Delinquency and Substance Use in Europe: Understanding Risk and Protective Factors [1st ed.] 9783030584412, 9783030584429

Citation preview

David P. Farrington Harrie Jonkman Frederick Groeger-Roth  Editors

Delinquency and Substance Use in Europe Understanding Risk and Protective Factors

Delinquency and Substance Use in Europe

David P. Farrington  •  Harrie Jonkman Frederick Groeger-Roth Editors

Delinquency and Substance Use in Europe Understanding Risk and Protective Factors

Editors David P. Farrington Institute of Criminology University of Cambridge Cambridge, UK

Harrie Jonkman Verwey-Jonker Institute Utrecht, The Netherlands

Frederick Groeger-Roth Department of Justice of Lower Saxony Crime Prevention Council of Lower Saxony Hannover, Germany

ISBN 978-3-030-58441-2    ISBN 978-3-030-58442-9 (eBook) https://doi.org/10.1007/978-3-030-58442-9 © Springer Nature Switzerland AG 2021 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Foreword

Communities That Care (CTC) is a system for empowering communities to promote healthy development and prevent problem behaviors in children and adolescents. The CTC system has been proven in a community-randomized trial and a statewide quasi-experimental trial to produce sustained population-level reductions in the initiation of delinquent and violent behaviors and drug misuse in communities (Chilenski, Frank, Summers, & Lew, 2019; Hawkins et al., 2012; Oesterle et al., 2018). A book by Fagan and colleagues (2019) describes CTC’s development, testing, and dissemination. The Center for Communities That Care at the Social Development Research Group, School of Social Work, University of Washington, directed by Dr. Kevin Haggerty, disseminates the CTC intervention (www.communitiesthatcare.net). Key to the effectiveness of Communities That Care is the Communities That Care Youth Survey (CTC-YS), a tool for assessing levels of antisocial behavior and drug use and the risk factors that predict and protective factors thought to prevent these behaviors among young people in a community. Typically administered anonymously in a community’s schools serving 12- to 18-year-old students, the CTC-YS enables the community’s young people to honestly report their own exposures, experiences, and behaviors without fear. The survey’s scales provide a baseline assessment of risk, protection, and behavioral problems that communities use to guide youth development and problem behavior prevention efforts and to measure progress over time through repeated administrations of the CTC-YS every two years. The development and psychometric testing of the CTC-YS were led by Dr. Michael W.  Arthur, a beloved member of the Social Development Research Group at the University of Washington who passed away this last year. His leadership in its development, with input from states and communities, resulted in a reliable and valid survey instrument whose results are easily scored, understood, and used by communities (Arthur et al., 2007; Arthur, Hawkins, Pollard, Catalano, & Baglioni, 2002; Briney, Brown, Hawkins, & Arthur, 2012; Glaser, Van Horn, Arthur, Hawkins, & Catalano, 2005). From the outset, Michael, Richard F. Catalano, and I agreed that the CTC-YS should be publicly available to communities without cost so that any community seeking an epidemiologically reliable and valid assessment v

vi

Foreword

of levels of risk exposure, protection, and problem behaviors of its young people could freely use the CTC-YS to guide its prevention work. The chapters in this exciting new book, reporting results from the Communities That Care Youth Survey from five European countries, indicate the wisdom of placing the CTC-YS in the public domain. These European investigators have used the CTC-YS to answer both shared and unique questions about risk exposure, protective factors, and problem behaviors among their own young people. The common foundations of the five studies in different European countries in this book are well described by the editors, Harrie Jonkman, David Farrington, and Frederick Groeger-­ Roth, in Chap. 1. In Chap. 8, the editors provide an excellent summary of conclusions across studies and the way forward for empirically based prevention of adolescent delinquency and drug use in Europe and beyond. Chapter 2 by David Farrington, David Utting, and Nick Axford provides an excellent history of the efforts to bring the CTC prevention system to Great Britain and describes previous efforts to conduct national surveys of delinquency and drug use in Great Britain. Elegant analyses of the nationally representative CTC-YS data in Tables 2.6 and 2.7 provide independent validation from Great Britain of the relationships between risk and protective factors and adolescent behavior outcomes found in the USA and other samples (Baheiraei et al., 2016; Hemphill et al., 2011). I appreciate the discussion in this chapter seeking to distinguish causal from co-­ occurring risk factors from a preventive perspective. The data from Great Britain underscore the continuing importance of family protective factors in adolescent development. Chapter 3 by Harrie Jonkman and Clemens Hosman, reporting on CTC in the Netherlands, used an internet rather than school survey and different analysis methods, dichotomizing at the median on each factor and counting those above the median as having that risk or protective factor. They also reported the Population Attributable Fraction (PAF) for each risk and protective factor. Tables 3.5 and 3.6 are worth studying regarding odds ratios and PAF for delinquency and violent behavior for measured CTC-YS risk and protective factors included in the Netherlands data set. The data in this chapter again point to the importance of protective factors of family opportunities, rewards, and bonding to family, in preventing delinquent and violent behavior. Chapter 4 by Frederick Groeger-Roth and Burkhard Hasenpusch provides an excellent example of planning, full implementation, and assessment of the CTC process in three pilot sites and then five sites in Lower Saxony, Germany, using all the CTC data collection instruments including the CTC-YS. The survey was administered online in schools to whole classrooms of students simultaneously. Both this chapter and Chap. 3 provide figures showing the relationship of exposure to increasing numbers of risk factors to increased prevalence of delinquent behavior and violence and the relationship of increased protection with less delinquency and violence. Continued work on the measurement of risk and protective factors and more recent use of empirically derived cutoff scores to identify those at risk on each specific factor are interesting German developments described in the chapter.

Foreword

vii

Note that researchers in the Netherlands and Germany elected to eliminate CTC-YS scales with Cronbach’s Alpha reliability less than 0.60. This resulted in different scales being eliminated from the data sets in these two countries, limiting the ability to compare the strength of relationships between risk and protective factors and behavior outcomes across countries. In such cases, it might be worth exploratory analyses using the less reliable scales to see how similar or different results are across studies even in the face of potential reliability problems which should lessen the likelihood of finding relationships. Chapter 5 by Josipa Basic, Miranda Novak, and Josipa Mihic describes the use of the CTC-YS in two cities in Croatia with a sample of 1424 secondary school and high school students. Table 5.5 is an interesting display of unique and common relationships of risk and protective factors as measured in the Croatian-developed CTC-YS with a range of youth outcomes including gambling, suicidal ideation, and depressive symptoms in addition to violence, drug use, and alcohol use. Like each of the preceding chapters, this chapter provides a clear description of the implications of the findings for policy. Chapter 6, by Andreas Kapardis, George Spanoudis, Constandina Kapardis, and Maria Konstantinou, provides a good example of adherence to the CTC system in seeking to involve local key leaders in CTC decisions regarding priority risks to be addressed and interventions to be used in Cyprus. The results of analyses show the strength of risk factors in predicting youth outcomes in a structural equation model. Chapter 7 provides interesting comparisons of the prevalence of delinquency and drug use behaviors across the samples from different countries. Note that even in the face of quite different prevalences of delinquent behavior and drug use across different samples and use of different analysis methods, many risk and protective factors are consistent in direction and strength of relationship with outcomes across samples from different countries. While risk factors are consistently more strongly related to delinquency and drug use outcomes than protective factors in these data sets, family protective factors of opportunities, rewards, and bonding during adolescence appear to have consistent protective effects across several studies and countries. I am very happy to see the presentation of the theoretical model underlying the CTC system and CTC Youth Survey, the social development model (Cambron, Catalano, & Hawkins, 2019; Catalano & Hawkins, 1996; Hawkins & Weis, 1985), described in Chap. 8. Different types of organizations undertook implementation and use of the CTC-YS in the different studies reported here, including universities, government agencies, and foundations. Yet all these studies were driven by a clear vision of the importance of measuring empirically derived risk and protective factors in addition to epidemiologically valid data on adolescent delinquency, violence, and substance use. All these chapters describe how valid and reliable measurement of risk and protection exposure in youth populations can be used to guide policies and preventive interventions in cities, counties, states, and nations, often with very specific recommendations based on the data presented.

viii

Foreword

This book demonstrates the power of international collaboration in a time when it appears in short supply in the face of the pandemic spread of the COVID 19 virus. These authors show the utility of designing, conducting, and reporting studies that include epidemiologically valid measures of risk and protection and adolescent problem behaviors based on a shared theoretical foundation and using a shared measurement tool, the CTC-YS. It is heartening to see how these brilliant scientists, researchers, policy-makers, community developers, and practitioners in five different countries have used the CTC-YS to improve understanding of youth behaviors and to create the empirical foundation for promoting prosocial behaviors and preventing adolescent problem behaviors. Michael Arthur would be very happy. J. David Hawkins Social Development Research Group School of Social Work, University of Washington Seattle, WA, USA

References Arthur, M. W., Briney, J. S., Hawkins, J. D., Abbott, R. D., Brooke-Weiss, B. L., & Catalano, R. F. (2007). Measuring risk and protection in communities using the Communities That Care Youth Survey. Evaluation and Program Planning, 30, 197-211. Arthur, M. W., Hawkins, J. D., Pollard, J. A., Catalano, R. F., & Baglioni, A. J., Jr. (2002). Measuring risk and protective factors for substance use, delinquency, and other adolescent problem behaviors: The Communities That Care Youth Survey. Evaluation Review, 26, 575-601. Baheiraei, A., Soltani, F., Ebadi, A., Cheraghi, M. A., Foroushani, A. R., & Catalano, R. F. (2016). Psychometric properties of the Iranian version of ‘Communities That Care Youth Survey’. Health Promotion International, 31, 59-72. Briney, J. S., Brown, E. C., Hawkins, J. D., & Arthur, M. W. (2012). Predictive validity of established cut points for risk and protective factor scales from the Communities That Care Youth Survey. Journal of Primary Prevention, 33, 249-258. Cambron, C., Catalano, R. F., & Hawkins, J. D. (2019). The social development model. In D. P. Farrington, L. Kazemian, & A. R. Piquero (Eds.), The Oxford handbook of developmental and life-course criminology (pp. 224-247). New York, NY: Oxford University Press. Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of antisocial behavior. In J. D. Hawkins (Ed.), Delinquency and crime: Current theories (pp. 149-197). New York, NY: Cambridge University Press. Chilenski, S. M., Frank, J., Summers, N., & Lew, D. (2019). Public health benefits 16 years after a statewide policy change: Communities That Care in Pennsylvania. Prevention Science, 20, 947-958. Fagan, A. A., Hawkins, J. D., Catalano, R. F., & Farrington, D.P. (2019). Communities That Care: Building community engagement and capacity to prevent youth behavior problems. New York, NY: Oxford University Press. Glaser, R. R., Van Horn, M. L., Arthur, M. W., Hawkins, J. D., & Catalano, R. F. (2005). Measurement properties of the Communities That Care® Youth Survey across demographic groups. Journal of Quantitative Criminology, 21, 73-102.

Foreword

ix

Hawkins, J. D., Oesterle, S., Brown, E. C., Monahan, K. C., Abbott, R. D., Arthur, M. W., & Catalano, R. F. (2012). Sustained decreases in risk exposure and youth problem behaviors after installation of the Communities That Care prevention system in a randomized trial. Archives of Pediatrics and Adolescent Medicine, 166, 141-148. Hawkins, J. D., & Weis, J. G. (1985). The social development model: An integrated approach to delinquency prevention. Journal of Primary Prevention, 6, 73-97. Hemphill, S. A., Heerde, J. A., Herrenkohl, T. I., Patton, G. C., Toumbourou, J. W., & Catalano, R. F. (2011). Risk and protective factors for adolescent substance use in Washington State, United States and Victoria, Australia: A longitudinal study. Journal of Adolescent Health, 49, 312-320. Oesterle, S., Kuklinski, M. R., Hawkins, J. D., Skinner, M. L., Guttmannova, K., & Rhew, I. C. (2018). Long-term effects of the Communities That Care trial on substance use, antisocial behavior, and violence through age 21 years. American Journal of Public Health, 108, 659-665.

Preface

The main aim of this book is to present information about delinquency and drug use and risk and protective factors that are related to these social problems, in five European countries: Great Britain, the Netherlands, Germany, Croatia, and Cyprus. This information was obtained by administering the same survey (the Communities That Care Youth Survey or CTC-YS) to samples of young people in all the different countries. This survey made it possible to compare these social problems, and influences on them, in these five countries. Communities That Care (CTC) is a community change process that targets risk and protective factors with preventive interventions that have been proved to be effective. The CTC-YS was developed as a tool to provide community-based partnerships with reliable information about the prevalence of youth behavior problems, as well as the prevalence of underlying risk and protective factors. This assessment enables communities to select top-priority risk and protective factors for targeting, based on the profile of the particular community, and to match the most elevated risk factors and/or most depressed protective factors with preventive strategies. Currently, little is known about the generalizability of this risk and protection approach in Europe. This book aims to fill this gap in knowledge. The work that led to this book was made possible through a grant by the “Prevention of and Fight against Crime Program” of the European Commission—Directorate-General Home Affairs for the project “Communities That Care (CTC) European Network: Making CTC work at the European level.” Partners in this project were: –– –– –– –– –– –– –– –– ––

Crime Prevention Council of Lower Saxony (Germany) Dartington Social Research Unit (United Kingdom) Verwey-Jonker Institute (The Netherlands) Seinpost Adviesbureau (The Netherlands) University of Applied Sciences Leiden (The Netherlands) Institute for the Prevention of Addictions and Drug Abuse (Austria) City of Malmö (Sweden) University of Cyprus (Cyprus) University of Zagreb (Croatia) xi

xii

Preface

Invaluable support and advice were provided by members of the project advisory board: –– David P. Farrington, Chair (University of Cambridge, UK) –– Gregor Burkhart (European Monitoring Centre for Drugs and Drug Addiction, Portugal) –– Carmel Cefai (University of Malta, Malta) –– David Foxcroft (Oxford Brookes University, UK; European Society for Prevention Research) –– J. David Hawkins (Social Development Research Group, USA) –– Clemens Hosman (Emeritus Professor of Mental Health Promotion and Prevention, The Netherlands) –– Matej Košir (UTRIP, Slovenia) –– Erich Marks (German Congress on Crime Prevention, Germany) –– Sebastian Sperber (European Forum for Urban Security, France) –– Karin Svanberg (National Council for Crime Prevention, Sweden) –– Elena Tryfonos (Ministry of Education, Cyprus) We would like to thank the European Commission for funding this project, all our collaborators for their contributions and their strong commitment, and in particular David Hawkins for his valuable advice and guidance and for permission to reproduce the CTC-YS questionnaire in this book. Cambridge, UK  David P. Farrington Utrecht, The Netherlands   Harrie Jonkman Hannover, Germany   Frederick Groeger-Roth

Contents

1 Introduction����������������������������������������������������������������������������������������������    1 Harrie Jonkman, David P. Farrington, and Frederick Groeger-Roth 2 Great Britain��������������������������������������������������������������������������������������������   21 David P. Farrington, David Utting, and Nick Axford 3 The Netherlands ��������������������������������������������������������������������������������������   41 Harrie Jonkman and Clemens M. H. Hosman 4 Germany ��������������������������������������������������������������������������������������������������   59 Frederick Groeger-Roth and Burkhard Hasenpusch 5 Croatia������������������������������������������������������������������������������������������������������   77 Josipa Bašić, Miranda Novak, and Josipa Mihić 6 Cyprus������������������������������������������������������������������������������������������������������   97 Andreas Kapardis, George Spanoudis, Constandina Kapardis, and Maria Konstantinou 7 Comparison of Countries������������������������������������������������������������������������  115 Harrie Jonkman, David P. Farrington, and Frederick Groeger-Roth 8 Conclusions����������������������������������������������������������������������������������������������  131 Harrie Jonkman, David P. Farrington, and Frederick Groeger-Roth Appendix: Communities That Care Youth Survey ��������������������������������������  143 Index������������������������������������������������������������������������������������������������������������������  155

xiii

Editors and Contributors

About the Editors David P. Farrington, O.B.E.  is Emeritus Professor of Psychological Criminology at Cambridge University. He has received the Stockholm Prize in Criminology and he has been President of the American Society of Criminology. His major research interest is in developmental criminology, and he is Director of the Cambridge Study in Delinquent Development, which is a prospective longitudinal survey of over 400 London males from age 8 to 61. In addition to 853 published journal articles and book chapters on criminological and psychological topics, he has published 115 books, monographs, and government publications and 162 shorter publications (total = 1130). Harrie  Jonkman  studied sociology and educational studies and worked in the field of youth development and education. He works at the Verwey-Jonker Institute, a social research institute in the Netherlands. His work focuses on the social and cognitive development of children and young people, social determinants, and prevention of health, developmental, and behavioral problems. He was national project leader of the Community That Care strategy in the Netherlands and supported this also in other countries. He supported prevention work in different countries and worked in steering committees of educational institutes. At the Verwey-Jonker Institute, he is involved in experiments and evaluations of social programs, longitudinal studies, and international comparative studies. Frederick  Groeger-Roth  studied Sociology, Psychology, and Political Sciences in Bielefeld and Berlin. He has researched on youth violence in deprived urban areas and worked for NGO’s on regional and national levels in the area of urban development. In 2009, he joined the Crime Prevention Council of Lower Saxony (CPC), located in the Ministry of Justice of Lower Saxony. He has headed the first Communities That Care pilot study in Germany and is at present senior advisor for the CPC unit on community-based prevention. He was appointed to the expert xv

xvi

Editors and Contributors

a­ dvisory board of the German Forum for Crime Prevention and of the National Centre for Crime Prevention in Germany.

Contributors Nick Axford  Peninsula Medical School, University of Plymouth, Plymouth, UK Josipa  Bašić  Faculty of Education and Rehabilitation Sciences, University of Zagreb, Zagreb, Croatia David  P.  Farrington  Institute of Criminology, University of Cambridge, Cambridge, UK Frederick Groeger-Roth  Ministry of Justice of Lower Saxony/Crime Prevention Council of Lower Saxony, Hannover, Germany Burkhard  Hasenpusch  Ministry of Justice of Lower Saxony, Langenhagen, Germany Clemens M. H. Hosman  Maastricht University and Radbout University Nijmegen, Groesbeek, The Netherlands Harrie Jonkman  Verwey-Jonker Institute Utrecht, Utrecht, The Netherlands Andreas Kapardis  Department of Law, University of Cyprus, Nicosia, Cyprus Constandina  Kapardis  Australian Graduate School of Policing and Security, Charles Sturt University, Goulborn, NSW, Australia Maria Konstantinou  Department of Psychology, University of Cyprus, Nicosia, Cyprus Josipa  Mihić  Laboratory for Prevention Research, Department of Behavioural Disorders, Faculty of Education and Rehabilitation Sciences, University of Zagreb, Zagreb, Croatia Miranda Novak  Laboratory for Prevention Research, Department of Behavioural Disorders, Faculty of Education and Rehabilitation Sciences, University of Zagreb, Zagreb, Croatia George  Spanoudis  Department of Psychology, University of Cyprus, Nicosia, Cyprus David Utting  Independent Writer and Researcher, St Albans, UK

Chapter 1

Introduction Harrie Jonkman, David P. Farrington, and Frederick Groeger-Roth

1.1  Background In recent decades, cross-national comparative research on youth substance use in European countries has increased significantly, owing to international self-report surveys. The European School Survey Project on Alcohol and Other Drugs and the Health Behaviour in School-aged Children (HBSC) survey are examples of these. These studies show that underage drinking and the use of drugs are important public health concerns in many countries. It is clear that alcohol is the drug most commonly used by youths; that adolescents who indulge in drinking are more likely to engage in risky behaviour such as drinking and driving; that underage drinking contributes to both unintentional and intentional injury deaths among adolescents and that adolescents who drink heavily are at increased risk of short- and long-term health problems (Steketee et al. 2012). The study of alcohol and drug use among youth and the consequences in the longer term has become an issue of growing importance. It is clear that these problems require a focus on prevention. In the twenty-first century, one of the biggest concerns in Europe is adolescents’ substance use and the increased use of drugs and

H. Jonkman (*) Verwey-Jonker Institute Utrecht, Utrecht, The Netherlands e-mail: [email protected] D. P. Farrington Institute of Criminology, University of Cambridge, Cambridge, UK e-mail: [email protected] F. Groeger-Roth Ministry of Justice of Lower Saxony/Crime Prevention Council of Lower Saxony, Hannover, Germany e-mail: [email protected] © Springer Nature Switzerland AG 2021 D. P. Farrington et al. (eds.), Delinquency and Substance Use in Europe, https://doi.org/10.1007/978-3-030-58442-9_1

1

2

H. Jonkman et al.

binge drinking.1 These changes have led some researchers to argue that a ‘new culture of intoxication’ has developed among European adolescents (Järvinen and Room 2007; Measham and Brain 2005). It is known that binge drinking has a close association with all types of delinquency, especially with violence (Felson et al. 2011; Gatti et al. 2013; Gatti et al. 2015). Juvenile delinquency can be defined as an underage person’s actions that violate the societal rules of behaviour embodied in law. Juvenile delinquency is a worldwide concern, but cross-national comparisons in the field of criminology have often been recognized as difficult to carry out (see, e.g. Farrington 2015). The European Sourcebook of Crime and Criminal Justice Statistics and the International Self-Report Delinquency Survey or ISRD are frequently conducted surveys that are considered to offer good insights into the delinquent world of juveniles. According to these studies, it is known that in today’s Europe delinquent behaviour among adolescents is not exceptional. Minors commit a significant proportion of all crime and especially of different kinds of property crimes. In Europe, juvenile delinquency is most common in the wealthiest Anglo-Saxon, Western European and Northern European countries, followed by the Southern European countries. However, according to the ISRD conducted between the years of 2005 and 2007, variations between the countries in relation to violent behaviour are large, and it is difficult to draw firm conclusions (Aebi et al. 2014; Junger-Tas 2010). It is commonly acknowledged that delinquency peaks in the adolescent years and that many juveniles who commit delinquent acts during their adolescence do not continue offending in adulthood. However, a significant proportion of adults who are committing crimes started offending in childhood or adolescence (see, e.g. Farrington 2015; Farrington and Wikström 1994). It is generally true that preventing juvenile delinquency is preventing a delinquent way of life (West and Farrington 1977).

