Youth Gangs in International Perspective: Results from the Eurogang Program of Research

Youth Gangs in International Perspective Finn-Aage Esbensen ● Cheryl L. Maxson Editors Youth Gangs in Internation...

Author: Finn-Aage Esbensen | Cheryl L. Maxson

13 downloads 729 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Youth Gangs in International Perspective

Finn-Aage Esbensen

●

Cheryl L. Maxson

Editors

Youth Gangs in International Perspective Results from the Eurogang Program of Research

Editors Finn-Aage Esbensen Department of Criminology and Criminal Justice University of Missouri-St. Louis St. Louis, MO, USA [email protected]

Cheryl L. Maxson Department of Criminology Law and Society University of California Irvine, CA, USA [email protected]

ISBN 978-1-4614-1658-6 e-ISBN 978-1-4614-1659-3 DOI 10.1007/978-1-4614-1659-3 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011943749 © Springer Science+Business Media, LLC 2012 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

We dedicate this volume of original research to our fearless leader, the individual who had the foresight to identify the need for, and importance of, multi-site, multi-method gang research: Old Silverback – Malcolm “Mac” Klein.

Contents

1

The Eurogang Program of Research and Multimethod Comparative Gang Research: Introduction ........................................ Finn-Aage Esbensen and Cheryl L. Maxson

Part I 2

3

4

5

6

7

1

Definitional Issues in the Comparative Context

Putting the “Gang” in “Eurogang”: Characteristics of Delinquent Youth Groups by Different Definitional Approaches ........................................................................ Kristy N. Matsuda, Finn-Aage Esbensen, and Dena C. Carson

17

Counting Gangs: Conceptual and Validity Problems with the Eurogang Definition ................................................................ Judith Aldridge, Juanjo Medina-Ariz, and Robert Ralphs

35

Gang Member: Who Says? Definitional and Structural Issues ............................................................................. Hannah Smithson, Leanne Monchuk, and Rachel Armitage

53

Five Decades of Defining Gangs in The Netherlands: The Eurogang Paradox in Practice ...................................................... Frank van Gemert

69

Gang Organization, Offending, and Victimization: A Cross-National Analysis..................................................................... David C. Pyrooz, Andrew M. Fox, Charles M. Katz, and Scott H. Decker Betwixt and Between Street and Prison Gangs: Defining Gangs and Structures in Youth Correctional Facilities............................................................................ Cheryl L. Maxson

85

107

vii

viii

Contents

Part II 8

Group Processes in the Comparative Context

Gang Dynamics Through the Lens of Social Identity Theory....................................................................................... Karen Hennigan and Marija Spanovic

127

9

Gang Membership: The Psychological Evidence ................................ Emma Alleyne and Jane L. Wood

10

The Onset of (Euro)Gang Membership as a Turning Point in the Life Course ......................................................................... Chris Melde and Finn-Aage Esbensen

169

The Sex Composition of Groups and Youths’ Delinquency: A Comparison of Gang and Nongang Peer Groups............................ Dana Peterson and Dena C. Carson

189

The Impact of Globalization, Migration, and Social Group Processes on Neo-Nazi Youth Gangs ........................................ Revital Sela-Shayovitz

211

Typically Moroccan? A Group Dynamic Explanation of Nuisance and Criminal Behavior ..................................................... Jan Dirk de Jong

225

11

12

13

Part III 14

15

16

Gang Depictions in Non-American Contexts

The Danish Gang-Joining Project: Methodological Issues and Preliminary Results ............................................................. Maria Libak Pedersen and Jonas Markus Lindstad The Stockholm Gang Intervention and Prevention Project (SGIP): Introducing a Holistic Approach to Gang Enforcement ............................................................................. Amir Rostami and Fredrik Leinfelt Are the Correlates and Effects of Gang Membership Sex-Specific? Troublesome Youth Groups and Delinquency Among Dutch Girls ................................................................................ Frank M. Weerman

Part IV

151

239

251

271

Future Directions

17

The Next Decade of Eurogang Program Research ............................. Malcolm W. Klein

291

18

The Intersection of Gang Definition and Group Process: Concluding Observations ...................................................................... Cheryl L. Maxson and Finn-Aage Esbensen

303

Index ................................................................................................................

317

Contributors

Judith Aldridge Centre for Criminology and Criminal Justice, School of Law, University of Manchester, Manchester, UK Emma Alleyne School of Psychology, Keynes College, University of Kent, Canterbury, Kent, UK Rachel Armitage Applied Criminology Centre, University of Huddersfield, Huddersfield, UK Dena C. Carson Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA Scott H. Decker School of Criminology and Criminal Justice, Arizona State University, Phoenix, AZ, USA Finn-Aage Esbensen Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA Andrew M. Fox School of Criminology and Criminal Justice, Arizona State University, Phoenix, AZ, USA Frank van Gemert Criminology Department, VU University, Amsterdam, The Netherlands Karen Hennigan Department of Psychology, University of Southern California, Los Angeles, CA, USA Jan Dirk de Jong Criminology Department, VU University, Amsterdam, The Netherlands Charles M. Katz School of Criminology and Criminal Justice, Arizona State University, Phoenix, AZ, USA Malcolm W. Klein Department of Psychology, University of Southern California, Los Angeles, CA, USA

ix

x

Contributors

Fredrik Leinfelt Stockholm Gang Intervention and Prevention Project, Stockholm County Police Department & Linnaeus University, Stockholm, Sweden Jonas Markus Lindstad Danish Ministry of Justice, Copenhagen, Denmark Kristy N. Matsuda Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA Cheryl L. Maxson Department of Criminology, Law and Society, University of California, Irvine, CA, USA Juanjo Medina-Ariz Centre for Criminology and Criminal Justice, School of Law, University of Manchester, Manchester, UK Chris Melde School of Criminal Justice, Michigan State University, East Lansing, MI, USA Leanne Monchuk Applied Criminology Centre, University of Huddersfield, Huddersfield, UK Maria Libak Pedersen Danish Ministry of Justice, Copenhagen, Denmark Dana Peterson School of Criminal Justice, University at Albany, Albany, NY, USA David C. Pyrooz School of Criminology and Criminal Justice, Arizona State University, Phoenix, AZ, USA Robert Ralphs Department of Sociology, Manchester Metropolitan University, Manchester, UK Amir Rostami Stockholm Gang Intervention and Prevention Project, Stockholm County Police Department & Linnaeus University, Stockholm, Sweden Revital Sela-Shayovitz David Yellin Academic College, The Hebrew University of Jerusalem, Jerusalem, Israel Hannah Smithson Department of Sociology, Manchester Metropolitan University, Manchester, UK Marija Spanovic Department of Psychology, University of Southern California, Los Angeles, CA, USA Frank M. Weerman Netherlands Institute for the Study of Crime and Law Enforcement (NSCR), Amsterdam, The Netherlands Jane L. Wood School of Psychology, Keynes College, University of Kent, Canterbury, Kent, UK

Chapter 1

The Eurogang Program of Research and Multimethod Comparative Gang Research: Introduction Finn-Aage Esbensen and Cheryl L. Maxson

The topic of youth gangs has received considerable scholarly attention in the United States as well as in other nations. Numerous publications have addressed the prevalence of, trends in, and responses to youth gangs within specific national contexts. Relatively missing from this field, however, is examination of the youth gang phenomenon across multiple societal and cultural contexts, although several notable exceptions exist (e.g., Covey 2003; Decker and Weerman 2005; Hagedorn 2008; Klein et al. 2001; van Gemert et al. 2008). In this edited volume, we hope to contribute to this evolving body of research. Specifically, this volume addresses important issues in gang research and represents work by a number of youth gang scholars from a variety of countries and disciplines. The chapters in this book (1) provide unique insights into definitional and measurement issues, (2) investigate group processes that distinguish youth gangs from other law-violating youth groups, and (3) describe interesting and contemporary gang research conducted in seven different nations (Denmark, Israel, the Netherlands, Sweden, Trinidad and Tobago, the United Kingdom, and the United States).

F.-A. Esbensen (*) Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA e-mail: [email protected] C.L. Maxson Department of Criminology, Law and Society, University of California, Irvine, CA, USA e-mail: [email protected] F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_1, © Springer Science+Business Media, LLC 2012

1

2

F.-A. Esbensen and C.L. Maxson

This book is the fourth in a series of edited volumes produced by the Eurogang Network, a cross-national collaboration of researchers, including European and American scholars, devoted to comparative and multimethod research on youth gangs and troublesome youth groups. The Eurogang web site (http://www.umsl.edu/~ccj/ eurogang/euroganghome.htm) notes three primary objectives of this network: 1. To build a foundation of knowledge regarding the socioeconomic conditions and institutional processes that foster or curtail the emergence and persistence/dissolution of youth gangs and problematic groups. 2. To construct an infrastructure for comparative, multimethod, cross-national research on youth violence in group contexts. 3. To disseminate and effectively utilize knowledge to inform the development of effective local, national, and international responses to emerging youth crime and violence issues (retrieved 7/26/11). Throughout our series of workshops, certain themes surface again and again in our discussions of definitional and methodological issues. We also engage in repeated discussions about the group processes that occur within youth gangs and how these might contribute to high levels of offending. As such, we devoted our tenth Eurogang meeting (June 2010) in Neustadt an der Weinstrasse, Germany, to presentations and discussions of these core issues associated with comparative research on youth gangs as well as site-specific descriptions of gang research in a variety of European settings. The presentations at that workshop provide the foundation for this book, supplemented by a few contributions from nonattending Eurogang members. All of the chapters represent original contributions to the field. Some of the chapter authors are well-known scholars in the field, while others are breaking new ground by being the first in their countries to raise interest in and conduct research on youth gang issues. Authors come from a variety of scholarly disciplines as well as cultural perspectives, providing rich and diverse views of the youth gang phenomenon. The works presented in the chapters also represent various and rich research traditions, including ethnographic methods, self-report surveys or interviews of youth, surveys or interviews with law enforcement, official records data, and victim interviews. Importantly, and unique among the few books that address gangs outside the United States, all of the authors utilize the Eurogang Program’s definition of a youth gang so that while the groups they describe are located in different countries, comparisons may be drawn between them.

1.1

Some Background Information on the Eurogang Program of Research1

The Eurogang Program was conceived during a small meeting in the béguinage2 in Leuven, Belgium, following a conference on restorative justice held in 1997. After publishing an exploratory article about the presence of gangs in Europe (Klein 1996), Malcolm Klein convened a small group of scholars to discuss how the study

1 The Eurogang Program of Research and Multimethod Comparative…

3

of street gangs in Europe might be fostered, and instigated another exploratory meeting later that year in San Diego, California. Following the enthusiastic response of researchers who were present at these meetings, and from others who were unable to attend, several of us organized the first Eurogang workshop in Schmitten, Germany, in September 1998. More than 40 people from 13 nations attended Eurogang I. Guided by the presentations of prepared papers, meeting participants learned about state-of-the-art gang research in the United States and Europe, as well as a number of European studies of other youth groups that might or might not be called street gangs. These presentations acted as a catalyst for the first of many lengthy discussions about gang definitions and research methods. Perhaps, the primary issue at this meeting was the sensitivity to the topic of gang existence in Europe: whether gangs existed in Europe. Some observers did not see the entities they expected based on sensationalized, mediadriven accounts of gangs in the United States of America. Adding to that were the understandable concerns that acknowledgement of European gangs might cause a “moral panic” that could stimulate a suppressive overreaction to the phenomenon. Nevertheless, there were many reports about troublesome youth groups that were recognized as clear examples of street gangs by the American gang researchers present at the meeting. Two important outcomes resulted from the Schmitten workshop. First was a spirit of enthusiasm for fostering systematic research on gangs in Europe. The second major product was the agreement to compile the presented papers and a few others into a published volume (Klein et al. 2001). The energy and intellectual curiosity of participants attending this initial gathering helped to establish a multinational collaborative that resulted in the Eurogang Program of Research. Since the Schmitten meeting, the group has engaged in a number of coordinated activities, including ten formal workshops (an 11th, known as EG XI, was convened in September 2011 in Hillerod, Denmark, and EG XII is scheduled in Stockholm, Sweden, during May 2012), numerous informal meetings and panels at professional conferences, the collaborative development of a package of instruments, two additional published volumes of research, and several funding applications for systematic, cross-national, multimethod studies of street gangs in Europe. An initial research design emerged from this first workshop that embraced multiple methods of gathering information about gangs, a commitment to implementing standardized designs across multiple sites, and a multilayered design structure that could provide city- and neighborhood-level contextual information to the more detailed accounts of gangs and youth groups. The fundamental characteristics of the Eurogang group process were also evident at this first workshop: numerous intense discussions of core issues in research and policy, respect for different perspectives, and a furtherance of social integration and communication among researchers. The second workshop convened in September 1999 in Oslo, Norway. This meeting proved pivotal in producing the organizational framework that would propel the methodological development of the instrument packages. Five instrument-based working groups were formed: city level descriptors, expert survey, youth survey, ethnography, and a program inventory. An additional group agreed to hammer out

4


the thorny definitional issues, and another took responsibility for seeking out funding opportunities for further meetings and research. There was little enthusiasm for a proposed workgroup on archival methods (using police or judicial data), and subsequent attempts to promote interest have not been successful. Oslo meeting participants aligned themselves with at least one, and sometimes several, working groups. Group facilitators volunteered to guide the groups toward draft instruments. We recruited others interested in participating that were not in attendance at the Oslo meeting. Within each working group, between 10 and 18 individuals participated in the instrument development discussions during the most active periods. It was understood that although much of the work would be accomplished within the working groups, it was critical to adopt common measures across different instruments. Therefore, the final decisions were made in plenary sessions. The work of the groups was provided extra momentum by the convening of the third Eurogang workshop in Leuven, Belgium, held only a month after the Oslo meeting. Substantial discussion revolved around the question of whether multi-investigator, multisite, systematic ethnography was possible or desirable. The ethnography working group tentatively agreed to draft guidelines for ethnographic researchers rather than the more specific instruments that were in development in the city, expert, and youth working groups. After this meeting, we used electronic communication to further refine instrument contents. Initial drafts of most of the instruments were introduced and discussed at the fourth Eurogang workshop that was held in the autumn of 2000 in the Netherlands, at the North Sea resort town of Egmond aan Zee. During this meeting, attention focused on the content of the instruments. Each instrument working group proposed a series of topics to be discussed by the larger group. Furthermore, a first attempt was made to arrive at a common definition of gangs/troublesome youth groups, the subject of study within the Eurogang Program. The following workshops, the fifth and the sixth Eurogang meetings, were very intensive. Both were held at the Correctional Training Facility of Bavaria in Straubing, Germany, in the summers of 2002 and 2003. During the fifth Eurogang meeting, we reached consensus after lengthy discussions on a Eurogang definition to be adopted in Eurogang work. The first drafts of the instruments were elaborated into complete questionnaires and protocols. The sixth Eurogang meeting served as a platform to (1) review the results of pretests of the instruments, (2) discuss translation issues, (3) coordinate common measures across different instruments, and (4) refine the content of the instruments. In addition, a draft of a gang prevention or intervention program inventory was introduced. During these meetings, participants began to address the issue of use policies for the instruments. Would use be limited to the Eurogang participants who had worked so hard to develop them, or would we make the instruments available to all who might want to use them? If there was free access, how could we learn from others’ experiences to refine the instruments? Another topic of discussion concerned the way in which different research traditions and different locations might handle informed consent procedures. Finally, we formalized the creation of the Eurogang electronic Listserv (currently there are over 235 subscribers) and the Eurogang web site.


5

Both meetings in Straubing included presentations on the first research studies which employed the Eurogang definition of gang/troublesome youth group and early versions of the proposed instruments. A number of these presentations, together with a few additional chapters, comprise the second volume of Eurogang research (Decker and Weerman 2005). We convened the seventh Eurogang workshop in 2004, for the first time in the United States in the city of Albany, New York. This meeting focused on a substantive issue: the role of violence in street gangs. Findings presented at this meeting were published in the European Journal of Criminology (Klein et al. 2006). Also, we discussed the results of a final round of pretests. The participants concluded that the instruments should be deemed “final” and made available to any interested researcher. Subsequently, the five instruments and the definitions essay were posted on the Eurogang web site, but electronic access remains contingent upon contact with the Eurogang’s “Use Master” so that the experience with the instruments can be tracked. The Albany workshop marked the close of the instrument development phase of the Eurogang Research Program and the beginning of Phase II, wherein Eurogang participants shifted focus to the conceptual and empirical issues that engage us as researchers and scholars. This change was evident in the eighth Eurogang workshop held in May 2005 in the Basque City of Oñati in Spain. This meeting included substantive presentations that addressed the issues of migration and ethnicity. Many of these presentations, together with others, comprise the third volume of Eurogang research (van Gemert et al. 2008). A ninth Eurogang workshop was convened in Los Angeles, California, in May 2008. This workshop had the purpose of raising the visibility of the Eurogang Program to US researchers and engaging a new generation of young scholars with Eurogang activities. To support these aims, Malcolm Klein produced a monograph which imagines a Eurogang study of the fictional community of Euroburg and illustrates many of the issues encountered while conducting cross-national, multimethod research on youth gangs (Klein 2009).

1.2

The Eurogang Definition of Gang Membership

Underlying all of the other instruments is the consensus Eurogang definition of a street gang/troublesome youth group. During the second Eurogang meeting in Oslo, a separate working group was established, called the Definitions group. This group was charged with the task of accomplishing what 70 years of American gang research had not been able to accomplish—providing a consensus definition of what constitutes a gang. Malcolm Klein spearheaded this daunting task. Following 3 years of discussions and a number of draft definitions, consensus was reached on the definitive wording of the definition during a plenary session at the fifth Eurogang meeting: A street gang is “any durable, street-oriented youth group whose involvement in illegal activity is part of its group identity.”

6

1.2.1


Development of the Definition: Definers and Descriptors

The process of developing a consensus definition began with agreement that the definition would have to be one that would attract widespread acceptance and one that would be suitable in multiple cultural contexts. Early in the deliberations, we made the important distinction between gang definers and gang descriptors. This distinction proved to be a significant step toward reaching agreement on a definition. Gang definers are those elements that are absolutely essential to characterize the group as a gang, while descriptors refer to those elements that help to describe specific characteristics of a particular group. For example, group names, colors, or symbols, and the use of tattoos are elements that are often ascribed to gangs and their members. However, does a group have to use a name to be considered a gang? Does a group have to adopt specific colors or symbols to make it a gang? And is it essential to have a tattoo in order to be a gang member? The answer to these questions is no. Although these characteristics might help to describe a gang or gang members, they are not essential elements of a gang. What then are the key components that define a gang? In the working group, the focus turned to such elements as size, age composition, location, stability, and group identity. To start with, it is clear that a gang is a group. Therefore, a gang must consist of more than one person. Although some law enforcement agencies consider two members sufficient to be a gang, most gang scholars agree that to constitute a gang, there must be at least three members. Given that our interest is in youth gangs, we must also place parameters on the age range of members. A group composed primarily of 20- and 30-year-olds would not fit well within the category of youth. Also, the types of behavior that evoke public and law enforcement concern need to be considered. We do not dispute the fact that many groups (primarily middle class and/or suburban youths) may be involved in troublesome and illegal behavior from time to time; however, the street-oriented aspect of gangs is what elicits fear and concern. In addition to this street-oriented nature of gangs, what other criteria help to differentiate a gang from other street-oriented groups such as mobs? One defining criterion is that gangs persist over some period of time. They do not assemble for just one day or one event. Groups that endure over time and those that dissipate virtually upon formation are qualitatively different. One other defining element of gangs is that they have a sense of “we-ness” or group identity. Without such an identity, we cannot speak of gang membership or gang activity. One persistent debate in the gang literature is the relevance of involvement in illegal activity. The working group was unanimous in its assessment that group involvement in illegal activity is a critical distinguishing element of youth gangs. Without this illegal activity, the group would not generate the policy interest that gangs currently and historically have. These defining elements, then, were combined to establish the Eurogang definition of a gang: “A street gang (or troublesome youth group corresponding to a street gang elsewhere) is any durable, street-oriented youth group whose involvement in illegal activity is part of its group identity.” In addition to the defining elements described above, we also included a phrase (“troublesome youth group corresponding to a street gang elsewhere”) that could be substituted for the word gang. We included this phrase because some researchers


7

are concerned that the word “gang,” or more specifically, its translation in their own language, like “bande” or “jeugdbende,” may not convey the same meaning as the word “gang.” It is possible that the public in one country has strong stereotypes or associations in mind that are connected to these words. The discussion about this problem was resolved by allowing local researchers to use the phrase “troublesome youth group,” or its translation, instead of the word “gang” if they want to prevent misconceptions by the general public. During Eurogang meetings and other conferences, however, researchers commonly refer to “gangs” as their subject of research. The defining elements just reviewed constitute what we refer to as the core or Level I measures. These should be incorporated into all of the methods of data collection. In addition to these elements, there are a number of descriptive characteristics that would be desirable to include in a study of street gangs. These descriptive items provide additional information about individual gang members and the gang as a group. We divided these descriptors into two additional categories, cleverly referred to as Level II and Level III, in which we ranked the desirability of their inclusion in a study of gangs or gang members. In other words, Level I consists of indicators that are required in all instruments, Level II are measures that are very desirable to collect but not strictly necessary, and Level III is a list of suggestions for interesting additional measures that can be included in a comparative gang study. We further grouped these variables as individual-level or group-level characteristics.

