E., & Chassin, L. (2015). The effects of violence exposure on the development of impulse control and future orientation across adolescence and early adulthood: Time-specific and generalized effects in a sample of juvenile offenders. Development and Psychopathology, 1 18. Nagin, D. S. (2005). Group-based modeling of development. Cambridge, MA: Harvard University Press. Nagin, D. S., & Land, K. C. (1993). Age, criminal careers, and population heterogeneity: Specification and estimation of a nonparametric, mixed Poisson model. Criminology, 31(3), 327 362. Nagin, D. S., & Piquero, A. R. (2010). Using the group-based trajectory model to study crime over the life course. Journal of Criminal Justice Education, 21, 105 116. Nylund, K. L., & Masyn, K. E. (2008). Covariates and latent class analysis: Results of a simulation study. Paper presented at the Society for Prevention Research Annual Meeting. Orpinas, P., Nahapetyan, L., Song, X., McNicholas, C., & Reeves, P. M. (2012). Psychological dating violence perpetration and victimization: Trajectories from middle to high school. Aggressive Behavior, 38(6), 510 520. Ozer, E. J., Tschann, J. M., Pasch, L. A., & Flores, E. (2004). Violence perpetration across peer and partner relationships: Co-occurrence and longitudinal patterns among adolescents. Journal of Adolescent Health, 34(1), 64 71. Piquero, A. R., Farrington, D. P., & Blumstein, A. (2003). The criminal career paradigm. Crime and Justice, 30, 359 506. Porter, L., & Vogel, M. (2014). Residential mobility and delinquency revisited: Causation or selection? Journal of Quantitative Criminology, 30, 187 214. Rees, C., & Pogarsky, G. (2011). One bad apple may not spoil the whole bunch: Best friends and adolescent delinquency. Journal of Quantitative Criminology, 27, 197 223. Rossman, R., Hughes, H., & Rosenberg, S. (2000). Exposure to interparental violence: Exposed children as psychologically abused. Children and Interparental Violence: The Impact of Exposure (pp. 5 6). New York, NY: Taylor & Francis. Rutter, L. A., Weatherill, R. P., Taft, C. T., & Orazem, R. J. (2012). Examining gender differences in the relationship between dating violence victimization and anger in college students. Violence and Victims, 27, 70 77. Schwab-Stone, M. E., Ayers, T. S., Kasprow, W., Voyce, C., Barone, C., Shriver, T., & Weissberg, R. P. (1995). No safe haven: A study of violence exposure in an urban community. Journal of the American Academy of Child & Adolescent Psychiatry, 34(10), 1343 1352. Silverman, A. B., Reinherz, H. Z., & Giaconia, R. M. (1996). The long-term sequelae of child and adolescent abuse: A longitudinal community study. Child Abuse & Neglect, 20(8), 709 723. Steffensmeier, D., & Allan, E. (1996). Gender and crime: Toward a gendered theory of female offending. Annual Review of Sociology, 459 487. Stith, S. M., Rosen, K. H., Middleton, K. A., Busch, A. L., Lundeberg, K., & Carlton, R. P. (2000). The intergenerational transmission of spouse abuse: A meta-analysis. Journal of Marriage and Family, 62(3), 640 654. Stouthamer-Loeber, M., Loeber, R., Homish, L., & Wei, E. (2001). Maltreatment of boys and the development of disruptive and delinquent behavior. Development and Psychopathology, 13, 941 955.

284

HENRIIKKA WEIR AND CATHERINE KAUKINEN

Straight, E. S., Harper, F. W., & Arias, I. (2003). The impact of partner psychological abuse on health behaviors and health status in college women. Journal of Interpersonal Violence, 18(9), 1035 1054. Straus, M. A. (2004). Prevalence of violence against dating partners by male and female university students worldwide. Violence Against Women, 10(7), 790 811. Sullivan, M., Egan, M., & Gooch, M. (2004). Conjoint interventions for adult victims and children of domestic violence: A program evaluation. Research on Social Work Practice, 14(3), 163 170. Swahn, M. H., Bossarte, R. M., Palmier, J. B., Yao, H., & Van Dulmen, M. H. (2013). Risk factors for multiple forms of violent experiences: Analyses of the 2009 youth risk behavior survey. Vulnerable Children and Youth Studies: An International Interdisciplinary Journal for Research, Policy and Care, 8, 225 236. Udry, J. R. (2003). The national longitudinal study of adolescent health add health, waves I and II, 1994 1996; wave III, 2001 2002 [machine readable data file and documentation]. Chapel Hill, NC: Carolina Population Center, University of North Caroline at Chapel Hill. Widom, C. S. (1989). The cycle of violence. Science, 244(4901), 160 166. Widom, C. S., & Maxfield, M. G. (2001). An update on the “Cycle of Violence.” National Institute of Justice Research in Brief. NCJ 184894. Widom, C. S., White, H. R., Czaja, S. J., & Marmorstein, N. R. (2007). Long-term effects of child abuse and neglect on alcohol use and excessive drinking in middle adulthood. Journal of Studies on Alcohol and Drugs, 68(3), 317. Wolfe, D. A., Crooks, C., Jaffe, P., Chiodo, D., Hughes, R., Ellis, W., & Donner, A. (2009). A school-based program to prevent adolescent dating violence: A cluster randomized trial. Archives of Pediatrics & Adolescent Medicine, 163(8), 692 699. Wolke, D., Copeland, W. E., Angold, A., & Costello, E. J. (2013). Impact of bullying in childhood on adult health, wealth, crime, and social outcomes. Psychological Science, 24(10), 1958 1970. Zurbriggen, E. L. (2009). Understanding and preventing adolescent dating violence: The importance of developmental, sociocultural, and gendered perspectives. Psychology of Women Quarterly, 33(1), 30 33.

GIRLS OFFENDERS PATHWAYS INTO THE SPANISH JUVENILE JUSTICE Rosario Pozo Gordaliza ABSTRACT Purpose The study presented is a starting point in Spain about the invisible phenomenon of female delinquency behavior and juvenile justice. Studying girls who break laws certainly provides insight into the standards, practices, and social customs affecting young women in a particular time and space, as well as providing clues about the expectations of their gender. Methodology/approach This research is being based on sociobiographic interviews and the analysis of individual life histories (totaling 16). However, in order to increase the validity and credibility of the information collected using this method, further complimentary methods of collecting data were also employed, leading to a triangulation of methods. This consisted of the analysis of social and judicial case files/dossiers (44) and first-hand observation undertaken during one-month stay in a female juvenile reform institution for young women convicted of committing crimes between the age of 14 and 18. The girls could, however, remain at the institution to the age of 21.

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 285 309 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009013

285

286

ROSARIO POZO GORDALIZA

Findings Girls are often the victims of what might be called multiple situation of marginality in that their gender race and class have placed them at the economic periphery of society. Understanding their lives and choices of girls who find themselves in the juvenile criminal justice systems also requires a broader understanding of the contexts and process within which their criminal behavior is lodged. Originality/value This research has looked closely at the existing theory on women and crime, as well as the forms of participation, the processes, factors and contexts of social exclusion, and the role of women offenders in the Spanish Juvenile Justice. Keywords: Girls; centers of detention; ethnography; crime

INTRODUCTION According to the definition of West and Zimmerman (1987), gender is the way people act in accordance with established behaviors in the light of normal concepts, attitudes, and activities appropriate to a sexual category. Gender activities emerge from the sex and define the membership of a sexual category. Gender is not a set of features, not a variable, not a role but the product of social action of some kind, which is constructed through interaction. In general when analyzing the differences between men and women you have to consider the control, identity, the meanings of actions, emotions and patterns as these make the distinction between men and women (male and female). Gender is relevant to explaining how girls and boys engage in delinquency trajectories, developmental processes, problem behaviors, troubling responses, etc. (Acoca, Countryman-Roswurm, Kaba, & Winn, 2015; Haney, 2010). Young offenders females have been ignored in the past and poorly researched as a social category (Chesney-Lind & Eliason, 2006; Zahn, 2009) especially in Spain. Any research that has been carried out has focused on crime in terms of juvenile sexuality (as prostitution) (Chesney-Lind & Pasko, 2004; Goodkind, Ng, & Sarri, 2006) as opposed to looking at criminal and delinquent behavior in terms of female gender (Haney, 2010; Zahn, Day, Mihalic, & Tichavsky, 2009). This has caused a distortion in the picture of juvenile crime and behavior. Recent developments in theorizing about gender differences in crime even though little is known about the lives of young women in Spain and current

Girls Offenders Pathways into the Spanish Juvenile Justice

287

information is based on theory. The small number of girls in the Spanish juvenile justice is not indicative of female juvenile delinquent behavior being a small problem (81.3% of the minors condemned in Spain were boys and only 18.7% girls, INE, 2013). Instead it reflects the limited investigation into their behavior and lack of access to specific programs and services (Alder & Worrall, 2004). As has happened with the study of adult female behavior in the past, the relatively low amount of Spanish female criminal behavior in meant that the subject was neglected and the situation made worse as a result. It has taken time to recognize that many of the multiple problems presented by young people whilst shared in part, will not be identical and will vary dependent on age, gender, ethnicity, and socioeconomic class and culture (Kempf-Leonard & Pernilla, 2007; O’Neill, 2005). To consider gender, ethnicity, and socioeconomic class as neutral and to be blind to it has, ultimately, had a negative impact on girls and women (Hannah-Moffat & Yule, 2011). The study presented is a starting point in Spain about the invisible phenomenon of female delinquency behavior and juvenile justice. The fundamental research being based on socio-biographic interviews and the analysis of individual life histories. However, in order to increase the validity and credibility of the information collected using this method; further complimentary methods of collecting data were also employed, leading to a triangulation of methods. This consisted of the analysis of social and judicial case files/dossiers and first-hand observation of the young women in the female juvenile reform institution. Forty-four dossiers were selected from the three provinces (and their respective courts) with the highest crime rates in Andalucı´ a. The interviews (totaling 16) and all observational studies were undertaken during a one-month stay in a female juvenile reform institution. The institution chosen is the only single sex juvenile reform center for young women convicted of committing crimes between the age of 14 and 18. The girls could, however, remain at the institution to the age of 21. The information collected on girls in institutions in the Andalucı´ an juvenile justice system. Finally, I have supplemented the research with a study of institutionalized young women and female juvenile delinquency. The research and theory is to be considered intersectoral in that it does not just look at gender but also at the factors of cultural, socioeconomic class, and ethnicity. This research has looked closely at the existing theory on women and crime, as well as the forms of participation of women offenders and, through the development of new research on institutionalized female delinquent youth, has arrived at new theories.

288

ROSARIO POZO GORDALIZA

LITERATURE REVIEW This paper begins with a consideration of boys and girls living in different worlds with different opportunities; a double standard exists in the social structure affecting how men and boys should behave and how women and girls should behave (Haney, 2010). In the same way that they have different ways of living their lives, girls also have different ways of committing crime.1 The strategies developed by young women within the home relate to those they display in the street, neither are exempt from the patriarchal culture that relegates girls to an inferior position and many of the behaviors that lead to their victimization are responses to this social exclusion (Hannah-Moffat, 2000). This situation is exacerbated among the most culturally deprived socioeconomic classes where there is a clear differentiation of roles. In addition to understanding the low profile of female criminality it is necessary to understand the social controls exerted on women in these contexts (Heidensohn, 2000). The man is portrayed as the creator or producer and has a public role, while the woman’s role is as a player or part, represented only in the private sphere. It is the informal controls exerted on women that mean they are relegated to the private sphere. Indeed, studies about the control exercised by educational institutions or the family, show that these have a greater effect on girls than on boys (Chesney-Lind & Eliason, 2006; Kempf-Leonard & Pernilla, 2007). Not only must the hierarchy of sex be taken into account, but also age, race and social class (Girlguiding, 2007; Moore & Padavic, 2010). The difference in gender roles exists as a structural difference (within society and the family) and as an ideological difference informing the girls’ beliefs about themselves and about their power and legitimacy in the patriarchal order. Childhood abuses and trauma experience is a significant factor risks for subsequent delinquency behavior. Recent studies on feminist pathways approach delinquent girls and incarcerated women report abuse victimizations much higher than the general population of women and girls (Chesney-Lind & Pasko, 2013). We know from empirical studies (Loucks, Malloch, McIvor, & Gelsthorpe, 2006) that the vast majority of the female population rarely enters contact with criminal justice agencies. Those that do are generally girls from poor backgrounds and ethnic minorities; these women in the juvenile and criminal justice systems have little in common with most women in mainstream society (Smith, 2002). As pointed out in (Worrall, 2001), female juvenile delinquents and criminals have been defined differently than boys; as the author has

Girls Offenders Pathways into the Spanish Juvenile Justice

289

acknowledged, female juveniles belong to a category that until now has been assessed as too small and low risk, to pay attention to. In a similar way, some authors have argued that policies and practices of intervention are a concern and risk, including what has been identified as “hybridization” (risk and need) in the discourse of criminal justice resulting in a change in the target classification/definition of delinquent girls and young women. This was first defined as women with “vulnerability,” “needs,” or those “at risk” requiring the intervention of the criminal justice system or the attention of criminology (Marutto & Hannah-Moffat, 2005). That is why there is now an urgent need to look at girls and the unique and distinct problems that they bring to the genre, taking into account the influence of age, ethnicity, and socioeconomic class/cultural, as these factors have different impacts, especially at an early age and in the performance of more traditional roles (Lee et al., 2008; Renzetti, Curran, & Maier, 2012). One of the most important discoveries of recent years in criminology is the way that crime differs dependent on gender (Godfrey, 2004; Malloch, 2004), motivated by the idea that you can use theories to explain male and female criminal behavior. Although boys and girls share some of the same pressures in their lives and sometimes the same access to learning of delinquent behavior the reality is that the reason why girls and boys become criminals and the paths that lead them to juvenile reform institutions are quite different (Estrada & Nilsson, 2012). In particular, scholars in the field have identified and documented how, in addition to stories of tremendously unfair lives, girls have been condemned from an early age as “predelinquent” or badly behaved (Smith, 2002). Although research into the differences in development among young men and women is a relatively new area of study in criminology, the information available so far allows you to see that gender differences in socialization and development do exist, and that these differences may, in fact, have an effect on patterns of crime (Alder & Worrall, 2004; Belknap, 2001). Women who disobey authority, who run away from home, who are sexually active, or who have become pregnant against the wishes of husbands or fathers and/or have become inadequate mothers are more vulnerable to formal controls exercised by the State through punishment or psychiatric labeling (MacDonald & Chesney-Lind, 2001). Accordingly, they are more likely to be imprisoned, segregated, or subject to formal checks by the authorities (Chesney-Lind & Shelden, 2013). When it comes to quantifying female juvenile delinquency and crime, there are a number of difficulties to face. The various available pieces of data represent different aspects of reported cases and are often inconsistent

290

ROSARIO POZO GORDALIZA

in terms of how cases are interpreted and how young offenders are classified. The result is that today even in countries that have greater development in the quantitative study of crime the data it is still rough and, in any case, focuses on specific times and geographical spaces. The quantitative strategy is extremely necessary, but unfortunately today is inadequate in Spain and we have no good data to rely on.2 In addition, there is conflict between the official crime data and data collected through independent surveys; the latter confirming that female crime is more common than official statistics show and that there are certain gender biases (and ethnicity biases)3 operating in juvenile justice, affecting the volume of female juvenile delinquency (Moore & Padavic, 2010). It looks like the numerical scarcity of institutionalized female juvenile delinquency is the result of different social projections of role of women. The most common crimes are committed by men and women from social groups without power; these groups do not reflect the old accepted roles of women but represent social inferiority and poverty (Watson & Edelman, 2012). In short, lack of power and opportunities for young people from the social margins (Burman, Batchelor, & Brown, 2001; Gottfredson & Kruttschnitt, 2008).

RESEARCH DESIGN AND METHODOLOGICAL ISSUES I gradually constructed, deconstructed, and reconstructed the objective of my study over the course of the research process. The design chosen led to the discovery of specific contexts and meanings that facilitated understanding of the phenomenon of female juvenile delinquency which reaches the institutionalization stage. For this purpose, the phenomenon was approached through methodological triangulation based on institutional ethnography that enabled the selection of the young women for in-depth interviews and life stories. Judicial and social records of 44 young women in three Andalusian provinces were analyzed. For the qualitative interviews, I maintained a flexible and open approach to selection, which enabled me to explore beforehand the context and the most appropriate subjects for the research before deciding on the final sample. The study was conducted with subjects that have been or are, to varying degrees, young women aged 14 21 with a history of institutionalization. To obtain the sample for the life stories and interviews, availability and willingness to cooperate, together with duly completed and signed permissions were taken into account. Delinquent young women with a record

Girls Offenders Pathways into the Spanish Juvenile Justice

291

of detention in Andalusia were selected. Sixteen detailed life stories were collected in an average of two or three sessions each. For the observation, the only female reform institution in Andalusia was chosen, and the observation period lasted a month with a full schedule from nine in the morning until 11 at night. All the information gathered was carefully recorded in a field notebook. In the case of the documentary analysis, 44 dossiers (with personal information judicial, social, police, institutional, etc.) from three provinces in Andalusia were selected. The criteria taken into account when choosing the dossiers were that the young women should be between 14 and 20 years with a recent institutionalization record between 2004 and 2008. However, selection of the provinces was determined by the representativeness of their female juvenile delinquency: the three provinces with the highest proportion in recent years were selected. Finally, we must point out that samples of judicial files were taken from the dossiers available in each court up to the point of theoretical saturation. The data collection period had an intermediate phase of preparations which included inspection, the contact phase, negotiations and contracts, the rationale for the meetings, introducing myself as a researcher, and dealing with the suffering in the stories. Each contact before the recruitment as well as meetings with each informant, were carefully prepared. This served as support for the narratives of the participants; it also contains information about the production of the stories which it is important to consider when the information that emerges is being analyzed. The procedures followed when conducting interviews (number of interviews, venue, time of interview, etc.), were extremely flexible and open, and were adapted to the circumstances of each young woman and her life story. Adjustments were also made according to the availability, time, and space of the employees or the institution, depending on particular situations that arose, but always mindful of their consistency with the objective and the general orientation of the research. The meetings followed this norm. First, we spent time in the girls’ institutional environment to make a preliminary connection. Secondly, and once we found a comfortable place, we had the first meeting and we opened the narrative by asking a general question. Then there was a second meeting in which probing questions were introduced. Finally, a meeting took place to conclude both the telling of the story and the listening relationship established between narrator and interlocutor and to evaluate the process of participation in the research. Everything was taped with two digital recorders, and always with the prior permission of the narrators. With regard to the transcription of the interviews, it must be noted that

292

ROSARIO POZO GORDALIZA

the stories were transcribed in full, preserving as much as possible all their richness: language mistakes, lapses, hesitations, silences, pauses, or interruptions.

Analysis and Interpretation of Data The basic research tool is the interpretative power of the study. For this we used various theories and analytical methods to cover the information from a type of discursive qualitative material in greater depth and fullness. Specifically, the analysis of the qualitative data consisted of the conceptual organization of information produced in the course of the empirical study based on significant categories, with priority being given to the content and interpretation of these categories to the detriment of the frequencies of the codes. The conditions of production have been included in this analysis by using descriptive data in the field notebook used in the process of collecting stories from the interviews. To carry out the analysis I followed these steps: Step I. Immersion in the material collected; Step II. Classification of the data; Step III. Coding and reorganization of the data; Step IV. Analysis and interpretation. The sociobiographical interviews were analyzed on three levels: a diachronic reading of the events narrated (identification of life trajectories), a thematic and synchronic reading of the issues related to the individual characteristics and life cycle of the young female offenders (methods of deciding on strategies and trajectories from the perspective of the characteristics of the young women) and a reading that takes into account the impact of structures (social, economic, political, institutional) on the development of the individual’s life: influence of the “context of reception” on the expectations and the strategies used by young female offenders. As we have seen, with respect to the principles of the analysis, priority was given to the uniqueness and particularity of each story narrated, and an attempt was made to be faithful to one of the fundamental premises of the biographical approach. In this regard, a single, intra-case logic was followed, whereby each story told was analyzed and studied in depth. The idea was to be able to obtain a reconstructed story from the analysis of listening to the story and of the main biographical milestones that constituted the life of the narrator. Secondly, a transversal inter-case logic was adopted, which, based on certain continuities and discontinuities in the single phase, facilitates identification of relevant thematic and analytical focal points as well as comprehensive transversal hypotheses to address the

Girls Offenders Pathways into the Spanish Juvenile Justice

293

phenomenon being studied. With these thematic and analytical axes as a point of departure, all the stories were reanalyzed from the perspective of gender mainstreaming that runs through them generally. Regarding the methods of analysis, I integrated different methods to cover both the content emerging from the stories (analysis of the content of the discursive material) and the structure of the narrative through which it is expressed (discourse analysis). The procedure for achieving the integrated analysis was supported by the Atlas.ti program for qualitative text analysis. The process involved identifying themes and subthemes, assigning codes to each of them, grouping pieces of text associated with each code, the separate analysis of each thematic module, producing findings, discovering significant relationships and preparing progress reports. The initial analysis was done at the same time as the interviews. Upon completion of the fieldwork, the information obtained was analyzed systematically backed by social theory and using relevant conceptual perspectives for each point.

Ethical Considerations of the Research This study scrupulously observed three ethical principles on which the rules of ethical conduct in research are based: the principles of kindness, respect for human dignity, and justice. For this reason, ethical issues took center stage in the whole process of my research. We must never forget that quality research is one that integrates research, reflection, critique, respect, and action. My study was torn between the rights of the individuals involved and the requirements of scientific rigor, which gave rise to contradictions and created multiple conflicts on various levels. It involved an emotional cost to the girls and young women themselves as well as to the very person constructing life stories characterized by a host of accumulated traumas in the construction of personal identity.

GIRLS’ PATHWAYS INTO THE JUVENILE JUSTICE SYSTEM: CONTEXTS AND PROCESS OF SOCIAL EXCLUSION This research found that young women are more likely to engage in dangerous, abnormal, or illegal activities in neighborhoods where there are concentrations of socially disadvantaged communities and are fewer

294

ROSARIO POZO GORDALIZA

resources to protect them. In understanding the socialization, life and crime of young girls it is important to take into account whether they have come from depressed and marginalized (particularly urban) areas that suffer from multiple problems; this is also an important element taken into account by the authorities when identifying areas that required their attention. It is seen in the analysis that the families of the young women that live within these neighborhoods are more likely to be controlled by formal state mechanisms. For example, from a young age, sometimes from birth, these young women will have been subject to interventions by social services. In addition to the formal interventions and controls, girls may also be subjected to other types of informal control on an informal or domestic level, intended to moderate behavior that is considered inappropriate or disruptive (albeit that the behavior is not necessarily wrong or a crime). Ultimately, and as pointed Worrall (2001), the closer the attention paid by the authorities to the life and activities of a young woman, the higher any ultimate sentence is likely to be. The results of the author’s doctoral thesis contain a series of interrelated elements or factors that are crucial to ascertaining the life stories of the young girls who end up in the juvenile justice system. Fig. 1 explains a large part of the results of this research: The following will explain several of the elements and factors set out above. Many of the young people in the study came from depressed, marginal slums in cities or peripheral areas with a multitude of problems, such as high rates of drug dependence, imprisonment, violence, crime (drug trafficking), unemployment, truancy, poor health, and underdeveloped basic public services (education, health, housing, and infrastructure); they belonged to ethnic minorities and/or were new immigrants with heavy socioeconomic disadvantages, among others. Crimes committed by young women from were related to neighborhood problems (e.g., fighting, theft, robbery, and drug trafficking). A second type of neighborhood can be found in areas near slums, which share the above problems, but to a lesser extent. It should be noted that violent crimes or domestic violence had occurred in the lives of young people living in areas near slums, adopted children and (mostly lower-middle class) young people who had faced frustrating events in their lives. Victimization (defined as the physical, psychological, or emotional neglect, abandonment, sexual abuse and bullying, etc., experienced by the young people in the study) and disproportionate rates of neglect and abuse: this occurred in the lives of all the young people, although it was more pronounced in young people from slums. This victimization was continuous

295

Girls Offenders Pathways into the Spanish Juvenile Justice

MAIN IMPORTANT RESULTS (2011)

FORMAL AND INFORMAL CONTROLS

VERY DEPRIVATED NEIGHBHORHOOD

SOCIALIZATION DRUGS PRISON AND CRIME

VICTIMIZATION

RUN AWAY

FELONIES CRIME AND GIRLS MISBEHAVIOR

PREVENTIVE INSTITUTION OR PREVENTIVE MESURES

POPULAR NEIGHBHORHOOD

CO-HABITATION BOYFRIEND OR “MARÍO”

PREGNANCY MOTHERHOOD ABORTIONS

AESTHETIC, BODY

SUICIDAL BEHAVIOR AND SELFHARM

PRIMARY OR SECONDARY EDUCATION

PROSTITUTION BEGGING HOMELESS

NEIGHBORHOOD

MULTIPROBLE MATIC DYSFUNCTIONAL FAMILY

LOW MIDDLE CLASS GIRLS FRUSTRATING EVENTS

Fig. 1.

MENTAL PROBLEMS

PEER GROUPS/ FAMILY

MINORS IMMIGRANTS

CONSUMPTION OF TOXIC SUBSTANCES

ADOPTED GIRLS

VIOLENCE

MENTAL OR PHYSICAL DISABILITY

The Process of Exclusions and Victimizations and Strategies for Survival (New Vulnerabilities).

and pervaded their life stories. Victimization within the family was a turning point that resulted in running away (on occasion, intermittently, definitively) so that the neighborhood (parks, squares, and streets are important elements of socialization), especially in marginal areas (where ethnicity appeared to be an important factor). This is important to look at these girls not only in terms of their perpetual criminal behavior or bad behavior, but also in terms of their experiences or victimization. The conditions that result from this physical and psychological victimization are unlikely to be picked up by parents and the authorities and, as such, many women are not formally diagnosed with psychological disorders and/or trauma. A disproportionate amount of neglect, abuse, and harassment is experienced by women who end up in juvenile reform schools. For a large proportion of youth, the neighborhoods themselves play an important role in their lives; not wanting to be seen at home and risk the

296

ROSARIO POZO GORDALIZA

emotional, physical, and sexual abuses that they face there, they spend a lot of their day on the streets. It is in this environment that the young women meet peers and partners and where, in the course of developing strategies for survival and subsistence, they enter into the spiral of criminal lifestyle, outside of the law. It is only by looking at the early and continued abuse suffered by these young women that it is understood why many spend most of their time on the streets. In situations where parents are not present or not properly able to care for their children, these neighborhoods and their spaces (parks, plazas, etc.) become an important factor in the socialization of young people. When considering the descriptions of victimization experienced by young people, it is difficult to identify events that could be described as turning points; their experiences appear to be continuous and overlapping. This is especially true of young people from the most excluded and marginalized areas/zones who are a good proportion of youth in the study. The other minority are the young people from neighborhoods near these marginalized areas who, whilst not living within the same neighborhoods, share many of the same public spaces and services, know the same people and have suffered similar early victimization in their homes. Their degree of participation in delinquency4 is lower than those who live within the marginalized areas and is more related to their “bad behavior” than serious criminal behavior. Their activities are likely to result in conditional control orders or probationary type controls that dictate that their behavior is controlled and monitored by their family, school, or social services. If their activities are of a criminal nature the courts may give them suspended sentences. Either way, it is the breaching of such orders that generally results in these girls being admitted to an institution. Past victimization was identified as a significant factor for all girls; manifesting itself most obviously in those in which the situation of victimization had been continuous and pronounced. In my study, the girls whose victimization was acute and pronounced in their homes spend an increasing amount of time outside of their homes to avoid the abuse or conflicts; they often run away spending time with friends or partners and some moving away permanently. When I speak of victimization I am referring to young girls who have, throughout their lives, experienced long periods of psychological abuse, physical abuse, abandonment, neglect and sexual abuse (among others). In most cases this abuse is not perpetrated on an occasional or accidental basis, but is an inherent part of their lives. This level of victimization generally occurs in the lives of girls who have lived and live in violent neighborhoods, where conflict exists within their families, their partnerships, their districts, and

Girls Offenders Pathways into the Spanish Juvenile Justice

297

their schools. In most cases these girls are both victims and victimizers; however, while the crimes against them are kept quiet, their behavior is criminalized. Many of these young people live daily with violence and conflicts (within their families and neighborhoods).5 That is why we need to look at abuse as a constant in their lives. Their stories are full of ups and downs in terms of the intensity of their victimization and, as such, it is difficult to pinpoint where the abuse, violence, and conflict starts and stops as they are always living with it. This is especially true in the cases of the girls from the marginalized areas and less so of girls from the neighboring areas. It can be seen within the young women of working-class neighborhoods that sexual abuse can lead to bad behavior ending at reform. Women who end up in juvenile reform schools have generally experienced disproportionate amounts of neglect and abuse. The girls who refuse to live in abusive homes run away and often end up collected by the juvenile justice system. In so doing, the system is punishing part of the survival strategies of these young people. It is not an accident that many of the survival strategies of these girls often lead to their becoming involved in the commission of criminal offenses. Although breaking the law is present in all social classes, it seems that girls who belong to socio-economically poor groups and who therefore have more social disadvantages are more likely to end up institutionalized in juvenile reform centers. Although poverty and socioeconomic disadvantage are not the only reasons, these could be related to the survival strategies of young women (Carlen, 1987, 1988). It is clear that there are young people without power who are very guarded by informal controls (including family) and, as such, invisible (Zahn et al., 2008). These young people tend always to be from marginalized neighborhoods or neighboring areas (Gilly & Gelsthorpe, 2009). The forms of female crime, therefore, are a way to survive; they are reactive and make sense in the context of their experiences. The other category of girls are those from the middle-classes and those who have been adopted out of the marginalized areas, who will exhibit behavior that is a reaction to early victimization (including sexual abuse, rape or sexual bullying and harassment). The actions of these young women must be understood in the context of their victimization Batchelor (2007), they are not necessarily criminals as such, but rather their actions are indicative of the frustrating events that have occurred to them; events that in many cases are not disclosed or known to the authorities or informal controls; events that are specifically connected to their gender. Sexual harassment, sexual abuse, bullying with sexuality as a component, pregnancy, young motherhood, and abortion are issues that particularly affect girls.

298

ROSARIO POZO GORDALIZA

Early socialization with toxic substances, crime and the world of prison and coping with conflict and violence (in the family, neighborhood, peers, secondary education, partner and others): This situation was more pronounced in slums. It is particularly striking how many families have histories of imprisonment or contact with the criminal justice system, whether this is a father, mother, sister, or brother in prison, prison reform, or protection. For many of these families their contact with the punitive system (specifically institutional) forms benchmarks in their lives, whilst being a constant in that it is a system in which they always return. These inputs and outputs repeat through generations: beginning with young children being in protection institutions during infancy, continuing into juvenile reform institutions in adolescence, and ending in prison in adulthood. The siblings of the young women will also have experienced situations of victimization and neglect in childhood and share many of the issues of the young women such as alcohol and drug abuse, lack of education, teen pregnancy, abuse of/from their partners, criminal activities, reliance on welfare centers and, ultimately prison reform. From this study it was clear that the girls considered their family lives and the multiple problems and conflicts associated with it, as normal. They also had an early socialization in the world drugs, alcohol, crime, and the prison world. This normalization occurs as a result of being brought up in an environment where imprisonment, drug abuse, criminal behavior, and the effects of these factors are considered the norm within peer groups and families from marginalized areas. Socialization in gender roles (ethnicity and socioeconomic class): gender roles are influenced by socioeconomic class and ethnicity, which explains the traditional standards of the young women in the study. From an early age, they engage in tasks that support, reinforce and replace paternal or maternal roles when parents are missing for various reasons that usually have to do with problems with toxic substances and imprisonment, etc. Young women also do the housework, are caretakers, participate in the informal economy (by working legally or illegally) and hold low-paid, precarious, temporary, subsistence-level jobs that are insufficient for their needs and are always prototypical of gender. Their mothers, sisters and grandmothers take on the sole financial responsibility for and care of their children. Poverty, exclusion, and loneliness engulfed part of the lives of the young Gitanas (especially in slums) who were totally dependent on social welfare. As for ethnicity, the group shared more features with excluded groups, rather than ethnic groups (high rates of mixed backgrounds). Gitanas and girls from mixed backgrounds

Girls Offenders Pathways into the Spanish Juvenile Justice

299

certainly make up a very heterogeneous group. This section gives a glimpse into the importance of “female problems,” including sexual abuse, sexual harassment, bullying, pregnancy, abortion, and motherhood, in the life stories of these girls and the juvenile justice system. All children are influenced by gender. It seems, however, that the differentiation of gender roles is in turn influenced by socioeconomic class and ethnicity. Children from working-class backgrounds will have more traditional views on gender roles than middle class children. Many of the youth from this study (from the marginalized and neighboring areas) had performed support tasks inside the home from an early age, substituting maternal and paternal roles when they were absent for various reasons (e.g., drug addiction, alcoholism, and prison); this was also true outside of the home, especially in the cases of girls who had gone to live with a partner. These support tasks mainly involved housework and care but also involved bringing money into the house, often through illegal means. Poverty and social exclusion dictate the course of these girls’ lives. In order to resist poverty, victimization, and lack of power the girls have to develop delinquent and illegal behavior. The schools the girls attended were public and located in slums or areas near them (they shared neighborhood problems and there were educational ghettos in which only male Gitanos or boys from mixed backgrounds went to school, accumulating many of the problems in the neighborhood). In this area, the characteristics the children presented were: irregular attendance, truancy, repeated courses, disruptive behavior in elementary or secondary school, expulsion, lower educational level than average, and dropping out of school without obtaining a secondary school diploma. The prototypes of the male or female role marked their families and social circles (few of whom had secondary school diplomas) and their jobs; they are particularly affected by unemployment and illegal, unstable and poorly paid markets. School did not mean much to those girls. Their peer groups or boyfriend(s) shared the same problems of the neighborhood and social spaces (schools and social resources, among others). Among young Gitanas and girls from mixed backgrounds, as well other girls, ethnicity was an important component. Their personal references were indeed Gitanos, Roma immigrants and people from mixed backgrounds. The study suggests that their peer groups were the most relevant to them, followed by their boyfriends and their boyfriends’ families. Furthermore, associating with a group of delinquent or drug-consuming peers or boyfriends increased the risk of criminal behavior or substance consumption. Preventive institutionalization (which acts as an informal control in

300

ROSARIO POZO GORDALIZA

addition to constituting a strong social capital): admission into these centers (in any form) was motivated by neglect, abandonment, abuse, homelessness, begging, misbehavior or the commission of misdemeanors or felonies before the age of 14, among other reasons. Running away, either temporarily or permanently: this was mainly caused by victimization at home, thus the neighborhood (streets, squares, and parks) was the scenario of socialization (in crime, drug abuse and violence, etc.). These are the places where young women met their peers and partners (more pronounced in the cases of girls from slums). The study showed how running away from home, either sporadically or definitively, was a tipping point that triggered other processes of vulnerability for young girls. Many Gitanas and girls from mixed backgrounds who run away were driven by the Gitano rite that considers escaping from home “la escapa´” a way to marry or cohabitate. Survival strategies: legal, illegal, extra-legal, or criminal strategies were used to cope with frustrating events, which almost always took place in slums or areas near them. Many of these strategies or ways to run away gave rise to new processes of vulnerability, such as substance consumption or abuse, partner abuse (without any formal or informal complaints), early pregnancy, abortions or motherhood. Living with a partner or getting married were other strategies commonly adopted to meet economic and emotional needs (this was true of girls from slums as well as those from neighborhood close to conflictive areas). Other methods that were detected but only in girls from slums were prostitution, begging, or homelessness (or intermediate situations). The body was an interesting element to analyze, especially for ascertaining the young women’s multiple strategies of resistance, control, and lack of control. Many of the girls responded to the traumatic events they had experienced with suicide attempts, anorexia, bulimia, self-harm, toxic substance consumption, and fights. I believe that there is a link between victimization in the home and the triggers that increase vulnerability of the young to delinquency. As a result of leaving home at a young age, young adults struggle to obtain economic independence and self-sufficiency. As such, they come to depend on the assistance of public institutions and social services. It is at this point when many young start to develop a variety of coping strategies, developing many of these strategies within the neighborhoods where they have grown up (often problem neighborhoods), by activities that are illegal (or on the brink of being) or by looking for a partner who can meet their basic needs (clothing, food, care, affection, shelter, safety). However, this form of escape is also the gateway to other means of vulnerability that will be addressed later

Girls Offenders Pathways into the Spanish Juvenile Justice

301

(abuse, ill-treatment by his family, pregnancy, abortion and risks associated with early motherhood). What we cannot ignore, in any case, is the existence of a sexual double standard reflected in parental authority and youth justice. The girls who complained of serious forms of parental abuse and the exertion of sexual and physical authority were almost always ignored by the authorities and the police. When the girls run away from these abusive homes, they are required to return and be institutionalized within them. It is important to remember that many of their survival strategies are linked to their living within a patriarchal society (and in neighborhoods where women are disenfranchized and have little power). It is no coincidence that the girls who come from environments of conflict and abuse within their homes and who lack opportunities, develop survival strategies related to their exclusion and the time that they spend on the streets. Such strategies may include prostitution and illegal or criminal activities. This is especially true of those young people who are in a more precarious financial situation, who live in the marginalized neighborhoods (or neighboring areas) and have experienced serious problems in their family. Thus, we must consider abuse and sexual abuse at a young age within home and subsequent escape from this environment, as a trigger for prostitution, begging, or homelessness. Whilst many girls have developed prostitution, begging, and homelessness as their survival strategies, some have also developed other variations, three in particular for example: instead of exercising pure and simple begging on the street, the girl would go home to their family, friends, or neighbors to eat or to get clothes and other basic necessities; instead of engaging in full time prostitution, some girls will sleep with people only to satisfy basic necessities and; those who would otherwise be homeless on their own will often spend periods of time living with friends in abandoned areas (boat, gates, beaches, portals, etc.). The other strategy, used by the girls as an alternative way of dealing with their economic and emotional situation at home, is to go and live with their boyfriends or Marios. It is very common for young people who are experiencing problems with their family at home to decide to move in with their boyfriends. This is seen as an opportunity to leave their homes. Some young people (usually Gypsy/Roma or Mestiza) decide to get married in the Gypsy tradition; in these cases the young women will leave school and concentrate on their new home and their role in caring and managing the household. In this regard, whilst reports and dossiers may record the young woman as officially living with her parents, it has been found that a number of young people before entering the juvenile reform center did not live with their parents but were living with their partners (where the girls had been

302

ROSARIO POZO GORDALIZA

with their parents, this was often a temporary arrangement linked to their probation and the requirement for parental control). Living with their partners means that these girls develop adult roles. However, they are not adults and I have found cases of girls aged 12 17 taking on such roles. In addition it is interesting how many of them identified in the reports their partner or boyfriend as their “Mario” (husband) and his house as their home. When young people live with their partners or are married they tend to leave school and devote themselves fully to the tasks of their new home, performing the functions of a housewife. Where these young people are still children, they will be the person primarily responsible for their own care as well as care of children from their partner’s previous relationships. The neighborhoods where they move to in order to live with their partners or Marios will also be difficult neighborhoods in marginalized areas. Some of the boyfriends or Marios will have a criminal record and may have been to jail, treatment centers, reform or protection, or have been homeless. These boyfriends will often be older than the young women, regularly exposing them to drug abuse and criminal behavior. Information taken from interviews with the girls showed a high number of incidences of jealousy on the part of their partners within their physical, psychological, sexual relationship or marriage, although the women very rarely complained about this behavior externally. Many of the girls in this situation are survivors of an abusive family environment and relationships. They are also generally victims of abuse in their new relationships. They often experience feelings of loneliness which they manage by fantasizing about different partners, idealizing the concept of starting a new family, or having suicidal thoughts. Pregnancies, abortions, motherhood: the young women in the study had no control over preventing unwanted pregnancies. They viewed getting pregnant as an opportunity to start a family or change their lifestyle. Pregnancies took place at an early age, before 16 (especially in young women from slums with a Gitano population) and in circumstances and stages in which the girls had consumed substances, had fights, committed crimes, and were abused (by their partners or families). Young girls also had high rates of abortions in situations that put their health at risk. For these young women, motherhood was an emotional or occupational strategy. Toxic substances: both the young women in the study from slums as well as those from neighboring areas consumed toxic substances (many had turned to consumption because of the frustrating events in their lives), especially cannabis, cocaine, alcohol, tobacco, and psychotropic drugs. Multi-drug

Girls Offenders Pathways into the Spanish Juvenile Justice

303

consumption or the use of hard drugs was low, but more pronounced in girls from slums (it must be noted that these drugs are present in these neighborhoods, which are predominated by the Gitano or mixedbackground population). This study suggests that “problem” consumption forced female consumers to commit criminal acts or “misbehave,” but was not necessarily a turning point in their lives; instead it was just another problem to add to their biographies. Furthermore, some of the girls received psychological or psychiatric treatment not because of their victimized past, but rather because of the beginning or suspicion of possible consumption. It cannot be concluded that young Gitanas or girls from mixed backgrounds consume more than other girls, but living close to or in marginal neighborhoods undoubtedly exposes them to contact with toxic substances more frequently and at younger ages. Violence and young women: domestic violence or abuse was related to misbehavior at home (breaking family rules, coming home late, running away from home, truancy, peers, and boyfriends that families disapprove of, suicide attempts or suicide, fights, verbal confrontations (insults), physical confrontations (scratching, biting, pushing, breaking household items, etc.) and isolation. The girls were characterized by fighting on both sides, being the subject (as an onlooker or participant) of on-going abuse at home and blaming their mothers for not having protected them from such abuse when they were young. The informal controls (social workers, teachers, parents and neighbors, etc.) were the ones who reported or denounced these behaviors to the juvenile justice system. The study perceived a quasi-relationship in these young women between having experienced domestic violence and the offense of domestic abuse among the girls. Grouped together here were young women from areas close to slums, adopted girls and lower-middle-class girls who had come up against a frustrating event in their lives. These young girls’ behavior was defined or redefined as misdemeanors or crimes. Judges normally impose preventive measures, which is why is so important the way to enter to the juvenile system for those girls that the gipsy girls are less represented. This study suggests that regular episodes of violence by young women in slums were rare and occurred in specific situations: on the street, at home, at primary or secondary school, with their peer group or partner, etc. Constantly present and habituation to present facts or violent situations and it was a survival strategy. The misdemeanors or felonies committed by the girls that led to custodial sentences were essentially breaches of preventive measures (community service or breach of the court few). What led them to preventive measures

304

ROSARIO POZO GORDALIZA

were: (a) misbehavior by young females redefined, for example, as domestic abuse or (b) economic crimes such as theft or robbery (mainly by girls from slums predominated by groups with low socioeconomic and cultural resources). Exceptionally, they sold drugs in circumstances that were widely known to the authorities and the rest of society (here ethnicity does seem to be an important component). Selling drugs was a source of family income for groups with a low potential for earning money legally. Some girls who had developed consumption habits began to sell drugs not only to pay for their own consumption, but also to meet their most basic needs. These girls committed more crimes than those usually involved in juvenile justice bodies (this issue has been very difficult to ascertain at the quantitative level). Many of the young girls from slums who began committing crimes at an earlier age went on to commit more crimes later on. Without a doubt, crime was an inherent element in their life stories or marginal style and these girls had suffered the most victimization (at home, in the neighborhood, by peers or partners, at school, etc.). Their crimes were more serious (officially and unofficially); they took less responsibility for and minimized the importance of their crimes. As for theft accompanied by force or violence (usually produced in a group or with their partners), their involvement was subsidiary (i.e., they performed functions such as removing valuables, keeping an eye out, intimidation, stealing, and covering up). Males (who were older than the girls) played a more active role in committing crimes.