1.1.1  The Importance of Prevention The early developmental phase of problem behaviour is important for future human development. We know, for example, that half of all lifetime cases of diagnosable mental illnesses begin by the age of 14 (Kessler et al. 2005). Problem behaviours are linked, in that a change in one type of problem behaviour may increase the development of another problem behaviour (Institute of Medicine 2009). For example, the more behavioural problems that youngsters have, the more likely it is that they will fail in school. This will decrease their chances of obtaining employment, which will increase their dependence on the social welfare system and increase their chances of coming into contact with the juvenile justice system.

1  Although often not accurately defined, ‘binge drinking’ has become a commonly used notion to describe drinking to get drunk.

1 Introduction

3

Rather than waiting until early alcohol consumption turns into alcohol dependence, early tobacco use causes cancer and adolescent antisocial behaviour turns into serious violence and depression, problem behaviour should be prevented at an early age (Hawkins et al. 2008). Here, important health gains are possible. It is crucial that youngsters who transform from being children into being young adults successfully pass the phase of adolescence and have ‘safe passage’ (Dryfoos 1998). In the last 20 years, people have sought more individually based than socially oriented answers in regard to youth problem behaviour. In recent times, authority has become professionalized, and practical answers are being provided more and more by doctors, lawyers and therapists, in comparison with parents, teachers and other important adults. This is at a time when more and more youngsters are required to function in specific contexts away from their family, school and neighbourhood. Nowadays, we live in a society with a lot of opportunities and risks (Boutellier 2015). For children who grow up in strong families, attend good schools and are socially supported and controlled by the communities in which they live, this is less of a problem. However, modern societies, which are defined by endless possibilities and an abundance of unexpected and unfamiliar social networks, can pose problems for children and youngsters who grow up in more chaotic and unstructured situations, without networks of structured relationships and social capital. The chance that they will develop problem behaviours during this time period is greater. These problem behaviours may not only have a negative influence on their lives at present, but they may also cause problems during their adult life later on. These problems can also have negative influences on the lives of people in their environment. Not all adolescents are successful and these problems (one or more, often in tandem) are part of their life story. With help, support and ‘nudges’ from people in their surroundings, and thoughtful preventive interventions at the right time and in the right place, their lives can be made more successful (Thaler and Sunstein 2009). It is clear that the incidence and prevalence of these problem behaviours, such as the use of substances, delinquency and other related behaviours like school dropout, depression and teen pregnancy, commence and/or increase significantly during the passing phase of adolescence and can lead to lifelong health-related problems, diseases and disorders. As a society, we have the responsibility to help all youngsters become independent and successful. Yet the big question remains: what can we do together and how should this be organized?

1.1.2  Behaviour in Social Contexts A crucial factor which influences human health and development is the social environment in which people live and work throughout their life course and how they cope with changing environments (Keating and Hertzman 1999). Humans are social, their lives are lived interdependently, and social influences are expressed through a network of shared relationships. Through social discourse and action, as well as interactions and relations with others, an individual person becomes human

4

H. Jonkman et al.

through social experiences (Dewey 1907; Habermas 1981; Mead 1934; Vygotsky 1978). Opportunities which cross their paths, as well as aspirations and ambitions for the future, are all defined by the social environments in which people live and work throughout their life cycle, which in effect are composed of interactions with inspirational others who provide them with direction (Damon 1990; Damon 1997; Damon 2008). Nonetheless, humans also require protection from different threats during specific moments throughout their lifetimes. In order to live a fruitful and optimal life, resistant to the risk of disease, development problems and an early death, protection is highly necessary (Catalano and Hawkins 1996). Individual social competencies, family skills, school quality, community characteristics and resources are all important for the development of adolescents, as prevention science has made clear in several studies (Weissberg et al. 2003). Prevention science has emerged as an interdisciplinary science created by the integration of developmental science and longitudinal studies, social and community epidemiology and research based on preventive and randomized trials (Coie et  al. 1993; Institute of Medicine 2009; Kellam and Rebok 1992; Mrazek and Haggerty 1994). Prevention science identifies two different groups of predictors in terms of individuals and their social environments. One group identifies which factors increase the likelihood of problems (risk factors), whilst the other focuses on factors which moderate and mediate exposure to risk, which in effect will decrease the likelihood of problems (protective factors). Through a number of experimental studies, it was found that tested and effective prevention programmes and policies could be developed, not only for individuals but also for families, schools and communities, in order to support the social and healthy development of youngsters (Elliott 1997; Sherman et al. 1997). In different European countries, preventive interventions have been developed and implemented during the last 20 years (Axford et al. 2016). These interventions aim to halt the development of problem behaviours and disorders and prevent a full-­ blown manifestation of these behaviours and disorders and other associated outcomes. For interventions, it is important that they take place at the right moment, in the right place and for the right reasons. Successful preventive interventions are based on the idea that risk factors should be minimized and protective factors should be enhanced (Pollard et al. 1999). We know more and more about the development of problem behaviours during the phase of adolescence and also about preventive possibilities that are available in the communities where the youngsters grow up and live over a longer period of time. Research questions have been derived from preventive and research activities in very diverse sociocultural contexts, mainly in the USA, Canada and Australia, but also in Europe. As a reflection of today’s consumer societies, in current research, adolescents’ and young adults’ substance use has been often interpreted as a way of forming identity and reaching different kinds of hedonistic goals. Whilst the individual choice has been emphasized, the research has not often taken account of the different kinds of social spheres in which adolescents are living (Measham and Shiner 2009).

1 Introduction

5

In one of the rare longitudinal delinquency studies conducted in Europe, the Cambridge Study in Delinquent Development (CSDD), it was noticed that the motives for committing a crime at younger ages were more hedonistic. During the adolescent years, delinquent acts were committed for excitement and enjoyment or just to relieve boredom (Farrington 2002a, pp.  146), which are typically noticed motives for adolescent substance use also. Most criminological theories, however, emphasize that the reasons behind delinquency and adolescent problem behaviour lay under the surface. For example, based on their longitudinal study, Sampson and Laub (2003) consider the role of social institutions, informal social control and routines as key factors also in desistance from crime in adulthood. In addition, as a social practice, substance use is guided by a variety of social norms that differ between countries (e.g. Katainen & Rolando 2015) and evidently between communities. Although different problem behaviours have been researched according to their own developmental patterns, there are similarities between these patterns of behaviour. As mentioned, problem behaviours tend to occur in tandem with one another (Dryfoos 1998). Research focusing on how all of these problem behaviours are connected and interrelated has been carried out in the past, but rarely in different countries and rarely in a comparative perspective (see Farrington 2002b; Loeber et al. 1998).

1.1.3  Development Over the Years We know that problem behaviours hardly ever spontaneously develop from one day to the next. Instead, these behaviour patterns generally develop over time, with differences but also similarities between them in which genes, social experiences, life courses and social circumstances play an interactive role. Social position is affected by what adolescents experienced in their early lives (conception, birth, early life and childhood), as is their social response to social circumstances (Marmot 2000). Adolescents are affected by their childhood experiences, where parts of their behaviour were already created. Since Freud, some scientists have focused on the importance of the early life exposure (Keating and Hertzman 1999; Tremblay 1999). Now, we know that exposure in early childhood influences cognitive, social and mental development. It is in this early phase that our brain develops with its great plasticity (Bruner 1990; Goldberg 2001). Different responses to critical early phases, for example, make individuals vulnerable or resistant to various diseases later on (Berkman and Kawachi 2000). With the cognitive revolution and great strides in brain research, we also now know the importance of the early childhood phase in terms of the origins of different diseases. In this sensitive period perspective, early social life conditions and early life exposure have causal influences on later health outcomes. However, it is not only the early phase which is important for human development. It is also the accumulation of exposures throughout childhood and

6

H. Jonkman et al.

adolescence and the cumulative advantages or disadvantages. Disadvantages are set in motion often as a result of a series of subsequent experiences that accumulate later during adolescence, which, for example, may be demonstrated through violence (Berkman and Kawachi 2000; Keating and Hertzman 1999; Sampson and Laub 1993). In the end, problem behaviours are an integral part of an individual’s life course, which takes place in the real world over a long period of time in a place where lives are lived and where people follow different paths and experience different stages and turning points in their personal development.

1.1.4  Daily Contexts Contemporary life is socially organized and the social context affects the way in which adolescents think, feel and act (Elder and Conger 2000; Furstenberg et al. 1999; Phelps et al. 2002). Overall, youngsters mainly grow up in four contexts in which they interact with others on a daily basis over a long period of time. These contexts include their family, school, peer group and neighbourhood. Most youngsters have a place or role in their family, which is the first social context in which they interact with others. In most cases, the family protects youngsters against risks and problems. Principles of love, protection and safety are important, and it is in this safe context that children and youngsters learn social and cultural rules, norms and values. Within this secure context, youngsters can practise their behavioural, social and personal skills (Damon 1997; Gardner 2006). In order to accomplish social and healthy maturity, the first years of development are crucial. Practices of monitoring and controlling are part of the parental role and are vital not only in this early phase but also, and perhaps especially, during adolescence when children’s lives broaden. During this time, the management qualities of parents are important (Furstenberg et al. 1999). The world of children expands once they begin attending school. Many young children have their first contacts within these structured institutions outside the family. Some children attend preschool classes, but in Europe nearly every European child attends primary school which starts roughly around the age of 5. After primary school, youngsters attend different types of secondary school. The school is the second important context of socialization for young people. Within the context of the school, they learn cognitive, social and creative knowledge and skills in a structured way. They spend thousands of hours in school during their lifetime (Rutter 1979). They meet similar and different peers and interact daily with students who have been evaluated as having similar academic ability. In addition, they are supervised by different teachers over the years. The organizational structure and social climate of schools also influence the development of youngsters. In recent times, the role of education has become more important in our society, and it seems that it has replaced the family in allocating and socializing youth (Gottfredson and Hirschi 1990).

1 Introduction

7

For children, but even more so for adolescents, the world broadens when they interact with peers. Activities with friends, especially informal activities during leisure time, are important in terms of their individual and social development. Friends are important, as they provide reference in regard to interests, perspectives and interaction with others. This interaction time is often called ‘experimental’. Youngsters’ behaviour, thinking, norms and values are confronted, and many receive new inputs during these years in interaction with peers. These ‘experiments’ are important in terms of identity development in adolescents (Erikson 1987). The influence of the family and the school will differ once adolescents interact more with friends. The neighbourhood or community is the social, physical, geographical and organizational unit in which youngsters grow up and develop (Kawachi and Berkman 2003). Neighbourhoods can often be identified by roads and physical features, but, especially in modern times, the borders are not always that clear. They can be identified as the surrounding area where youngsters are born and live and where they often go to their first school. It is where they play with their friends on the street. When youngsters are 11, 12 or older, their world expands, and they begin to attend schools outside their neighbourhood. The influence of the neighbourhood on the development of youngsters is complex and difficult, and our knowledge is still in its infancy (Sampson 2011; Sampson et al. 1997). However, the social demographic position of the inhabitants and the social-cultural structure (poverty and socioeconomic differences) of the neighbourhood can directly influence child development. Nonetheless, this context also has mediating influences on other contexts in which children grow up, such as familial regimes (Pinkster 2009).

1.1.5  Risk and Protective Factors Overall, the level of health, its distribution and social determinants (risk and protective factors) are essential for understanding problem behaviours, monitoring development and progress and assessing the effects of actions. Risk and protective factors are the best researched social determinants of problem behaviours at this moment, and they have been researched over a long time in sciences like epidemiology, criminology, sociology and prevention (Farrington 2002b; Loeber et al. 1998). Risk factors, on one hand, include those factors related to the child, family, school, peer group or neighbourhood, which are associated with an increased probability of different youth problem behaviours (Hawkins et  al. 1998; Loeber and Farrington 1998). Experimental, observational, longitudinal and aetiological studies have revealed these associations over and over again in different studies during the last decades (Arthur et al. 2007; Dryfoos 1998; Junger-Tas 1996; Junger-Tas 1997; Loeber et al. 2008; Loeber and Farrington 1998). Studies show that several risks in different contexts can contribute to the development of minor and major problems such as bullying and fighting to violence, drugs and alcohol (Farrington 2004; Hawkins et al. 1998; Loeber and Farrington 1998). It is especially the accumulation

8

H. Jonkman et al.

of risk factors which is important. Risk factors have to be detected and are important targets for prevention. Although still less researched compared to risk factors, there is more interest and knowledge in public health and prevention science about the importance of protective factors. These factors not only protect against problem behaviour, but they also increase desirable outcomes, for example, positive adjustment and positive mental health (Catalano et al. 2004). Behaviours are not randomly distributed within the population, but rather they are socially patterned and often clustered together. Poverty, low socioeconomic status and low education are all factors that increase the likelihood of risky behaviours. The social position in which you are born, grow up and live places individuals at ‘risk of risks’ (Rose 1992). That is the reason why individual development can be placed in an ecological context. Environments place constraints on individual behaviour, and norms, social control and opportunities can improve the quality of life (Berkman and Kawachi 2000). There is increasing interest and activity in promoting a more multilevel approach in behavioural, social and health sciences. Development should not only be researched on the individual level but on multiple levels (from the genetic up to the sociocultural and political level of analysis). Individual outcomes are more and more researched in the context of higher levels in which these outcomes operate (Galea 2007; Luke 2004; Viner et al. 2012).

1.1.6  Communities That Care (CTC) Given the fact that different problem behaviours of youngsters are connected and interrelated and have their developmental patterns, the importance of the four daily contexts in which they grow up and the importance of risk and protective factors, what can be done? Would it be realistic to consider building or rebuilding environments that have a desirable effect on a child’s healthy and social development, when this development in itself is so complex? Is it possible to reduce this knowledge and complexity into workable steps and goals for practitioners and politicians? Communities That Care (CTC) is an example of a prevention system, which aims to support the healthy social development of youngsters at the community level, taking into account this complexity (Fagan et al. 2019). It is a manualized system, which seeks to develop and transform prevention work within communities to address alcohol, drug use and delinquency and also other problem behaviours. CTC mobilizes and empowers coalitions of diverse community stakeholders to collaborate in community assessment, planning and action and to implement and institutionalize science-based prevention service systems. The premise of CTC is that a reduction in the prevalence of adolescent alcohol and drug use, delinquency and other problem behaviours in a community can be achieved through the identification of elevated risk factors and depressed protective factors. It addresses those risk factors found in scientific studies which have been known to increase the likelihood

1 Introduction

9

of adolescent substance abuse, including the consumption of alcohol, cigarettes, cannabis and hard drugs, violence and delinquency, sexual-related problem behaviour and depression. Yet, it also addresses protective factors which help to reduce the likelihood of these outcomes. Based on this knowledge, the CTC process involves assessing the prevalence of the above-mentioned problem behaviours. CTC also relates both risk and protective factors within a particular community to the identified problem behaviours. Based on this local profile, communities can identify and implement tested and effective preventive interventions to address the underlying factors. A strategic, community-specific process has been designed to increase communication, collaboration and ownership among service providers and community members. During this process, communities receive technical assistance and specific training courses from licensed CTC experts. Although it remains a community intervention where different parties bear different responsibilities, one person will be assigned as the local project leader who has specific responsibilities during the 3-year implementation period. After the implementation period, the community will be strong enough to stand on its own legs, still using the CTC prevention framework (Hawkins and Catalano 1992; Ince et al. 2005; Jonkman 2012). The CTC prevention strategy can be summarized as follows. First, all residents of a particular city, community or neighbourhood, as well as all of those involved in the upbringing and development of the young, will be mobilized. The second step aims to create a common vision and language, and to set up a coherent planning structure, which combines all the different area-specific efforts in order to secure a safe future for the young people. This is followed by a prioritization of efforts based on scientific research in regard to risk and protective factors. Next, clear and quantifiable results are analysed and defined, which can be followed up over time. Subsequently, gaps and overlaps within the selection of programmes for youngsters are identified. At the next stage, effective and promising programmes will be deployed. Finally, the development of youth will be monitored and assessed. The starting point of the approach is the analysis of the situation and problems within a city or neighbourhood. The CTC Youth Survey (CTC-YS) enables cities and communities (municipalities and neighbourhoods) to chart the development of both young people and the quality of their living environment (Arthur et al. 2002). With the help of these insights, municipalities and neighbourhoods can get a firm grip on the development of their young people, and they will be able to follow this development over time. This local epidemiological tool enables them to launch a systematic campaign for the improvement of the social and educational environment. Furthermore, these insights help clarify which communities are eligible for the deployment of effective programmes. CTC is a community-orientated prevention strategy. In order to prevent something, you must have significant insight about its root causes. The CTC approach is based on a theoretically and empirically grounded model of risk and protective factors, related to the origins of young people’s problematic behaviour (Catalano and Hawkins 1996; Hawkins and Weis 1985). This model provides the basis for the development of a precautionary approach.

10

H. Jonkman et al.

Over a long period of time, different parties will consistently cooperate to reduce problem behaviour in a specific city or community. There are four core elements that characterize this intervention (Jonkman et al. 2008): 1. The use of similar implementation processes. The implementation of CTC is a process, which takes place over a long period of time, where at specific moments, specific targets are reached and necessary steps are taken for the successful implementation of the CTC Prevention Support System at the local level. Special training sessions and technical assistance are delivered to the community, and specially developed tools and important scientific concepts of prevention are transferred to communities (Ince et al. 2005; Jonkman et al. 2006); see www. communitiesthatcare.net. 2. The use of epidemiological data. Actual research on the distribution and determinants of the health and behaviour of youngsters is important for the improvement of their environment and lives. Analyses of prevalence, the social contexts in which they grow up, their development over the years and the use of risk and protective factors are all important. The use of epidemiological data is essential for CTC. 3. The use of promising and effective programmes. When the situation of a community has been mapped out and the risk and protective factors have been prioritized, it is important to vigorously tackle them. Within CTC, this is achieved by the implementation of tested and effective programmes. CTC provides guidelines about which programmes should be deployed where, when and how, in order to support the healthy and social development of children and adolescents. A guide to tested and effective programmes (e.g. Groeger-Roth and Hasenpusch 2011; Ince et al. 2005) gives communities an overview identifying which programme is suitable for each domain (family, school, individual/peer, community), different ages (0–4 years, 4–12 years and 12–18 years) and different risk and protective factors; see Blueprints for Healthy Youth Development: www. blueprintsprograms.org. 4. The use of ongoing evaluation of results. The effects of preventive interventions must be made clear over a long period of time. Here, the epidemiological data of the CTC-YS are used as well. By routinely administering such tests (e.g. once in every 3 years), the development of problem behaviour and changes in risk and protective factors can be tracked. At the same time, the operation of individual programmes and the total programme supply in communities is clarified as well.

1.1.7  The CTC Youth Survey (CTC-YS) Politicians ask more and more which policy, programme or intervention shows results, what are their conditions and what should be done to disseminate them on a broader scale. Social policy and decision making is an important topic on different levels. On the local and city level, politicians ask, for example, what is the level of

1 Introduction

11

antisocial behaviour among youngsters about which inhabitants complain, which areas should we target and what are the results we can expect from prevention strategies? Politicians, for example, might be confronted with a high level of binge drinking among youngsters. Questions which are important for them are as follows: Is binge drinking highly prevalent among youngsters in specific areas? Is it a typical national problem when we compare it with other countries? How can we explain this binge drinking, and what can be done to decrease the problematic drinking behaviour of youngsters? Alcohol and cannabis use among youngsters is high in Western countries nowadays, especially in Europe. European politicians, for example, ask themselves what are the similarities and differences between several countries, how to explain this high level of substance use and what can be done internationally and at an early stage to reduce this behaviour. Practitioners want to work with instruments and tools which improve their work. Their work has to be professionalized and they want to show that their work is important. For researchers, good analytic research is also important because good research adds to the knowledge of social programmes, informs social action and improves social conditions and human development. With the use of good instruments, researchers can define the problem, the seriousness of the problem and the location of the problem. For everyone, it is important that the policy, intervention and action target the right population. For researchers, it is important to know the intermediate as well as the final outcomes and whether the costs of the intervention are reasonable in relation to its benefits. Their work should be placed in ‘a tradition that has aspired to improve the quality of our physical and social environments and enhance our individual and collective well-being through the systematic creation and application of knowledge’ (Rossi et al. 2004, p. 2). Supporting families, schools and communities in bringing up children and adolescents in this time frame is an important scientific topic, especially when this topic is tackled in the context of communities. Conceptualizing, implementing and evaluating co-ordinated prevention programming is seen as a promising perspective (Nation et  al. 2003; Wandersman and Florin 2003). CTC is such a co-ordinated community perspective that is useful for politicians at different levels and for practitioners as well as researchers. CTC is applied to communities which suffer from the burden of problem behaviours of youngsters, want to prevent the development of problem behaviour of youngsters or want to build up strategic youth policy in their area. CTC is targeted on adolescent youth but also tackles this age group with early interventions. CTC combines the analyses of community problems and strengths, effective prevention programming, concerns for collaboration and monitoring of results. These elements can be seen as the preventive input. With the use of the CTC instruments, coaching and the local implementation plan, communities themselves work out their community prevention strategy. The aim is to decrease risk factors and increase protective factors in families, schools, peer groups and communities. These factors can be seen as intermediate outcomes. Decreasing risk factors and increasing protective factors should, in the end, reduce alcohol use (and other substance use) and

12

H. Jonkman et al.

antisocial behaviour, but also have a significant influence on school dropout, sexually related problem behaviour and depression. The CTC-YS assesses problem behaviours and risk and protective factors in a community (see Arthur et al. 2002, 2007). The latest (2014) version of the CTC-YS is reproduced in our Appendix, and it can be freely downloaded from www.communitiesthatcare.net. We are very grateful for David Hawkins for giving us permission to reproduce this. Users of the CTC-YS are encouraged to also use the CTC-YS Item Construct Dictionary, which can also be freely downloaded from the Internet.