1.2.2

Individual-Level Characteristics

• Level I: Demographics • Level II: Family background, parental schooling, employment, prevention, intervention, suppression experiences, victimization history outside of gang, selfreported delinquency, illiteracy, parental monitoring and supervision, girl/ boyfriends, legal status/immigrant, proportion close friends in gang, ex-gang status, and siblings • Level III: SES background, personal networks beyond the gang, mental health, school and family attachment, and residence of family

1.2.3

Group-Level Characteristics

• Level I: Age, common group crimes, drug, alcohol use, duration, ethnic composition, gang or not, negative peer commitment, sex, group size, illegal activity, immigrant composition, name (what is it and who gave it), reasons for joining, street orientation, subgroups, term used, and territory • Level II: Attachment to group, entry and exit criteria, external antagonists, other groups that are present in location, fights with other groups, group values, history of gang, key events/incidents, roles, symbols and colors, and structured narratives • Level III: Class composition of gang, hanging out together, kinship, political orientation, and proportion of members colocated


8

1.2.4

Eurogang Instruments

As described in the preceding sections, each of the Eurogang working groups produced an instrument to be used in comparative gang research. Contributors to this volume rely on one or more of the following three instruments in their research: the expert survey, the youth survey, and the ethnography guidelines. Each of these instruments is described below. Expert survey. This compact questionnaire was developed to survey or interview (telephone or in person) local experts on the presence of street gangs/troublesome youth groups. These experts may be police officers, youth workers, or anybody else who has sound knowledge about the gangs in a neighborhood or city. The questionnaire covers the existence of gangs according to the Eurogang definition, information about the demographics and other characteristics of these gangs, and the gang type to which they belong. It provides a general picture of the amount and nature of gangs in a certain area. Youth survey. This questionnaire is intended to collect quantitative individual data from young people. The youth survey is designed to be administered in classrooms in secondary schools using paper-and-pencil methods, but it can also be administered in other settings as well (e.g., community or institutional samples). Several items determine whether respondents belong to a gang/troublesome youth group according to the Eurogang definition. Additional items cover structural and cultural characteristics of the group to which respondents belong. Ethnography guidelines. This is a set of guidelines and advice on how to collect qualitative information on one or more street gangs/troublesome youth groups employing ethnographic approaches. The guidelines are flexible, as researchers may prefer different methods (observational methods and/or in-depth interviews with gang members or key informants). Another part of the document offers a list of topics that needs to be addressed in a Eurogang ethnographic research project. These topics are focused on group characteristics, gang culture, individual members, and the historical and local context of the gang under study.

1.3 1.3.1

Overview of the Book Definitional Issues in Comparative Context

The first section of the book consists of six chapters that focus on definitional and methodological issues. Matsuda, Esbensen, and Carson’s chapter, “Putting the ‘Gang’ in ‘Eurogang’: Characteristics of Delinquent Youth Groups by Different Definitional Approaches,” offers a descriptive analysis of gang members resulting from different definitions of gang membership. They compare and contrast the qualities of gang members as defined by (1) self-nomination as a gang member, (2) the


9

Eurogang definition of gang membership, and (3) youth who report that their friends are gang members. Of particular interest in their research is the lack of overlap (or concurrence) in classifying survey participants as gang members. However, regardless of the definition used, gang members comprise a qualitatively different group than those not identified as gang-involved, suggesting a considerable robustness in each of these measures. Two of the chapters in this section provide a critique of the Eurogang definition, maintaining that it includes some groups that are “not gang like.” Aldridge, MedinaAriz, and Ralphs’ (“Counting Gangs: Conceptual and Validity Problems with the Eurogang Definition”) chapter reflects on the utility of the Eurogang definition across a number of British research projects. They suggest that “street orientation” should more properly be considered a descriptive—rather than defining—criterion for the categorization of gang members, and they raise validity concerns in relation to the key aspect of the Eurogang definition: that the group’s involvement in illegal activity is part of its group identity. Specifically, they cast doubt on the utility of this criterion as it would include classification of three groups included in their studies as gangs that the authors argue would be falsely identified as such, including a group of pot-smoking adolescents, clubbers, and illegal ravers. Along a similar vein, Smithson, Monchuk, and Armitage (“Gang Member: Who Says? Definitional and Structural Issues”) question whether experts can accurately identify gang and nongang groups. They provide a discussion of the structure and formation of gangs in Great Britain’s “North City” from the point of view of the young people identified as gang members. Based on interviews with the young people as well as those individuals responsible for the classification of these young people as gang members (i.e., police officers and multiagency panels), the authors question the extent to which the youth groups could or should be labeled as gangs. Findings demonstrated that few of the young people viewed themselves as belonging to a gang. In fact, many of the youth were resistant to the application of the label to their group. The next two chapters tackle the role of group organization as a defining feature of gangs. In the chapter by van Gemert, “Five Decades of Defining Gangs in the Netherlands: The Eurogang Paradox in Practice,” the author reviews five decades of publications on gangs in the Netherlands, with a focus on the manner in which gangs are conceptualized and described. Utilizing police reports and academic research, van Gemert examines the qualities and group characteristics that are used to classify groups as gangs, including, for example, members’ age, whether or not the group is territorial, whether or not the group has leaders, and group size and duration. He notes that the tendency among the police to emphasize the role of structure (i.e., leaders, hierarchy, formal rules, etc.) in conjunction with reliance upon American gang stereotypes has resulted in a lack of recognition of gang problems in the Netherlands. Pyrooz, Fox, Katz, and Decker examine the extent to which gang organization is related to individual levels of offending and victimization. Utilizing data from three studies (two from the United States and one from Trinidad and Tobago), the authors are able to explore the role of organizational characteristics on behavior within different national contexts. Among the findings reported in their chapter, “Gang

10


Organization, Offending, and Victimization: A Cross-National Analysis,” is the fact that youth gangs in the United States tend to exhibit more organizational features than the groups in Trinidad and Tobago. They also report that “while gang organization was associated with increased levels of delinquency and victimization, the findings were mixed across the research contexts,” thereby highlighting the importance of continued cross-national research to examine the interplay of cultural context and group structure. The last chapter in this section, “Betwixt and Between Street and Prison Gangs: Defining Gangs and Structures in Youth Correctional Settings,” Maxson describes her effort to gather systematic depictions of gangs in correctional institutions by utilizing the Eurogang research framework. The Eurogang definition and methods needed minimal alteration to make them useful in this context. In contrast with previous research, she finds that gangs in these institutional settings are more similar to street gangs than prison gangs.

1.3.2


Hennigan and Spanovic (“Gang Dynamics Through the Lens of Social Identity Theory”) describe how the influence of group norms can be magnified or marginalized by variations in an individual’s social identification and perception of the entitativity (the degree of “groupness”) of his/her own group. They explore the implications for behavioral choices made by gang- and non-gang-affiliated youth. Following the social psychological framework proposed in the Hennigan and Spanovic chapter, Alleyne and Wood argue that the study of gangs, which has been largely restricted to criminological or sociological perspectives, would benefit by incorporation of group psychological processes. In their chapter, “Gang Membership: The Psychological Evidence,” the authors examine the specific roles of moral disengagement and antiauthority orientations on behavior. They utilize a sample of London students to compare gang members, peripheral youth, and nongang youth to assess the social-cognitive processes associated with gang membership and varying rates of involvement in delinquency. Their findings suggest a developmental process whereby youth who become more gang-involved gradually adopt gang norms in terms of language and attitudes that facilitate offending. For instance, consistent with moral disengagement theory, gang members are more likely to assign blame to victims or to displace responsibility. Melde and Esbensen employ a life-course perspective to examine the effect of gang joining on attitudinal, emotional, and behavioral outcomes. Their chapter, “The Onset of (Euro)Gang Membership as a Turning Point in the Life Course,” views gang joining as a turning point in the adolescent experience that is associated with a change in life-course trajectories. They suggest that it is the change in attitudes, emotions, and routine activities associated with gang membership that account for the documented increase in offending among members. Utilizing prospective longitudinal data from a three-wave, multisite sample of over 1,400 youth,


11

the authors use the Eurogang definition of gang membership to replicate their earlier findings based on a different measure of gang membership (i.e., consider their group of friends to be a gang). Their findings lend further support to the robustness of the Eurogang definition of gang membership. The chapter by Peterson and Carson, “Group Sex Composition and Youths’ Delinquency: A Comparison of Gang and Non-gang Peer Groups,” examines the role of the sex composition of gangs and non-gang groups. Prior research has found that group sex composition may be a stronger predictor of delinquency than individual sex. In an earlier study, Peterson and colleagues, for instance, found that the delinquency of girls in sex-balanced gangs is suppressed by boys in those gangs. Boys in those gangs seek to maintain their dominant standing by excluding females from the activities that confer status in the gang. On the other hand, girls in majority-male gangs were found to commit crime at levels comparable to the boys. In the current chapter, Peterson and Carson extend this earlier study by employing the Eurogang definition of gang membership and by examining the effect of group sex composition of non-gang groups. They suggest that gender dynamics prevalent in youth gangs should be a reflection of those found in the larger society: that is, in non-gang peer groups. To test their hypotheses, they rely on self-report data from a multisite longitudinal study of 3,820 youths in the United States. In the next chapter, “The Impact of Globalization, Migration, and Social Group Processes on Neo-Nazi Youth Gangs,” Sela-Shayovitz reports on her study of a neoNazi group in Israel. Drawing on her in-depth interviews with gang members, their mothers, social workers, and police officers, Sela-Shayovitz examines the extent to which migration and the associated marginalization of these Russian born youths help to explain the emergence of their neo-Nazi beliefs. She concludes that “the experience of enduring social rejection as non-Jewish immigrants in a Jewish country generated a reaction formation and cultural resistance identity, namely the neo-Nazi ideology.” A galvanizing element for these young men was their exposure to neo-Nazi music and ideology via the internet. Sela-Shayovitz suggests that the role of the Internet was a central feature in the shaping of their gang identity and illegal activity. In the last chapter in this section, “Typically Moroccan? A Group Dynamic Explanation of Nuisance and Criminal Behavior,” de Jong tackles an important question regarding the labeling of Moroccan youth in the Netherlands as gang members. Over the course of the past 20 years, Moroccan immigrants, especially young males, have been targeted as the source of a disproportionate amount of juvenile crime and “nuisance” behavior. They congregate in public places and intimidate passersby. Relying upon one particularly widely publicized case, de Jong assesses the extent to which the group of Moroccan boys participating in the event could be classified as a gang. He further examines two competing explanations for the emergence of the behavior in question: a cultural perspective (the media portrayed the behavior as “typically Moroccan” and no further explanation would be required) and group dynamic/identity perspective (a collective need to protect themselves from outsiders contributes to a shared group identity). De Jong conducted years of ethnographic research with Moroccan youth in Amsterdam and is thereby able to provide intriguing insights into this particular case and other “typically Moroccan” events.

12

1.3.3


Gang Depictions in Non-American Contexts

The first chapter in this section, “The Danish Gang Joining Project: Methodological Issues and Preliminary Findings,” presents interesting research conducted by researchers at the Danish Ministry of Justice. Responding to concerns among citizens and politicians that motorcycle gangs were recruiting young adolescents into gangs, the ministry was commissioned to assess recruitment strategies into adult gangs. To achieve this objective, Pedersen and Lindstad adopted the Eurogang definition and methodologies to survey youths attending 18 schools in greater Copenhagen and also used the expert survey to interview more than 40 individuals knowledgeable about youth issues, including street workers, police officers, teachers, and youth workers. Findings reported in this chapter reveal remarkable similarities with regard to gang prevalence and gang characteristics to studies conducted in the United States and several other nations. Of significance is their finding that the youth identified as gang-involved using the Eurogang methodology were statistically and substantively more delinquent than youth classified as serious criminals (but not gang-involved). The second chapter in this section also hails from Scandinavia, a part of Europe that appears to be experiencing significant gang problems. In their chapter (“The Stockholm Gang Intervention Project: Introducing a Holistic Approach to Gang Enforcement”), Rostami and Leinfelt rely on police data to describe the recent development of gangs in Sweden. Interestingly, there has been a tendency among Swedish sociologists and criminologists to deny the presence of street gangs in their country, yet the police data suggest that local youth gangs in Stockholm (the capital city) have posed a problem at least since the late 1990s. In 1999, for example, the Stockholm County Police established a “gang commission” to address the growing gang problem in the city. This early attempt to respond to gangs was largely suppressive in nature, but it did lay the ground work for the eventual development in 2009 of a more balanced police response to gangs, the Stockholm Gang Intervention Program (SGIP). This strategy is based upon three core elements (1) research and science, (2) policing, and (3) collaboration with other authorities. The authors conclude their chapter with a detailed description of a central component of the SGIP, the PANTHER (Preventive Analysis about Network Targets for a Holistic Enforcement Response) model for gang intervention, which incorporates police suppression and enforcement with the various social science intervention and prevention concepts into one fully operational model against gang crime. The concluding chapter in this section provides an examination of the extent to which biological sex influences the gang experience. While a number of studies conducted in the United States have examined the role of sex and sex composition on gang behavior, this is an area that has not been subject to much empirical work in Europe. In his chapter, “Are the Correlates and Effects of Gang Membership SexSpecific? Troublesome Youth Groups and Delinquency among Dutch Girls,” Weerman replicates findings from the American studies. Among the findings from a study involving survey data from 1,830 Dutch adolescents are the following: risk


13

factors for gang membership are similar for boys and girls; gang girls are active in a variety of delinquent activities, similar to gang boys; and the sex composition of the gang appears to influence the extent to which both gang boys and gang girls engage in illegal activity.

1.3.4

Future Directions for the Eurogang Program

Prior to the concluding/summary chapter by Maxson and Esbensen, there is an invited chapter written by Malcolm Klein, “The Next Decade of Eurogang Program Research,” in which he reviews the short history of the Eurogang Program, discusses obstacles confronting truly comparative gang research, and identifies areas of emphasis for future research. He acknowledges that the Eurogang project has come a long way: from a small exploratory group of a handful of interested individuals to a multinational group of researchers and policymakers. The group has grown in size, has been durable, has a group identity, and, with this volume, has produced four edited volumes of gang research. But the underlying goal of the Eurogang program has not yet been met. The research to date, while using the Eurogang definition and methods, continues to be primarily single-site, singlemethod studies conducted in only one nation. Missing are multisite, multimethod, cross-national collaborations. Klein encourages such large-scale efforts that continue to utilize a common definition and also includes an evaluative component of gang policies and programs.

Notes 1. Portions of the following sections were previously drafted by the coeditors for publication of the Eurogang Manual Weerman et al. (2009). Eurogang Program Manual: Background, development, and use of the Eurogang instruments in multisite, multi-method comparative research. http://www.umsl.edu/~ccj/eurogang/euroganghome.htm. 2. Lacking suitable men following the Crusades, single women chose to live in communities called beguinages.

References Covey HC (2003) Street gangs throughout the world, 2nd edn. Thomas, Springfield Decker SH, Weerman FW (2005) Europeans street gangs and troublesome youth groups. Altamira Press, Walnut Creek Hagedorn JM (2008) A world of gangs: armed young men and gangsta culture. University of Minnesota Press, Minneapolis Klein MW (1996) Gangs in the United States and Europe. European Journal of Crime Policy and Research 4:63–80

14


Klein MW (2009) The street gangs of Euroburg: a story of research. Universe, Bloomington Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (2001) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Amsterdam Klein MW, Weerman FM, Thornberry TP (2006) Street gang violence in Europe. Eur J Criminol 4:413–437 van Gemert F, Peterson D, Lien I-L (2008) Street gangs, migration, and ethnicity. Willan, Uffculme, Devon Weerman FM, Maxson CL, Esbensen F, Aldridge J, Medina J, van Gemert F (2009) Eurogang program manual background, development, and use of the Eurogang instruments in multi-site, multi-method comparative research. Retrieved from the Eurogang Network website: http:// www.umsl.edu/~ccj/eurogang/ Eurogang_20Manual.pdf

Part 1

Definitional Issues in the Comparative Context

Chapter 2

Putting the “Gang” in “Eurogang”: Characteristics of Delinquent Youth Groups by Different Definitional Approaches Kristy N. Matsuda, Finn-Aage Esbensen, and Dena C. Carson*

2.1

Introduction

While there seems to be consensus that gangs are distinct from other groups, there is less agreement on the characteristics necessary or sufficient to define a gang. Despite decades of attention, policymakers, researchers, and law enforcement have not agreed on a universal definition of a “street gang” (Ball and Curry 1995; Esbensen et al. 2001; Klein 1969; Klein and Maxson 2006; Miller 1975). A recent review of state policies by Barrows and Huff (2009) revealed that only two states used the same definition of a “gang member.” Gang researchers have long lamented that the lack of a consensus definition leads to overestimations, underestimations, and depictions of gangs that may not be comparable (Klein and Maxson 2006). The implication of the definitional issue for law enforcement and policymakers is not trivial. It is not possible to identify and respond to gangs if one cannot identify gangs and gang members. Social scientists have been putting forth gang definitions since the 1920s (Thrasher 1963), but the discourse around definitions remains timely (Curry and Decker 1998; Esbensen et al. 2010; Klein 1971). Scholars have assessed the degree to which changing definitional criteria affects the depiction of gang members (Esbensen et al.

* This research was made possible, in part, by the support and participation of seven school districts, including the School District of Philadelphia. This project was supported by Award No. 2006-JV-FX-0011 awarded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. The opinions, findings, and conclusions or recommendations expressed in this manuscript are those of the authors and do not necessarily reflect the views of the Department of Justice or of the seven participating school districts. K.N. Matsuda • F.-A. Esbensen (*) • D.C. Carson Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA e-mail: [email protected] F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_2, © Springer Science+Business Media, LLC 2012

17

18

K.N. Matsuda et al.

2001; Winfree et al. 1992). Research suggests that as the gang definition changes, the qualities of the gang and gang members also shift. One study (Winfree et al. 1992) showed that an inclusive question like “Do you belong to a gang?” captures individuals at all levels of gang membership (i.e., wannabes, active members, and former gang members). A more restrictive definition of gangs, one that requires details or characteristics about the gang, yields more fringe or “wannabe” members. In contrast, a study by Esbensen et al. (2001) applied increasingly restrictive definitions to a sample of youth. They began with a global measure of “have you ever been in a gang” and narrowed down to youth that reported being current, core members in more structured gangs. They found that as focus narrowed to more central gang members, the demographic characteristics of the group did not change, but the level of delinquency and risk factors increased significantly. It is often not possible to collect in-depth detail about an individual’s gang history or extensive characteristics about their gangs. Acquiring a response to a simple question like “Are you a gang member?” may be an achievement. Alternatively, individuals may be willing to give extensive information on their peer group, but be reluctant to admit to being in a “gang.” The result of relying on one source of data as opposed to another has not been adequately examined despite the frequent use of variant methods of defining gang membership. The purpose of this chapter is to investigate the depiction of gang members that results from different definitions of gang membership. This research should help inform whether different methods of identifying gang members result in similar or dissimilar depictions of youth, and importantly, whether gangs can reliably be identified without the use of the term “gang.”

2.1.1

Different Definitional Approaches

American gang researchers often use the self-nomination approach to identify gang members (e.g., “Have you ever been a gang member?” or “Are you currently a gang member?”) (Esbensen et al. 2001; Thornberry et al. 2003). This method is both parsimonious and straightforward. It does, however, hinge on each respondent’s unique perception of what constitutes a “gang member.” The notion of a “gang member” may evoke stereotypical images of necessary characteristics that may or may not be grounded in reality. Respondents may be reluctant to admit to being or considering themselves like those depictions. Alternatively, some researchers have asked respondents if their friends are gang members (Melde et al. 2009; Melde and Esbensen 2011). This approach removes the individual’s own personal investment (either stigma or posturing) and allows for a measure of a group dynamic. A gang is, after all, a group, and the dynamics associated with such a delinquent entity is generally of central interest. This method, however, still relies on a respondent’s own gang definition and assumes that youth who associate with gangs are gang-involved. While prior research has shown that peer groups typically consist of a mix of prosocial and antisocial youth, the studies do not specifically examine gang youth (Elliott and Menard 1996; Haynie 2002;

2

Putting the “Gang” in “Eurogang”…

19

Warr 1993). The current study will provide a better understanding of whether youth who consider their group of friends to be a gang also report the same attitudes and behaviors as other gang youth. Issues related to gang definitions further complicate the likelihood of crossnational gang research. The word “gang” appears to be a concept with a shared understanding across the United States and in other English-speaking countries (though geographical, national, and cultural differences may influence results). The term “gang,” however, may not translate well (or at all) in other languages, even if groups that share similar characteristics to American gangs are present in the culture. The Eurogang definition was created in response to this issue (for a history see Sect. 1 in this volume and Klein et al. 2001). The Eurogang definition defines and measures gang membership without using the term “gang.” Instead, the definition includes qualities believed to reflect the central gang characteristics (i.e., factors that are necessary definers of a gang).1 Earlier work utilizing this approach found predictive and discriminant validity of the Eurogang approach (Esbensen et al. 2008a; Weerman and Esbensen 2005) in both European and American samples. These three approaches to defining gang membership (i.e., self-nomination, friends are gang, and Eurogang) have produced similar depictions of youth involved in gangs, but, to date, the effect of employing these different definitional criteria has not been adequately explored. Regardless of gang definition used, some behavioral characteristics have been consistently shown to distinguish gang members from non-gang youth. For example, gang members are more delinquent than non-gang youth (Esbensen and Winfree 1998; Miller 2001; Thornberry 1998). Gang members are also more delinquent than non-gang youth with delinquent friends (e.g., Battin et al. 1998). Gang-involved youth are also more likely to experience increased victimization (Taylor et al. 2007). In addition, these three definitional approaches have identified gang members who score significantly higher than non-gang youth on a variety of risk factors that have been theoretically and empirically associated with gang membership and delinquent offending, including commitment to negative peers (Burgess and Akers 1966; Sutherland 1947), techniques of neutralization (Sykes and Matza 1957), and parental monitoring (Klein and Maxson 2006). To date, research has been restricted to studies utilizing one of these definitional approaches. In this chapter, we compare gang youth defined by the three approaches discussed above (1) the self-nomination approach (i.e., “Are you currently a gang member?”), (2) the friends are a gang method (i.e., “Do you consider your group of friends to be a gang?”), and (3) the Eurogang approach (i.e., the application of a set of criteria to determine whether the group is a gang). We explore similarities and differences associated with employing all three definitional standards to the same set of respondents. We will thereby be able to address the following questions: 1. Are gang members identified by different gang definitions similarly or dissimilarly situated along behavioral and attitudinal dimensions believed to be associated with gang membership? 2. Does each definition produce the same gang sample; that is, regardless of definition, do we identify the same youths as gang members?

20

K.N. Matsuda et al.

To address the questions posed above, we will compare the demographic characteristics, risk factors associated with gang membership, and the behavioral responses of individuals in each of the three gang groups. We will also examine the extent to which these definitional approaches identify the same individuals and the extent to which there is overlap between and consistency across the three definitional approaches.