Limitations and Difficulties Encountered Any study has limitations and difficulties in its design, delimitation, execution and analysis. The main difficulty encountered in carrying out this study, as noted previously, has been the invisibility of young female offenders inside and outside the juvenile justice system in Spain, which has resulted in the lack of interest shown by scientific and literary publications. First, in terms of its design I can say that I selected only young girls who have been identified as offenders by the system and who have been subjected to some form of detention. In addition, the highly qualitative orientation of my study undoubtedly entails the limitations of using this technique in terms of representativeness and universality. Secondly, regarding its delimitation, the decision was taken to conduct this research in Andalusia in the only female juvenile reform center, as well as in the three courts where the most cases of registered delinquency were identified.

Girls Offenders Pathways into the Spanish Juvenile Justice

305

Thirdly, with regard to its execution, I must emphasize the tedious bureaucratic tangle, due to which I encountered particular difficulties in the execution of the empirical phase. It was a tedious process in terms of time (more than one year to obtain the first official permits) and bureaucratically speaking (constant problems accessing information due to the very nature of my subject). Other difficulties arose with finding a space to conduct the in-depth interviews without interruption despite the interviews in progress some staff personnel came in without prior warning. Difficulties were also encountered in conducting the interviews with those young girls who were actually ready and who, due to problems with parents’ permission, were denied the opportunity to do the interview, to tell their story and participate in the study. Also during the early stage of my stay in the center, I found it problematic to get used to my interlocutors (minors using very extreme slang) both linguistically and culturally. And finally, I must emphasize the solution for finding some of the dossiers of the young women even in the juvenile courts (loss, disorganization of documents, changes in name or issues, etc.).

DISCUSSION AND FINAL CONSIDERATIONS This paper specifically notes that one cannot understand the crime of young people without understanding their life and socialization in terms of gender, ethnicity, and socioeconomic class and culture inside and outside of the home (April Bernard The Intersectional, 2013; O’Neill, 2005). Undoubtedly, studying girls who break laws certainly provides insight into the standards, practices and social customs affecting young women in a particular time and space, as well as providing clues about the expectations of their gender. Girls are often the victims of what might be called multiple situation of marginality in that their gender race and class have placed them at the economic periphery of society. Understanding their lives and choices of girls who find themselves in the juvenile criminal justice systems also requires a broader understanding of the contexts and process within which their criminal behavior is lodged. The situation of vulnerability and marginality is an important link between the delinquency and crime. The strategies that young marginalized women use to deal with situations of exclusion and vulnerability may not seem attractive, but in neighborhoods and families that are dealing with concentrated social disadvantage that have few resources to protect the young

306

ROSARIO POZO GORDALIZA

women, these women look for alternative strategies often becoming involved in dangerous and criminal activities. As for the young people from middle class families (including those who have been adopted) who live in suburbs and share many of the problems of neighborhoods of the marginalized areas, their degree of exclusion is not as intense and they are identified primarily by having bad and delinquent behaviors which only become crimes at the time that they violate the conditions of their probation. These girls have generally suffered (or continue to suffer) early victimization on an ad hoc or permanent basis (including sexual abuse, emotional abuse, and neglect) and have previously been in institutions or the protection of the authorities; as such, it can be seen that this history of moving from institution to institution ends up in spending time at the reform center. The young woman’s initial experiences of exclusion breed survival strategies in a parallel process of quasi-permanent vulnerability, defining much of what they are and what they become. Undoubtedly, the information and analysis provided in the study helps us understand the conditions that lie behind female juvenile delinquency and institutionalization, by looking at their life histories and the everyday strategies they employ to get by. Social exclusion experienced by these young women results from victimization within the family, which sends them out into the neighborhood with limited choices and life chances. Continued research and studies into the behavior of girls is needed so that they can come out of the invisibility in which they lead their lives, in terms of their “bad behavior” inside and outside the family and their criminal acts. It is necessary to continue to learn more about the relationship between women and crime today and also what society considers “bad behavior in girls.” The study also considers the ideas and concepts of those people within the juvenile justice system who are making the interventions (both directly and indirectly, i.e., police, social services, psychologists, and judges). The report is concerned that social policy needs to be introduced that recognizes these marginalized girls with specific reference to their trauma and takes into account their life experiences (Acoca et al., 2015; Wesely, 2006). The specific principles listed here are intended for use in the development of gender-responsive programs for women (Bloom & Covington, 1998): Comprehensive, Safe, Empowering, Community and Family Focused and Relational. The independent commission on youth crime and antisocial behavior; time for a fresh start (2010) this commission proposed prevention, resaturation and integration considering effectiveness and evidence. Also the Spanish criminal justice need a “new breed” on

Girls Offenders Pathways into the Spanish Juvenile Justice

307

girls and juvenile justice. I consider that much has come out of this study that needs further review. It is essential that the processes, factors, and contexts of social exclusion continue to be analyzed to ensure better understanding of female juvenile delinquency and the role on Juvenile Justice. The phenomenon continues and there are a lot of things that are still left to be learnt about these girls. I hope this paper and its particular focus on institutionalized youth in Spain and their unique collective identity, will serve as a starting point for future contributions and theoretical reflections that will reduce adverse impacts of the invisible phenomenon of female juvenile behavior.

NOTES 1. The children are socialized into roles and this seems to influence the inputs, outputs, and length of their sentences. 2. In the article on juvenile delinquency, Rechea Alberola and Ferna´ndez Molina (2003) explained how the problem of successive legislative changes made it difficult to have a single definition of youth and young offenders in recent years. Statistics have not properly taken into account these changes and, as such, it is not possible to look at the statistics and identify accurately whether or not there has been an increase in young offending; it also means that it is necessary to use alternative data for the interpretation of the phenomenon of juvenile delinquency. 3. A large group of ethnic mestizos (people who have a mix of Roma/Gypsy or Gitano ethnicity) were identified in the discourse. The matter of ethnicity and ethnic identity, however, is more complex to address than it first appears and needs more specific in-depth study. Regarding the ethnicity of the young girls, I found a great diversity in behavior from the traditional to the more modern. As far as the dossiers and case studies were concerned, however, anything that did not match the stereotypical image of gypsy was not picked up and, as such the statistics of mestizos and Gypsy’s in juvenile crime were unreliable. 4. In this paragraph I am referring to the official records of crime and delinquency. The study, however, has revealed that many more crimes are committed than are recorded. 5. During my fieldwork I found two realities: the realities of dossiers and the reality of the stories from the girls. These two realities should not be regarded as contradictory but as complementary. As regards the dossiers, many lacked depth and explanatory information, information was often confusing or misplaced. With the life stories, however, more cases of mistreatment and abuse was identified than had previously been reported or recorded (information denied to the social services by the girls). Despite the fact that very little of the information was visible in the dossiers, there were many indications that made the seriousness and extent of the problem quite clear, not only during childhood, but in late adolescence.

308

ROSARIO POZO GORDALIZA

REFERENCES Acoca, L., Countryman-Roswurm, K., Kaba, M., & Winn, M. T. (2015). Girls in Justice. Alder, C., & Worrall, A. (2004). A contemporany crisis? In C. Alder & A. Worrall (Eds.), Girls’ violence: Myths and realities (pp. 1 20). Albany, NY: State University of NY Press. April Bernard The Intersectional. (2013). Alternative: Explaining female criminality. Feminist Criminology, 8(1), 3 19. Batchelor, S. (2007). Prove me the Bam! Victimisation and agency in the lives of young women who commit violent offences. Unpublished Ph.D. thesis. University of Glasgow. Belknap, J. (2001). The invisible woman: Gender, crime, and justice (2nd ed.). Belmont, CA: Wadsworth. Bloom, B., & Covington, S. (1998). Gender-specific programming for female offenders: What is it and why is it important? Paper presented at the 50th Annual Meeting of the American Society of Criminology, Washington, DC. Burman, M., Batchelor, S., & Brown, J. (2001). Researching girls and violence: Facing the dilemmas of fieldwork. British Journal of Criminology, 41(3), 443 459. Carlen, P. (1987). Out of care, into custody. In P. Carlen & A. Worrall (Eds.), Gender, crime and justice. Milton Keynes: Open University Press. Carlen, P. (1988). Women, crime and poverty. Milton Keynes: Open University Press. Chesney-Lind, M., & Eliason, M. (2006). From invisible to incorrigible: The demonization of marginalized women and girls. Crime, Media, Culture, 2, 29. Chesney-Lind, M., & Pasko, L. (2004). The female offender: Girls, women, and crime (2nd ed.). Thousand Oaks, CA: Sage. Chesney-Lind, M., & Pasko, L. (2013). The female offender girls, women, and crime (3rd ed.). London: Sage Publications, Inc. Chesney-Lind, M., & Shelden, R. G. (2013). Girls delinquency and juvenile justice (4th ed.). USA: Wiley-Blackwell. Estrada, F., & Nilsson, A. (2012). Does it cost more to be a female offender? A life-course study of childhood circumstances, crime, drug abuse, and living conditions. Feminist Criminology, 7(3), 196 219. Gilly, S., & Gelsthorpe, L. (2009). Engendering the agenda: Girls, young women and youth justice youth. Girlguiding. (2007). Girls shout out! A UK-wide research report by Girlguiding UK. London: The Guide Association. Godfrey, B. (2004). Rough girls, 1880 1930: The “Recent” history of violent young women. In C. Alder & A. Worrall (Eds.), Girls’ violence: Myths and realities. Albany, NY: State University of New York Press. Goodkind, S., Ng, I., & Sarri, R. C. (2006). The impact of sexual abuse in the lives of young women involved or at risk of involvement with the juvenile justice system. Violence against Women, 12(5), 456 477. Gottfredson, D. C., & Kruttschnitt, C. (2008). Violence by teenage girls: Trends and context: Office of justice programs. USA: Office of Juvenile Justice and Delinquency Prevention, Girls Study Group. Haney, L. (2010). Offending women: Power, punishment, and the regulation of desire. Berkeley, CA: University of California Press. Hannah-Moffat. (2000). Prisons that Empower. British Journal of Criminology, 40(3), 510 531.

Girls Offenders Pathways into the Spanish Juvenile Justice

309

Hannah-Moffat, K., & Yule, C. (2011). Gaining insight, changing attitudes, and managing “risk”: Parole release decisions for women convicted of violent crimes. Punishment & Society, 13, 149 175. Heidensohn, F. (2000). Sexual politics and social control. Buckingham: Open University Press. Instituto nacional de estadı´ stica. (2013). Madrid: INE. [Consulta: 10 junio 2013]. Disponible en. Retrieved from http://www.ine.es/ Kempf-Leonard, K., & Pernilla, J. (2007). Gender and runaways risk factors, delinquency, and juvenile justice experiences. Youth Violence and Juvenile Justice, 5(3), 308 327. Lee, E., Hershberger, S., Field, E., Wersinger, S., Pellis, S., et al. (2008). Sex differences: Summarizing more than a century of scientific research. Loucks, N., Malloch, M., McIvor, G., & Gelsthorpe, L. (2006). Evaluation of the 218 Centre. Edinburgh: Scottish Executive Justice Department. Scottland. MacDonald, M. J., & Chesney-Lind, M. (2001). Gender bias and juvenile justice revisited: A multiyear analysis. Crime & Delinquency, 47(2), 173 195. Malloch, M. (2004). Missing out: Gender, drugs and justice. Probation Journal, 51(4), 295 308. Marutto, P., & K. Hannah-Moffat (2005). Assembling risk and the restructuring of penal control. British Journal of Criminology, 45, 1 17. Moore, L. D., & Padavic, I. (2010). Racial and ethnic disparities in girls’ sentencing in the juvenile justice system Feminist Criminology, 5(3), 263 285. O’Neill, T. (2005). Girls in trouble in the child welfare and criminal justice system. In G. Lloyd (Ed.), Problem girls: Understanding and supporting troubled and troublesome girls and young women. Abingdon: Routledge Falmer. Rechea Alberola, C. Y., & Ferna´ndez Molina, E. (2003). Juvenile justice in Spain: Past and present. Journal of Contemporary Criminal Justice, 19(4), 384 412. Renzetti, C. M., Curran, D. J., & Maier, S. L. (2012). Women, men, and society. CRVAW Faculty Book Gallery. Smith, A. M. (2002). The sexual regulation dimension of contemporary welfare law: A fifty state overview. Michigan Journal of Gender & Law, (8), 121. Watson, L., & Edelman, P. (2012). Improving the juvenile justice system for girls: Lessons from the states. USA: Georgetown Center on Poverty. Wesely, J. (2006). Considering the context of women’s violence: Gender, lived experiences, and cumulative victimization. Feminist Criminology, 1(4), 303 328. West, & Zimmerman (1987). Doing gender. Gender and Society, 1(2), 125 151. Worrall, A. (2001). Girls at risk? Reflections on changing attitudes to young women’s offending. Probation Journal, 48(2), 86 92. Zahn, M. A. (2009). The delinquent girl. UK. http://www.temple.edu/tempress/titles/1966_reg.html Zahn, M. A., Day, J. C., Mihalic, S. F., & Tichavsky, L. (2009). Determining what works for girls in the juvenile justice system. Crime & Delinquency, 55(2), 266 293. Zahn, M. A., Hawkins, S. R., Chiancone, J., Whitworth, A., et al. (2008). The girls study group Charting the way to delinquency prevention for girls. USA: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

This page intentionally left blank

A BINARY MODEL OF BROKEN HOME: PARENTAL DEATHDIVORCE HYPOTHESIS OF MALE JUVENILE DELINQUENCY IN NIGERIA AND GHANA Suleman Ibrahim ABSTRACT Purpose In terms of the concept of broken home as a juvenile delinquency risk factor, whilst Nigeria and Ghana are culturally different from western nations (Gyekye, 1996; Hofstede, 1980; Smith, 2004), parental death (PDE) and parental divorce (PDI) have been previously taken-for-granted as one factor, that is ‘broken home’. This paper aims to deconstruct the singular model of ‘broken home’ and propose a binary model the parental death and parental divorce hypotheses, with unique variables inherent in Nigerian/Ghanaian context. Methodology/approach It principally deploys the application of Goffman’s (1967) theory of stigma, anthropological insights on burial rites and other social facts (Gyekye, 1996; Mazzucato et al., 2006; Smith, 2004) to tease out diversity and complexity of lives across

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 311 340 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009014

311

312

SULEMAN IBRAHIM

cultures, which specifically represent a binary model of broken home in Nigeria/Ghana. It slightly appraises post-colonial insights on decolonization (Agozino, 2003; Said, 1994) to interrogate both marginalized and mainstream literature. Findings Thus far, analyses have challenged the homogenization of the concept broken home in existing literature. Qualitatively unlike in the ‘West’, analyses have identified the varying meanings/consequences of parental divorce and parental death in Nigeria/Ghana. Originality/value Unlike existing data, this paper has contrasted the differential impacts of parental death and parental divorce with more refined variables (e.g. the sociocultural penalties of divorce such as stigma in terms of parental divorce and other social facts such as burial ceremonies, kinship nurturing, in relation to parental death), which helped to fill in the missing gap in comparative criminology literature. Keywords: Parental death; parental divorce; juvenile delinquency; broken home; Nigeria and Ghana

INTRODUCTION Nigeria and Ghana are Anglophone Sub-Saharan nations separated and surrounded by Francophone nations. Despite multiple ethnic variations within and across these two countries, they have similarities: British colonization, English language, relatively, similar time of independence (Amponsah, Akotia, & Olowu, 2006; Feit, 1968; Mivanyi, 2006). In part, these similarities offer a logical basis to examine them together without the surrounding nations. In terms of male juvenile delinquency, adult males in these nations have been implicated in the bulk of banditry crimes such as kidnapping, armed robbery, terrorism (Ajaegbu, 2012; Adu-Mireku, 2002) as well as cybercrimes (Warner, 2011; Tade, 2013; Doyon-Martin, 2015), especially1 socioeconomic cybercrimes. As a result, many facets of development such as tourism are weakened (Adegoke, 2014; Aderibigbe & Olla, 2014; Otoo, 2014). Given that, juvenile offenders often ‘metamorhosized’ into hardened criminals (Igbinovia, 1988; Loeber, Stouthamer-Loeber, Farrington, & Kalb, 2001) and there is a convergence of numerous developmental theories on stability of delinquency across lifespan

A Binary Model of Broken Home

313

(Farrington, 2005; Loeber, 1982; Moffitt, 1993; Patterson & Yoerger, 1993), it is reasonable to consider male juvenile delinquency in Nigeria/ Ghana. Delinquency itself, like beauty, is in the eye of the beholder (Agozino, 2003; Becker, 1963/1997; Durkheim, 1965/1885). Yet researchers have generally defined juvenile delinquency as a set of behaviours prohibited by law (Akers & Burgess, 1966; Dugdale, 1887; Moffitt, 1993, Loeber et al., 2001; Farrington, 2012a) and consistent with these researchers’ views, this present endeavour operationalizes juvenile delinquency as a range of rule breaking behaviours. A principal feature of most studies is that juvenile delinquency is broadly age-graded (Moffitt, Lyam, Silva, & Bartusch, 1997; Sampson & Laub, 2005). However, some studies have pointed out that age is lived differently and age-related behaviours are constituted differently across cultures (Brathwaite, 1996; Cain, 2000). Therefore, a universal assumption is that juvenile delinquency is moderated by risk and protective factors (Derzon, 2010; Mulder, Brand, Bullens, & Van Marle, 2010). A risk factor, as Kazdin, Kraemer, Kessler, Kupfer, and Offord (1997, p. 377) clearly explained, is ‘a characteristic, experience, or event that, if present, is associated with an increase in the probability (risk) of a particular outcome over the base rate of the outcome in the general (unexposed) population’. This suggests that whilst risk factors have direct association with delinquency, they are not deterministic due to their interaction with protective factors. On the flipside, protective factors such as having a support network only moderate the potential consequences of risk variables. They have no direct association with delinquency (Derzon, 2010; Piquero, Welsh, Tremblay, Farrington, & Jennings, 2009). Therefore, this paper will lean its weight more on risk variables that have direct association with delinquency than protective factors. In terms of risk variables, children-behaviours are primarily ruled by immediate gratification (Erikson, 1995; Freud, 1991/1927). Thus it is the responsibility of ‘care-givers’ to see that they adapt to societal norms (Bowlby, 1988). Yet the likelihood of this adaptation concurrently depends on environmental variables as well as children’s characteristics (Belsky, 1984). Consistent with this assumption, Wissink, Asscher, Dekovic, Prinzie, and Stams (2014) conceptualized juvenile delinquency risk factors into tripartite branches: familial, individual and structural factors. These tripartite branches are illustrated in Fig. 1. A more important issue is that the vulnerability effect of any single risk-variable on these branches is magnified only in the presence of other variables (Derzon, 2010; Miller, 2014; Welsh & Farrington, 2007a, 2007b).

314

SULEMAN IBRAHIM

The Tripartite Factors within the ‘Glass’ are in a Dialectic Relationship in Predicting Delinquency

Broken Home, Lax Parenting, Physical Punishment, Family Conflict, Low Family Income (Derzon, 2010; Piquero et al., 2009; Moffitt et al., 2013; Sanni et al., 2010)

Family Factors

Individual Factors

Structural Factors

Juvenile Delinquency

Fig. 1.

Impulsivity, Low IQ, Academic Difficulty, Aggression, Substance Abuse (Farrington et al. 2010; Loeber et al., 2012; Boakye, 2013)

Social Disorganisation, High Crime Rates, Delinquent Peers, Low Social Buffers, Gang Membership (Skinner et al., 2014; Smith and Mc Vie, 2003; Weinberg, 1964; Shaw and MacKay, 1948)

The Tripartite Branches of Risk Factors.

Historically, the legal concept of ‘juvenile delinquency’ was introduced in about 1900 in Nigeria/Ghana through colonialization the British Government (Ebbe, 1992). Drawing from Cohen’s (2008/1988) analysis of African history and western criminology, Agozino (2003, p. 6) constructed a link between colonialism and criminology. He argued that criminology is a ‘discipline for disciplining the Other’ and not to benefit the ‘Other’ (see also Agozino, 2010; Kitossa, 2012). This link between colonialism and criminology perhaps underpins the total absence of empirical study in Nigeria/Ghana, until official colonialization ended in the 1960s (Bankole & Chipaca, 2014). In part, whilst juvenile delinquency has been extensively investigated in western societies (Loeber et al., 2001), Nigerian/Ghanaian studies are scanty (Boakye, 2013; Tankebe, Hills, & Bankole, 2014). Also, various studies conducted in western nations have alongside familial and structural factors covered individual factors, whereas Nigerian/Ghanaian literature focused mainly on familial variables, apart from Boakye’s (2013) study, one of the most comprehensive study so far in this region. Boakye (2013) situated predictors of juvenile delinquency more to individual factors than to familial variables and identified

A Binary Model of Broken Home

315

‘academic-difficulty’ as one of the strongest predictors of delinquency in Ghana. However, academic difficulty is not the main focus of this paper and the overall concentration of Nigerian/Ghanaian literature converged more (in a descending order of significance) on ‘broken home’ than others. It is reasonable therefore that these patterns of existing literature dictate where this paper will lean its weight, that is where Nigerian/Ghanaian literature mainly overlaps with that of the ‘West’ broken home. Broken home has been homogenized by existing literature, whereas this paper proposed a binary model parental divorce and parental death hypotheses with more refined variables. Given that every experience counts in cross-cultural research (Cain, 2000) and Sub-Saharan African is a critical site for comparative criminology knowledge-mining (Tankebe et al., 2014), it is compelling therefore that this paper aims to highlight the importance of widening the focus of comparative criminology beyond European, American and Australian continents (Tankebe et al., 2014). This is not to concede that claims made in these world regions may have applicability in African nations (LaFree, 2007, p. 16), but on the contrary, ‘criminology’s claims in western nations might just be nothing more than local truths’ (Nelken, 2010, p. 14). Therefore, are the consequences of broken home in the ‘West’ universalizable to Nigerian and Ghanaian context? Are the consequences of parental death and parental divorce the same across cultures? A principal goal of this paper is to deconstruct the homogenization of parental death and parental divorce in most existing literature and sharpen the distinction between the two constructs, specifically in Nigerian/Ghanaian context. The need to address these issues is reinforced by sociocultural variations across these world regions.

Theoretical Background Apparently, there are huge and long-standing cultural variations between western nations broadly categorized under individualism, and Nigeria/ Ghana relatively under collectivism (Hofstede, 1980; Smith, 2001). Culture is an interwoven set of beliefs, practices, and ‘common sense’ that define people’s way of life (Skeggs, 2013). According to Fincher and Thornhill (2014, p. 114), ‘[C]ollectivism is characterised by conformity, kinship and in-group dependency, whereas individualism places a lesser emphasis on in-group reliance and kinship’. Even if modernity blurs the boundary between collectivism and individualism (Sampson, 2000),

316

SULEMAN IBRAHIM

according to Smith’s (2001, 2004 and 2011) anthropological insights, whilst modernity privileges individuality in western societies, people’s socioeconomic and political insecurity in contemporary Sub-Saharan region reinforce a strong rather than less reliance on ties to family/community of origin (see also McCauley, 2012 Ghana; Akanle & Olutayo, 2011 Nigeria). Given that mainstream criminology-claims are reflective of western culture (Kalunta-Crumpton, 2004), suggests that some claims made in western world region may not be universalizable. This paper will assess the validity of this assumption in terms of the concept of broken home. Also, it will help to reveal the ‘superordination-subordination’ relationship between the ‘West’ and cultural others Nigeria/Ghana (Hall, 1990; Fanon, 1967/ 2008; Onyeozili, 2004; Otu, 1999; Pieterse, 1989; Ranuga, 1986; Said, 1994), which perpetuates claims from western studies as ‘universal-truths’ (Smith, 1999). For Foucault (1977/1991), ‘superiority’ or ‘inferiority’ in knowledgepower relationship, is distinguishable only by unequal power resources, exercisable and contestable in varying levels (see also Cohen, 2008/1988). This is consistent with Staples’s (1960) idea that, having the hegemonic power over ‘knowledge’ is vulnerable to hegemonic tendencies towards cultural others (see also Althusser, 1971/2001). Accordingly, this paper will also deploy a post-colonial perspective as a tool of enquiry, because of its emphasis on decolonization. ‘Decolonization’ is the deployment of a more critical examination at mainstream assumptions and research practices that appear to be ‘universal-truths’ and manifestly marginalize/shape the practices/claims of the ‘Other’ (Smith, 1999).

METHODOLOGY In order to address the above questions raised above, library-based research was used. Keywords such as juvenile delinquency, broken home, family and delinquency, parental death, parental divorce and delinquency, etc., were used in academic search engines such as Google scholar, project MUSE and psycINFO, to identify relevant journal articles. Every relevant paper found (in line with the research topic and after reading the abstracts) in Nigerian/Ghanaian contexts was selected, whereas only the most recent ones from the ‘West’ were included due to the varying volume and other aspects of literature in these world regions as mentioned. Accordingly, studies in these two world regions in terms of broken home were summarized

A Binary Model of Broken Home

317

and juxtaposed on the basis of their unique axes of differentiations, which will help to set off points for synthesis. The rationale is to expose ‘what is true/false of one society in relation to the other’ (Bendix, 1963). Also, it will interrogate some contemporary academic endeavours in Nigeria/Ghana, which are often as Fanon (1967/2008) and Myers (2006) suggested, rooted in a colonial framework. As Agozino (2003) pointed out, in criminology and the social sciences in general, the deficiency of original theorizing among the colonized; the monopolization of theory-production by imperialist countries, are the conditions that perpetuate overreliance of the former on the latter. This suggests that theoretical originality is the arena where marginalized voices are mostly vulnerable. This paper therefore takes a provocative stance and aims to stimulate contemporary scholarly endeavours from formerly colonized nations to think for themselves and drop the ‘colonial baggage’ the way poets and novelists from these nations have hitherto demonstrated. This theoretical stance will provide avenues for discourse deconstruction and evoke new interpretations (Go, 2013; Myers, 2006; Uday, 2013). It will lead to the deconstruction of a singular model of broken home, and offer more refined components that constitute parental death and divorce, which sharpen the contrast between them across regions (see Figs. 3 and 4).

An Overview of Literature: Broken Homes Familial variables have been implicated since the 19th century to impact on children’s susceptibility to crime (Akers & Burgess, 1966; Dugdale, 1887; Farrington, 2011; Fergusson, McLeod, & John Horwood, 2014; Price & Kunz, 2003; Shaw & McKay, 1932). However, Derzon’s (2010) metaanalysis of 119 works in western societies pointed out that the ill or protective effects of familial factors such as broken homes, are only evident in combination with other branches of factors illustrated in Fig. 1. Specifically, Wells and Rankin (1991) summarized 50 studies on brokenhomes and explained that the ill-effect of a broken home requires the presence of other risk factors to significantly predict delinquency (see also Price & Kunz’s, 2003 summary of 72 studies). Consistent with the above position, the majority of literature from Nigeria/Ghana highlighted that a broken home is a predictor of juvenile delinquency (Aniche, 2015; Boakye, 2013; Ebbe, 1992; Oloruntimehin, 1973; Sanni, Modo, Uduh, Ezeh, & Okediji, 2010; Weinberg, 1964). However, unlike most studies from the West with large sample sizes and

318

SULEMAN IBRAHIM

based on longitudinal data, Nigeria/Ghana based studies are all crosssectional and have relatively small sample sizes. Another key area of divergence is on family size and socio-economic status (SES). Whilst a large family size and low SES have an unfavourable impact on children’s risk of involvement in criminality (Dugdale, 1887; Loeber et al., 2001), most Nigeria/Ghana-based studies suggest that a large family size has beneficial impact on children and a low SES has no significant effect on children’s susceptibility to be involved in criminality (Boakye, 2013; Ebbe, 1992; Oloruntimehin, 1973; Weinberg, 1965). These distinct slices of data will be considered later in the analysis and critique of evidence. A more important issue is that, unlike most western literature that has attached a relatively small amount of importance to broken homes in comparison to variables such as individual factors shown in Fig. 1, most regional literature from Nigeria and Ghana situated broken home as a principal driver of juvenile delinquency (Aniche, 2015; Oloruntimehin, 1973; Sanni et al., 2010; Ugwuoke & Duruji, 2015; Weinberg, 1964). This is indicative of the assumed centrality of the family in this region as a major determinant of children’s behaviour. Specifically, Sanni et al. (2010, p. 27) pointed out that ‘[A] healthy home environment is the single most important factor necessary to keep children from becoming delinquent’ and Ugwuoke and Duruji (2015, p. 44) argued that ‘children from broken homes are at the greatest risk of becoming delinquents’. This suggests that a broken home is a more important index of the rate of criminality in children in Nigeria/ Ghana than in western societies. Having said that, whilst numerous studies have ‘clumped together’ different forms of family disruptions and compared them with two parents families (Biblarz & Gottainer, 2000; Fergusson et al., 2014; Loeber et al., 2001; Oloruntimehin, 1973) few have specifically compared divorced parents with intact families (Akanbi et al., 2015; Amato, 2014; Animasahun, 2014; Aughinbaugh et al., 2005; Cherlin, McRae, & Chase-Lansdale, 1998; Li, 2007) and fewer have attempted to distinguish between parental death and parental divorce in varying degrees (Amato & Anthony, 2014; Free Jr, 1991; Gregory, 1965a, 1965b; Juby & Farrington, 2001; Rutter, 1971). Whilst recent Nigerian and Ghanaian studies from other fields of study have considered the relationship between orphanhood and educational attainment (e.g. Hampshire et al., 2015), others have investigated the causes and consequences of divorce in isolation (Aniche, 2015; Asante, Osafo, & Nyamekye, 2014). As far as this paper is concerned, none has considered the varying variables that constitute and contrast parental death (PDE) and parental

A Binary Model of Broken Home

319

divorce (PDI) within the Sub-Saharan region (Nigeria/Ghana) and also between the Sub-Saharan region and the West. The specific variables that define the type of family disruption in different world regions may vary, meaning that this a crucial entry point in the understanding of the varying impacts of parental divorce and parental death on juvenile delinquency in these world regions. Even if marital conflicts that accompany most divorces are broadly universalizable, divorce is a bud, which unfolds overtime and not a one-time event. It is inseparable from sociocultural ‘contaminations’. This suggests that it is unreasonable to separate divorce from major underlying social facts that may buffer or amplify its negative effects across cultures. Bearing in mind that, whilst most western societies are drifting towards elevated individualism at the expense of existing social structure and traditional values (Hofstede, 1980; Lesthaeghe, 2010), Nigeria and Ghana are in the zone of collectivism as Smith’s (2004) and McCauley’s (2012) ethnographic works have highlighted. These sociocultural variations suggest that divorce may not be a uniform experience for individuals across cultures. This begs the question, are the variables that constitute parental divorce in western society the same in Nigeria and Ghana? Unlike studies from both world regions that locate marital conflict as a crucial aspect of differentiation between parental divorce and parental death (Amato & Anthony, 2014; Boakye, 2013; Free Jr, 1991; Gregory, 1965a, 1965b; Juby & Farrington, 2001; Rutter, 1971; Sanni & Oyibo, 2013), this paper, alongside marital conflict, aims to situate stigma as a principal element of distinction between parental divorce and parental death on one hand, and on the other hand, invoke some anthropological insights on burial rites, to further sharpen the contrast between the two constructs, unlike existing literature.

PARENTAL DEATH AND PARENTAL DIVORCE Two-parent families can be disrupted for reasons other than parental death and parental divorce. Creighton, Park, and Teruel (2009) found that international migration can separate spouses (see also Jampaklay, 2006). Similarly, Paris, DeVoe, Ross, and Acker (2010) found that overseas deployment of a parent for military service can lead to single-parenthood (see also Gorman, Eide, & Hisle-Gorman, 2010). Finally, Lee, Porter, and Megan (2014) found that parental incarceration often disrupts families (Wildeman, Wakefield & Turney, 2013). Generally, all five routes to

320

SULEMAN IBRAHIM

single-parenthood negatively impact on children (Kelly & Emery, 2003) by their association with high psychological stress, low parental supervision and/or low parental income in comparison to two-parent families (Amato & Keith, 1991; Savage, 2014). Arguably, all five forms of lone-parenthood could be risk factors in children’s involvement in anti-social behaviours. However, given that to date, Nigeria/Ghana-based literature has principally focused on parental death and parental divorce as elements of broken homes, it is reasonable then for this paper to consequently centre on these two factors. Parental divorce and parental death lead to a loneparent household, they both generally carry similar baggage of lower parental supervision and lower income than two-parent families. Arguably, to paraphrase Amato and Anthony (2014, p. 383), parents with high socio-economic resources ‘are more likely to buffer their children from the detrimental impacts of family disruptions’, because they are more likely to afford childcare services instrumental social support, and more immune to child rearing psychological distress than others. Apart from the psychological impact of family disruptions, which are ‘impossible’ to quantify accurately (Kelly & Emery, 2003) the two main consequences of parental separation low supervision and poverty, are moderated by parents’ socio-economic positions (Savage, 2014). Based on this assumption, which is based on a western context (Farrington, 2012a, 2012b; Loeber et al., 2001; Smith & McVie, 2003) this paper therefore asks, can these mechanisms hold true across cultures, that is Nigeria and Ghana? According to Nigeria/Ghana-based cross-sectional studies, a family’s socio-economic status is weakly related to the likelihood of children’s involvement in criminality (Akinlabi & Olatunji, 2013; Boakye, 2013; Oloruntimehin, 1973; Weinberg, 1964). The sociological underpinning is, the majority of children live in poverty and deprived communities (Akinlabi & Olatunji, 2013; Boakye, 2013). Consistent with this assumption, Owusu, Agyei-Mensah, and Lund (2008) elaborated that, the majority of those in poverty in this region generally have high aspirations, which they called ‘poverty-of-hope’. It is only a minority of people in poverty who have low aspirations, which Owusu et al. (2008) coined ‘poverty-ofdespair’. The latter group unlike the former are more likely to offend due to their low aspirations. Also, unlike in western societies, which are characterized by individualism, Nigerian and Ghanaian cultural values broadly put huge emphasis on collectivism (Gyekye, 1996; Kuyini, Alhassan, Tollerud, Weld, & Haruna,

A Binary Model of Broken Home

321

2009). Perhaps as a consequence of this, Okpako (2004), Okorodudu (2010), Boakye (2013) and Annor (2014) noted that, the burden of parenting is often distributed among members of one’s extended families and communities. Specifically, Okorodudu (2010, p. 61) explained that, in the Sub-Saharan region, the term ‘parents expand beyond immediate mother and father to include members of the extended family, neighbours and every other person who in one way or the other is involved in upbringing of the child’. This suggests that parenting responsibility is not solely bestowed on biological parents/guardians as in Western societies. Parents in need of social support can easily benefit from a close-knit network that often characterizes collectivist societies (Hofstede, 1980). Even if large family size is common place, it has no detrimental effect on children in Nigeria/Ghana as compared to Western societies (Annor, 2014; Boakye, 2013). The underpinning being, unlike in western nations, the extended family network and older siblings serve as an additional layer of juvenile supervision in Nigeria/Ghana. Drawing from Durkheim’s (2001/1912) theory of social bond, Stets and Carter (2012, p. 122) explained that ‘the motive underlying behaviour in interaction is to maximize and replenish emotional energy’. This is most conducive in collective societies. It is required to buffer the unavoidable emotional distress in everyday parenting. What affects parent-wellbeing also affects parents’ capacity to parent well or otherwise (Belsky, 1984). This highlights that the accessibility of social support in a Nigerian/ Ghanaian context is not significantly influenced by peoples’ economic positions. Shared supervision of children found in Nigeria/Ghana weakens the impact of possible low supervision by direct parents. Support for this is found in Akinlabi and Olatunji’s (2013, p. 21) study, which confirmed that ‘respondents from rich and poor homes do not differ in their upbringing due to economic status’. This suggests that, in contrast to findings in western societies, economic deprivation does not appear to have a negative impact on child rearing in Nigeria/Ghana. This suggests that the buffering impact of high SES on parental divorce and parental death in western societies is not applicable to the Nigeria/Ghana. What does this tell us about the differences between parental death and parental divorce, in terms of their efficacy in predicting juvenile delinquency across these world regions? Given that most western societies can be generalized to be fairly individualistic (Gyekye, 1996; Hofstede, 1980; Selin, 2014), PDI and PDE can be argued to have relatively similar consequences in terms of juvenile delinquency, as both factors reduce parental supervision and parental income.

322

SULEMAN IBRAHIM

Relatedly, given that the negative effects of PDI and PDE are moderated by a family’s socio-economic position, suggests that it is the minority of families (affluent families) to whom most negative outcomes are buffered in terms of parental supervision and poverty. Therefore, it can be deduced that in western societies, the departure of one parent due to death or divorce, is relevant in predicting delinquency in children as illustrated in Fig. 2. This aligns with a multitude of research studies in Western societies according to Amato and Keith’s (1991) meta-analysis (see also Derzon, 2010). Contradictorily, whilst these geopolitical spaces are relatively welfareoriented (Barker, 2013; Pratt, 2008), welfare benefits have minimal impact on families’ economic position in relation to paid childcare (Koopmans, 2010). This suggests that, the buffering effect of welfare on lone-parent families in western societies, if any, is dissolved by the individualistic components of western societies. As Borgers, Dronkers, and Van Praag (1996) have pointed out, the interactions of these two mechanisms, on one hand explain why the negative impacts of parental death in relation to divorce is less in continental Europe more welfare-based, than in the

Conflicting Mechanisms

Lone Parent Family

Welfare State Support

J U V E N I L E

Fig. 2.

D E L I N Q U E N C Y

Broken Home

Individualist Society

Lower income Lower supervision

Parental Divorce or Parental Death

Parental Death Hypothesis in Western Societies.