1.2  Aims and Outline of This Book This book brings together the findings of scientists and practitioners who work on CTC in different countries in Europe. They have all worked, often over a long time, on this community strategy. In their work, they have all used the CTC-YS and they have compared their results from this survey. In Chap. 2, Farrington, Utting and Axford present the results of the large nationally representative CTC-YS conducted in Great Britain (England, Wales and Scotland), with over 14,500 students. GB was the first country in Europe where the CTC-YS was used. This chapter shows how self-reported offending rates and the prevalence of substance use are relatively high in GB in the European context. In relation to these problem behaviours, the relationships of the peer and individual factors were the strongest. The authors note that the relationship of protective factors to problem behaviours was weaker. These findings are in line with the results discovered in other European contexts. The Netherlands has a relatively long history of CTC studies conducted since the beginning of the 2000s. The data from the Netherlands in Chap. 3 (Jonkman and Hosman) were collected as part of an effectiveness study, and the CTC-YS was completed in five middle-sized cities located in the South and South-West of the Netherlands. Not surprisingly, the results show that boys tend to commit more violent and delinquent acts than girls and that older adolescents commit these acts more often than younger ones. Jonkman and Hosman researched violence and delinquency in a comprehensive perspective and show how these behaviours are correlated with other problem behaviours and with risk factors and protective factors. They try to define the predictive power of risk and protective factors and propose adequate indicators for this. They also show how this knowledge can be used to develop strategic targets of public health policy, crime prevention and social policy. Chapter 4 is written by Groeger-Roth and Hasenpusch, and their data were collected in three sites in Lower Saxony (Germany). Their analyses show how the probability that an adolescent commits violence or delinquency or drinks heavily grows as the number of risk factors increases. According to this study, risk and protective factors had the strongest relation with delinquency, out of the different types of problem behaviours analysed. Surprisingly, the analysis shows that, in

1 Introduction

13

communities in the case of binge drinking, the level of protection has a small impact for most adolescents. In Chap. 5, Bašić, Novak and Mihić present and analyse data collected in Croatia in a crime prevention project supported by the Croatian Ministry of the Interior (MUP) and the United Nations Development Programme (UNDP). In an effort to build local prevention structures, the CTC-YS was completed in two different types of areas. Their analysis is theoretically driven. Based on earlier studies, the problem behaviours are classified as externalizing or internalizing. According to the results, adolescents from mixed communities tend to have more serious externalizing problem behaviours, including the use of alcohol and violence, and also more internalizing depressive symptoms. Their analysis shows that Belief in the Moral Order predicts less externalizing and internalizing problem behaviours among adolescents in Croatian communities. In Cyprus, the CTC-YS was completed in three junior secondary schools in the capital of Nicosia. In Cyprus, epidemiological research in the field of criminology is still quite rare, and there exist inconsistent results regarding the current frequency of youth delinquency. The biggest concern about juvenile delinquency is the number of serious juvenile cases. The analyses of Kapardis, Spanoudis, Kapardis and Konstantinou in Chap. 6 show that, in Cypriot communities, where multiple significant societal changes have taken place in recent decades, the most important protective factor is a family with certain positive features. However, at the same time, the research findings suggest that investing only in protective factors is not sufficient to prevent delinquency in Cyprus. In Chap. 7 (Comparison of Countries), we present the scientific properties of the CTC-YS, followed by the comparison of the results, in the different countries. Here we compare the prevalences of different problem behaviours in different countries, we look at gender and age differences, and we compare the associations between risk and protective factors and the problem behaviours. We look at differences and similarities between the countries and the consequences for the CTC-YS in the future. Chapter 8 (Conclusions), we present the integrated results from the forgoing chapters, mention several limitations of this comparative study and discuss the implications for practice, prevention science and future research. Some key issues are what do we know about delinquency and substance abuse, what does cross-­ national comparative research tell us, and how should we research risk and protective factors in the future?

1.3  Conclusions Here is a summary of the main conclusions of the CTC-YS studies:

14

H. Jonkman et al.

1.3.1  The CTC-YS in the European Context • The CTC-YS studies presented in this book are based altogether on over 25,000 valid surveys collected in five countries from different kind of communities. • The countries included are Great Britain, the Netherlands, Germany, Croatia and Cyprus. • The CTC-YS that was implemented in different European contexts was culturally adapted. • In the European context, the surveys were implemented with somewhat shorter version of the CTC-YS than the latest US version. • The prevalence of some, mainly the same, problem behaviours was examined in each country. These behaviours included violent and delinquent behaviour, the use of substances (typically of alcohol and some drugs) and depressive symptoms. • The ability of the CTC-YS as a prevention survey tool to examine both externalizing and internalizing problems and possible influences on them may be considered unique. • Problem behaviour was examined in country reports in relation to the communities where the CTC-YS study has been conducted. • Differences in problem behaviour and between genders and age groups were examined. • In the Netherlands and Germany, cut-off points were used to organize the data. • The Netherlands chapter shows how the population attributable fraction (PAF) statistical tool can be used to show what kind of reduction in the prevalence of problem behaviour might occur if adolescents were not exposed to some specific risk factor. Likewise, the PAF can show how much problem behaviour might be reduced by the increase in some specific protective factors.

1.3.2  Problem Behaviours 1.3.2.1  Violent and Delinquent Acts • In the European context, both violent and delinquent acts seem to increase with age, but the increase in delinquent acts is greater. • Boys tended to commit more violent and delinquent acts than girls, but the gender difference in delinquent acts is not as big as in violent acts. • The most common type of delinquent act in the data is stealing, and the most common type of violent act is being involved in a fight. • In Germany in the younger age group, the prevalence of violence was the second highest after depression, but in the older age group, the prevalence of both violent and delinquent acts decreased to the bottom of the frequency list of all examined behaviours.

1 Introduction

15

• The increase in violent acts with age in the GB data is more significant than in the data collected in other European contexts. 1.3.2.2  Alcohol Use • According to CTC-YS studies conducted in Europe, the use of alcohol is very common among adolescents in Europe. • The use of alcohol does not necessarily mean binge drinking. • However, with age both the use of alcohol and binge drinking increase significantly. • Differences between genders are not typically large in the use of alcohol or in binge drinking in European communities. However, with the exception of GB, the prevalence of drinking is higher in all age groups among boys. • The high prevalence of drinking in rural contexts is typical in European CTC-YS studies. • At the same time as the social norms may encourage or even push young people to use alcohol, they may be protective against drug use. 1.3.2.3  Use of Drugs • • • •

In most of the European communities, the use of illegal drugs is rare. The use of cannabis is more common than the use of other drugs. Typically, older adolescents use drugs more than younger ones. Likewise, in most European contexts, the use of drugs is more common among boys. • However, in GB the prevalence of drug use is relatively high, and gender differences in some cases “favour” girls. 1.3.2.4  Depressive Symptoms • In most of the European countries, the prevalence of depressive symptoms is high among adolescents. • It can be concluded that depressive symptoms seem to be more common among older adolescents, but the differences between age groups are not huge. • In all European contexts, depressive symptoms are significantly more common among girls. • When interpreting the results, it should be noted that the CTC-YS is not a diagnostic tool and that the findings only provide recommendations about the need for prevention related to mental health in populations.

16

H. Jonkman et al.

1.3.3  R  isk and Protective Factors Influencing Problem Behaviours 1.3.3.1  Risk Factors • Peer and individual risk factors are the most influential in relation to externalizing problem behaviour including substance use and violent and delinquent acts. • The dominance of peer and individual risk factors may be partly explained because antisocial attitudes, early problem behaviour and peer deviance may possibly all reflect the same underlying theoretical construct as delinquency, at least to some extent. • Risk factors in the family domain had comparatively strong relationships with violence. • The dominance of peer and individual factors was more evident in relation to delinquency than in relation to violence. • Risk factors have different effects on different types of problem behaviour. 1.3.3.2  Protective Factors • The effect of protective factors is much weaker compared to the effect of risk factors. • Out of all types of externalizing behaviour, protective factors have the strongest effect on delinquency. • Family protective factors have the strongest effect in relation to delinquent and violent behaviour. • The need to develop CTC-YS studies in European contexts is particularly evident when it comes to protective factors.

1.3.4  D  ifferences Between Communities in Problem Behaviours • In Croatia, the type of community was significantly related to all externalizing problem behaviours examined, which occurred more in rural/urban settings than in strictly urban contexts. • More comparative information, comparing different kinds of communities, is needed about adolescent problem behaviours and risk and protective factors influencing them.

1 Introduction

17

1.3.5  Final conclusions • Risk factors tend to predict a high rate of problem behaviours, whereas protective factors tend to predict a low rate. • It is important to relate CTC-YS results to gender and age differences. • The results show the overall usefulness of the CTC-YS in Europe. • It would be desirable to use the CTC-YS repeatedly with national samples of youth in different countries to assess changes in problem behaviours and in risk and protective factors that influence them.

References Aebi, M.  F., Akdeniz, G., Barclay, G., Campistol, C., Caneppele, S., Gruszczyńska, B., et  al. (2014). European sourcebook of crime and criminal justice statistics. Helsinki: European institute for crime prevention and control (HEUNI). Arthur, M.  W., Briney, J.  S., Hawkins, J.  D., Abbott, R.  D., Brooke-Weiss, B.  L., & Catalano, R. F. (2007). Measuring risk and protection in communities using the communities that care youth survey. Evaluation and Program Planning, 30, 197–211. Arthur, M. W., Hawkins, J. D., Pollard, J. A., Catalano, R. F., & Baglioni, A. J. J. (2002). Measuring risk and protective factors for substance use, delinquency, and other adolescent problem behaviors: The communities that care youth survey. Evaluation Review, 26, 575–601. Axford, N., Sonthalia, S., Wrigley, Z., Webb, L., Mokhtar, N., Brook, L., et al. (2016). What works in Europe? Developing a European communities that care database of effective prevention programmes. Dartington: Dartington Social Research Unit. Berkman, L.  F., & Kawachi, I. (Eds.). (2000). Social epidemiology. New  York: Oxford University Press. Boutellier, H. (2015). Het seculiere experiment. Meppel: Boom. Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University Press. Catalano, R. F., Berglund, M. L., Ryan, J. A. M., Lonczak, H. S., & Hawkins, J. D. (2004). Positive youth development in the United States: Research findings on evaluations of positive youth development programs. Annals of the American Academy of Political and Social Science, 31, 98–124. Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of antisocial behavior. In J.  D. Hawkins (Ed.), Delinquency and crime: Current theories (pp.  149–197). New York: Cambridge University Press. Coie, J. D., Watt, N. F., West, S. G., Hawkins, J. D., Asarnow, J. R., Markman, H. J., et al. (1993). The science of prevention: A conceptual framework and some directions for a national research program. American Psychologist, 48, 1013–1022. Damon, W. (1990). The moral child: Nurturing children’s natural moral growth. New York: The Free Press. Damon, W. (1997). The youth charter. How communities can work together to raise standards for all our children. New York: The Free Press. Damon, W. (2008). The path to purpose: How young people find their calling in life. New York: The Free Press. Dewey, J. (1907). The school and society. Chicago: University of Chicago Press. Dryfoos, J. G. (1998). Safe passage: Making it through adolescence in a risky society. New York: Oxford University Press.

18

H. Jonkman et al.

Elder, G. H., & Conger, R. D. (2000). Children of the land: Adversity and success in rural America. Chicago: University of Chicago press. Elliott, D.  S. (1997). Blueprints for violence prevention. Boulder, CO: Centre for the Study of Prevention of Violence, Institute of Behavioral Science, University of Colorado. Erikson, E. H. (1987). A way of looking at things: Selected papers of Erik H. Erikson 1930-1980. New York: W.W. Norton & Company. Fagan, A. A., Hawkins, J. D., Catalano, R. F., & Farrington, D. P. (2019). Communities that care: Building community engagement and capacity to prevent youth behavior problems. Oxford: Oxford University Press. Farrington, D. P. (2002a). Key results from the first forty years of the Cambridge study in delinquent development. In T. P. Thornberry & M. D. Krohn (Eds.), Taking stock of delinquency (pp. 137–184). New York: Kluwer Academic/Plenum. Farrington, D. P. (2002b). Multiple risk factors for multiple problem violent boys. In R. R. Corrado, R.  Roesch, S.  D. Hart, & J.  K. Gierowski (Eds.), Multi-problem violent youth (pp.  23–34). Amsterdam: IOS Press. Farrington, D. P. (2004). Conduct disorder, aggression and delinquency. In R. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology (2nd ed., pp. 627–664). Hoboken, NJ: Wiley. Farrington, D.  P. (2015). Cross-national comparative research on criminal careers, risk factors, crime and punishment. European Journal of Criminology, 12(4), 386–399. Farrington, D.  P., & Wikström, P.-O.  H. (1994). Criminal careers in London and Stockholm: A  cross-national comparative study. In E.  G. M.  Weitekamp & H.-J.  Kerner (Eds.), Cross-­ national longitudinal research on human development and criminal behaviour (pp.  65–89). Dordrecht: Kluwer. Felson, R.  B., Savolainen, J., Bjarnason, T., Anderson, A.  L., & Zohra, T. (2011). The cultural context of adolescent drinking and violence in 30 European Countries. Criminology, 49(3), 699–728. Furstenberg, F. F., Cook, T. D., Eccles, J., Elder, F., & Sameroff, A. (1999). Managing to make it: Urban families and adolescent outcome. Chicago: University of Chicago Press. Galea, S. (Ed.). (2007). Macrosocial determinants of population health. New York: Springer. Gardner, H. (2006). Multiple intelligence: New horizons. New York: Basic Books. Gatti, U., Rocca, G., & Verdem, V. (2013). Delinquency, vctimization and alcohol involvement. In M. Steketee, H. Jonkman, H. Bertel, & N. Vettenburg (Eds.), Alcohol use among adolescents in Europe: Environmental research and preventive actions (pp.  139–154). Utrecht: Verwey-­ Jonker Institute. Gatti, U., Soellner, R., Bräker, A.-B., Verde, A., & Rocca, G. (2015). Delinquency and alcohol use among adolescents in Europe: The role of cultural context. European Journal of Criminology, 12, 362–377. Goldberg, E. (2001). The executive brain: Frontal lobes and the civilized mind. New York: Oxford University Press. Gottfredson, M.  R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press. Groeger-Roth, F., & Hasenpusch, B. (2011). Green list prevention: Inclusion- and rating-criteria for the CTC programme – Databank. Hannover: Crime Prevention Council of Lower Saxony. Habermas, J. (1981). Theorie des kommunikativen Handelns. Frankfurt am Main: Suhrkampf Verlag. Hawkins, J. D., & Catalano, R. F. (1992). Communities that care: Action for drug abuse prevention. San Francisco: Jossey-Bass. Hawkins, J. D., Catalano, R. F., Arthur, M. W., Egam, E., Brown, E. C., Abbott, R. D., et al. (2008). Testing communities that care: The rationale, design and behavioral baseline equivalence of the community youth development study. Prevention Science, 9(3), 178–190. Hawkins, J.  D., Herrenkohl, T., Farrington, D.  P., Brewer, D., Catalano, R.  F., & Harachi, T. W. (1998). A review of predictors of youth violence. In R. Loeber & D. P. Farrington (Eds.), Serious and violent juvenile offenders: Risk factors and successful interventions (pp. 106–146). Thousand Oaks, CA: Sage.

1 Introduction

19

Hawkins, J. D., & Weis, J. G. (1985). The social development model: An integrated approach to delinquency prevention. Journal of Primary Prevention, 6(2), 73–97. Ince, D., Beumer, M., Jonkman, H., & Vergeer, M. (2005). Veelbelovend en effectief: Overzicht van preventieprojecten en -programma's in de domeinen gezin, school, kinderen en jongeren, wijk. Amsterdam: SWP. Institute of Medicine. (2009). Report on preventing mental, emotional and behavioral disorders among young people: Progress and possibilities. Washington, DC: Institute of Medicine. Järvinen, M., & Room, R. (2007). Youth drinking cultures: European experiences. Aldershot: Ashgate. Jonkman, H.(2012). Some years of communities that care. Learning from a social experiment. Vrije University: Amsterdam. Jonkman, H., Boers, R., Dijk, B. V., & Rietveld, M. (2006). Wijken gewogen. gedrag van jongeren in kaart gebracht. Utrecht/Amsterdam: NIZW/SWP. Jonkman, H., Haggerty, K., Steketee, M., Fagan, A., Hanson, K., & Hawkins, J.  D. (2008). Communities that care, core elements and context: Research of implementation in two countries. Social Development Issues, 30(3), 42–58. Junger-Tas, J. (1996). Jeugd en gezin: Preventie vanuit een justitieel perspectief. Den Haag: Ministerie van Justitie. Junger-Tas, J. (1997). Jeugd en gezin: II. Preventie vanuit een justitieel perspectief. Den Haag: Ministerie van Justitie. Junger-Tas, J. (2010). Delinquent behaviour in thirty countries. In J.  Junger-Tas, I.  Marshall, D.  Enzmann, M.  Killas, M.  Steketee, & B.  Gruszczyńska (Eds.), Juvenile delinquency in Europe and beyond: Results of the second international self-report delinquency study. New York: Springer. Katainen, A., & Rolando, S. (2015). Adolescents’ understandings of binge drinking in Southern and Northern European contexts: Cultural variations of ‘controlled loss of control’. Journal of Youth Studies, 18(2), 151–166. https://doi.org/10.1080/13676261.2014.933200 Kawachi, I., & Berkman, L.  F. (2003). Neighbourhoods and health. New  York: Oxford University Press. Keating, D. P., & Hertzman, C. (1999). Developmental health and the wealth of nations: Social, biological and educative dynamics. New York: Guilford Press. Kellam, S. G., & Rebok, G. W. (1992). Building developmental and etiological theory through epidemiologically based preventive intervention trials. In J. McCord & R. E. Tremblay (Eds.), Preventing antisocial behavior: Interventions from birth through adolescence (pp. 162–195). New York: Guilford Press. Kessler, R.  C., Berglund, P., Demler, O., Jin, R., Merikangas, K.  R., & Walters, E.  F. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62(6), 593–602. Loeber, R., & Farrington, D. P. (Eds.). (1998). Serious and violent juvenile offenders: Risk factors and successful interventions. Thousand Oaks, CA: Sage. Loeber, R., Farrington, D. P., Stouthamer-Loeber, M., & Van Kammen, W. B. (1998). Multiple risk factors for multi-problem boys: Co-occurrence of delinquency, substance use, attention deficit, conduct problems, physical aggression, covert behavior, depressed mood and shy/withdrawn behavior. In R. Jessor (Ed.), New perspectives on adolescent risk behavior (pp. 90–149). New York: Cambridge University Press. Loeber, R., Slot, N.  W., Laan, P.  V. D., & Hoeve, M. (Eds.). (2008). Tomorrow's criminals. The development of child delinquency and effective interventions. Farnham: Ashgate. Luke, D. (2004). Multilevel modeling. Thousand Oaks, CA: Sage. Marmot, M. (2000). Multilevel approaches to understanding social determinants. In L. F. Berkman & I. Kawachi (Eds.), Social epidemiology (pp. 349–367). New York: Oxford University Press. Mead, G. H. (1934). Mind, self and society from the standpoint of a social behaviorist. Chicago: University of Chicago Press.

20

H. Jonkman et al.

Measham, F., & Brain, K. (2005). Binge drinking, British alcohol policy and the new culture of intoxication. Crime, Media, Culture, 1(3), 262–283. https://doi.org/10.1177/1741659005057641 Measham, F., & Shiner, M. (2009). The legacy of ‘normalisation’: The role of classical and contemporary criminological theory in understanding young people’s drug use. International Journal of Drug Policy, 20(6), 502–503. https://doi.org/10.1016/j.drugpo.2009.02.001 Mrazek, P. J., & Haggerty, R. J. (Eds.). (1994). Reducing risks for mental disorders: Frontiers for preventive intervention research. Washington, DC: National Academy Press. Nation, M., Crusto, C., Wandersman, A., Kumpfer, K. L., Seybolt, D., Morrissey-Kane, E., et al. (2003). What works in prevention: Principles of effective prevention programs. American Psychologist, 58(6/7), 449–457. Phelps, E., Furstenberg, F. F., & Colby, A. (Eds.). (2002). Looking at lives: American longitudinal studies of the twentieth century. New York: Russell Sage Foundation. Pinkster, F. M. (2009). Living in concentrated poverty. Amsterdam: University of Amsterdam. Pollard, J. A., Hawkins, J. D., & Arthur, M. A. (1999). Risk and protection: Are both necessary to understand diverse behavioral outcomes in adolescence? Social Work Research, 23(3), 145–158. Rose, G. (1992). The strategy of preventive medicine. Oxford, UK: Oxford University Press. Rossi, P. H., Lipsey, M. W., & Freeman, H. E. (2004). Evaluation: A systematic approach (7th ed.). Thousand Oaks, CA: Sage. Rutter, M. (1979). Fifteen thousand hours: Secondary schools and their effect on children. Cambridge, MA: Harvard University Press. Sampson, R.  J. (2011). Great American city: Chicago and the enduring neighborhood effect. Chicago: University of Chicago Press. Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press. Sampson, R. J., & Laub, J. H. (2003). Shared beginnings, divergent lives: Delinquent boys to age 70. Cambridge, MA: Harvard University Press. Sampson, R.  J., Raudenbusch, S.  W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 914–918. Sherman, L. W., Gottfredson, D., MacKenzie, D., Eck, J., Reuter, P., & Bushway, S. (Eds.). (1997). Preventing crime: What works, what doesn’t, what is promising: A report to the United States congress. College Park, MD: University of Maryland Department of Criminology and Criminal Justice. Steketee, M., Jonkman, H., Mak, J., Aussems, C., Huygen, A., & Roeleveld, W. (2012). Communities that care in Nederlandse steden. Utrecht: Verwey-Jonker Instituut. Thaler, R., & Sunstein, C. (2009). Nudge: Naar betere beslissingen over gezondheid, geluk en welvaart. Amsterdam: Business Contact. Tremblay, R. E. (1999). When children’s social development fails. In D. P. Keating & C. Hertzman (Eds.), Developmental health and the wealth of nations: Social, biological and educational dynamics (pp. 55–71). New York: Guilford Press. Viner, R.  M., Ozer, E.  M., Denny, S., Marmot, M., Resnick, M., Fatusi, A., et  al. (2012). Adolescence and the social determinants of health. The Lancet, 379(9826), 1641–1652. https:// doi.org/10.1016/s0140-6736(12)60149-4 Vygotsky, L. S. (1978). Mind in society: The psychology of higher mental functions. Cambridge, MA: Harvard University Press. Wandersman, A., & Florin, P. (2003). Community interventions and effective prevention. American Psychologist, 58(6/7), 441–449. Weissberg, R. P., Kumpfer, K. L., & Seligman, M. E. P. (2003). Prevention that works for children and youth: An introduction. American Psychologist, 58(6/7), 425–432. West, D. J., & Farrington, D. P. (1977). The delinquent way of life. London: Heinemann.