2.2

Methods

Data for this study originate from the national evaluation of the Gang Resistance Education and Training (G.R.E.A.T.) program (Esbensen et al. 2011). The G.R.E.A.T. program is a national school-based gang prevention program taught by local law enforcement to (primarily) middle school classrooms. The evaluation is a longitudinal, panel design study that followed a cohort of students in seven diverse cities across the United States for 5 years.2 The seven cities are Albuquerque, New Mexico; Chicago, Illinois; a Dallas–Fort Worth area, Texas; Greeley, Colorado; Nashville, Tennessee; Philadelphia, Pennsylvania; and Portland, Oregon.

2.2.1

Sample Selection

School districts and police departments (that teach the G.R.E.A.T. program) in each of the seven cities agreed to be included in the evaluation. The process produced a final sample of 31 schools and 195 classrooms, and 4,905 students during the 2006– 2007 school year. Sixth grade students were included from 26 schools; seventh grade students comprised the sample in the remaining five schools. Classrooms in the participating schools were randomly assigned to receive or not receive the G.R.E.A.T. program. Active parental consent was required for student participation (for a more detailed description of the active consent process, consult Esbensen et al. 2008b). Overall, 89.1% of youths (N = 4,372) returned a completed consent form, with 77.9% of parents/guardians (N = 3,820) allowing their child’s participation.3

2.2.2

Methods

Students in this research completed a confidential group-administered pretest questionnaire, a posttest survey after program administration, and annual follow-up surveys. The completion rate for the 1-year follow-up was 87% and for the 2-year follow-up was 83%. In the current study, we rely primarily on data from Wave 4 (or the 2-year follow-up) but also draw on cross-sectional results from Wave 1 and Wave 3 (pretest and 1-year follow-up) when illustrative.

2


2.2.3

21

Sample Characteristics

The first column in Table 2.1 presents the characteristics of the analysis sample (n = 3,162) at Wave 4. The demographic profile of the Wave 4 sample did not differ substantively from the original Wave 1 sample (N = 3,820). The sample was equally male and female. The sample was racially/ethnically diverse, with Hispanic (39.7%), white (26.2%), and black (16.7%) youths accounting for nearly 83% of the sample. The remaining youth in the sample identified as another race/ethnicity or bi/multiracial (“other”) at Wave 4. The average age of the sample was 13.5 years at this wave. It is important to keep in mind that age range is significantly truncated because 26 of the 31 schools delivered the G.R.E.A.T. program in sixth grade and the remaining schools in seventh grade. Over 44% of the sample reported living with both their mother and their father (i.e., “nuclear family structure”). A majority of the sample reported other living arrangements (e.g., single parent, step-parents present, with other relatives).

2.2.4

Measures

This study explores the relationship between demographic, attitudinal, and behavioral measures across different definitions of gang membership. The means for all measures are reported for the full sample in the first column of Table 2.1. A description of each measure is detailed in this section. Demographics. Survey respondents were asked to indicate their sex, ethnicity/race (white, black, Hispanic/Latino, or other), and age. Youth who indicated that they lived with both their mother and father were coded as “nuclear family structure.” All other living structures were classified as “non-nuclear.” Behavior. Respondents’ survey indicated the frequency of their involvement in 14 different types of delinquency in the previous 6 months.4 The frequency of their involvement in these delinquent offenses was used to create a measure of “delinquency.” Respondents’ volumes (i.e., zero times, one to two times, about once a month, about once a week, or everyday in the previous 6 months) of their use of tobacco, alcohol, illegal drugs (marijuana or other illegal substances), and inhalants (i.e., paint, glue, or other things inhaled to get high) were used to create a measure of “substance use.” Youth also indicated the frequency of their “victimization” in the previous 6 months.5 Youth were asked whether they spent time with their friends where no adults were present (yes or no) as a measure of time “unsupervised” with peers. Finally, they were asked if they spent time with their friends where drugs and alcohol are available (yes or no) as a measure of “drugs and alcohol present.” Attitudes. The survey included attitudinal scales strongly linked to both gang membership and general delinquency. Each scale was created from items measured on a five-point scale. Lower scores on each variable indicate less of the given measure (e.g., a lower score on the “anger” scale indicates less anger). “Parental monitoring”

Table 2.1 Characteristics of full sample, gang sample, and by definition at Wave 4 Full sample Not in gang Any gang N (% of sample) 3,162 2,775 (87.8) 379 (12.2) M(SD)/% M(SD)/% M(SD)/% Male 49.3 48.7 50.9 Race/ethnicity White 26.2 27.5 16.9 Black 16.7 16.7 16.4 Hispanic 39.7 38.2 50.3 Multiracial/other 17.4 17.5 16.4 Age (Wave 4) 13.51 (.72) 13.48 (.70) 13.73 (.80) Nuclear family 44.2 54.4 46.4 Delinquency 7.95 (17.76) 4.80 (10.41) 30.68 (34.95) Substance use 0.91 (2.18) 0.53 (1.45) 3.64 (3.93) Victimization 9.05 (14.76) 7.85 (12.52) 17.77 (24.03) Parental monitoring 4.02 (.80) 4.10 (.75) 3.46 (.92) Negative peer commitment 2.04 (1.04) 1.91 (.94) 2.99 (1.22) Hitting neutralizations 3.45 (1.19) 3.34 (1.18) 4.23 (.93) School commitment 3.64 (.73) 3.71 (.69) 3.14 (.77) Guilt 2.40 (.62) 2.48 (.58) 1.83 (.55) Anger 3.01 (1.02) 2.94 (1.00) 3.57 (.95) Unsupervised 51.5 58.7 89.2 Drugs and alcohol present 12.5 9.2 59.9 Eurogang 215 (6.8) M(SD)/% 46.0 20.7 10.8 55.4 13.1 13.78 (.78) 49.3 33.10 (33.22) 4.60 (3.94) 19.28 (22.78) 3.43 (.92) 3.26 (1.21) 4.34 (.81) 3.02 (.75) 1.72 (.50) 3.63 (.98) 94.0 76.3

Self-nominate 151 (4.8) M(SD)/% 60.0 11.7 20.0 51.0 17.2 13.76 (.85) 37.1 41.02 (42.26) 3.95 (4.34) 21.23 (28.92) 3.27 (.96) 3.04 (1.27) 4.28 (.98) 3.11 (.84) 1.76 (.58) 3.68 (.95) 86.4 54.8

9.3 21.5 52.3 16.9 13.72 (.83) 41.1 32.49 (37.81) 3.33 (4.12) 17.20 (25.46) 3.45 (.95) 2.97 (1.22) 4.27 (.97) 3.12 (.80) 1.83 (.59) 3.59 (.98) 82.2 52.7

Friends gang 175 (5.5) M(SD)/% 55.8

22 K.N. Matsuda et al.

2


23

was a 4-item scale measuring the extent of parental awareness of youth’s activities and location (Wave 4 alpha = .81). “Negative peer commitment” was created from three items measuring the likelihood the respondent would continue to hang out with their friends even if they were getting the respondent into trouble (alpha = .86). “Hitting neutralizations” was a 3-item scale that captured a respondent’s belief that physical violence is acceptable under certain circumstances (alpha = .88). Seven items captured the respondent’s “school commitment” (i.e., attitudes specific to aspects of the educational experience and general feelings about education) (alpha = .81). The level of respondent’s “guilt” for engaging in seven different delinquent offenses was measured on a 3-point response scale (alpha = .91). The final attitudinal scale captured the respondent’s level and control over “anger” (4 items, alpha = .83). Gang membership. Gang membership was measured in three ways. First, respondents could “self-nominate” as a gang member by answering “yes” to the question “Are you now in a gang?” Second, the Eurogang approach captured youth that had a group of friends that fit the following criteria (1) had 3 or more people, (2) with people mostly between the ages of 12 and 25, (3) that spends a lot of time in public places, (4) been in existence for more than 3 months, and (5) accepted and actually participated in illegal activity. Third, respondents were asked “Do you consider your group of friends to be a gang?” Respondents that answered affirmatively were classified as having “friends in gang.”

2.3 2.3.1

Results Non-gang Members Versus Gang Members

Youth that fit at least one of the three gang definitions were compared with youth that were not classified as gang involved using any of the three definitions. At Wave 4, 12.2% of the entire sample was classified as gang involved according to at least one of the three gang definitions. There was a general increase in the proportion of gang members over time (Wave 1 = 9.5%, Wave 3 = 11.2%). Demographic, behavioral, and attitudinal differences between the two groups of youth (i.e., any gang and not in gang) were compared using bivariate tests of association. For ease of presentation, the results from Wave 4 are presented in Table 2.1, but similarities and differences across waves are discussed in the text. In general, the results across waves were similar. There were no significant differences in sex or family structure between gang and non-gang youth at Wave 4. Sex differences were found at Waves 1 and 3 with males more likely than females to be gang involved in the earlier waves. At Wave 4, however, a near equal proportion of males were gang involved as females. Bivariate analysis also failed to show any significant difference in the proportion of youth from two-parent households (i.e., “nuclear”) that were gang involved at Wave 4. The same pattern holds at Wave 3. At Wave 1, however, youth classified as gang involved were significantly less likely to report living with both their mother and father.

24

K.N. Matsuda et al.

There were significant differences in the age and race/ethnicity of gang members and non-gang members at each wave. As stated above, data used in this analysis are from a longitudinal, panel design study in which youth were recruited during the early years of middle school. Therefore, the age range of the sample at each wave is restricted. Regardless, at each of the three waves, gang members were significantly older than non-gang members (at Wave 4, gang = 13.73, non-gang = 13.48). With respect to the racial/ethnic composition of the gang and non-gang groups, there was considerable variation across waves. At Wave 1, there were no differences between the gang and non-gang youth for Hispanics (36.3% non-gang, 38.3% gang) or other youth (19.1% non-gang, 18.7% gang). More black youth were gang involved (16.8% non-gang, 28.4% gang group), while fewer white youth were gang members (27.8% non-gang, 14.4% of gang group). By Wave 3, the difference between the black gang and non-gang youth had diminished (16% non-gang, 21.1% gang), and by Wave 4, there was no difference among the black youth with regard to gang membership. For Hispanic youth, the proportion of gang members increased so that by Wave 4 approximately 50% of the gang youth were classified as Hispanic compared with 38% of the non-gang youth. Across all waves of data, white youth were consistently less likely to be gang members, and youth of other race/ethnicities were equally represented in the gang and non-gang groups. Gang and non-gang groups were compared across three dimensions of behavior. Consistent with prior research, gang members reported significantly more delinquency, substance use, and victimizations at every wave than did the non-gang youth. Gang members at Wave 4 reported an average of 30.68 delinquent offenses as compared to 4.80 for non-gang members (see Table 2.1). The rate of substance use was seven times higher for gang members (7.49) than non-gang members (1.03). Gang members were also victimized more frequently (17.77) than non-gang members (7.86) in the previous 6 months. We also compared gang and non-gang youth across a variety of attitudinal and other behavioral dimensions. Gang membership was significantly related to a number of risk factors. Gang members reported lower levels of parental monitoring, school commitment, and guilt than their non-gang counterparts and higher levels of commitment to negative peers, neutralizations toward hitting, self-reported anger, and time spent unsupervised with peers and where alcohol and drugs were present (see Table 2.1 for specifics). For example, 9% of non-gang youth reported spending time where alcohol and drugs were present as compared to 60% of gang-involved youth. Taken as a whole, these cross-sectional, bivariate results suggest that gang membership, as defined by any or all of the three definitions, is correlated with attitudinal and behavioral differences.

2.3.2

Similarities and Differences by Gang Member Definition

We now turn to our main concern in this chapter, examination of whether the three different definitions of gang membership identify the same group of respondents and

2


25

Fig. 2.1 Overlap between gang member definition at Wave 4

the extent to which they exhibit similar risk factors and behaviors associated with gang membership. In the following sections, we compare the characteristics of youth, their attitudes, and their behaviors by definition to explore the validity of the three gang measures. As stated previously, 12.3% (n = 384) of the Wave 4 sample was gang involved using at least one of the three definitions. Some of these youth were captured under two or all of the definitions. The final three columns of Table 2.1 present the prevalence of gang membership by definition at Wave 4. Examining each unique definition reveals that the Eurogang definition produced the most gang members (6.8% of sample, n = 215). Almost 5% of youth self-reported being a gang member (4.8%, n = 151), and 5.5% (n = 175) of youth reported their friends were in a gang. While the percent of gang members produced by each definition was consistently under 10%, the distribution of gang members varied by definition and wave. The proportion of youth that self-report gang membership was consistently around 5% at each wave (Wave 1 = 4.5%, Wave 3 = 5.2%, Wave 4 = 4.8%), although the same youth did not necessarily report gang membership from one wave to the next. The Eurogang definition produced more gang members with each subsequent wave (Wave 1 = 2.3%, Wave 3 = 5.1%, Wave 4 = 6.8%). The percent of youth that reported their friends were in a gang decreased at each wave (Wave 1 = 7.2%, Wave 3 = 6.7%, Wave 4 = 5.5%). From a policy perspective, these varying prevalence rates suggest that definition is important. The magnitude of the “gang problem” varies substantially by definition. From a methodological standpoint, it is important to assess the degree to which these various operationalizations of gang members converge. Figure 2.1 illustrates the degree of concordance between the samples captured by the different definitions of gang membership. In total, 33% of youth identified as gang by any definition are defined as gang members by more than one definition. Only 9% of youth in this sample fit all three gang definitions. At Wave 4, the largest proportion of youth was defined solely by the Eurogang definition. The same general pattern of definitional overlap exists for the previous two waves, but at Wave 1, the largest proportion of youth was defined solely because their friends were gang members (40%). At Wave

26

K.N. Matsuda et al.

3, the proportion of youth with no overlap in gang membership was more evenly distributed (only self-nominate = 16.0%, only Eurogang = 23.0%, and only friends are gang = 24.0%). Despite the fact that individuals defined as gang by any definition were significantly different than non-gang members, Figure 2.1 shows that each definition captures different youth in the sample. The remainder of this chapter explores whether the different subsets of youth defined by the three definitions are substantively different from each other. This will inform the validity of using any and all of the three approaches. As stated previously, there are a number of reasons that certain definitions may be advantageous under specific conditions. For example, cultural differences may make the use of the word “gang” problematic. Alternatively, individuals may believe that a significant segment of their friends are gang members, but may be reluctant to consider themselves “like them” despite shared activities or attitudes. Regardless of the differences in reasons to use one method of gang identification over the other, there is no a priori reason to believe that the gang members that are revealed should differ by definition. The following paragraphs examine demographics, attitudes, and behaviors by gang definition. Because youth could be included in multiple definitional categories, tests of differences were not conducted. This section includes a descriptive overview of the proportion of youth in each category (see Table 2.1). Approximately half the sample (50.9%) classified as gang involved using any of the three definitions was male. Analysis by definition shows that the self-nomination approach (i.e., “I am a gang member”) yielded more males than either of the other two approaches, and the Eurogang definition produced more females. Across racial/ethnic dimensions, the self-nomination approach and the friends are gang approach produced a similar racial/ethnic distribution. The Eurogang definition produced twice as many white gang members as the other two definitions. About 12% of youth that self-nominated as a gang member were white compared with 9.3% of those that said their friends were gang members, and 20.7% of those identified as gang members using the Eurogang definition. In contrast, the proportion of Eurogang youth that were black was about half that of the other two gang groups. Across definitions, the average age of gang members across the three groups was similar. Youth defined as gang involved by the Eurogang definition were more likely to be from two-parent households than the other two groups. Almost half of the Eurogang youth were from nuclear families (49.3%) compared to those whose friends were in a gang (41.1%) and those who self-nominated as a gang member (37.1%). But these observed differences may be heavily influenced by age of the participants and/or data collection wave. For example, the sex of gang members so classified using the Eurogang definition varies across waves. At Wave 1, 67.1% of Eurogang gang members were male, but at Wave 4, there were more female gang members using this definition. In contrast, the proportion of male gang members produced by the other two definitions remained relatively stable over time. For race/ ethnicity, the Eurogang definition always produced a smaller proportion of black gang members, but across all definitions, the proportion of black gang members decreased by wave. Conversely, the proportion of white gang members increased significantly over time using the Eurogang definition.

2


2.3.3

27

Behavior and Attitudes by Definition

Taken as a whole, youth that were classified as gang-involved by any of the three definitional approaches were significantly more delinquent (general delinquency and substance use) and more likely to be victimized than youth that were not gang involved. Recall that the average frequency of events in the past 6 months for nongang-involved youth was 4.80 delinquent offenses, 1.03 illegal substance use, and 7.85 victimizations. Looking across each definition (see Table 2.1), gang members, however defined, are more delinquent and more likely to be victimized than nongang members. Youth who self-admit gang membership at Wave 4 reported a mean of 41 delinquent offenses in a 6-month period, and Eurogang and youth whose friends were in gangs had over 30 delinquent offenses during the same period (33.1 and 32.5, respectively). Self-nominated gang members also reported the most victimization, while Eurogang youth reported the highest mean of substance use. By restricting this analysis to Wave 4 data, it appears that the self-nomination method produces the most highly delinquent type of “gang member.” Examining earlier waves, however, uncovers more complicated dynamics. At the earliest wave, the Eurogang definition produced the most delinquent group of gang-involved youth. Eurogang youth report an average of 34.4 delinquent acts as compared with the self-nomination youth who report 25.1 delinquent acts. The same pattern holds for substance use and victimization. At Wave 3, the Eurogang and self-nominated gang members report nearly identical rates of delinquency, substance use, and victimization. It is important to keep in mind that these similarities in offending were found in spite of the modest overlap of youth that fall within these two categories. Youth whose friends were in gangs reported less troublesome behaviors than the other two gang groups. The differences were modest, and those with gang friends still engaged in significantly more problem behaviors than the non-gang group. Respondents were asked to indicate whether they spent peer time unsupervised and, if so, whether alcohol and drugs were present. A large proportion of gang youth spent peer time unsupervised (self-nominate = 86.4%, Eurogang = 94.0%, friends are gang = 82.2%). The proportion of youth that indicated they spent unsupervised time with alcohol and drugs present increased dramatically across all groups from Wave 1 to Wave 4. By Wave 4, over three quarters of Eurogang youth (76.3%) indicated they hung out unsupervised around alcohol and drugs. This is in contrast to 54.8% of self-admitted gang members and 52.7% of youth whose friends were in gangs. A comparison of attitudinal risk factors associated with delinquency and/or gang membership is also presented in Table 2.1. In general, the mean score for each gang group is roughly equivalent. With the exception of commitment to negative peers, the mean of every attitudinal measure differed minimally across each gang definition. Eurogang youth had stronger commitment to negative peers than the other two gang groups (gang = 3.0, Eurogang = 3.3, friends are gang = 3.0). This pattern was consistent across waves.

28

2.3.4

K.N. Matsuda et al.

Multivariate Logistic Regression Models

We used a series of multivariate logistic regressions to test which demographic, attitudinal, and behavioral measures significantly distinguished gang members from non-gang members according to each definition. Because of the dependence of the subsamples, patterns of significance across the three models, as opposed to direct comparisons, will be discussed. Odds ratios will be discussed when noteworthy for each model separately but are not appropriate or intended to be compared across models (Menard 2010). Table 2.2 presents the results of multivariate logistic regressions for each of the gang definitions (i.e., gang, Eurogang, friends are gang) and for youth that were defined as gang by any of the three definitions (i.e., any gang). There were a number of consistent findings across each type of gang membership. For each definition, negative peer commitment, neutralizations for hitting, guilt, and unsupervised peer time spent where drugs and alcohol are available significantly distinguished gang from non-gang members. The effects are in the expected direction. Youth who had a stronger commitment to negative peers, used more neutralizations about violence, felt less guilty, and were more likely to spend time unsupervised where drugs and alcohol are present were more likely to be classified as a gang member, by any definition. For example, youth who spent peer time unsupervised where drugs and alcohol are present increased their odds of being classified as Eurogang by 763%. It is also noteworthy that these general trends were found at Wave 1 and Wave 3 for every definition of gang membership (not shown in table). There were also definitional specific trends. Higher level of parental monitoring decreased the likelihood of self-nomination as a gang member, but not Eurogang membership or gang friends (see Table 2.2). This trend was consistent for the selfnomination group across multiple waves of data. Increased level of anger was also correlated with self-nominated gang membership. Every unit of increase in the anger scale increased the odds of self-nominated gang membership by 30%. Consistent with bivariate results, being female increased the likelihood of being a Eurogang member. The odds of Eurogang membership were 91% higher for females than males at Wave 4. Older youth were also more likely to be classified as Eurogang. Youth who were less committed to school and who spent time unsupervised were also significantly more likely to be classified as Eurogang. These results were not replicated in any of the other groups. This general pattern of results was not consistent for the Eurogang samples at Wave 1 and Wave 3 as previously suggested by the bivariate analysis. Black and Hispanic youth (as compared to white youth) were more likely to report their friends were in a gang. The odds of youth reporting their friends were in a gang increased 149% for black youth and 99% for Hispanic youth as compared to white youth (see Table 2.2). The greater likelihood of black youth reporting their friends as gang was consistent across all three waves of data. Significant ethnic differences between Hispanic and white youth were present only at Waves 3 and 4. Finally, while higher levels of anger distinguished both the self-nomination and friends as gangs groups, it did not significantly distinguish the Eurogang group.