A Binary Model of Broken Home

323

United States less welfare-based. On the other hand, it reinforces the standpoint that welfare-state-support and the individualistic elements of a society are conflicting mechanisms in the moderation of the effect of loneparenthood due to either divorce or parental death in western nations unlike in Nigeria/Ghana. Even if some studies have noted a slight differential effect of parental death and parental divorce on juvenile delinquency in the West (Free Jr, 1991; Gregory, 1965a, 1965b; Juby & Farrington, 2001; Rutter, 1971), Wells and Rankin’s (1991, p. 88) meta-analysis emphasized that their differences are statistically non-significant. Sharp contrasts between parental death and parental divorce, and also between existing literature and this paper are required for clarity of thought and precision in illustrations. In Nigeria/Ghana, parental death and parental divorce lead to lone-parent families as in western societies. Despite a long standing convergence of Nigerian/Ghanaian literature in homogenizing the differential consequences of parental death and parental divorce as one factor (Boakye, 2013; Ebbe, 1992; Oloruntimehin, 1973; Sanni et al., 2010; Ugwuoke & Duruji, 2015; Weinberg, 1964), this paper proposes a binary model of broken home, because parental divorce and parental death yield distinct and unique slices of realities in Nigeria and Ghana as illustrated in Figs. 3 and 4. Even if some studies in western societies have reported slight differences between the two constructs as aforementioned, unlike these studies that locate a crucial area of their differences in divorce conflicts, this paper, alongside the conflict element of divorce, situates its hypothesis on the unique social facts inherent in Nigeria/Ghana such as elaborate burial rituals as illustrated in Fig. 3, in relation to parental death and other factors such as stigma as illustrated in Fig. 4 in terms of parental divorce. In Nigeria/Ghana, unlike parental death, recent studies have labelled parental divorce as national threats, and public health problem (e.g. Akanbi et al., 2015; Aniche, 2015). Data from the most comprehensive study in this area (Boakye, 2013, p. 268) specifically illustrated that, PDI (OR = 5.1) was more strongly associated with self-reported delinquency than PDE (OR = 1.6.). This suggests that parental divorce, significantly, and to a greater extent than parental death, may induce some compounding variables that underpin children’s involvement in delinquent behaviour as illustrated in Figs. 3 and 4. However, these studies have not offered viable explanations in line with contextual social facts to tease out these more refined variables, but rather generally deployed ‘marital status’ or ‘broken home’ to explain children’s likelihood to be involved in delinquent behaviour. Despite these scholars identifying differences between PDE and

324

SULEMAN IBRAHIM

Buffers

Extended Collectivist Society

Lone Parent Family

Burial

Family

Rites,

Lower income

Support

Empathy

Lower supervision

Buffers low income & low supervision

J U V E N I L E

Fig. 3.

D E L I N Q U E N C Y

Broken Home

Parental Death

Parental Death Hypothesis in Nigeria and Ghana.

PDI, they could have put the findings into context to a greater extent, reflecting more deeply on them in relation to the sociocultural contours of this world region. This invokes Agozino’s (2003, p. 7) idea that unlike a range of marginal voices (poets, novelists) not trained in criminology such as Frantz Fanon, Wole Soyinka, Kofi Awoonor, Ngugi Wa Thiong’o, Chinua Achebe, etc., those trained in criminology and the social sciences in general, rarely challenge mainstream familial definitions and predefined models upon the lived experience of marginal groups. The implication is that in social science, there is little original theory coming out of colonized nations compared to the imperialist countries. These theoretical oversights (the use of the concept of broken home to clump together PDE and PDI when considering the possible link with juvenile delinquency and also contextless generalization) perpetuate a longstanding over-reliance on western insights and claims by colonized nations. As Greenwald, Pratkanis, Leippe, and Baumgardner (1986, p. 227)

325

A Binary Model of Broken Home

Lone Parent Family Stigma Lower income Society School

J U V E N I L E

D E L I N Q U E N C Y

Peers

Lower supervision

Disruption of bonds Distortion of relationships Residential mobility

Parental Divorce Patriarchal context

Fig. 4.

Broken Home

Parental Divorce Hypothesis in Nigeria/Ghana.

beautifully summarized, ‘[T]heory obstructs research progress when the researcher has more faith in the correctness of the theory than in the suitability of the procedures that were used to test it’ and also the fundamental social-cultural fabric of situation, as Igbinovia (1988, p. 154) once emphasized. This suggests that the over-reliance on western criminological theories by former colonized nations, in an imitative fashion rather than in an innovative way is unlikely to decolonize knowledge within criminology (Agozino, 2003, p. 7; Uday, 2013, p. 450). Therefore, in order to overcome this limitation, these particular contours of Sub-Saharan complex social facts (PDE and PDI) require the application of insights from other disciplines such as anthropology. An anthropological lens is crucial to examine the complexities that surround the subjective and socio-cultural ways of life in Nigeria/Ghana that tend to differentiate the impacts of PDE from those of PDI. Whilst PDE is a risk factor, its effects are strongly buffered by tripartite mechanisms: burial rites, extended family support and collectivism. African cultural beliefs and practices emphasize collectivism (Gyekye, 1996), as evidenced by elaborate African burial rites. Burial rites in Nigeria/Ghana unlike in western nations, are major community events and generally last for about two

326

SULEMAN IBRAHIM

weeks (McCauley, 2012; Smith, 2004). This therefore induces authentic empathy towards the bereaved children in the event of the death of a parent and sociologically, ‘unite its adherents in a single moral community called the church’ (Durkheim, 2001/1912, p. 46). Consequently, these sociocultural processes reinforce extended family support. Unlike divorce, death in any family necessitates the ‘coming home’ or ‘being there’ of all extended family members for the immediate bereaved. The dialectic interactions of these mechanisms a gap overlooked in criminology literature (Nigerian/ Ghanaian) are prominent features of the anthropological literature on burial rites and social cohesion (Geschiere & Gugler, 1998; Gugler, 2002; Lentz, 1994; Smith, 2004). In line with these anthropologists, in most African societies including Nigeria and Ghana, it is culturally mandatory to be buried ‘at home’ in people’s places of origin. The elaborateness of burial rites acknowledge the need to ‘be there’ for the children of the deceased and to develop concrete plans on how to implement this support (Mazzucato, Kabki, & Smith, 2006; Smith, 2004). This is one of the ‘most powerful symbolic indicators of the continuing strength of ties to place of origin’ (Smith, 2004, p. 569). It also serves as a mechanism, which absorbs the impact of PDE on children as the burden of parenting is mutually and more enthusiastically shared among the deceased relatives (Lentz, 1994; Gugler, 2002). Given that unlike in western nations, elaborate burial rites reinforce extended family support for children of a lone parent due to parental death, they reduce the likelihood of children’s involvement in criminality as illustrated in Fig. 3. Even if such rituals are not immune to familial conflicts (Geertz, 1977), most burial-based conflicts are primarily geared to maximize the socio-economic benefits for the immediate bereaved (Mazzucato et al., 2006; Smith, 2004). This implies that the two major consequences of parental death poverty and low parental supervision are minimized by these mechanisms available in collectivist societies such as in Nigeria/Ghana. For example, Hampshire et al. (2015), although covering a slightly different field of study parental death and educational attainment, recently noted that in relation to non-orphans, orphans were not significantly disadvantaged due to kinship nurturing available in this world region. One the other hand, whilst the negative impacts of PDE are buffered by sociocultural mechanisms, the numerous detrimental effects of PDI (as shown in Fig. 4) which interpenetrate one another, are amplified via stigmatization: stigmatization at school, at play with peers and in society. This invokes the theory of stigma. According to Goffman’s (1967/1990, p. 3)

A Binary Model of Broken Home

327

theory, stigma can be defined as an ‘attribute that is deeply discrediting [and that lessens the holder] from a whole and usual person to a tainted, discounted one’. Stafford and Scott (1986, p. 80) elaborated on Goffman’s idea and postulated that stigma is a feature of individuals that is in opposition with the collective norm of a social unit. Crocker, Major, and Steele (1998, p. 505) pointed out that ‘stigmatized individuals posses (or are believed to possess) some attribute that conveys a social identity that is devalued in a particular context’. Thus this paper applies the term stigma as Link and Phelan (2001, p. 367) suggested, when its elements: segregation, labelling, stereotyping, status loss, and discrimination co-occur to unfold their outcomes on the stigmatized. Historically, despite the erosion of some African traditions due to colonization, urbanization and industrialization (Igbinovia, 1988), in line with the theory of stigma, parental divorce in Nigeria/Ghana is still seem as a violation of African traditional norm and ‘common-sense’ (Akinlabi & Olatunji, 2013, p. 22). According to socio-cultural ‘common-sense’, it is preferable to be a second or third wife than to bear the agony of being mocked and labelled as prostitute, infertile or evil-spirit possessed when a woman is of age and unmarried. This sets the desperation of unmarried women in motion, pedalling and reinforcing a patriarchal system and polygamous marriages. Consequently, desperations are often soaked-up by and condition some contemporary women in this world region into polygamous systems of marriage. This socio-cultural-platform shapes the parameter with which ‘marriage’ and ‘divorce’ are measured. Therefore, while a marriage accords social prestige, divorce connotes loss of social status (Akinlabi & Olatunji, 2013; Aniche, 2015; Asante et al., 2014; Denga, 1981, 1982). Support for this is reinforced by recent Nigerian/Ghanaian studies on divorce, which revealed that the negative societal reactions levelled against divorce prevent its supposed benefits such as personal freedom and liberty (Aniche, 2015; Asante et al., 2014). How divorce is culturally tainted is evident in Ebbe’s (1992, p. 356) analysis of African family structure as mainly patriarchal. Most importantly, ‘family-name’ is generally seen as supreme, which must be protected and cherished more than members’ rights, tendencies and instinctual inclinations. Arguably, any transgression against what Durkheim (2001/1912) called the collective conscience in the form of divorce, activates the devaluing of not only the direct transgressors’ statuses but anyone associated to them such as the children. This suggests that the conservative sentiments of Nigerian/Ghanaian society that view divorce almost as a taboo, charge transgressors with the sociocultural penalties of

328

SULEMAN IBRAHIM

divorce: stigmatization and humiliation. The sociocultural penalties of divorce can be conceptualized as the culturally defined and location specific negative consequences of divorce, which could be intentional or otherwise bestowed on divorced individuals by other members of a given society. No wonder Durkheim (2001/1912, p. 322) once reminded us, ‘[N]o society can exist that does not feel the need to reaffirm the collective ideas that constitute its personality’. Given that what affects parents’ psychosocial position often affects their children (Belsky, 1984) it is reasonable therefore to construct a link between the sociocultural penalties of divorce received by parents and the negative impacts on their children. Based on the patriarchal and paternalistic nature of the Sub-Saharan sociocultural context (Ajayi & Owumi, 2013; Fourchard, 2005), the social weight of stigma bestowed on children from divorced parents is greater for boys than girls. Conversely, women suffer more than men from the sociocultural penalties of divorce (Asante et al., 2014), possibly due to the legacy of gender inequalities in Nigeria/Ghana (Akotey & Adjasi, 2015; Fawole & Adeoye, 2015). In terms of male juvenile delinquency, the sociological underpinning can be deduced from the principles of paternalism. According to Akinlabi and Olatunji (2013, p. 21) a father is a ‘figure of identification and a significant agent of discipline’ (see also Animasahun, 2014, p. 598). This supports Ebbe’s (1992, p. 356) view: unlike a mother, ‘the father is the supreme judge of the family and the chief enforcer of the norms of the society’. Therefore, boys from divorced parents are generally perceived as ‘incapable’ of possessing ascribed manly attributes in African ways of being, due to as Ebbe (1992) noted, the static hierarchical model of gender relations in Sub-Saharan Africa. This entrenched tradition that promotes machismo in boys, combined with boys’ need as Freud (1991/1927) once suggested, to identify with their ideal models, which are usually fathers, increases their propensity to involve in delinquency if the ‘father-figure’ is absent due to divorce. Also, if a man fails in his marriage(s), he could be seen from a sociocultural subjective lens as having attributes of a ‘failure’, who could not ‘manage’ his wife/wives. According to Achebe (1994), there is a proverb that, if a finger touches oil, it can easily spread to other fingers. This implies that any attributes of unmanliness affixed to a man also affect his sons (Ajayi & Owumi, 2013; Ebbe, 1992; Fourchard, 2005). As illustrated in Fig. 4, if society distinguishes boys based on their parents’ marital status boys from divorced parents and those from ‘healthy-homes’, such a society labels them on the basis of these stereotypes. Society therefore ‘armed with these facts’, stigmatizes the less cherished ones, who then lose their statuses and become

A Binary Model of Broken Home

329

‘abnormal’. The peer group of the stigmatized, ‘orchestrated’ by their significant others, is more likely to distance themselves from them, so as not to be ‘contaminated’, thereby creating social isolation for the bearers of stigma, because as Thomas and Thomas (1928, p. 572) noted, ‘if children believe situations are real, they are real in their consequences’. Given that schools are crucial social contexts for children, peer groups and teachers intentionally or otherwise participate in this circle of ‘othering’ (Goffman, 1967/1990). These are processes that sustain the predicament of stigma, as the stigmatized are most likely to regress educationally, dropout and become involved in anti-social behaviour. This problem is compounded, because as Joireman (2008, p. 1238) explained, unlike in western cultures, in Sub-Saharan African ‘women do not own property but rather, under customary law, women are perceived as properties’. Although women can own property under public law, the enforcement of public law at the expense of customary law is against the grain of Sub-Saharan Africa cultural practice (Mamdani, 1996, p. 50). The implication is that parental divorce necessitates residential mobility for women/children, unlike parental death (Amato & Anthony, 2014, p. 382). Given that mothers who often have custody of children have no property rights if they are lone parents via divorce (Joireman, 2008), this may result in higher levels of residential mobility, which could induce stress on mothers and children and a distortion of relationship between them, which could concomitantly as Hoeve et al. (2012) noted, further put a strain on the bonds between them (see also Mack, Leiber, Featherstone, & Monserud, 2007). What does this tell us about the predicament of boys from divorced parents in terms of delinquency? Given that, men rather than women in this context, are culturally and predominantly raised to be bread winners, implies that the possibility of having a decent accommodation after divorce for women/children is limited unless an ex-husband or a woman’s family of origin has a fleet of houses. It is only the minority of people who possess such wealth in Sub-Saharan Africa (Joireman, 2008; Owusu et al., 2008). This phenomenon unlike parental death merged with stigmatization and other sociocultural penalties of divorce, place children at a higher risk of being stereotyped, labelled, bullied, isolated by peers, new neighbours and even discriminated intentionally or otherwise by teachers. In a nutshell, whilst the predictive strength of parental death as a risk factor is shielded by collectivism, extended family support and elaborate burial ceremonies, that of parental divorce alongside other variables as illustrated in Fig. 4, is amplified by stigma on multiple levels: school, peers, society. Beyond the amplification of the negative

330

SULEMAN IBRAHIM

impacts of parental divorce on children’s involvement in criminality, the scars of stigma often accompany the stigmatized from childhood to adulthood. For example, marriage proposals could be accepted or rejected based on certain sociocultural conditions, which include among other variables, if a prospective spouse is from a disorganized subcultural home, that is if an individual has divorced parents or otherwise (Aniche, 2015). This is because marriage in this context is not just between two persons, but between two families, and lineages, and even distant communities (Amponsah et al., 2006; Mivanyi, 2006). Arguably, the scars of stigma conferred on juveniles in this world region could potentially have long-lasting effects.

CRITIQUE OF EVIDENCE However, in critiquing the evidence, caution should be applied in the interpretation of the above culture-induced differences. Whilst cross-cultural comparison of parental death and parental divorce is not immune to limitations, it however boosts causal inferences bestowed on the impacts of parental divorce across difference geopolitical spaces (Amato & Anthony, 2014). In other words, cross-national comparisons tend to neglect the complexity and diversity of lives within targeted countries. Nevertheless, there is consistent evidence across cultures in support of the negative effect of parental divorce, specifically due to the ‘marital-conflict’ element of most divorces. The main limitation of the above comparison is evident in the methodological differences between western studies and Nigeria/Ghana studies. Whilst all Nigerian/Ghanaian studies are based on cross-sectional data, longitudinal data is central to the majority of western studies. Given that correlational studies based on self-report and their conclusions are not certainties, they require extensive empirical confirmation, ideally longitudinal data, to confirm to their efficacy. As Mitchell, Vella-Brodrick, and Klein (2010, p. 37), although on a different tangent well-being, once stressed, ‘how meaningful research evidence in social science is will need to be determined by future longitudinal research’. Besides, all four studies (Amato & Anthony, 2014; Aughinbaugh et al., 2005; Cherlin et al., 1998; Li, 2007) specifically on effects of parental divorce on children outcomes that deployed ‘fixed effect models’ are based on Western societies. According to Amato and Anthony (2014, p. 371) unlike all models available to cross-sectional studies, ‘[F]ixed effects models have the advantage of controlling for all time-invariants variables’. Studies

A Binary Model of Broken Home

331

based on cross-sectional data addressed these issues, by putting the interviewees on the power-chair in the interview process, that is allowing participants to develop the trend of the interview itself and/or controlling for a huge amount of potential confounding variables. A key limitation of these methods includes the impossibility of knowing if all relevant factors have been addressed and in the case of quantitative studies, if all variables have been included in the statistical model. Secondly, unlike longitudinal studies, cross-sectional studies are unable to measure the effect of marital conflict before the event of divorce to ascertain if similar outcomes can be establish at both times or otherwise. Support for this is found in Miller’s (2014, p. 276) study that concluded that, parental-divorce juvenile-delinquency linkages are often present before the occurrence of the actual divorce itself (see also Buehler et al., 1997; Sun, 2001).

CONCLUSION This paper has interrogated mainstream assumptions and Nigerian/ Ghanaian scholarly endeavours on the taken-for-granted singular model of broken home. Specifically, anthropological insights on burial rites and Goffman’s (1967/1990) theory of stigma provided a unique window to contrast the differential impacts of parental death and parental divorce with more refined variables unlike existing data, that is this paper has deployed the application of the sociocultural penalties of divorce such as stigma in terms of PDI and burial ceremonies, kinship nurturing, etc., in relation to PDE. This has helped to fill in the missing gap in comparative criminology literature as illustrated in Figs. 3 and 4, because the above oversights are not confined to mainstream literature, it is a feature of regional literature from Nigeria/Ghana. It has also exposed the uncritical imitation of criminology claims in western nations by colonized nations, which had hitherto been the norm, perhaps due to lack of theoretical originality coming out of colonized nations (Agozino, 2003). The implication is the obscurity of ‘what is true of all societies and what is true of one society at one point in time and space’ (Bendix, 1963, p. 532). If the acknowledgement of diversity is taken seriously, it could, and perhaps should broaden the geographical scope of criminological endeavours that will, as Tankebe et al. (2014) suggested, take the experience of non-western regions unconditionally serious. Finally, the main limitation of this paper is patterned by the existing literature. Whilst most western assumptions are informed by longitudinal

332

SULEMAN IBRAHIM

data, Nigerian/Ghanaian literature is cross-sectional based. However, even if this pattern of existing literature across these distinct world regions is considered as a weakness of this paper in drawing a ‘balanced’ comparative analysis, it by no means undermine the importance of its main achievement: the deconstruction of the singular model of ‘broken home’ and proposal of a rather binary model the parental death and parental divorce hypotheses based on more refined variables. Future research empirical studies, are needed to substantiate this claim beyond theoretical realm.

NOTE 1. Socioeconomic cybercrimes, primarily driven by commercial benefits, can be conceptualized as intentional fraud-based crimes that are computer or/and internetmediated, such as online fraud, romance scam, and e-embezzlement.

ACKNOWLEDGEMENTS I am grateful to Claire P. Monks for proofreading this paper. I thank Lucinda Platt, Onwubiko Agozino, Justice Tankebe and Sally Mann, for their academic advice. I also thank the anonymous reviewers for useful comments on earlier draft of this paper. My profound gratitude goes to God Almighty for his grace, my family, and friends (Dave, Jovan, Craig, Ben, Sophie) for their emotional support. I am also grateful to all CDT members, the EPSRC and the UK government as part of the Centre for Doctoral Training in Cyber Security at Royal Holloway, University of London.

REFERENCES Achebe, C. (1994). Things fall apart. London: Everyman’s Library. Adegoke, N. (2014). Kidnapping, security challenges and socio-economic implications to the Niger delta region of Nigeria. Centrepoint Journal (Humanities Edition), 16(2), 205 216. Aderibigbe, A. M., & Olla, J. O. (2014). Kidnapping terrorism and political violence, implication for strategic security management in Nigeria. Developing Country Studies, 4(6), 16 19. Adu-Mireku, S. (2002). Fear of crime among residents of three communities in Accra, Ghana. International Journal of Comparative Sociology, 43(2), 153 168.

A Binary Model of Broken Home

333

Agozino, B. (2003). Counter-colonial criminology: A critique of imperialist reason. London: Pluto Press. Agozino, B. (2004). Imperialism, crime and criminology: Towards the decolonisation of criminology. Crime, Law and Social Change, 41(4), 343 358. Agozino, B. (2010). Editorial: What is criminology? A control-freak discipline! African Journal of Criminology and Justice Studies, 4(1), 1 20. Ajaegbu, O. O. (2012). Rising youth unemployment and violent crime in Nigeria. American Journal of Social Issues and Humanities, 2(5), 315 321. Ajayi, J., & Owumi, B. (2013). Socialization and child rearing practices among Nigerian ethnic groups. Academic Journal of Interdisciplinary Studies, 2(2), 249 256. Akanbi, M., Gbadebo, B., Adekola, P., Olawole-Isaac, A., Sowunmi, A., & Godwin, H. (2015). Influence of family structure on adolescent sexual behaviour in Mushin local government area of Lagos state, Nigeria. Australian Journal of Social Science Research, 2(4), 1 16. Akanle, O., & Olutayo, O. A. (2011). Kinship construction variability among Nigerian international migrants: The context of contemporary diaspora. Human Affairs, 21(4), 470 480. Akers, R., & Burgess, R. (1966). A differential association-reinforcement theory of criminal behavior. Social Problems, 14, 363 383. Akinlabi, F. B., & Olatunji, B. F. (2013). Social, economic and educational implications of parental separation on upbringing of school children in Ekiti state. Global Advanced Research Journal of Peace, Gender and Development Studies, 2(2), 021 025. Akotey, J. O., & Adjasi, C. K. (2015). Exploring the effect of microinsurance on asset inequality among households in Ghana. The Journal of Developing Areas, 49(2), 373 398. Althusser, L. (1971/2001). Lenin and philosophy and other essays. Delhi: Monthly Review Books. Amato, P., & Anthony, C. (2014). Estimating the effects of parental divorce and death with fixed effects models. Journal of Marriage and Family, 76(2), 370 386. Amato, P. R. (2014). The consequences of divorce for adults and children: An update. ˇ asopis za opc´a drustvena ˇ ˇ Drustvena istrazivanja-C ˇ pitanja, 1, 5 24. Amato, P. R., & Keith, B. (1991). Parental divorce and the well-being of children: A metaanalysis. Psychological Bulletin, 110(1), 26. Amponsah, B., Akotia, C., & Olowu, A. (2006). Ghana. In J. Georgas, J. W. Berry, F. J. van de Vijver, C¸. Kagitc¸ibasi, & Y. H. Poortinga (Eds.), (2006), Families across cultures: A 30-nation psychological study. Cambridge: Cambridge University Press. Aniche, A. I. (2015). Sociological approach to study divorce as a social problem and Nigerian situation: Imperatives for sustainable solutions. Global Journal of Applied, Management and Social Sciences, 8, 13 18. Animasahun, R. A. (2014). Marital conflict, divorce and single parenthood as predictors of adolescent antisocial behaviour in Ibadan. British Journal of Education, Society and Behavioural Science, 4(5), 592 602. Annor, F. (2014). Managing work and family demands: The perspectives of employed parents in Ghana. In Work family interface in Sub-Saharan Africa (pp. 17 36). New York, NY: Springer International Publishing. Asante, K. O., Osafo, J., & Nyamekye, G. K. (2014). An exploratory study of factors contributing to divorce among married couples in Accra, Ghana: A qualitative approach. Journal of Divorce & Remarriage, 55(1), 16 32.

334

SULEMAN IBRAHIM

Aughinbaugh, A., Pierret, C. R., & Rothstein, D. S. (2005). The impact of family structure transitions on youth achievement: Evidence from the Children of the NLSY79. Demography, 42, 447 468. Bankole, C., & Chipaca, A. (2014). Juvenile delinquency in Angola. Criminology and Criminal Justice, 14(1), 61 76. Barker, V. (2013). Nordic exceptionalism revisited: Explaining the paradox of a Janus-faced penal regime. Theoretical Criminology, 17(1), 5 25. Becker, H. (1963/1997). Outsiders: Studies in sociology of deviance. New York, NY: Simon and Schuster Ltd. Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 55(1), 83 96. Bendix, R. (1963). Concepts and generalizations in comparative sociological studies. American Sociological Review, 28(4), 532 539. Biblarz, T. J., & Gottainer, G. (2000). Family structure and children’s success: A comparison of widowed and divorced single-mother families. Journal of Marriage and Family, 62(2), 533 548. Boakye, K. (2013). Correlates and predictors of juvenile delinquency in Ghana. International Journal of Comparative and Applied Criminal Justice, 37(4), 1 22. Borgers, N., Dronkers, J., & Van Praag, B. M. (1996). The effects of different forms of twoand single-parent families on the well-being of their children in Dutch secondary education. Social Psychology of Education, 1(2), 147 169. Bowlby, J. (1988). A secure base. Abingdon: Routledge. Brathwaite, F. (1996). Some aspects of sentencing in the criminal justice system of Barbados. Caribbean Quarterly, 42(2 3), 101 118. Buehler, C., Anthony, C., Krishnakumar, A., Stone, G., Gerard, J., & Pemberton, S. (1997). Interparental conflict and youth problem behaviors: A meta-analysis. Journal of Child and Family Studies, 6, 233 247. Cain, M. (2000). Orientalism, occidentalism and the sociology of crime. British Journal of Criminology, 40(2), 239 260. Cherlin, A. J., McRae, C., & Chase-Lansdale, P. L. (1998). Effects of parental divorce on mental health throughout the life course. American Sociological Review, 63, 239 249. Cohen, S. (2008/1988). Against criminology. New Brunswick, NJ: Brunswick Publishers. Creighton, M. J., Park, H., & Teruel, G. M. (2009). The role of migration and single motherhood in upper secondary education in Mexico. Journal of Marriage and Family, 71, 1325 1339. Crocker, J., Major, B., & Steele, C. (1998). Social stigma. In D. T. Gilbert & S. T. Fiske (Eds.), The handbook of social psychology (Vol. 2, pp. 504 553). Boston, MA: McGraw-Hill. Denga, D. (1981). Juvenile delinquency among polygynous families in Nigeria. The Journal of Social Psychology, 114(1), 3 7. Denga, D. (1982). Childlessness and marital adjustment in Northern Nigeria. Journal of Marriage and the Family, 44(3), 799 802. Derzon, J. H. (2010). The correspondence of family features with problem, aggressive, criminal, and violent behavior: A meta-analysis. Journal of Experimental Criminology, 6, 263 292. Doyon-Martin, J. (2015). Cybercrime in West Africa as a result of transboundary e-waste. Journal of Applied Security Research, 10(2), 207 220.

A Binary Model of Broken Home

335

Dugdale, R. (1887). “The jukes”: A study in crime, pauperism, disease and hereditary. New York, NY: G. P. Putman and Sons. Durkheim, E. (1965/1885). The rules of sociological method. New York, NY: The Free Press. Durkheim, E. (2001/1912). The elementary forms of religious life. Oxford: Oxford University Press. Ebbe, O. N. (1992). Juvenile delinquency in Nigeria: The problem of application of Western theories. International Journal of Comparative and Applied Criminal Justice, 16(1 2), 353 370. Erikson, E. H. (1995). Childhood and society. New York, NY: Vintage. Fanon, F. (1967/2008). Black skin, white mask. New York, NY: Pluto Press. Farrington, D. (2005). Childhood origins of antisocial behaviour. Clinical Psychology and Psychotherapy, 12, 177 190. Farrington, D. (2012a). Childhood risk factors for young adult offending: Onset and persistence. In F. Lo¨sel, A. E. Bottoms, & D. P. Farrington (Eds.), Young adult offenders: Lost in transition? London: Routledge. Farrington, D. (2012b). Predictors of violent young offenders. In B. C. Feld & D. M. Bishop (Eds.), The Oxford handbook of juvenile crime and juvenile justice. Oxford: Oxford University Press. Farrington, D. P. (2011). Families and crime. In J. Q. Wilson & J. Petersilia (Eds.), Crime and public policy (pp. 130 157). Oxford: Oxford University Press. Farrington, D. P., Piquero, A. R., Nagin, D. S., & Moffitt, T. E. (2010). Trajectories of offending and their relation to life failure in late middle age: Findings from the Cambridge study in delinquent development. Journal of Research in Crime and Delinquency, 47(2), 151 173. Fawole, O. I., & Adeoye, I. A. (2015). Women’s status within the household as a determinant of maternal health care use in Nigeria. African Health Sciences, 15(1), 217 225. Feit, E. (1968). Military coups and political development: Some lessons from Ghana and Nigeria. World Politics: A Quarterly Journal of International Relations, 20(2), 179 193. Fergusson, D. M., McLeod, G. F., & John Horwood, L. (2014). Parental separation/divorce in childhood and partnership outcomes at age 30. Journal of Child Psychology and Psychiatry, 55(4), 352 360. Fincher, C., & Thornhill, R. (2014). Collectivism individualism, family ties, and philopatry, In The parasite-stress theory of values and sociality (pp. 113 170). New York, NY: Springer International Publishing. Foucault, M. (1977/1991). Discipline and punish: The birth of the prison. London: Penguin. Fourchard, L. (2005). African urban history: Past and present perspective. Lagos Historical Review, 5(1), 1 21. Free, M. D., Jr. (1991). Clarifying the relationship between the broken home and juvenile delinquency: A critique of the current literature. Deviant Behavior, 12(2), 109 167. Freud, S. (1991/1927). Civilization, society and religions: Group psychology and the analysis of the ego, future of an illusion and civilization and its discontents. New York, NY: Penguin Books Ltd. Geertz, C. (1977). The interpretation of cultures. New York, NY: Basic Books. Geschiere, P., & Gugler, J. (1998). Introduction: The urban rural connection: Changing issues of belonging and identification. Africa, 68(03), 309 319. Go, J. (2013). For a postcolonial sociology. Theory and Society, 42(1), 25 55.

336

SULEMAN IBRAHIM

Goffman, E. (1967/1990). Stigma: Notes on the management of spoiled identity. London: Penguin. Gorman, G. H., Eide, M., & Hisle-Gorman, E. (2010). Wartime military deployment and increased pediatric mental and behavioral health complaints. Pediatrics, 126, 1058 1066. Greenwald, A., Pratkanis, A., Leippe, M., & Baumgardner, M. (1986). Under what conditions does theory obstruct research progress? Psychological Review, 93, 216 229. Gregory, I. (1965a). Anterospective data following childhood loss of a parent: I. Delinquency and high school dropout. Archives of General Psychiatry, 13(2), 99 109. Gregory, I. (1965b). Anterospective data following childhood loss of a parent: II. Pathology, performance, and potential among college students. Archives of General Psychiatry, 13(2), 110 120. Gugler, J. (2002). The son of the hawk does not remain abroad: The urban rural connection in Africa. African Studies Review, 45(01), 21 41. Gyekye, K. (1996). African cultural values: An introduction. Accra: Sankofa. Hall, S. (1990). Cultural identity and diaspora. Retrieved from http://sites.middlebury.edu/ nydiasporaworkshop/files/2011/04/D-OA-HallStuart-CulturalIdentityandDiaspora.pdf. Accessed on June 20, 2014. Hampshire, K., Porter, G., Agblorti, S., Robson, E., Munthali, A., & Abane, A. (2015). Context matters: Fostering, orphanhood and schooling in Sub-Saharan Africa. Journal of Biosocial Science, 47(02), 141 164. Hoeve, M., Stams, G. J. J., van der Put, C. E., Dubas, J. S., van der Laan, P. H., & Gerris, J. R. (2012). A meta-analysis of attachment to parents and delinquency. Journal of Abnormal Child Psychology, 40(5), 771 785. Hofstede, G. (1980). Culture’s consequences: International differences in work related values. Beverly Hills, CA: Sage. Igbinovia, P. (1988). Perspectives on juvenile delinquency in Africa. International Journal of Adolescence and Youth, 1(2), 131 156. Jampaklay, A. (2006). Parental absence and children’s school enrollment: Evidence from a longitudinal study in Kanchanaburi, Thailand. Asian Population Studies, 2, 93 110. Joireman, S. F. (2008). The mystery of capital formation in Sub-Saharan Africa: Women, property rights and customary law. World Development, 36(7), 1233 1246. Juby, H., & Farrington, D. P. (2001). Disentangling the link between disrupted families and delinquency sociodemography, ethnicity and risk behaviours. British Journal of Criminology, 41(1), 22 40. Kalunta-Crumpton, A. (2004). Criminology and orientalism. In B. Agozino & A. KaluntaCrumpton (Eds.), Pan-African issues in crime and justice. Hants: Ashgate Publishing. 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. Kelly, J. B., & Emery, R. E. (2003). Children’s adjustment following divorce: Risk and resilience perspectives. Family Relations, 52, 352 362. Kitossa, T. (2012). Criminology and colonialism: Counter colonial criminology and the Canadian context. Journal of Pan African Studies, 4(9), 204 226. Koopmans, R. (2010). Trade-offs between equality and difference: Immigrant integration, multiculturalism and the welfare state in cross-national perspective. Journal of Ethnic and Migration Studies, 36(1), 1 26.

A Binary Model of Broken Home

337

Kuyini, A. B., Alhassan, A. R., Tollerud, I., Weld, H., & Haruna, I. (2009). Traditional kinship foster care in northern Ghana: The experiences and views of children, carers and adults in Tamale. Child & Family Social Work, 14(4), 440 449. LaFree, G. (2007). Expanding the domain of criminology: 2006 presidential address for the American society of criminology. Criminology, 45(1), 1 31. Lee, H., Porter, L., & Megan, C. (2014). Consequences of family member incarceration impacts on civic participation and perceptions of the legitimacy and fairness of government. The Annals of the American Academy of Political and Social Science, 651(1), 44 73. Lentz, C. (1994). Home, death and leadership: Discourses of an educated elite from NorthWestern Ghana. Social Anthropology, 2(2), 149 169. Lesthaeghe, R. (2010). The unfolding story of the second demographic transition. Population and Development Review, 36(2), 211 251. Li, J. A. (2007). The kids are OK: Divorce and children’s behavior problems. Working Paper No. WR-489, RAND Corporation, Santa Monica, CA. Link, B. G., & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27, 363 385. Loeber, R. (1982). The stability of antisocial and delinquent child behavior: A review. Child Development, 53(6), 1431 1446. Loeber, R., Farrington, D. P., Jolliffe, D., Stouthamer-Loeber, M., & Kalb, L. M. (2001). The concentration of offenders in families, and family criminality in the prediction of boys’ delinquency. Journal of Adolescence, 24(5), 579 596. Loeber, R., Menting, B., Lynam, D. R., Moffitt, T. E., Stouthamer-Loeber, M., Stallings, R., & Pardini, D. (2012). Findings from the Pittsburgh youth study: Cognitive impulsivity and intelligence as predictors of the age-crime curve. Journal of the American Academy of Child & Adolescent Psychiatry, 51, 1136 1149. Mack, K. Y., Leiber, M. J., Featherstone, R. A., & Monserud, M. A. (2007). Reassessing the family-delinquency association: Do family type, family processes, and economic factors make a difference? Journal of Criminal Justice, 35(1), 51 67. Mamdani, M. (1996). Citizen and subject: Contemporary Africa and the legacy of late colonialism. Princeton, NJ: Princeton University Press. Mazzucato, V., Kabki, M., & Smith, L. (2006). Transnational migration and the economy of funerals: Changing practices in Ghana. Development and Change, 37(5), 1047 1072. McCauley, J. F. (2012). Africa’s new big man rule? Pentecostalism and patronage in Ghana. African Affairs, 112(446), 1 21. Miller, L. (2014). Juvenile crime and juvenile justice: Patterns, models, and implications for clinical and legal practice. Aggression and Violent Behavior, 19(2), 122 137. Mitchell, J., Vella-Brodrick, D., & Klein, B. (2010). Positive psychology and the internet: A mental health opportunity. Sensoria: A Journal of Mind, Brain & Culture, 6(2), 30 41. Mivanyi, Y. C. (2006). Nigeria. In J. Georgas, J. W. Berry, F. J. van de Vijver, C¸. Kagitc¸ibasi, & Y. H. Poortinga (Eds.), (2006), Families across cultures: A 30-nation psychological study. Cambridge: Cambridge University Press. Moffitt, T., Lyam, D., Silva, P., & Bartusch, D. (1997). Is age important? Test a general theory of developmental and anti-social behaviour. Criminology, 35(1), 13 48. Moffitt, T. E. (1993). Adolescence-limited and life-course persistent antisocial behaviour: A developmental taxonomy. Psychological Review, 100(4), 674 701.

338

SULEMAN IBRAHIM

Moffitt, T. E., Jaffee, S. R., Bowes, L., Ouellet-Morin, I., Fisher, H. L., Merrick, M. T., & Arseneault, L. (2013). Safe, stable, nurturing relationships break the intergenerational cycle of abuse: A prospective nationally representative cohort of children in the United Kingdom. Journal of Adolescent Health, 53(4), S4 S10. Mulder, E., Brand, E., Bullens, R., & Van Marle, H. (2010). A classification of risk factors in serious juvenile offenders and the relation between patterns of risk factors and recidivism. Criminal Behaviour and Mental Health, 20(1), 23 38. Myers, G. A. (2006). The unauthorized city: Late colonial Lusaka and postcolonial geography. Singapore Journal of Tropical Geography, 27(3), 289 308. Nelken, D. (2010). Comparative criminal justice: Making sense of difference. London: Sage. Okorodudu, G. N. (2010). Influence of parenting styles on adolescent delinquency in delta central senatorial district. Edo Journal of Counselling, 3(1), 58 86. Okpako, J. E. F. (2004). Parenting the Nigerian adolescents towards smooth transition to adulthood. In I. A. Nwazuoke, O. Bamgbose, & O. A. Morokola (Eds.), Contemporary Issue and Research in Adolescents (pp. 275 288). Ibadan, Nigeria: Ibadan Omoade Printing Press. Oloruntimehin, O. (1973). A study of juvenile delinquency in a Nigeria City. British Journal of Criminology, 13, 157 169. Onyeozili, E. C. (2004). Gunboat criminology and the colonialization of Africa. In B. Agozino & A. Kalunta-Crumpton (Eds.), Pan-African issues in crime and justice. Hants: Ashgate publishing. Otoo, F. E. (2014). Constraints of international volunteering: A study of volunteer tourists to Ghana. Tourism Management Perspectives, 12, 15 22. Otu, N. (1999). Colonialism and the criminal justice system in Nigeria. International Journal of Comparative and Applied Criminal Justice, 23(2), 293 306. Owusu, G., Agyei-Mensah, S., & Lund, R. (2008). Slums of hope and slums of despair: Mobility and livelihoods in Nima, Accra. Norwegian Journal of Geography, 62, 180 190. Paris, R., DeVoe, E. R., Ross, A. M., & Acker, M. (2010). When a parent goes to war: Effects of parental deployment on very young children and implications for intervention. American Journal of Orthopsychiatry, 80, 610 618. Pieterse, J. (1989). Empire and emancipation: Power and liberation on a world scale. New York, NY: Praeger. Piquero, A., Welsh, B., Tremblay, R., Farrington, D., & Jennings, W. (2009). Effects of early family/parenting training programs on antisocial behavior and delinquency. Journal of Experimental Criminology, 5, 83 120. Pratt, J. (2008). Scandinavian exceptionalism in an era of penal excess part I: The nature and roots of Scandinavian exceptionalism. British Journal of Criminology, 48(2), 119 137. Price, C., & Kunz, J. (2003). Rethinking the paradigm of juvenile delinquency as related to divorce. Journal of Divorce & Remarriage, 39(1 2), 109 133. Ranuga, T. (1986). Frantz Fanon and black consciousness in Azania (South Africa). Phylon, 47(3), 182 191. Rutter, M. (1971). Parent-child separation: Psychological effects on the children. Journal of Child Psychology and Psychiatry, 12(4), 233. Said, E. (1994). Culture and imperialism. London: Vintage. Sampson, E. E. (2000). Reinterpreting individualism and collectivism: Their religious roots and monologic versus dialogic person other relationship. American Psychologist, 55(12), 1425 1432.