Chapter 2

Great Britain David P. Farrington, David Utting, and Nick Axford

2.1  Introduction Communities That Care (CTC) is a method of preventing offending, drug use and other types of problem behavior (Fagan et al. 2019; Hawkins et al. 2002). It is based on risk-focused prevention (see, e.g., Farrington 2000, 2020). In this type of prevention, risk factors are identified and reduced, while protective factors are identified and enhanced. However, most knowledge about effective interventions is based on high-quality evaluations carried out in the USA. It is important to investigate the extent to which these interventions are effective in the UK and Europe more broadly (Axford et  al. 2016). In the UK, the Early Intervention Foundation Guidebook (https://guidebook.eif.org.uk) provides information about over 100 early intervention programs that have been evaluated and, in many cases, shown to improve outcomes for children and young people. The CTC Youth Survey (CTC-YS) was devised to measure risk and protective factors and problem behavior by young people (Arthur et al. 2007; Monahan et al. 2014). Based on these data, CTC helps communities to identify priority risk and protective factors and to select and implement tested and effective prevention programs and policies that address those factors and that fill gaps in existing resources. Communities monitor and evaluate these interventions, measuring results and tracking progress to ensure that improvements are achieved. D. P. Farrington (*) Institute of Criminology, University of Cambridge, Cambridge, UK e-mail: [email protected] D. Utting Independent Writer and Researcher, St Albans, UK e-mail: [email protected] N. Axford Peninsula Medical School, University of Plymouth, Plymouth, UK e-mail: [email protected] © Springer Nature Switzerland AG 2021 D. P. Farrington et al. (eds.), Delinquency and Substance Use in Europe, https://doi.org/10.1007/978-3-030-58442-9_2

21

22

D. P. Farrington et al.

In this chapter, we describe the results of applying the CTC-YS to a large representative national sample of secondary school students (aged 11–16) in Great Britain, which is the island consisting of England, Wales, and Scotland (see Beinart et al. 2002). The survey was funded by the Joseph Rowntree Foundation (JRF) and undertaken by Communities That Care Ltd. It was primarily intended to provide national comparison data for local data that was being gathered by communities which were implementing the CTC approach and using the same questionnaire to survey local secondary school students. David Farrington was an advisor to the survey and David Utting was one of the primary authors of the report. When the results were first reported, there were plans—never realized—for them to provide the baseline for a series of regularly repeated, national surveys using the same questionnaire. The report’s authors also expressed a hope that the survey data would be further analyzed to advance knowledge about the effects of risk and protective factors in the lives of children and young people (Beinart et al. 2002, p. 48). The main aim of this chapter is to report on the prevalence of different types of problem behavior and on the relationships between risk and protective factors and problem behavior. The previous report by Beinart et al. (2002) is not easily obtainable nowadays. Also, in the original report, the results were mainly presented graphically, whereas we present exact figures which might be useful in future systematic reviews and meta-analyses.

2.1.1  The Development of CTC in the UK The development of CTC in the UK began in 1995 when David Farrington was commissioned by JRF to advise them about how best to reduce youth crime. His report, entitled “Understanding and Preventing Youth Crime,” was published by JRF in 1996 (Farrington 1996), and it recommended that CTC should be implemented and evaluated in the UK. JRF was supportive of this, and set up a Study Group on CTC, which included David Farrington and David Utting, which met during 1996–1997. Farrington (1997) then published an article recommending how to evaluate CTC in the UK.  He proposed that CTC should be implemented in three experimental communities and that each of them should be compared with a similar control community. He also recommended that there should be before and after surveys in secondary schools and households to measure risk and protective factors and offending outcomes. This classic design did not meet with universal approval (see Farrington 1998; Pawson and Tilley 1998a, 1998b). However, JRF implemented the main features of the design in three towns of England and Wales in 1998–2003. There were before and after secondary school surveys (but not household surveys) in experimental and control areas in 1999–2000 and 2002, and the before surveys (based on the CTC-YS) were used to inform the choice of interventions. David Farrington was not involved in this evaluation, but he continued to advocate the use of CTC, for example, in his

2  Great Britain

23

1999 Presidential Address to the American Society of Criminology (Farrington 2000). Also, the CTC team published a review, commissioned by the UK Youth Justice Board, of risk and protective factors for youth crime and effective interventions (Anderson et al. 2001). Unfortunately, the evaluation of CTC by Sheffield University researchers encountered many problems. In particular, because the experimental and control schools were drawn from the same cities, many children in experimental schools did not live in CTC areas, and many children in control schools did live in CTC areas. Therefore, the researchers decided to abandon the control schools and compare children in experimental schools who lived in CTC areas with children in the same schools who did not live in CTC areas. This design obviously raised problems of contamination of control children by experimental children. There were many other problems. Most selected interventions focused on the parents of young children, but the CTC approach was evaluated by surveying secondary school students. Also, CTC was poorly implemented in the two English towns. For example, in “Northside,” the coordinator left and was not replaced for 6 months, and only three interventions were implemented. CTC was adequately implemented only in the one Welsh town, and the results were generally encouraging; 14 out of 20 tests showed desirable effects of CTC on risk factors, especially community and family factors, in response to the community and family interventions (Crow et al. 2004, p. 31). Despite these not wholly positive results, CTC was soon expanded to over 20 sites in England, Wales, and Scotland. However, sites found it difficult to prioritize risk and protective factors for interventions, because they had no national comparison data to tell them whether their area was unusually high or low on particular factors. This was the main reason why JRF decided to fund a national survey in 2000–2001. Communities That Care Ltd. was established to support the introduction of CTC in the UK, and it organized the national survey. However, it was subsequently de-­ registered and merged with the Rainer Foundation, which became part of another nonprofit organization called Catch22, which aimed to build resilience in people and communities. A subsequent initiative saw CTC combined with another approach to planning preventive services, Common Language (devised by the Dartington Social Research Unit, now called the Dartington Service Design Lab), to form Evidence2Success, which was tested in Scotland (Utting 2016).

2.1.2  UK National Self-Reported Offending Surveys The first national self-reported offending survey in the UK was carried out by David Farrington in 1981 (see Farrington 2001). In this, a nationally representative household sample of 433 persons aged 17–64 was interviewed and asked about their offending. This was intended to inform the design of the first British Crime Survey in 1981, which focused on victimization (see Mayhew and Elliott 1990, p. 94). It

24

D. P. Farrington et al.

was soon followed by the first national self-reported delinquency survey in 1983, with 751 persons aged 14–15 (Riley and Shaw 1985). This focused on parental supervision in relation to juvenile delinquency. The next self-reported offending survey in 1992–1993 was larger (1,648 persons) and covered the age range from 14 to 25 (Graham and Bowling 1995). This focused on prevalence, onset, and desistance. An even larger Youth Lifestyles Survey (of 4,848 persons aged 12–30) was carried out in 1998–1999 and was the first to use a computer-assisted self-completion method (Flood-Page et al. 2000). While this survey was mainly concerned to yield national prevalence estimates, it also studied risk factors for serious and/or persistent offending. Next, the Offending, Crime and Justice Survey was carried out annually between 2003 and 2006 with persons aged 10–25 (e.g. Roe and Ashe 2008). This was mainly concerned with offending behavior, and longitudinal analyses were carried out by Hales et al. (2009). None of these national surveys was very comparable to the present CTC-YS, because they were based on household samples (as opposed to school samples) and because they focused mainly on the prevalence of offending rather than on risk and protective factors. The most comparable surveys are the On Track Youth Lifestyles Surveys, which were based on the CTC-YS. These were used to evaluate the effects of the UK Government’s “On Track” initiative, which was implemented and evaluated in 2000–2006 with children aged 4–12 and their families in high-crime, high-­ deprivation areas. Like CTC, On Track targeted risk and protective factors, and two surveys were completed in schools, of about 30,000 students in 2001 (Armstrong et al. 2005) and about 20,000 students in 2004 (Bhabra et al. 2006). However, these were not nationally representative surveys. The aforementioned 1992–1993 national self-reported offending survey in England and Wales was analyzed as part of the first International Self-Report Delinquency Study or ISRD-1 (Bowling et al. 1994). The UK did not participate in the second study (ISRD-2) in 2005–2007 (Junger-Tas et  al. 2010). England and Scotland did participate in the third study (ISRD-3) in 2014–2015, and school students completed the survey either online or on paper. In England, 367 students took part from 11 schools in Birmingham and 533 students took part from eight schools in Sheffield (Herlitz et  al. 2016a). In Scotland, 841 students took part from 10 schools in Glasgow and 445 students took part from 6 schools in Edinburgh (Herlitz et  al. 2016b). Also, a book on Minority Youth and Integration was published by Roche and Hough (2018). However, ISRD-3 was a city-based survey, not a national survey. It is expected that ISRD-4 will collect data in 2021–2022. Farrington (1973) published the first review of the psychometric properties of self-reported delinquency surveys and concluded that they were concurrently and predictively valid. More recently, Gomes et al. (2018) reviewed the history of these surveys, and Jolliffe and Farrington (2014) presented up-to-date information on their reliability and validity. Both concluded that these surveys have been proven to be useful methods of measuring offending. The CTC-YS that is described in this chapter is unique in being based on a very large nationally representative sample of schools and in its extensive measurement of risk and protective factors as well as problem behavior outcomes.

2  Great Britain

25

2.2  Method 2.2.1  The Survey Design In July 2000, Communities That Care Ltd. commissioned Ipsos-RSL, a market research company, to collect data from a random sample of secondary schools in England, Wales, and Scotland during the first two terms of the 2000–2001 school year. Pilot work was first carried out to test the survey materials. Fifteen schools from three local education authorities (LEAs) were approached by three experienced interviewers, and eight schools agreed to participate. This pilot work confirmed that the survey procedures were satisfactory, and the results from these eight schools were included in the final dataset. In the main data collection, 41 LEAs were selected with probability proportional to size, from a list that was stratified by region. Within each selected LEA, five schools were chosen, again with probability proportional to size. Of the total 220 selected secondary schools, two were removed from the list because they had recently participated in other CTC research. Of the remaining 218 schools, 89 participated in the survey, for a response rate of 41%. In each selected school, one class in each of years 7–11 (age 11–12 to age 15–16) was randomly selected to take part in the survey. The self-completion questionnaires were administered in the classroom by the class teacher. In total, 14,666 students completed paper questionnaires. The completed questionnaires were computer-scanned, and the resulting dataset was edited for inconsistencies and inappropriate answers. In an extensive validity check, students were removed if they met any of the following conditions: (a) They claimed to have taken all of the listed drugs in the previous four weeks. (b) They claimed to have taken a fictitious drug (“derbisol”). (c) They gave two or more inconsistent responses to questions about alcohol or smoking. (d) They gave two or more inconsistent responses to questions about illegal drugs. (e) They gave two or more inconsistent responses to questions about age. (f) In total, 221 students (1.5%) were identified by one or more of the above criteria and excluded from the dataset. Thus, there were 14,445 valid survey forms. The data were weighted to correct for unequal student selection probabilities and to correct for nonresponse bias. For example, selective grammar schools were slightly overrepresented (7.9%) in the achieved sample of schools, while nonselective comprehensive schools up to age 18 were slightly underrepresented (55.1%). The weighted sample size was 14,529. All analyses in this chapter are based on the weighted sample.

26

D. P. Farrington et al.

2.2.2  Risk and Protective Factors Family, school, community, and individual or peer risk factors were measured in the CTC-YS, in addition to protective factors in the same domains. The constituent items were worded as follows: (A) Family Risk Factors (A1) Poor parental supervision • • • • • • • •

My family has clear rules about using alcohol and drugs. If I drank alcohol, my parents would catch me. If I played truant, my parents would catch me. The rules in my family are clear. My parents want me to phone if I am late home. My parents ask me about my homework. When I am not at home, my parents know where I am. My parents would know if I was not home on time.

Each item was answered yes or no, and the score was the number of no answers (from 0 to 8), indicating poor parental supervision and discipline. A higher score indicates greater risk. (A2) Family conflict • • • • •

People in my family often insult or yell at each other. People in my family have serious arguments. We argue about the same things over and over again. Adults in my home sometimes try to hurt me. Adults in my home sometimes try to hurt each other.

Each item was answered agree or disagree, and the score was the number of agree answers (from 0 to 5), indicating family conflict. (A3) Family history of problem behavior • • • • • •

Have siblings been excluded? [from school] Have siblings drunk regularly under age 18? Have siblings smoked regularly under age 16? Have siblings smoked cannabis? Have siblings taken other drugs? Has a family member had a serious drug/alcohol problem?

Each item was answered yes or no, and the score was the number of yes answers (from 0 to 6), indicating a family history of problem behavior. (A4) Parental attitudes condoning problem behavior Parental attitudes to: • • • •

Stealing Picking a fight Drawing graffiti Drinking alcohol

2  Great Britain

• • • • •

27

Smoking cigarettes Smoking cannabis Using other drugs Playing truant Getting pregnant or getting a girl pregnant

Each item was answered wrong or not wrong, according to the parent’s attitude, and the score was the number of not wrong answers (from 0 to 9), indicating favorable parental attitudes to problem behavior. (B) School Risk (B1) Low achievement • What are your marks like this year? This was scored on a five-point scale from 0 (very good) to 4 (very poor). (B2) Bullying • Have you bullied other pupils in the past year? This was scored on a five-point scale from 0 (never) to 4 (very often). (B3) Low commitment to school • • • • • •

How interesting are school subjects? How important is schoolwork for later life? How often did you enjoy school? How often did you try your best at school? How often did you hate school? How often did you play truant in the past year?

Each answer was scored as positive or negative commitment, and the score was the number of negative answers (from 0 to 6), indicating low commitment to school. (B4) School disorganization • • • • • •

I feel safe at school. Teachers use punishments to keep control. School has clear rules about being late. School has clear rules about being absent. School has clear rules about bullying. It is easy to play truant from my school.

Each answer was scored as positive or negative to school, and the score was the number of negative answers (from 0 to 6), indicating school disorganization. (C) Community Risk (C1) Community disorganization • • • • •

There are lots of fights in my neighborhood. There is crime and/or drug selling in my neighborhood. There are lots of empty buildings in my neighborhood. There are lots of graffiti in my neighborhood. How safe do you feel in your neighborhood after dark?

28

D. P. Farrington et al.

Each answer was scored as positive or negative to the neighborhood, and the score was the number of negative answers (from 0 to 5), indicating community disorganization. (C2) Availability of drugs How easy is it to get hold of: • • • • •

Alcohol? Cigarettes? Cannabis? Cocaine, LSD, or ecstasy? Heroin?

Each answer was scored as easy or not easy, and the score was the number of easy answers (from 0 to 5), indicating availability or drugs. (D) Individual/Peer Risk (D1) Alienation and low social commitment • • • • • • •

I do the opposite of what people tell me, just to make them mad. I like to see how much I can get away with. It is OK to beat people up if they start the fight. It is OK to take something without asking if you can get away with it. I ignore rules that get in my way. It is sometimes OK to cheat at school. It is important to be honest with your parents.

Each answer was scored as desirable or undesirable, and the score was the number of undesirable answers (from 0 to 7), indicating alienation and low social commitment. (D2) Attitudes condoning problem behavior How wrong is it to: • • • • • • • • • •

Play truant? Drink regularly? Smoke cigarettes? Smoke cannabis? Use LSD? Take a weapon to school? Steal something? Pick a fight? Attack someone? Get self or other pregnant?

Each answer was scored as wrong or not wrong, and the score was the number of not wrong answers (from 0 to 10), indicating favorable attitudes to problem behavior. (D3) Early involvement in problem behavior Age that you were first: • Excluded?

2  Great Britain

• • • • •

29

Smoked cigarette? Arrested? Drank regularly? Sniffed glue? Smoked cannabis?

The score was the number of acts committed by age 13 (0 to 6), indicating early involvement in problem behavior. (D4) Peer involvement in problem behavior In the past year, have your friends: • • • • • • • • • •

Been excluded? Regularly played truant? Tried alcohol? Smoked cigarettes regularly? Smoked cannabis? Used other illegal drugs? Sold or dealt illegal drugs? Carried a weapon to school? Stolen or tried to steal a vehicle? Been arrested?

The score was the number of yes answers (from 0 to 10), indicating peer involvement in problem behavior. (E) Protective Factors (E1) School opportunities for prosocial involvement Do you have lots of chances to: • • • •

Help decide class activities or rules? Talk to a teacher one to one? Be part of class discussions or activities? Get involved in sports/clubs, etc.?

The score was the number of yes answers (from 0 to 4), indicating school opportunities for prosocial involvement. (E2) School rewards for prosocial involvement • Teachers notice when I do well and tell me. • School tells my parents when I do well. • Teachers praise me when I work hard. The score was the number of yes answers (from 0 to 3), indicating school rewards for prosocial involvement. (E3) Family attachment Do you: • Feel very close to your mother? • Share thoughts and feelings with your mother? • Feel very close to your father?

30

D. P. Farrington et al.

• Share thoughts and feelings with your father? The score was the number of yes answers (from 0 to 4), indicating family attachment. (E4) Family opportunities for prosocial involvement • My parents give me chances to have fun with them. • My parents ask me before making family decisions. • I could talk to my parents about a personal problem. The score was the number of yes answers (from 0 to 3), indicating family opportunities for prosocial involvement. (E5) Family rewards for prosocial involvement • • • •

Do you enjoy spending time with your mother? Do you enjoy spending time with your father? How often do your parents show that they are proud of you? How often do your parents notice you doing something well?

The score was the number of yes or often answers (from 0 to 4), indicating family rewards for prosocial involvement.

2.3  Results 2.3.1  The CTC-YS Sample Table 2.1 shows some characteristics of the sample. Just over half of the students (51.8%) were girls. (All analyses in this chapter are based on persons who were known on a variable; thus, 14,285 students were known on gender.) The students were distributed fairly equally over the five school years, but somewhat unequally over age. For example, only 11.9% of students were age 11 because year 7 covered age 11–12, but 19.8% of students were age 12 because they were drawn from year 7 (age 11–12) and year 8 (age 12–13). In regard to ethnic origin, 92% of students were White. The largest ethnic minority groups in the sample were Indian (3%) and Pakistani (1.9%). Most students (80.45%) were surveyed in England, while 9.48% were in Wales and 10.07% were in Scotland.

2.3.2  The Prevalence of Problem Behavior Students were asked whether they had ever stolen anything. Table 2.2 shows that 39% of boys and 32.2% of girls reported doing this. The ever prevalence of stealing increased with age, from 29% of boys aged 11–12 to 46.5% of boys aged 15–16 and

2  Great Britain

31

Table 2.1  CTC youth survey sample Gender Male Female Total Age 11 12 13 14 15 16 Total Ethnic origin White Black Caribbean Black African Black Other Indian Pakistani Bangladeshi Chinese Total Region England Wales Scotland Total

N

%

6882 7403 14,285

48.2 51.8 100

1697 2828 3033 2727 2847 1165 14,297

11.9 19.8 21.2 19.1 19.9 8.1 100

12,899 149 93 110 419 273 29 49 14,021

92.0 1.1 0.7 0.8 3.0 1.9 0.2 0.4 100

11,689 1377 1463 14,529

80.5 9.5 10.1 100

Weighted N = 14,529 Unweighted N = 14,445

from 21.1% of girls aged 11–12 to 39.5% of girls aged 15–16. Students were also asked whether they had ever been arrested, and 11% of boys and 4.5% of girls reported that this had happened to them. Again, the ever prevalence of arrests increased with age, from 5.8% of boys aged 11–12 to 15.2% of boys aged 15–16 and from 2.1% of girls aged 11–12 to 7% of girls aged 15–16. Students were also asked about committing nine specified offenses in the previous year. Table 2.2 shows that vandalism (25.3% of boys and 22.9% of girls), shoplifting (22.7% of boys and 19.6% of girls), and receiving stolen property (22.1% of boys and 15.3% of girls) were the most common types of offenses. In contrast, the least prevalent types of offenses were stealing a vehicle (4.6% of boys and 1.3% of girls), stealing from a vehicle (4.9% of boys and 1.5% of girls), and burglary (6.9% of boys and 2.4% of girls). The prevalence of offending generally increased with age. The biggest difference between boys and girls was in carrying a weapon (20.3% of boys versus 5.4% of girls).