*

p < .05, **p < .01

C²(14) = 677.58** −2LL = 1,414.14 Nagelkerke R² = .41 n = 2,851

C²(14) = 314.13** −2LL = 691.831 Nagelkerke R² = .35 n = 2,634

Table 2.2 Logistic regression results for correlates of gang membership by definition at Wave 4 Any gang Self-nominate B SE Odds B SE Odds Sex .23 .15 1.26 −.29 .23 .75 Black .20 .24 1.22 .54 .38 1.72 Hispanic .09 .19 1.09 .52 .32 1.69 Other .13 .23 1.14 .51 .38 1.67 Age .14 .10 1.15 .11 .14 1.11 Nuclear family −.00 .15 1.00 −.41 .22 .66 Parent monitoring −.20 .10 .82* −.44 .14 .65** ** Neg. peer commit. .30 .07 1.35 .36 .10 1.43** Hit neutralization .35 .08 1.43** .39 .13 1.48** School commit. −.17 .11 .85 −.07 .166 .93 Guilt −.75 .13 .47** −.86 .18 .42** Anger .20 .08 1.22* .26 .12 1.30* Unsupervised .73 .20 2.07** .532 .29 1.70 Drugs and alcohol 1.49 .16 4.45** .98 .24 2.67** Constant −4.87 1.47 −4.79 2.18 Eurogang SE .21 .34 .25 .32 .13 .20 .13 .09 .12 .15 .18 .11 .33 .21 1.99 C²(14) = 638.78** −2LL = 793.69 Nagelkerke R² = .51 n = 2,710

B .64 −.35 −.21 −.51 .26 .16 .04 .37 .45 −.40 −.99 .12 1.02 2.16 −7.78

Odds 1.91** .71 .82 .60 1.30* 1.18 1.04 1.44** 1.57** .67** .37** 1.13 2.77** 8.63**

Friends as gang SE Odds .20 .90 .35 2.49** .30 1.99* .35 1.94 .13 1.08 .20 .80 .12 .81 .09 1.46** .11 1.33** .14 .77 .16 .52** .10 1.23* .23 1.17 .21 3.34** 1.92 C²(14) = 308.43** −2LL = 868.36 Nagelkerke R² = .31 n = 2,663

B −.10 .91 .69 .66 .07 −.22 −.21 .38 .28 −.26 −.66 .21 .16 1.21 −4.18

2 Putting the “Gang” in “Eurogang”… 29

30

2.4

K.N. Matsuda et al.

Discussion

The purpose of this chapter was to explore the validity of different definitional approaches to identify gang members. We applied three definitions to the same diverse sample of youth in the United States. The use of any or all of the three definitions reveals a subsample of youth that are behaviorally and attitudinally distinct from non-gang youth. Youth defined as gang by any method have more risk factors previously shown to be correlated with gang membership and/or delinquency. They also report significantly higher levels of problem behaviors. Gang members report mean levels of delinquency, for example, five to six times the average of non-gang members. The extent of definitional overlap, however, is far from perfect. In fact, less than 10% of the sample of gang youth at Wave 4 was classified as gang members by all three definitions. In other words, each definition captured a unique group of individuals. The majority of youth were only considered gang members by one definition. This result was stable across multiple waves of data. Further investigation showed that while different youth were captured by each definition, the attitudes and behaviors of all groups were relatively similar. Most of the strongest correlates associated with gang membership were common to all three definitions. Importantly, each definition of gang membership produced a subsample that was distinct from non-gang youth. In short, each definition identifies youth who express attitudes and report behaviors that are consistent with our understanding of and expectations for gang members. Each definition, however, categorizes a different group of youth. The Eurogang definition at Wave 4, for example, captured a larger proportion of youth that are often underrepresented by other methods (at the same wave) and not stereotypically perceived to be gang members. More females, white youth, and individuals from nuclear family structures were classified as Eurogang than the other two definitions. We conclude that there is utility and validity in each gang definition. Ideally, all three could be applied to the same sample to provide the fullest and most nuanced interpretation of gang youth. Each definition adds legitimacy to a claim that someone is a gang member. Self-admission is important and parsimonious, but some may be reluctant to admit such a stigmatizing status. Identification with gang peers and a confirmation of involvement in a group that exhibits gang-like features is a useful way to triangulate reality and increase confidence in a determination of “gang involvement.” This research shows, however, that far fewer gang members will be defined by using this most restrictive definition. Future research should further explore the qualities and conditions that lead youth to fall in different definitional categories. For example, do members from gangs with specific structures self-report as gang members, while members from other structures fall under Eurogang? From a policy and law enforcement perspective, it is important to ask what impact the willingness to self-nominate as a gang member has on the effectiveness of intervention. The Eurogang definition in the

2


31

United States does not require youth to adopt the gang identity and/or stigma. Are Eurogang-defined youth better targets for intervention? They exhibit similar problem behaviors and attitudes as self-nominated gang members, but there may be important differences in group identity and/or cohesion that affect their likelihood to self-nominate. Also, the Eurogang definition was created for use in contexts in which the word “gang” is unavailable or inappropriate. Therefore, the generalizability of these findings to a broader international context should not be made without further replication. Differences in prevalence or the picture of gangs in the countries that use only the Eurogang definition may look distinct from the American research that only uses the self-nomination. Additional research in English-speaking European countries would be an important contribution to the discussion of measurement error and validity (Sharp et al. 2006). These findings implicate an important policy issue. States have increasingly designated sentencing enhancements for crimes committed by individuals identified as “gang members.” The method by which states have defined their gang members has varied (Barrows and Huff 2009). Self-nomination is often used to legally label someone a gang member, but affiliation with gang members (i.e., friends are gang) and involvement in a group that fit a specified criterion (i.e., Eurogang like definition) are also commonly used (Barrows and Huff 2009). In light of the reality of gang policies, these findings can lead to two opposing, yet logical, recommendations. One could read these findings as support for the notion that any of the three approaches are appropriate and valid methods to identify gang members (and apply gang enhancements). This is consistent with current policy and implicates a broad definition of gang membership. Alternatively, the demonstrated consensus across the three definitions may lead some to advocate for a more restrictive criterion for membership to diminish the possibility of widening the net of gang-involved youth and the serious consequences that follow. Given these findings, we offer a policy recommendation. For prevention strategies, intended for a wide audience with minimal consequences for a type II error, we recommend that a broad definition of gang membership be used (i.e., any of the three definitions). For intervention and suppression strategies, specifically targeted to gang members, we recommend a focus on youth that fit multiple definitions because of the serious consequences associated with making a type II error.

Notes 1. The Eurogang approach defines a street gang as “any durable, street-oriented youth group whose involvement in illegal activity is part of its group identity” (Weerman et al. 2009). 2. Sites were selected based on three criteria (1) existence of an established G.R.E.A.T. program, (2) geographic and demographic diversity, and (3) evidence of gang activity. Other considerations included length of time the program had been in operation, number of G.R.E.A.T. trained officers, number of schools in which the program was offered, and the components of the program implemented.

32

K.N. Matsuda et al.

3. It should be noted that while Esbensen et al. (2008) reported a 79% consent rate, the addition of two schools to the evaluation after the publication of that article resulted in the 78% overall consent rate reported here. 4. Respondents were asked about their involvement in the following delinquent activities: skipped classes without an excuse, lied about age to get into some place or buy something, avoided paying for things, purposely damaged or destroyed others’ property, carried a hidden weapon for protection, illegally spray painted a structure, attempted or stole something worth less than $50, attempted or stole something worth more than $50, attempted or gone into a building to steal something, hit someone, attacked someone with a weapon, used a weapon or force to get something from someone, been involved in gang fights, and sold marijuana or other illegal drugs. 5. Youth were asked if and how often they were attacked or threatened on their way to or from school, had their things stolen at school, been attacked or threatened at school, had mean rumors or lies spread about them at school, had sexual comments or gestures made at school, been made fun of at school, been bullied at school, been hit, had a weapon or force used to get their money or property, been seriously attacked or with a weapon, and had things stolen.

References Ball RA, Curry GD (1995) The logic of definition in criminology: purposes and methods for defining gangs. Criminology 33:225–45 Barrows J, Huff CR (2009) Gang and public policy: constructing and deconstructing gang databases. Criminol Public Policy 8:675–703 Battin SR, Hill KG, Abbott RD, Catalano RF, Hawkins JD (1998) The contribution of gang membership to delinquency beyond delinquent friends. Criminology 36:93–115 Burgess RL, Akers RL (1966) A differential association reinforcement theory of criminal behavior. Soc Probl 14:128–147 Curry GD, Decker SH (1998) Confronting gangs: crime and community. Roxbury, Los Angeles Elliott DS, Menard S (1996) Delinquent friends and delinquent behavior: temporal and developmental patterns. In: David Hawkins J (ed) Delinquency and crime: current theories. Cambridge University Press, New York Esbensen FA, Winfree LT Jr (1998) Race and gender differences between gang and non-gang youth: results from a multi-site survey. Justice Quart 15:505–526 Esbensen FA, Winfree LT, He N, Taylor TJ (2001) Youth gangs and definitional issues: when is a gang a gang, and why does it matter? Crime Delinq 47:105–130 Esbensen FA, Brick B, Melde C, Tusinski K, Taylor TJ (2008a) The role of race and ethnicity in gang membership. In: van Gemert F, Peterson D, Lien IL (eds) Street gangs, migration, and ethnicity. Willan, Uffculme, Devon, UK Esbensen FA, Melde C, Taylor TJ, Peterson D (2008b) Active consent in school-based research: how much is enough and how do we get it? Eval Rev 32:335–362 Esbensen FA, Peterson D, Taylor TJ, Freng A (2010) Youth violence: sex and race differences in offending, victimization, and gang membership. Temple University Press, Philadelphia Esbensen F-A, Peterson D, Taylor TJ, Freng A, Osgood DW, Carson D, Matsuda KN (2011) Evaluation and evolution of the Gang Resistance Education and Training (G.R.E.A.T.) Program. J Sch Violence 10:53–70 Haynie DL (2002) Friendship networks and delinquency: the relative nature of peer delinquency. J Quant Criminol 18:99–134 Klein MW (1969) Violence in American juvenile gangs. In: Mulvihill DJ, Tumins MM, Curtis LA (eds) Crime of violence, volume 13, a staff report to the National Commission on the Causes and Prevention of violence. GPO, Washington, DC Klein MW (1971) Street gangs and street workers. Prentice-Hall, Englewood Cliffs

2


33

Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, New York Klein MW, Kerner H-J, Maxson CL, Weitekamp EM (eds) (2001) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Dordrecht, the Netherlands Melde C, Esbensen FA (2011) Gang membership as a turning point in the life course. Criminology 49:513–552 Melde C, Taylor TJ, Esbensen FA (2009) I got your back’: an examination of the protective function of gang membership in adolescence. Criminology 47:565–594 Menard S (2010) Logistic regression: from introductory to advanced concepts and applications. Sage, Thousand Oaks Miller WB (1975) Violence by youth gangs and youth groups as crime problem in major American cities. U.S. Government Printing Office, Washington, DC Miller J (2001) One of the guys: girls, gangs, and gender. Oxford University Press, New York Sharp C, Aldridge J, Medina J (2006) Delinquent youth groups and offending behavior: findings from the 2004 offending, crime and justice survey. Research Development and Statistics Directorate, Home Office, London Sutherland EH (1947) Principles of criminology, 4th edn. J.B. Lippincott, Philadelphia Sykes G, Matza D (1957) Techniques of neutralization: a theory of delinquency. Am J Sociol 22:664–670 Taylor TJ, Peterson D, Esbensen FA, Freng A (2007) Gang membership as a risk factor for adolescent violent victimization. J Res Crime Delinq 44:351–80 Thornberry TP (1998) Membership in youth gangs and involvement in serious and violent offending. In: Loeber R, Farrington DP (eds) Serious and violent juvenile offenders: risk factors and successful interventions. Sage, Thousand Oaks, pp 147–66 Thornberry TP, Krohn MD, Lizotte AJ, Smith CA, Tobin K (2003) Gangs and delinquency in developmental perspective. Cambridge University Press, New York Thrasher FM (1963) The gang: a study of one thousand three hundred thirteen gangs in Chicago. University of Chicago Press, Chicago Warr M (1993) Age, peers, and delinquency. Criminology 31:17–40 Weerman FM, Esbensen FA (2005) A cross-national comparison of gangs: The Netherlands and the United States. In: Decker SH, Weerman FM (eds) European street gangs and troublesome youth groups. Altamira Press, Walnut Creek Weerman FM, Maxson CL, Esbensen FA, Aldridge J, Medina J, van Gemert F (2009) Eurogang program manual: Background, development, and use of the Eurogang instruments in multi-site, multi-method comparative research. Available at http://www.umsl.edu/~ccj/eurogang/euroganghome.htm Winfree LT, Fuller K, Vigil T, Mays GL (1992) The definition and measurement of gang status: policy implications for juvenile justice. Juv Family Court J 43:29–37

Chapter 3

Counting Gangs: Conceptual and Validity Problems with the Eurogang Definition Judith Aldridge, Juanjo Medina-Ariz, and Robert Ralphs*

3.1

Introduction

There is considerable academic debate and disagreement regarding the definition(s) of gangs. Such definitional disputes have exercised researchers from the seminal work of Thrasher (1927), and they explain, probably to a considerable degree, contrasting research findings. The Eurogang definition has been praised, if not necessarily as the most conceptually suitable definition of a gang, for the attempt it represents to provide researchers with a consistent starting point in identifying, measuring and characterising youth gangs (Wood and Alleyne 2010). And given the Eurogang commitment to developing a common framework for comparative international research, the construction of a consensus definition in the context of such diverse available definitions and approaches is critical. The development of such a definition is no mean feat, however, and does not (simply) involve the development and operationalisation of theoretically informed concepts appropriate to the definition of a gang. It was also imperative that the product meet the cultural and linguistic nuances of a group representing more resulting therefore in the kind of compromise that is inevitable when seeking to create a consensus definition.

* This research was supported in part by a grant from the UK Economic and Social Research Council (RES-000-23-0615). J. Aldridge (*) • J. Medina-Ariz Centre for Criminology and Criminal Justice, School of Law, University of Manchester, Manchester, UK e-mail: [email protected] R. Ralphs Department of Sociology, Manchester Metropolitan University, Manchester, UK F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_3, © Springer Science+Business Media, LLC 2012

35

36

J. Aldridge et al.

According to this international group of researchers, a gang is a “durable street-oriented youth group whose involvement in illegal activity is part of its group identity” (Klein and Maxson 2006; Weerman et al. 2009). This chapter provides us with the opportunity to reflect upon the utility of the definition across a number of research projects in which we and others have employed it. We bring together analyses from the ESRC-funded Youth Gangs in an English City (YOGEC) project (Aldridge and Medina 2008), and two large youth surveys. We raise questions primarily about two aspects of the Eurogang definition. Firstly, we question whether gangs must be “street-oriented” (spend time in public places). In support of our position, we draw on our YOGEC ethnographic research findings that at least some gangs in Research City did not typically spend time gathered in public places, so that according to the EG definition, they would not be considered to be gangs, even though we argue they ought to be labelled as such. Secondly, we wish to bring to the table a series of problems we encountered with what we see as the key aspect of the Eurogang definition: that the group’s “involvement in illegal activity” is part of its “group identity.” Our evidence here addresses four separate problems (1) the definition includes a number of group types that, arguably, should not be considered gangs; (2) the definition requires researchers to draw conclusions based on their observations or questioning about group identity, and the theoretical and conceptual basis required to determine group identity is insufficiently developed; (3) we encountered problems with the current operationalisation of this aspect of the definition in the survey instrument in relation to its validity and the extent to which it is accurately comprehended by respondents; and (4) empirical evidence in the form of cross-tabulated analyses using survey data suggests that not all groups identified by the EG definition are involved in offending activity, and a substantial proportion of those that are, are involved in drug taking alone. Finally, we identify and discuss a feature of all the gangs we encountered in our research—whether street-based or not—common to all: a reputation for a willingness to resort to violence or its threat, even if rarely enacted.

3.2

Data and Methods

We reflect in this chapter on a number of data sources relevant to the definitional concerns outlined above.

3.2.1

ESRC funded ethnographic study, Youth Gangs in an English City (YOGEC) between 2005–2008

We conducted our study in a large city where authorities had recognised the existence of a violent “gang problem” and implemented explicit gang suppression and prevention measures. The methods for this research are explained in greater detail elsewhere (Aldridge et al. 2008, 2010; Medina et al. 2011; Ralphs et al. 2009), including our decision not to reveal the name of the city in order to protect the

3 Counting Gangs: Conceptual and Validity Problems with the Eurogang Definition

37

identity of individuals we engaged with throughout the research, and also to avoid further stigmatisation of certain areas within the city (referred to as “Research City”). Data were primarily collected over 26 months between 2005 and 2008 by way of direct observation and informal conversation, 130 formal interviews1 and 9 focus groups (with parents, with non-gang youth and with community and agency representatives). Our ongoing contacts and continued presence in the field have resulted in additional opportunities for data collection since this time, including interviews with gang parents (Aldridge et al. 2009) and in relation to a gang mentoring project (Medina et al. 2010). We worked in two broad areas of Research City “Inner West,” an ethnically diverse corridor of neighbourhoods where gangs were primarily Black Caribbean/ African/Asian/mixed race, and “Far West,” a housing estate on the outskirts of the city where the population and the gangs there were predominantly white. We established access to six gangs that differed in terms of their longevity, ethnic composition, public profile and (to some degree) nature of criminal activity. Our fieldwork and observations occurred over a longer period in Inner West than in Far West, and our understanding of the gangs in the former, therefore, became somewhat better developed. We set out to use the Eurogang definition in order to construct our sample. The Eurogang definition provided us with a sensible starting point given both its international salience and its conceptual latitude, allowing us to engage with the diversity of groups that we found in Research City. Three additional interviews were conducted in Research City with individuals who were members of groups who met the Eurogang definition for the purposes of reflecting on the definition, and their results are reported for the first time here. The 2004 Offending Crime and Justice survey (Sharp et al. 2006) is a nationally representative household survey of 10–25-year-olds (N = 4,259) in England and Wales and had a response rate of 69.9%. The gang questions were only asked to respondents aged 10–19 (N = 3,827). The 2008 Madrid Youth Survey is a representative community survey of 14–29-year-olds (N = 1,209) covering the autonomous community of Madrid, including Madrid city and surrounding towns and villages. Both surveys relied on face-to-face interviews in the household of the respondents. Both surveys included key questions from the Eurogang Youth Survey instrument.

3.3 3.3.1

Results Were All Gangs “Street-Based”?

As Tita et al. (2005) have argued, “gang set spaces” become associated to a gang because these are places gang members collect to “hang out” (p. 273). We observed gang affiliated young people in Far West spending time like this in public places, often in large numbers and typically outside shopping areas, similar to youth gangs documented in the USA (e.g. Vigil 2002). The young people we interviewed spoke about hanging out in public places, and we regularly observed them doing so. The parents who took part in our Far West focus group reported feeling intimidated by

38

J. Aldridge et al.

these groups of young people, who would sometimes threaten or hassle those they encountered. Interestingly, in spite of their conspicuous street presence, Far West was not officially recognised by police and other local authorities as having a gang problem, and the city’s dedicated gang/gun units had no presence there.2 Over the two years we spent conducting fieldwork in Inner West—the area of Research City renowned for its high-profile gangs—however, we never observed large groups of gang-involved (or indeed other) young people in public places there, in spite of our regular observations (almost daily at times) in its neighbourhoods during the research. Our observations confirming a lack of street presence amongst these gangs today is particularly compelling because during the late 1980s and early 1990s, these Inner West gangs had in fact been street-based, with many members participating in open-market drug sales (Medina et al. 2011). Young people in Inner West gangs only very occasionally referred to being in public places, primarily parks and immediately outside their houses. Some gang members made use of particular local youth centres, and older members referred to being associated with a particular bar or gym. But this kind of public presence was not the norm amongst the Inner West gangs of Research City, even though it was elsewhere in the city, and as we documented in Far West. More often, gang members spent their time inside one another’s houses and flats. Where we did observe the use of public space by small numbers of gang members, this occurred more for younger members than older members, the latter being more likely to have access to residential accommodation they could occupy with privacy. We can speculate as to some possible explanations for our finding that Inner West gang members seem to be less likely than some of their American counterparts to be street-oriented. The weather was often mentioned by gang members themselves and by the police and other local authorities as a reason for not spending time on the streets. The short days in the winter months, and regularity of rain throughout the year, perhaps make the outdoors less appealing. In spite of the regularity with which this reason for staying indoors was offered, however, this is not for us the most compelling explanation. Non-gang young people living in “gang” neighbourhoods (see Ralphs et al. 2009), as well as youth affiliated with gangs, reported experiencing “hassle” from the police in the form of policing that involved formally recorded “stop and search” as well as the unrecorded “stop and account” where a search is not conducted (also known as “stop checks”). This perception by young people of police hassle is corroborated by the police themselves: If we suddenly see Crestside Crew emerging on Long Lane, bang, get over there, get our uniform lads to absolutely hammer them, harass them, do them for anything they can. So they basically think, “We’ve had enough of this” and we can dampen things down […] but [hanging around on the streets] from our point of view makes them a target. Lowerside lads can drive by, so if we’re dispersing them and displacing them, it’s diminishing the problem again. [KI-16]

This style of policing was also evident to us in our observations. Where youth gangs lack a street presence, therefore, this may result when young people have withdrawn as a result of the police actively hassling them when they occupy public spaces. It was not only the police who observed the supposed vulnerability of gang


39

members in public places: gang members did too, and regularly referred to the fear of gang-connected drive-by shootings, scores of which had occurred over the years, some fatal (see Aldridge et al. 2011b). So it is likely that Inner West gang members tended not to occupy public spaces, in contrast to their Far West counterparts, because of their desire to avoid what they feared might be dangerous encounters with other gang members or with the police. Yet anoter possible explanation lies with the trend amongst young people generally to spend more time indoors and online. In fact, we found considerable evidence that gang-related conflict and dramas were played out online, particularly on the most popular social networking site at the time—Myspace.3 As well, we observed that young people on curfew, sometimes via court-mandated electronic “tags,” were required to be at home between 7:00 pm and 7:00 am. This resulted in gang members congregating to “chill” at the residence of the young person on curfew. In some instances, neighbours complained at the regular comings and goings of young people that resulted. Finally, we note a further possible explanation for the non-street-based orientation of these groups. As we discuss elsewhere (Aldridge et al. 2011b), as Inner West gangs have evolved over a number of decades, they included increasing numbers of “residential outsiders” as members. We noted a range of ways in which gang members come to reside outside the neighbourhoods that their associated gang is identified with, by name or origin. In short, some never lived in the gang associated area, others originated there but were moved out by officials, whilst others made a calculated decision to move out due to unwanted attention from rival gangs and the police, to less “hot” areas or as part of a process of desistance. Rather than gang members residing and spending most of their time in their “gang” neighbourhood, we have shown here how gangs in the Inner West area of Research City have members who reside in areas across the city and indeed, beyond the city’s boundaries. The existence of residential outsiders in gangs—the numbers of which we contend may increase as a gang’s existence continues over generations—in itself may mitigate against gang members being street-oriented. In short, as places of residence of members spread out geographically, the ease of gathering in public places diminishes. We do not wish to imply here that youth gangs in Research City are strictly nonstreet-based groups. Some groups spent more time in public places than others (e.g. groups in Far West seemed to spend more time in this way than those in Inner West; and younger members more generally were more likely to be street-based than older members, who are more likely to have their own, or access to, flats or homes in which to spend time). However, for at least some Inner West gangs (Upperside and Lowerside in particular), we contend that most observers would not designate them as street-based. We suggest that this defining criterion of the EG definition may unwittingly, in some instances, exclude groups we would want to consider gangs. The question remains, however, whether—by evolving into primarily non-streetbased groups—these groups may more properly be considered organised criminal gangs—a different kind of entity to the youth street gang. However, we do not believe that the defining nature of these groups has changed, even though their street orientation has. These groups continue to be primarily youthful, even though the groups have existed under their current names for about three

40

J. Aldridge et al.

decades. They continue to engage in the range of illegal activities typically carried out by young people in youth gangs in the USA, the rest of the UK, and elsewhere, including drug sales, robbery, and violence, as well as in more low-level delinquency. These groups have not moved into higher level organised crime—but on the contrary have become somewhat less organised over the past two decades, albeit from an atypically high level of organisation in the late 1980s and early 1990s, when they were more properly considered what we have termed “quasi-specialist” drug gangs (see Medina et al. 2011). Finally, we argued above that the shift for young people in these gangs to spending more time indoors has in part resulted from intense policing of these groups in public spaces (evidenced by the fact that we still find a street presence amongst the groups we observed in areas not being policed in this way). So long as this kind of policing does not change the nature and activities of these groups in relevant ways, we are not inclined to conclude that these groups are no longer gangs—just that the location of much of their daily activity has changed.