A Binary Model of Broken Home

339

Sampson, R., & Laub, J. (2005). A general age-graded theory of crime: Lessons learned and the future of life-course criminology. Integrated developmental and life course theories of offending, 14, 165 182. Sanni, K., Modo, F., Uduh, N., Ezeh, L., & Okediji, A. (2010). Family types and juvenile delinquency issues among secondary school students in Akwa Ibom state, Nigeria: Counseling implications. Journal of Social Science, 23(1), 21 28. Sanni, L., & Oyibo, J. (2013). Housing and deviants among civil servants’ adolescent children in Lokojah, Nigeria. African Journal for Psychological Studies of Social Issues, 16(1), 37 50. Savage, J. (2014). The association between attachment, parental bonds and physically aggressive and violent behavior: A comprehensive review. Aggression and Violent Behavior, 19(2), 164 178. Selin, H. (2014). Parenting across cultures: Childrearing, motherhood and fatherhood in nonwestern cultures. Dordrecht: Springer. Shaw, C., & Mckay, H. (1948). Juvenile delinquency and urban areas. Chicago, IL: University of Chicago Press. Shaw, C. R., & McKay, H. D. (1932). Are broken homes a causative factor in juvenile delinquency? Social Forces, 10(4), 514 524. Skeggs, B. (2013). Class, self, culture. London: Routledge. Skinner, A. T., Bacchini, D., Lansford, J. E., Godwin, J. W., Sorbring, E., Tapanya, S. … Pastorelli, C. (2014). Neighborhood danger, parental monitoring, harsh parenting, and child aggression in nine countries. Societies, 4(1), 45 67. Smith, D., & McVie, S. (2003). Theory and method in the Edinburgh study of youth transitions and crime. British Journal of Criminology, 43(1), 169 195. Smith, D. J. (2001). Kinship and corruption in contemporary Nigeria. Ethnos, 66(3), 344 364. Smith, D. J. (2004). Burials and belonging in Nigeria: Rural urban relations and social inequality in a contemporary African ritual. American Anthropologist, 106(3), 569 579. Smith, D. J. (2011). Stretched and strained but not broken: Kinship in contemporary Nigeria. In A. M. Gonza´lez, L. de Rose, & F. Oloo (Eds.) Frontiers of globalization: Kinship and family structures in Africa (pp. 31 69). Trenton, NJ: Africa World Press. Smith, L. T. (1999). Decolonizing methodologies: Research and indigenous peoples. Sidney: Zed books. Stafford, M. C., & Scott, R. R. (1986). Stigma deviance and social control: Some conceptual issues. In S. C. Ainlay, G. Becker, & L. M. Coleman (Eds.), The dilemma of difference. New York, NY: Plenum. Staples, R. (1960). White racism, black crime, and American justice: An application of the colonial model to explain crime and race. Phylon, 36(1), 14 22. Stets, J., & Carter, M. (2012). A theory of self for the sociology of morality. American Sociological Review, 77(1), 120 140. Sun, Y. (2001). Family environment and adolescents’ well-being before and after parents’ marital disruption: A longitudinal analysis. Journal of Marriage and Family, 63, 697 713. Tade, O. (2013). A spiritual dimension to cybercrime in Nigeria: The ‘yahoo plus’ phenomenon. Human Affairs, 23(4), 689 705. Tankebe, J., Hills, A., & Bankole, C. (2014). Emerging issues of crime and criminal justice in sub-Saharan Africa. Criminology and Criminal Justice, 14(1), 3 7. Thomas, W. I., & Thomas, D. S. (1928). The child in America: Behaviour problems and programmes. New York, NY: Alfred Knopf.

340

SULEMAN IBRAHIM

Uday, C. (2013). The case for a postcolonial approach to the study of politics. New Political Science, 35(3), 479 491. Ugwuoke, C. U., & Duruji, O. U. (2015). Family instability and juvenile delinquency in Nigeria: A study of Owerri municipality. Journal of Humanities and Social Science, 20(1), 40 45. Warner, J. (2011). Understanding cyber-crime in Ghana: A view from below. The International Journal of Cyber Criminology, 5, 736 749. Weinberg, S. K. (1964). Juvenile delinquency in Ghana: Comparative analysis of delinquents and nondelinquents. Journal of Criminal Law, Criminology, and Police Science, 55, 471 481. Weinberg, S. K. (1965). Urbanization and male delinquency in Ghana. Journal of Research in Crime and Delinquency, 2, 85 94. Wells, L. E., & Rankin, J. H. (1991). Families and delinquency: A meta-analysis of the impact of broken homes. Social Problems, 38(1), 71 93. Welsh, B. C., & Farrington, D. (2007a). Saving children from a life of crime: Early risk factors and effective interventions. New York, NY: Oxford University Press. Welsh, B. C., & Farrington, D. P. (2007b). Save children from a life of crime. Criminology & Public Policy, 6(4), 871 879. Wildeman, C., Wakefield, S., & Turney, K. (2013). Misidentifying the effects of parental incarceration? A comment on Johnson and Easterling (2012). Journal of Marriage and Family, 75, 252 258. Wissink, I. B., Asscher, J. J., Dekovic´, M., Prinzie, P., & Stams, G. J. J. (2014). Delinquent behavior, poor relationship quality with parents, and involvement with deviant peers in delinquent and nondelinquent adolescents different processes, informant bias, or both? International Journal of Offender Therapy and Comparative Criminology, 58(9), 1001 1019.

DELINQUENCY, GENDER, AND SOCIAL CONTROL WITHIN THE FAMILIAL CONTEXT Melissa A. Menasco ABSTRACT Purpose This study is a comparison of male and female delinquent behaviors utilizing a large sample in a test of social control theory. Methodology/approach A sample of 8,363 adolescent 10th graders was drawn from the U.S. National Educational Longitudinal Study (NELS). The sample consisted of 3,774 males and 4,589 females. This work utilized logistic regression and ordinary least squares to determine whether adolescent behaviors such as smoking cigarettes, drinking alcohol, using marijuana, and other delinquent behaviors are linked to weak social bonds. Findings The findings of this study provide limited substantiation of social control theory for both sexes. Females who were found to have strong social bonds were less likely to engage in delinquent behaviors than males. For both sexes, the strongest element of the social bond was the element of belief.

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 341 382 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009015

341

342

Research limitations/implications United States.

MELISSA A. MENASCO

These results may be specific to the

Practical implications Understanding contributing factors to adolescent substance use and delinquency will assist in developing social policy that will support families. Originality/value This study provides insight into the differentiated nature of the social bond for males and females. Knowing that belief is an integral component in determining the strength of the social bond will aid in the development of social policies. Keywords: Social control; substance use; adolescence; delinquency

As originally proposed by Hirschi (1969), social control theory focuses on the nature of the relationship, or the social bond, between the individual and society. Hirschi’s paradigm suggests that it is the quality of the relationship between adolescents and their social environment as well as their attitudes towards society that determine delinquent behavior. The social bond provides a means for measuring these relationships and attitudes. Attachment, commitment, involvement, and belief comprise what Hirschi described as the elements of the social bond. The social bond was the means of instilling in the individual a sense of their relationship to society. Each element operates in its own way toward the successful integration of the societal norms by the individual. Successful integration of social norms, according to Hirschi, produced individuals who conformed to societal expectations. Those individuals who had weakened social bonds either through lack of attachment, commitment, involvement, belief, or all of the elements of the social bond, were he believed, those inclined to a life of deviant behavior. The four components of the social bond consist of: (1) attachment, (2) commitment, (3) involvement, and (4) belief. Attachment is described as the affectional tie between parent and child. Commitment is described as the desire to attain societal goals. Involvement refers to the activity level of children. Belief refers to the child’s system of core values. These four components of the social bond operate in concert to develop the child’s sense of conformity to societal norms. According to this theory, weakened social

Delinquency, Gender, and Social Control

343

bonds for adolescents are the precursors of juvenile delinquency. Conversely, adolescents with strong social bonds believe in the importance of conforming to social norms and, consequently, exhibit socially approved behaviors. Previous studies have focused on cross-sectional approaches to determining the impact of the social bond (Cernkovich & Giordano, 1992; Eve, 1978; Hindelang, 1973; Huebner & Betts, 2002; Jenkins, 1995; Jensen & Brownfield, 1983; Johnson, 1987; Krohn & Massey, 1980). With few exceptions, these studies have substantiated, at least partially, the strength of the social bond in predicting delinquent behavior (Cernkovich & Giordano, 1992; Eve, 1978; Foshee & Bauman, 1992; Hindelang, 1973; Huebner & Betts, 2002; Jenkins, 1995; Jensen & Brownfield, 1983; Johnson, 1987; Krohn & Massey, 1980; Liska & Reed, 1985; Rankin & Kern, 1994; Wiatrowski, Griswold, & Roberts, 1981; Wright & Cullen, 2001). Utilizing a sample from the 2000 National Education Longitudinal Study (NELS), this study will address the following topics. First, the research will examine the tenth grade respondents’ attitudes toward family, school, life goals, and delinquent acts. Second, the research will examine the relationship between the male/female social bond and delinquency during the same timeframe. This will serve to test the original thesis of social control theory: that the weakened social bond is associated with delinquent behavior as well as provide insight into its effect upon the sexes. The validity of social control theory is further substantiated by a demonstration of the breadth of its insights concerning delinquency.

THE SOCIAL BOND The social bond is developed and strengthened through component parts, the first component Hirschi identified as attachment. Attachment describes the quality of the relationship the individual has with others in society. He reasoned that the strength of attachment, or more precisely, the affectionate ties that the individual felt toward parents, peers, and school was inversely related to delinquency. That is, the more closely, more intimately, more sharing, better communicating, more often the child interacted with his parents, the less probable that the child would become delinquent. Hirschi’s study focused on attachment of children to their parents, peers, and school relationships.

344

MELISSA A. MENASCO

As described by Hirschi, developing a strong social bond to society suggests that the individual has internalized the social and cultural norms of society. Those who have a strong bond with society are less inclined to deviate from the norms of that society because they have internalized and accepted these norms. Those who have a weakened bond with society are more inclined to deviate from societal norms because they have not internalized the social and cultural norms. Johnson (1987) focused on gender differences between attachment to mothers and attachment to fathers. His results challenged Hirschi’s conclusion that children’s attitudes toward one parent were reflective of their attitudes toward the other parent. In a sample of over 700 adolescent boys and girls, he found that both males and females expressed stronger attachment to mothers than fathers. Utilizing three measures of attachment: affective and behavioral attachment, feeling loved by their parent, and level of anger toward their parent; he found that affective and behavioral attachment as defined by Hirschi was the weakest measure of attachment. Supporting the literature that boys are more delinquent than girls (Chesney-Lind & Shelden, 2003; Hirschi, 1969; Jensem & Eve, 1976; Smith & Visher, 1980; U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention, 2004), Johnson also found that anger toward father was the most predictive measure of delinquency. Referring to anger toward father as a measure of detachment, he found a positive relationship between increased detachment toward father and delinquency for both males and females. Male delinquency was associated with all three of the attachment measures while female delinquency was only associated with anger toward father Wiatrowski et al. (1981) analyzed data from the Youth in Transition Study (Bachman, 1975) in order to determine the strength of each of the elements of the social bond. Hirschi’s work implied that each element of the social bond was equally responsible for the ties that children feel toward society; Wiatrowski et al. questioned whether this was true. They found that parental and school attachment had the strongest negative effects on delinquency. They further found substantiation that commitment, involvement, and belief were also correlated with delinquency, although their effects were not as strong as the attachment component Jensen and Brownfield (1983) found that children who expressed attachment to their parents and who thought that their parents had taken illicit drugs were not inhibited from drug use. Children, who expressed attachment to their non-drug using parents as Hirschi’s theory predicted, were found to have lower drug use. While Jensen and Brownfield found support

Delinquency, Gender, and Social Control

345

of Hirschi’s theory of the element of attachment with regard to non-drug using parents, they found that attachment to drug using parents to have little impact on the drug use of their children. They conclude by suggesting that adolescent perceptions of their parents’ behavior do impact the quality of attachment. Commitment of the child to long-term goals such as attainment of a college degree, getting married and obtaining a well-paid position were also key to the measurement of this element of the social bond. Hirschi argued that in order for the system of social control to function properly, a system of rewards for maintaining adherence to societal norms was necessary. Individuals who commit deviant acts, he reasoned, did not worry about consequences of their behavior because they had lost the motivation to attain these goals. Those individuals who aspired to societal goals would exercise restraint from deviating. Hirschi specified three societal goals for the adolescent: education, occupation and passage to adult status. Realizing that some adolescents may complete their education, that is, finish high school, but remain unable to begin their occupation, Hirschi noted, that this places them in the position of reaching adult status without access to all of the rewards of adult status, namely a good job. Hirschi reasoned that for these adolescents who were partially admitted to adult status by virtue of having completed their formal education, but who did not have the responsibilities of family and work, would view this period as a happy time. This freedom from the restraints posed by school would be conducive to delinquency. Jenkins (1995) studied delinquency and school commitment in a sample of 754 seventh and eighth graders. Defining delinquency in terms of school crime, school misconduct and school nonattendance, she confirmed Hirschi’s observations of an inverse relationship between school commitment and delinquency. Involvement was Hirschi’s term for describing the extracurricular activity level of the adolescent. Examples of involvement would be after school activities, including sports, church, and volunteer work. Children who are very busy, Hirschi reasoned, would not have time to commit delinquent acts. In terms of delinquent acts committed, he found that boys who were more frequently bored were more likely to have engaged in delinquency. Additionally, when he compared feelings of boredom with time spent on homework per day, he found that these two variables were inversely related. The more time spent on homework or the more involved the boys were in school activities, the less likely they were to engage in delinquent acts. Huebner and Betts (2002) found confirmation of two elements of the social bond: attachment and involvement. They found that attachment of

346

MELISSA A. MENASCO

girls was stronger than boys in predicting delinquency and substantiated that involvement positively impacted the academic achievement of boys and girls as well as inversely impacting delinquency. Boys and girls who were involved in structured activities were more likely to have better grades than those who were not. According to control theory, delinquency is not caused by a belief in a delinquent subculture, but rather a lack of belief in the norms of society. When he examined boys who had no police record and who reported that they had never been picked up by the police, he found that with respect to self-reported acts of delinquency, boys who had respect for police committed fewer delinquent acts than boys who indicated little respect for police, which led him to conclude that boys who had no respect for authority figures also had no respect for the law. In addition, there have been numerous studies to replicate his work (Foshee & Bauman, 1992; Friedman & Rosenbaum, 1988; Hindelang, 1973; Matsueda, 1982; Seydlitz, 1991; Wiatrowski et al., 1981). Most of the subsequent research has confirmed his findings although some results were not as strong as the theory would predict (Agnew, 1985; Krohn & Massey, 1980).

FEMALE DELINQUENCY Since males commit the majority of delinquent acts, delinquency research has traditionally been focused on males, the assumption being that delinquency theories applied to both males and females, yet researchers understood that the focus of concern was male offenders. In addition, the paucity of female delinquents was found to be an obstacle to the collection of meaningful information. Yet within the last forty years, with the overall rates of delinquency in decline, there has been an increase in the rate of female delinquency, in fact, serious violent delinquency. Males still commit a majority of the crimes as listed in the Federal Bureau of Investigation’s Uniform Crime Reports (2015), but it is noticeable that female delinquency has been increasing, most especially in terms of serious violent crime (2015). A number of issues arise with regard to these observations. First, there are more delinquent males than females. One of the debates that continues in various theoretical frameworks is whether delinquents are born or socialized. Second, males and females engage in different types of delinquent behavior. Girls, for example, run away from home at a rate far greater

Delinquency, Gender, and Social Control

347

than boys, while boys commit the majority of violent crimes. Finally, comparative male and female delinquency studies on social control theory will provide context for this research. Theories of delinquency have discussed societal norm violation in terms of male deviancy because of the evidence that juvenile crime is overwhelmingly male dominated. One of the questions that arises is whether the theoretical framework for any theory of deviancy is applicable to female delinquency. Since female delinquency has always been very low in comparison to male delinquency, it simply has not seemed to matter whether females were included in the theoretical framework. In his work, The Criminality of Women, Pollak (1950) claimed that female law violation occurs more often than reported, but the criminal justice system is more lenient towards female offenders. Proponents of the chivalry hypothesis cite as evidence that police officers are less likely to arrest females (Visher, 1983) and even when arrested, females are more likely to have charges dismissed (Bernstein, Cardascia, & Ross, 1979). Furthermore, when comparing similar crimes, female sentencing has been found to be less severe than for males (Reckless & Kay, 1967). Corley, Cernkovich, and Giordano (1989) did not find support for the chivalry hypothesis influencing the imposition of sanctions by the courts in their study of treatment of delinquent males and females. From a biosocial and psychological perspective, the reasons for female delinquency are attributed to physical and emotional traits. Theorists in this tradition believe that biology, psychological factors, and the social environment are all factors influencing delinquency. Some of the more prominent ideas promoted by these theorists are: precocious sexuality (Buchanan, Eccles, & Becker, 1992; Caspi, Lyman, Moffitt, & Silva, 1993; Glueck & Glueck, 1934), hormonal differences between males and females (Gove, 1985), premenstrual syndrome (Fishbein, 1971; Harry & Balcer, 1987; Horney, 1978), and aggression (Ellis, 1988; Frodi, Maccauley, & Thome, 1977; Maccoby & Jacklin, 1974). Since the primary adult figures in children’s lives are their parents, this perspective places a good deal of emphasis on children’s home life. Children from troubled families, including broken homes or those who lack supervision, are at greater risk for delinquency. As early as 1923, W. I. Thomas noted that female delinquency was linked with the desire for material wealth and excitement (Thomas, 1923). According to Thomas, girls from lower socioeconomic classes had not been schooled with appropriate middle-class values and in their desire for wealth and thrills, were forced to engage in sexual activity as a means

348

MELISSA A. MENASCO

of acquiring material possessions. In The Adolescent Girl in Conflict (1966), Gisela Konopka pointed out that love deprivation and isolation were the basis for female delinquency. According to Konopka, an important aspect to the maturation process of adolescent girls is the need for acceptance by males. If the adolescent girl is unable to develop these relationships with family and friends, she may turn to sexual relationships as a means of satisfying her need for male approval. Pursuing sexual relationships with males would then lead to disapproval of her behavior by both family and community. This disapproval, in turn, would serve to increase her feelings of isolation from family and friends. Konopka posited that girls who lived in fatherless homes were especially disadvantaged because they had no means to develop the basis of a healthy relationship with a male figure. Vedder and Somerville (1970) attributed female delinquency to family pressures. They found that 75% of institutionalized females had family problems. In their work, The Delinquent Girl, they note the impact of unfair social practices on girls as well as the effects of living in a male-dominated culture. Subsequent work of socialization theorists continues to link female delinquency with a troubled home life. In their study of incarcerated girls, Belknap, Holsinger, and Dunn (1997) report that delinquent girls spoke about physical and sexual abuse as well as degrading and embarrassing social situations. As well as citing an association with abuse within the home and female delinquency, Chesney-Lind (1987) notes that many of these girls end up running away from home only to become victims of the system that set out to protect them. In her book, Sisters in Crime (1975), Freda Adler stated that females commit fewer crimes than males because of their limited access to opportunities. By their nature, sex roles are restrictive for women, but as women enter the ranks of male-dominated professions, she predicted that the crime rate for females would begin to increase to match that of males. Simon asserted that as women gained more economic and social power, they would be more likely to engage in male-dominated crimes (Simon, 1975). Her prediction is supported by the increase in the female crime rate over the past 35 years. In fact, while male delinquency rates still far exceed those of females, the patterns of crimes between the sexes are more similar than different. James Messerschmidt has developed a theoretical model that attributes inequities between the sexes as a consequence of power in the patriarchal capitalist system (1993). Female delinquency is created by male exploitation of females, either through abuse, harassment, or undue influence.

Delinquency, Gender, and Social Control

349

FEMALE DELINQUENCY TYPOLOGIES While girls have traditionally engaged in delinquent acts less frequently than boys, there has been a steady increase in the delinquency rates of girls over the last thirty years. Boys still engage in criminal activity at a far greater rate than girls, but girls are closing the gap between the sexes. The 2010 Uniform Crime Reports (UCR), which is a compilation of crime statistics from over 12,000 law enforcement agencies as reported to the Federal Bureau of Investigation (FBI), confirms that, overall, boys were three times more likely to be arrested than girls (U.S. Federal Bureau of Investigation, 2015). The statistics reported by the UCR show that the arrests for girls under age 18 years in 2001 constituted 19.09% of all juvenile indexcrime arrests, and they now constitute approximately 14.60% (U.S. Federal Bureau of Investigation, 2015). Of particular note are the increased arrest rates of female over male delinquency for aggravated assault (12.02% vs. 10.69%), simple assault (21.19% vs. 14.36%), and weapons law violations (25.36% vs. 19.31%) over the period of 2001 through 2010. Regarding types of offenses, boys are far more likely to be arrested for violent crimes and serious property offenses. Boys continue to be more likely to commit violent crimes, including homicide, robbery, and aggravated assault, than are girls. For serious index property crimes, boys are eight times more likely to be arrested for weapons violations, seven times more likely to be arrested for burglary, and five times more likely to be arrested for motor vehicle theft than girls. Furthermore, boys are more than twice as likely to be arrested for drunkenness and curfew violations than girls. Boys are arrested over four times more often than girls for drug law violations. Types of offenses for which girls are typically arrested include running away from home and prostitution (Chesney-Lind & Shelden, 2003). Over half (60%) of the arrests for running away from home are girls. Girls are far more likely to be arrested for status offenses, which in addition to running away from home include truancy, curfew violation, incorrigibility, and loitering. Interestingly, the 2010 UCR reports that curfew/loitering account for 5.9% of all boys’ and girls’ arrests. Furthermore, the information gleaned from the UCR describes several key factors regarding male and female delinquency. First, male delinquents continue to overwhelmingly commit the greatest number of juvenile offenses, with arrests exceeding female delinquents by more than 2:1 (U.S. Federal Bureau of Investigation, 2010). Males commit most of the violent crimes and females commit the most status offenses. Second, there has

350

MELISSA A. MENASCO

been a steady increase in the rates of female delinquency over the past thirty years. Third, when ranking the most common delinquent behaviors, male and female delinquents exhibit similar patterns of criminal behavior.

MALE/FEMALE SOCIAL CONTROL STUDIES Smith and Visher (1980) conducted an empirical review of studies comparing sex and involvement in crime from 1940 through 1975. In their review they gathered information such as self-report questionnaires, official reports of arrest, class, age, race, offense type, level of family intactness, and level of urbanization. Their review of forty-four studies found the gap narrowing between male and female deviancy in general and delinquency in particular, depending upon type of offense. In terms of minor offenses, males and females exhibited similar propensity levels while the more serious criminal behaviors continued to be dominated by males. Of particular interest was their finding that African American male and female involvement in criminal activity appears to be converging while European Americans do not exhibit a similar convergence. In other words, African American women have advanced faster than European American women, thus lessening the gap between the sexes and creating a similar propensity toward crime as predicted by liberal feminist theory. Although this study was conducted 35 years ago, the trends of female and male delinquency have continued to the present. Boys continue to participate in the most serious offenses, but girls, by their increasing rates of participation in delinquent acts, have narrowed the gap that existed between the less serious forms of delinquent behavior. Comparing social control theory with that of masculinity theory, Shover, Norland, James, and Thornton (1979) studied 1,002 students from 8th through 12th grades. Masculinity theory asserts that criminal behavior is associated with traditional masculine behavior. Aggressiveness, physical strength, and proof of legal or illegal achievements are associated with masculine traits and criminal behavior. This perspective argues that there exists a direct link between traditional gender roles and criminal behavior. In this model, feminine behavior is non-aggressive, de-emphasizes physical strength, and is lacking in the achievements to which males aspire, thus producing less criminal activity. In contrast to the masculinity theory model, the authors posit that with social control theory, an indirect link exists between gender roles and

Delinquency, Gender, and Social Control

351

delinquency. Parents and significant authority figures more closely control females, they reason, thus giving females fewer opportunities to engage in delinquent behavior. Furthermore, females are more likely to form closer and longer lasting relationships than males (Bardwick & Douvan, 1972) as well as develop a more receptive belief in moral standards. Jensem and Eve (1976), utilizing a self-report instrument, studied over 4,000 seventh through twelfth graders in and around Richmond, California as a test of social control theory. Their study compared African Americans with whites although it was initiated primarily to assess the differences between male and female delinquency. Consistent with previous research and official statistics, they found that girls were less delinquent than boys. Jensem and Eve tested the social control variable of attachment to parents from an instrumental and non-instrumental perspective. They found that both African American and white males were more delinquent than females regardless of level of attachment to their parents, that is, boys who indicated that they had high levels of maternal and paternal support admitted to more delinquent acts than girls who also indicated close relationships with their parents. However, they did find support for social control theory with this variable. Both boys and girls who had strong relationships with their mothers and fathers were less delinquent than boys and girls who indicated that their relationships were weak.

METHODS The data for this study is drawn from The National Education Longitudinal Study (NELS). Collected by the United States Department of Education National Center for Educational Statistics (NCES), NELS is a panel survey of students, parents, teachers, and school principals. Information was gathered through self-report from students covering all aspects of their life including such topics as family, school, extracurricular activities, self-perceptions, delinquent activity and future aspirations. Information concerning the student, the school curriculum, the school district, parental involvement, and a multitude of cultural factors are some of the areas about which NELS queried parents, teachers, and school principals. A stratified, probability sample, NELS consists of five waves beginning in 1988. Student respondents were eighth graders in 1988, tenth graders in 1990, twelfth graders in 1992, twenty years of age in 1994, and twenty-six years of age in the final wave in 2000. Only students who

352

MELISSA A. MENASCO

participated in the five waves of the NELS collection process were included in this study. The resulting sample totaled 8,363 individuals. By sex, the sample consisted of 4,589 females and 3,774 males.

DEFINITION OF VARIABLES Social Bond Variables The four elements of the social bond were each determined by an indexed and scaled variable composed of appropriately related questions for each element. The attachment component of the social bond was defined as an indexed variable comprised of four questions. In order to assess the quality of the parent child relationship, and hence determine strength of attachment the following questions were asked of parents: “Looking over the last year, how frequently did you and your teenager participate in the following activities together? Attending school activities (sports, plays). Working on homework or school projects. Spending time just talking together. Doing something else fun together.” Possible responses for each question ranged from: 1 = never, 2 = rarely, 3 = sometimes, 4 = frequently. Higher values imply stronger levels of attachment between parent and child, and lower values imply weaker levels of attachment. Commitment was defined as an indexed and scaled variable. Children were asked the following questions, “Think about how you see your future. What are the chances you will have graduated from high school? What are the chances that you will go to college? What are the chances that you will have a job that pays well?” Possible responses were: 1 = very low, 2 = low, 3 = fifty-fifty, 4 = high, 5 = very high. Higher values indicate optimistic expectations of future accomplishments and lower values indicate a pessimistic outlook on future accomplishments. Involvement was measured by asking children the question, “In a typical week, how much total time do you spend on all SCHOOL-SPONSORED extracurricular activities?” Possible responses were: 0 = none, 1 = less than 1 hour/week, 2 = 1 4 hours/week, 3 = 5 9 hours/week, 4 = 10 19 hours/ week, 5 = 20 or more hours/week. Belief was defined as an indexed, scaled variable. Children were asked the following questions, “How often do you feel it is ‘ok’ for you to: Be late for school? Cut a couple of classes? Skip school for a whole day? Cheat on tests? Copy someone else’s homework? Get into physical fights? Belong

Delinquency, Gender, and Social Control

353

to gangs? Make racist remarks? Make sexist remarks? Steal belongings from school, a student, or a teacher? Destroy or damage school property? Smoke on school grounds? Drink alcohol during the school day? Use illegal drugs during the school day? Bring weapons to school? Abuse teachers physically? Talk back to teachers? Disobey school rules?” Scores ranged from 1 to 4, with low scores indicating “often” or agreement with delinquent attitudes and high scores indicating “never” or disagreement with delinquent attitudes. Delinquency Variables Five delinquency variables were selected from participants’ 10th grade responses to include a range of delinquent behaviors. Designated variables were: school problems, substance abuse, running away from home, physical fights, and arrests. School problems were defined as a continuous variable. Children were asked the following questions, “How many times did the following things happen to you in the first half of the current school year? I was late for school. I cut or skipped classes. I got in trouble for not following school rules. I was put on an in-school suspension. I was suspended or put on probation from school. I was transferred to another school for disciplinary reasons.” Responses were summed and the mean score was calculated, with scores ranging from 1 to 4. Low scores of “0” indicated no school problems and high scores of “4” indicated multiple school problems. Substance abuse was defined as a continuous variable. Questions were based on drug and alcohol use. Children were asked the following questions concerning alcohol consumption, “Next we want to ask you about drinking alcoholic beverages including beer, wine, wine coolers and liquor. On how many occasions (if any) have you had alcoholic beverages to drink? During the last 30 days.” To measure drug use two questions were asked, “On how many occasions (if any) have you used marijuana (grass, pot) or hashish (hash, hash oil)? During the last 30 days. On how many occasions (if any) have you used cocaine in any form (including ‘crack’)? During the last 30 days.” Low scores of “0” indicated no substance abuse and high scores of “9” indicated multiple instances of drug and alcohol use. Running away from home is a categorical variable derived from the question, “Did you run away from home for a week or more at any time during the last two years?” Responses were recoded to: 0 = no, 1 = yes. Involvement in physical fights was derived from the question, “In the first half of the current school year, how many times did any of the following things happen to you at school? I got into a physical

354

MELISSA A. MENASCO

fight at school.” Responses were recoded to the categorical variable: 0 = never, 1 = once, twice, more than twice. Determining whether students had been arrested came from the question, “How many times did the following things happen to you the first half of the current school year? I was arrested.” Responses were recoded to the categorical variable: 0 = never, 1 = one or more times.

Individual Characteristics Variables In addition to studying the social bond and its relationship with delinquency variables, the following predictors were included in order to determine their level of influence. Ethnicity was coded as a categorical variable. Based on student responses three variables were coded to determine ethnicity. Students were classified as Black, Hispanic, or Asian. All students had three variables coded by ethnicity. Parents were asked the following question, “What was your total gross family income from all sources before taxes in 1991? (If you are not sure of the amount, please estimate.)” Possible responses were: 1 = none, 2 = less than $1,000, 3 = $1,000 $2,999, 4 = $3,000 $4,999, 5 = $5,000 $7,499, 6 = $7,500 $9,999, 7 = $10,000 $14,999, 8 = $15,000 $19,999, 9 = $20,000 $24,999, 10 = $25,000 $34,999, 11 = $35,000 $49,999, 12 = $50,000 $74,999, 13 = $75,000 $99,999, 14 = $100,000 $199,999, 15 = $200,000 or more. Parent’s education was defined to be the highest level of education of either mother or father. Possible responses were: 1 = didn’t finish high school, 2 = high school grad or GED, 3 = greater than high school and less than 4 year degree, 4 = college graduate, 5 = M.A. or equivalent, 6 = PhD, MD, or other. Students were asked to provide information regarding their parents’ occupations. The question asked was, “Which of the categories below comes closest to describing your father’s (stepfather’s or male guardian’s) current job? If he is unemployed, retired, or disabled, select the answer that best describes his most recent job. Also, if your father works for more than one job, please answer for the job you consider to be his major activity.” From their responses a categorical variable was created to indicate whether their fathers were professionals (doctors, lawyers, advanced professional degree holders). Responses were coded as: 0 = non-professional, 1 = professional. A similar procedure was used to create a categorical variable indicating whether their mothers were professionals. Responses were coded as: 0 = non-professional, 1 = professional.

Delinquency, Gender, and Social Control

355

Students were asked for information regarding the employment status of both parents. The question they were asked was, “Is your father, stepfather or male guardian currently working, unemployed, retired, or disabled?” Responses were coded as: 0 = father unemployed, retired, or disabled, 1 = father employed. Mother’s employment status was also designated a categorical variable, in the same manner. Parents were asked for their marital status in the second NELS follow-up (1992). Responses were recoded into a dummy variable designating whether they were single parents: 0 = two-parent household, 1 = single parent household. Parents were asked to provide information regarding the number of siblings that the student had. Possible responses were: 0 = none, 1 = one, 2 = two, 3 = three, 4 = four, 5 = five, 6 = six or more.

HYPOTHESES Based on the results of prior research (Eve, 1978; Cernkovich & Giordano, 1992; Hindelang, 1973; Huebner & Betts, 2002; Jenkins, 1995; Johnson, 1987; Krohn & Massey, 1980; Rankin & Kern, 1994; Wiatrowski et al., 1981) it is expected that a relationship between the social bond and delinquency will exist. There are three objectives for this study. The first objective is to determine whether the social bond has a differential impact on males and females. The second objective is to test the efficacy of the social bond by varying the seriousness of delinquent activity. It is known through official crime statistics that certain offenses are more likely to be committed by one sex over the other. Social control theory would predict that the social bond for females is weaker than the social bond for males when comparing the act of running away from home. Similarly, a weak social bond for males should be associated with getting arrested or getting into fights. The final objective is to test each of the elements of the social bond and its effect on the sexes. Hypothesis 1. All elements of the social bond, attachment, commitment, involvement, and belief, will contribute to the social bond for females and only attachment and belief will be significant for males. Hypothesis 2. Attachment and belief will be stronger for females than males. Hypothesis 3. Females will demonstrate stronger social bonds than males.

356

MELISSA A. MENASCO

Table 1 describes delinquency characteristics for males and females. Consistent with the literature, males exhibit higher levels of delinquent behavior with the exception of running away. While the difference between female and male runaway responses is slight (1%), it is nonetheless important to note that females report higher incidences of running away than males, which is also consistent with the literature. In terms of school problems, 19% of the females and 15% of the males report no incidences of misbehavior. Thus, the majority of students in this sample (81% of the females and 75% of the males) have had at least one reported school problem. Both sexes report an average of two or three incidences of school misbehavior. Slightly under two-thirds of the females (61%) report no drug or alcohol involvement in the thirty days prior to taking the survey while 57% of the males report no involvement with drugs or alcohol. Of those who did admit to taking drugs or alcohol, both males and females averaged less than one occasion in the thirty days prior to survey administration with males reporting occasions slightly more often. There is a substantial difference between male and female reporting of physical fights. Males are far more likely to have engaged in fights than females (22% vs. 7%). This finding is substantiated in official crime statistics when comparing male and female participation in violent crime. When comparing arrest rates, one in three arrests is a female. The overwhelming majority of students have never been arrested (99% of the females and 96% of the males). Not surprisingly, this table substantiates information reflected in official crime statistics. Males report more delinquent activity than females. Furthermore, the only delinquent offense that was higher for females than for males was running away from home. Although males commit more delinquent acts than females, the reporting demonstrates similarities between the sexes. When comparing school problems and substance abuse, both males and females report an average of less than one offense during the reporting period. The next section is a comparison of the relationship of each element of the social bond for tenth grade boys and girls. Respondents were asked a series of questions relating to the parent child relationship and an indexed variable was created which was the measurement of attachment. The means for females and males are 3.12 and 3.09 respectively, indicating that both sexes report parent child activities as occurring between “sometimes” and “frequently” with girls reporting slightly higher frequencies of joint parent child activities. Both sexes have high expectations for their future as demonstrated in the measure of commitment. An indexed variable was created which included three questions relating to future expectations. Here

357

Delinquency, Gender, and Social Control

Table 1.

Mean Levels of Delinquency, Social Bond, and Individual Characteristics, by Sex.

Delinquency School problems Substance abuse Runaway from home Physical fights Ever arrested Social bond Attachment Commitment Involvement Belief Individual School performance Black Hispanic Asian Household income Parent’s education Mother professional Father professional Mother employed Father employed Single parent # Siblings N

Females

N

Males

N

.42 .61 5% (.23) 7% (.26) 1% (.11)

3721 1792 247

.49** .74** 4%** (.19) 22%** (.42) 4%** (.21)

3209 1625 142

331 60

3.12 (.54) 4.51 (.53) 1.54 (1.42) 3.68 (.29)

3.09** (.53) 4.46** (.57) 1.63** (1.53) 3.54** (.4)

54.87 (21.90) .09 (.28) .11 (.32) .07 (.26) 10.25 (2.35) 3.13 (1.23) .07 (.25) .12 (.33) .88 (.32) .87 (.33) .30 (.46) 2.27 (1.47) 4,589

58.92** (22.59) .07** (.25) .11 (.31) .07 (.26) 10.48** (2.26) 3.23** (1.20) .07 (.25) .13 (.33) .87 (.33) .88 (.33) .27** (.45) 2.12** (1.38) 3,774

Note: N = 4,589 females, 3,774 males. **p < .01.

836 169

358

MELISSA A. MENASCO

again, boys’ and girls’ responses are very similar. The mean boys’ response is 4.46 and the mean girls’ response is 4.51, indicating that the likelihood of completing future goals is between “high” and “very high” with girls indicating slightly higher optimism. Involvement in school activities was measured with a single question asking how much time on average students spent on extracurricular activities. Boys reported average participation in extracurricular activities as 1.63 and girls reported average participation as 1.54. This student sample participates in school extracurricular activities from less than one hour per week and up to four hours per week with boys reporting slightly higher school involvement. The belief measure was an indexed variable made up of questions relating to delinquent attitudes. These questions asked whether students agreed with statements relating to tardiness, cutting classes, using illegal weapons, smoking, belonging to gangs and various other delinquent activities. Girls (3.68) and boys (3.54) agree that it is “rarely” or “never” appropriate to destroy school property, use drugs and alcohol, fight, smoke, cheat, and generally engage in socially disapproved behaviors. These responses are indicative of a strong belief in social norms. The difference in the means between boys and girls is statistically significant for all four of the social bond measures with girls indicating stronger attachment, commitment and belief than boys, and boys reporting higher levels of involvement than girls. The last section of this table is a comparison of the means of individual characteristics between males and females. School performance for each student was calculated by averaging the tenth grade standardized mathematics, reading, science, and history test scores. The average score for males was higher than females. The average annual income of the families in this study was between $25,000 and $34,999. Boys’ families had a slightly higher average annual income than girls’ families. Parents of children in this study, on average, had some college education. Mothers who are professionals, that is, hold professional degrees such as MD, JD, or PhD, account for 7% of girls’ families and 7% of boys’ families while fathers who are professionals account for 12% of girls’ families and 13% of boys’ families. The majority of parents in this sample are employed. Eighty-eight percent of the girls’ mothers and eighty-seven percent of the boys’ mothers are working outside the home. Eighty-seven percent of the girls’ fathers and eighty-eight percent of the boys’ fathers are employed. Thirty percent of the girls come from a single parent home while twenty-seven percent of the boys come from a single parent home. Girls are slightly more likely to have more than two siblings than boys.

Delinquency, Gender, and Social Control

359

These characteristics provide an informative profile of the participants in this study. The typical 10th grade girl in this sample is white and middle class. Her parents have had some college education and it is most likely that both parents are employed outside the home. She probably lives with both parents and has two siblings. Her academic performance as measured by standardized test scores is at the 55th percentile. The typical 10th grade boy has similar characteristics. He is white and middle class although his family may have a slightly higher income than the girls’ families. His parents have some college education and may, in fact, have more college education than the girls’ parents. As with the girls, both his parents are more likely to work outside the home. He most likely lives with both parents and has two siblings. His academic performance as measured by standardized test scores is at the 59th percentile which is slightly higher than the girls’ scores. Table 2 examines the correlation of the elements of the social bond with each of the delinquency variables. The initial observation is the inverse relationship between each of the elements of the social bond with all of the delinquency variables, that is, for both males and females the stronger the attachment, involvement, commitment and belief, the less likely they are to have school problems, engage in substance abuse and fighting, run away from home, and been arrested. Belief appears to be the element of the social bond which has the strongest negative correlation with delinquent acts. Children who agree with social norms are less likely to engage in delinquent acts. Conversely, children who do not conform to societal standards are more likely to engage in delinquent acts. Furthermore, all measures are significant with two exceptions: boys fighting and boys running away from home.

School Problems All of the elements of the social bond for both sexes have an inverse relationship with school problems, that is, the stronger the social bond, the less likely for either boys or girls to have school problems. Belief exerts the strongest influence on girls with 28% of the variance of school problems accounted for by belief. The correlation between belief and school problems is significant, r(4587) = −.53, p < .001. The effect of commitment, involvement, and attachment on school problems is small, but significant. Attachment has the weakest correlation with school problems, r(4587) = −.1, p < .001. Similar to the girls, belief for the boys exerts the strongest

360

Table 2.

Correlations between Elements of the Social Bond and Delinquency, by Sex.

Social Bond

Attachment Commitment Involvement Belief

Females

Males

School problems

Substance abuse

Runaway

Fights

Arrested

School problems

Substance abuse

Runaway

Fights

Arrested

−.098** −.144** −.158** −.533**

−.066** −.063** −.109** −.438**

−.070** −.076** −.102** −.186**

−.047** −.113** −.103** −.185**

−.050** −.053** −.058** −.187**

−.072** −.163** −.157** −.546**

−.051** −.080** −.091** −.456**

−.022 −.074** −.052** −.156**

−.004 −.089** −.067** −.250**

−.066** −.075** −.099** −.269**

Note: N = 4,589 females, 3,774 males. **p < .01.

MELISSA A. MENASCO

Delinquency, Gender, and Social Control

361

influence on school problems. Belief accounts for 30% of the variance in school problems. The correlation between belief and school problems is significant, r(3772) = −.55, p < .001.

Substance Abuse As with school problems, there is an inverse relationship between each of the elements of the social bond and substance abuse. The correlation between belief and substance abuse is not as strong as the correlation between belief and school problems for either sex. Belief accounts for 19% of the variance in substance abuse for the girls and 21% of the variance in substance abuse for the boys. Commitment for the girls has the weakest correlation with substance abuse, r(4587) = −.063, p < .001. Attachment is the weakest correlation for the boys, r(3772) = −.051, p < .01.

Running Away Attachment, commitment, involvement, and belief for both sexes have a weak, inverse relationship with running away from home. The strongest correlation for boys is belief, r(3772) = −.156, p < .001. Belief is also the strongest correlation for girls, r(4587) = −.186, p < .001. Attachment, commitment, and involvement are significant for girls with the correlation of attachment being the weakest, r(4587) = −.070, p < .001.

Fights The correlations between commitment, involvement, and belief for boys and girls are weak, but significant. Belief is the strongest correlation for girls, r(4587) = −.185, p < .001, as well as for the boys, r(3772) = −.250, p < .001. The weakest correlation for girls was attachment, r(4587) = −.047, p < .001.