32

D. P. Farrington et al.

Table 2.2  Self-reported offending

Ever stolen Ever been arrested Shoplifted last year Stole vehicle last year Stole from vehicle last year Stole elsewhere last year Burglary last year Receiving last year Vandalism last year Attacked someone last year Carried weapon last year

% of boys All 11–12 39.0 29.0 11.0 5.8 22.7 14.8 4.6 2.1 4.9 2.5 19.9 15.4 6.9 3.4 22.1 12.3 25.3 16.6 15.9 10.2 20.3 14.4

13–14 41.4 12.0 24.6 4.4 5.0 21.8 6.9 23.5 27.0 17.6 21.4

15–16 46.5 15.2 28.4 7.6 7.4 21.9 10.6 30.5 32.1 19.5 25.1

% of girls All 11–12 32.2 21.1 4.5 2.1 19.6 10.9 1.3 0.6 1.5 1.0 12.4 9.0 2.4 1.4 15.3 7.9 22.9 12.0 7.4 4.8 5.4 2.9

13–14 35.4 4.5 23.4 1.2 1.4 14.6 1.9 16.4 27.0 7.9 5.8

15–16 39.5 7.0 23.2 2.1 2.1 12.8 4.2 21.6 28.7 9.3 7.6

% of girls All 11–12 77.8 58.4 29.5 9.4 29.9 8.4 46.7 24.9 9.3 1.9 19.3 8.4 3.1 0.5 13.3 1.4 1.9 0.5 1.3 0.3 9.5 6.0 1.6 0.2 1.3 0.7 1.8 0.4 1.7 0.9 0.6 0.3 8.0 1.0 3.6 2.8

13–14 82.6 28.3 29.8 50.8 9.4 19.5 2.7 12.9 1.4 1.0 10.9 1.3 1.0 1.8 1.8 0.4 7.3 4.2

15–16 91.9 52.7 52.8 64.2 17.2 30.2 6.2 26.3 4.2 2.8 11.2 3.5 2.4 3.2 2.6 1.1 16.4 3.7

Total N about 6300 boys and 7000 girls Table 2.3  Substance use

Ever drunk alcohol 5+ drinks in last 4 weeks Ever seriously drunk Ever smoked cigarette Regular smoker Ever taken drugs Ever sold drugs Ever taken marijuana Ever taken amphetamine Ever taken barbiturate Ever taken solvent Ever taken ecstasy Ever taken LSD Ever taken mushrooms Ever taken cocaine Ever taken heroin Cannabis in last 4 weeks Solvent in last 4 weeks

% of boys All 11–12 79.8 67.1 31.3 12.9 28.8 11.6 36.9 22.1 6.9 1.9 20.9 11.8 5.4 1.9 14.5 2.8 2.1 0.8 1.3 0.8 8.4 7.7 1.8 0.6 1.4 0.6 3.2 1.3 2.4 2.0 1.2 1.1 8.9 1.4 3.2 3.8

13–14 82.2 27.8 26.2 38.7 6.1 19.2 4.5 12.4 1.3 0.9 8.9 1.0 0.9 2.8 1.6 0.9 7.6 3.2

15–16 90.6 56.4 51.4 50.9 13.8 33.3 10.4 30.2 4.7 2.5 8.4 4.3 3.0 6.0 4.0 1.9 19.2 2.6

Total N 6200–6600 boys and 6800–7200 girls

Students were asked whether they had ever drunk alcohol, and Table 2.3 shows that 79.8% of boys and 77.8% of girls said that they had done this. At age 15–16, the ever prevalence increased to 90.6% of boys and 91.9% of girls. They were also asked about binge drinking in the previous 4 weeks, defined as consuming five or more drinks in a single session; 31.3% of boys and 29.5% of girls admitted this. At

2  Great Britain

33

age 15–16, the prevalence increased to 56.4% of boys and 52.7% of girls. Students were also asked about ever being seriously drunk, and 28.8% of boys and 29.9% of girls admitted this. Again, the ever prevalences were highest at age 15–16 (51.4% of boys and 52.8% of girls). Regarding ever smoking cigarettes, 36.9% of boys and 46.7% of girls said that they had done this. At age 15–16, this was admitted by 50.9% of boys and 64.2% of girls. However, only 6.9% of boys and 9.3% of girls said that they were regular smokers. Students were asked whether they had ever taken drugs, and 20.9% of boys and 19.3% of girls said that they had done this. The prevalence was 33.3% of boys aged 15–16 and 30.2% of girls aged 15–16. In regard to selling and dealing in drugs, 5.4% of boys and 3.1% of girls said that they had done this. Again, the prevalence was highest at age 15–16 (10.4% of boys and 6.2% of girls). Students were also asked whether they had ever taken nine specific drugs, and the most commonly used drug was cannabis or marijuana (14.5% of boys and 13.3% of girls; the terms cannabis and marijuana are used interchangeably in this chapter). At age 15–16, 30.2% of boys and 26.3% of girls said that they had used marijuana. The next most common type of drug was glue or solvents; 8.4% of boys and 9.5% of girls had used this, but it was not noticeably more prevalent at age 15–16 than at age 13–14. Other types of drugs, such as magic mushrooms (3.2% of boys and 1.8% of girls), cocaine (2.4% of boys and 1.7% of girls), and amphetamines (2.1% of boys and 1.9% of girls) were less prevalent. Students were also asked about drug use in the previous 4 weeks. Only marijuana (8.9% of boys and 8.0% of girls) and solvents (3.2% of boys and 3.6% of girls) had prevalences greater than 1%. While the prevalence of marijuana was highest at age 15–16 (19.2% of boys and 16.4% of girls), the prevalence of solvent use was highest at age 11–12 for boys (3.8%) and at age 13–14 for girls (4.2%). Students were also asked if they had ever been excluded from school. Table 2.4 shows that 13.6% of boys and 4.9% of girls said that they had been excluded. However, only 2.3% of boys and 0.6% of girls said that they had been excluded in the previous year. Larger numbers had been suspended from school in the previous year (8.5% of boys and 2.8% of girls). Even more students said that they had played Table 2.4  School problems

Ever excluded Excluded in last year Suspended in last year Truant in last year Truant in last 4 weeks Bully often in last year Bullied often in last year

% of boys All 11–12 13.6 8.9 2.3 1.8 8.5 5.0 24.0 12.6 14.4 9.3 4.9 5.4 12.8 15.4

Total N about 6600 boys and 7200 girls

13–14 14.8 2.7 9.9 23.1 13.9 5.0 14.1

% of girls 15–16 All 11–12 13–14 15–16 17.4 4.9 2.5 4.3 8.3 2.3 0.6 0.3 0.7 0.9 10.5 2.8 1.3 2.8 4.4 38.1 24.3 9.5 23.9 41.1 21.1 13.2 7.7 12.7 19.7 4.2 2.9 4.1 2.5 2.3 7.8 12.9 16.4 12.7 9.4

34

D. P. Farrington et al.

truant in the previous year (24% of boys and 24.3% of girls). At age 15–16, the prevalence of truancy was 38.1% of boys and 41.1% of girls. In the previous 4 weeks, 14.4% of boys and 13.2% of girls had truanted. Students were also asked about bullying and being bullied. Regarding bullying, 4.9% of boys and 2.9% of girls said that they had often bullied someone in the previous year. Interestingly, the prevalence of bullying decreased with age for both boys and girls. It was highest at age 11–12 (5.4% of boys and 4.1% of girls). Regarding being bullied, 12.8% of boys and 12.9% of girls said that they were often bullied in the previous year. Again, the prevalence of being bullied was highest at age 11–12 (15.4% of boys and 16.4% of girls).

2.3.3  Developing Delinquency and Drug Scores In order to investigate the relationship between risk and protective factors and delinquency and drug use, it was necessary to construct delinquency and drug scales. No scales were constructed for school problems because these were sometimes included among the risk factors (see later). The delinquency score was based on nine offenses committed in the previous year: shoplifting, stealing a vehicle, stealing from a vehicle, other stealing, burglary, receiving stolen property, vandalism, attacking someone (intending to seriously hurt them), and carrying a weapon. The score was the simple sum of how many of these nine types of offenses had been committed. Table 2.5 shows that 31.6% of boys and 21% of girls had committed two or more of these offenses. These students were operationally defined as “delinquents” in future analyses, while those who had committed no offenses or only one offense were defined as “nondelinquents.” At age 15–16, 38.9% of boys and 26.8% of girls were defined as “delinquents.” The drug score was based on nine drugs ever taken: marijuana, amphetamines, barbiturates, solvents, ecstasy, LSD, magic mushrooms, cocaine, and heroin. The score was the simple sum of how many of these nine types of drugs had been taken. Table 2.5 shows that 16.4% of boys and 15.8% of girls had taken one or more of these drugs. These students were operationally defined as “drug users” in future analyses, while those who had taken no drugs were defined as “nondrug users.” At age 15–16, 28.9% of boys and 26.6% of girls were defined as “drug users.”

2.3.4  Risk and Protective Factors Versus Delinquency Table 2.6 shows the relationships between risk and protective factors and the delinquent/nondelinquent classification explained earlier. Cohen’s d is used as the measure of strength of relationships. At least partly because of the large number of

2  Great Britain

35

Table 2.5  Delinquency and drug scores Boys All 11–12 Delinquency Score 0 49.1 59.7 1 19.4 19.9 2 11.4 9.9 3 6.3 4.3 4–5 7.7 3.5 6–9 6.2 2.7 0–1 68.4 79.6 2–9 31.6 20.4 Drug Score 0 83.6 92.7 1 9.8 4.9 2+ 6.6 2.4 0 83.6 92.7 1+ 16.4 7.3

13–14

15–16

Girls All

11–12

13–14

15–16

46.8 18.6 12.2 7.0 9.1 6.3 65.4 34.6

41.0 20.0 11.7 7.4 10.3 9.6 61.1 38.9

59.7 19.2 9.3 5.3 4.7 1.7 79.0 21.0

74.3 14.3 5.8 2.2 2.9 0.5 88.5 11.5

54.0 21.9 10.2 6.9 5.1 1.8 75.9 24.1

52.6 20.6 11.6 6.2 6.1 2.9 73.2 26.8

85.1 9.0 5.9 85.1 14.9

71.1 16.5 12.5 71.1 28.9

84.2 9.7 6.1 84.2 15.8

93.5 5.3 1.2 93.5 6.5

84.8 8.9 6.3 84.8 15.2

73.4 15.5 11.1 73.4 26.6

N about 6000 boys and 6700–6800 girls

participants, all relationships are statistically significant. For example, for boys, the mean score on poor parental supervision was 1.09 for nondelinquents and 2.21 for delinquents, with a d value of 0.75. For girls, the corresponding means were 1.06 for nondelinquents and 2.33 for delinquents, with a d value of 0.84. These d values indicate large effects. All relationships were in the expected direction. It is clear that the largest d values (strongest relationships) were between the individual/peer risk factors and delinquency (all greater than d = 1.0). However, this could be because antisocial attitudes, early problem behavior, and peer deviance all reflect the same underlying theoretical construct as delinquency. It is not clear that these variables have causal effects on delinquency. For example, much delinquency is committed in groups, so anyone who commits delinquent acts is likely to have delinquent peers. The same is true of the bullying variable. The other, more explanatory, risk factors may have causal effects on delinquency. For boys and girls, the strongest relationships were for availability of drugs, parental attitudes condoning problem behavior, low commitment to school, a family history of problem behavior, and poor parental supervision. The weakest relationships were for low achievement, school disorganization, and community disorganization. However, relationships were usually stronger for girls than for boys. Also, the negative relationships between protective factors and delinquency were usually weaker than the positive relationships between risk factors and delinquency. The most important protective factors were family opportunities for prosocial involvement, family rewards for prosocial involvement, and family attachment.

36

D. P. Farrington et al.

Table 2.6  Risk and protective factors versus delinquency Scales (No. of items in parentheses) Family risk Poor parental supervision (8) Family conflict (5) Family history of problem behaviour (6) Parental attitudes condoning problem behaviour (9) School risk Low achievement (1) Bullying (1) Low commitment to school (6) School disorganization (6) Community risk Community disorganization (5) Availability of drugs (5) Individual/peer risk Alienation and low social commitment (7) Attitudes condoning problem behaviour (10) Early involvement in problem behaviour (6) Peer involvement in problem behaviour (10) Protective factors School opportunities for prosocial involvement (4) School rewards for prosocial involvement (3) Family attachment (4) Family opportunities for prosocial involvement (3) Family rewards for prosocial involvement (4)

Boys ND

DEL

d

Girls ND

DEL

d

1.09 0.74 0.54 0.57

2.21 1.24 1.46 1.66

0.75 0.44 0.76 0.87

1.06 0.87 0.70 0.47

2.33 1.58 1.92 1.62

0.84 0.60 0.95 1.00

1.13 0.37 0.87 0.90

1.51 0.89 1.89 1.61

0.48 0.60 0.78 0.57

1.05 0.27 0.78 0.88

1.44 0.71 1.85 1.49

0.50 0.59 0.88 0.53

0.60 1.15

1.25 2.43

0.56 0.95

0.70 1.30

1.40 2.66

0.59 1.02

1.30 1.27 1.00 1.44

3.25 3.67 2.46 4.01

1.28 1.04 1.21 1.13

1.18 1.32 1.03 1.31

3.41 3.96 2.56 3.85

1.42 1.24 1.42 1.36

2.98 2.21 3.14 2.52 3.23

2.55 1.81 2.61 2.14 2.72

0.40 0.42 0.47 0.45 0.49

3.04 2.22 3.04 2.54 3.23

2.57 1.78 2.39 2.04 2.58

0.45 0.45 0.58 0.60 0.60

ND = Mean score of non-delinquents DEL = Mean score of delinquents; d = Cohen’s d

2.3.5  Risk and Protective Factors Versus Drug Use Table 2.7 shows the relationships between risk and protective factors and drug use. As before, the largest d values were for individual/peer risk factors, but these relationships may not be causal. Otherwise, the most important risk and protective factors for drug use tend to be the same as those for delinquency. Also, as with delinquency, relationships between protective factors and drug use were usually weaker than relationships between risk factors and drug use.

2.4  Conclusions This chapter has described results obtained from the application of the CTC-YS to a very large school-based British nationally representative sample of students aged 11–16. Although the research was carried out some time ago, this is the first use of

2  Great Britain

37

Table 2.7  Risk and protective factors versus drug use Scales (No. of items in parentheses) Family risk Poor parental supervision (8) Family conflict (5) Family history of problem behaviour (6) Parental attitudes condoning problem behaviour (9) School risk Low achievement (1) Bullying(1) Low commitment to school 6) School disorganization (6) Community risk Community disorganization (5) Availability of drugs (5) Individual/Peer risk Alienation and low social commitment (7) Attitudes condoning problem behaviour (10) Early involvement in problem behaviour (6) Peer involvement in problem behaviour (10) Protective factors School opportunities for prosocial involvement (4) School rewards for prosocial involvement (3) Family attachment (4) Family opportunities for prosocial involvement (3) Family rewards for prosocial involvement (4)

Boys ND

DRG

d

Girls ND DRG

d

1.20 0.82 0.60 0.69

2.54 1.18 1.82 1.94

0.91 0.31 1.04 1.01

1.08 0.93 0.72 0.50

2.44 1.52 2.09 1.64

0.92 0.49 1.07 1.01

1.19 0.47 1.01 1.00

1.52 0.80 1.99 1.66

0.42 0.38 0.75 0.54

1.05 0.30 0.82 0.90

1.51 0.59 1.82 1.53

0.58 0.39 0.83 0.55

0.70 1.25

1.30 2.92

0.51 1.27

0.75 1.29

1.34 2.94

0.49 1.27

1.64 1.59 1.06 1.66

3.06 3.99 3.12 4.71

0.86 1.02 1.86 1.37

1.35 1.45 1.01 1.36

2.96 3.78 2.83 3.98

0.97 1.07 1.81 1.40

2.93 2.15 3.07 2.48 3.17

2.50 1.75 2.52 2.07 2.66

0.40 0.41 0.50 0.48 0.48

3.03 2.19 3.02 2.54 3.20

2.55 1.78 2.36 1.99 2.52

0.47 0.42 0.58 0.65 0.64

ND = Mean score of non-drug-users DRG = Mean score of drug-users d = Cohen’s d

the CTC-YS in Europe with a large national sample. Ideally, the survey should be repeated to investigate changes over time. The survey provides valuable information about the prevalence of delinquency, smoking, alcohol use, drug use, and school problems such as truancy and bullying. It would be very useful for local areas that are using the CTC-YS to compare their results with these national figures in order to determine which problems are unusually prevalent in their areas. The survey also shows how the problem behavior ­outcomes vary with age and gender and the strength of relationships between risk and protective factors and delinquency, substance use, and school problems. The results of the analyses of risk and protective factors suggest important targets for interventions designed to reduce delinquency and drug use. Regarding risk factors, it would be desirable to improve parental supervision, family conflict, antisocial parental attitudes, commitment to school, school disorganization, community disorganization, the availability of drugs in communities, and antisocial student

38

D. P. Farrington et al.

attitudes. It would also be desirable to enhance family attachment and opportunities and rewards for prosocial involvement in the family and the school. These are exactly the aims of CTC intervention programs. Acknowledgments  We gratefully acknowledge the valuable contributions of Sarah Beinart, Barry Anderson, and Stephanie Lee to the CTC-YS.

References Anderson, B., Beinart, S., Farrington, D. P., Langman, J., Sturgis, P., & Utting, D. (2001). Risk and protective factors associated with youth crime and effective interventions to prevent it. London: Youth Justice Board for England and Wales. Armstrong, D., Hine, J., Hacking, S., Armaos, R., Jones, R., Klessinger, N., & France, A. (2005). Children, risk and crime: The On Track youth lifestyles surveys (Research Study No. 278). London: Home Office. Arthur, M. W., Briney, J. S., Hawkins, J. D., Abbott, R. D., Brooke-Weiss, B. L., & Catalano, R. F. (2007). Measuring risk and protection in communities using the communities that care youth survey. Evaluation and Program Planning, 30, 197–211. Axford, N., Sonthalia, S., Wrigley, Z., Webb, L., Mokhtar, N., Brook, L., et al. (2016). What works in Europe? Developing a European Communities That Care database of effective prevention programmes. Dartington: Dartington Social Research Unit, Report submitted to the European Commission. Beinart, S., Anderson, B., Lee, S., & Utting, D. (2002). Youth at risk? A national survey of risk factors, protective factors and problem behaviour among young people in England, Scotland and Wales. London: Communities That Care Ltd.. Bhabra, S., Dinos, S., & Ghate, D. (2006). Young people, risk and protection: A major survey of secondary schools in On Track areas (Research Report No. RR 728). London: Department for Education and Skills. Bowling, B., Graham, J., & Ross, A. (1994). Self-reported offending among young people in England and Wales. In J. Junger-Tas, G. J. Terlouw, & M. W. Klein (Eds.), Delinquent behavior among young people in the Western world: First results of the International Self-Report Delinquency Study (pp. 42–64). Amsterdam: Kugler. Crow, I., France, A., Hacking, S., & Hart, M. (2004). Does Communities That Care work? An evaluation of a community-based risk prevention programme in three neighbourhoods. York: Joseph Rowntree Foundation. Fagan, A. A., Hawkins, J. D., Catalano, R. F., & Farrington, D. P. (2019). Communities that care: Building community engagement and capacity to prevent youth behavior problems. New York: Oxford University Press. Farrington, D. P. (1973). Self‑reports of deviant behavior: Predictive and stable? Journal of Criminal Law and Criminology, 64, 99–110. Farrington, D.  P. (1996). Understanding and preventing youth crime. York: Joseph Rowntree Foundation. Farrington, D.  P. (1997). Evaluating a community crime prevention program. Evaluation, 3, 157–173. Farrington, D. P. (1998). Evaluating Communities that Care: Realistic scientific considerations. Evaluation, 4, 204–210. Farrington, D. P. (2000). Explaining and preventing crime: The globalization of knowledge – The American Society of Criminology 1999 Presidential Address. Criminology, 38, 1–24.

2  Great Britain

39

Farrington, D. P. (2001). What has been learned from self-reports about criminal careers and the causes of offending? (Online Report). London: Home Office. Reprinted in D. Canter (Ed.) (2014). Criminal psychology, Vol. 2: Criminals’ characteristics (pp. 21–53). London: Sage. Farrington, D. P. (2020). The developmental evidence base: Prevention. In D. A. Crighton & G. J. Towl (Eds.), Forensic psychology (3rd ed.). Chichester: Wiley. Flood-Page, C., Campbell, S., Harrington, V., & Miller, J. (2000). Youth crime: Findings from the 1998/99 Youth Lifestyles Survey (Research Study No. 209). London: Home Office. Gomes, H. S., Maia, A., & Farrington, D. P. (2018). Measuring offending: Self-reports, official records, systematic observation and experimentation. Crime Psychology Review, 4, 26–44. Graham, J., & Bowling, B. (1995). Young people and crime (Research Study No. 145). London: Home Office. Hales, J., Nevill, C., Pudney, S., & Tipping, S. (2009). Longitudinal analysis of the Offending, Crime and Justice Survey 2003-06 (Research Report 19). London: Home Office. Hawkins, J. D., Catalano, R. F., & Arthur, M. W. (2002). Promoting science-based prevention in communities. Addictive Behaviors, 27, 951–976. Herlitz, L., Hough, M., McVie, S., & Murray, K. (2016a). Understanding and preventing youth crime in England: Key findings. London: Institute for Criminal Policy Research. Herlitz, L., Hough, M., McVie, S., & Murray, K. (2016b). Understanding and preventing youth crime in Scotland: Key findings. London: Institute for Criminal Policy Research. Jolliffe, D., & Farrington, D.  P. (2014). Self-reported offending: Reliability and validity. In G. J. N. Bruinsma & D. Weisburd (Eds.), Encyclopedia of criminology and criminal justice (pp. 4716–4723). New York: Springer-Verlag. Junger-Tas, J., Marshall, I. H., Enzmann, D., Killias, M., Steketee, M., & Gruszczynska, B. (2010). Juvenile delinquency in Europe and beyond: Results of the Second International Self-Report Delinquency Study. Dordrecht: Springer. Mayhew, P., & Elliott, D. (1990). Self-reported offending, victimization, and the British Crime Survey. Violence and Victims, 5, 83–96. Monahan, K. C., Oesterle, S., Rhew, I., & Hawkins, J. D. (2014). The relation between risk and protective factors for problem behaviors and depressive symptoms, antisocial behavior, and alcohol use in adolescence. Journal of Community Psychology, 42, 621–638. Pawson, R., & Tilley, N. (1998a). Caring communities, paradigm polemics, design debates. Evaluation, 4, 73–90. Pawson, R., & Tilley, N. (1998b). Cook-book methods and disastrous recipes. Evaluation, 4, 211–213. Riley, D., & Shaw, M. (1985). Parental supervision and juvenile delinquency (Home Office Research Study No. 83). London: Her Majesty’s Stationery Office. Roche, S., & Hough, M. (2018). Minority youth and integration: The ISRD-3 study in Europe and the US. Cham: Springer. Roe, S., & Ashe, J. (2008). Young people and crime: Findings from the 2006 Offending, Crime and Justice Survey (Statistical Bulletin 09/08). London: Home Office. Utting, D. (2016). Building better outcomes for children? A process evaluation of the Evidence2Success project in Perth and Kinross. York: Joseph Rowntree Foundation.