3.3.2

Involvement in Illegal Activity Is Part of Its Group Identity

Whilst the emphasis on street presence may unwittingly exclude some of the more violent gangs, conversely, as we go on to outline below, the ‘group identity’ element of the Eurogang definition result in the potential of including non-gang groups. In the course of our ethnographic research (and since), we encountered groups that (arguably) we would not wish to consider as youth gangs, but who meet all the criteria in the EG definition: durable, street-based group of three or more mostly young people, including the key involvement in illegal activity is part of its group identity. Examples of three such groups presented as case studies follow. The public park pot smokers. We encountered groups of young people who spent time hanging out in public parks smoking cannabis. For some, this particular illegal activity—drug taking—was “defining” of the group—what they were “about.” We learned about one such group from now 17-year-old Libby, who, when she started going out with boyfriend Luke, became a regular part of one of these groups when she was 14. Luke had been a regular cannabis smoker for some time, but up until then, Libby had no experience with cannabis. The group would meet at a park near Libby’s school, when school let out in the afternoons when the weather was pleasant. There were about five core members who met regularly, and a few others who came less often. Libby recalls playing “taxi” her first time with the group. Everyone sat in a circle and passed round a joint so that people in the group could inhale from the joint in turn. The “game” was that each smoker was not allowed to exhale until the joint came back round to them again. Libby said that these park meetings were about having fun and socialising, but smoking weed was always a part of this, and the key reason for meeting where and when they did. They organised themselves in such a way as to facilitate their drug taking. They located themselves in a remote part of the park so that they could smoke without being seen by other park users. Each member of the group took turns procuring and providing the drug, and Libby


41

herself did this on one occasion. After about a month, Libby stopped attending these after-school meetings. The fact that she was not enjoying the effects of the drug was critical in making her decision, but so too was her new-found reputation amongst her close friends for being a “druggie.” This was not a reputation she was happy with, because her school friends at that time looked down on drug taking (although only a year later, cannabis smoking was considered fairly unremarkable activity amongst Libby’s friends, most of whom had become at least opportunistic users). The group existed prior to Libby’s involvement, and continued afterwards, for a total of a few years, although the location of their meetings changed to indoor athome locations as the group got older. We are aware of numerous groups of this kind. They tend to be dominated by school-age young people, usually in the range of 12–16 years of age, and exist in both middle class and working class neighbourhoods. It would be unusual for group members to be threatening or violent. Their meetings in public places satisfy the “street presence” criterion, and many, like Libby’s, were sufficiently durable to meet the “3 months” criterion too. And like Libby’s group, for some, involvement in cannabis smoking was a central activity for the group. But would we wish to consider this kind of group a “gang”? Probably not. They do not see themselves as a gang, are not seen by other groups or young people as a gang, are not responded to as a gang by individuals or groups in the community, or by the police, and generally are not seen as much of a problem. Particularly as this kind of decidedly “recreational” drug taking has moved from the margins to the mainstream (a process of normalisation), it has become less likely to be understood as unambiguously deviant, and more likely to be socially accommodated, even by abstainers (Aldridge et al. 2011a). The mere fact of the widespread prevalence of drug taking in this age group—in annual surveys over the past decade, between four and five in ten 15-year-olds in England had ever taken an illicit drug (Omole 2011)—coupled with the fact that the bulk of this activity will occur in groups and in public locations, suggests that the inclusion of drug taking as one of the defining criteria for gang membership might mean that researchers potentially identify vast swathes of ordinary young people as gang members. The illegal ravers. Our second case study is a group of young people whose activities are connected to illegal raves. Illegal raves are unlicensed free parties held in urban and countryside locations: fields, abandoned warehouses and industrial sites. Dance music is played and many or even most who attend take illegal drugs (typically stimulant drugs including MDMA, cocaine, amphetamines and ketamine). In the UK, after the Criminal Justice and Public Order Act 1994 was passed, holding events like these became illegal, and so began the game of cat-and-mouse between those who organise and attend illegal raves on the one hand, and the police and other local authorities on the other. Hayley, 21 and currently a second-year law student, has been a member of a group of young people who attended illegal raves from the age of 14. The core group of five were mostly white middle class boys from the city in which she grew up. There were many factors that appealed to Hayley about illegal raves: she liked the happy vibe of the crowd of people that events like these drew, the lack of age

42

J. Aldridge et al.

restrictions on attendance and of aggressive lads or security staff looking for fights, the ability to take drugs (including tobacco) freely and openly and the fact that all of this was hers for the price of bus fare and “a few quid on pill.” Mass intoxication on illegal drugs was the key appeal of raves for Hayley and her friends: “Not just me on the drugs. Everybody all being there, all being just as fucked as each other. Going west [being intoxicated] on pills wasn’t the point, it was everybody doing it together, I reckon.” Hayley still attends raves when she can, but now restricts her attendance at these mostly outdoor events to the summer only. Groups like Hayley’s involved in organising or just attending illegal raves may be seen to fit the definition of a gang: the particular illegal activities are defining of the groups’ existence—they come together to organise or attend events that are by definition illegal, and to take illegal drugs in the process of doing so. It is therefore these illegal activities that give these kinds of groups meaning. But like the public park pot smokers, we suggest, these groups should not be considered gangs: they do not see themselves as gangs, and are not seen by others in this way. Although the police generally have an intolerant attitude to these kinds of groups and attempt to disrupt their activities, the groups themselves are usually not only non-violent and nonthreatening, but often espouse peace-loving attitudes and behaviours. Hayley’s group was for the most part law-abiding outside of these events and their drug taking. The clubbers. Our third case study is a group of clubbers who take stimulant-based dance drugs when they attend licensed dance club venues. Although going to night clubs (unless underage) is not illegal, their use of illegal drugs at these venues is. As with the public park pot smokers and the illegal ravers, these groups can be seen to fit the EG definition of a gang because taking illegal drugs in these licensed venues is defining of the group’s existence, providing both identity and meaning for the group members. Twenty-nine-year-old Ph.D. student Jessica has been going clubbing with friends she met at university for about 10 years now. They are all in professional jobs: primary school teacher, fundraiser, manager of a care home, manager of a travel agent, web designer and urban regeneration manager. “There [is] a core group of about seven who meet up regularly and have a shared interest in clubbing and drug taking. Drug taking—it’s just part of what we do rather than the sole reason we always meet up. I think it is a major part of why we are closer to each other than the rest. We were introduced to drugs at university in Sheffield and Gatecrasher was just coming to the end of its peak. So those of us who enjoyed taking drugs and going clubbing have continued to do that afterward.” When asked if taking illegal drugs is seen as “okay” in the group, Jessica replied: “Absolutely. It’s a shared interest. It’s why we are going: music, dancing, drugs are all important to us. We wouldn’t go clubbing without drugs, just because it is a major part of the experience.” When asked about the group’s orientation to other kinds of criminal activity, Jessica states: “I think that minor criminal acts we see as ok. Anything that involves violence would not be ok. Bit of middle class fraud is accepted such as claiming on the insurance for your mobile phone if you have lost it.”


43

Like the other groups discussed here, Jessica’s group does not see itself as a gang and is not seen by others in this way. Her group is not threatening or violent, and is generally law-abiding outside of drug use. And although the identity of her group is clearly oriented around illegal drug taking, and has been for 10 years, we would be unlikely to consider her group a gang. One might question whether spending time in a licensed club qualifies as a “public place” or not, although doing so might be seen as consistent with the EG definition of “street-oriented” as including indoor shopping malls (Weerman et al. 2009, p. 20). After each of these interviews, Libby, Hayley and Jessica were presented with the EG Youth Survey item: “Is doing illegal things accepted by or okay for your group?” All responded “yes,” although each (independently, and without being asked) requested to specify which illegal things their group accepted. We will return to a more detailed consideration of this Youth Survey item below. The findings from these case studies are echoed in research by Maher (2007), who observed 95 street-based youth groupings in urban and rural locations of South Wales. Maher deemed about half of these groups to have “gang-like” attributes; specifically, these corresponded to Yablonsky’s (1960) “contemporary” or “traditional” gang; or to Maxson and Klein’s (1995) “collective” gang. Amongst the remainder of the groups Maher observed (which included subcultural groups like skaters and goths, or her younger, aged 10–15 “wannabees,” who attempted to mimic the activities of older groups), however, were some that met the Eurogang definition. That is, some of Maher’s remaining “non-gang” groups were those for which illegal activity could take on a “defining” quality, as it appears to for our public park pot smokers, illegal ravers and clubbers. Amongst Maher’s groups, these activities typically included underage drinking, drug taking or graffiti. Thus, although not all of the groups Maher considered to be non-gang groups corresponded to the Eurogang definition, some of them clearly did. The three case studies reported above, and Maher’s distinctions between gang and non-gang groups, suggest that there may be a threshold or “tipping point” in deviance or criminality that a group must engage in for the group to be considered a gang. What this threshold might consist in is discussed in the final section of this chapter.

3.3.3

What Is “Group” Identity and How Do We Determine It?

Above, we described three non-gang groups for which a particular illegal activity was centrally important. Was this illegal activity part of their “group identity”? The EG definition requires researchers to draw conclusions based on their observations or questioning about group identity. But what exactly is “group identity”? Gang researchers such as Vigil (1988), drawing on the classic work of Erikson (1963), have written about gang group processes and “gang identity.” However Vigil’s purpose, consistent with Erikson’s, is to understand how interpersonal relationships—including relationships sustained through membership in a gang—are implicated in the formation of an individual’s identity. Thus, terms used by Vigil

44

J. Aldridge et al.

like “gang identity” and “group identity” refer not to the identity of the collective, but to how membership in a particular collective, with its particular set of norms and values, is implicated in the identity of the individual. The explanatory focus, therefore, is how individuals make sense of themselves in relation to the group. Gang researchers have often referred to group level processes—group norms and group values. But to the extent that the term “group identity” has been applied, it seems to boil down mainly to group norms and values. Vigil’s discussion of the gang as “ego ideal” (Vigil 1988, p. 432) nevertheless hints at recent attempts by social psychologists drawing on the work of Social Identity Theorists such as Tajfel (1978). Vigil describes the way in which, for some gang members, concerns about the self can become submerged under group concerns. For a discussion of social identity perspectives and their application to gang dynamics, see Hennigan and Spanovic (2012). This kind of work is likely to have considerable utility in understanding more precisely how (and when) individual gang members begin to favour and enact group norms and values over (possibly conflicting) individual ones, and provide clues, therefore, about key locations and timings for desistance-focussed interventions. It is clear that some relevant theoretical and conceptual work in relation to gang group processes is occurring, therefore. However, we are less clear that this work is likely to result in the conceptual clarity that would allow researchers and others to identify unambiguously the identity of the group. We turn now to problems we have found with existing attempts to operationalise the conceptually and theoretically ambiguous/under-developed “group identity” concept.

3.3.4

The Youth Survey Instrument: Items That Operationalise the “Illegal Activity” Aspect of the EG Definition

There are two questions used in the youth survey instrument to operationalise the aspect of the EG definition requiring the group’s involvement in illegal activity to be part of its identity as a group: 1. Do people in your group actually do illegal things together? 2. *Is doing illegal things accepted by or okay for your group? It is in relation to the second (*) of these questions that we turn our attention now. Illegality versus criminality. Only a minority of acts that are illegal are also criminal or delinquent, but the legal status of certain behaviours as criminal or simply administrative infractions varies considerably across jurisdictions and countries that belong to different legal traditions. In common law countries (including the USA, the UK and Canada), there is a greater tendency to classify as criminal offences acts that in civil law countries (most of Europe) would not meet the criteria to be deemed a criminal offence. This distinction may be important in relation to the kinds of activities in which young people might engage, and particularly in relation to some


45

of the activities discussed above. A range of activities in which young people engage may, in particular jurisdictions, be legal, illegal-but-not-criminal or criminal: skateboarding, playing certain sports in specific public places, hanging out in ways that cause alarm to others, truancy, downloading copyrighted material from the internet, smoking tobacco in some public places, drinking alcohol and drug taking. We can probably all agree that there are some activities—illegal in at least some jurisdictions—that we do not want respondents to have in mind when they answer the question (“Is doing illegal things accepted by or okay for your group?”); the result of doing so would be to include groups we would not wish to consider gangs (e.g. skateboarders who are all about skating in prohibited locations). However, there is no easy solution to this problem, for a few reasons. What is defined as illegal-but-not-criminal in some jurisdictions may be actually criminal in others, and only an administrative infraction in yet others. Drug taking by young people in the USA, for example, is a criminal activity, but in Spain is neither criminal nor illegal, and only an administrative infraction if done in outdoor locations. Thus, replacing the word “illegal” (“Is doing illegal things accepted by or okay for your group?”) with “criminal” does not resolve our problem. Alternatively, the term “criminal” may carry connotations of seriousness that might result in the exclusion of some individuals we might wish to say “yes” to this question. On the other hand, young respondents may not distinguish between “criminal” and “illegal” acts at all. One solution might be to specify the particular illegal activities that concern us in relation to our defining-of-a-gang purposes by way of a list of examples. The youth survey does include items that specify particular kinds of illegal activities that the group does together; but these are not used as particular examples to guide the respondent in relation to the item: “Is doing illegal things accepted by or okay for your group?” In any case, it is not clear whether there is consensus within Eurogang as to which illegal activities are relevant in relation to this question. For example, in discussions amongst Eurogangsters over the years, some have regarded an identity around drinking alcohol underage as an illegal activity defining of a gang and others have not; the same is true for some drug taking. The additional clarity such a list of activities provides needs to be considered in relation to the trade-off of added complexity, and perhaps even more critically, length, both particularly important considerations when designing survey questions for use with young people, and where the final wording of any question must be amenable to translation into a range of languages. Activities that are criminal or illegal vary across jurisdictions, and this makes instruments intended for cross-national comparisons problematic. However, the key point in relation to this survey item remains: what transgressive activities, in relation to group identity, are agreed to be conceptually relevant? How mild might these activities be for a group to be a gang? Or put another way, how severe must they be? Validity for item: “Is doing illegal things accepted by or okay for your group?” It is possible that respondents may have a number of interpretations in mind when answering this question. We refer to the first possibility as the “values/morality question,” involving respondents in expressing whether doing illegal things is acceptable within their group. This is the intended function of the question as it

46

J. Aldridge et al.

stands. However, we suggest another possible interpretation for this question; what we refer to as the “difference between right and wrong question.” Under this interpretation, respondents may be demonstrating in their answer that they (or members of their group) “know” right from wrong. The 2004 Offending Crime and Justice Survey (OCJS) in the UK included the item—“Is doing illegal things accepted by or okay for your group?”—but found that “seventeen percent of those who said that their group did not believe it was OK to engage in illegal activities then went on to admit that the group had in fact committed illegal activities together when presented with a list of various activities” (Sharp et al. 2006, p. 39). There are a number of possible explanations for this anomalous finding,4 but the explanation that concerns us here relates to the possibility that respondents may be interpreting the question to be asking if they (or members of their group) know the difference between right and wrong. Thus, respondents may wish to present themselves as a “particular kind of person”—one that knows right from wrong, even if they are themselves morally at ease engaging in the illegal activities they recognise others perceive as wrong. This possible interpretation is consistent with Matza (1964) and others who have suggested that most criminals do not think that committing crime is acceptable, even if they do go on to engage in criminal activities. This same argument has been taken one step further when applied to the kinds of rationalising or “neutralising” statements made by young people interviewed about their drug taking. These statements have been interpreted by some (e.g. Shiner and Newburn 1997) as evidence of young drug users holding wider consensus values rather than subcultural values. Other commentators (e.g. Aldridge et al. 2011a) have interpreted these statements differently. These neutralising statements made by young drug takers to an interviewer—an interviewer possibly perceived as judgemental—can equally be interpreted as the interviewee demonstrating an awareness of wider anti-drug consensus values. Thus, statements like these are instead healthy and appropriate rationalisations that provide a coherent and acceptable personal narrative to a possibly judgemental observer in the form of an interviewer. In other words: “I know right from wrong, even if I choose to do otherwise,” or “I know that others view my actions as wrong, even if I see things differently.” Thus, a negative response to the question—“Is doing illegal things accepted by or okay for your group?”—may reflect perceived demand characteristics of the interview or questionnaire situation. The fact that cognitive testing of the Eurogang Youth Survey questions carried out on behalf of the OCJS revealed confusion about this item amongst pilot respondents is further suggestive that alternative interpretations are possible for this item, calling into question its validity. Another validity-related issue arises when we ask whether this item is in fact an accurate operationalisation of its underlying concept: “a group for whom involvement in illegal activity is part of its group identity.” On the face of it, a question about the normative orientation of the group (“doing illegal things is okay”) does not appear to be operationalising the concept it purports to measure, whether doing illegal things is part of its group identity. A young person could be a part of a group that thinks that doing illegal things (say, shoplifting) is “okay,” but shoplifting is neither part of what they do as a group nor part of what the group is about. This observation may call into question the construct validity of the item.


47

We can also reflect on the validity of the item in relation to the three case studies presented above. All those interviewed about their participation in three youth group formations (the public park pot smokers, the illegal ravers, and the clubbers) responded “yes” to the question: “Is doing illegal things accepted by or okay for your group?” The fact that each interviewee asked to specify the activities they were including in deciding their response (e.g. taking but not selling drugs, claiming insurance for a “stolen” mobile phone that was actually lost, but not large scale insurance fraud, and so on) suggests that a range of interpretations of “illegal things” is available to, and used by, respondents. More importantly, however, their positive replies to this question suggest that their groups would be considered gangs according to the EG definition had they been youth survey respondents. We would not consider their groups to be gangs, and suggest that to identify them as such is to overestimate the prevalence of these groups.

3.3.5

EG Definition Gangs in Surveys: Unexpected Findings in Relation to Offending

Here we report on survey data from Madrid (Spain) and England and Wales to illustrate some of these problems. We have cross-tabulated gang membership5 with the question about the type of criminal behaviour the youth groups (in which the respondents are embedded) engage. The Offending Crime and Justice Survey (2004), for example, asked whether the group has threatened or frightened other people, used force or violence, had broken, damaged or destroyed things, stolen things or sold drugs, inter alia. The Madrid Survey (2008) used a similar battery of questions about group behaviour. We created a new variable based on responses to these questions with three response levels: the group engaged in none of these activities, the group used drugs together (and no other form of illegal/criminal activity), or the group engaged in other forms of illegal/criminal behaviour (violence, property crimes, etc.). The cross-tabulations (see Table 3.1) suggest that in both samples similar percentages of young people (about 12%) that are counted as gang members belong to groups that do not engage in any of the listed criminal behaviours. This raises the question about which illegal acts respondents are thinking of when they report that their groups do engage and approve of illegal behaviour (e.g. downloading copyrighted material from the internet), whether we may want to think of these illegal behaviours as significant or relevant enough for us to count these

Table 3.1 Types of delinquency carried about by the gangs of respondents classified as gang members (using EG definition) No delinquency Taking drugs together only Other delinquency Madrid regions 2008 11.9% 33.3% 54.8% N 427 354 313 OCJS 2004 11.8% 13.4% 74.8% N 1,952 144 519

48

J. Aldridge et al.

individuals as gang members, or whether there is an element of measurement error here that we simply do not yet understand. Equally, in both samples, a proportion of the so-called gang members belong to groups with the only recognised “antisocial/ illegal/criminal” behaviour of using drugs. In England and Wales, this amounts to 13%, and in the Madrid sample, about a third of the gang members.6 This suggests that depending on the context a significant proportion of the so-called gang members may simply be groups of young people that take illegal drugs together. It is clear that a range of definitions employed for counting gangs has predictive validity, including self-nomination as well as the Eurogang measure (e.g. Esbensen and Weerman 2005). These different definitions therefore scoop up subsamples of young people that are behaviourally and attitudinally distinct from non-gang youth in important and predictable ways. Recent research (Matsuda et al. 2012) suggests that these different definitions do not seem to converge, instead picking up different subsets of delinquent youths. This suggests that we may still have some way to go in formulating a consensus definition that links up conceptual and operational definitions of youth gangs.

3.4

Alternatives and Discussions

So far, we have questioned two aspects of the Eurogang definition. We provided evidence that the key defining criteria in the Eurogang definition—involvement in illegal activity is part of the group’s identity—may be problematic on a number of levels. We argued that groups that we would not wish to consider youth gangs (public park pot smokers, illegal ravers, clubbers) would be defined as gangs because of this definitional criterion. We suggested further that “group identity” may be a problematic concept, and identified some validity issues that concern us. And we have provided evidence that at least some groups that we would wish to consider youth gangs (they and others consider themselves to be gangs, encounter official and unofficial responses consistent with this label, and met the Eurogang definition criteria) are not particularly street-oriented. We turn now to an alternative to the key defining criteria in the EG definition that we found to distinguish gang from non-gang groups during our time in the field for our ethnographic YOGEC research. The feature of these groups that separated them from other youth group formations that we would not wish to consider gangs (like those in our case studies presented here) was this: the group has a reputation for a willingness to resort to violence or its threat, even if rarely enacted. The key features of this definition, in more detail, are as follows: 1. The reputation in question is for a willingness to resort to violence, rather than violence itself. As a result, particularly established gangs may have little need to engage in violence where its reputation precedes it, hence the qualifier “even if rarely enacted.” Other gangs and community members may be unwilling to challenge gang activity by virtue of this reputation alone.