Arrests All elements of the social bond were shown to have an inverse and weakly correlated relationship with being arrested. Out of the four elements of the

362

MELISSA A. MENASCO

social bond, belief is the strongest correlation for both boys and girls. Attachment is the weakest relationship for being arrested for both of the sexes. When comparing males and females by type of delinquency, it is clear that belief has the strongest effect on inhibiting delinquent behavior. While a strong belief in social norms is associated with less delinquent behaviors, it appears that more seriously delinquent behaviors such as fighting or being arrested are not as strongly influenced by beliefs as the less serious delinquent behaviors of school problems and substance abuse. The correlation between belief and the less serious delinquent behaviors (school problems and substance abuse) indicate that boys who conform to social norms are less likely to engage in delinquent acts than girls who conform to social norms. Furthermore, males and females were remarkably similar in the correlations between the social bond and delinquency. With few exceptions, when females displayed a strong correlation between an element of the social bond and a delinquent act, males similarly displayed a strong correlation between the same element of the social bond and the same delinquent act. This pattern continues when comparing the weakest correlations. It would seem that attachment, commitment, involvement and belief have a similar effect on boys and girls. In terms of comparing the elements of the social bond and their impact on delinquency, attachment, commitment, involvement, and belief demonstrated an inverse relationship with delinquency. The strongest inverse relationship for both males and females was that of belief. The weakest correlation was found to be attachment for all delinquency measures. Table 3 describes the correlation between the individual characteristics of children and delinquency. For all individual characteristics, the correlations with each of the delinquency measures are weak. The strongest correlation is that of girls with single parents and school problems (.111), which is still a weak correlation. Boys and girls with school problems had similar and weak correlations for each of the individual characteristics. Of the three racial groups: black, Hispanic, and Asians, Hispanic children had the strongest correlation (.077 for girls and .061 for boys) with school problems and Asian children reported an inverse relationship with having school problems. The strongest correlation with school problems was found to be children of single parents followed closely by number of siblings in the family. Substance abuse among girls was found to be weak and inversely correlated with each of the racial groups. Boys’ substance abuse was negatively

Correlations of Individual Characteristics by Delinquency, by Sex.

Individual

Black Hispanic Asian Household income Parent’s education Mother professional -.003 Father professional Mother employed Father employed Single parent # Siblings

Females

Males

School Substance Runaway problems abuse

Fights

Arrested

−.101** −.016 −.097** .021 −.008 −.019 −.005 −.013 −.007 .044** .004

.065** .029* −.045** −.119** −.133** −.008 −.065** −.042** .051** .047** .048**

−.029* .008 .005 −.009 −.035* −.013 −.031* .024 .003 0 .028

.018 .077** −.041* −.073** −.075** .007 −.040** −.027 −.052** .111** .088**

−.020 .007 −.010 −.045** −.042** −.035* −.042** −.014 −.037* .054** .030*

School Substance Runaway problems abuse

Fights

Arrested

−.055** .007 −.065** .020 −.032* −.004 −.017 −.015 .006 .009 .006

.015 .004 −.042** −.020 −.059** −.003 −.020 −.025 −.021 .013 .023

.023 .009 −.022 −.014 −.033*

.040* .061** −.035* −.052** −.071** −.013 −.018 −.031 −.056** .083** .082**

−.015 .014 −.019 −.043** −.053** −.013 −.034* −.012 −.046** .025 .019

Delinquency, Gender, and Social Control

Table 3.

−.006 −.009 −.049** .025 .018

Note: N = 4,589 females, 3,774 males. **p < .01; *p < .05.

363

364

MELISSA A. MENASCO

correlated with black and Asian boys, but very weakly correlated to Hispanic boys. Running away from home was weakly correlated for both boys and girls with all of the individual characteristic measures. Household income was inverse and weakly correlated with running away from home for girls and living with a single parent was positive and also weakly correlated with running away from home. Boys were found to have weak and negatively correlations with parent’s education (−.053) and father’s employment (−.046). All of the individual characteristics for girls were found to be significant and weakly correlated with fighting. Parent’s education was found to have the strongest correlation for boys and girls, that is, the higher the education level of the parents the less likely the involvement with fighting. Girls’ arrests were very weakly correlated with all individual characteristics with one exception. Boys’ arrests were also weakly correlated with all of the individual characteristics. The strongest correlation was found to be an inverse relationship between father’s employment and boys’ arrests (−.049). Overall, the correlation between each of the individual characteristics and delinquency was weak for both sexes. Girls from families of single parents have a higher correlation with school problems and running away from home than boys from single parent families with school problems and running away from home. Table 4 presents the results of testing the probability that the social bond is predictive of delinquency. Model 1 results were derived by a logistic regression of the social bond with each of the three delinquency measures. Model 2 results were derived by a logistic regression of the social bond as well as the individual characteristics of school performance, ethnicity (Black, Hispanic, Asian), household income, parent’s education, mother and father professional, mother and father employment, single parent family, and number of siblings. For Model 1, all elements of the social bond are significant for girls who have run away from home. As social control theory predicts, there is an inverse relationship with attachment (−.315), commitment (−.259), involvement (−.261), and belief (−1.639) with running away. This model suggests that girls are less likely to run away from home as their levels of attachment, commitment, involvement, and belief increase. Comparing Model 1 with Model 2, attachment increases (−.345), commitment loses its significance, while involvement (−.213) and belief (−1.599) decrease. The Model 1 results suggest that each unit increase of attachment decreases the odds of running away approximately 27% while the Model 2 results suggest that

Table 4.

Logistic Regression Coefficients for Social Bond and Individual Characteristics by Delinquency, by Sex.

Social Bond

Females Model 1 Runaway

Attachment Commitment Involvement Belief

Fights

−.315** −.084 −.259* −.462** −.261** −.217** −1.639** −1.511**

Individual School performance Black Hispanic Asian Household income Parent’s education Mother professional Father professional Mother employed Father employed Single parent # Siblings -2 log likelihood 1762.776 Nagelkerke R square .101

538.333 .168

Note: N = 4,589 females, 3,774 males. **p < .01; *p < .05.

Males Model 2

Arrested −.461* −.284 −.238* −2.415**

2201.742 .093

Runaway −.345** −.129 −.213** −1.599**

Fights

Model 1 Arrested

−.045 −.420 −.183 −.137 −.108** −.200 −1.644** −2.460**

−.013** −.015** −.006 −.637* .361 −1.527 −.202 −.093 .052 −.048 −.456 .493 −.011 −.066* .033 .096 −.242** −.096 −.103 −.099 .166 −.325 −.150 −.918 .081 −.098 1.206 −.212 −.061 −.009 .338* −.041 −.121 .043 .022 .141 1736.09 2096.764 521.707 .117 .147 .195

Runaway .010 −.372** −.100 −1.233**

1130.315 .076

Model 2

Fights

Arrested

.141 −.231** −.054 −1.296**

−.281 −.201 −.246** −1.814**

3759.385 .091

1169.890 .177

Runaway −.023 −.171 −.051 −1.225**

Fights

Arrested

.102 −.081 −.032 −1.293**

−.385* −.076 −.221** −1.841**

−.015** −.011** −.011** −.713 .017 .526 −.087 −.085 .118 −.291 −.300 −.397 −.037 .021 .048 −.068 −.066 −.039 .333 .052 .257 −.152 .151 .365 .048 −.152 .054 −.448 −.049 −.606* .071 −.006 .112 .012 .010 .017 1102.739 3717.350 1145.677 .102 .107 .197

366

MELISSA A. MENASCO

each unit increase in attachment decreases the odds of running away approximately 29%, controlling for other variables in the model. Belief demonstrated the most striking results in both models. Each unit increase of belief decreased the odds of running away approximately 81% in Model 1 and approximately 80% in Model 2, controlling for other variables in the models. The individual characteristics which are significant for Model 2 are school performance (−.013), being Black (−.637), and living in a single parent home (.338). Model 2 suggests that each unit increase in academic performance decreases the odds of girls running away from home. This supports social control theory which proposes that students with higher academic performance in school will have stronger social bonds. This model also suggests that girls who are Black are less likely to run away from home while girls from single parent families are more likely to run away. Black girls, perhaps, realize how difficult living as a runaway is because their life circumstances are more difficult than middle class girls. One possible reason for the odds of girls running away from single parent homes being higher than those from two-parent households is the inability of a single parent to reinforce the direct and indirect controls as well as a two-parent home. This result is problematic for social control theory, since Hirschi argued that attachment to one parent, not necessarily both parents, is sufficient to increase the strength of the social bond. If attachment to one parent is sufficient, then single-parent households would be expected to have comparable levels of delinquency as two-parent households, which is clearly not the case. While it seems unlikely that all singleparent households could expect to have developed strong levels of attachment to their daughters, this result provides a counter argument to Hirschi’s assumption. Model 1 for boys who have run away from home has only two significant elements of the social bond: commitment (−.372) and belief (−1.233). Model 1 suggests that each unit increase in commitment decreases the odds of running away approximately 31%, controlling for other variables in the model. Furthermore, each unit increase in belief decreases the odds of running away from home approximately 71%, controlling for other variables in the model. The odds of running away from home are less likely for boys who expect that they will finish high school, attend college and find a good paying job. Similarly, the odds of running away from home are less likely among boys who agree with the social norms. When comparing Model 1 with Model 2, commitment loses its significance and belief continues to maintain its significance. The only individual characteristic which is significant for boys is that of school performance. Each unit increase in school

Delinquency, Gender, and Social Control

367

performance decreases the odds of running away from home approximately 71%, controlling for other variables. When comparing the models of boys and girls who have run away from home, only belief was found to be significant for both sexes and models. Attachment is significant in Model 1 and Model 2 for girls but it is not significant for the boys in either model. Commitment is significant in Model 1 for boys and girls. Involvement is significant for both of the girls’ models, and not significant for either of the boys’ models. Belief is the only element of the social bond to be significant for both male and female models. When comparing the full models of running away from home, school performance is significant for both sexes. Ethnicity, specifically being Black, was found to be significant for girls. Living in a single parent household was also found to be significant for girls who reported running away from home. Since official crime statistics consistently report higher numbers of girls running away from home than boys, the fact that the social bond seems to play a much larger role in girls’ lives than boys’ lives is indicative of its importance when discussing issues related to children running away from home.

Fights Commitment (−.462), involvement (−.217), and belief (−1.511) were all significant in Model 1 for girls who engaged in physical fights. Each unit increase in commitment decreases the odds of girls fighting approximately 37%, controlling for other variables. Each unit increase in belief decreases the odds of girls fighting approximately 78%, controlling for other variables. When comparing girls in Model 2 with Model 1, commitment is no longer significant, with involvement and belief still significant. The odds of girls’ fighting increase when comparing belief in Model 2 with belief in Model 1. In Model 2, each unit increase in belief decreases the odds of girls’ fighting approximately 81%, controlling for other variables. Individual characteristics of school performance (−.015), household income (−.066), and parent’s education (−.242) are all significant in Model 2. The odds of girls getting into fights decrease as their household income increases. Similarly, with increased levels of parent’s education, the odds of girls getting into fights decrease. Both of these results are consistent with observations that Hirschi made regarding the association of social class and educational attainment, and provide the basis for future research into status attainment measures.

368

MELISSA A. MENASCO

Model 1 for boys engaged in fights has only two significant predictors: commitment (−.231) and belief (−1.296). As each unit of commitment increases, the odds of boys fighting decrease approximately 21%, controlling for other variables. Belief (−1.293) is the only significant predictor of the social bond variables in Model 2. As each unit of belief increases, the odds of boys fighting decrease approximately 73%, controlling for other variables. Boys’ school performance (−.011) is the only individual characteristic found to be significant. Again, support for social control theory was in the expected direction. When comparing Model 1, commitment and belief are significant for boys and girls who fight. Furthermore, involvement is significant for girls, but not for boys in this model. When comparing Model 2, both sexes show an inverse relationship between belief and getting into fights. What this appears to indicate is that the odds of getting into fights decrease as acceptance and agreement with social norms increase. Involvement is significant for girls in Model 2, but not for boys. The only significant individual characteristic for boys is school performance. Girls, too, show a significant, inverse relationship between school performance and getting into fights. While boys are more frequently involved in fights than girls, this result suggests that as academic performance improves, the odds of getting into fights decrease. In addition, household income and parent’s education are significant for girls, but not significant for boys.

Arrests Commitment is the only social bond variable that is not significant for girls who reported getting arrested in Model 1. Attachment (−.461), involvement (−.238), and belief (−2.415) are all significant. When comparing Model 1 with Model 2 for girls, attachment and involvement are no longer significant while belief remains significant. For Model 1, each unit increase in belief decreases the odds of getting arrested approximately 91%, controlling for other variables, and for Model 2, the odds of getting arrested decrease approximately 92%, controlling for other variables. None of the individual characteristics are significant for girls in Model 1. For boys, involvement (−.246) and belief (−1.814) are the only two significant predictors of getting arrested in Model 1. Comparing Model 2 with Model 1, attachment (−.385) becomes significant, with involvement (−.221) and belief (−1.841) remaining significant. School performance (−.011) and father’s employment (−.606) are the only significant individual

Delinquency, Gender, and Social Control

369

characteristics in Model 2. The odds of boys whose fathers are employed of getting arrested decrease approximately 46%, controlling for other variables. When comparing boys’ and girls’ results for Model 1, involvement and belief are significant for both sexes while attachment is significant for only girls. Belief for girls in Model 2, is the only significant social bond variable while boys indicate that attachment, involvement, and belief are significant. For individual characteristics, school performance for boys in Model 2 shows a significant inverse relationship with getting arrested, while for girls, school performance is not significant. Father’s employment is also significant for boys, but not so for girls. Individual characteristics are not significant for girls. Table 5 presents an analysis of the relationship of the social bond with the delinquency variables of school problems and substance abuse. Ordinary least squares regression (OLS) was utilized with the social bond variables of attachment, commitment, involvement, and belief defined as predictors of each of the dependent variables of school problems and substance abuse. Beginning with girls, the next section of the analysis will explain school problems in terms of the social bond (Model One) and the social bond with the individual characteristics as previously noted (Model Two). Attachment (−.020), commitment (−.051), involvement (−.023), and belief (−.756) are all significant for girls in Model One. The inverse relationship of these elements of the social bond are in the direction that social control theory would predict, that is, girls who care about their parents’ opinions; girls who expect to graduate from high school and attend college; girls who are involved in extracurricular activities; and girls who agree with social norms are less likely to have school problems. Model One predicts that 29.7% of the variance in girls’ school problems is explained by the social bond. Furthermore, belief in social norms exerts the strongest influence with girls in terms of school problems. When individual characteristics are added to the social bond as predictors, the model improves in terms of explaining school problems. Model Two predicts that 32% of the variance in girls’ school problems is explained by commitment (−.024), involvement (−.015), belief (−.761), school performance (−.002), being Hispanic (.080), single parent household (.068), and number of siblings (.015). The impact of the social bond variables on girls’ school problems changes in Model Two: attachment is no longer significant, commitment and involvement are not as strong as in Model One, and belief increases. Girls who believe that cheating, stealing, smoking and drinking on school grounds, and disobeying school rules are

370

Table 5.

OLS Regression Coefficients for Social Bond and Individual Characteristics by Delinquency, by Sex.

Social Bond

Female Model 1

Attachment Commitment Involvement Belief

Hispanic Asian

Model 2

Model 1

Model 2

School problems

Substance abuse

School problems

Substance abuse

School problems

Substance abuse

School problems

Substance abuse

−.020* (−.026) −.051** (−.063) −.023** (−.078) −.756** (−.513)

−.023 (−.013) .004 (.002) −.036** (−.054) −1.416** (−.430)

−.019 (−.024) −.024* (−.030) −.015** (−.050) −.761** (−.516)

−.063** (−.036) .024 (.013) −.035** (−.052) −1.368** (−.416)

−.004 (−.004) −.071** (−.084) −.024** (−.076) −.633** (−.527)

−.007 (−.003) −.038 (−.020) −.025* (−.034) −1.255** (−.450)

−.012 (−.013) −.028* (−.033) −.018** (−.057) −.633** (−.528)

−.031 (−.014) .020 (.010) −.018 (−.025) −1.233** (−.442)

−.002** (−.089) .022 (.015) .080** (.059) −.017 (−.010)

−.003** (−.066) −.367** (−.110) −.115** (−.038) −.372** (−.101)

−.003** (−.140) .066* (.035) .073** (.047) −.019 (−.010)

−.005** (−.107) −.227** (−.051) −.008 (−.002) −.249** (−.058)

MELISSA A. MENASCO

Individual School performance Black

Male

# Siblings F R-square N

484.316 .297 4589

277.639 .195 4589

.002 (.009) .006 (.017) .021 (.012) .022 (.017) .001 (.000) −.017 (−.013) .063** (.068) .015** (.054) 134.265 .320 4589

.012 (.029) .020 (.026) .023 (.006) .062 (.021) −.044 (−.015) −.019 (−.007) .092** (.044) .005 (.007) 79.458 .218 4589

431.731 .314 3774

250.823 .210 3774

.005 (.023) .004 (.009) .019 (.010) .044* (.031) −.019 (−.013) −.031 (−.021) .053** (.049) .016** (.046) 123.582 .345 3774

.017 (.035) .000 (.000) .013 (.003) .012 (.004) −.095 (−.028) .019 (.005) .020 (.008) −.009 (−.011) 67.950 .224 3774

Delinquency, Gender, and Social Control

Household income Parent’s education Mother professional Father professional Mother employed Father employed Single parent

**p < .01; *p < .05.

371

372

MELISSA A. MENASCO

not “ok” are less likely to have troubles. School performance is significant for girls with school problems. Girls who are good students are less likely to have school problems and, conversely, girls who are poor students are more likely to have school problems. This supports social control theory’s argument that academic performance and delinquency are associated with the social bond. Being Hispanic (.080) is significant for girls’ school problems meaning that girls who are Hispanic are more likely to have had school problems. Living in single parent households (.068) was another significant predictor of girls’ school problems. This result also implies that girls in two-parent households are less likely to have had school problems. The single parent household is a problem for social control theory because Hirschi proposed that attachment to one parent was sufficient to satisfy development of the social bond. This result would appear to contradict his findings. Number of siblings (.015) was also significant for girls’ school problems. This relationship was positive, meaning that the larger the family, the more likely that girls would have school problems. Conversely, fewer brothers and sisters were associated with a reduced likelihood of school problems. These results are not surprising, as it seems evident that parents in larger families have less time to spend time with each child than parents in smaller families. If parents are unable to spend time with their children, the ability to develop attachment between parent and child is reduced. When analyzing boys’ school problems in Model One, commitment (−.071), involvement (−.024), and belief (−.633) are significant and inversely related to school problems. This is in the expected direction proposed by social control theory. Boys who are involved in extracurricular activities are less likely to have school problems. Furthermore, belief exerts the strongest negative influence over school problems. Comparing Model Two with Model One for boys, commitment (−.028), involvement (−.018), and belief (−.633) are still significant although commitment and involvement are not as influential as predictors of school problems. Belief remains unchanged and exerts the strongest influence of any predictor in Model Two. School performance (−.003), being Black (.066), being Hispanic (.073), having a father who is a professional (.044), living in a single parent household (.053), and number of siblings (.016) are all significant individual characteristics for boys in Model Two. Boys’ school performance was inversely related with their school problems. Boys who are good students are less likely to have had school problems while boys who do poorly in school are more likely to have had school problems. This result in conjunction with the inverse relationship of

Delinquency, Gender, and Social Control

373

the social bond variables in Model Two supports social control theory, that is, the social bond for boys with poor academic achievement is weakened. In Model Two, boys who are Black or who are Hispanic were more likely to have had school problems. From a social control theory perspective, this result would suggest that these boys had weakened social bonds. The question that then arises is why Black and Hispanic boys have weak social bonds. A closer look at attachment, commitment, involvement, and belief for this group of boys would be useful in determining the quality of the social bond. It is also possible that gender roles within the Black and Hispanic cultures impact the social bond, and more specifically attachment and belief. Further research into this issue is warranted, however the current sample size (Black males = 258, Hispanic males = 398) precludes that investigation from the current analysis. Boys living in single parent households are more likely to have school problems than boys from two-parent families. This would appear to indicate that boys from single parent homes do not develop as strong a social bond as boys from two-parent homes. Hirschi demonstrated that attachment towards one parent was sufficient for developing a strong social bond. While it does not necessarily follow that developing attachment to one parent in a two-parent household is synonymous with developing attachment to the only parent in a single-parent household; it is also not unreasonable to expect that attachment in a single-parent household is any less likely than in a two-parent household. For this reason, this result is somewhat surprising. Another factor which may have a negative impact on the development of attachment in the single-parent household is parental availability. Out of economic necessity, single parents may not be readily available to spend time with their children. While they may want to, they may be unable to attend school-sponsored events, help with homework, or spend time discussing their children’s problems. The number of siblings within the family is positively associated with school problems. This result would suggest support for social control theory by directing attention to the parent child relationship. It seems obvious to point out that parents have less time to spend with each child as the number of children in the family increases. This may impair the quality of the social bond in much the same way as that of the single parent. Model Two for boys with school problems is more robust than Model One, 34.5% of the variance in school problems is explained by the social bond variables of commitment, involvement, and belief as well as school performance, being Black, being Hispanic, whether father is a professional, whether living in a single-parent household, and number of siblings.

374

MELISSA A. MENASCO

Comparing girls and boys results for Model One, attachment is significant for girls, but not for boys; commitment and belief are stronger for boys; and involvement is almost identical for boys and girls. When Model Two for girls is compared with Model Two for boys, commitment, involvement, and belief are significant for both sexes. Both boys and girls report a significant relationship between school performance, being Hispanic, living in a single parent household, and number of siblings with school problems. In addition, boys who are Black and sons of fathers who are professionals also report a significant relationship with school problems. The results of the two models are very similar in terms of significant predictors and their standardized betas which might lead to the conclusion that similar factors are at work regardless of sex. With the social bond variables as predictors of substance abuse for girls, involvement (−.036) and belief (−.756) were found to be the only significant variables for Model One. Their relationship to substance abuse was found to be in the expected direction. The more active girls were in extracurricular activities, the less likely that they were to have engaged in the use of alcohol or drugs. Belief was found to have the greatest influence in deterring girls from engaging in drug and alcohol-related activities. In Model Two, attachment (−.063) is significant where it was not significant in Model One; commitment remains non-significant, both involvement (−.035) and belief (−1.368) decrease very slightly from their Model One counterparts. Belief exerts the strongest influence in this model as in Model One. Model Two, in fact, provides more information than Model One, with 21.8% of the variance in substance abuse being explained by attachment, involvement, belief, school performance (−.003), being Black (−.367), being Hispanic (−.115), being Asian (−.372), and living in a single parent household (.092). School performance continues to be significant in terms of the delinquency variables, in this case, substance abuse. Girls with higher academic achievement are less likely to be involved in substance abuse according to Model Two. Girls who are Black, Hispanic, or Asian are less likely to have engaged in drug or alcohol-related activities. These results suggest a cultural component to the behavior of girls. While it is important to recognize that the representation of each ethnicity is small (Black females = 405, Hispanic females = 515, Asian females = 328) it seems unmistakable that culture has an impact on limiting the involvement of girls in substance abuse. The betas indicate that Black girls (−.110) are the least likely to engage in substance abuse, followed by Asian girls (−.101), and Hispanic girls (−.038). Parents are seen as the primary source of cultural beliefs and attitudes and

Delinquency, Gender, and Social Control

375

given that attachment to parents and belief in social norms are also significant factors in limiting engagement in substance abuse, it makes sense that this group of girls would be less likely to misbehave. While each culture may vary in terms of its practices, gender roles for girls with regards to delinquent behaviors are fairly consistent. The gendered socialization of girls, that is, the learning of gender roles through interaction and socialization with others (Giddens, Duneier, & Applebaum, 2005) appears to have a deterrent effect on substance abuse. Girls who live in single parent households are more likely to engage in drug and alcohol use. In terms of discussing the day-to-day activities within a single-parent household, it is clear that unless the parent works at home, there is less direct control of their daughters. It seems reasonable to assume that girls will then have increased opportunities to drink and do drugs. The implication of girls living in a single-parent household is that attachment to the single parent does not occur. Since attachment and belief, in particular, are significant, it appears that girls in single parent homes have a weak social bond. Involvement (−.025) and belief (−1.255) are the only significant predictors of substance abuse for boys in Model One. The inverse relationship with substance abuse is in the expected direction proposed by social control theory. Boys who are more involved in extracurricular activities and who agree with social norms of acceptable behaviors are less likely to engage in drinking and drug use. For Model One, 21% of the variance in substance abuse is explained by involvement and belief. When individual characteristics are added to the social bond variables, the resulting Model Two is slightly better, 22.4% of the variance in substance abuse is explained by belief (−1.233), school performance (−.005), being Black (−.227), and being Asian (−.249). Although belief is the only element of the social bond to be significant, its direction supports social control theory. School performance is significant and inversely related to substance abuse demonstrating that boys who do better in school are less likely to engage in alcohol and drug use. Black and Asian boys are less likely to engage in drug and alcohol use in Model Two. It should be noted that these ethnic groups are small (Black males = 258, Asian males = 281). The social bond is strong enough to deter Black and Asian boys from engaging in drinking and drug use. Social control theory would suggest that attachment and belief for this group of boys is strong, although it is difficult to predict the role of involvement and commitment. However, their ethnicity would further suggest that the social bond is working in conjunction with gender socialization in the Black and Asian cultures.

376

MELISSA A. MENASCO

When comparing the substance abuse models for boys and girls, it is clear that the social bond exerts greater influence over girls. Attachment, involvement, and belief are inversely related to drug and alcohol use for girls. Belief, school performance, being Black, and being Asian are significant for boys and girls with belief being the strongest element of the social bond for both. Hirschi reasoned that children who expected to graduate from high school, pursue a college education, and obtain a job that paid well, would have higher levels of commitment. Since school performance was significant, indicating that children with higher levels of academic achievement were less likely to engage in substance abuse, it also seems reasonable to conclude that children with higher academic achievement would also have higher expectations for themselves. Girls who live in single-parent households are more likely to engage in drug and alcohol use, while this predictor was not significant for boys. From a social control theory perspective, this indicates that girls in singleparent households are more likely to have a weak social bond. Considering the gender role assigned to girls versus the gender role assigned to boys, will perhaps provide a better understanding of this result. Girls are expected to be dependent, submissive, sympathetic, and, in general, nurturing. Boys, on the other hand, are expected to be aggressive, outspoken, independent, and dominant. Parents may be more protective of their daughters than their sons, exercising stronger instrumental control over their daughters than their sons (Kim, 2003). These gender roles are conveyed to children through socialization, with parents acting as the primary conveyors of gender socialization. Single-parent households may be less able to exercise the instrumental control that exists in a two-parent household, thus affording girls greater freedom. Increased freedom in conjunction with a weak social bond gives these girls the opportunity to take drugs and drink alcohol.

CONCLUSIONS Beginning with a comparison of male and female delinquency, this analysis found substantiation of generally higher levels of delinquency among males. The mean levels of school problems, substance abuse, fighting, and arrests were higher for males than females. Only the mean level of running away from home was higher for females than for males. This result is substantiated by official crime statistics which report that females are more

Delinquency, Gender, and Social Control

377

likely to run away from home than males. When comparing the mean levels of the components of the social bond, females indicated that their levels of attachment, commitment and belief were stronger than those of males. Only involvement was stronger for males than females. This result suggests that the social bond has a stronger impact on females than males. For both males and females, the social bond was shown to be negatively correlated with delinquent behavior. With increased strength of attachment, commitment, involvement, and belief, there was a decrease in school problems, substance abuse, running away from home, fighting, and getting arrested. The only non-significant relationship to be found was between attachment and males running away from home. Overall, females reported stronger social bonds than males. Belief was significant and inversely related to engaging in all delinquent acts for both males and females. School or academic performance was significant and inversely related to engaging in all delinquent acts for both males and females with the exception of females who reported being arrested. When comparing females who ran away from home with males who ran away from home, attachment, involvement, and belief were all significant for females while belief was the only element of the social bond to be significant for males. At this level of analysis, ethnicity appeared to have a negative impact on delinquent behaviors. Household income and level of parent’s education also indicated an inverse relationship with females who had engaged in fighting. Single parent households had a positive relationship with females who ran away from home, had school problems, and engaged in substance abuse. Number of siblings was positively related to both males and females who reported school problems. Males reported a positive relationship between their fathers being professionals and school problems. Males also reported that if their fathers were employed, they were an associated probability in getting arrested. The first objective was to determine whether the social bond had a differential impact on males and females. The results of this study suggest that the social bond differentially affects males and females. For most delinquent acts, not only were more elements of the social bond significant for females than for males, but also the strength of the relationship between delinquent acts and the social bond were stronger for females than males. This study was unable to find full support for Hypothesis 1. Belief was the only element of the social bond found to be significant for all levels of delinquency and for both sexes. Involvement was only significant for males who were arrested and had school problems while for females, involvement was significant for all delinquent acts with the exception of getting arrested.

378

MELISSA A. MENASCO

Attachment was significant for females who ran away from home and had engaged in substance abuse and for males who had been arrested. Commitment was only significant for both males and females who had school problems. It is important to point out that where attachment, commitment, involvement, and belief are significant, the direction of the reported relationship supports social control theory. For both males and females, all significant elements of the social bond demonstrated an inverse relationship with delinquency, that is, with increasing levels of attachment, commitment, involvement and belief, there was a decrease in delinquent acts. Belief was significant for all delinquency variables for both sexes. In all cases, females reported stronger beliefs than the males. Attachment was significant for females who ran away and who were involved in substance abuse while for males attachment was only significant for boys who were arrested. This study found limited support for Hypothesis 2. Belief was significant for all delinquency variables for both sexes. In all cases, females reported stronger beliefs than the males. Attachment was significant for females who ran away and who were involved in substance use while for males attachment was only significant for boys who were arrested. This study found strong support for Hypothesis 3. Females reported more significant elements of the social bond than did males. When males and females reported the same significant elements of the social bond for the same delinquent acts, females demonstrated stronger social bonds than males. The only exception to this finding was for the delinquent act defined as school problems where males reported stronger commitment and involvement than females. The claim of social control theory is its ability to explain delinquency. By operationalizing the four elements of the social bond, researchers retrospectively study as well as prospectively project delinquent activity. From the social control theory perspective, ethnicity, class, parents’ occupation, parents’ education, and sex are factors which are secondary to the development of the social bond. The argument underlying this position is foundational to the social control theorists understanding of human nature. Human behavior for the traditional social control theorist is based on choice. Individuals may be influenced by their ethnicity, for example, but social control theory argues that it is still the individual’s decision to conform to societal norms. The results of this study would appear to challenge this position. In particular, the demonstrated strength of the social bond of females compared with the social bond of males. Males reported weaker

Delinquency, Gender, and Social Control

379

social bonds than the females in terms of running away from home, a traditionally female-dominated delinquent act as well as a contradiction of social control theory. However, when comparing the social bond between the sexes for fighting, males demonstrated the expected weaker social bond. While not conclusive, these contradicting results appear to point toward the effects of gender roles and gender socialization. Females are traditionally kept closer to the family in terms of instrumental (parental knowledge of children’s whereabouts) and relational (children’s concern for parental approval when parents are not present) parental controls, whereas males are encouraged to be independent via weaker instrumental and relational parental controls. It seems difficult if not impossible to untangle the role of gender socialization from the attachment that develops between parent and child. However, gender socialization provides answers concerning the seeming inconsistency of these results. From the gender socialization perspective, girls are encouraged to be passive, to avoid conflict, and to please others while boys are encouraged to be risk-takers, to be challengers, and to be independent. It is easy to see that the social bond and specifically the attachment measure are developed in conjunction with the definitions of gender roles. It is not hard to imagine that as girls are taught to please their parents through gender socialization, attachment to parents increases. Similarly, as boys are encouraged to be independent, risk-takers, attachment to parents decreases. Returning to the contradicting results of running away from home and getting into fights, the difference in socialization that occurs between males and females readily explains the inconsistencies of the social bond. Since girls are socialized in a manner that encourages the development of prosocial behavior, the strength of the social bond for females should be higher than that of males, indeed, expected. In addition to partially explaining the variation in the social bond, it seems reasonable to suggest that gender socialization has a role in explaining differences in delinquent behaviors. Girls are more likely to run away from home because they may be ill equipped to deal with the emotional and/or physical conflicts that may occur. Single-parent homes were also positively associated with girls who ran away from home. Thinking about the stress associated with single parenthood work, raising an adolescent, finances, and other children in the household all may negatively stress the parent child relationship such that running away appears to be a viable alternative. Boys’ participation in certain types of delinquent acts appears to be influenced by gender socialization. As evidenced in this and previous studies, they are more likely to engage in fighting and getting arrested than

380

MELISSA A. MENASCO

girls. The nature of these delinquent acts is confrontational and challenging to authority, characteristics drawn from male role definitions. Boys’ social bonds are weaker than those of girls, which social control theory predicts. However, the social bond cannot explain the selection of a particular delinquent act, it can only explain why it occurs. By incorporating the ideas of gender socialization and gender roles, social control theory can explain why delinquency occurs as well as provide a better understanding of why the sexes select specific forms of delinquent behavior. Several conclusions about the nature of the social bond are evident from the results of this study: (1) Each of the elements of attachment, commitment, involvement, and belief operate differentially in the development of the social bond, (2) females appear to develop stronger social bonds than males, and (3) individual and family characteristics play a role in the development of delinquent behaviors.

REFERENCES Adler, F. (1975). Sisters in crime. New York, NY: McGraw-Hill. Agnew, R. (1985). Social control theory and delinquency: A longitudinal test. Criminology, 23(1), 47 61. Bachman, J. (1975). Youth in transition. Ann Arbor, MI: Inter-University Consortium for Political and Social Research. Bardwick, J., & Douvan, E. (1972). Ambivalence: The socialization of women. In J. M. Bardwick (Ed.), Readings on the psychology of women. New York, NY: Harper & Row. Belknap, J., Holsinger, K., & Dunn, M. (1997). Understanding incarcerated girls: The results of a focus group study. Prison Journal, 77, 381 405. Buchanan, C. M., 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, 63 107. Caspi, A., Lyman, D., Moffitt, T., & Silva, P. (1993). Unraveling girls’ delinquency: Biological, dispositional, and contextual contributions to adolescent misbehavior. Developmental Psychology, 29, 283 289. Cernkovich, S. A., & Giordano, P. C. (1992). School bonding, race, and delinquency. Criminology, 30(2), 261 291. Chesney-Lind, M. (1987). Girls’ crime and women’s place: Toward a feminist model of female delinquency. Paper presented in November at the American Society of Criminology meeting in Montreal, Canada. Chesney-Lind, M., & Shelden, R. G. (2003). Girls, delinquency, and juvenile justice (3rd ed.). Belmont, CA: Wadsworth. Corley, C. J., Cernkovich, S., & Giordano, P. (1989). Sex and the likelihood of sanction. The Journal of Criminal Law and Criminology, 80(2), 540 556.

Delinquency, Gender, and Social Control

381

Ellis, L. (1988). The victimful-victimless crime distinction and seven universal demographic correlates of victimful criminal behavior. Personality and Individual Differences, 9, 525 548. Eve, R. A. (1978). A study of the efficacy and interactions of several theories for explaining rebelliousness among high school students. The Journal of Criminal Law & Criminology, 69(1), 115 125. Fishbein, D. (1971). Selected studies on the biology of antisocial behavior. Psychosomatic Medicine, 33, 515 537. Foshee, V., & Bauman, K. E. (1992). Parental and peer characteristics as modifiers of the bond behavior relationship: An elaboration of control theory. Journal of Health and Social Behavior, 33(1), 66 76. Friedman, J., & Rosenbaum, D. P. (1988). Social control theory: The salience of components by age, gender, and type of crime. Journal of Quantitative Criminology, 4(4), 363 381. Frodi, A., Maccauley, J., & Thome, P. R. (1977). Are women always less aggressive than men? A review of the experimental literature. Psychological Bulletin, 84, 634 660. Giddens, A., Duneier, M., & Appelbaum, R. P. (2005). Introduction to sociology (5th ed.). New York, NY: W.W. Norton. Glueck, S., & Glueck, E. (1934). Five hundred delinquent women. New York, NY: Knopf. Gove, W. (1985). The effect of age and gender on deviant behavior: A biopsychosocial perspective. In A. S. Rossi (Ed.), Gender and the life course (pp. 115 144). New York, NY: Aldine. Harry, B., & Balcer, C. (1987). Menstruation and crime: A critical review of the literature from the clinical criminology perspective. Behavioral Sciences and the Law, 5, 307 322. Hindelang, M. J. (1973). Causes of delinquency: A partial replication and extension. Social Problems, 20, 471 487. Hirschi, T. (1969). Causes of delinquency. Berkeley, CA: University of California Press. Horney, J. (1978). Menstrual cycles and criminal responsibility. Law and Human Nature, 2, 25 36. Huebner, A. J., & Betts, S. C. (2002). Exploring the utility of social control theory for youth development: Issues of attachment, involvement, and gender. Youth & Society, 34(2), 123 145. Jenkins, P. H. (1995). School delinquency and school commitment. Sociology of Education, 68(3), 221 239. Jensem, G. J., & Eve, R. (1976). Sex differences in delinquency: An examination of popular sociological explanations. Criminology, 13, 427 448. Jensen, G. F., & Brownfield, D. (1983). Parents and drugs: Specifying the consequences of attachment. Criminology, 21, 543 554. Johnson, R. E. (1987). Mother’s versus father’s role in causing delinquency. Adolescence, 22(86), 305 315. Kim, J. G. (2003). The mechanism of power and control on adolescent sexual behavior: Explaining variation of parental control, religiosity, delinquent peers, and delinquency. Unpublished doctoral dissertation, University at Buffalo, Buffalo, NY. Krohn, M. D., & Massey, J. L. (1980). Social control and delinquent behavior: An examination of the elements of the social bond. The Sociological Quarterly, 21, 529 543. Liska, A. E., & Reed, M. D. (1985). Ties to conventional institutions and delinquency: Estimating reciprocal effect. American Sociological Review, 50, 547 560. Maccoby, E., & Jacklin, C. (1974). The psychology of sex differences. Stanford, CA: Stanford University Press.

382

MELISSA A. MENASCO

Matsueda, R. L. (1982). Testing control theory and differential association: A causal modeling approach. American Sociological Review, 47, 489 504. Messerschmidt, J. (1993). Masculinities and crime: Critique and reconceptualization of theory. Lanham, MD: Rowman & Littlefield. Pollak, O. (1950). The criminality of women. Philadelphia, PA: University of Pennsylvania Press. Rankin, J. H., & Kern, R. (1994). Parental attachments and delinquency. Criminology, 32, 495 515. Reckless, W. C., & Kay, B. A. (1967). The female offender. President’s Commission on Law Enforcement and Administration of Justice (n.p.). Seydlitz, R. (1991). The effects of age and gender on parental control and delinquency. Youth & Society, 23, 175 201. Shover, N., Norland, S., James, J., & Thornton, W. E. (1979). Gender roles and delinquency. Social Forces, 58(September), 162 175. Simon, R. J. (1975). The contemporary woman and crime. Washington, DC: U.S. Government Printing Office. Smith, D. A., & Visher, C. A. (1980). Sex involvement in deviance/crime: A quantitative review of the empirical literature. American Sociological Review, 45, 691 701. Thomas, W. I. (1923). The unadjusted girl. Boston, MA: Little, Brown. U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. (2004). Juvenile justice bulletin. Washington, DC: Government Printing Office. U.S. Federal Bureau of Investigation. (2015). Uniform crime reports. Washington, DC: Government Printing Office. U.S. National Criminal Justice Reference Service. Office of Justice Programs. (1979). In I. N. Bernstein, J. Cardascia, & C. E. Ross (Eds.), Defendant’s sex and criminal court decisions. San Francisco, CA: Jossey-Bass Publishers. Vedder, C., & Somerville, D. (1970). The delinquent girl. Springfield, IL: Charles C. Thomas. Visher, C. A. (1983). Gender, police arrest decisions, and notions of chivalry. Criminology, 21(1), 5 28. Wiatrowski, M. D., Griswold, D. B., & Roberts, M. K. (1981). Social control theory and delinquency. American Sociological Review, 46, 525 541. Wright, J. P., & Cullen, F. T. (2001). Parental efficacy and delinquent behavior: Do control and support matter? Criminology, 39, 677 706.