Chapter 3

The Netherlands Harrie Jonkman and Clemens M. H. Hosman

3.1  Introduction During the 1990s there was a growth of delinquency and violence among youngsters in the Netherlands, about which the Dutch Ministry of Justice was worried. There was interest in alternative ways to deal with serious offenders as well as in investigating the possibility of preventing serious offending at an early stage. The Ministry commissioned two reports that would deal with different options for prevention, and Communities That Care (CTC) was presented as a promising approach. CTC was thought to be one of the more positive socio-political responses to the significant increase in violence and youth delinquency in certain local settings in the Netherlands during these years. CTC was seen as a structured, community-­orientated and effective answer to the social consequences of the youth problem behaviours which had disruptive effects in a number of Dutch areas, cities and neighbourhoods. This coherent and planned initiative was expected to positively affect different contexts (family, school, peers and neighbourhood) and to add a more rational approach to local youth policy as well as stimulating more effective methods of raising children in these areas (Jonkman 2012). Two developments played a key role in why CTC was seen as an important development in the Netherlands. These were the growing importance of a developmental perspective in social and health sciences and the growth of effective interventions. Both were based on a wealth of scientific research outcomes. In the 1990s there was a clear scientific move towards a more developmental perspective, as could be seen in the fields of health, psychiatry, economy, criminology and H. Jonkman (*) Verwey-Jonker Institute Utrecht, Utrecht, The Netherlands e-mail: [email protected] C. M. H. Hosman Maastricht University and Radbout University Nijmegen, Groesbeek, The Netherlands e-mail: [email protected] © Springer Nature Switzerland AG 2021 D. P. Farrington et al. (eds.), Delinquency and Substance Use in Europe, https://doi.org/10.1007/978-3-030-58442-9_3

41

42

H. Jonkman and C. M. H. Hosman

sociology. Considerable emphasis was placed on the importance of early development, the upbringing of children and youngsters and the importance of family, community conditions and the relationship and interaction with antisocial peers. Empirical youth studies and local and national monitors which used different indicators provided information about health and well-being in different times and places. Local youth monitors of the city of Rotterdam and national youth monitors under the authority of the Ministery of Health, Welfare and Sport became examples of these. Studies like Kinderen in Tel, Generation R and the international HBSC (Health Behaviour in School-aged Children) study were other empirical youth studies of this kind. In addition, an important number of studies were published at that time showing the growing availability of evidence-based interventions and that some preventive interventions are more effective than others (Elliott 1997; Sherman et  al. 1996). There was a growing recognition that preventive interventions might have a positive impact on the emotional and social development of children and youngsters if they were implemented well and early enough. So, the problems with youth on the one hand and the scientific and practical developments on the other hand encouraged the Dutch government to adopt CTC as a preventive experiment (Jonkman 2012). The universal community prevention strategy CTC, with its emphasis on the empirical study of youth in communities and the use of effective programme(s), started in the Netherlands around 2000. The translation and introduction of the original US program to the Dutch context was made possible by two Dutch ministries, the Ministry of Justice and the Ministry of Health, Welfare and Sport, as they were called at that time. The experiment was set up in four pilot areas (disadvantaged areas in Amsterdam, Arnhem, Rotterdam and Zwolle). The CTC instruments were adapted and tested for use in Dutch schools and communities, including the Communities That Care Youth Survey (CTC-YS) and the Manual of Promising and Effective Programs (Ince et  al. 2004). This was the first book on effective youth programs in the Netherlands. Meanwhile training sessions were set up, and the pilot cities were guided by prevention workers in implementing the CTC strategy. After this first phase of CTC in the Netherlands, it was concluded that the strategy increased the quality of planning and decision taking; that it stimulated collaboration among service providers and coordination of the input in programming preventive interventions; that it put more focus on risk and protective factors; that it demonstrated more use of effective and promising approaches; and, in the end, that it stimulated the participation of young people and other citizens in preventive interventions. The years 2005–2008 can be seen as the second phase of CTC. The government funded CTC prevention work in some cities. Meanwhile, other parties (provinces, cities) were interested and started to invest strongly in CTC. These were the most successful years of CTC in the Netherlands, in which more than 25 communities worked with this strategy for their youth policy, and the instruments were refined. The CTC approach was also important in the development and use of effective programs in the Netherlands and was one of the main resources of the Dutch ‘Databank of Effective Interventions’ (DEI) which informs Dutch cities and institutes about

3  The Netherlands

43

effective and promising programs. The CTC-YS was implemented in cities and communities to show the prevalence of different youth problem behaviours and detect the most important risk and protective factors. Based on these profiles of problems, risks and protection, effective and promising programs were chosen from the CTC manual and the DEI. The years 2008 and afterwards represent the third phase of CTC. The national government withdrew because it emphasized more and more the importance of local policies and the responsibility of cities. The interest of cities in the prevention strategy decreased, however, because they had to pay for it themselves in a period of severe economic crisis. But during this period a quasi-experimental study of five experimental and control communities started on the effects and the process of the CTC prevention system. This study examined whether this prevention system had effects on the development of problem behaviours and on the specific risk and protective factors among youngsters (12–18  years). This study could not detect any impact on any problem behaviour, but it showed results on collaboration and community commitment. The disappointing results were placed in the context of the threats to internal validity (e.g. the lack of tested and effective programs in the Netherlands at that time, delayed and partial implementation and contamination) and design limitations (among them non-randomization and small sample size; see Jonkman 2012; Steketee et al. 2013). City councils repeatedly ask experts and national institutes for insights and instruments that help them to focus on the main youth problems like antisocial behaviour and substance use. They are interested in scientific support to choose areas where the problems are most prevalent, to choose problems which are most prevalent and to choose effective interventions which can help them to deal with these problems. Information about the underlying factors (especially risk and protective factors) offers them good prospects for targeted social investment and a successful prevention policy. Sound research into prevalence and the underlying risk and protective factors can form the basis for rational and effective prevention strategies. Over the years the CTC-YS was used many times in the Netherlands, and the results helped the city councils in finding their way in youth policy. In 2011 the CTC-YS had already been implemented in the Netherlands. At this time, the collection of data on youngsters between 12 and 18 years was part of the experimental study in which five Dutch cities participated. This chapter is based on that large dataset collected in five middle-sized cities located in the provinces of Zuid-Holland and Zeeland. We examined the impact of predictors of violence and delinquency for a general population of youngsters in the Netherlands (Jonkman 2012). This kind of research is important for national public health, crime prevention and social policy in the Netherlands as well as for public health and social issues in Dutch cities and communities. In this study, we examined the prevalence of violence and delinquency, other problem behaviours and predictors (risk factors, protective factors) in youngsters aged between 12 and 18 years and looked for variation in the contexts of communities. We researched the correlates of violence and delinquency of youngsters in a comprehensive perspective. With the use of various

44

H. Jonkman and C. M. H. Hosman

groups of factors (socio-demographic variables, risk and protective factors), problem behaviours were analysed. We asked ourselves: what is the prevalence of violence and delinquency; how do they correlate with demographic covariates, with other problem behaviours and with risk and protective factors; and what is the predictive power of these underlying factors? Ultimately, how can we use this knowledge to develop strategic targets for public health policy, crime prevention and social policy? Adequate indicators will be proposed in this chapter.

3.2  Methods 3.2.1  Settings The data we used in this study are based on a study which was conducted in 2011 as part of an effectiveness study on CTC in the Netherlands (Jonkman 2012; Steketee et al. 2013). Five cities participated in this study. The cities are located in the provinces of Zuid-Holland and Zeeland (South-west of the Netherlands). The five cities (Capelle aan den IJssel, Gouda, Middelburg, Spijkenisse, Zwijndrecht) are middle-­ sized with an average population of 61,633 (between 44,511 and 77,096 inhabitants). With the support of the city councils, we wrote a letter to the youngsters and asked them to fill in the CTC-YS on the Internet. From the five cities, youngsters participated across 52 communities. To make this Dutch sample comparable to samples in different European countries, we used in this research only the 12–18-year-­ old youngsters (Table 3.1). To make our study and data comparable to the other EU datasets in this book, not all the information gathered from our dataset could be used. For example, information on ethnicity, home situation and socioeconomic background was not used, because in the other countries this was not asked or it was not comparable.

Table 3.1  Participants in cities and communities City Capelle aan den Ijssel Gouda Middelburg Spijkenisse Zwijndrecht Mean

Youngsters 12–18 years 4731

Number of participants 1326

Response rate 28.0%

Number of communities 9

Min-max No. in communities 39–342

4903 3310 4980 2423 4069

1721 1142 995 550 1147

35.1% 34.5% 20.0% 22.7% 28.1%

11 11 14 7 10

21–450 23–162 16–218 38–109 27–256

3  The Netherlands

45

3.2.2  Participants Names and addresses of youngsters in a specific age group were provided by the city council. All the eligible youngsters received an invitation letter to participate, and their parents received an information letter. Youths were asked to fill in a questionnaire on the Internet and could use a personal code for this. The youngest were 11 years old and the oldest were 18 and 19 years. We took 92 youngsters outside the range of 12–17 years out of the dataset of 5887 youngsters to make the data internationally comparable. In total in this study 5795 youngsters participated; the 14-year-­olds were the biggest group (20.7%), and the 17-year-olds were the smallest group (12.2%). Concerning gender, 54.9% of the youngsters were female (Table 3.2).

3.2.3  Measures The research instrument used for this research is adapted from the CTC-YS (Arthur et  al. 2007; Glaser et  al. 2005; Jonkman 2012; Jonkman et  al. 2006; Pollard et al. 1999). The survey was developed from the 1990s onwards to measure problem behaviours in adolescent youngsters and their risk and protective factors. It can be conducted relatively easily in schools and communities. With this instrument the quality of the social domains in which youngsters grow up (family, school, friends and communities) can be measured. In addition, the youth behaviour profiles of cities, communities, neighbourhoods or other areas can be created and compared with each other. This instrument was tested for reliability and validity (Arthur et al. 2007; Glaser et al. 2005). In recent years the CTC-YS has been used in various countries like Australia, Austria, Canada, Croatia, Cyprus, Great Britain, Germany and the Netherlands (Jonkman 2012; Jonkman et  al. 2006). In the Netherlands, the instrument has been used since 2000 and renewed several times.

Table 3.2  Age and gender of participants

Age 12 years 13 years 14 years 15 years 16 years 17 years Gender Female

Percentage 18.1 20.1 20.7 15.2 13.7 12.2 54.9

46

H. Jonkman and C. M. H. Hosman

3.2.3.1  Instrument The outcome indicators for this study concerned violence and youth delinquency. Violence, as behaviour with physical force exerted for the purpose of causing damage or injury, was in this research assessed by four questions regarding last year events: (1) Did you carry a weapon (weapons)? (2) Were you involved in a fight (fighting)? (3) Did you hit someone with the intention of hurting (hitting)? (4) Did you threaten someone to get money (assault)? Youth delinquency is a generic term covering various punishable acts by youngsters. Besides violent offenses, crimes against property, arson and destruction are also part of youth delinquency (Jonkman et al. 2006; Elliott and Tolan 1999; Junger-Tas et al. 2013). We asked five questions regarding delinquency: (1) Did you intentionally destroy anything on the street (vandalism)? (2) Did you steal anything from a shop (theft shops)? (3) Did you steal anything from school, e.g. from another student (theft schools)? (4) Did you sell stolen items to someone else, e.g. to another student (selling stolen items)? (5) Were you arrested by the police (contact with police)? For both indicators we used binary data. Students scored a ‘yes’ on violence or delinquency when they answered yes to one or more of these problem behaviour items (last year). They got a ‘no’ when they were involved in none of the items. For this study we also researched substance abuse (alcohol, smoking, soft drugs) and depression, which are frequently associated with violent and delinquent behaviour in youngsters (Catalano et al. 2012). We asked the youngsters different questions about substance use, every time with yes and no questions: Smokever (ever smoked cigarettes), smoking now (smoked a cigarette during the last month) and smokeheavy (smoke more than 10 cigarettes a day) are the questions asked concerning smoking. We used three alcohol-related questions in this comparative research: alcever (ever drink alcohol), alcnow (did you drink alcohol during the last month) and alcbinge (drunk five or more glasses of alcohol during the last 2 weeks). On drug use we measured hashnow (did you use hash, marihuana or weed during the last month?). On depression we asked the youngsters four questions about their feelings during the last year (Sometimes I think that life is not worth it; At times I think I am no good at all; All in all, I am inclined to think that I am a failure; In the past year, have you felt depressed or sad most days, even if you felt ok sometimes?). We made a binary variable out of these questions. Youngsters who scored a yes on one or more of these four questions got a yes on depression, and otherwise they got a no. Although we also asked questions on school dropout and sexually related problem behaviour, this information is not used in this chapter because it was not used in other European countries. We also measured risk and protective factors as predictors of outcomes in this study. Risk factors are defined as possible causes of problem behaviour that have an undesirable influence on behaviour according to the scientific literature (Catalano et  al. 2012; Garmezy and Masten 1985; Hawkins et  al. 1992, 1995; Loeber and

3  The Netherlands

47

Farrington 1998, 2001; Werner and Smith 1992). Protective factors are factors that positively influence prosocial behaviour and/or buffer against the impact of risk factors (Catalano et al. 2004; Hosman and Jané-Llopis 1999; Pollard et al. 1999; Rutter 1985). All risk and protective factors included in the CTC-YS were selected based on their influential role found in numerous longitudinal, prospective and experimental studies. Risk and protective factors are associated with the domains in which youngsters grow up on a daily basis: family, school, friends and communities. Altogether we asked youngsters questions about 20 risk factors. For the family we asked them about five risk factors (Family History of Problem Behaviour; Poor Family Management; Family Conflict; Parent Attitudes Favourable Towards Drug Use; Parents’ Attitudes Favourable to Antisocial Behaviour). For the school we asked them about two risk factors (Early Academic Failure; Low Commitment to School). For peer/individual we asked about eight risk factors (Rebelliousness; Gang Involvement; Early Initiation of Antisocial Behaviour; Early Initiation of Drug Use; Favourable Attitude Towards Drugs; Favourable Attitude Towards Antisocial Behaviour; Friends’ Use of Drugs; Interaction with Antisocial Peers). For the domain of the community, we asked about five risk factors (Low Neighbourhood Attachment; Community Disorganization; Transition and Mobility; Perceived Availability of Drugs and Weapons; Norms Favourable to Antisocial Behaviour). We asked the youngsters also about ten protective factors spread over the same four domains of family, school, peers and community. For the family we asked about three factors (Attachment; Opportunities for Prosocial Involvement; Rewards for Prosocial Involvement), for the school two factors (Opportunities for Prosocial Involvement; Rewards for Prosocial Involvement), for peers three factors (Belief in the Moral Order, Social Skills, Religiosity) and for the community two factors (Opportunities for Prosocial Involvement; Rewards for Prosocial Involvement). In this study we included in further analyses only the 14 risk and 7 protective factors that each were measured by multiple items and had a Cronbach alpha of 0.60 or higher (Table 3.3). We also dichotomized the risk and protective factors at the median score. Youngsters who scored higher than the median were scored one on this factor. If they scored lower than the median, they got a zero for the factor. Based on this we counted for each individual the number of risk factors and protective factors and constructed the two variables Riskcount and Protectcount. For the definition of communities, we used the administrative boundaries of census block groups with their own Dutch zip code (CBS 2011). By defining the community, we could research the influence of the community on the outcomes. We defined 52 communities (containing a minimum of 16 students and a maximum of 450 students).

48

H. Jonkman and C. M. H. Hosman

Table 3.3  Risk and protective factors, number of items and Cronbach alpha Risk factors FR1: Family History of Problem Behaviour FR2: Poor Family Management FR3: Family Conflict FR4: Parental Attitudes for Drug Use FR5: Parental Attitudes for Antisocial Behaviour SR1: Academic Failure SR2: Low Commitment to School IR1: Rebelliousness IR2: Gang Involvement IR3: Early Initiation of Antisocial Behaviour IR4: Early Initiation of Drug Use IR5: Favourable Attitudes to Drug use IR6: Favourable Attitudes to Antisocial Behaviour IR7: Friends’ Use of Drugs IR8: Interaction with Antisocial Peers CR1: Low Neighbourhood Attachment CR2: Community Disorganization CR3: Transition and Mobility CR4: Perceived Availability of Drugs and Weapons CR5: Laws and Norms Favourable to Antisocial Behaviour Protective factors FP1: Attachment FP2: Opportunities for Prosocial Involvement FP3: Rewards for Prosocial Involvement SP1: Opportunities for Prosocial Involvement SP2: Rewards for Prosocial Involvement IP1: Belief in Moral Order IP2: Social Skills IP3: Religiosity CP1: Opportunities for Prosocial Involvement CP2: Rewards for Prosocial Involvement

Items

Alpha

4 8 3 3 3 2 5 3 1 4 4 4 5 4 5 3 6 4 2 3

0.68 0.80 0.77 0.59 0.66 0.54 0.75 0.31 NA 0.29 0.52 0.79 0.75 0.69 0.75 0.85 0.85 0.45 0.95 0.67

6 3 2 4 3 4 4 1 2 3

0.85 0.73 0.83 0.60 0.62 0.51 0.52 NA 0.38 0.79

Note: NA = not applicable. Bold alpha values are 0.60 or greater and these factors are use in the analyses

3.2.4  Analyses First, we describe the sample and show the results for violence and delinquency and their associations with other problem behaviours (smoking ever, smoking now, smoking heavy, alcohol ever, alcohol now, binge drinking and depression). In this first part of the study, we also research the influence of the community on violence

3  The Netherlands

49

and delinquency and consider if it is necessary to research violence and delinquency of the youngsters in a multilevel perspective. For this we used multilevel analysis techniques (Hox et  al. 2018; Gelman and Hill 2007; Rabe-Hesketh and Skondal 2012) which can account for the clustering effects of the sampled youngsters within communities, and we modelled the relationships between the individual and the community if this was necessary. Second, for violence and delinquency, we study the separate influence of each risk and protective factor. First, we looked at the odds ratio (OR) which is an adequate measure of the association between exposure to a risk or protective factor and an outcome (in this case violence and delinquency). It represents the odds that violence or delinquency will occur given an exposure to the risk or protective factor and compared to the odds of the outcome when there is non-exposure. An OR of 2.0 or more (or conversely less than 0.5) is seen as fairly large (Cohen 1996). But, for the purpose of prevention, it may be more important to know what it means for the outcomes (violence, delinquency) when the exposure to the risk factor decreases or when the exposure to the protective factor increases. For this the population attributable fraction (PAF) is important. The PAF is the proportional reduction in population violence or delinquency (in our case) that would occur if exposure to a risk factor is eliminated. It shows the proportion of violent/delinquent cases which occurs in the total population and which might be explained by exposure to the risk factor. The PAF equals the preventable fraction or the incidence of cases that might be prevented if preventive measures were able to completely eliminate that risk factor. We also researched the OR and PAF of protective factors for violence and delinquency. As a third and last part of the study, we investigated the clustering effects of risk and protective factors. For studying this, two age groups are defined (12–14 and 15–17 years). Again, we dichotomized each factor, but now within each of these age groups, and used the median within each group as the cut-off point. Each individual had a binary risk factor and a binary protective factor. We count the number of risk factors for each individual (‘Riskcount’) to analyse the clustering effect of risk factors. We also count the number of protective factors (‘Protectcount’). For this part we used logistic regression to investigate the results. For the three parts of the study, we conducted statistical analyses using Stata 15.1 (Stata Corporation 2017).

3.3  Results 3.3.1  Descriptive Results Of the youngsters, 6.9% admitted one or more violent acts; 1.7% carried guns, 5.0% were involved in a fight, 2.5% attacked someone, and 0.2% threatened someone. 5.6% of the youngsters admitted delinquent acts such as steal from a shop (1.6%), steal at school (1.4%), contact with police (2.0%), selling stolen things (0.3%) and

50

H. Jonkman and C. M. H. Hosman

vandalizing on the street (0.4%). Boys and the older age group had higher prevalence rates of violence (11.1%, 7.3%) and delinquency (9.1%, 7.2%) than girls and the younger age group on violence (3.8%, 6.4%) and delinquency (3.1%, 4.4%). We also researched the prevalence of smoking ever (20.7%), smoking last month (9.3%), smoking 10 or more cigarettes a day (1.4%), drinking alcohol ever (42.3%), drinking alcohol last month (15.9%), binge drinking (1.8%) and use of hash last month (9.4%); 38.6% of the youngsters reported depression. Smoking, alcohol drinking, hash use and depression were positively associated with violence and delinquency. When youngsters were violent or delinquent, their odds ratios are significantly positive, for instance, OR = 1.51 for violence versus hash use and OR = 7.48 for delinquency versus smoking 10 cigarettes or more a day (Table 3.4). We also researched the influence of the community on violence and delinquency using hierarchical multilevel analysis. We found for violence and for delinquency a low intraclass correlation of 0.01, which is negligible. Because of this low level of variability on outcome indicators between Dutch communities, we did not use multilevel analysis for violence and delinquency of individual youngsters clustered in higher levels of communities.