49

2. It is the group that has this reputation, rather than each of its members. Thus, not all members are required to be violent themselves, or individually to have a reputation for violence. 3. The group’s reputation is acknowledged not only by group members themselves but also by others outside of the group, and can include: other young people in the area, other gangs, residents of the wider neighbourhood or community, police and other statutory or community agencies, or as expressed in media reports. In our research, this “reputation for violence” criterion actually distinguished what we would want to consider the “gang” from the “non-gang” groups we observed, in particular, those who regularly and centrally engaged in illegal activities like drug taking, illegal raves and clubbing. It also distinguishes amongst the gang and non-gang groups identified in Maher’s (2007) research (Maher, personal communication). This criterion does not seem to result in (at least some of the) observational over-definition we found using the EG definition—unless these groups simultaneously held a reputation for violence or its threat, they would not have gang status. A reputation for a willingness to resort to violence does not require the gang to specialise in violence, as none of the gangs in Research City did, and as other gang researchers have suggested is rare (e.g. Decker and Van Winkle 1996). It does not require the group to be especially violent, or violent very often—indeed, some of the more entrenched and notorious gangs in Research City did not often have to prove their reputations; their reputations preceded them. It was often the less established gangs in Far West, or the members of younger cliques within the more established gangs in Inner West, who were more regularly and spectacularly violent. Our proposed criterion has the advantage of allowing us to define and understand gangs not just from the perspective of what members think about themselves, but in relation to how they are responded to from without—by other groups and gangs, by others in the community and by police and other statutory and community agencies. This is significant because it gives purchase on the gang in its social context. A gang that “thinks it’s tough” but is not viewed as such from the outside may not be of practical importance or significance in terms of whether the group is a problem, or worthy of intervention. This can simultaneously be seen to present a potential measurement problem in survey research. The requirement to understand this reputation from without as well as from within means that making this determination using self-report surveys nearly impossible (although researchers might ask if respondents believe that their group holds this reputation more widely, and, as some social network analysts have done, link replies amongst those in a network, thereby assessing external views of a group). This “reputation for violence” criterion, when operationalised, may be more appropriate to research that involves observation in a social setting (e.g. ethnographic research). There is a range of features connected to the “reputation for violence” criterion that is beyond the scope of the discussion here, but that we intend to elaborate in future publications. (Some of these include: its utility in distinguishing between youth gangs and organised criminal groups, accommodation of the finding that

50

J. Aldridge et al.

members engage in illegal opportunities with those outside the gang, and trade on their gang reputations even decades after desistance.) We hope that the “reputation for violence” criterion can be debated and discussed by gang researchers in relation to its utility in defining, and therefore better understanding, gangs in a crossnational context.

Notes 1. Forty gang members ranging from individuals in their teens to others in their 40s, 59 “associates” and 21 key informants (police, youth workers, community activists). Only those individuals for whom we had clear evidence of gang involvement were classified as members. The gang sample included many of the individuals considered to be “original gangsters” and lead figures in Research City gangs. 2. In the year prior to writing, this official position has changed, with police and local government authorities for the first time recognising gangs in Far West. 3. During fieldwork, we became aware of young people enacting gang conflict (within and between gangs) with a specifically “gang” symbolic context on social networking sites. Gang names, names of individuals in gangs, names of girls/women affiliated to these gangs, names of places (streets, schools) and events (arguments, violence and other conflict) were invoked. Some posted videos of themselves issuing or responding to threats. 4. Sharp et al. (2006) suggest that respondents may have different perceptions of illegality, and that these varying perceptions may have resulted in this finding; alternatively (or additionally) respondents may find it difficult to comment on the orientation of the group as a whole, especially where variation in perceptions amongst members exists. 5. The Eurogang definition of membership relies on a set of questions about the characteristics of the groups of friends of respondents including its street orientation, its durability, its age composition and its delinquent orientation. For more details see Weerman et al. (2009). 6. In Spain, the consumption of drugs in public places is illegal, but not criminal. It could result in an administrative sanction, typically a small fine. Consumption in private is not considered illegal. In the UK, however, the possession of some of the very same drugs for personal use is a criminal offence. This, as well as other sampling differences, may explain the different proportions. As we have highlighted, although the Eurogang concept demands we pay attention to groups involved in criminal or delinquent activities, or behaviour that elicit “fear and concern” (Weerman et al. 2009, p. 19), the actual survey operationalisation uses the broader notion of “illegality” that may capture many activities that are neither criminal nor inspire much if anything in the way of fear and concern.

References Aldridge J, Medina J (2008) Youth gangs in an English city: social exclusion, drugs and violence (No. ESRC RES-000–23–0615) Aldridge J, Medina J, Ralphs R (2008) Dangers and problems of doing ‘gang’ research in the UK. In: van Gemert F, Peterson D, Lien I-L (eds) Street gangs, migration and ethnicity. Willan, Cullompton, pp 31–46 Aldridge J, Shute J, Ralphs R, Medina J (2009) Blame the parents? Challenges for parent-focussed programmes for families of gang-involved young people. Child Soc 25:371–381 Aldridge J, Measham F, Williams L (2011a). Routledge, London


51

Aldridge J, Ralphs R, Medina J (2011b) Collateral damage: gang territory and policing in an English city. In: Goldson B (ed) Youth in crisis? ‘Gangs’, territoriality and violence. Routledge, London, pp 72–88 Decker SH, Van Winkle B (1996) Life in the gang. Cambridge University Press, New York Erikson EH (1963) Childhood and society. Norton, New York Esbensen FA, Weerman FM (2005) Youth gangs and troublesome youth groups in the United States and the Netherlands: a cross-national comparison. Eur J Criminol 2:5 Hennigan KM, Spanovic M (2012) Gang dynamics through the lens of social identity theory. In: Esbensen FA, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, Oxford Maher, J (2007) Angels with dirty faces: youth gangs and troublesome youth groups in South Wales. PhD thesis, University of Glamorgan Matsuda KN, Esbensen FA, Carson DC (2012) Putting the “gang” in “Eurogang”: characteristics of delinquent youth groups by different definitional approaches. In: Esbensen FA, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Matza D (1964) Delinquency and drift. Wiley, New York Maxson CL, Klein M (1995) Investigating gang structures. J Gang Res 3:33–40 Medina J, Ralphs R, Aldridge J (2010) Mentoring siblings of gang members: a template for reaching families of gang members? Child Soc. doi:10.1111/j.1099-0860.2010.00307.x, Online First Medina J, Aldridge J, Ralphs R (2011) Gang transformation, changes or demise: evidence from an English city. In: Hazen JM, Rodgers D (eds) Global gangs: comparative perspectives. University of Minnesota Press, Minneapolis Omole T (2011) Drug use. In: Fuller E (ed) Smoking, drinking and drug use among young people in England in 2010. Information Centre for Health and Social Care, London, NATCEN Ralphs R, Medina J, Aldridge J (2009) Who needs enemies with friends like these? The importance of place for young people living in known gang areas. J Youth Stud 12:483–500 Sharp C, Aldridge J, Medina J (2006) Delinquent youth groups and offending behaviour: findings from the 2004 offending, crime and justice survey. Home Office, London Shiner M, Newburn T (1997) Definitely, maybe not? The normalisation of recreational drug use amongst young people. Sociology 31:511–529 Tajfel H (1978) Differentiation between social groups. Academic, London Thrasher F (1927) The gang. University of Chicago Press, Chicago Tita GE, Cohen J, Engberg J (2005) An ecological study of the location of gang “set space”. Soc Probl 52:272–299 Vigil JD (1988) Group processes and street identity: adolescent chicano gang members. Ethos 16:421–445 Vigil JD (2002) A rainbow of gangs, street cultures in the mega-city. University of Texas Press, Austin Weerman F, Maxson CL, Esbensen FA, Aldridge J, Medina J, Van Gemert F (2009) Eurogang Program Manual: Background, development, and use of the Eurogang instruments in multi-site, multimethod comparative research. http://www.umsl.edu/~ccj/eurogang/Eurogang_20Manual.pdf Wood J, Alleyne E (2010) Street gang theory and research: where are we now and where do we go from here? Aggress Violent Behav 15:100–111 Yablonsky L (1960) The violent gang. Commentary 30:125–130

Chapter 4

Gang Member: Who Says? Definitional and Structural Issues Hannah Smithson, Leanne Monchuk, and Rachel Armitage

4.1 Introduction The last decade has witnessed increased media interest and consequently political interest in the UK in youth gangs. This has mainly been a result of high-profile shootings and stabbings attributed to gang activity. Despite this, British academics remain divided over the existence of delinquent gangs within the UK. There is a certain amount of reluctance to use the word “gang” and “gang member” in the UK (Alexander 2008; Hallsworth and Young 2008). Resolving this debate is made more difficult by the challenges of defining gangs and identifying their members and associates, a problem which has also plagued gang research in the USA. The conflation of labels such as “gangs,” “delinquent youth groups,” and “organized crime networks” has added further confusion to the debate. The purpose of this chapter is to outline the number of definitions which are being used across the UK and to explore how these definitions are perceived by practitioners and applied in practice. The chapter reports the findings of research conducted in a city in the North of England referred to as “North City.”1 The aim of the study was to answer a number of broad ranging questions surrounding youth violence, including an exploration of young people’s motivations to join a gang and become involved in the use of weapons, including firearms. This chapter reports the findings from this study and provides a discussion of the structure and formation of gangs in North City and encompasses the views of the young people identified as gang members and those responsible for this identification, i.e., police officers and representatives from a range of agencies and organizations (e.g., Probation Service, Children’s Services, Youth Services, Neighborhood Management).

H. Smithson (*) Department of Sociology, Manchester Metropolitan University, Manchester, UK e-mail: [email protected] L. Monchuk • R. Armitage Applied Criminology Centre, University of Huddersfield, Huddersfield, UK F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_4, © Springer Science+Business Media, LLC 2012

53

54

H. Smithson et al.

Formal interviews were undertaken with 45 “gang members,” their associates, and key informants such as senior and operational police officers working in North City’s specialist gang and firearm response unit. Findings demonstrated that few of the young people viewed themselves as belonging to a gang and contested the applicability of being labeled as such.

4.1.1

Defining the Gang: The UK Experience

In 2004, the Home Office defined delinquent youth groups as “youthful groups which have durability and structure and whose members spend time in public places and engage in delinquent activities together” (Home Office 2004, p. 1). By 2008, the Home Office, through the Tackling Gangs Action Programme (TGAP), was beginning to place greater emphasis on the use of guns with the definition: A group of three or more people who have a distinct identity (e.g. a name or badge/emblem) and commit general criminal or anti-social behavior (ASB) as part of that identity. This group uses (or is reasonably suspected of using) firearms, or the threat of firearms, when carrying out these offences. (Home Office 2008, p. 23)

More recently, in 2009, the Centre for Social Justice Working Group, a group comprised of “…prominent academics, practitioners and policy makers” (The Centre for Social Justice 2009, p. 2), sought to apply a universal definition to be adopted by all those tackling gangs in order to end the confusion surrounding the terminology used within this arena and to allow comparative analysis between different studies. The group defined gangs as: A relatively durable, predominately street-based group of young people who (1) see themselves (and are seen by others) as a discernible group, (2) engage in a range of criminal activity and violence, (3) identify with or lay claim over territory, (4) have some form of identifying structural feature and (5) are in conflict with other, similar gang. (The Centre for Social Justice, 2009, p. 21)

In 2010, a joint thematic review by the three HM Inspectorates of Prisons, Probation, and Constabulary found that there was no agreed working definition or common understanding of what constituted a gang among those working in a prison setting or in the community within the UK (HM Inspectorate of Prisons 2010). Of course we must also consider the development of the Eurogang definition and the implications of this definition for the development of UK policy and practice. They define a gang as “…any durable, street-oriented youth group whose involvement in illegal activity is part of its group identity” (Weerman et al. 2009 p. 20). The difference between the Eurogang definition and the definitions provided above is that the Eurogang approach is being used to produce comparable cross-national data sets through the development of youth surveys which allow the measurement of each of the discerning characteristics within the definition. The UK’s movement toward a common definition is promising, but the extent to which this is being applied by organizations and agencies, such as the police, is questionable and remains inconsistent

4 Gang Member: Who Says? Definitional and Structural Issues

55

across the country. This incoherent approach to defining what constitutes a gang is highlighted by a number of UK-based research projects which have been conducted to establish the extent and nature of gang involvement in the UK.

4.1.2

The Extent and Nature of Gang Involvement in the UK

A number of UK-based research projects have identified the presence of gangs. Stelfox (1998) for example, surveyed all UK police forces to explore the scope of gang activity across the UK. It is important to return to the notion of the complexities around defining a gang. These complexities were encountered by Stelfox (1998) while attempting to design the surveys to be sent to each of the police forces in the UK: “There proved to be no generally agreed definition of a gang which was applicable to the UK situation which would serve as the basis for gathering data for this study” (p. 398). Stelfox (1998) argues it was important the definition used in this study encompassed the use of criminal activity, and consequently, the definition used for his study included “any group who uses violence or the threat or fear of violence to further a criminal purpose, but excluding football hooligans and terrorists” (Stelfox 1998, p. 398). Of the 48 forces that replied to the surveys, 16 identified gangs were operating in their area. Across these forces, profiles of 71 gangs were returned. The majority of these gangs were composed of adult males. There was only one female gang. Two thirds of gangs were predominately white, one quarter was ethnically mixed and the remainder was predominately a single ethnic group. The average age range of the gang members was between 25 and 29 years old. Gang structures were typically loose, with no identifiable leader. Most engaged in a wide range of offenses, although 17% were described as offense specialists. Three quarters of gangs were involved in some sort of drug dealing. Most forces reported violence as the main problem associated with gangs. Sixty percent of gangs allegedly possessed firearms. Decker (2001) argues that the concept of the gang has been distorted by the dominance of the view that gangs are well organized and tightly structured. Most UK research presents a picture of gangs that are disorganized and typically do not have leaders. Mares’ (2001) ethnographic study of two gangs in Manchester illustrates this pattern. Both gangs were loosely organized and had no formal leaders. Mares (2001) does not outline the definition used within his study; however, they “…were mostly ethnic street gangs involved in the sale of drugs with strong intergang rivalry and high levels of violence” (Mares 2001, p. 154). He describes the heavy involvement of both gangs in drug trading including heroin, crack, and cocaine. There were about 90 members in each gang, and the large majority was Afro-Caribbean in origin, which was representative of the ethnic composition of the area. The members were aged between 10 and 30 years old. Gangs found elsewhere in Manchester were all white, and most gang members were aged less than 25 years old, and some were as young as 10 years old.

56

H. Smithson et al.

Aldridge and Medina (2008), researching gangs in a Northern UK city,2 defined a gang as “durable, street-orientated and have a group identity for which involvement in criminal activity is key” (Aldridge and Medina 2008, p. 3). Similar to Mares’ research, they found that gangs were “fluid, loose, messy and interlinked networks” (Aldridge and Medina 2008, p. 4) very much like informal friendship groups. Interviews with the young people identified that they did not perceive themselves as having a “membership” to a gang—but rather belonging to a social network where they would socialize with their peers. The ethnic composition of the gangs in this study reflected the areas in which they lived, although the authors note that only those gangs from areas with a proportionally higher black minority received media and police attention. This study found evidence of violence, weapon carrying, and drug trading, but the authors argue that gangs were in no way specialized in these activities. In the research Bullock and Tilley (2002) conducted in Manchester, they defined a gang as “relatively enduring identifiable groups of young people who see themselves as members of those groups, and who commit crime as part of that membership” (p. 23). Each of the four Manchester gangs that they studied had a core group of main players and a number of additional and associate members. A large majority of gang members were black and male and were heavily involved in criminal behavior, having on average, 12 prior arrests and two convictions. They committed a wide range of offenses, including serious violent offenses and property offenses. Weapon carrying was common. Bennett and Holloway (2004) adopt Klein’s (2001) definition of a “street gang” and “youth gang” in their research. They summarize their research based on arrestees in Britain by suggesting that there are a variety of gangs but that some common themes may be identified. These include the likelihood that gang members will be male and criminally active (particularly with regard to robbery and drug supply) and have a tendency to carry weapons including guns. Research also indicates that gang members are likely to be white, and the dominant ethnic minority groups are Caribbean and Bangladeshi. However, these differences may be explained by the observation that gangs tend to reflect the ethnic composition of the areas from which they are drawn (Aldridge and Medina 2008; Esbensen and Weerman 2005). For the purposes of this chapter and in an attempt to explain and position our findings, we will focus on structural typologies of gangs as well as on definitional issues. Typologies are invaluable for understanding and delineating distinctly different categories of criminal groups. Klein and Maxson’s (2006) typology of gangs as set out below will be utilized to explore the extent to which the young people involved in our study fit into any of these categories. Traditional gang. A traditional gang has usually been in existence for 20 or more years and continues to regenerate itself. A traditional gang is a large, enduring, territorial gang with a wide age range and several internal cliques based on age or area. Neotraditional gang. A neotraditional gang resembles the traditional gang and is very territorial. However, a neotraditional gang has not been in existence as long, often for 10 years or less.


57

Compressed gang. A compressed gang represents a small gang comprising of up to 50 members. Unlike the traditional and neotraditional forms, there is an absence of territoriality and subgroups in a compressed gang. It is also unclear whether they would grow into the more traditional forms. Collective gang. The collective gang looks like the compressed form but is bigger and with a wider age range. The collective form comprises 100 members or less; they do not have developed subgroups and may or may not be territorial. The form resembles a shapeless mass of adolescents and young adults and has not developed the distinguishing characteristics of other gangs. Specialty gang. The specialty gang tends to be small with 50 or fewer members. It has developed a well-defined territory which may be based on residency or the opportunities for particular forms of crime. The principal purpose of a specialty gang is more criminal than social (pp. 177–178).

4.2 4.2.1

Research in North City Overview

The research was a 10-month study commissioned by a City Partnership Group responsible for North City’s gun and gang crime strategy. Four areas in North City were identified by the commissioning body as the focus of the research. The areas will be referred to as areas A, B, C, and D. It is important to note that North City itself is among the most deprived cities in the UK with relatively high crime rates, low life expectancy, and poor social cohesion. Each of the four research areas included within the study are characterized by these factors. Prior to outlining the methodology, it is important to contextualize the four areas: Area A. Area A has a population of over 15,000, has a lower-than-average household income for North City, and a relatively high level of unemployment. It has a comparatively high proportion of children compared to the other areas in North City, and nearly half of the households comprise lone parent households. The crime rate in this area is slightly lower than average compared to the average crime rate for North City. Area B. Area B has a population of over 14,000 and displays varying levels of deprivation mixed with pockets of affluence. Unemployment levels in this area are lower than other areas of North City, and nearly 30% of the households comprise lone parent households. The crime rate in the area is lower than average compared to the average crime rate for North City. Area C. Area C has a population of 14,000 and is one of the more deprived areas in North City. This area has the highest unemployment rate in North City, and over half of the households comprise lone parent households. The crime rate in the area is higher than the average crime rate for North City.

58

H. Smithson et al.

Area D. Area D has a population of over 14,500 and, like area C, is one of the more deprived areas in North City. It has a lower-than-average income and higherthan-average unemployment rate, and nearly half of the households comprise lone parent households. The crime rate in area D is higher than the average crime rate for North City.

4.2.2

Methodology

The data described in this chapter were derived from conducting semistructured interviews with a total of 45 individuals across North City. The interviews were conducted by two researchers. Fifteen interviews were conducted with practitioners, in the main, representatives from the multiagency partnership group established to try and reduce levels of gang and gun crime in North City. The group consisted of senior police officers, frontline youth workers, and representatives from North City’s Youth Service and Children’s Service. Those interviewed were identified as the best positioned to provide information about North City’s gang and gun crime strategy and information about each individual agency’s role within the strategy. Interviews were also conducted with thirty young people identified as being involved with gangs by the agencies outlined above. The young people were recruited for participation in several ways: through referrals from North City’s Youth Service and Probation Service, through ethnographic fieldwork and undertaking detached youth work, and finally through prison interviews. The semistructured interviews were undertaken using a narrative approach, which allowed respondents to present their life story. Prompts were used to gather specific information about life in the gang, motivations to become involved, methods of making money, use of weapons including firearms, levels of violence, and desistance from the gang. The recruitment of the young people was undertaken in the four areas (areas A, B, C, and D). The authors believe that the number of participants interviewed including practitioners and young people was considered to be sufficient for this study. It is important to reiterate at this juncture that the research study was commissioned, conducted, and completed within a 10-month time period. Due to the limited number of respondents, we are not suggesting that the findings be generalized to other cities in the UK, but nevertheless, we believe that they add an important dimension to the paucity of UK research. It should be noted that a considerable amount of time was used to identify, include, and explore possible methods of recruiting young people. As discovered by Aldridge and Medina (2008), we also found that it was important to devote a considerable amount of time to establishing rapport and trust with the young people to be included in the research. It was evident that some of the young people were apprehensive as to whether or not we were actually police informants. To alleviate any concerns, we spent time establishing contact with the young


59

people, attending youth clubs and accompanying youth workers on their detached work. We also reassured them that any information provided would remain anonymous and confidential. The interviews with the young people lasted between 30 min and 2 h and were undertaken in a variety of venues, such as on the street, youth centers, houses, staff offices, and prison rooms. All respondents were asked to read an information sheet and to sign a consent form prior to commencing the interview. All interviews were recorded, transcribed, and then stored electronically. Due to the sensitivity of the research, a number of security safeguards were implemented. For example, passwords and encryption software were used to prevent unauthorized access to the files. As described previously, the notion of gang membership is problematic. However, the sample of the 30 young people interviewed included 18 young people labeled as gang members by police officers, youth workers, or probation staff. The majority of these young people were either serving or had served prison sentences for gangrelated offenses including firearm offenses. Twelve “associates” were also identified by the above agencies as being on the periphery of gangs and/or whose friends were involved in gang activity. Six of the sampled young people were, at the time, known to the gang and firearm specialist unit. They were individuals whom the gang and firearm specialist unit were observing or pursuing for gang- and firearm-related offenses. All of the young people were aged between 16 and 29 years old. Twentynine were male and one was female. All were White British in ethnic origin. None of the young people were identified by us as gang members, nor did they self-identify as gang members. It was our task, as researchers, to explore the views of the young people with regard to the labels attached to them by agencies and their views on the existence of gangs and their membership in these gangs. The following sections draw together the conflicting views of young people and practitioners in North City.