THE SELF AND THE ‘SELFIE’: CYBER-BULLYING THEORY AND THE STRUCTURE OF LATE MODERNITY Shahid Alvi, Steven Downing and Carla Cesaroni ABSTRACT Purpose This paper addresses the lack of conceptual and theoretical consensus around cyber-bullying and problems associated with overreliance on mainstream criminological thinking to explain this phenomenon. Methodology/approach The paper offers a critical criminological perspective on cyber-bullying encouraging scholars to engage with fundamental complications associated with the relationship between latemodernity, neo-liberalism and cyber-bullying. It argues for an approach that contextualizes cyber-bullying within the realities and consequences of late-modernity and neo-liberalism. Findings The paper argues that a robust understanding of cyberbullying entails contextualization of the problem in terms of the realities

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 383 406 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009016

383

384

SHAHID ALVI ET AL.

of consumption, individualism, youth identity formation and incivility in late modern society. Originality/value In addition to challenging extant theoretical approaches to cyber-bullying, the paper has important implications for intervention that surpass the limitations of law and order policies which tend to focus on criminalizing poorly understood bad behaviour or indicting internet technologies themselves. Keywords: Cyber-bullying; critical criminology; youth identity; internet technologies

INTRODUCTION The internet has transformed our experiences of the world around us, reinvented social relationships, and shone the spotlight of academic scrutiny onto social problems associated with its ubiquitous presence. One aspect of the internet human relations couplet garnering a great deal of public and media attention is cyber-bullying, typically, yet contentiously, defined as repeated and intentional electronic aggression against those who cannot easily defend themselves, perpetrated by people seeking power and control over others (Agatston, Kowalski, & Limber, 2007; Hinduja & Patchin, 2008; Kowalski, Giumetti, Schroeder, & Lattanner, 2014). In addition to a host of negative outcomes such as depression, lowered self-esteem, substance abuse, decreased school performance and other physical and psycho-social problems, tragic but nevertheless sensationalized cases of young persons’ suicide as a consequence of cyber-bullying have generated popular and scholarly concern (Beran & Li, 2007; Patchin & Hinduja, 2006; Privitera & Campbell, 2009; Ybarra, Diener-West, & Leaf, 2007). Interest in the topic has also been driven by the pervasiveness of new technologies and efforts to understand their impact, even as those technologies shift constantly. Upwards of 95% of 12 to 17-year-olds in the United States report using the internet, a proportion that has remained constant for nearly a decade (Madden, Lenhart, Duggan, Cortesi, & Gasser, 2013; Ybarra, 2004). The omnipresence of these technologies has also driven scholarly interest in explaining and responding to cyber-bullying, as illustrated by a recent Google scholar search for the term yielding nearly 19,000 results, almost 3,000 of which were from 2014 alone. Further, the constant

The Self and the ‘Selfie’

385

metamorphosis of communication technologies make it nearly impossible to study the impact of such developments on behaviour utilizing methodologies only permitting a surface and often incomplete snapshot of a phenomenon over which there is little conceptual consensus (Kowalski, Limber, & Agatston, 2008; Livingstone & Haddon, 2008; Schrock & Boyd, 2008). In effect, despite this growing attention, we know very little about the phenomenon, or even whether it deserves the kind of frenzied attention it tends to receive. It is crucial that the idea of cyber-bullying be subject to theoretical scrutiny because agreed upon definitions are foundational to any attempt to understand the incidence, prevalence and contours of cyber (or any kind of) misbehaviour (Tokunaga, 2010). Yet the vast majority of existing empirical and theoretical studies still struggle to achieve agreement on the fundamental characteristics of cyber-bullying (Cesaroni, Downing, & Alvi, 2012; Langos, 2012), a problem which confounds our understanding of it, and leads, among other things to wide variation in estimates of the incidence and prevalence of the problem (Juvonen & Gross, 2008). Most analyses of cyber-bullying characterize it as a growing menace with often catastrophic consequences. Others argue that it is simply an extension of ‘traditional’ bullying (schoolyard fights, shunning, taunting, etc.) and as such, a normal part of growing up (Campbell, 2005; Li, 2007). Some theoretical and empirical explorations imply that new digital technologies are directly or indirectly responsible for negative outcomes (Fryling, Cotler, Rivituso, & Mathews, 2014; Raskauskas, 2009), but by drawing attention to social structural factors in explaining online malfeasance, others maintain that cyber-bullying should not be blamed on internet technologies themselves (boyd & Marwick, 2011). In many cases then, discussions of cyber-bullying see it as one element of the ‘dark side’ of the internet itself an ‘ephemeral space of creation and destruction’ (Adler & Adler, 2008, p. 50) that baffles adults struggling to keep up with new technologies as used by children and youth (Bellah et al., 1985; Juvonen & Gross, 2008; Smith et al., 2008), in a world where ostensibly unlimited perils lurk around every digital corner. As we will argue in this paper, the confusion over cyber-bullying and what to do about it stems from the limitations of theoretical approaches that decontextualize the problem by assuming that the causes of ‘cyber-bullying’ lie primarily within the attributes of individuals or technologies themselves. Our arguments here are juxtaposed with those of a dominant strand of the criminological literature, what many have called ‘administrative’ or ‘orthodox’ criminology (Young, 1992), that relies on conceptions of human behaviour that ignore the crucial conditioning effects of social

386

SHAHID ALVI ET AL.

structure on everyday life, and demonstrate a ‘disconnection from power, [and] loss both of structure and meaning’ (Young, 2011, p. 198). Thus, this paper examines the phenomenon of cyber-bullying through the lens of critical criminology, a diverse set of perspectives that share at their core the notions that both inequality and social structure should be taken seriously if we are to begin to understand and respond appropriately to crime and deviance. In this regard, we draw attention to what we consider to be an under-examined but integral element of cyber-bullying theory the structure and consequences of late modernity.

Is Cyber-Bullying Just Another Type of Bullying? Extant research on cyber-bullying has been dominated by attempts to delineate its incidence, prevalence and correlates, the individual attributes and demographic characteristics of cyber-bullies and victims, and the role of group (bystander) behaviour in cyber-bullying. The variance in perpetration estimates is considerable, with some studies suggesting rates of 20% (reported in Hinduja & Patchin, 2008) and other studies reporting rates as high as 50% (as reported in Gibb & Devereux, 2014). Much of this variance can be explained by the lack of theoretical agreement on how to operationalize cyber-bullying (for a detailed discussion see Cesaroni et al., 2012). Empirical studies have identified a range of cyber-bullying correlates. These studies have suggested that cyber-bullying may or may not be a gendered phenomenon, with females and males appearing to have equal rates of victimization and perpetration in some studies (Slonje & Smith, 2008), other results suggesting that girls are more likely to be victimized than boys (Erdur-Baker, 2010; Lapidot-Lefler & Dolev-Cohen, 2014; Li, 2006), and some investigations suggesting no significant gender differences (Kowalski, Giumetti, Schroeder, & Reese, 2012; Smith et al., 2008). Another study suggests that men are more likely to be confrontational on social networking sites while women more likely to ‘stalk’ (obsessively check people’s status updates and learn about them) and less likely to bully or be confrontational (Turkle as cited in Price, 2011). One of the strongest correlations is the relationship between cyber-bullying and traditional bullying, with both victims and perpetrators of the latter indicating some experiences with the former (Ko¨nig, Gollwitzer, & Steffgen, 2010; Steffgen & Ko¨nig, 2009; Wright, 2014). Other studies have only begun to study cyber-bullying in relation to socio-economic classes or race (Tynes, Giang, Williams, & Thompson,

The Self and the ‘Selfie’

387

2008), although investigations of traditional bullying have shown very weak correlations with socio-economic status (Moore, Guntupalli, & Lee, 2010; Tippett & Wolke, 2014). These assessments constitute attempts to delineate factors distinguishing cyber-bullying from traditional bullying, and from other forms of deviance. In short, as a first step towards mitigating the problem, researchers are asking what is different and new about cyber-bullying and what factors might help us to explain the phenomenon. There are some elements of cyber-bullying that make it qualitatively different from traditional bullying, the most salient of which are its tendency to be perpetrated anonymously, yet often with consequences that are not anonymous. The internet is ‘a-spatial,’ and therefore lends itself to the kinds of ‘emotional dumping’ and ‘other outpourings of personal selfexpression that would never be tolerated in physical space’ (Erdur-Baker, 2010; Hayward, 2012; Suler, 2004, p. 141). Further, cyber-bullying differs from off-line bullying in its form, consisting of types such as harassment, flaming, trickery, ‘outing,’ exclusion, impersonation, cyberstalking, catfishing, trolling and sexting, to name a few types. Unlike other forms of bullying, the study of cyber-bullying also constitutes the pursuit of a moving target. As the noted physicist Heisenberg pointed out, our observations of phenomena are both influenced by the observer and shift the moment we try to pin those observations down (Henry & Milovanovic, 1996). This view helps to partially explain why there is no widely agreed upon definition of the problem, because unlike other harms, the mode of delivery (which also impacts on the definition of the infraction) keeps changing (Monks & Coyne, 2011). For instance, while it is still a dominant technology, Facebook is declining in popularity among teens, and is now used primarily by 25 to 34-year-olds; essentially, once well-established networking technologies are becoming passe´ as new communication technologies emerge virtually overnight (Global Web Index, 2014a, 2014b). Many studies identify cell phones and internet communication technologies such as email as the vehicles by which bullying is perpetrated, yet by the time these studies are published, newer communication technologies such as Instagram, Twitter and Snapchat, and now wearable technologies such as Google Glass and Kapture have already superseded technologies like Facebook, rendering them old and unfashionable, but still raising the potential of such tools to facilitate bullying of others. The consequences and contours of these newer technologies are virtually unknown. Consider for instance the possibility that a person wearing Google glass could be recording every motion, conversation, and social interaction of

388

SHAHID ALVI ET AL.

another person or persons, upload these instantaneously to a social networking site, complete with commentary, for consumption by a ‘networked public’ of large audiences1 (boyd, 2008). In all likelihood, the capacity of Google glass to facilitate covert observation of human interaction opens up all kinds of possibilities for forms of cyber-bullying we have yet to witness, or for that matter, for outcomes that have positive intent and consequences. The point is that we do not know to what uses, positive or negative, emerging technologies will be put. The dominant forms of explanation for cyber-bullying suggest that it is both new and insidious, yet in terms of explanation, do not differ from other theories of deviant behaviour, including traditional bullying. These studies argue that cyber-bullying has the same characteristics and motives as traditional bullying except that it is mediated and carried out via technology. Indeed, one of the pioneers of cyber-bullying research, Dan Olweus, argues that cyber-bullying has been overblown by the media, and is in fact a low-prevalence phenomenon (Olweus, 2012). Many scholars have argued that cyber-bullying is simply an extension of traditional bullying. Li (2007) opines that correlations between the two forms mean that cyber-bullying is the ‘same old wine in a new bottle,’ while others have suggested similarly that it is an old problem in a new guise (Campbell, 2005). They maintain that there are similarities between traditional bullying as a mode of control and cyber-bullying as a mode of control mediated and delivered through the mechanisms of electronic technologies. Thus, the differences between traditional bullying and cyberbullying are that the latter is often, but not always, anonymously perpetrated, and is expansive in its potential harm because the networking capacity of the internet means that the audience is theoretically unlimited, notwithstanding research findings suggesting that most victims know, or at least believe to know, who is bullying them and that typically the bully is often located within their social network (Hinduja & Patchin, 2010). Regardless of the differences in form, then, both cyber-bullying and traditional bullying have real and sometimes serious consequences in terms of harm (Raskauskas & Stoltz, 2007). It is not surprising, then, that within orthodox criminology, equating cyber-bullying with forms of criminal behaviour has encouraged the use of explanatory frameworks that have themselves had mixed success in explaining crime in general. A brief survey of the burgeoning literature on cyberbullying includes tests of control, strain, general strain, routine activities, techniques of neutralization, and other widely tested sociological and social psychological theoretical approaches (Holt & Bossler, 2013). Among the

The Self and the ‘Selfie’

389

more promising of these approaches include theories of low self-control, and social learning theories because both emphasize the relationship between individual behaviour, accountability, cooperation, empathy and social identity (Cesaroni et al., 2012). The broader field of deviance studies has been confronted with similar challenges. As Goode (2004) has pointed out, the sociology of deviance has, to a large extent abandoned the study of power and social structural concerns, positioning itself instead within the realm of micro level, symbolic interactionist studies, and ‘abdicating’ to criminologists questions of the etiology of deviance. This development does not mean that symbolic interactionism’s established studies of stigma, labelling and identity cannot add to our understanding of crime in general, and cyber-bullying in particular. For instance, labelling theory and work on the negative consequences of stigma have had an enormous impact on criminal justice policy concerning young offenders, in that they form one of the core rationales for diversionary policies (Alvi, 2012). Nevertheless, the core principles of symbolic interactionism focus attention on the motivations of reflexive individuals interacting to produce social reality, not on the significance of external social forces (Meltzer, Petras, & Reynolds, 1975; Rafferty & Vander Ven, 2014). Yet another view within the study of micro relations and in particular around questions of identity can be found in the debate between postmodern thought and symbolic interactionism and philosophical pragmatism generally. As distinguished from the notion of the ‘condition of postmodernity,’ which we identify as synonymous with late capitalism (see Harvey, 1990), the postmodern ‘turn’ in the social sciences promoted the idea that there can be no self, and that the concept must be abandoned. Adler and Adler (1999, p. 32, 2008) summarize this genre of thought as a rejection of modernist views of the real self substituted by a view in which the ‘quest for self-presentation has replaced the quest for meaning ….[it is] devoid of a core, it is decentered and ultimately dissolved’. Against this view we propose that there is nothing ‘post’ about the ‘postmodern self’. Rather, our position is that the constitution of identity and the self is a ‘joint accomplishment, neither completely determined by the social world nor pre-given at birth’ (Callero, 2003, p. 121). It is a product of both reflexive agency and the modern material realities inhering in structural conditions, an argument that has been supported by the research of numerous scholars (see, e.g. Adler & Adler, 2008; Kendall, 1998; Lemert, 1997). If some varieties of postmodernism wish to propose that there can be no ‘grand narrative,’ no ‘position,’ (ironically, that seems to be their position), that everything, including identity must be deconstructed, and that the realities of modernist

390

SHAHID ALVI ET AL.

society no longer prevail, then surely there is no point in analysing cyberbullying, or any other type of potential harm as rooted in real values and material relations transmitted from the social world. Accordingly, our argument here is that harms are not ethereal, ‘hyperreal’ constructs. They are real. As MacKinnon (1999, p. 706) asks: Can postmodernism identify fascism if power only exists in microcenters and never in systematic, fixed, and determinate hierarchical arrangements? How can you oppose something that is always only in play? How do you organize against something that isn’t even really there except when you are thinking about it? Can postmodernism hold the perpetrators of genocide accountable? If the subject is dead, and we are dealing with deeds without doers, how do we hold perpetrators accountable for what they perpetrate?

Neither postmodern theories of identity and the self nor orthodox criminology provide us with satisfactory answers to the question of what is to be done about cyber-bullying because they portray such behaviour simply as fluid ‘performances’ in cyberspace (Kendall, 1998) in the former case, or decontextualized malfeasance in the latter. Indeed, as most critical criminologists point out, the policy and intervention implications of most of these approaches emphasize some combination of early intervention, punishment or control, responses that do not address the crucial role of broader social and cultural spheres in creating environments in which problematic behaviour is inevitable. Consider, for example, the problem of mens rea with respect to cyberbullying, and the corollary that cyber-bullying may not be seen as harmful by perpetrators, particularly teen perpetrators. Since they have been recognized in most Western jurisdictions as having diminished culpability and blameworthiness, what does this mean for theoretical approaches that implicitly or explicitly assume a utilitarian foundation for criminal or deviant behaviour? Both classical and neo-classical traditions in criminology view deviance and crime as rational choices made by reasoning actors who weigh the utility of those choices on a scale maximizing pleasure and minimizing pain. Yet the vast majority of empirical quantitative studies of criminal or deviant motivation do not explain enough variance to be confident in the strength of the perspective, and we would certainly not be continuing the search for an overarching and fully explanatory model if we had found one already. Similarly, the family of control theories in criminology are based on the premise that individuals are in need of formal and informal social controls to curb external and internal triggers that might impel them towards deviant behaviour. In both perspectives, the level of analysis is

The Self and the ‘Selfie’

391

individual, with a cursory nod towards the role of social forces in conditioning and setting limits on the choices people make. Another example from the vast literature on men’s violence against women suggests similar challenges to theories that do not consider broader contexts. That literature has repeatedly pointed to the endemic nature of violence in society, noting in defence of this proposition that violence is historically and culturally invariant. In short, this body of work reminds us that as a rule, men do not beat, sexually harass and psychologically abuse women because they are psychologically abnormal, because of alcohol or drugs, or because they had a bad day. They do these things because they seek power and control in a world they think entitles them to dominance, but that they perceive has stripped them of the capacity to exercise that control, and because the cultural web they live in grants them the psychological and emotional resources to choose violence as a means of negotiating everyday life. They abuse because they can, and because they have been given ‘permission’ to do so. We know that cyber-bullying is a relatively new construction, forged by adults who are trying to come to grips with a phenomenon that defies easy categorization and analysis. This inability, amplified by media generated anxiety, has resulted in the coupling of cyber-bullying with terms such as ‘epidemic,’ ‘pandemic,’ and ‘dangerous’. Yet the problem is not that cyberbullying is an epidemic, it is that its antecedents are endemic to modern societies. Yet there has, to the best of our knowledge, been no explicit recognition of the roles played by structural, cultural and institutional arrangements in explanations of cyber-bullying, and by extension, in law. One very good example of a misguided legal approach to dealing with an alleged problem is found in the recent debates in Canada over Bill C-13, legislation that will come into force in March of 2015. This bill, entitled the Protecting Canadians from Online Crime Act essentially tries to control cyber-bullying by providing police with greater powers to use ‘modern tools’ such as expanded investigative powers (with proper judicial oversight), and the creation of new production orders for specific information and the criminalization of non-consensual distribution of intimate images and gender based hate information (Canada Parliament House of Commons, 2014). Similar legal considerations exist in the United States, where the Supreme Court ruled in Roper vs. Simmons and in Miller vs Alabama based on amicus curiae of the American Psychological Association, that young people simply do not have the same moral compass we would expect adults to navigate by. Thus, if it appears to be clear in criminal law that young people are not developmentally ready to make

392

SHAHID ALVI ET AL.

many decisions that can result in potentially harmful consequences, what does this tell us about the moral culpability of young people when it comes to potential harms to others caused by their use of the internet? Internet associated behaviours seem to have simultaneously prompted adults, legislators and law enforcement to attempt to control and contain behaviour within the technological social spaces occupied by youth, while ignoring young peoples’ right to developmental considerations. To base responses to cyber-bullying on the principles of general and specific deterrence is to take an approach to dealing with a problem (whose very definition nobody can agree on) that refuses to acknowledge that cyber-bullying is rooted in much more fundamental problems that go beyond the limitations of the reach of law. It is not surprising that sensationalized media accounts of cyber-bullying do little to inform the public and politicians about the real nature of the problem. The expectation seems to be, if we just criminalize bad behaviour on the internet, our children will be safer and people will behave better the problem is solved, and the epidemic is over. It is tempting to explain cyber-bullying through the lens of mainstream theories of crime, deviance and bullying, but it is important to move carefully in this direction, and with a clear commitment to engagement with deeper, more fundamental sociological problems. Theoretical canons offer the opportunity to publish without engaging in deeper political, sociological and philosophical questions that may challenge what Hall and Winlow refer to as a ‘preferred image of reflexivity [which is] largely cosmetic’ (2012, p. 7). Nor can explanations of cyber-bullying be read off from traditional explanations of crime, when, as it is contended here, cyber-bullying itself may only be one expression of a larger set of problems. By focusing on the technological advances and individual pathologies that have allegedly facilitated the ‘rise’ of cyber-bullying, we run the risk of rendering invisible the cultural and political contexts that give rise to it in the first place. Let us consider, then, the outlines of these contexts, and their role in creating the conditions in which phenomena like cyber-bullying can exist.

The Self and the ‘Selfie’ in Late Modernity According to theorists such as Giddens (2013), Beck (1992), Bauman (2000), and Lash (1999) and Young (Young, 2007), late modernity is tied inextricably to neoliberal social forces including globalization, welfare state decline, increased social mobility, flexible employment, insecurity in

The Self and the ‘Selfie’

393

personal relationships, and the emergence of the ‘market’ as a template for social values (Buckingham, 2008; Giroux, 2003, p. 2). One aspect of these transformations involves the loss of influence of customary or traditional values which include kindness, civility, restraint and empathy, towards a situation in which identities and the rules governing them are ‘liquid’ and negotiable (Bauman, 2000). It is true that in some cases, the internet opens up the possibility of new virtual communities that might fill the gaps left by the demise of traditional community and the cohesion that accompanied it (Adler & Adler, 2008), or may provide spaces for collective action or dissent (Fernback, 2002). However, it is also true that when people go online, they do not completely abandon any material identity or the culture in which it is embedded in favour of a radically different ‘self’. Rather, as Adler and Adler (2008, p. 48) point out, ‘the majority of people who venture into cyber worlds remain firmly grounded in the reality outside, content to lurk and interact sporadically’. Thus, young people do not, as a rule, completely abandon the material world in which they are grounded; both their online persona and occasional missives are conditioned by the values they live by and have been socialized to internalize. Moreover, the fact of the internet and its enclosure within late modernity has meant that significant others are no longer only available as reference points in face to face encounters, but also as a web of electronic ‘generalized elsewheres’ (Young, 1999). Notions of the self, including feelings of shame, empathy, respect and other norms become open to discourses that are global rather than localized in communities that are weakening or fragile, and these globalized discourses are now dominated by the self-centred language and norms of late modern capitalism. Late modern capitalist societies embody a set of values emphasizing consumption, identity, ‘success’ and hyper-individualism, which in turn can foster and nourish uncivil, hurtful behaviour (Young, 2011). As France (2007, p. 125) points out, the internet can enable the construction of the virtual self invisibly, in terms of ‘who you are, where you live and what you look like’. It is a place where you can ‘write yourself into being’ (boyd, 2008, p. 129), a location where one can create a contrived ‘self’ via the immediacy and fictitiousness of the ‘selfie’.2 Despite even fairly mainstream theoretical frameworks acknowledging relationships between macro forces and identity (e.g. Durkheim’s theory of anomie), research applying a ‘self-creation’ framework to cyber-bullying remains focused on individual, cognitive differences, tests of individual-focused strain rather than structural anomie (e.g. Hay, Meldrum, & Mann, 2010) or to the link between bullying and suicidal ideation (e.g. Henry,

394

SHAHID ALVI ET AL.

Lovegrove, Steger, Chen, & Cigularov, 2014) rather than systemic and structural forces (see, e.g. Suler, 2004). Challenging the concept that the process of creating the self as stemming solely from individual factors, critical observers of the dynamics of late modernity note shifts from forms of social organization in which certainty, behavioural norms and culturally and generationally transmitted values morph into a world in which those traditions and certainties are potentially diminished in favour of ‘self-made’ identities (Bauman, 2000; Beck, 1992; Giddens, 2013). Giddens observes that neoliberal societies breed a cultural fabric within which we have extraordinary freedom to choose who we are and how we act, while tacitly surrendering to norms of responsibilization for actions circumscribed by ‘expert’ guidance around the quest for selfimprovement, success and individuality (Giddens, 2013). As Young (1999, p. 14) argues, the late modern world is, in effect, a world of ontological insecurity, characterized by heightened choices, constant interrogation of established beliefs and certainties, self-reflexivity, a lack of embedded biography and the incessant confrontation with a plurality of world views and beliefs. Internet technologies also facilitate and advertise mass consumption of tangible and nontangible capital (e.g. ‘friends’ on Facebook). This consumption-orientation yields strong potential to focus institutional attention away from formal and informal relational dynamics (such as the family or workplace) to ones centred on economic success. Essentially, this overwhelming focus on consumption and economic prosperity amount to what Messner and Rosenfeld (2007) have identified as institutional anomie. Within the context of an anomic late modernity, new information technologies permit us to create our own ‘selves,’ to disseminate those selves across spatial borders, instantaneously. These technologies encourage ‘disembeddedness’ from social context, the invention of the self by the self as contrasted with the manufacture of identity as shaped by tradition and certainty. It should not be inferred that such traditions were in any sense ‘better’ than the options afforded people today. Indeed, they were much worse on a number of dimensions, including, but not limited to standards respecting sexuality, race and gender that were and continue to be oppressive. The point is that there were limited and circumscribed choices in relation to norms of conduct, of civility, and what was acceptable behaviour in light of the inevitable (face-to-face) conflict that is a hallmark of being human. As Durkheim (1893) suggested, a mechanical division of society holds the potential to alienate and isolate individuals, uprooting them from the grounding of norms, values and mores. Not long ago, one could have a conversation with another, in person, and never have the temerity to call

The Self and the ‘Selfie’

395

that individual names or trumpet their inferiority. In contrast, life in the Internet age permits us to do and say whatever we like, anonymously if we wish, often with impunity. Given such shifts, one is prompted to consider the orientation of the individual within larger social structural dynamics, of which the Internet is undeniably a central component. If this placement orients the individual as a normless (i.e. anomic) actor, even mainstream criminological theories that acknowledge the importance of culturally transmitted goals such as Merton’s (1938) will likely be unable to clearly define what constitutes economic, social and culturally bound ‘goals’ and ‘means’ for youth, much less to what extent these youth navigate the pursuit and acquisition of such goals through mechanical or organic divisions within their increasingly digital social spheres. In the end, neoliberal discourses inherent in late modernity involve an emphasis on ‘compulsory individuality’ shaped by political and economic discourses aligned with marketplace values of consumption, success and competition (Willett, 2008, p. 55) essentially, the ‘American dream’ of unbounded individuality (Messner & Rosenfeld, 2007). There is some good evidence for such claims. A recent survey of Internet users reported that 61% of youth aged 16 to 24 said that they use the internet to take on a different personality, underscoring the pervasiveness of the search for identity through electronic anonymity (Global Web Index, 2014a, 2014b). It is well known that adolescence is a time characterized by the search for identity and independence, and the quest for inclusion, popularity and status. In late modernity, however, search is complicated by the plurality of choices and the associated lack of certainty with respect to standards and values, a relegation of bygone ‘moral compasses’ which for better or worse constituted a limited range of choices available to young people, to the potential for numerous choices among an infinitely long menu of options. And it is likely that the menu of identity choices young people are confronted with today emphasizes the holy grail of ‘success’ in terms that are individualistically rather than collectively focused. Indeed, such theoretical conceptualizations are not far afield from those employed even by many mainstream criminologists who employ, for example, concepts of collective efficacy with respect to neighbourhood influences on crime (Sampson, Raudenbush, & Felton, 1997). In the case of late modernity and the Internet, neighbourhoods are, especially for youth, often online. There is more than a hint of empirical evidence for this contention in research examining values and happiness. For instance, in their survey of 10,000 middle and high school students from 33 schools, Weissbourd & Jones (2014) and his colleagues have documented that a large majority (80%) of

396

SHAHID ALVI ET AL.

young people value aspects of personal success, defined as a combination of achievement and happiness, over concern for others (20%), and that this finding does not vary across the spectrum of race, class, or culture. Interestingly, as Western societies have gotten richer, their people have become no happier (Mohr, 2011). Further, as Bryant and Bryant (2005, p. 415) note, teens are subject to both branded and viral marketing influencing their purchasing decisions to the tune of billions of dollars, while at the same time promoting ‘underdeveloped or superficial relationships with online communication partners, [decreasing] face-to-face communication with friends and family, and [promoting] lower quality, online conversations instead of ‘richer’ face-to-face and telephone ones’. Similarly, data on attitudes towards the state of civility in the United States point out that 95% of Americans believe incivility to be a problem, 70% believe that the internet encourages uncivil behaviour, and 66% feel that social media has allowed people to engage with each other less civilly (WeberShandwick & Powelltate, 2014). But is this the fault of the internet and information communication technologies, as so many have implied? Is it the consequence of psychological or social deficiencies, as some mainstream criminological thought would suggest? Or is cyber-bullying a manifestation of what some people want to say, could not in the past, but now can? Is technology merely the lever by which we can efficiently and anonymously oppress others? Should the positive correlations between cyber-bullying and being a victim of traditional bullying (Steffgen & Ko¨nig, 2009), the act of cyber-bullying and the presence of bystanders (Lapidot-Lefler & Dolev-Cohen, 2014), and the link between being a traditional bully and a cyber-bully surprise us? (Jang, Song, & Kim, 2014; Tokunaga, 2010). With respect to the role of bystanders, the extent to which cyber-bullying is perpetrated as a group rather than as individuals is still unclear, as is the role of ‘contagion’ or deindividuation (Le Bon, 1897). Further research on the relationship between anonymity, group behaviour, de-individuation and online bullying may prove useful in understanding the phenomenon, particularly if such research accounts for the social milieu that conditions and shapes the correlates of de-individuation, such as lack of self-monitoring, and diminished attention to the individuality of group members (Haines & Cheney Mann, 2011), and the potential of emotions in mitigating or encouraging negative collective behaviour (Chmiel et al., 2014). More broadly, research is needed on cyber-bullying and the significance of youth peer culture particularly in relation to the ways in which these cultures both influence, and are influenced by the vagaries of the adult world. A rich body of literature on youth

The Self and the ‘Selfie’

397

peer culture indicates that children create meaning from their interpretations and reformulations of adult culture, through which they create identities around constructs such as gender and ethnicity (Eder & Kinney, 1995; Kyratzis, 2004). At the same time, such investigations must account for the influence of social structure, culture and power as crucial elements of the socialization of youth (Corsaro & Eder, 1990). Survey data from Global Web Index indicates that nearly two thirds of youth agree that people who bully online also bully in person, and 3 out of 5 agree that people bully because it’s ‘cool’. Almost half believe that people bully because they’ve been bullied by someone else. Further: In general, 67% believe people bully online because they can do it without getting caught. More than 75% thought because people bully because it is easier than face-toface confrontation, 58% believed it is because the bully is not happy with their own life, and 25% feel it’s because there are no emotional consequences. (Global Web Index, 2014a, 2014b)

These data highlight the importance of taking seriously the issue of how norms around civility are instilled and practiced in social relationships, and certainly, there has been some promising scholarship calling attention to the significance of moral disengagement, normative beliefs and perceptions of what is and is not acceptable behaviour in relation to cyber-bullying (Bussey, Fitzpatrick, & Raman, 2014; Kowalski et al., 2014; Renati, Berrone, & Zanetti, 2012; Robson & Witenberg, 2013; Williams & Guerra, 2007). Indeed, the findings from such studies may help to explain consistent significant positive correlations between traditional bullying and cyberbullying, in that the Internet is really just providing new opportunities to extend conventional bullying into cyberspace (Shariff, 2005; Shariff & Hoff, 2007; Wolak, Mitchell, & Finkelhor, 2007). Consider also the tendency to pathologize youth misbehaviour as demonstrated in work attempting to identify the characteristics delineating ‘problem’ youth internet behaviour from those who are ostensibly ‘normal’ bullies. For example, some scholars have noted that internet aggressors face multiple psychosocial challenges, including poor relationships with parents, substance use and other forms of delinquency (Ybarra & Mitchell, 2004a). Others have noted correlations between cyber-bullying behaviour and low scores on social skills indexes (measuring cooperation, assertion, empathy and selfcontrol) (Lapidot-Lefler & Dolev-Cohen, 2014), smoking, low school commitment and ‘anti-social’ behaviour (Cappadocia, Craig, & Pepler, 2013; Erdur-Baker, 2010; Hinduja & Patchin, 2008; Ybarra, Espelage, & Mitchell, 2007; Ybarra & Mitchell, 2004b). Clearly, these factors are not

398

SHAHID ALVI ET AL.

unique to internet bullies, but are instead functions of social environments that both condition and set limits on the possibilities of behaviour. Another complication is that children’s definitions of what constitutes bullying and cyber-bullying may be challenging traditional cultural and legal definitions of childhood. Take for example the uncertainty over phenomena such as ‘sexting,’ which some argue is to varying degrees a perfectly normal form of sexual expression albeit mediated by new forms of communication, and perhaps over policed by legal norms and strictures whose purpose is to prevent the victimization of children by adults. Those in this camp also often seek to balance responses to sexting by taking account of levels of coercion versus voluntary consent. Others express concern over appropriate sexual behaviour as interpreted through the lens of adult anxiety anchored in their own personal experiences growing up and the conviction that the role of parents and the law is to protect the best interests of the child, or with respect to girls, that sexting reflects the hyper sexualization of girls and women in modern cultures (Simpson, 2013). Here, we suggest the focus shifts from the characteristics of pathologized individuals reflecting methodological individualism, to the characteristics of the social world in which individuals are embedded (methodological holism). It is clearly important then, from this point of view, that any account of cyber-bullying also take into account changing perceptions of childhood and adolescence, with respect to the tendency in modern societies to control and manage away the risks supposedly embodied by wayward youth (Alvi, 2012), towards the late modern preoccupation with what might happen to youth in the hostile digital world. This might, for example, include concerns around young women’s behaviour in the digital world. ‘Slut shaming,’ anxiety around sexting may be a reversion to fears around the morality of young women and the need to control their sexuality and bodily integrity. Most of the available data on cyber-bullying is quantitative rather than qualitative, with few researchers attempting to determine, in particular, the perspective of youth themselves on the nature, normality and consequences of what adults call cyber-bullying. In short, while what we think we know about cyber-bullying is contestable in terms of incidence, prevalence and the short or long term consequences of cyber-bullying, we know comparatively very little about its meaning from the point of view of those who are victims and perpetrators because we have not really asked them (deLara, 2012; Olweus, 2012, p. 6; but see Agatston et al., 2007), and more important, we may not be asking them the right questions (Vandebosch & Van Cleemput, 2008). In effect, the internet and its technologies are probably

399

The Self and the ‘Selfie’

not the problem here, and perhaps we should be focusing our attention on the ways in which the normal behaviours of young people, such as seeking attention, self-validation, normative direction and status are related to the internet as an empty space in which anything is possible (boyd & Marwick, 2011). They are behaviours embedded in a social world that is unable to provide guidance because it suffers itself from an ambiguity of meanings posing as a finite, finished world governed by the value template of the market emphasizing success and happiness through consumption. As Weissbourd and Jones remind us (2014, pp. 1 3): Any healthy civil society also depends on adults who are committed to their communities and who, at pivotal times, will put the common good before their own. We don’t seem to be preparing large numbers of youth to create this society ….Sooner or later, we will need to take on the large and fundamental problem of the messages that our society sends to our children about the definition of success and about what it means to be an ethical member of a community.

CONCLUSION Cyber-bullying is a social problem mediated by technology, not an individual problem of maladjustment or pathology. If we are to understand the issue theoretically, we must begin with an attempt to explicate a conceptual framework serving as a heuristic device that prepares the groundwork for a substantive theory lending itself to empirical investigation (Mouzelis, 2003). The quest for a criminology focusing not solely on the efficacy of extant theory but on the ways in which we conceptualize criminality in light of the new ‘dark age’ we live in now (Giroux, 2004) is the key to advancing our ideas about how to prevent and mitigate the very real damage cyberbullying can cause. But we do not see this dark age in pessimistic terms, though there are good grounds for that. Rather, we see it as a dark age in the sense that there is too much that we do not know, and much that we have blinded ourselves to. A project of recalibrating the lens through which we examine cyberbullying will entail a serious analysis of the economic, political and cultural workings and consequences of consumer capitalism. In turn, such concerns would benefit from a critical reading of late modernity that attempts to illuminate the meanings attached to identity, civility and ‘getting along’ in the current era. As we have alluded to, this pursuit need not necessarily oppose

400

SHAHID ALVI ET AL.

theories such as anomie (in its various forms), and can instead act as an adhesive for theoretical integration. Nevertheless, we continue to struggle, blindfolded, with theoretical approaches that are simply not equipped to address central and pivotal transformations in modern social life, including technological innovation (warts and all), and neoliberalism with all its accompanying pathologies. Bryce and Fraser (2013, p. 785) note that youth understand ‘the role of mediated communication in facilitating cyberbullying and accept this as a routine feature of online interaction that is contextualized within their everyday lives’. It is precisely the banal acceptance of the inevitability of the ‘routine’ that requires theoretical attention here. We need to refocus attention on what matters, including the underlying issues of racism, sexism, homophobia, derision of physical and mental abilities, sexual orientation, poverty, inequality, social exclusion and the like, and as we have suggested in this paper, issues around the superficiality of much of everyday life, embodied in the reification of popularity, success, fashion and ‘coolness’. We do not accept that such hollowness is ‘just the way it is,’ or as some postmodern analysis might suggest, that ‘society happens in your head’ (MacKinnon, 1999, p. 701). Some scholars have suggested that cyberspace is a testing ground in which people can experiment with identities that might be instantiated in the real world (Kitchin, 1998), but we would argue that the opposite is also true. If the sources of online malfeasance can be located in the pathological, uncivil and criminogenic skeleton of late modernity, then a preventative strategy must take seriously the ways in which the late modern world has unfolded and at the same time limited the behavioural options available to all of us, the ways in which it has commodified those choices, and the potential consequences of those choices. Should children be taught, therefore to use the internet in productive, progressive and empathetic ways? Of course they should, but they should also be taught the cultural competencies offered through the lens of critical media literacy and the foundational principles of critical thinking, so that they may begin to question and resist the salience and legitimacy of a society that has made it so easy to embrace, hold and express hurtful attitudes and behaviours.

NOTES 1. Indeed, the tragic case of ‘cyber-spying’ involving Tyler Clementi is a good illustration of what is possible with older technologies, and portends what may come very soon.

401

The Self and the ‘Selfie’

2. A selfie is a usually flattering portrait of oneself uploaded to a social networking site.

REFERENCES Adler, P. A., & Adler, P. (1999). Transience and the postmodern self. The Sociological Quarterly, 40(1), 31 58. Adler, P. A., & Adler, P. (2008). The cyber worlds of self-injurers: Deviant communities, relationships, and selves. Symbolic Interaction, 31(1), 33 56. Agatston, P. W., Kowalski, R., & Limber, S. (2007). Students’ perspectives on cyber bullying. Journal of Adolescent Health, Youth, Identity and Digital Media, 41, s59 s60. Alvi, S. (2012). Youth criminal justice policy in Canada: A critical introduction. New York, NY: Springer. Bauman, Z. (2000). Liquid modernity. Cambridge: Polity Press. Beck, U. (1992). Risk society: Towards a new modernity. Thousand Oaks, CA: Sage. Bellah, R., Madsen, R., Sullivan, W. M., Swidler, A., & Tipton, S. (1985). Habits of the heart. Berkeley, CA: University of California Press. Beran, T., & Li, Q. (2007). The relationship between cyber-bullying and school bullying. Journal of Student Wellbeing, 1(2), 15 33. boyd, D. (2008). Why youth ! social network sites: The role of networked publics in teenage social life. In Youth, identity, and digital media (pp. 119 142). Cambridge, MA: MIT Press. boyd, D., & Marwick, A. (2011). Why cyber-bullying rhetoric misses the mark. The New York Times. Bryant, P., & Bryant, A. (2005). Adolescents and the internet. Adolescent Medicine Clinics, 16(2), 413 426. Bryce, J., & Fraser, J. (2013). It’s common sense that it’s wrong: Young people’s perceptions and experiences of cyber-bullying. Cyberpsychology, Behavior and Social Networking, 16(11), 783 787. Buckingham, D. (2008). Introducing identity. In D. Buckingham (Ed.), Youth, identity and digital media (pp. 1 24). Cambridge, MA: MIT Press. Bussey, K., Fitzpatrick, S., & Raman, A. (2014). The role of moral disengagement and selfefficacy in cyber-bullying. Journal of School Violence, 1 17. Callero, P. L. (2003). The sociology of the self. Annual Review of Sociology, 29(1), 115 133. Campbell, M. A. (2005). Cyber bullying: An old problem in a new guise? Australian Journal of Guidance and Counselling, 15(01), 68 76. Canada Parliament House of Commons. (2014). Protecing Canadians from online crime act. House of Commons Debates, Hansard. Cappadocia, M. C., Craig, W. M., & Pepler, D. (2013). Cyber-bullying: Prevalence, stability, and risk factors during adolescence. Canadian Journal of School Psychology, 28(2), 171 192. Cesaroni, C., Downing, S., & Alvi, S. (2012). Bullying enters the 21st century? Turning a critical eye to cyber-bullying research. Youth Justice, 12(3), 199 211.

402

SHAHID ALVI ET AL.

Chmiel, A., Sienkiewicz, J., Paltoglou, G., Buckley, K., Sckowron, M., & Thelwall, M. et al., (2014). Collective emotions online. In N. Agarwal (Ed.), Online collective action, lecture notes in social networks (pp. 59 74). New York, NY: Springer. Corsaro, W. A., & Eder, D. (1990). Children’s peer cultures. Annual Review of Sociology, 16, 197 220. deLara, E. W. (2012). Why adolescents don’t disclose incidents of bullying and harassment. Journal of School Violence, 11(4), 288 305. Durkheim, E. (1893). 1964. The division of labor in society. New York, NY: Free Press. Eder, D., & Kinney, D. A. (1995). The effect of middle school extra curricular activities on adolescents’ popularity and peer status. Youth & Society, 26(3), 298 324. Erdur-Baker, O. (2010). Cyber-bullying and its correlation to traditional bullying, gender and frequent and risky usage of internet-mediated communication tools. New Media & Society, 12(1), 109 125. Fernback, J. (2002). The individual within the collective: Virtual ideology and the realization of collective principles. In S. Jones (Ed.), Virtual culture: Identity and communication in cybersociety (pp. 36 53). London: Sage. France, A. (2007). Understanding youth in late modernity. Milton Keynes: Open University Press. Fryling, M., Cotler, J., Rivituso, J., & Mathews, L. (2014). Cyber-bullying or normal game play? Impact of age, gender, and experience on cyber-bullying in multi-player online gaming environments: Perceptions from one gaming forum. Paper presented at the Proceedings of the Conference for Information Systems Applied Research ISSN: 21671508, Baltimore, MD. Gibb, Z. G., & Devereux, P. G. (2014). Who does that anyway? Predictors and personality correlates of cyber-bullying in college. Computers in Human Behavior, 38, 8 16. Giddens, A. (2013). Modernity and self-identity: Self and society in the late modern age. New York, NY: Wiley. Giroux, H. (2003). Racial injustice and disposable youth in the age of zero tolerance. International Journal of Qualitative Studies in Education, 16(4), 553 565. Giroux, H. A. (2004). The terror of neoliberalism: Authoritarianism and the eclipse of democracy. Hemdon, VA: Paradigm Publishers. Global Web Index. (2014a). GWI social summary. London: Trendstream Limited. Global Web Index. (2014b). Facebook rooms reflects strong demand for anonymity Globalwebindex Q3, 2014. Globalwebindex Q3, 2014. Retrieved from http://insight.globalwebindex.net Goode, E. (2004). Is the sociology of deviance still relevant? The American Sociologist, 35(4), 46 57. Haines, R., & Cheney Mann, J. E. (2011). A new perspective on de-individuation via computermediated communication. European Journal of Information Systems, 20(2), 156 167. Hall, S., & Winlow, S. (2012). New directions in criminological theory. London: Routledge. Harvey, D. (1990). The condition of postmodernity: An enquiry into the conditions of cultural change. Oxford: Blackwell. Hay, C., Meldrum, R., & Mann, K. (2010). Traditional bullying, cyber bullying, and deviance: A general strain theory approach. Journal of Contemporary Criminal Justice, 26(2), 130 147. Hayward, K. (2012). Cultural geography, space and crime. In S. Hall & S. Winlow (Eds.), New directions in criminological theory (pp. 123 144). London: Routledge.