3.3.2  R  isk and Protective Factors Versus Violence and Delinquency We researched the association of violence and delinquency with risk factors in the domains of family, school, friends and community (Table 3.5). We found that all the risk factors were significantly associated with violence and delinquency. We found high associations for Early Initiation of Antisocial Involvement (OR violence = 4.88 and OR delinquency = 7.13). We found lower associations for Problems with Family Management (only for violence), Low Neighbourhood Attachment and Norms Favourable to Antisocial Behaviour. But even here we see that the odds of violence is significantly higher for youngsters who got a yes on these risk factors. The other Table 3.4  Relation between violence and delinquency and other problem behaviours in odds ratios

Problem behaviours Violence Smoke ever 4.05 (3.29/5.00) Smoke now 4.78 (3.77/6.07) Smoke heavy 13.4 (8.54/21.09) Alcohol ever 3.12 (2.49/3.91) Alcohol use now 1.94 (1.50/2.52) Alcohol binge 3.08 (1.96/4.84) Hash use now 1.51 (1.11/2.05) Depression 1.11 (1.06/1.16)

Delinquency 4.73 (3.76/5.98) 6.40 (4.98/8.21) 7.48 (4.74/12.67) 5.67 (4.25/7.57) 3.17 (2.41/4.17) 3.55 (2.27/5.57) 2.16 (1.60/2.93) 1.15 (1.10/1.21)

Note: 95% confidence intervals in parentheses

51

3  The Netherlands

Table 3.5  Odds ratio (OR) and population attributable fraction (PAF) of risk factors for violence and delinquency Risk factors History of Problem Behaviour in the Family (FR1)

OR violence 3.59 (2.69/4.69) Problems with Family Management (FR2) 1.75 (1.41/2.16) Conflicts in the Family (FR3) 2.39 (1.93/2.96) Parental Attitudes Favourable Towards Antisocial 4.43 Behaviour (FR5) (3.53/5.57) Low Commitment to School (SR2) 2.79 (2.26/3.45) Early Initiation of Alcohol and Drug Use (IR4) 4.88 (3.90/6.11) Favourable Attitudes Towards Drug Use (IR5) 2.51 (2.04/3.10) Favourable Attitudes Towards Antisocial Behaviour 3.30 (IR6) (2.66/4.09) Friends’ Use of Drugs (IR7) 3.84 (3.09/4.80) Interactions with Antisocial Peers (IR8) 4.43 (3.51/5.58) Low Neighbourhood Attachment (CR1) 1.33 (1.08/1.64) Community Disorganization (CR2) 2.41 (1.94/3.00) Norms Favourable Towards Antisocial Behaviour 1.41 (CR5) (1.14/1.75) Riskcount 1.38 (1.33/1.44)

OR PAF delinquency 0.13 4.97 (3.69/6.70) 0.24 2.98 (2.31/3.84) 0.35 2.10 (1.66/2.65) 0.53 3.21 (2.53/4.08) 0.38 3.47 (2.73/4.42) 0.53 7.13 (5.45/9.33) 0.36 4.25 (3.30/5.48) 0.44 5.29 (4.08/6.87) 0.45 5.24 (4.07/6.76) 0.55 7.21 (5.39/9.66) 0.12 1.37 (1.09/1.72) 0.11 2.32 (1.82/2.94) 0.02 1.79 (1.42/2.25) 1.51 (1.44/1.58)

PAF 0.19 0.46 0.30 0.43 0.47 0.65 0.60 0.60 0.55 0.70 0.13 0.11 0.19

risk factors show large effects for violence and delinquency defined as OR > 2.0 (History of Problem Behaviour in the Family, Conflicts in the Family, Parental Attitudes Favourable Towards Antisocial Behaviour, Low Commitment to the School, Early Initiation of Alcohol and Drug Use, Favourable Attitudes Towards Drug Use, Favourable Attitudes Towards Antisocial Behaviour, Friends’ Use of Drugs and Interaction with Antisocial Peers and Community Disorganization). From a prevention perspective, the population attributable fraction (PAF) is important because it is the proportion of violent/delinquent cases occurring in the total population which might be explained by the exposure to preventable risk factors, not as a causal factor but as an indicator. For example, 11% of the violence might be eliminated if youngsters were not exposed to Community Disorganization, but 72% of the violent cases might be reduced by taking away the influence of Early Initiation of Antisocial Behaviour. The influences on delinquency are comparable.

52

H. Jonkman and C. M. H. Hosman

Table 3.6  Odds ratio (OR) and population attributable fraction (PAF) of protective factors for violence and delinquency Protective factors Attachment (FP1) Opportunities for Prosocial Involvement (FP2) Rewards for Prosocial Involvement (FP3) Opportunities for Prosocial Involvement (SP1) Rewards for Prosocial Involvement (SP2) Religiosity (IP3) Rewards for Prosocial Involvement (CP2) Protectcount

OR violence 0.49 (0.39/0.62) 0.53 (0.43/0.67)

PAF OR delinquency 0.22 0.44 (0.34/0.56) 0.19 0.37 (0.28/0.48)

PAF 0.20 0.24

0.61 (0.49/0.76) 0.68 (0.55/0.85)

0.15 0.45 (0.35/0.59) 0.12 0.57 (0.44/0.73)

0.22 0.16

0.74 (0.60/0.92) 1.14 (0.93/1.41) ns 0.85 (0.69/1.05) ns 0.85 (0.80/0.90)

0.10 0.51 (0.40/0.66) 0.05 0.94 (0.74/1.18) ns 0.06 0.69 (0.54/0.88)

0.19 0.02 0.12

0.75 (0.70/0.80)

Here 70% of the delinquent cases might be prevented by eliminating the Interaction with Antisocial Peers. For protective factors we see also significant results (Table 3.6). Only the influence of Religiosity is nonsignificant for violence and delinquency, and Rewards for Prosocial Involvement in the Community is nonsignificant for violence. Violence and delinquency are negatively associated with the protective factors. For violence, Attachment with the Family decreases the odds by 51%, and for delinquency it is the Opportunities for Prosocial Involvement in the Family which decreases the odds by 63%. For delinquency we detected large effects for Attachment, Opportunities for Prosocial Involvement (on delinquency) and Rewards for Prosocial Involvement (on delinquency). The highest protective factor contribution to violence comes from Attachment (PAF = 0.22), and the highest contribution to delinquency comes from Opportunities for Prosocial Involvement in the Family (PAF = 0.24).

3.3.3  The Cumulative Effect of Risk and Protective Factors In the third and last part of the study, we researched the cumulative effect of risk and protective factors which are clustered together. The probability of violence increases by 34% with every new risk factor, and the probability of delinquency increases by 42.5%. The probability of violence and delinquency decreases with every protective factor by 14% and 23%, respectively. The chance of violence is 0.02% and delinquency is 0.02% when youngsters have two or less risk factors and is 38% and 35% when youngsters have ten or more risk factors. For protective factors, the influence is just the other way round. When they have six or seven protective factors, the chance of violence is 0.05% and of delinquency is 0.02%. When they have no or one protective factor, they have a chance of 14% of both violence as delinquency (see Fig. 3.1).

3  The Netherlands

.6 0

0

.2

.4

mean of delinquency

.4 .2

mean of violence

.6

.8

.8

53

0

1

2

3

4

5

6

7

8

9

10 11 12 13

0

1

2

3

4

5

6

7

8

9

10 11 12 13

Delinquency vs. no. of risk factors

.1

mean of delinquency

0

0

mean of violence .05 .1

.05

.15

.15

Violence vs. no. of risk factors

0

1

2

3

4

5

6

Violence vs. no. of protective factors

7

0

1

2

3

4

5

6

7

Delinquency vs. no. of protective factors

Fig. 3.1  Cumulative effects of number of risk factors and protective factors on probability of violence and delinquency

3.4  Conclusion Three decades of research showed us the potential and promise of preventing youth problem behaviours and disorders worldwide. Effective interventions for delaying or preventing the early onset of these problem paths can be implemented based on multiple factors that contribute to the development. A variety of risk and protective factors could be detected that, respectively, increase or decrease violence and delinquency. These factors also influence substance use, school dropout, depression and teen pregnancy. During the years, knowledge has been built up about these specific problem behaviours. Also, knowledge has been built up about the underlying risk and protective factors, how to define them and how to influence them in effective interventions. These predictors become core elements in the most effective interventions. We know that we can find them in the different social contexts in which youngsters grow up, that they are correlated and cumulative, that they have specific and overall effects and that their effects continue over time (Coie et  al. 1993; Catalano et al. 2012; Catalano and Hawkins 1996; Hawkins et al. 2002; Hawkins and Weis 1995; IOM 2009; Kellam and Rebok 1992; Lancet 2012; Rose 1992).

54

H. Jonkman and C. M. H. Hosman

The present study provides a unique opportunity to investigate problem behaviours and their predictors in Dutch community research, and we have presented the results for violence and delinquency. This study is a confirmation of what we know but it throws fresh light on it. We also showed the strong association of these two problem behaviours with other common youth problem behaviours: smoking ever, smoking now, smoking heavy, alcohol use ever, alcohol use now, binge drinking and depression. In this study we also determined the influence of risk and protective factors on violence and delinquency. We are not the first researchers who have done this, but this study demonstrates the strong association between problem behaviours, risk factors and protective factors. We looked at odds ratios of 14 risk factors and 7 protective factors (not all studied) in the domains of family, school, peers and community. We also determined the magnitudes of their impact, and we showed what the potential effects of preventive work might be if risk factors could be eliminated and protective factors enhanced. We also looked at the cumulative association of risk factors and protective factors. The findings of this prevention study suggest what we know from other studies; there are strong influences of risk factors on different problem behaviours. In the Netherlands, especially the Parental Attitudes (in the Family domain) and the Early Initiation and Interaction with specific Peers (in the Individual/Peer domain) seem to be important. The influence of protective factors is also seen (especially Attachment and Opportunities for Prosocial Involvement in the Family are important in the Netherlands), but this influence is smaller. In this study we not only investigate the associations between different problem behaviours, and risk and protective factors. We show also how strong the magnitude of these associations is and, especially interesting from a prevention perspective, what outcomes we can expect when our preventive reactions are effective (as the population attributable fraction). In the end we underlined the importance of the cumulative effect of risk and protective factors. Figure 3.1 shows that the effect of risk factors is not linear but exponential. This finding of a curvilinear relation between the number of risk factors and both violence and delinquency reveals an interesting implication for prevention policy. As we showed, reducing the number of risk factors from 13 to 11 (taking out the strongest predictors) could already reduce the risk of violence and delinquency by 40–50%. This suggests that prevention policy that targets only a small number of important risk factors, and is able to successfully reduce their prevalence, could result in a rather large preventive effect. In translating the results of this study into effective and cost-effective prevention policies, the causal relationships between risk and protective factors and problem behaviours are crucial. From well-designed developmental, experimental and meta-­ analysis research, it is known that risk factors are causally related over the life span and between system levels. Early risk factors could trigger risk factors during later age periods. By targeting early risk factors with a long-term broad-spectrum effect (i.e. multifinality) in the end, a much wider range of risk factors could be reduced over time. This present study is unique because it investigates violence and delinquency in a comprehensive perspective. These problem behaviours are related to other

3  The Netherlands

55

common youth problems and related to the most important factors from the daily domains in which youngsters grow up. We not only surveyed individual influences but also took into account the influences of the community. In this study we used reliable and valid instruments which are proven and can be compared to each other. We used modern statistical methods to say something about influences as well as something about prevention possibilities. However, we know that this research is limited for several reasons. Firstly, it is important to mention that this is not a nationally representative study. We used data collected in 1  year from quite similar communities in the south and west of the Netherlands. In the future we need more representative data. Secondly, based on earlier longitudinal and effect research, we assume that the studied risk factors and protective factors have a causal influence on the development of violence and delinquency. However, the CTC-YS does not provide prospective data. The associations with outcome indicators that we found in this study point strictly only at possible causal factors. Thirdly, in community studies the influences of risk and protective factors are often separately researched, as is done in this study, and less studied in connection with each other. New studies should explicitly study interactions between both types of factors, such as different kinds of buffering effects. Fourth, in further studies we should also take into account more seriously the influence of structural indicators of communities, such as poverty, socioeconomic status of the environment and social capital, in these community studies. The influence of the broader social context on risk factors, protective factors and problem behaviours has been given too little attention in prevention science up until now. Violence and delinquency are targets for social policy and prevention. The burden of suffering cannot be solved only by individual care, which receives a big part of the money. For successful social policy and prevention, we need a clear perspective on where, how and to whom early care should be given. Community intervention which influences social determinants of behavioural problems is important. Risk and protective factors are important targets in successful community work as this study confirms once again. Evidence-based policy is based on the premise that there is enough research done to guide community decision-making. The best available knowledge should be incorporated into what is done in practice. This knowledge is important for evidence-based practice, for the advancement of prevention science and to support the healthy development and future of all children. Acknowledgements  We like to thank Majone Steketee and Claire Aussems for their work on the Dutch CTC-YS and research.

References Arthur, M.  W., Briney, J.  S., Hawkins, J.  D., Abbott, R.  D., Brooke-Weiss, B.  L., & Catalano, R. F. (2007). Measuring risk and protection in communities using the communities that care youth survey. Evaluation and Program Planning, 30, 197–211.

56

H. Jonkman and C. M. H. Hosman

Catalano, R. F., Berglund, M. L., Ryan, J. A. M., Lonczak, H. S., & Hawkins, J. D. (2004). Positive youth development in the United States: Research findings on evaluations of positive youth development programs. Annals of the American Academy of Political and Social Science, 591, 98–124. Catalano, R. F., Fagan, A. A., Gavin, L., Greenberg, M. T., Erwin, C. E., Ross, D. A., et al. (2012). Worldwide application of prevention science in adolescent health. The Lancet, 379(9826), 1653–1664. Catalano, R.  F., & Hawkins, J.  D. (1996). A theory of anti-social behavior. In J.  D. Hawkins (Ed.), Delinquency and crime: Current theories (pp.  149–197). New  York: Cambridge University Press. CBS. (2011). Kerncijfers wijken en buurten 2011. Den Haag: Centraal Bureau voor de Statistiek. Cohen, P. (1996). Childhood risks for young adult symptoms of personality disorder: Method and substance. Multivariate Behavioral Research, 31(1), 121–148. Coie, J. D., Watt, N. F., West, S. G., Hawkins, J. D., Asarnow, J. R., Markman, H. J., et al. (1993). The science of prevention: A conceptual framework and some directions for a national research program. American Psychologist, 48, 1013–1022. Elliott, D. S. (Ed.). (1997). Blueprints for violence prevention (volumes 1–11). Boulder: Centre for the Study and Prevention of Violence, Institute of Behavioral Science, University of Colorado. Elliott, D. S., & Tolan, P. H. (1999). Youth violence, prevention, intervention and social policy. In H. Flannery & C. R. Huff (Eds.), Youth violence: Prevention, intervention and social policy (pp. 2–47). Washington, DC: American Psychiatric Association. Garmezy, N., & Masten, A. S. (1985). Risk, vulnerability, and protective factors in developmental psychopathology. Advances in Clinical Psychology, 8, 1–52. Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models. Cambridge: Cambridge University Press. Glaser, R. R., Van Horn, M. L., Arthur, M. W., Hawkins, J. D. & Catalano, R. F. (2005). Measurement properties of the communities that care youth survey across demographic groups. Journal of Quantative Criminology, 21(1), 73–102. Hawkins, J. D., Catalano, R. F., & Arthur, M. W. (2002). Promoting science-based prevention in communities. Addictive Behaviors, 27(6), 951–976. Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112(1), 64–105. Hawkins, J. D., & Weis, J. G. (1995). The social development model: An integrated approach to delinquency prevention. Journal of Primary Prevention, 6(2), 73–97. Hosman, C., & Jané-Llopis, E. (1999). Political challenges 2: Mental health. In The evidence of health promotion effectiveness: Shaping public health in a new Europe (pp.  29–41). Paris: International Union for Health Promotion and Education. Jouve Composition and Impression. Hox, J. J., Moerbeek, M., & Schoot, R. (2018). Multilevel analysis: Techniques and applications (3rd ed.). Mahwah: Lawrence Erlbaum. Ince, D., Beumer, M., Jonkman, H., & Vergeer, M. (Eds.). (2004). Veelbelovend en effectief: Overzicht van preventieve projecten en programma’s in de domeinen gezin, school, jeugd. Utrecht: NIZW. Institute of Medicine. (2009). Report on preventing mental, emotional, and behavioural disorders among young people: Progress and possibilities. Washington, DC: Institute of Medicine. Jonkman, H. (2012). Some years of communities that care: Learning from a social experiment. Amsterdam: VU-University. Jonkman, H., Boers, R., van Dijk, B., & Rietveld, M. (2006). Wijken gewogen: Gedrag van jongeren in kaart gebracht. Utrecht: NIZW/SWP. Junger-Tas, J., Marshall, I. H., Enzmann, D., Killias, M., Steketee, M., & Gruszcunka, B. (2013). The many faces of youth crime: Contrasting theoretical perspectives on juvenile delinquency across countries and cultures. New York: Springer.

3  The Netherlands

57

Kellam, S. G., & Rebok, G. W. (1992). Building developmental and etiological theory through epidemiologically based preventive intervention trials. In J. McCord & R. E. Tremblay (Eds.), Preventing antisocial behavior: Interventions from birth through adolescence (pp. 162–195). New York: Guilford Press. Lancet. (2012). Editorial: Putting adolescents at the Centre of health and development. The Lancet, 379, 1561. Loeber, R., & Farrington, D. P. (Eds.). (1998). Serious and violent juvenile offenders: Risk factors and successful interventions. Thousand Oaks, CA: Sage. Loeber, R., & Farrington, D. P. (Eds.). (2001). Child delinquents: Development, intervention and service needs. Thousand Oaks, CA: Sage. Pollard, J. A., Hawkins, J. D., & Arthur, M. W. (1999). Risk and protection: Are both necessary to understand diverse behavioural outcomes in adolescence? Social Work, 23, 145–158. Rabe-Hesketh, S., & Skondal, A. (2012). Multilevel and longitudinal modeling using Stata (2nd ed.). College Station: Stata Corporation. Rose, G. (1992). The strategy of preventive medicine. Oxford: Oxford University Press. Rutter, M. (1985). Resilience in the face of adversity: Protective factors and resistance to psychiatric disorder. British Journal of Psychiatry, 147(6), 598–611. Sherman, L. W., MacKenzie, D., Gottfredson, D. C., Eck, J., Reuter, P., & Bushway, S. D. (1996). Preventing crime: What works, what doesn’t, what is promising. Washington, DC: National Institute of Justice. Stata Corporation. (2017). Stata statistical software: Release 15. College Station: Stata Corporation. Steketee, M., Jonkman, H., Berten, H., & Vettenburg, J. (2013). Alcohol use among adolescents in Europe: Effective environmental strategies in prevention. Utrecht: Verwey-Jonker Institute. Werner, E. E., & Smith, R. S. (1992). Overcoming the odds: High risk children from birth to adulthood. Ithaca: Cornell University Press.

Chapter 4

Germany Frederick Groeger-Roth and Burkhard Hasenpusch

4.1  Introduction In Germany, CTC started in the year 2008 as an initiative of the state government in 1 of the 16 “Länder”, the State of Lower Saxony (Groeger-Roth 2012). The leading organization was the Crime Prevention Council of Lower Saxony (CPC, located in the Ministry of Justice). Other ministries were involved in a steering committee for the CTC pilot in 2009–2012 (Ministries of Interior, Education and Social Affairs). Some ministries were quite reluctant at the beginning. Stronger support was coming from the Ministry of Education, looking for opportunities to implement better prevention measures in schools. The Ministry of Social Affairs provided additional funding in the pilot phase. Scientific institutions were involved as contractors and were important partners in conducting the CTC-YS (Arpos Institute in Hannover) and for conducting an external evaluation of the pilot project (University for Applied Science in Cologne). Several key persons from the community level (local authorities) were also involved in the pre-planning phase to discuss the need for CTC and possible frameworks where CTC could be used. In particular the chief officers of the youth departments of some of the pilot cities played an important role by facilitating CTC in their respective municipalities. The reasons for starting with CTC were mainly due to prevention policy issues: the lack of coordination of services and decision-­making inside the diverse “silos” of health, school, police and youth care sectors. Already established prevention network structures at the local level were partly beginning to regress, because of the perceived ineffectiveness by the participants (Schreiber F. Groeger-Roth (*) Ministry of Justice of Lower Saxony/Crime Prevention Council of Lower Saxony, Hannover, Germany e-mail: [email protected] B. Hasenpusch Ministry of Justice of Lower Saxony, Langenhagen, Germany © Springer Nature Switzerland AG 2021 D. P. Farrington et al. (eds.), Delinquency and Substance Use in Europe, https://doi.org/10.1007/978-3-030-58442-9_4

59

60

F. Groeger-Roth and B. Hasenpusch

2007). The establishment of new prevention coalitions at the local level was rare in previous years, and the community (crime) prevention “movement” seemed to slowly weaken. Additionally at that time in Germany, several prevention programmes with a stronger empirical base of effectiveness were available (Beelmann et  al. 2014). However, these programmes were not widespread. This is particularly the case in areas where the available data was showing a persistence of behavioural youth problems: underage drinking (binge drinking), bullying in schools, mental health problems and the proportion of youth served by the youth care system. Communities That Care was first implemented in the years 2009 to 2012 in the context of a pilot project in three sites in Lower Saxony—a quarter within the city of Hannover (Mühlenberg), a quarter within the city of Göttingen (Weststadt) and four communities in the County of Emsland (Schubert et al. 2013). The pilot project was funded largely by the European Union (Prevention of and Fight against Crime Programme, European Commission—Directorate-General Home Affairs), together with the Ministries of Justice and Social Affairs of Lower Saxony and the Klosterkammer Foundation. Since the completion of the pilot project in 2012, several other communities have adopted the CTC approach in Lower Saxony with the technical and financial support of the Crime Prevention Council of Lower Saxony (Jonkman 2015). Process evaluation data shows that CTC was implemented in the sites to a great degree according to the original model (Schubert et al. 2013; Jonkman 2015). The CTC-YS results were used to prioritize elevated risk factors and depressed protective factors in the intervention areas to be targeted by the local stakeholders. This decision process could be facilitated by the establishment of a statewide representative CTC-YS every 2 years, starting in 2013 (Ludwig and Soellner 2013; Soellner et  al. 2016, 2018). Community stakeholders are using the statewide results as a baseline to determine if their factors are above or below average. After prioritizing up to five risk and protective factors, communities are trained to analyse their existing prevention efforts with regard to service gaps or duplications related to the prioritized factors (e.g. in specific age groups or service sectors). New interventions were selected to fill service gaps from a list of tested prevention programmes provided by the Crime Prevention Council (the “Green List of Prevention”, Groeger-­ Roth and Hasenpusch 2011). This web-based programme registry is organized in such a way that available prevention programmes can be selected according to the risk and protective factor and gap profile that is specific to a given community. All CTC sites are implementing new programmes, defined in their “CTC Action Plans”, as a result of the CTC process (examples are available on the German CTC website www.ctc-info.de). An evaluation of the effectiveness of CTC could not be completed to date. Two studies of CTC implementation in Germany are available at present. Schubert et al. (2013) summarized the results of the evaluation of the CTC pilot project (2009–2012) in Lower Saxony, and Jonkman (2015) studied the implementation of CTC in five sites in Lower Saxony after the pilot phase (2013–2014). Both studies focus on CTC process results and not on youth outcomes.