4.3

The Challenges of Defining the Gang Within North City: The Practitioner’s Perspective

The challenges of defining gangs and gang-involved young people were reflected by the practitioners interviewed. North City’s partnership group does not have a common agreement about what constitutes a gang as the following quote demonstrates: Not only is there not an agreed definition here in [North City], there’s not an agreed definition in the country. Common sense would tell you that the TGAP [Tackling Gangs Action Programme] definition is a reasonable explanation of what a gang is, but in this area you know there’s a lot of dysfunctional, disorganisation. Our groups of young males who we are terming “gangs” and they themselves see themselves as being part of a gang but quite frankly we’ve got young kids who do nothing more than cause ASB [anti-social behaviour] and they’re just being risk taking kids. Now would you say they’re being a gang? Well the definition might lead you that way. So I’m not so sure that definition is crucial to looking at what we’re doing. (Police Officer 1)

60

H. Smithson et al.

The difficulties in defining and identifying gangs are borne out by the fact that several practitioners admitted that they did not really know the extent of gang involvement in North City. One police officer was quite candid in this respect: If you want my honest opinion I don’t actually think we’ve got to the bottom of that. I have been a purveyor of an argument for quite some time now that we don’t understand the gun crime and gang problem. (Police Officer 3)

A number of other practitioners were particularly robust in their claims that gangs did exist in North City, with one police officer stating the number of gang members the city was currently dealing with: We’re roughly dealing with about 102 people associated with gangs. That’s it in terms of numbers. In terms of incidents, well we’ve seen a 47% reduction in firearm discharge that doesn’t mean we’re complacent; there are still some firearms out there. (Police Officer 5)

The above remarks return us to what for many respondents was a central issue, that the extent of gang and gun involvement in North City was heavily dependent upon how one defined a gang. Our findings echo those of Decker and Kempf-Leonard (1991). Their study attempted to find differences in the understanding of gangs among a number of different persons: police officers, members of an antidrug/antigang task force, and juvenile detainees. Interestingly, the police used the most narrow and restrictive definition of gang membership. Despite the UK government’s encouragement of practitioners and agencies to subscribe to the universal definition of a gang as set out in the “Dying to Belong Report” published by The Centre for Social Justice (2009), it is clear in North City that this has not occurred. Of greater concern is the fact that practitioners in North City do not think that definition is crucial. This begs the question as to how practitioners identify young people as gang members and once they do, how gang members are then dealt with. Part of North City’s gang and gun crime strategy was to devise intervention projects for gang-involved and gang-related youths. Without a clear definition, it is difficult to discern how such projects could be developed. The following sections demonstrate that although a common working definition was not in place, some young people living within the research areas in North City were identified as gang members. Moreover, practitioners were even creating names for groups of young people congregating together, based upon where they lived or where they chose to gather with their friends.

4.4

What’s in a Name?—The Young People’s Perspective

When young people were asked about the names given to local groups, their responses were very different. However, the majority felt strongly that these names had been given to groups of young people by the police and the media and that the young people themselves did not use these names. More importantly, the young people did not view themselves as being part of a gang.


61

Despite the police and media attention afforded to gangs in areas A and B, not one of the young people interviewed from these areas referred to their groups by these names. In fact, they showed a great deal of contempt for the names: There is no gangs. All that going on the “…” Gang and that like, there’s no “…”. Gang, do you think it is? Where do you think you are? We don’t think we’re fucking Bloods and Crips and Fucking Dutch and all this and that, what are you like? There are no gangs, you are all Muppets! (Young Person 4) I laugh at it all the time I hear it, like the notorious “…” Gang. It was just a gang who hanged around in the… and they called it the “…” Gang. Everyone get mixed into them just cos you lived in…. Someone would say “him there” and then the police would say, he’s part of the “…” Gang. There might have been two or three groups at the…. (Young Person 26)

As mentioned above, many of the young people interviewed spoke about the fact that in their opinion, the police themselves had named gangs. It was mainly young people from areas A and B who had held this opinion: No one goes round saying “…Gang” and “… Gang” and all that. You know what I mean? It’s the police that make them up. Do you think kids are going to walk round like “yeah, I’m in the… Gang” and all that?…look like a little clown? (Young Person 4)

It is instructive to note that despite the acknowledgement that the youths do not appear to view themselves as constituting a gang, they have still been labeled and received a name and are referred to by this name by the police and perhaps other agencies. This raises the question of how much labeled gang activity is due to the (mis)application of naming of what are otherwise young people using public space. This is discussed in the following quote from a young person: As if kids are going to walk round doing that “I’m a gang” you’re mad….police just drive past and see kids standing outside the [name of shopping centre] shops and think “yeah that’s the ‘….’ gang because they are in the [name of shopping centre] and there’s loads of them.” (Young person 7)

This was also confirmed during an interview conducted with a police officer. As the police officer below alludes, most young people do not belong to gangs but can be mistaken for gang members or associates by simply being observed in a particular area with other young people: …the police are very quick to sort of say that these people are members of [a gang] because perhaps they’re seen together when say just walking home from school or something and suddenly he is now a part of a gang. And there is nothing to base that on other than observations that people have seen but they haven’t actually gone that stage further where they have been stopped because of some behaviour, it’s that they have been seen together. (Police 3)

Ralphs et al. (2009) provide an enlightening account of this phenomenon in their findings from research undertaken in an English city. Young people in this study were seemingly harassed and labeled as gang members simply because of the public space they occupied and the friends and associates they kept. Another police officer argued that the uniformity of dress and fashion among many young people made them appear similar for stylistic reasons rather than for gang membership. However, it was clear from the detached youth work we undertook and from the interviews with the young people that the young people were not

62

H. Smithson et al.

interested in any distinctive styles of dress and dressed much as any other young person in North City. The young people interviewed felt labeling groups as gangs served to glamorize violence and encourage young people to aspire to be part of a gang. As one practitioner commented: Someone was telling me a tale about a police officer asking a young person why he had joined a group and he said well, before I was in the group I was nothing, now I’m in the group I’m part of something. I am someone. So the kids are seeing it as getting an identity and respect. (Practitioner 6)

As noted by Howell (2007), media reporting can also exacerbate a gang problem. The media’s portrayal of a high-profile fatality, for example, may exacerbate the situation, contribute to the creation of a “moral panic,” and consequently exaggerate and misrepresent the scale of the actual gang problem. Furthermore, publicizing gang activity can perversely make that gang appear strong and thereby enhance their ability to intimidate and multiply, a consequence that was noted by one of the young people interviewed: “…” never went round saying “we’re a gang” there were just fights know what I mean? Someone got jumped and then all of a sudden guns started being used and then the papers and the police started saying “….” gang. I’ve not once said I’m “…” or “…” Crew, know what I mean? They’re the ones that called us gangs, now all these kids in school are just looking and then all they see any time something happens in the paper is there’s, like all the lads round by ours now they’re all driving round in fast cars and fast motor bikes and they’re only 13 and they’re thinking I want a go. So it’s their fault really with the gangs ‘cos we never called our self a gang. (Young Person 1)

There was also an honest admission that agencies had added to the problem by inadvertently publicizing gangs and enhancing their reputations: They [the police] should never have made us into gangs so they’ve got to take some responsibility. (Young Person 1) It comes back to the issue I said right at the beginning when I said we never understand the problem. So how can we say we’ve solved it when we don’t understand it? We gave the gangs credibility calling them “…” Crew and the “…” Gang. We gave them the self publicity and credibility they required to associate. (Police Officer 2)

The young people’s rebuttals to gang membership cannot be ignored. It would suggest that the gang label is indeed being either misapplied in some cases or overused.

4.5

Gang Structures in North City

Decker (2001) argues that the concept of the gang has been distorted by the dominance of the view that gangs are well organized and tightly structured. Most UK research presents a picture of gangs that are disorganized and loosely structured which typically do not have leaders (Stelfox 1998; Mares 2001; Bullock and Tilley 2002; Bennett and Holloway 2004).


63

It is a similar situation in North City with practitioners believing that where gangs do exist in North City, they tend to be rather disorganized and unstable with unstructured membership patterns. The picture in North City is one of rather chaotic young people who have access to firearms rather than mature organized criminal gangs: There is a natural assumption that they have a hierarchical structure which they can then attack using a variety of tactics. I actually believe in [North City] specifically we haven’t got that. We have got groups of disenfranchised youths who associate on the basis of geography and past history between families… (Police Officer 4)

These perceptions of the loose structures of gang activity were reflected in the majority of responses from young people, few of whom talked about their group as organized with a hierarchical structure. The vast majority felt strongly that their group was not organized, had no structure, and was simply a group of friends doing what they choose to do. This was particularly found in areas A and B. Those who spoke of leaders or hierarchy were generally discussing this in the context of drug dealing, where someone would be in charge of supplying the drugs, leaving them to “graft” (work), and then return the profits to their “boss.” However, others did speak of “top men” and “leaders” when referring to group rivalries. This was more evident in areas C and D. Respondents from mainly areas A and B expressed the view that groups within their area were not organized or structured and that there was no leader or “boss” directing other members. The vast majority of participants highlighted that their gang was just a group of friends with no one in charge of activity: Not as organised as they think really, it’s all over the place to be honest with you. (Young person 14) No that’s what I mean we’re, we’re like, we’re all of us, we’re all equal you know what I mean? You don’t get told by someone like the leaders to go and do something. (Young person 13) It’s just your mates innit? Just chill like, there is no leaders it’s just like where you’re from and that. (Young person 12)

Although their responses suggest that the groups are loosely organized and have no identifiable leader, a large number referred to “older ones,” who were usually in their late teens and early twenties and had expensive cars and nice clothes. The issue of leadership is ambiguous as although a named “leader” may not exist, there may be a slightly more structured hierarchy than becomes apparent through the young people’s responses: There’s no leader who tells us what to do like, but there’s the older people who have us doing stuff for them innit? (Young person 9)

If we use Klein and Maxson’s (2006) typology, it could be concluded that the structures in areas A and B are best described as collective gangs: Size can be under 100 but is probably larger…it has not developed subgroups and may or may not be a territorial gang….resembles a kind of shapeless mass of adolescents and young adult members and has not developed the distinguishing characteristics of other gangs. (pp. 177–178)

64

H. Smithson et al.

In contrast to areas A and B, young people living in areas C and D spoke of fairly rigid structures and of hierarchies: Interviewer 1: Respondent: Interviewer 1: Respondent:

So the only way gangs would stop, you were saying, is if you take their…? [Agreement noise] If their top man… Yeah, when we take down their leaders (Young person 16) and Yeah, yeah they [the leaders] own everything, they own every little kid and everything, every little kid you see running round and all that they own everything. (Young person 18)

Young people in areas C and D spoke of “top” families “owning” the area. Hobbs (2001) refers to family firms as “…long established families who had retained a considerable degree of hegemony within the neighborhood…” (p. 551). He describes family firms as owning areas in the 1960s but suggests that these firms have greatly diminished (Hobbs 2001). However, our findings indicate that these family firms may still be in existence in parts of North City, particularly in areas C and D. This extract between a young person and one of the interviewers is the clearest example of this finding: Respondent: Interviewer 1: Respondent: Interviewer 1: Respondent: Interviewer 1: Respondent: Interviewer 1: Respondent:

They’re the second from top [name of family]. Oh okay so they’re not the top because a lot of people say that they’re like the top but they’re not quite? No They’re just underneath the top? Just underneath yeah So if there’s someone else there, then the [name of family], so where does your family come? I wouldn’t even have a clue innit, I’d myself, I’d rank myself higher innit but I don’t know. You’d rank yourself higher than? Not the [name of family] no. They’re just crazy man. (Young person 16)

Klein and Maxson’s (2006) traditional gang typology may also be a useful way of explaining the apparent rigid structure of areas C and D: …generally been in existence for 20 or more years; they keep regenerating themselves….a large, enduring, territorial gang with a wide age range and several internal cliques based on age or area (pp. 176–177).

This contrasts with much of the research undertaken in the UK with the groups in areas C and D being quite different from the gangs typically depicted in UK research which can be found in areas A and B of North City. Owing to this finding, it is clear that in North City a “one size fits all” approach to prevention and intervention cannot


65

be adopted as there appears to be at least two distinctly different categories of groups. Undertaking further research in North City over a greater period of time may indeed identify more categories.

4.6

Group Territories and Rivalries Within North City

Kintrea et al. (2008) in their study of “Young People and Territoriality in British Cities” defined territoriality as “a social system through which control is claimed by one group over a defined geographical area and defended against others” (p. 1). Case studies showed that territoriality was important in the lives of many young people, although it was manifested in various forms—from young people who socialized on the streets; to groups with a stronger territorial affiliation, some of whom identified themselves as a gang; to more highly organized, criminally oriented territorial gangs. Territoriality often gave rise to physical conflict between groups of young people. Although we are not suggesting that territoriality is a defining characteristic of gangs, our research certainly identifies that in many instances, territory was the rationale behind rivalries and violence among young people in each of the four areas. We found as did Kintrea et al. (2008) that territoriality was manifested in various forms. Practitioners reported that in areas A and B, gangs were involved in dealing drugs (primarily cannabis according to one source) and were primarily motivated by a historic rivalry of two groups who were loosely associated. The importance of local rivalries was apparent from the interviews with young people. The ethnographic nature of the study and the time spent in the areas by the research team enabled us to build a geographical picture of these territorial issues. It became abundantly clear that in most instances, territory and rivalry could be explained by living on one side of a road compared to another or being from adjacent housing estates. Many of the young people from areas A and B were adamant that it was too dangerous for them to venture into the opposing area. Many stated that the only way that they could would be to travel by car. The extract below illustrates this mentality clearly: Interviewer 1: Respondent:

So you wouldn’t go to [area A]? I wouldn’t go there no. Not unless I was in a car or something like that, I wouldn’t walk through there. Interviewer 1: Right. And is it because people would know that you lived in [area B]? Respondent 1: Yeah. I can go there but I just, I wouldn’t just walk through it, know what I mean? Because I know I’m going to get fucked. (Young Person 16) Territorially was discussed by a number of young people who were interviewed from areas C and D, confirming their classification as a “traditional gang” as

66

H. Smithson et al.

described earlier. The following quotes from respondents in these two areas clearly highlight this: Respondent: Interviewer 1: Respondent: Interviewer 1: Respondent:

The people shooting at us who we’ve got a beef with. We’ve got our own little patch and they’ve got their own little patch. So where’s your patch? “….” Road and “….” How far away from one other are these two patches? About 5 min. (Young person 24)

One young person from area D agreed that the issue of territoriality was an instigator for violence and even described how this territorial claim could be determined by which road you travel on: Interviewer 1: Respondent:

Is it more like an area? Roads. You are talking roads not areas. It’s scary…I have been in situations, I have been there and someone has drove past me on a bike and pointed and the only reason why they wouldn’t let that thing off [and use a firearm] was because their cousin was standing next to me. (Young person 25)

The aspect of territoriality was epitomized by a comment made by a young person from area D: Ninety-nine percent of the time they remain in their area, in their segregated little piece of land…they’re living in and most of their little brothers do, they’re all in one area…it’s very rare you’ll see them in [in another area]…you probably will see them occasionally but it’s very rare you see them branching out of their area, it’s unusual… (Young person 2)

Without doubt, the closed environments that many of the young people occupied in each of the four areas generated a strong parochial identification, local loyalty, and sense of belonging, and this without question promoted territorial disputes between rival areas.

4.7

Conclusion

Current gang definitions point to a degree of durability and structure to groups but are vague regarding the degree of structure that constitutes a gang. Most UK and US research presents a picture of gangs that are disorganized and typically have no leaders—hence, the Eurogang paradox wherein street gangs in Europe are not recognized as such because they do not fit the inaccurate stereotypes of US gangs (Klein 2001). Our findings suggest that North City is, in the main, no different. Overall, respondents suggested that gang activity tended to be low level, disorganized, and limited to small geographical areas.


67

Crucially, in contrast to currently accepted definitions, we found little evidence that groups of young people living in the four areas of North City had a “distinct identity” but perhaps more crucially that they did not define themselves as a discernable “gang.” Rather they were defined and labeled as gang members by the police and other relevant organizations. Young people in some of the areas may have resigned themselves to accepting the gang label, but the question remains as to whether they were engaging in what agencies were labeling as gang activity. Our findings have led us to question (not conclude), given the current UK political and policy fixation with youth gangs, whether consideration should be given to dropping the use of the term gang as a helpful explanation of youth violence and delinquency in the UK. This difficulty was highlighted by UK Youth Offending Service workers, interviewed as part of the Youth Justice Board (YJB)’s report into gangs, who expressed grave concerns over what they saw as the indiscriminate use of the term gang. They argued that many young people take part in group offending, but they would not necessarily class themselves as being part of a gang (YJB 2007). The Youth Justice Board suggested the use of the term “delinquent youth groups” rather than “gangs” to more accurately describe youth networks (YJB 2007). It is unlikely that the term gang will be replaced, which makes the issue of employing a robust and consistent definition of central importance. It is clear that any definitions employed will have a significant impact on estimates of the size and nature of the gang problem; how young people are targeted and selected to become involved in a range of interventions designed to curtail their gang involvement and perhaps more importantly, the means by which gang-involved young people are treated within the criminal justice system.

Notes 1. The name of the city has been changed to protect the anonymity of the city and all the participants who took part in the research. 2. This city was not North City.

References Aldridge J, Medina J (2008) Youth gangs in an English city: social exclusion, drugs and violence. University of Manchester. Research report ESRC RES-000–23–0615. 16–01–2008 Alexander C (2008) Re-thinking “gangs”. Report prepared for the Runnymeade Trust. Retrieved from http://www.blackeducation.info/upload/docs/RethinkingGangs-2008.pdf Bennett T, Holloway K (2004) Gang membership, drugs and crime in the UK. Br J Criminol 44:305–323 Bullock K, Tilley N (2002) Shootings, gangs and violent incidents in Manchester: developing a crime reduction strategy. Crime reduction research series paper 13, Home Office, London

68

H. Smithson et al.

Decker SH (2001) The impact of organizational features on gang activities and relationships. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Academic Publishers, London, pp 21–39 Decker S, Kempf-Leonard K (1991) Constructing gangs: the social definition of youth activities. Crim Just Policy Rev 5:271 Esbensen F-A, Weerman FM (2005) Youth gangs and troublesome youth groups in the United States and the Netherlands: a cross national comparison. Eur J Criminol 2:5–37 Hallsworth S, Young T (2008) Gang talk and gang talkers: a critique. Crime, Media, Culture 4:175–195 HM Inspectorate of Prisons (2010) The management of gang issues among children and young people in prison custody and the community: a joint thematic review. HM Inspectorate of Prisons, London Hobbs D (2001) The firm: organizational logic and the criminal culture on a shifting terrain. Br J Criminol 4:549–560 Howell JC (2007) Menacing or mimicking? Realities of youth gangs. Juvenile Fam Court 58:39–48 Kemshall H, Mackenzie G, Wood J, Bailey R, Yates, J (2005) Strengthening multi-agency public protection arrangements (MAPPAs). Home Office Development and Practice Report. Home Office, London Klein M (2001) Resolving the Eurogang paradox. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the US and Europe. Kluwer Academic Publishers, London Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, New York Klein MW, Weerman F, Thornberry TP (2006) Street gang violence in Europe. Eur J Criminol 3:413–437 Office H (2004) Delinquent youth groups and offending behaviour: findings from the 2004 offending, crime and justice survey. HMSO, London Office H (2008) Tackling gangs: a practical guide for local authorities, CDRPs and other local partners. HMSO, London Ralphs R, Medina J, Aldridge J (2009) Who needs enemies with friends like these? The importance of place for young people living in known gang areas. J Youth Stud 12:483–500 Stelfox P (1998) Policing lower levels of organised crime in England and Wales. Howard J 37:393–406 The Centre for Social Justice (2009) Dying to belong: an in-depth review of street gangs in Britain. The Centre for Social Justice, London Weerman FM, Maxson CL, Esbensen F, Aldridge J, Medina J, Gemert F (2009) Eurogang program manual: background, development, and use of the instruments in multi-site, multi-method comparative research. Retrieved from http://www.umsl.edu/~ccj/eurogang/Eurogang_20Manual.pdf Youth Justice Board (2007) Groups, gangs and weapons. Youth Justice Board, London

Chapter 5

Five Decades of Defining Gangs in The Netherlands: The Eurogang Paradox in Practice Frank van Gemert

5.1

Introduction

Cities in The Netherlands have to deal with problematic youth groups. Sometimes, these groups are referred to as “jeugdbendes,” the Dutch equivalent of gangs, but mostly, social scientists and policy makers choose other wording. For example, in September 2005, a report on national television opened with the headline: “Rotterdam is struck by gangs.” Especially in certain quarters of Rotterdam-South, groups of youth committed serious acts of violence and sexual assault. The last remark of the report was that city council avoided speaking about “jeugdbendes,” because these groups were not like “the deadly gangs known from the American ghettos.” There appears to be confusion regarding what terminology is appropriate, and to settle the argument, American gangs are used as a point of reference. Reference to the American situation is by no means a solution to definition problems. This became clear in the mid-1990s when Dutch criminological research on organized crime intensified. Questions about the essence of organized crime were, either implicitly or explicitly, answered by pointing to the well-known American images. The Italian mafia of New York, as portrayed in the movie The Godfather by Francis Ford Coppola, was often used as an example (Bovenkerk and Husken 2005). It has now become clear that Don Corleone and his clan should not be used as a model for understanding the vast majority of organized crime. Different dynamic network structures are currently the starting point for studying organized crime (Fijnaut et al. 1996). Two perspectives on organized crime rely on different definitions. The mafia perspective is paired with a definition which emphasizes solid family ties and hierarchy. Conversely, the network perspective uses a definition with a focus on content and frequency of interactions. Both perspectives have consequences, not in the least for combating organized crime. The first will lead to a F. van Gemert (*) Criminology Department, VU University, Amsterdam, The Netherlands e-mail: [email protected] F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_5, © Springer Science+Business Media, LLC 2012

69

70

F. van Gemert

strategy of deep infiltration to attack the organization from within, hoping to blow it up at once; the second opens up possibilities to strike more frequently and at different locations in the network. Modern media, especially movies, have created stereotypes of organized crime. But this mechanism goes beyond organized crime. Starting with The Westside Story, numerous movies have also portrayed American gangs in specific ways that has led to stereotyping. Gangs from the west coast of the USA are often the source of these specific images. Still, stereotypical gangs such as Crips and Bloods have become the main model for what a gang is all over the world. So, when people in France, Russia, or The Netherlands talk about gangs, their point of reference is not as realistic as they may think. This is what Malcolm Klein has called “The Eurogang Paradox” (Klein 2001). A specific perspective and its associated definition have consequences for policies and strategies used to fight criminal phenomena. Gangs and organized crime both can be considered group crime, and in recent years, in The Netherlands, both have provoked considerable attention. While the picture for organized crime has become clearer, and criminologists and policy makers have more or less agreed on the perspective and the strategies concerning organized crime (Fijnaut 2010; KLPD 2008; Nelen and Huisman 2008), this remains unresolved for gangs. This chapter will explore the depictions and definitions of gangs that have been used over the years in The Netherlands. Of what elements are they comprised? How, and in what context, did they evolve? Is there currently a consensus on what gang definition to use? To what extent does the Eurogang Paradox play a role? The starting point will be gang depictions in Dutch publications from the past five decades. In this period, a wide number of social scientists and other professionals (e.g., lawyers, police, policy makers) have used depictions with several diverging elements.1 In the following sections, 16 of these are presented, systematically ordered, and analyzed.2 The last paragraphs will discuss the process of defining as it took place in The Netherlands and comment on a definition that is currently in use by the police.