The Self and the ‘Selfie’

403

Henry, K., Lovegrove, P., Steger, M., Chen, P., & Cigularov, K. (2014). The potential role of meaning in life in the relationship between bullying victimization and suicidal ideation. Youth and Adolescence, 43(2), 221 232. Henry, S., & Milovanovic, D. (1996). Constitutive criminology: Beyond postmodernism. London: Sage. Hinduja, S., & Patchin, J. W. (2008). Cyber-bullying: An exploratory analysis of factors related to offending and victimization. Deviant Behavior, 29(2), 129 156. Hinduja, S., & Patchin, J. W. (2010). Bullying, cyberbullying, and suicide. Archives of Suicide Research, 14(3), 206 221. Holt, T. J., & Bossler, A. M. (2013). An assessment of the current state of cybercrime scholarship. Deviant Behavior, 35(1), 20 40. Jang, H., Song, J., & Kim, R. (2014). Does the offline bully-victimization influence cyberbullying behavior among youths? Application of general strain theory. Computers in Human Behavior, 31, 85 93. Juvonen, J., & Gross, E. F. (2008). Extending the school grounds? Bullying experiences in cyberspace. Journal of School Health, 78(9), 496 505. Kendall, L. (1998). Meaning and identity in cyberspace: The performance of gender, class, and race online. Symbolic Interaction, 21(2), 129 153. Kitchin, R. (1998). Cyberspace: The world in the wires. New York, NY: Wiley. Ko¨nig, A., Gollwitzer, M., & Steffgen, G. (2010). Cyber-bullying as an act of revenge? Australian Journal of Guidance and Counselling, 20(02), 210 224. Kowalski, R. M., Giumetti, G. W., Schroeder, A. N., & Lattanner, M. R. (2014). Bullying in the digital age: A critical review and meta-analysis of cyber-bullying research among youth. Psychological Bulletin, 140(4), 1073 1137. Kowalski, R. M., Giumetti, G. W., Schroeder, A. N., & Reese, H. H. (2012). Cyber bullying among college students: Evidence from multiple domains of college life. Cutting-Edge Technologies in Higher Education, 5, 293 321. Kowalski, R. M., Limber, S. P., & Agatston, P. W. (2008). Cyber bullying: Bullying in the digital world. Malden, MA: Blackwell Publishing. Kyratzis, A. (2004). Talk and interaction among children and the co-construction of peer groups and peer culture. Annual Review of Anthropology, 33(1), 625 649. Langos, C. (2012). Cyber-bullying: the challenge to define. Cyberpsychology Behavior and Social Networking, 15(6), 285 289. Lapidot-Lefler, N., & Dolev-Cohen, M. (2014). Differences in social skills among cyberbullies, cybervictims, cyberbystanders, and those not involved in cyber-bullying. Journal of Child & Adolescent Behavior, 2(4), 1 9. Lash, S. (1999). Another modernity, a different rationality. Malden, MA: Blackwell Oxford. Le Bon, G. (1897). The crowd: A study of the popular mind. New York, NY: MacMillan. Lemert, C. C. (1997). Postmodernism is not what you think. Malden, MA: Blackwell. Li, Q. (2006). Cyber-bullying in schools: A research of gender differences. School Psychology International, 27(2), 157 170. Li, Q. (2007). New bottle but old wine: A research of cyber-bullying in schools. Computers in Human Behavior, 23(4), 1777 1791. Livingstone, S., & Haddon, L. (2008). Risky experiences for children online: Charting European research on children and the internet. Children & Society, 22(4), 314 323.

404

SHAHID ALVI ET AL.

MacKinnon, C. A. (1999). Points against postmodernism. Chicago-Kent Law Review, 75, 687 712. Madden, M., Lenhart, A., Duggan, M., Cortesi, S., & Gasser, U. (2013). Teens and technology 2013. Washington, DC: Pew Internet & American Life Project. Meltzer, B. N., Petras, J. W., & Reynolds, L. T. (1975). Symbolic interactionism: Genesis, varieties and criticism. London: Routledge & Kegan Paul. Merton, R. (1938). Social structure and anomie. American Sociological Review, 3, 672 682. Messner, S., & Rosenfeld, R. (2007). Crime and the American dream (4th ed.). Belmont, CA: Thomson Wadsworth. Mohr, J. (2011). From an affluent society to a happy society. Hamburg: Diplomica Verlag. Monks, C. P., & Coyne, I. (2011). Bullying in different contexts. Cambridge: Cambridge University Press. Moore, R., Guntupalli, N. T., & Lee, T. (2010). Parental regulation and online activities: Examining factors that influence a youth’s potential to become a victim of online harassment. International Journal of Cyber Criminology, 4(1 2), 1 2. Mouzelis, N. (2003). Sociological theory: What went wrong? Diagnosis and remedies. London: Routledge. Olweus, D. (2012). Cyber-bullying: An overrated phenomenon? European Journal of Developmental Psychology, 9(5), 520 538. Patchin, J. W., & Hinduja, S. (2006). Bullies move beyond the schoolyard: A preliminary look at cyber-bullying. Youth Violence and Juvenile Justice, 4(2), 148 169. Price, M. (2011). Alone in the crowd. American Psychological Association, 42(6), 26. Privitera, C., & Campbell, M. A. (2009). Cyber-bullying: The new face of workplace bullying? Cyberpsychology & Behavior, 12(4), 395 400. Rafferty, R., & Vander Ven, T. (2014). “I hate everything about you”: A qualitative examination of cyberbullying and on-line aggression in a college sample. Deviant Behavior, 35(5), 364 377. Raskauskas, J. (2009). Text-bullying: Associations with traditional bullying and depression among New Zealand adolescents. Journal of School Violence, 9(1), 74 97. Raskauskas, J., & Stoltz, A. D. (2007). Involvement in traditional and electronic bullying among adolescents. Developmental Psychology, 43(3), 564 575. Renati, R., Berrone, C., & Zanetti, M. A. (2012). Morally disengaged and unempathic: Do cyberbullies fit these definitions? An exploratory study. Cyberpsychology, Behavior, and Social Networking, 15(8), 391 398. Robson, C., & Witenberg, R. T. (2013). The influence of moral disengagement, morally based self-esteem, age, and gender on traditional bullying and cyber-bullying. Journal of School Violence, 12(2), 211 231. Sampson, R., Raudenbush, S., & Felton, E. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918 924. Schrock, A., & boyd, d. (2008). Online threats to youth: Solicitation, harassment, and problematic content. Internet safety technical task force, enhancing child safety and online technologies: Final report of the internet safety technical task force to the multi-state working group on social networking of state attorneys (pp.73 145). Shariff, S. (2005). Cyber-dilemmas in the new millennium: School obligations to provide student safety in a virtual school environment. McGill Journal of Education/Revue des sciences de L’e´ducation de McGill, 40(3), 457 477.

The Self and the ‘Selfie’

405

Shariff, S., & Hoff, D. L. (2007). Cyber bullying: Clarifying legal boundaries for school supervision in cyberspace. International Journal of Cyber Criminology, 1(1), 76 118. Simpson, B. (2013). Challenging childhood, challenging children: Children’s rights and sexting. Sexualities, 16(5 6), 690 709. Slonje, R., & Smith, P. K. (2008). Cyber-bullying: Another main type of bullying? Scandinavian Journal of Psychology, 49(2), 147 154. Smith, P. K., Mahdavi, J., Carvalho, M., Fisher, S., Russell, S., & Tippett, N. (2008). Cyberbullying: Its nature and impact in secondary school pupils. Journal of Child Psychology and Psychiatry, 49(4), 376 385. Steffgen, G., & Ko¨nig, A. (2009). Cyber bullying: The role of traditional bullying and empathy. In The good, the bad and the challenging. Conference proceedings. Luxembourg. Suler, J. (2004). The online disinhibition effect. Cyber Psychology and Behavior, 7, 321 326. Tippett, N., & Wolke, D. (2014). Socioeconomic status and bullying: A meta-analysis. American Journal of Public Health, 104(6), e48 e59. Tokunaga, R. S. (2010). Following you home from school: A critical review and synthesis of research on cyber-bullying victimization. Computers in Human Behavior, 26(3), 277 287. Tynes, B. M., Giang, M. T., Williams, D. R., & Thompson, G. N. (2008). Online racial discrimination and psychological adjustment among adolescents. Journal of Adolescent Health, 43(6), 565 569. Vandebosch, H., & Van Cleemput, K. (2008). Defining cyber-bullying: A qualitative research into the perceptions of youngsters. Cyberpsychology and Behavior, 11(4), 499 503. WeberShandwick, & Powelltate. (2014). Civility in America. Retrieved from www.webershandwick.com Weissbourd, D., & Jones, S. (2014). The children we mean to raise: The real messages adults are sending about values. Cambridge, MA: Harvard Graduate School of Education. Willett, R. (2008). Consumer citizens online: Structure, agency, and gender in online participation. In D. Buckingham (Ed.), Youth, identity, and digital media (pp. 49 70). Cambridge, MA: MIT Press. Williams, K. R., & Guerra, N. G. (2007). Prevalence and predictors of internet bullying. Journal of Adolescent Health, 41(6 Suppl. 1), S14 21. Wolak, J., Mitchell, K. J., & Finkelhor, D. (2007). Does online harassment constitute bullying? An exploration of online harassment by known peers and online-only contacts. Journal of Adolescent Health, 41(6 Suppl. 1), S51 S58. Wright, M. F. (2014). Longitudinal investigation of the associations between adolescents’ popularity and cyber social behaviors. Journal of School Violence, 13(3), 291 314. Ybarra, M. L. (2004). Linkages between depressive symptomatology and Internet harassment among young regular internet users. Cyberpsychology & Behavior, 7(2), 247 257. Ybarra, M. L., Diener-West, M., & Leaf, P. J. (2007). Examining the overlap in internet harassment and school bullying: Implications for school intervention. Journal of Adolescent Health, 41(6 Suppl. 1), S42 S50. Ybarra, M. L., Espelage, D. L., & Mitchell, K. J. (2007). The co-occurrence of internet harassment and unwanted sexual solicitation victimization and perpetration: Associations with psychosocial indicators. Journal of Adolescent Health, 41(6 Suppl. 1), S31 S41. Ybarra, M. L., & Mitchell, K. J. (2004a). Youth engaging in online harassment: Associations with caregiver-child relationships, internet use, and personal characteristics. Journal of Adolescence, 27(3), 319 336.

406

SHAHID ALVI ET AL.

Ybarra, M. L., & Mitchell, K. J. (2004b). Online aggressor/targets, aggressors, and targets: A comparison of associated youth characteristics. Journal of Child Psychology and Psychiatry, 45(7), 1308 1316. Young, J. (1992). Ten points of realism. In R. Matthews (Ed.), Rethinking criminology: The realist debate (pp. 24 68). London: Sage. Young, J. (1999). The exclusive society. London: Sage. Young, J. (2007). The vertigo of late modernity. London: Sage. Young, J. (2011). The criminological imagination. Cambridge: Polity Press.

SETTING THE PROBLEM OF MISSING CHILDREN IN THE REPUBLIC OF IRELAND AGAINST CONTEMPORARY INTERNATIONAL RESEARCH PERSPECTIVES Michael Rush ABSTRACT Purpose This paper sets a case study of missing children in the Republic of Ireland against a review of international research to explore broader understandings and responses to the problem. Methodology/approach The study begins by reviewing the literature on pioneering American initiatives dating back to the 1970s and more recent literature from Great Britain where a series of high-profile scandals involving sexual exploitation of teenage girls provoked a number of controversial inquiries into the police and social work professions. The present study was prompted by an evaluation of the 116 000 Missing Children Hotline which was introduced to Ireland in 2012 under the

Violence and Crime in the Family: Patterns, Causes, and Consequences Contemporary Perspectives in Family Research, Volume 9, 407 433 Copyright r 2015 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 1530-3535/doi:10.1108/S1530-353520150000009017

407

408

MICHAEL RUSH

auspices of the European Union (EU) Daphne III Programme by the Irish Society for the Prevention of Cruelty to Children (ISPCC). Findings The central conclusion emerging from analysis of the evidence is that Missing Children Hotlines remain rooted in representations of ‘stranger danger’ and disconnected from repeat runaway children who feature prominently in police reports from formal care settings or family homes and who are actively targeted by sexual predators and criminal gangs. The implications are that systemic change requires grounding in research strategies which combine police data with anthropological studies to give legitimacy to the voices of runway and sexually exploited children. Originality/value The study offers original international perspectives on missing children to epistemological research communities in the fields of social work, criminology and policing with recommendations that Missing Children and Runaway Safe-lines are targeted systemically at keeping runaway children, homeless children and at-risk-youth safe and off the streets. Keywords: Missing children hotlines; repeat runaway children; child sexual exploitation; social work; policing

INTRODUCTION The problem of missing children has become a highly politicised topic of contemporary European social policy debates. In addition, the issue has given rise to a new international academic sub-field focused on related topics including the prevention of children repeatedly going missing from care (Nemeth, 2015), gangs and missing children (Holmes, 2014, 2015), the exponential cost of policing the problem of missing children (Greene & Pakes, 2013) and the risks of missing children falling victim to Child Sexual Exploitation in groups and gangs (Berelowitz, Firmin, Edwards, & Gulyurtlu, 2013, Office of the Children’s Commissioner, 2012). One resource rich response to the problem was the introduction of a shared 24hour Missing Children Hotline number 116 000 by European Union member states. The present study was prompted by the introduction of the 116000 hotline to the Republic of Ireland by the Irish Society for the Prevention of Cruelty to Children (ISPCC) in 2012. That initiative was funded under the auspices of the European Union (EU) Daphne III

The Problem of Missing Children in Ireland

409

Programme which was aimed at preventing and combating violence against children, young people and women at risk. The study sets the context for the Irish Missing Children Hotline by reviewing international literature on contemporary ways of thinking about the problem of missing children. The central conclusion emerging from the review is that Missing Children Hotlines are rooted in conventional representations of ‘stranger danger’ and remain disconnected from the day to day problem of repeat runaway children from care who feature prominently in missing children reports to the police and are actively targeted by sexual predators and criminal gangs. The study recommends that 24-hour Missing Children Hotline answering services be re-conceptualised along the lines of the National Runaway Safe-lines in the USA and be re-targeted at repeat runaway children and their families. It also recommends that the service be actively promoted by the police, social work and social care professions to keep runaway children, homeless children and at-risk-youth safe and off the streets. In light of the pioneering role of American initiatives in this field, this study begins with a review of the epistemological and public policy literature from the USA which illustrates the importance of moving beyond fears of ‘stranger danger’ and towards the more clear-headed task of classifying different categories of missing children. The study then focuses on the British experience where a series of high-profile scandals involving the organised sexual exploitation of teenage girls by organised gangs of men provoked a number of inquiries into the failings of child protection agencies, most notably the police and social work professions (Coffey, 2014; Office of the Children’s Commissioner, 2012; Wishart, 2014). It then turns to Ireland and the contemporary initiatives of the European Union, and a final section concludes.

THE UNITED STATES OF AMERICA The USA pioneered the setting up of a 24-hour missing children hotline in 1984 under the management of the newly formed National Center for Missing and Exploited Children (NCMEC). The establishment of the NCMEC with an annual $3.3 million budget followed the introduction of the Missing Children Act (1982), during the presidency of Ronald Reagan. President Reagan’s proclamation of May 25th as ‘Missing Children Day’ called upon law enforcement and child protection agencies to protect

410

MICHAEL RUSH

children who had ‘lost their homes’ and urged American families to ‘protect their children’ against the ‘stark terror’ and ‘unique tragedy’ faced by ‘thousands of American families’, which was ‘that of a missing child’ (www.reagan.utexas.edu/archives/speeches/1984/51584). The NCMEC distinguished between six categories of missing children, namely: • • • • • •

non-family abduction (detaining a child for at least one hour); stereotypical kidnapping; family abduction; runaway/thrownaway (where a child is thrown out of his/her home); missing involuntary (whereabouts unknown) and missing benign explanation (none of the above).

NCMEC outreach teams work in communities to make child safety a daily concern (www.missingkids.com). In addition, the NCMEC has trained law enforcement officers, prosecutors and child care professionals. NCMEC publications include volumes such as A child is missing: Providing Support for families of missing children (Bowers, 2007), which examined the emotional and physical shocks for families and the special needs of siblings of runaway children. Moreover, the NCMEC and the missing children hotline played a vital role in reuniting families with missing children after Hurricane Katrina and other national or natural disasters (Broughton, Allen, Hanneman, & Pertrikin, 2006). An overview of national estimates of missing children in the USA by Sedlak, Finkelhor, Hammer and Schultz summarised findings from the Second National Incidence Studies of Missing, Abducted, Runaway, and Thrown-away Children (NISMART) for the Office of Juvenile Justice and Delinquency Prevention (2002). The NISMART study was undertaken under the auspices of the Missing Children’s Assistance Act (1984) and the first study was reported on by Finkelhor et al., (1996) under NISMART-1 (1990). The NISMART-2 estimates were based on four separate studies, namely, the National Household Survey of Adult Caretakers (NHSAC) involving interviews with 16,111 primary carers; the National Household Survey of Youth (NHSY), which was carried out with 5,015 randomly selected young people identified by respondents to the NHSAC; the Law Enforcement Study (LES), which involved a nationally representative mail survey sample of 400 counties, including 400 sheriff’s departments, and was focused on stereotypical kidnappings opened for investigation in 1997. The fourth source of data was the Juvenile Facilities Study (JFS), which involved

The Problem of Missing Children in Ireland

411

a nationally representative telephone interview survey sample of 74 facilities to determine the number of runaways in 1997 (Sedlak et al., 2002, p. 2). The NISMART-2 data still provides the most up to date picture and shows that the total number of missing children in 1999 was between 1,100,000 and 1,500,000 children of which between 645,000 and 949,000 had been reported missing. The missing children rate was between 16.1 and 21.4 per 100,000 if all missing children are counted and between 9.2 and 13.5 per 100,000 for reported missing children (Sedlak et al., 2002, p. 5). However, by the time the NISMART-2 study was published 99.8% of nonreported missing children had returned home and a major finding from the study was that the vast majority of missing children were runaways from care institutions or youth facilities (2002, p. 7). Most non-reported missing children were classified as ‘caretaker missing’ in the sense that they continued to be cared-for even though their exact location was unknown to their original carers, for example, they had run away to relatives’ or friends’ houses. The NISMART-2 study found that although there were a large number of ‘caretaker-missing’ children and a majority came to the attention of the police or missing children’s agencies, all but a small number were recovered very quickly (Sedlak et al., 2002, p. 9). A Federal Grant for NISMART-3 was announced by the Office of Juvenile Justice and Delinquency in 2010, with $1 million funding to measure the contemporary scale of the missing children in the USA. The grant was awarded to the Rockville Institute (Fernandes-Alcantara, 2014, p. 5). Meanwhile, data from the Federal Bureau of Investigation’s National Crime Information Center (NCIC) for 2014 showed that 466,949 entries from a total of 635,155 missing person’s entries were for children under the age of 18. However, the FBI records also revealed that when law enforcement agencies make an entry into the NCIC there was an optional field for completion called the Missing Person Circumstances (MPC) field. The MPC field was utilised only in 49.5% of entries in 2014 amounting to 314,523 cases. From these 314,523 cases, 96% (301,851) were coded as runaways, 7% (2,249) as abducted by a non-custodial parent and 0.1% (332) as abducted by a stranger. The FBI records for 2013 showed that the MPC field was utilised in 48.8% of 627, 911 records. Apart from the relative stability in these numbers over the two years, the main feature to observe is that the overwhelming majority of missing children entries into the NCIC concerned runaway children (www.fbi.gov/about-us/cjis/ncic/ncic-missingperson-and-unidentified-person-statistics-for-2014). Whilst recognising that ‘a child can be exploited regardless of whether he or she goes missing’, the problem of child sexual exploitation was

412

MICHAEL RUSH

inextricably linked by U.S. researchers to the problem of missing children (Fernandes-Alcantara, 2014, p. 8). The true incidence of child sexual exploitation was recognised as unknown because it often went undetected but the main three sources of data were NISMART-2, the National Survey of Children’s Exposure to Violence and data collected by NCMEC. In addition, the FBI’s National Incident-Based Reporting System illustrated through analysis of 418,000 victims between 2000 and 2001 that over half of crimes committed against juveniles involved sexual assaults (FernandesAlcantara, 2014, p. 44). Moreover the National Child Abuse and Neglect Data System (NCANDS) and the National Survey of Child and Adolescent Well-being (NSCAW) tracked the share of children who enter the foster care system as a result of sexual abuse and found that children in the foster care system were vulnerable to sex trafficking and were targeted by traffickers because they ‘have characteristics that make them vulnerable to manipulation and control’ (Fernandes-Alcantara, 2014, p. 45). The prevalence of sexual exploitation and victimisation of children in the child welfare system was enumerated from reports to the NCMEC in 2012 which revealed that 67% of sex trafficking victims were in foster care at the time of their victimisation (Fernandes-Alcantara, 2014, p. 45). The National Runaway Safe-line estimated that one-third of all teenage runaways would ‘be approached by a potential exploiter within 48 hours of leaving home’ (Kelly, 2014, p. 30). The National Runaway Safe-line is based in Chicago and was established in 1971 as the federally designated 24-hour volunteer answering service offering non-judgemental expertise to youths at risk and their families. It seeks to keep runaway children, homeless children and atrisk-youth safe and off the streets. Kelly (2014) highlighted the nature of sexual exploitation in gangs and groups in the USA. A study titled Children Missing from Care: the Law Enforcement Response emphasised the NECMEC’s focus on prevention and reported that the NCMEC was ‘proud to be a part of a larger effort addressing the issue of children missing from out-of-home care’ (Smith, Bunlak, Condon, & Reed, 2005, p. ix). That study was a joint response to the issue of children missing from out-of-home care by the NCMEC and the Child Welfare League of America (CWLA), which is the USA’s largest and oldest membership-based child welfare organisation. It concluded that the most effective way of preventing children going missing was to improve ‘the quality of out-of-home care services … especially in regard to stopping repeated runaway incidents’ (2005, p. 25). This type of sober reflection on prevention stood in stark contrast to urban and media myths in the USA about child-kidnappings by ‘trolls and

The Problem of Missing Children in Ireland

413

strangers’ which were warned against two decades earlier by Fritz and Altheide (1987). The Fritiz and Altheide study suggested that media panics about trolls and strangers were unhelpful in framing ways of understanding the missing children issue. They reasoned that the American public would be able to reassess the nature of the problem if they were provided with a better picture (1987, p. 488). Best’s work in the 1980s illustrated the scale of ‘stranger danger’ panics that gripped the USA fuelled by the ubiquitous visibility of missing children photographs on milk cartons, grocery bags, billboards and televised public service images (Best, 1988, p. 102). Best suggested that ordinary Americans encountered more reminders about missing children than about any other social problem in the USA (1988, p. 103). The panics in the 1980s followed a series of child abductions and murders. Best’s work (1987, 1988, 1990) focused on the rhetoric of policy-makers and NGOs and the way ‘atrocity tales’ might fuel more sophisticated and cynical ways of mobilising and maintaining public support for the issue of missing children (1988, p. 115). As a way of illustrating the more commonplace nature of the problem of missing children to law enforcement systems, a study on fatigue and improved performance among police officers in the USA found that 35% of overtime was for off duty court appearances, 11% was for filling in for officers off sick or on vacation, 20% was for late arrests or report writing and 9% was for missing children and crowd control of parades (Vila, Morrison, & Kenney, 2002, p. 12). Put simply, the study by Vila et al. illustrated that by 2002, cases of missing children were considered to be a significant cost of everyday police and a drain on budgets, rather than exceptional occurrences (2002). Previously, David Finkelhor, from the Crimes against Children Research Center who worked on the National Incidence Studies of Missing, Abducted, Runaway, and Throwaway children revealed ‘Five myths about missing children’ in an article for the Washington Post. Finkelhor reiterated that only 1% of missing children are abducted by strangers, and that between 1999 and 2011 there was a 31% fall in the number of missing person reports. He warned that instead of teaching children not to speak to strangers it would be better to teach children about inappropriate adult behaviour and how to avoid people they know who behave badly and inappropriately. Moreover, Finkelhor highlighted a systemic shortcoming of the child welfare system in the USA, which was that dealing with missing children cases by bringing them back a task primarily for police officers does nothing to address the circumstances which make children run away. The data from the USA clearly shows that the majority of missing children are runaways from

414

MICHAEL RUSH

family and care homes and from histories of ‘conflict and abuse that eat away at their mental health and well-being’ (Finkelhor, 2013). Finkelhor’s central message was that social policy responses to missing children needed to be less about law enforcement and the work of police officers in bringing them home and more about improving child welfare services, family therapy and child protection services targeted at runaway and repeat runaway children.

SUMMARY OF U.S. LITERATURE The American literature illustrates that the missing children problem in the USA was predominantly a problem involving runaway children and that the establishment of the National Runaway Safe-line in 1971 predated the establishment of the Missing Children Hotline in 1984 by over a decade. The targeting of the National Runaway Safe-line at keeping runaway children, homeless children and at-risk-youth safe and off the streets and the recommendations by Smith et al. to improve the quality of out-of-home care (2005, p. 25) combined to emphasise the complex welfare and child protection issues involved in managing the growing independence of young people who repeatedly go missing or run-away from home or from care institutions. One lesson for Ireland and other European Union member states from three decades of American experience is the importance of missing children reports to the police and national surveys of formal care facilities as sources of data for understanding the problem. However, the main lesson was that the problem of runaway children and the risks they face on the streets, including child sexual exploitation, were serious enough to merit dedicated national responses involving law enforcement agencies, child protection agencies and voluntary agencies.

GREAT BRITAIN A series of Child Sexual Exploitation (CSE) scandals in Great Britain has served to spotlight the failings of child protection services, administered by local authorities, and their contiguous police forces. Typically, these scandals involved the sexual abuse of hundreds of teenage girls over a number of years by organised gangs of men of Pakistani or Muslim heritage and shared a pattern of the branding of underage

The Problem of Missing Children in Ireland

415

victims by police and child protection agencies as sex-workers or deviants when they originally reported the crimes. The authorities under scrutiny included Rotherham Metropolitan Borough Council and South Yorkshire Police Force (2010), Rochdale County Council and Greater Manchester Metropolitan Police Force (2012) and Oxfordshire County Council and Thames Valley Police Force (2015). These scandals provoked the publication of the Joint All Party Parliamentary Group (APPG) Inquiry on Children Who Go Missing from Care (2012) and the Accelerated Report of the Office of the Children’s Commissioner’s inquiry into Child Sexual Exploitation in Gangs and Groups, with a special focus on children in care (2012). These reports identified three overarching issues. First, while there were no exact figures for how many children go missing the total was estimated at 100,000 per annum. Second, children in residential care were at particular risk of going missing and were vulnerable to sexual and other exploitation. Third, Local Safeguarding Children Boards had an important role to play in monitoring and interrogating data on children who go missing (2013). Local Safeguarding Children Boards were established under the Children Act (2004) which defined the safeguarding and welfare of children as a key duty of local authorities in Great Britain. Moreover, the scandals and ensuing reports revealed that although looked-after children in institutional care were particularly vulnerable when they went missing, the majority of children who went missing did so from their family home (2013) (House of Commons Education Committee 2014). The Chair of the Joint All Party Parliamentary Group (APPG) Inquiry on Children Who Go Missing from Care (2012) was Ann Coffey MP, a qualified social worker who was also the author of a report commissioned by the Greater Manchester Police Force entitled, Real Voices: child exploitation in Greater Manchester an independent report by Ann Coffey MP (2014). The Greater Manchester Police Force report expressed concern about the high number of looked after children who generated multiple missing reports, because these children were ‘particularly vulnerable to Child Sexual Exploitation’ (2014, p. 45). The report also noted, but without critical comment, that police forces introduced a new recording system in 2013 which separated reports of missing children into two categories of missing, consisting of cases which received an immediate response, and the lower risk category of absent, which didn’t receive an immediate response. Coffey also expressed strong concerns that some privately run children’s homes were ‘flouting government guidance’ on keeping local authorities informed about children’s nationwide movements

416

MICHAEL RUSH

from one local authority jurisdiction to another, especially when it was known that sexual predators targeted the movements of vulnerable children who repeatedly went missing from care (2014, p. 45). In November 2014, the Birmingham City Council published We Need to get it Right: A health check into the Council’s role in tackling child sexual exploitation (Wishart, 2014). The main findings of this report were that 83 children were victims of CSE or were at risk and 49 children were being offered early help (2014, p. 29). Significant and medium risk factors of CSE included periods of going missing overnight, entering or leaving cars driven by adults, unexplained amounts of money, multiple unknown callers, disclosure of sexual assault followed by withdrawal of the allegation, reports of having been sexually assaulted, and accident or hospital attendance due to alcohol consumption. Vulnerability factors included ‘sofa surfing’, a history of living in formal care, involvement in criminal or gang related activities, a history of engagement with child protection services, a family history involving abuse, sexual abuse, addiction or mental health issues. Only 3 of the 83 victims of CSE were boys, 1 was under 13, 57 were aged between 14 and 16 years old and 25 were 17 or older. Significantly, over 50% (44) of the victims of CSE were living in formal residential care and only 11 of the 44 children in care had Full Care Orders while 33 were being cared for on a voluntary basis (Wishart, 2014, p. 29). The front cover of the Birmingham City Council report was taken from the front cover of The Vanishing: why young people go missing through the eye of a young person, which was a short booklet published by the Children’s Society. The booklet was written in the words of ‘children in care who run away’ and consisted of simple drawings each accompanied by a short reason given by a child as to why he or she ran away. The reasons included ‘I just wanted to get high so I could forget’ and ‘I don’t want to be here’ and ‘I sold it, I smoked it and now I am it’ and ‘I was gang related it’s too late for me’ and the simple declaration ‘I was abused’. The booklet was designed to give residential children’s homes, foster carers, social workers and other professionals an insight in to what young people ‘say about why they go missing from care homes’. The booklet went on to explain that the reasons for running away were due to many factors, including the quality of their friendships, their feelings about school, or their family experiences prior to leaving home. The booklet also explained that running away wasn’t necessarily a reflection of poor quality care but that the disruption inherent in being ‘looked after’ by the formal care system could compound the problems faced by

The Problem of Missing Children in Ireland

417

young people often with devastating effects. The Vanishing booklet encouraged carers, social workers and other professionals within the formal care system to maximise young people’s contact with social workers and warned professionals working within the child protection and care systems that children who did not go missing were much less likely to be at risk than those children that do run away. A report by Greene and Hayden from the Centre for the Study of Missing Persons at the University of Portsmouth focused specifically on the location and characteristics of repeat reports to the police of missing people (Greene & Hayden, 2014). The study utilised the COMPACT data-base of a single anonymised police force, which held all reports of people reported missing for more than two hours. To identify repeat missing cases the study focused specifically on cases where the same address was linked to three cases or more, which resulted in 1,321 cases being examined involving 149 addresses (2014, p. 3). The results showed that private care homes were the most common locations of missing people reports (57.1%) with the next most common location being a family home address (16%) followed by mental health units (9.9%) and hospitals (7.7%) (2014, p. 3).The overall cost to the police force of responding to reports from the top 10 locations was estimated to be between £482,250 and £879,060 (2014, p. 3). Another study by Greene and Pakes (2013) estimated from a survey of 407 police officers and case assessments undertaken by 33 police officers that a realistic cost minimum per reported case was £1,325 and £2,415 for medium-term and medium risk cases (Greene & Pakes, 2013, p. 27). In addition, the study suggested that care homes and children’s homes should be made to pay £1,000 towards the police costs of investigating runaway cases from formal care institutions (Greene & Pakes, 2013, p. 34, Woodford, 2009). The Greene and Pakes study argued that focusing on the escalating costs of policing missing children could help to drive dramatic change in how missing children investigations were carried out. The study also raised questions about whether the agencies responsible for the formal care of children should contribute to the cost of policing and returning repeat runaway children (2013, p. 34). Put simply, British epistemology began to highlight the duty for improved management of the growing independence of young people in formal care settings and to put a strong focus on runaway prevention and cost reduction. Moreover, the Greene and Hayden report recommended that the Quality Care Commission should request police summaries of missing people reports from establishments under review and that placement strategies for children in care should include polices

418

MICHAEL RUSH

for information exchange between care homes, local authorities and the police (2014, p. 4). Rees (2011, p. 5) explained that the first report on runaways was published by the Children’s Society charity in 1999, which estimated 77,000 children aged under 16 ran away for the first time and that overall there were 129,000 incidents. Rees reported that the Children’s Society, with the support of the Department of Children, Schools and Families, conducted a survey of local authorities and police forces (Evans, HoughtonBrown, & Rees, 2007) which demonstrated ‘patchy implementation of government guidance’. This account prompted the government to produce a Young Runaways Action Plan in 2008 (HM Government, 2008). The Running Away 3 report found that 36% of runaways in 2011 had runway before their 13th birthday, which was a 6% increase on the number of first time runaways under 13 (30%) in 2005 (2011, p. 12). Overall, however the reports showed that since 1999, when data-collection began, the runaway problem has remained at a relatively consistent level.

SUMMARY The review of British literature spotlighted the links between children reported missing, child sexual exploitation and the problem of repeat runaways from formal care and family homes. In particular Birmingham City Council illustrated the importance of data-sharing between social work agencies and Metropolitan Police Forces, since their analysis which showed that over 50% of children who were identified as victims of CSE or at risk of CSE were living in formal residential care. The literature also illustrated that protecting the anonymity of individual children did not act as a barrier to the scrutiny of the care settings of missing children by researchers, politicians and the general public. The overarching lesson to be learned from the British literature is that the systematic grooming of school-age girls involving drugs, alcohol and gifts followed by forced sex with multiple men had become part of everyday life for some young girls. This pattern was so prevalent that when girls were openly harassed by groups of men in public ‘nobody came up and said what is happening’ (Coffey, 2014, p. 10). The exposure of the prevalence of grooming and child sexual exploitation which shocked the British public, politicians and researchers also drew attention to the social problem of runaway and missing children and the shortcomings of child protection and out-ofhome care systems.

The Problem of Missing Children in Ireland

419

THE EUROPEAN UNION The consensus among EU member state on the need for a coordinated Missing Children Hotline 116 000 response was mainly driven by civic advocacy campaigns and in particular by the coming together of Missing Children Europe (MCE) as an EU-funded umbrella NGO in 2008. The MCE umbrella group estimated that 250,000 children were reported missing every year, or as the slogan goes ‘one child every two minutes’ (Rego, 2013, p. 1). Moreover, the MCE umbrella group emphasised the link between the exponential growth of missing children reports and corresponding growth of Child Sexual Exploitation. On foot of this connection the MCE morphed into the European federation for missing and sexually exploited children. The Missing Children Europe (MCE) alliance is the leading organisation in the European Union concerning missing and runaway children. In 2005, the MCE established a common European telephone number for missing children and so far 23 European Union member states have introduced the 116 000 hotline. The MCE annual report for 2012 set out four short-term priorities, namely, to ensure that all hotlines provide comparable support to families calling the hotline, to further extend the 116 000 hotline to non-participating EU member states, to launch annual campaigns and promote International Missing Children’s Day, and to promote a network approach to funding in the EU for missing children hotlines (Moralis, 2012, p. 24). A recent study entitled Missing Children in the European Union, Mapping, data-collection and statistics identified issues concerning the definition of missing children with only Ireland, Estonia and Hungary actually having a legal definition of the category of missing person (Cancedda, Day, Dimitrova, & Gosset, 2013, p. 2). Five specific categories of missing children were defined by Missing Children Europe: (i) runaways; (ii) abduction by third person; (iii) international parental abduction; (iv) missing unaccompanied migrant minors, and (v) lost, injured or otherwise missing children. However, differentiation between third person abductions and international parental abductions was not always applied at the national level (Cancedda et al., 2013, p. 3). Some countries had additional categories such as missing children with mental health issues. The EU study received data from 25 countries (two excluded countries were Austria, which deletes data once cases are solved, and Sweden which did not distinguish between missing adults and children). The data revealed inconsistences and gaps. For example three countries with similar populations provided very different figures with 50,326 reported cases in France, 5,513 in Italy and 96,341 in the United Kingdom

420

MICHAEL RUSH

(the latter cases arose from just two-thirds of the United Kingdom’s police forces, suggesting a total of somewhere in the region of 150,000 cases for the United Kingdom). Potential explanations for differences included (a) the nature of irregular immigrant populations and transit migration, (b) the different nature of care organisations from which children runaway, (c) the effectiveness of preventative measures and (d) different procedures for reporting runaways from care institutions. For example in Ireland all runaways from care are reported as missing children whereas in the United Kingdom about one third of all missing children from care reports were classified as the lower risk of ‘absent’ rather than ‘missing’. European wide data revealed that Hungary with 852 cases per 100,000 children and Ireland with 557 cases per 100,000 have the highest reported rates of missing children (Cancedda et al., 2013, p. 4). Two distinct sources of data were calls to hotlines and police reports. Where both sets of data were available it was possible to work out a ratio between hotline calls received and the number of cases reported to the police. That ratio provided an ‘idea of the current ability of hotlines to intercept a significant part of the missing children phenomenon’ (Cancedda et al., 2013, p. 39). However, calls received and answered by the NGOs were not correlated directly with missing children reports to the police. The EU study recommended that ‘a coordinated and multi-agency effort’ was needed between ‘116 000 hotline operators, police, judicial services, child welfare associations, NGOs and citizens’ (Cancedda et al., 2013, p. 25). Recommendations on raising public awareness included targeted publicity campaigns aimed at social workers and care workers and establishing protocols for reporting by care institutions. Making annual statistics publicly available was also recommended. Ireland did not provide data on runaways from care institutions, among those countries that did, the countries with high figures for 2012 included the Czech Republic (3996), France, (30,794), Italy (2,362), Finland, (3,080), the United Kingdom (1,490). The overall missing children data showed 6,615 children reported missing in Ireland (population 4.6 million) in 2012. Countries with similarly small populations included Denmark, (pop 5.6 million) with 1,039 children reported missing in 2012 and Estonia (pop, 1.3 million) with 1,038 children reported missing. Indeed, Ireland stood out alongside Hungary (pop, 9.9 Million) 14, 419 children reported missing in 2011, as one of the two countries with the highest number of children per 100,000 population who were reported missing (Cancedda et al., 2013, p. 71). For example, the number of children reported missing in Ireland (6,615) was 600% higher than the number of children reported missing in Denmark (1,039),

The Problem of Missing Children in Ireland

421

despite the fact that Denmark (pop 5.6 million) is a more populous country than Ireland (pop 4.6 million).