4 Germany

61

Schubert et al. (2013) also analysed the early planning stages in Lower Saxony. The three pilot sites were chosen to test the feasibility of implementing CTC under different circumstances: disadvantaged communities within a big city (Hannover) or in a middle-sized city (Göttingen) and communities in a more affluent rural area (County of Emsland). The evaluation provided information about community characteristics that may influence the CTC implementation, as well as data on the implementation process in the pilot sites and the use of CTC-specific instruments and tools. In all three pilot areas, CTC mobilized networks or coalitions of mainly professionals and some volunteers who were working with youth and families. It was beneficial to use already existing networks as the basis for the CTC effort. On some readiness issues, the evaluation found quite good starting conditions (quality of collaboration, willingness to support CTC, available financial resources), but there was also a challenge regarding the time resources of the professionals because of their regular workload. The focus on local networks and coalitions of mainly professionals facilitated the decision-making process. To what extent this lack of citizen involvement may be also a challenge in implementing the CTC action plans was beyond the scope of the evaluation study. Some problems regarding the involvement of local politicians in the key leader boards were also detected as a possible challenge for sustaining the CTC approach. The study concluded that the CTC method and structure was highly suitable within already established working routines and the regulatory framework within the German youth care system. The second study (Jonkman 2015) focused on CTC implementation in Lower Saxony after the pilot phase. The report provided an analysis and evaluation of CTC in five German communities: County of Osnabrück, Cities of Stadthagen, Oldenburg, Hameln and Nordstemmen. Three instruments that were also used in the US and Netherlands evaluation (Hawkins et al. 2008) were adapted to the German context: 1. In the “Community Key Informant Interview”, changes in thinking and working in prevention among the key leaders of the community were researched. The study detected an increased adoption of science-based prevention within 1 year of CTC implementation on a small scale. Key leaders found the analysis of the underlying problems more important, they started using this knowledge in their work, and they tried to combine this with the use of effective programmes in their community. 2. With the “Milestones and Benchmarks” instrument, by which the quality of CTC implementation (programme integrity) can be researched, the study found that CTC was implemented in Germany with high standards. Although not all the implementation phases could be researched, because of the limited time frame of the evaluation, the study showed positive results. Also it became evident that, with the success of the CTC implementation, the challenges increased. Success and effort seemed to be two sides of the same coin. 3. The third part was the evaluation of the community boards by the use of the “Community Board Interview”. This questionnaire covers different aspects of effective coalitions (e.g. board participation, effective collaboration, knowledge and competencies, leadership quality, tensions, etc.). The study identified

62

F. Groeger-Roth and B. Hasenpusch

p­ ositive developments regarding strengthening the work of the local coalitions through CTC, although not for all of the researched aspects. One of the major decisions to be made in the pilot phase was to decide on the CTC-YS to be used in the German context. One reason for this was the use of another school-based survey in some cities throughout the country, developed by the Criminological Research Institute of Lower Saxony (KFN). This was a paper and pencil survey developed for ninth graders in public schools, and it had widespread use in Germany (Baier et al. 2010). There are some overlaps between it and the original CTC-YS: measurement of problem behaviours like violence, delinquency and substance misuse, as well as some risk factors like family problems and delinquent peer associations. The reasons against the use of the KFN questionnaire included the length (two school hours instead of 1 h for the CTC-YS), with increased problems of facilitating the survey in the school context and the somewhat unsystematic collection of factors included in the questionnaire. In the following sections, we describe the adaptation and administration of the CTC-YS and some of its results, the bivariate and multivariate relations between the risk and protective factors and selected forms of problem behaviour by youths.

4.2  Method 4.2.1  Translating, Adapting and Administering the CTC-YS For the development of the German version of the CTC-YS, we translated the US version of 2006 and the Dutch version of 2004. Each individual item was scrutinized regarding its face validity in the new cultural context, since a literal translation might often not have captured the intended meaning of the item. This process took about 3 months and was documented in detail. This German synthesis of the US and Dutch questionnaires was administered in May 2009 to two classes (grades 7 and 9) in Hannover. Following the analysis of this pretest, several items that had obviously been misunderstood by some students were re-formulated, and the questionnaire was shortened so that it could be administered within one lesson of 45 min. After a second pretest with another class in February 2010, the questionnaire was shortened again slightly. The final questionnaire can be used in grades 6–12 and contains 156 items in 46 groups of questions, on socio-demographic variables, 5 forms of problem behaviour and scales measuring 16 risk and 10 protective factors. On the whole, this questionnaire is very similar to the one described in Chap. 3 on the Dutch CTC-YS, as was the procedure for dichotomizing the scores for the risk and protective factors at the median. The students’ responses regarding problem behaviour—violence, delinquency, use of drugs (alcohol, tobacco and hash), early pregnancy and depression— were recoded into binary variables, again similar to the Dutch approach.

4 Germany

63

The scales on risk and protective factors differ to some extent from the Dutch questionnaire for two reasons: 1. Some of the factors used in the Netherlands were not applicable in Germany (e.g. “Gang Involvement”) or were split into two separate factors (e.g. “Perceived Availability of Drugs and Weapons”), and some factors were added from the US version (e.g. “Peer Rewards for Problem Behaviour”, “Sensation Seeking” and “Interaction with Pro-social Peers”). 2. As a rule, in this comparative endeavour, scales with Cronbach’s alpha lower than 0.60 were excluded from the analysis. In our case, this applied to “Parents’ Attitudes Favourable to Drug Use”, “Transition and Mobility”, “Opportunities for Pro-social Involvement at School”, “Rewards for Pro-social Involvement at School”, “Belief in the Moral Order”, “Social Skills”, “Interaction with Pro-­ social Peers”, and “Opportunities for Pro-social Involvement in the Community”. In the Netherlands, this rule eliminated other scales, so that unfortunately using equal criteria for quality of measurement led to different sets of risk and protective factors in different countries, making cross-national comparisons difficult. Administering the survey as a “paper and pencil questionnaire” aiming at representative results for small geographic units would have been far too expensive for our budget. Therefore, we conducted an online survey once we were certain that all potentially participating schools had the facilities needed for administering the survey online for a whole class at one time. The survey was conducted by an external research institute, which also performed the initial screening of the data and some analyses of the prevalence of problem behaviour. To our knowledge, this was the first time that a large-scale survey of students was conducted online in Germany. This method saves time and money, since it avoids the laborious importing of filled­in questionnaires.

4.2.2  The Sample Originally, the sample consisted of 4364 respondents, of whom 500 came from the city of Hannover, 384 from the city of Göttingen and 3480 from the largely rural county of the Emsland. A total of 47 schools in these 3 sites agreed to participate in the survey, with at least 1 class in the grades 6–10 or 12, depending on the kind of school. In some schools, however, not all teachers actually conducted the survey in their class. Some of the participating students did not indicate which school they were attending or which grade they were in, and others gave obviously erroneous answers to these questions. Some schools were vocational schools for which we do not have data on the size of the classes. Thus, we could only use 35 schools with 606 classes in which we could identify 3734 students by school and grade for estimating the rate of participation. This rate was about 58% in Hannover, 60% in Göttingen and 74% in the County of Emsland.

64

F. Groeger-Roth and B. Hasenpusch

Table 4.1  Breakdown of the sample by demographic criteria

Demographic subgroup Boys  Age group 12–14 years (N = 1012, 52.4%)  Age group 15–17 years (N = 920, 47.6%)  Boys total (N = 1932, 50.5%) Girls  Age group 12–14 years (N = 966, 50.9%)  Age group 15–17 years (N = 931, 49.1%)  Girls total (N = 1897, 49.5%) Age group 12–14 years total (N = 1978, 51.7%) Age group 15–17 years total (N = 1851, 48.3%) Total (N = 3852)

Hannover % % N (r) (c) 57

Göttingen % % N (r) (c)

Emsland % N (r)

5.6 12.7 113 11.2 31.2 842

% (c)

All sites % N (c)

83.2 27.7 1012 26.3

178 19.3 39.7 57

6.2 15.7 685

235 12.2 52.5 170

8.8 47.0 1527 79.0 50.2 1932 50.2

69

7.1 15.4 102 10.6 28.2 795

141 15.1 31.5 88

9.5 24.3 702

74.5 22.5 920

23.9

82.3 26.1 966

25.1

75.4 23.1 931

24.2

210 11.1 46.9 190 10.0 52.5 1497 78.9 49.2 1897 49.2 126

6.4 28.1 215 10.9 59.4 1637 82.8 53.8 1978 51.3

319 17.2 71.2 145

7.8 40.1 1387 74.9 45.6 1851 48.1

448 11.6

9.4

362

3042 79.0

3852

Notes: Subgroups do not add up due to missing values. %(r) = row percent. %(c) = column percent

From the original sample of 4364, we excluded 87 students who were younger than 12 and 399 who are older than 17, as well as 26 who had not indicated their age, so that a total of 3852 cases remained in the sample. Of these, 279 students were 12 years old, 348 were 13, and 385 were 14, forming the younger age group in further analyses, and 384 were 15 years old, 339 were 16, and 197 were 17, forming the older age group. In some analyses, the sample was smaller due to missing values. Table 4.1 shows that, on the whole, both age groups and sexes were equally represented in the sample. However, in Hannover the older age group is clearly over-­ represented (71.2%). More importantly, with 79% of all cases, the students from the County of Emsland clearly dominate the sample.

4.3  Results 4.3.1  B  ivariate Analyses of the Relations Between Risk and Protective Factors and Problem Behaviours Table 4.2 shows the prevalence of the problem behaviours in the demographic subgroups of the sample. Use of hash was omitted because it was admitted too rarely to be analysed (only 1.1% of all cases). As one might expect, most of the problem

8.1 8.6 8.3 19.6 20.7 24.8 16.0 19.8 20.0

78 80 158 387 384 111 58 602 771

84 113 197 245 393 103 44 491 638

159 279 438

29.8 32.8 31.3

302 302 604 8.7 12.1 10.4 12.4 21.2 23.0 12.2 16.1 16.6

15.7 30.3 22.7

Delinquency N %

Violence N %

Note: Subgroups do not add up due to missing values

Demographic subgroup Boys  Age group 12–14 years (N = 1012)  Age group 15–17 years (N = 920)  Boys total (N = 1932) Girls  Age group 12–14 years (N = 966)  Age group 15–17 years (N = 931)  Girls total (N = 1897) Age group 12–14 years total (N = 1978) Age group 15–17 years total (N = 1851) City of Hannover (N = 448) City of Göttingen (N = 362) County of Emsland (N = 3042) Total (N = 3852)

Table 4.2  Prevalence of problem behaviour by age group, sex and site

58 228 286 118 490 109 24 475 608

60 261 321 6.0 24.5 15.1 6.0 26.5 24.3 6.6 15.6 15.8

5.9 28.4 16.6

Smoking N %

87 472 559 224 1065 152 57 1080 1289

135 590 725 9.0 50.7 29.5 11.3 57.5 33.9 15.7 35.5 33.5

13.3 64.1 37.5

Alcohol use N %

158 478 636 348 1084 158 61 1213 1432

188 603 791

16.4 51.3 33.5 17.6 58.6 35.3 16.9 39.9 37.2

18.6 65.5 40.9

Binge drinking N %

418 412 830 722 650 157 118 1097 1372

300 237 537

43.3 44.3 43.8 36.5 35.1 35.0 32.6 36.1 35.6

29.6 25.8 27.8

Depression N %

4 Germany 65

66

F. Groeger-Roth and B. Hasenpusch

behaviours are more prevalent among boys, with the exception of smoking (with very similar prevalences) and depression (which was much more prevalent among girls). In terms of odds ratios, being male increases the risk of violence by a factor of about 5 and the risk of being delinquent by a factor of 2 (in younger students) or 3 (in older ones). Higher prevalences in the older age group are also not surprising. This applies in particular to delinquency, smoking, the use of alcohol and binge drinking. The high prevalence of deviance among students from Hannover is easily explained by the dominance of the older age group there. Table 4.3 shows the relations between the problem behaviours. Most of these are statistically significant, with the exception of depression, which was only weakly related to the other problem behaviours. Smoking, the use of alcohol and depression were neglected in further analyses due to limitations of space. Intraclass correlation coefficients for problem behaviours in relation to 22 communities out of which we had at least 20 cases were computed in order to assess how strongly the communities resembled each other. They ranged from 0.01 (depression), 0.02 (delinquency) and 0.03 (violence) to 0.06 (use of alcohol), 0.09 (binge drinking), and 0.11 (smoking). The first three correlations were considered to be negligible, but it seems that some of the variance in smoking was explained by the communities in which the students lived, a finding which may warrant further research in the future. Of the 22 risk and 11 protective factors included in the CTC-YS, only 20 risk and 5 protective factors were retained, because the others had insufficient reliability (Cronbach’s alpha below 0.60). These 25 factors had a mean odds ratio of 3.3 (over all factors, age groups and problem behaviours). The most prominent risk factors were those in the peer and individual domains. Their mean odds ratio was 5.6 compared to 2.0 for the remaining factors. Among them, “Early Initiation of Antisocial Behaviour” and “Early Initiation of Drug Use” had the highest mean odds ratios (8.0 and 9.9), confirming again that past behaviour is the best predictor of future behaviour. Taken together, the protective factors had a considerably lower mean odds ratio (1.6) than the risk factors (3.7). Intraclass correlation coefficients for community risk and protective factors in relation to 22 communities out of which we had at least 20 cases were also computed. They ranged from 0.01 (CR 4, Perceived Availability of Drugs, and CP 2, Rewards for Pro-social Involvement) and 0.02 (CR 1, Low Neighbourhood Attachment, and CR 7, Norms Favourable to Antisocial Behaviour) to 0.04

Table 4.3  Relations between problem behaviours (odds ratios) Violence Delinquency Smoking Alcohol use Binge drinking

Delinquency 8.59

Smoking 4.53 5.95

Alcohol use 2.40 4.29 10.20

Binge drinking 2.46 4.02 8.73 31.04

Depression 1.55 1.82 1.63 0.99 n.s. 1.06 n.s.

67

4 Germany

(CR 6, Perceived Availability of Weapons) and even 0.13 (CR 2, Community Disorganization). The latter finding may warrant further research in the future. Table 4.4 shows the odds ratios for risk and protective factors versus problem behaviour outcomes, separately for younger and older age groups. Of the 150 odds ratios displayed in Table 4.4, only 12 were not significant, 5 of them referring to binge drinking in the older age group, where the mean odds ratio of 2.1 was comparatively low. Figures 4.1, 4.2, 4.3 and 4.4 show the relative percentages of violence, delinquency and binge drinking in relation to the number of risk and protective factors to which a person is exposed. On the x-axis, the number of students is shown above the number of factors present in this category. All three percentages in Fig. 4.1 show a marked increase as the number of risk factors rises; the prevalence of violence increases from 3% to 57%, delinquency increases from 1% to 58%, and binge drinking increases from 12% to 74%. Figure 4.2 shows the relationship between the sum of protective factors and the prevalence of problem behaviours. The effect here is much weaker; violence and binge drinking are reduced only by a factor of about 2.5, and delinquency is reduced by a factor of 5.2. Figure 4.3 shows the same results as Fig. 4.1, but is restricted to the group of individual and peer factors, which according to Table 4.4 have quite strong relations to the problem behaviours. In the case of violence, however, the increase is comparable to Fig. 4.1. In relation to binge drinking, it is about twice as strong, and in the case of delinquency, it is difficult to compare because, in the group without any of these factors, there is not one case reporting any delinquent behaviour. Of the individual and peer protective factors, only one, Religiosity, was retained in the analysis. Figure 4.4 portrays a rather weak connection between religiosity and 80 70 60 50 Violence

40

Delinquency

30

Binge Drinking

20 10 0

419

424

455

474

421

447

330

0-3

4-5

6-7

8-9

10 - 11

12 - 14

15 +

Fig. 4.1  Percent of problem behaviours versus number of risk factors

68

F. Groeger-Roth and B. Hasenpusch

Table 4.4  Odds ratios for risk and protective factors for problem behaviours versus age group (younger/older) Violence OR(y) (OR(o) Factor fr1—History of Problem 2.6 2.4 Behaviour fr2—Poor Family 2.0 1.8 Management fr3—Family Conflict 1.7 1.3 3.5 2.4 fr5—Parents’ Attitudes Favourable to Antisocial Behaviour sr1—Early Academic 1.3 1.5 Failure sr2—Low Commitment to 2.6 2.2 School ir1—Rebelliousness 3.4 3.4 10.8 17.4 ir3—Early Initiation of Antisocial Behaviour ir4—Early Initiation of 3.3 4.6 Drug Use 3.0 3.6 ir5—Favourable Attitude Towards Drugs 4.4 4.5 ir6—Favourable Attitude Towards Antisocial Behaviour ir7—Friends’ Use of Drugs 3.0 4.9 5.0 5.7 ir8—Interaction with Antisocial Peers ir9—Peer Approval for 2.9 1.7 Antisocial Behaviour ir10—Sensation Seeking 5.3 4.4 cr1—Low Neighbourhood 1.4 1.1 Attachment cr2—Community 2.8 2.3 Disorganization cr4—Perceived Availability 3.6 2.7 of Drugs cr7—Norms Favourable to 1.4 1.5 Antisocial Behaviour cr6—Perceived Availability 3.1 3.4 of Weapons fp1 * Family Attachment 1.7 1.6 fp2—Opportunities for 1.7 1.5 Pro-social Involvement fp3—Rewards for 1.3 1.3 Pro-social Involvement

Delinquency Binge drinking Mean Alpha/ OR(y) (OR(o) OR(y) (OR(o) OR items 3.4 2.7 2.4 1.4 2.5 0.71/4 3.0

2.3

2.8

1.7

2.3

0.75/7

2.9 2.8

1.4 2.6

1.9 2.1

1.1 1.4

1.7 2.5

0.70/3 0.70/4

1.7

1.4

1.5

1.3

1.5

0.64/2

3.5

2.5

3.1

1.7

2.6

0.72/6

4.7 8.0

4.1 6.8

4.7 3.4

2.2 1.9

3.8 8.0

0.72/3 0.62/5

8.9

5.0

31.5

6.1

9.9

0.68/5

0.47

4.9

10.8

5.4

5.6

0.74/4

0.31

5.2

3.5

2.1

4.4

0.80/5

5.7 7.6

5.0 4.7

14.9 3.2

5.4 1.9

6.5 4.7

0.61/4 0.81/8

3.6

2.4

5.2

2.0

3.0

0.78/4

6.7 2.0

4.9 1.1

4.0 1.6

3.7 0.9

4.8 1.3

0.63/2 0.72/2

2.7

1.9

1.9

1.0

2.1

0.83/6

4.6

2.8

5.0

2.4

3.5

0.84/4

1.7

1.4

1.2

1.0

1.4

0.69/3

3.3

2.7

2.8

1.7

2.8

2.8 3.1

1.7 1.7

2.4 2.2

1.5 1.3

1.9 1.9

1 item only 0.83/6 0.66/3

1.9

1.5

1.9

1.3

1.5

0.83/2 (continued)

69

4 Germany Table 4.4 (continued)

Violence Delinquency Binge drinking Mean Alpha/ OR(y) (OR(o) OR(y) (OR(o) OR(y) (OR(o) OR items 1.4 1.1 1.4 1.6 1.6 1.8 1.5 1 item only 1.1 1.1 1.5 1.2 1.6 1.1 1.3 0.73/3

Factor Ip4—Religiosity

cp2—Rewards for Pro-social Involvement Mean OR for problem 2.96 behaviour within age group

3.17

4.00

2.94

4.69

2.13

3.32

* Protective factors were recoded to facilitate comparison of their odds ratios with those of the risk factors: 1—protective factor not present, 0—protective factor present Note: OR in bold: significant at p