IRELAND: CASE STUDY The evaluation case study of the Irish Missing Children Hotline 116000 was carried out by the present author between March and December 2014 and involved examination of Client Relationship Management (CRM) data and other organisational records kept the Irish Society for the Prevention of Cruelty to Children (ISPCC). It also involved interviews with ISPCC staff and volunteers, officers from the An Garda Sı´ ocha´na (police) missing persons unit, the Tusla family and children’s service agency with statutory responsibility for child protection and children in care, a representative from Missing Children Europe, a representative from the Department of Children and Youth Affairs and a representative from the Directorate-General for Justice in the European Commission. The starting point for the Irish study occurred when the Irish Government under the auspices of a Joint Oireachtas (parliamentary) Committee held a public hearing in 2012 on the issue of missing persons. The Committee recommended that ‘a National Missing Persons Day be established in Ireland and that a single national telephone helpline contact be set up’ (Joint Committee on Justice, Defence and Equality, 2012, p. 4). It reported that over 40,500 missing persons were reported to the Garda Sı´ ocha´na (police) during the five-year period 2007 2012. The report explained that missing persons were categorised as: missing children, missing people with mental health conditions, and adults who wished to start a new life and did not want to be found. It also found that missing children were always classified as high risk (2012, p. 7). The Joint Oireachtas Committee also recommended that a national forum consisting of Gardaı´ , relevant State bodies and recognised NGOs that work in the area of missing persons should be established to ensure that a viable, agreed and coordinated response system is put in place (2012, p. 7). Mr John O’Mahoney, the assistant commissioner of An Garda Sı´ ocha´na, explained to the Joint Oireachtas Committee that 40,500 people were reported missing between 2007 and 2012 and that ‘the vast majority related to children who go missing from care’ (2012, p. 5). When asked to put the Irish statistics in a comparative context the Assistant Commissioner explained that Ireland had very strict procedures generally and particularly

422

MICHAEL RUSH

strict procedures for reporting missing persons from care and young persons who entered the State unaccompanied, which might differ from those in operation in other jurisdictions (2012, p. 5). Moreover, the Assistant Commissioner explained that children in the 16 to 17-year-old age category who were repeatedly reported missing were understood to be ‘testing the waters’ especially those in care who are subject to a ‘curfew’. However, the Assistant Commissioner reported that while one the one hand repeat runaways who were understood to be testing ‘the boundaries’ generally turned up ‘safe and well’, on the other hand they were also ‘very likely’ to be dealing with people who often take advantage of such children (2012, p. 11). The Tusla representative who was interviewed for the ISPCC missing children hotline evaluation explained that 92% of the 6,500 children in formal care in Ireland were in stable foster care settings leaving about 450 in residential care settings and that it was this latter group who were repeatedly subject to missing children reports to the Gardaı´ . Official Tusla statistics show that as of August 2014 there were 6,489 children in care in Ireland (Table 1) which reflects growth from 5,727 in 2010 and 6,282 in 2012 (Tusla, 2014). In response to a question from Senator Martin Conway, the Assistant Commissioner explained to the Joint Oireachtas Committee in 2012 that the implementation of the pan-European 116000 missing children hotline to Ireland had been awarded to the Irish Society for Prevention of Cruelty to Children (2012, p. 21). The ISPCC was established in 1889 and provided a range of child welfare services including Childline, a well-established and high-profile 24-hour non-directive listening service for children. The ISPCC secured funding under the European Union Daphne III Programme in February 2012, which stated that the objective of the 116000 hotline was to take calls from the general public, parents, professionals and missing children. A co-funding application to the Department of Children and Table 1.

Children in Care: Tusla Monthly Management Data Activity Report.

Type of Care Foster care general Foster care relative Out of state care (UK) Secure care (only 17 beds) Residential care Other care Total

Number of Children 4,133 1,893 6 17 324 116 6,489

The Problem of Missing Children in Ireland

423

Youth Affairs explained that the scope of 116000 service covered taking calls, providing a confidential text support by sending automated generic SMS to missing children, managing data for the purposes of statutory requirements and to gain ‘a clearer picture of how the issue of missing children is manifested in Ireland’ (ISPCC, 2012, p. 4). The evaluation of the service revealed that responding to the existing ISPCC Childline service and the new Missing Children Hotline 116000 service involved two very distinct skills sets for volunteers and two very distinct services for children, families and the general public. This was because the Childline service offered a ‘non-directive answering service’ to children while on the other hand the missing children hotline offered a ‘directive answering service’ to children and adults who sought help. ISPCC volunteers and call answering facilitators explained that they worked in shifts and that they found the work was both challenging and rewarding. They explained that the times of the day to expect a high volume of Childline calls was between 8.00am and 9.30am before school, at lunch time, in the afternoon after school, at the weekends when children are getting ready to go out with friends, on school trips. Calls also came from children who were off-sick from school or out-of-school for other reasons and from children in residential care or foster care. The call-answerers explained that they often had regular callers from both family homes and residential or foster care settings and that calls from formal care settings made up about 10% of all calls. The volunteers explained that although calls from groups might be sexualised and considered ‘prank calls’, that it was not uncommon for children involved in group or prank calls to call back later and present real concerns. The volunteers also identified five categories of callers to Childline who could be considered missing, namely: homeless children; abandoned children; children looking for shelter, runaways and other homeless children. They explained that calls from missing children represented about 5 10% of all calls received. A major problem spoken about by child callers, according to the front-line call answerers, was the children’s complaint ‘that we don’t give advice’. However, the ISPCC volunteers prided themselves in being ‘non-directive’ and had a strong belief in the non-directive nature of the Childline service. The ISPCC had a modern CRM computerised data management system. This was used to record the necessary details of calls and the number of calls taken. The CRM data was collated by the ISPCC and reported to the Department of Children and Youth Affairs on a quarterly basis. Table 2 illustrates the total numbers of calls received during the period January 2013 December 2013.

424

MICHAEL RUSH

Table 2. Missing Children Hotline Calls January 2013 December 2013. Total Calls October December 2013 116 000 Hotline calls received 116 000 Hotline calls answered Calls received to the Childline service regarding missing children Average call waiting time for calls answered Average call waiting time for calls abandoned Calls with Gardaı´ regarding broadcast alerts of missing children

Total Calls 2013

131 80 0

523 305 126

2

34 seconds 50 seconds 12

In 2014 the ISPCC began encouraging volunteers to write up short case studies describing level 3 type calls they had answered. Level 3 calls were those where the caller engages in conversation with the call facilitator or volunteer. The three case studies below were provided as examples. It is significant that one of the three examples is from a child who has run away from residential care and another is from a child with a social worker.

Volunteer Case Study Profile 1: Homelessness Sarah (16 years) said that she had been kicked out of home. She was crying throughout the call. She told Childline that she had been living with her mom and dad and brother (25 years). Sarah said that she was beaten up by her brother while her parents stood by. Sarah was given information about her rights and options were explored with Sarah who said that she didn’t want to go to the Gardai because she had taken a lot of alcohol that morning. She said she wanted to get in touch with her social worker because she doesn’t know where to go or what to do. All her friends have gone off and just left her. She feels upset and feels that everyone is against her. She tried to ring her mom to get the social worker’s number but her mom kept hanging up. Caller continued crying and then hung up.

Volunteer Case Study Profile 2: Runaway Caller (12 years) said she is a little bit down. Her mam was giving out to her more than ever last night. She was screaming at her for no reason. She

The Problem of Missing Children in Ireland

425

never did this before. She even punched her in the face. Caller stated that she ran away and spent the night at a community centre leader’s house. This morning the leader said she’d have to call her mam, so she left. She’s spent the day in a park and isn’t sure where she is. She called her cousin who told her to come over to her house later. Caller is worried about going there as her mam might be there and she’s scared of her mam now. She’s afraid to go home as well.

Volunteer Case Study Profile 3: Runaway Leah (13 years) told Childline that she has run away from care home. She said that she stayed in a place last night with four boys aged 19. They offered her drugs and alcohol. She said that she took alcohol but no drugs. She said that she does not want to go back to the care home. Leah said that she will stay out again tonight. Childline gave Leah information about her rights and expressed concern about Leah’s safety. Call ended abruptly. Later on Leah called back to say she had thought about things and was going back to her care home. She said thanks and ended call. In addition to the three examples of Childline cases, the ISPCC provided 12 text boxes from level 3 Missing Children Hotline 116000 calls. Text boxes are filled in after every call by volunteers as a record of the call. Each text box is about one paragraph or about 100 words. The information provided required to be anonymised and is presented in Table 3. The text boxes from the 116000 hotline were all from level 3 calls. A Level 3 call signifies that a discussion took place between the caller and the call answerer, as opposed to a hang up call or a call where only one person talks and the other listens. The 12 level 3 text boxes illustrated that two of the calls involved people from other EU jurisdictions. One from a missing mother and one from an ex ISPCC volunteer concerning a teenage runaway. Both seem to have been calls from or about highly publicised cases. Other calls involved within-neighbourhood family abductions, homelessness and sexual abuse, child sexual exploitation (CSE), runaways and selfharm, teenage callers worried about friends whose parents are problem drinkers, adult callers looking for information and parents of missing children looking for emotional support. There was no evidence provided to suggest that these calls generated further responses from the ISPCC and they are provided here to illustrate the nature of calls and the lesson that that children and adults who were already known to the statutory social

426

MICHAEL RUSH

Table 3. Missing Children 116000 Text Boxes. Time and Date

Gender

14:04 July 2014 16:05 August 2014

Female

21:27 August 2014

Female adult

20:02 September 2014

Female adult

October 2014

Female

22:08 February 2014 11:01 February 2014 17:02 November 2014

Female

16:35 May 2014

Female

16:51 May 2014

Male adult

Female adult

Female Male

Nature of Call Caller homeless and previously sexually abused but was reluctant to talk to the police Caller whose family have abducted her son to protect him is afraid of her physically abusive boyfriend. Caller says she and birth father of child will contact local Garda station Ex ISPCC volunteer witnessed inappropriate sexual behaviour between an older man and a young girl who bore a resemblance to a girl reported missing from an EU jurisdiction. Caller said she would phone Garda Caller and her children had been reported missing from an EU jurisdiction and appeared on the news but wished to say she was on holiday and refused to give her name, didn’t want to contact Garda Child having suicidal thoughts, self-harming and thinking about running away from parental home, won’t say why, won’t go to counselling Same as above repeat caller 6 months earlier, has already seen counsellor didn’t help Caller wishes to know at what age she should show her daughter the ISPCC running away video Caller wonders whether this number is the fostering helpline ISPCC offered him Health board number and ISPCC support line Caller worried about her friend who is underfed and thinking about running away because parents have alcoholism problems Caller’s daughter is missing and he is calling for emotional support to get through the next few days

services used the ISPCC services for support, which in the case of Childline was ‘non-directional’ support and in the case of the missing children hotline 116000 was a ‘directional’ answering service. The cases studies and textboxes also highlighted that callers wished to remain anonymous, which does not prevent the ISPCC from reporting such calls to the statutory agencies but does leave the organisation open to charges of over-reporting without adequate identifying information. An overarching lesson to be learned from the data was that instead of being comprehended as a source of new child protection cases for the statutory child protection services, the ISPCC might be better understood as a provider of confidential and

The Problem of Missing Children in Ireland

427

universal call-answering services, which were frequently accessed by children known to the statutory social services. The Health Service Executive and Tusla Child and Family Services agency are responsible for a child missing from care, and liaise with Gardaı´ (police) about procedure. The Irish Foster Care Association (IFCA) information booklet on Safe Care in Foster Care explained that a Health Service Executive Absence Management Plan (AMP) is compiled for every child placed into formal care to assess the risk of a missing incident occurring, as part of an overall Placement Plan (2013, p. 20). The IFCA explained that ‘a child is considered missing when his/her whereabouts are unknown and the AMP indicates concern for the child’s safety’ (2013, p. 20). A missing incidence report includes the completion of a Missing Child from Care Report Form with a photograph of the child and a notification of each report to the relevant District Officer (Police Superintendent) and Social Worker. The annual HSE Review of Adequacy explained that the reasons children runaway were varied and complex and could not be understood in isolation from their family circumstances and their experiences of care (2012, p. 46). The Garda Missing persons unit was established in 1982 and is staffed by a Garda Sergeant and a civilian. The functions of the office are to maintain accurate and up to date records on missing persons within Ireland, assist District Officers in their management of High Risk Missing Person cases, develop missing person policy and procedures, provide assistance in training and policy implementation, assist in the identification of human remains, upload missing persons to the various internal and external websites, assist with international enquiries from/to Interpol; finally, to assist operational management of the Child Rescue Ireland (CRI) Alert policy. The data provided on missing persons between 2003 and 2011 show a significant and steady increase from 3,987 in 2003 to 8,511 in 2011, which is an increase of well over 100% in 7 years and suggests that missing person incidents in Ireland are an exponentially growing problem for An Garda Sı´ ocha´na. Overall between 2003 and 2011 there were 62,426 reported incidents and 335 outstanding cases. Statistics supplied by the Gardai isolate missing children from the overall missing people data and are presented in Table 4. The missing children data in Table 4 illustrated that missing children reports represented over 75% of all missing people reports and that children were disproportionately represented in Ireland’s rapidly increasing problem of missing people reports. In essence, the exponential number of missing people reports in Ireland consists largely of missing children

428

MICHAEL RUSH

Table 4.

Missing Children Reports.

Year

Missing Children Reports

2012 2011 2010 2009 2008

6,615 6,388 6,244 5,614 5,841

reports. The Missing Children in the European Union: Mapping, Data collection and statistics report stated in relation to the Irish case that all missing child cases are reported to the national Missing People’s Bureau (MPB), which is located within the national police (Garda Sı´ ocha´na). The MPB places all information about missing person incidents on its PULSE information system and that the PULSE reports were reviewed on ‘a daily basis, with relevant information communicated directly back to local police units’ (Cancedda et al., 2013, p. 22). The Children Missing from Care booklet included details of A Joint Protocol between An Garda Sı´ocha´na and the Health Service Executive Child and Family Services and set out procedures for children who runaway ‘five, ten, fifteen or twenty times in a thirty day period’ (2012, p. 13). The protocol also drew attention to a strong correlation between the frequency with which a child goes missing and his or her ‘risk of experiencing harm’ (2012, p. 6). The Children Missing from Care protocol also established that where it was apparent on the return of a child from a period of absence that the child had been victim of crime or exposed to risk then An Garda Sı´ ocha´na were called to attend both for child protection protocols and for recording of evidence (2012, p. 12). Two representatives from An Garda Sı´ ocha´na, at the rank of Superintendent and Sergeant, who were interviewed for the present study explained that a Strategic High Level Group involved Tusla and that the information recorded on the Pulse data system was more comprehensive than any other records. The Gardaı´ referred to the tragic case of Daniel McAnaspie who was murdered in June 2013 while missing from care, and expressed concern that children missing from care were vulnerable to becoming being victims of crime or to becoming involved in crime. Moreover, the Gardaı´ expressed concern that many runaways from care who were reported missing were the subjects of voluntary care orders (as was the case in Birmingham see Wishart, 2014). The nature of Ireland’s ‘contracted-in’ residential care system, which consists of small-scale homes with small numbers of children was

The Problem of Missing Children in Ireland

429

presented as a barrier to releasing data to the European Union study because there was ‘concern that releasing figures could potentially lead to the identification of individuals, especially when the number of children in care and the number of institutions is small’ (Cancedda et al., 2013, p. 42). The Health Information and Quality Authority (HIQA) is the agency in Ireland which monitors and regulates the standards of formal care. A HIQA report for the Gleann Alain special care unit in Cork for 2011, which accommodated seven girls caused controversy when it was reported in the Irish print and television media that there were 25 incidents of girls ‘absconding’ for periods of up to six days in the previous year (Hough, 2012). The HIQA report on the Gleann Alain special care unit noted ‘an extended period of absence for three children’ and that ‘records were incomplete and inconsistent’ (2011, p. 12). Overall, the HIQA report concluded that ‘the delivery of a good standard of care to children required improvement’ (2011, p. 16). Another HIQA report published in January 2015 on an unnamed care home catering for five children of mixed gender between the ages of the 13 and 18 stated that children regularly went missing and returned ‘under the influence of illegal substances’ (HIQA/694, 2015, p. 16). This was reported in the press under the headline ‘Teens in state-run home were leaving regularly to take drugs’ (Hennessy, 2015). The overarching lesson from the Irish case study was over 6,000 missing children reports annually were generated mainly by a group of repeat runaway children from residential care numbering less than 450 children in total. It was also clear that the ISPCC Childline 24-hour answering service and the Missing Children Hotline 116000 service were being utilised by children in care, yet not specifically targeted at children in care. More, the incidence of school-age girls reported missing featured prominently in HIQA reports on private care homes. Although, children were returned safe it was highlighted that children were at risk crime and exploitation while missing. More, they were frequently missing for days rather than hours and were often returned under the influence of drugs or alcohol. The ISPCC Missing Children Hotline service was successfully implemented in accordance with requirements of the Department of Children and Youth Affairs. However, the recommendation of a national forum consisting of Gardaı´ , relevant State bodies and recognised NGOs was yet to be put in place to ensure a viable, agreed and co-ordinated response system to address the risks children face while they are missing. Without such a system is in place, it seems child protection and policing systems in the Republic of Ireland may yet be faced with the type of grooming and child

430

MICHAEL RUSH

sexual exploitation disclosures that have already shocked British society, and that meanwhile repeat runaway children face unidentified risks.

CONCLUDING DISCUSSION The Irish case illustrates that the high level of missing children reports is indicative that reporting mechanisms in Ireland are fit-for-purpose when compared to other European Union member states and that the Gardai are vigilant at keeping ‘a full record of all actions taken in respect of a missing child from care investigation’ (2012, p. 11). The case study also illustrated that a key element of the Irish protocols was that Tusla Children and Family Services and An Garda Sı´ ocha´na ensured full compliance, even in the case of nationwide alert, with legal requirements not to publicly disclose when a child was reported missing from formal care (see section 31, Child Care Act, 1991) (2012, p. 9). Regrettably, these protocols act as a barrier to analysing and making public some of the most comprehensive member state data on missing children in the European Union, and as a barrier to gaining valuable lessons about the missing children problem. Meanwhile, HIQA reports on Irish residential care homes for children revealed serious systemic deficiencies but were prevented from publicly identifying the care providers. Alternatively, official reports in Great Britain by Local Authorities and Metropolitan Police services were able to illustrate the importance of listening to the ‘voice of the child’ as a key systemic qualitative informant and provide valuable detailed quantitative data on correlations between missing children reports from care and incidents or risks of child sexual exploitation. In a similar fashion the agencies and epistemological communities in the USA placed a strong emphasis on data recording and publishing for critical and transparent analysis of the problems of missing and runaway children. Contemporary reports from Great Britain including the observation by Ann Coffey MP that child sexual exploitation had become a ‘social norm’ in Greater Manchester (2012) combined with reports from the USA that runaways will be approached by a potential exploiter within 48 hours of leaving home (Kelly, 2014, p. 30) to illustrate that child protection agencies and formal care systems are faced with ever increasing challenges. This lesson is not lost on the Missing Children Europe (MCE) lobby group which represents itself as the European federation for missing and sexually exploited children (http://missingchildreneurope.eu).

The Problem of Missing Children in Ireland

431

However, the American model of supplementing the establishment in 1971 of the National Runaway Safe-line with the establishment of a National Missing Children Hotline in 1984 seems be unnecessarily duplicative for the European Union and serves to obscure the reality that most missing children are runaways. The introduction of the National Missing Children Hotline coincided with moral panics about stranger danger and child abductions in the USA when the problem of missing children achieved extraordinary visibility (Best, 1988, p. 102). However, the call in the USA nearly 30 years ago by Fritz and Altheide for a rejection of the ‘trolls and strangers’ type of moral panics and for a more sober social construction of the missing children issue still bears resonance today (1987, p. 488). Locating the European Union wide 116 000 within a reassessment of the social construction of the missing children issue requires recognition that vast majority of missing children reports to the police in Europe and in the USA concern children who go missing or runaway from residential care. These children are usually the subject of voluntary care orders indicating that their parents are looking for support from state. However, the central hypothesis of this paper was that police reports of missing children across the USA and the European Union serve to illustrate that contemporary residential care systems (for challenging children) are manifestly unfit for the purpose unless more supports are offered to repeat runaway children, and the Irish case served to support this hypothesis. Therefore, if there is a case to be made for missing children hotline services it is within broader directive and supportive child welfare systems that bring highly specialised child development resources to incident management plans, action plans, and individual plans to manage children’s growing independence in care settings. However, systemic change and policy change requires grounding in user-led anthropological research which gives primacy and legitimacy to the voices of runway and sexually exploited children. If the Missing Children Hotlines 116000 are to play a part across EU member states they would be better modelled on the National Runaway Safe-line in the USA and renamed as Missing Children and Runaway Safe-lines and systemically targeted at ‘keeping runaway children, homeless children and at-risk-youth safe and off the streets’.1

NOTE 1. This is the slogan of the National Runaway Safeline http://www.1800runaway. org/

432

MICHAEL RUSH

ACKNOWLEDGEMENTS Acknowledgements to Professor Tony Fahey, School of Social Policy, Social Work and Social Justice, University College Dublin and to Dr Nicola Carr, School of Sociology, Social Policy and Social Work, Queens University Belfast for comments and proof-reading

REFERENCES Berelowitz, S., Firmin, C., Edwards, G., & Gulyurtlu, S. (2013). If only someone had listened, office of the children’s commissioner’s inquiry into child sexual exploitation in gangs and groups. London: Office of the Children’s Commissioner. Best, J. (1987). Rhetoric in claims-making: Constructing the missing children problem. Social Problems, 34(2), 101 121. Best, J. (1988). Missing children: Misleading statistics. Public Interest, 92, 84 92. Best, J. (1990). Threatened children: Rhetoric and concern about child victims. Chicago, IL: University of Chicago Press. Bowers, D. T. (2007). A child is missing: Providing support for families of missing children. Alexandria, VA: NCEMC. Broughton, D. D., Allen, E. E., Hanneman, R. E., & Pertrikin, J. E. (2006). Reuniting fractured families after a disaster: The role of the national center for missing and exploited children. Pediatrics, 117, 5. Cancedda, A., Day, L., Dimitrova, D., & Gosset, M. (2013). Missing children in the European Union: Mapping, data collection and statistics. Rotterdam: ECORYS Nederland BV. Child Care Act. (1991). Irish Statute Book, Dublin. Coffey, A. (2014). Real voices: Child exploitation in Greater Manchester An independent report by Ann Coffey MP. Manchester: Greater Manchester Police. Evans, K., Houghton-Brown, M., & Rees, G. (2007). Stepping up: The future of runaways services. London: The Children’s Society. Fernandes-Alcantara, A. L. (2014). Vulnerable youth: Background and policies. Washington, DC: Congressional Research Service. Finkelhor, D. (2013). Five myths about missing children. Washington Post. Retrieved from http://www.washingtonpost.com/opinions/five-myths-about-missing-children/2013/05/ 10/efee398c-b8b4-11e2-aa9e-a02b765ff0ea_story.html. Accessed on May 10, 2015. Finkelhor, D., Asdiigian, N., & Hotaling, G. (1996). New categories of missing children: Injured, lost, delinquent, and victims of caregiver mix-ups. Child Welfare, 75(4), 291 310. Fritz, N. J., & Altheide (1987). The mass media and the social construction of the missing children problem. The Sociological Quarterly, 28(4), 473 492. Greene, K. S., & Hayden, C. (2014). Repeat reports to the police of missing persons. Portsmouth: University of Portsmouth. Greene, K. S., & Pakes, F. (2013). The cost of missing person investigations: Implications for current debates. Policing, 8(1), 27 34. Hennessy, M. (2015, January 12). Teens in state-run home were leaving regularly to take drugs. The Journal.ie. Retrieved from http://www.thejournal.ie/state-run-home-teensdrugs-1877618-Jan2015/

The Problem of Missing Children in Ireland

433

HIQA. (2011). Gleann Alainn special care unit. Inspection report ID number 506. Dublin: HIQA. HIQA. (2015). Children’s residential centre report ID 694 for centre 38. Dublin: HIQA. HM Government. (2008). Young runaways action plan. London: HMSO. Holmes, L. (2014). When the search is over: Reconnecting missing children and adults. London: Oak Foundation and Missing People. Holmes, L. (2015). Gangs and missing children. Paper presented to the 2nd international academic conference on missing children and adults, July 8 10, Vrije Universiteit Brussel, Brussels. Hough, J. (2012). Care unit staff powerless to stop runaways. Irish Examiner, April 23, 2012. (p. 3) House of Commons Education Committee. (2014). Residential children’s homes. Sixth Report of Session 2013 2014. The Stationary Office, London. Health Service Executive. (2012). HSE review of adequacy 2012. Dublin: HSE. Irish Society for the Prevention of Cruelty to Children. (2012). Application for funding to the department of children youth and family affairs for the 116000 missing children Hotline. Dublin: ISPCC. Joint Committee on Justice, Defence and Equality. (2012). Report on hearings in relation to missing persons. Dublin: Government Publications. Kelly, A. (2014, November 16). I carried his name on my body for nine years. The Observer Magazine, pp. 25 32. Moralis, D. (2012). Missing children Europe Annual report. Brussels: MCE. Nemeth, E. (2015, July 8 10). Comparison of the feasibility of a short-term and a long-term preventative program in child care institutions against running away among youngsters. Paper presented to the 2nd international academic conference on missing children and adults, Vrije Universiteit Brussel, Brussels. Office of the Children’s Commissioner. (2012). Accelerated report on the emerging findings of the OCC’s inquiry into child sexual exploitation in gangs and groups with a special focus on children in care. London: OCC. Rees, G. (2011). Still running: Early findings from our third national survey of young runaways. London: The Children’s Society. Rego, G. (2013). Missing children Europe Annual report. Brussels: MCE. Sedlak, A. J., Finkelhor, D., Hammer, H., & Schultz, D. J. (2002). National estimates of missing children: An overview. Washington, DC: OJJDP. Smith, T. B., Bunlak, K., Condon, L., & Reed, L. (2005). Children missing from care: The law enforcement response. Alexandria, VA: NCMEC. The Irish Foster Care Association. (2013). Information booklet on safe care in foster care. Dublin: IFCA. Tusla. (2014). Quarter 3 2014 national performance activity report. Dublin: Tusla Child and Family Agency. Vila, B., Morrison, G. B., & Kenney, D. J. (2002). Improving shift work and work-hour policies and practices to increase police officer performance, health and safety. Police Quarterly, 5(1), 4 24. Wishart, B. (2014). We need to get it right: A health check into the Council’s role in tackling child sexual exploitation. Birmingham: Birmingham City Council. Woodford, G. (2009). Care homes could be charged £1,000 for runaways by police; children’s homes in Notts could be asked to pay £1,000 towards police searches for repeat runaways. Nottingham Evening Post, July 13, p. 13.

This page intentionally left blank

ABOUT THE AUTHORS Ebenezer Bayode Agboola is an Assistant Lecturer at the College of Social and Management Sciences, Afe Babalola University Ado Ekiti (ABUAD), Nigeria. He studied Sociology at the Ondo state University, Ado Ekiti, now Ekiti State University, from 1994 to 1999 and the University of Ibadan from 2005 to 2007 where he obtained his B.Sc. and M.Sc., respectively. He has applied for his Doctorate Degree in the University of Ilorin, Nigeria. His research interests include gender, criminology, and social justice. Muaweah Ahmad Alsaleh is a Psychologist in the Department of Psychology at the University of Caen Lower Normandy. He previously worked at the University of Aleppo, Faculty of Informatics, where he assisted the adaptation of adolescents with physical handicaps to their environment, particularly in the use of technologies. Most recently, he spoke on depression at the International Mental Health Congress (IMHC) in Lille, France. I taught at the University of Aleppo in Syria and at the University of Caen. Most recently, I’m working with patients in multiple sclerosis at the University Hospital Center of Caen. Shahid Alvi is Professor in the Faculty of Social Science and Humanities at the University of Ontario Institute of Technology. He received his PhD from Carleton University. His current research interests include violence against immigrant women and the elderly, youth crime, communities and crime, digital technology in society, and the links between violence and culture. His recent publications focus on young offender law in neo-liberal society, divorced and separated men in transition, and the commodification of university and college education. He is the author or co-author of numerous articles, book chapters, and five books. He has also consulted to a number of communities on public safety issues. In 2002, he was the recipient of the Critical Criminologist of the Year award from the American Society of Criminology’s Critical Criminology Division. Hila Avieli is an Assistant Professor of Criminology at the University of Ariel, Faculty of Social Sciences, and a Postdoctoral Fellow at the Faculty of Welfare and Health Sciences, University of Haifa. Also, she is Research 435

436

ABOUT THE AUTHORS

Fellow at the Center for the Study of Ageing. Being both a social worker and a criminologist, Dr. Avieli’s career involves research in issues that are relevant to social work and criminology such as family violence, elder abuse, mental illness, and aggression. She views herself as a researchpractitioner who is applying research skills and principles into practice. Her future projects include exploring the field of disability and crime within the family from a gerontological perspective thereby improving our understanding of these excluded groups of individuals. Tova Band-Winterstein is a Senior Lecturer at the Department of Gerontology at the University of Haifa, Faculty of Social Welfare and Health Sciences, and a Research Fellow at the Center for the Study of Ageing. Her career combines ongoing involvement in clinical practice and supervision in the field of gerontology and elder abuse and neglect with applied research. She views herself and functions as a research-practitioner who is applying clinical knowledge in research and research skills and principles in practice. She participated in the research team that wrote the First National Elder Abuse and Neglect grant. Her writing is phenomenological and expresses the much-needed insider’s view of elder abuse by giving voice to the old persons who experience it. Carla Cesaroni is an Associate Professor in the Faculty of Social Science and Humanities at the University of Ontario Institute of Technology. She received her PhD from the University of Toronto. Her research seeks to examine the stress and adjustment of incarcerated individuals. Additionally, she is interested in the role of punishment in the criminal justice system and in the public psyche. For the last decade, she has studied the experiences of adolescent males and females serving custodial sentences. She is currently conducting a comparative study of incarcerated young adults in Canada and Scotland. Aime´e X. Delaney earned a PhD from the University of New Hampshire, where she studied under Dr. Murray A. Straus and Dr. Cesar Rebellon. Her research focuses on family and community violence, risk factors associated with juvenile crime, risky behaviors among youth, and the victimization of youth. She has developed and analyzed data from three different Youth Risk Behavior Surveys. In addition to writing policy publications and monographs related to work on a Drug Free Community Grant (2010 2015), Dr. Delaney has published in the Encyclopedia of Community Corrections (2012). Dr. Delaney regularly presents research at the Academy of Criminal Justice Sciences, the American Society of Criminology, and the

About the Authors

437

Northeast Association of Criminal Justice Sciences. She is currently the Secretary on the executive board for the Northeast Association of Criminal Justice Sciences. Dr. Delaney is an Assistant Professor of Criminal Justice at Worcester State University, where she teaches courses in the areas of victimization, juvenile justice, theories of crime, and research methods. Steven Downing is currently an Associate Professor in the Faculty of Social Science and Humanities at the University of Ontario Institute of Technology. He received his PhD in Criminology from The University of Texas at Dallas in 2007. Allison Dwyer Emory is a PhD Candidate in the departments of Sociology and Policy Analysis and Management at Cornell University. She received her B.A. in Political Science from Wellesley College. Prior to attending graduate school, she worked at the Justice Policy Center at the Urban Institute, focusing on jail reentry and justice reinvestment issues. Her research focuses on the interactions between families and the legal system, including both family and criminal courts. Her dissertation investigates the implications of parental incarceration for children and families. Olufemi Adeniyi Fawole is a Faculty Member in the Department of Sociology at the University of Ilorin. His research has focused on developmental issues for adolescents and young adults in Nigeria. His recent books include Family Risk Factors and Juvenile Delinquency in Nigeria: A Study of Senior Secondary School Students in Ilorin, Kwara State, Nigeria (with Ayodele Adeniyi and Segun Awoniyi) and Personality and Communication Styles of Married Adults in Nigeria: Effects of Personality Types and Communication Styles on Marital Stability among Couples in Nigeria (with Blessing Fawole). Frank D. Fincham obtained a Doctoral Degree in Social Psychology as a Rhodes Scholar at Oxford University. He then completed Postdoctoral Training in Clinical Psychology at Stony Brook University before assuming a position as Assistant Professor at the University of Illinois, where he ultimately became a Professor and the Director of Clinical Training. Following a brief period as a professor in the United Kingdom, he became a SUNY Distinguished Professor at the University at Buffalo before assuming his current position as Eminent Scholar and Director of the Family Institute at The Florida State University. A Fellow of five different professional societies, Fincham has been listed in the Association for Psychological Science Observer as among the top 25 psychologists in the world in terms of impact (defined as number of citations per paper).

438

ABOUT THE AUTHORS

Peggy C. Giordano is a Distinguished Research Professor of Sociology at Bowling Green State University. Her research interests include gender and crime, social network influences on delinquency and adult criminal behavior, intergenerational transmission, mechanisms underlying criminal desistance, and relationship dynamics associated with intimate partner violence. Suleman Ibrahim is a Research Assistant and a PhD Student (EPSRC funded) at the Centre for Doctoral Training in Cyber Security, where he is co-supervised by the Information Security Group and the Centre for Criminology and Sociology, Royal Holloway University of London. He received a M.Sc. in Criminal Justice Policy from London School of Economics and Political Science (LSE) and a B.Sc. in Sociology and Psychology from the University of Greenwich. He was the winner of the prestigious University Merit Award, School of Humanities and Social Science in 2013. Prior to his university education, he attended Lambeth College, Clapham Centre, in Southwest London. His research interests include but not limited to the sociology of cyber-crime and cyber security, juvenile delinquency, cross-cultural marriage and family studies, “race”, ethnicity and crime, postcolonial perspectives, and counter colonial criminology. Beyond his research endeavors in social science, he is also a poet, and some of his poems have appeared in the following books (German): Wortstu¨rmer: Das Buch zum Literaturpreis and Ueberleben am Abgrund: Psychotrauma und Menschenrechte. He was awarded a poetry prize in Austria with a poem published in the former, and he has recently authored an unpublished collection of poems: Blown into a Stream. Catherine Kaukinen, PhD, is a Professor and Chair in the Department of Criminal Justice at the University of Central Florida. Dr. Kaukinen received her PhD in Sociology in 2001 from the University of Toronto. Her research interests include intimate partner violence, risk and protective factors for violent victimization, the history of Title IX and Federal initiatives to address sexual violence against college women, and the evaluation of campus-based violence against women prevention and intervention programs. With funding from the Office on Violence Against Women, she has developed a multi-campus victim service intervention and prevention program addressing dating violence, intimate partner violence, sexual assault, and stalking. She also holds a contract with the Colorado Department of Corrections to conduct research on the effectiveness of institutional programming to prepare offenders for their release into the community. Her research has appeared in Criminology, Journal of Marriage

About the Authors

439

and Family, Journal of Research in Crime & Delinquency, Journal of Interpersonal Violence, and Violence and Victims, among other outlets. Preethi Krishnan is a PhD Student in Sociology at Purdue University. She has an M.A. in Personnel Management and Industrial Relations from Tata Institute of Social Sciences, Mumbai, India. Her research areas are gender, violence, and social movements particularly in South Asia. Her Masters project at Purdue focused on laws and domestic violence and which examined archival data on decisions, on domestic violence cases, made by India’s Supreme Court for the period 1995 2011. She has worked in a nonprofit research organization in India where she documented best practices in women’s empowerment programs in India. Monica A. Longmore is a Professor of Sociology at Bowling Green State University. Her interests include social psychological processes, including the nature and consequences of dimensions of the self-concept, and the impact of intimate partner violence on depression. Wendy D. Manning is a distinguished Research Professor of Sociology at Bowling Green State University, Director of the Center for Family and Demographic Research, and Co-Director of the National Center for Family and Marriage Research. Her research examines family, relationship and marriage patterns, with a focus on young adulthood. Melissa A. Menasco is an Assistant Professor in the Criminal Justice Department at The State University of New York, College at Buffalo, Buffalo, NY. Her early research interests include the study of juvenile delinquency in terms of social control theory, academic performance, and gender. More recently, her studies have included self-control theory, crosscultural delinquency, and juvenile psychopathy. She has given numerous conference presentations at the annual meetings of the American Society of Criminology, Eastern Sociological Society, Academy of Criminal Justice Sciences, and the American Sociological Association. Her most recent work includes studying the long-term effects of adolescent substance use upon marriage, divorce, and remarriage. Her postdoctoral work under Dr. Mark G. Frank (University at Buffalo, Department of Communication) was in the area of nonverbal communication and deception detection. Dr. Menasco received her undergraduate degree from U.C.L.A. in Mathematics and her master’s and doctorate from the University at Buffalo in Sociology.

440

ABOUT THE AUTHORS

Mallory D. Minter is a Doctoral Student in Sociology at Bowling Green State University. Her research interests include peer influences on young adult intimate partner violence, parenting within the context of violent relationships, intimate partner violence among the aging population, and the impact of arrests and incarceration on intimate relationships. Sonali Mukherjee is a Research Fellow with Institute of Social Sciences, New Delhi. She is the Project Director of ICSSR funded project “Structural Basis of Gender Violence: Analysis of Implications of Adverse Sex Ratio in Haryana and West Bengal.” She presented this paper at XVIII ISA World Congress of Sociology, held in Yokohama, Japan, from July 13 to July 19, 2014. Her primary interest is in gender studies. Her doctoral work was on “Economic Development and the Changing Position of Tribal Women: A Study of Transformation in Maria Society,” an empirical study based in Dantewada district of Chhattisgarh. She is also working on the manuscript of her book, Marias of Bastar: The People, Their Lives and Livelihood strategies. She has been leading research projects on various subjects like land leasing and agriculture; marketing development of tribal products/produce; women and agriculture; women’s self-help groups, and social security schemes for tribal women. She taught for a couple of years at the graduate and postgraduate level. She taught Social Anthropological theories, Political Anthropology and Research methodology at the PG level. She taught papers on Research Methodology, Economic Sociology, Gender, Human Rights and Environment at the graduate level. Amanda Murray received her Bachelor of Arts degree in Psychology from American University in 2015. She plans to earn a dual Masters degree in Clinical Counseling and Forensic Psychology. She has previously served as a Health Educator in DC Public Schools with Peer Health Exchange. Her research interests include developmental psychology, psychopathology, and legal decision-making. Rosario Pozo Gordaliza is an Assistant Lecturer at Universitat de les Illes Balears, in the Department of Pedagogy and Specific Didactics. She has her PhD in Sociology from the University of Granada (2011), Diploma in Social Education (UVA), and Bachelor of Psychology (ULE) with various specializations Masters in social problems, qualitative social research methodology, ethnicity, and gender at national and international level. She has been hired FPU scholarship and training program or research staff of the MEC in the Department of Sociology at the University of Granada (2004 2009). From 2009 until 2011, she has worked in international

About the Authors

441

cooperation projects, consulting and social research in the AECID, and various NGOs in Southeast Asia. Peter M. Rivera is a Doctoral Student in the Family and Child Sciences Department at The Florida State University (FSU). Prior to joining FSU, Rivera studied at Seattle Pacific University. He has been part of numerous research teams, and is currently part of FSU’s Family Institute. His research has focused on understanding the mechanisms and pathways to adverse long-term effects of maltreatment. Currently, he is a Fellow in the American Association for Marriage and Family Therapy Minority Fellowship Program. Rivera holds an M.S. in Marriage and Family Therapy from Seattle Pacific University, and a B.A. and B.S. in Psychology and Social Science from Central Washington University. Michael Rush, PhD, is a College Lecturer in the School of Social Policy, Social Work and Social Justice in University College Dublin whose latest book is entitled Between Two Worlds of Father Politics: USA or Sweden, which is published by Manchester University Press (2015). With Dr. Marie Keenan he was co-author of The Social Politics of Social Work: AntiOppressive Social Work Dilemmas for 21st Century Welfare Regimes (2014) for the British Journal of Social Work. Recent work also includes An Evaluation of the Missing Children Hotline 116 000 for the ISPCC (2014). Joanne Savage is an Associate Professor in the Department of Justice, Law and Criminology at American University. She is primarily interested in the “big picture” of violence in society and, has researched a wide variety of causes of violent offending. Her recent interests emphasize the development of violent behavior including abuse, parental attachment, warmth and rejection, and academic achievement and school bonds. She has also studied the effects of media violence on violent behavior, and social factors such as race, inequality, and violent crime trends in Washington, DC. Yael Smeloy is a Social Worker, Head of the Elderly Department in a Social Services Agency in Israel. She is also a graduate of gerontology (M.A.) from the University of Haifa. During 20 years of working with senior citizens, she developed specialty in working with those who are exposed to elder abuse. She views herself as a research-practitioner who is applying clinical knowledge in research and vice versa. As a social worker in the field of elderly abuse and neglect, she developed insights and skills about those who are dealing with their harmful and abusive child.

442

ABOUT THE AUTHORS

Mangala Subramaniam is Associate Professor in the Department of Sociology at Purdue University. Educated at Delhi University, the University of Pennsylvania, and the University of Connecticut, Mangala’s research is in the broad areas of gender (and intersections with caste, race, and class) and social movements. She focuses on issues that are at the center of gender politics, development, basic needs (such as health, water, and food), and the state. The key issues addressed in her research include how the disadvantaged organize to make claims; the gendered basis of interpretation of the law in judgments on violence against women (VAW) cases; and the innovative forms of organizations, including community-based groups, that emerge to collectively seek access to basic needs. She has recently authored/co-authored articles in journals such as International Sociology and Current Sociology. Her current research projects focus on violence against women, water rights, and distribution of improved seed technologies, and HIV prevention strategies in India. She has presented papers at international conferences in Accra (Ghana), Delhi (India), and Tokyo (Japan). She has been a member of the International Team of Experts for Poverty Alleviation of the United Nations Economic Commission for Europe (Geneva) from 2001 to 2004. Mangala has held offices in national professional associations including the ASA and SSSP. She is currently the Treasurer-elect of Sociologists of Women in Society (SWS). Henriikka Weir, PhD, is an Assistant Professor in the School of Public Affairs at the University of Colorado at Colorado Springs. Dr. Weir received her B.A. in Psychology (2008), M.S. in Criminology (2009), and Ph.D. in Criminology (2013) from the University of Texas at Dallas. She is passionate about children’s rights and substance abuse treatment for offenders. She also enjoys teaching and interaction with students. Dr. Weir’s research interests include child maltreatment, substance abuse, policing, biosocial and developmental explanations of crime, as well as quantitative research methods. Dr. Weir’s research has been published in Journal of Criminal Justice and Behavior, Journal of Criminal Justice, Journal of Criminal Justice Education, and Youth Violence and Juvenile Justice, among others. Xing Zhang is a PhD Student in the Department of Policy Analysis and Management at Cornell University. She received her B.A. in Economics from the University of Illinois at Chicago. Her research interests are in family demography, race, and ethnicity.