5.1.1

Half a Century of Gang Definitions

Sixties. “Gangs, like in America, cannot be found in The Netherlands (…) in spite of sensational reports” (Cnoop Koopmans 1961, p. 279). This is the opening remark of an article in a juridical journal from 1961 that was a reaction to reports in the media. The author, a judge, describes a gang as “a pugnacious group of young persons, united in a formal relation, in which they seek status, safety that they cannot find in their environment. The gang is durable and exists longer than membership of individuals. The gang is a formal organization with a clear structure of rank and file, the use of symbols, and a territory that is defended for intruders of other gangs” (p. 279). Four years later, the same journal published an article, again instigated by journalistic reports. This time, the press mentions “organized youth gangs,” but the

5

Five Decades of Defining Gangs in The Netherlands…

71

author concludes that because of timely interference by the police, groups like these did not evolve. In this article, gangs are described as “two or more youth united in a group, more or less oriented towards a common goal, that is not legal. The gang has a hierarchical structure with a leader on top. It has a more or less pronounced division of tasks. The group has rules of conduct, offences of which will be sanctioned. There is a we-feeling, esprit de corps, solidarity. Information circulates through the group, members know what other members do, and they are about the same young age” (Piek 1965, p. 185). A few years later, in two subsequent volumes of a police journal, again a gang depiction is used (Van Helten 1970). This time, three groups of youth in Rotterdam have committed crimes and were stopped by the police. The author, a policeman, links the offenses of the groups to the emergence of a subculture and adds that this is not an abnormal phenomenon for youth coming of age. In these publications, the emphasis is on criminal behavior by youth groups and possible approaches for policing. These depictions are a reaction to media reports on gangs. They are comforting in the sense that they all refer to a serious American phenomenon that, according to the authors, does not exist in The Netherlands. This seems to have set the tone insofar as this line of reasoning returns over the years. Seventies. In this decade, for the first time, the link between gangs and ethnic minorities is discussed. Migration and integration emerge as themes that will become more important in the future. The specific migration history of Surinamese in The Netherlands explains why in this period these youth come to the forefront. Surinam becomes independent in 1975, and inhabitants of the former colony now must decide on their nationality. As a consequence, those that chose to be Dutch leave South America and move to Europe. A study of young Surinamese migrants in Amsterdam reveals 12–15 gangs in several neighborhoods in the period from 1975 to 1982 (Sansone 1992). Also in Haarlem, in the late 1970s, gangs composed mainly of Surinamese youth are reported (Van Niekerk et al. 1989). Neither of these publications present a definition. The first explicitly speaks about size of the groups, the age of members, leadership, and to the fact that these gangs are territorial, while in the second study, age, relation to the neighborhood, and territoriality are mentioned. Eighties. Up until the 1980s, the numbers of youth from ethnic minorities are low, and serious problems do not arise. But then things change, and Surinamese youngsters are the first to find themselves in the spotlights (Buiks 1983; Sansone 1992; Van Gelder and Sijtsma 1988a). Later, Moroccan boys take their place (Kaufman and Verbraeck 1986; Van Gelder and Sijtsma 1988b; Werdmölder 1986, 1990). Most of the research in this period is focused on drugs because addiction, especially to heroin, is a new and serious problem that involves the lives of many young migrants. In the big cities of Amsterdam and Rotterdam, black Surinamese youngsters, that introduce the South American street corner life in certain inner-city quarters, become the visual exterior of the drug trade on street level. In these quarters, the drug problem seems to have a color.

72

F. van Gemert

A second important migrant group that is mentioned concerning gangs is Moroccans. Their migration history is quite different from the Surinamese because there never were colonial ties with Morocco. Moroccan immigration was initiated with the active recruitment of guest workers by Dutch companies in the earlier decades. In times of a booming economy, employers needed manual labor power that was not to be found in The Netherlands. Guest workers came from all around the Mediterranean area, but when the economy shrank again, they were no longer needed. Those who stayed were the immigrants who found the Dutch welfare system to offer better living conditions than the mother country. These generally were Moroccans and Turks, who then reunited their families on Dutch soil. The sons of these first-generation guest workers find themselves on the streets in the 1980s. The content of an unofficial report from the City Council of Amsterdam is published by a newspaper. This brief report is mainly based on impressions and estimations by the police and clearly is not an academic source, but it contains an alarming depiction of a Moroccan gang. Leadership and a hierarchical structure are mentioned, as is the number of 200–300 members. This group is said to be responsible for various criminal activities in the center of town (Loef 1988). Next, an ethnographic study of a Moroccan gang in the capital provides detail about such groups (Werdmölder 1986, 1990). The group studied is described as a “gang,” and the author points to leadership, the meaning of criminal activities and conflicts with citizens from the same neighborhood that strengthen the group ties (Werdmölder 1986, p. 34). Nineties. In this decade, the number of publications that address the problem of gangs grows. Antillean youth are the next ethnic minority to receive attention (de Jong et al. 1997; Van Hulst and Bos 1993; Van San 1998). At the same time, a new aspect is highlighted. Youth groups that copy gangs like Crips and Bloods from the American West Coast are found in Dutch big cities, especially in The Hague. International hiphop youth culture plays an important role in the distribution of the “gangsta” style. Because of the fact that the new role models mainly are African-Americans, in the beginning, especially black Surinamese and Antillean youth seem to be attracted. This link to American gangs draws attention, and several studies are conducted. In archival research on three Crips gangs (Van Gemert 1998a), a definition by Klein and Maxson (1989) is applied that focuses on three elements “…community recognition as a group or collectivity, recognition by the group itself as a distinct group of adolescents and young adults, and enough illegal activities to get a consistent negative response from law enforcement and/or neighborhood residents” (Klein and Maxson 1989, p. 205; see also Van Gemert 2001). A second publication derives from research on Antillean youth (de Jong et al. 1997). Here too, groups of Crips and Bloods are reported, but whether these are gangs remains unclear. The gang depiction includes a high level of organization, leadership, and a variety of roles. Starting in the mid-1990s, the police in the region of The Hague show particular interest in youth groups. Several years in a row, they make an inventory of youth groups in the area by using a survey; policemen on the beat are respondents. In this

5


73

research, for the first time, a distinction is made between groups that are troublesome, groups that cause nuisance, and criminal groups (Van Oosterwijk et al. 1995). This distinction will surface again. “Youth gangs” are seen as groups that use Americanstyle elements, and they can be found in all three categories (Van Oosterwijk et al. 1995, p. 44). A small number of the groups are labeled “jeugdbende,” the common equivalent of gang. Over the years, however, the definitions of “jeugdbende” (gang) used by the Hague police change gradually. At first, a “jeugdbende” is defined as “a group of three or more youth, that are connected because of race, country of origin, culture or territory, that meet regularly and (among other things) have the goal to commit criminal activities” (Van Oosterwijk et al. 1995, p. 44). Here, shared identity and criminal activities are key elements, and a link to migration is implicit. The next year, we see a change: “…. that meet regularly and primarily have the goal to commit criminal activities in order to have financial gain. Compared to other forms of youth groups, the gang has a relative high level of organization” (Gruter et al. 1996, p. 7). Yet another year later, new elements are added. “… has a relative high level of organization. Even though troublesome and nuisance behavior can also occur in these groups, the activities of these groups concentrate on property offences. Apart from that, organized drug trade and instigation to prostitution occur” (Gruter 1997, p. 20; Solm and Rotteveel 2000, p. 2). Along with Antillean youth, Moroccan boys remain in the gang picture. In 1998, there are “riots” in Amsterdam-West, a site that continues to be known for the problems of Moroccan youth (see de Jong 2012). In an ethnographic study of Moroccan youth in Rotterdam, the Klein and Maxson (2006) definition that stresses group ties, criminal actives, and subsequent negative response is used, and no gangs are found (Van Gemert 1998b). Three other studies with a more general view on youth crime use different definitions. Angenent (1997) describes gangs as hierarchical groups committing serious crimes, with a leader, stable positions and rituals, a strict organization, and a territory. Bruinsma discusses the link between interpersonal relations and criminal activities. In his view, Dutch gangs can be found in the big cities, but they are smaller than in the USA. According to him, gangs are characterized by organized behavior. They have a hierarchical structure, with set rules and codes. New members go through initiation rites during which they have to prove to be capable of doing criminal acts (Bruinsma 1991). Finally, a study by Terpstra (1997) describes the emergence of a collective lifestyle of youth in a deprived city. This group should not be considered a gang, because gangs are defined as tight groups with clear leadership and a division of tasks to accomplish collective, planned illegal actions (Terpstra 1997). 2000 and later. In the new millennium, the seriousness of problems with youth groups seems to have grown. In particular, nuisance by groups that hang around on the streets is perceived as an important cause for citizens to feel unsafe. This has been registered not just in the big cities, Amsterdam (Hoenson 2000) and Rotterdam (SAMS 2004), but throughout the country. The case of the Diamantbuurt, a neighborhood in

74

F. van Gemert

Amsterdam where a couple left their home because of persistent conflicts with Moroccan boys, gets national exposure and becomes a cause célèbre. Tapping into current fears, Geert Wilders, the leader of new right-wing populist party PVV, has labeled these youngsters “street terrorists.” The Ministry of Justice ordered research to develop an instrument to take inventories of youth groups. This instrument is, to a certain extent, a newer version of the instrument that was used earlier by the police in The Hague. It is a survey among policemen who regularly work in a neighborhood; it can be compared to the Eurogang Expert Survey. Data are gathered on youth groups, on risk factors, and on criminal activities of its members (Beke et al. 2000; Ferwerda and Kloosterman 2004). One of the first pilots is, again, The Hague (Van Wijk et al. 2000). Like before, the distinction between troublesome, nuisance and criminal groups is used in the new survey. “Straatbendes en jeugdbendes” (possible translations: street gangs and youth gangs) are so called “plus variants” from respectively nuisance groups and criminal groups (Beke et al. 2000, pp. 132–3; Ferwerda and Kloosterman 2004, p. 15). Criteria for a plus variant are two dichotomies (1) tight, organized versus loose, unorganized and (2) open, not hierarchical versus closed, hierarchical. Questions on a number of characteristics operationalize these concepts.3 If the majority of the answers on both dichotomies are positive, the group is considered a plus variant, e.g., a straatbende or a jeugdbende. This means that a gang (jeugdbende) is defined as “a criminal youth group that does not operate in public, but goes underground and ‘starts to belong to organized crime.” (Beke et al. 2000, p. 133). In the city of Amsterdam, two studies are conducted using the Eurogang definition: A gang is any durable street-oriented youth group whose involvement in illegal activity is part of their group identity. The first is an ethnographic study of a Moroccan gang in Amsterdam-West (Van Gemert and Fleisher 2002, 2005). Here, the Eurogang Ethnography Guidelines are used (Weerman et al. 2009). The second study (Van Gemert 2005) is an inventory of youth groups and gangs in the capital using the Eurogang Expert Survey (Weerman et al. 2009) on police officers.

5.1.2

Systematic Overview

Gangs are groups, but not all groups are gangs. The depictions and definitions described above differ. In the table below, the elements from all 16 depictions and definitions are presented. Distinguished are elements on age (members have to be youth), size (the minimum number of members), durability (the minimum time that the group has to exist), and use of symbols (the group can be recognized by external attributes). Other elements are subdivided: location (where the group can be found and what is the relation to public space), structure (is the group organized), criminal activities (what kind, mode, motivation), and the relation between individual and group (does the group have a function for the individual). Age, size, duration, and symbols. Table 5.1 shows that all depictions and definitions include the element of age. All speak about youngsters or youth, but as a rule, specific

5


75

age is not mentioned. Less than half (seven) say something about the size of the group; usually the minimum is three persons. In the other nine depictions, the use of the word “group” seems self-evident. Duration or the use of symbols are seldom included, respectively twice and once. Even though the use of gangsta style elements has made groups easy to recognize and drew attention especially in The Hague, this is not considered a vital element of gangs. Location. Location is a gang element in 10 of the 16 depictions and definitions. Location is of importance to the group because of the meaning it derives from territory (four times), provider of identity (three times), or reason for conflict (twice). Structure. Twelve of the 16 depictions and definitions refer to structure, but they mention different subelements. Leadership is considered the most relevant (in nine depictions and definitions); high level of organization, hierarchy and tasks, functions and rules are seen as less important (respectively six, five, four times). The element of initiation is mentioned in only two depictions and definitions. Bruinsma (1991) and Angenent (1997), in particular, emphasize structure; they each use five subelements. Also Beke et al. (2000) point to these elements (see note 3). Crime. The element of crime is also present in 75% of the depictions and definitions. Most authors agree that without crime, there is no gang. Again we see variation in subelements. Five times the group is said to have the goal of committing crime, and four times authors point to the meaning of crime for the group. Group and individual. The relation between group and individual is mentioned in seven of the depictions and definitions. The group can provide safety, status, or ways to spend free time. It can also provide cohesion via a feeling of us and them or because of ethnicity. Over the years, the elements of age, structure, and crime are presented as key criteria for determining what a gang in Dutch publications is. Most authors considered other elements to be less important. While one can easily agree on the elements of age and crime, structure is a different matter. We will return to this issue later.

5.1.3

Depictions Becoming Definitions

In The Netherlands, the 1960s is a postwar decade that stands for rebuilding, optimism, and economic growth. At that time, Dutch society is segmented because by birth, people belong to one of four “pillars”: social democrats, liberals, Catholics, and Protestants (Buruma 2007). Most people are religious and go to church every week, and in general, there is strong social control. In the 1960s, this starts to change, and the 1970s really bring new élan. Youth break free and start experimenting with sex and drugs. New ideas are explored, and institutions in health care and justice seek and apply new free forms. The country rapidly becomes secularized. Youth that draw the most attention because of their activist ideas were called “Provos” (Buikhuisen 1966). They have a middle-class background, and they often are students. Compared

F. van Gemert

x x x

x x

Age of members Size Duration Symbols

Depictions/definitions by author

Elements and subelements

x

x x x x x x x x x x x

x

x x

x

Territory Orientation to street Conflict Identity Leader High-level organization Hierarchy Tasks/functions/rules Initiation

Cnoop Koopmans 61

x

x x

Piek 65

x x x

x x

Sansone 92

x

x

Van Niekerk 89

x

x x

Loef 88

x

x

Werdmölder 90

x

x

Angenent 91

x x x x x

x

Bruinsma 91

x x

Van Oosterwijk et al. 95

x

x x

Gruter et al. 96

Table 5.1 Elements and subelements of gang depictionsa

x x x

Gruter 97

x

Van Gemert 98a 98b

x

de Jong, Steijlen, Masson 97

x

x

Terpstra 97

x

Beke Ferwerda 00 04

x

Van Gemert Fleisher 02 05

x

Numbers

x x x

16 7 2 1 4 1 10 2 3 9 5 6 26 4 2

76

Location

Structure

Crime x x

x

x

x

x

x

The gray areas mark the depictions/definitions that have none of the items per element

Group/individual

a

Sort Amount Meaning Goal Co-offense Financial gain Norms values Underground Planned Ethnicity Us vs. them Status Safety Pastime x

x x x x

x

x x

x

x

x x

x

x

x

x

x

x

x

x

x

x

x

x x

3 2 4 5 1 2 1 1 1 2 5 1 1 3 12

20

5 Five Decades of Defining Gangs in The Netherlands… 77

78

F. van Gemert

to young people in this somewhat revolutionary movement, working-class youth groups active in crime remain in the background (e.g., Kok 2011). None of the publications mentioned above from the 1960s to the 1970s are the result of studies on gangs. In the beginning, these publications are written by professionals that comment on gangs being reported in the media, while in the 1970s, these texts start focusing on new migrant groups, and researchers stumble on groups that could be gangs. During the 1980s and 1990s, new migrant youth get noticed. Surinamese, Moroccan, and Antillean each seem to take their turn under the microscope of youth crime. Until then, youth groups or gangs never were on the research agenda, and the use of the term gang (“jeugdbende”) is criticized. For example, the unofficial report from City Council of Amsterdam (Loef 1988) calls for sharp reactions. It is not based on solid methodology, but it still plays an important role in a very sensitive discussion (Bovenkerk 1989; Werdmölder 1989). Later Werdmölder is criticized because in his dissertation, he describes a Moroccan group as a gang (Werdmölder 1990). Critics say that by referring to American literature, Werdmölder creates a false image of the group. De Haan (1991) points out that the characteristic structure with a leader is lacking, while Van Gelder (1992) states that Moroccan members normally have no strict ties and do not form formal groups. These criticisms come from academics, and they agree that the gang label does not fit the groups presented in Dutch publications. The terminology is rejected, because structural elements are said not to apply to these groups. At first, problems with migrant youth were associated with drugs. Especially Surinamese youth were unlucky to find their migration history coincide with the expansion of heroin use in The Netherlands. Political correctness seems to have played a role because the gang label can easily lead to making minority groups a scapegoat, and blaming the victim. There were youth groups that may have been gangs, but well into the 1980s, in the general perception, this was not an issue. There was no “gang problem.” In the 1990s, we see a change, probably due to a new youth culture that brings the gangsta style and adds visibility to the phenomenon. Researchers start focusing on these groups, and it is not a coincidence that the police in the Hague initiated research because in this city, we see the first Dutch groups that call themselves “Crips” (Van Gemert 1998a, 2001; Van Oosterwijk et al. 1995; Van Stapele 2003). Through films and hip-hop music, American gang members have become role models, and throughout the world, their music and appearance are copied. Even though in America, police gang squads seriously study gang symbols to keep track of gang activity, in The Netherlands, keeping an eye on baggy trousers, colored bandanas, or hand signs would be ineffective. The gang image is so popular that it is impossible to determine if young people that look “gangsta” really are gang members (Van Gemert 2002). Because of this possible confusion, in their first inventory, the police in The Hague decide to use the word “jeugdbende” only if certain criteria are met (see above) and speak of “gang” when pointing to groups that use the popular gang symbols (Van Oosterwijk et al. 1995). Even though the gangsta style clearly worked as a catalyst, most youth groups from recent years differ from US gangs in their reason for being. Thrasher (1927)

5


79

has defined gangs as “groups in conflict,” and most American gangs are involved in conflicts with other gangs because of territory, drug money, girls, etc. In The Netherlands, youth groups seldom fight each other; much more often, these groups have conflicts within their own neighborhoods. Their opponents are residents, shopkeepers, or the police (Van Gemert and Stuifbergen 2008). Even though a number of these groups have criminal activities, the vast majority attract attention because of less serious offenses and nuisance. Still, these conflicts can have very negative impacts because they make people feel unsafe in their own neighborhoods. These conflicts are reported repeatedly in the media, and over the years, especially Moroccan groups become known for their troublesome behavior (Bervoets 2006; de Jong 2007; see also de Jong 2012; Van Gemert 1998b; Van Gemert and Fleisher 2005). The question remains: Are these groups gangs?

5.1.4

Confusing Consensus

In the 1990s, youth groups become the subject of more research. Depictions of gangs that derive from American examples or even American literature are replaced by new definitions formulated to fit the Dutch situation. Dutch youth groups become better documented. The Ministry of Justice funds research to construct an instrument for inventories of youth groups. The result is a survey known as “Shortlistmethod,” that draws the distinction between groups as troublesome, cause nuisance, and criminal groups (Beke et al. 2000). The experiences of The Hague police are the first source of data, and maybe because of this, the use of the term “jeugdbende” is problematic. As was described above, “jeugdbende” can only be used when strict criteria for structure elements are met, resulting in the identification of few gangs. This instrument is used to study groups, but “jeugdbende” is dealt with as a very specific category. The consequence of this definition surfaces when results from the Shortlistmethod are compared with those from the Eurogang Expert Survey. Both surveys use the same respondents: policemen on the beat. The City Council of Amsterdam published numbers of problematic youth groups using the former instrument, and not even once were the terms “jeugdbende” or “gang” used (Gemeente Amsterdam 2005). The Eurogang instrument (Van Gemert 2005) identified a smaller number of youth groups (85 compared to 118), but 39 of these were classified as gangs. Data for the Eurogang study were gathered a year earlier, but a striking difference like this (0–39) can only be understood by taking into account the difference in definition. The Shortlist-method was tested and later became obligatory in all police regions throughout the country. At first, this was for local use only. The fact that official police reports seldom or never mention “jeugdbende” had an interesting boomerang effect. RTL, a broadcast company, became interested in the news on youth groups, and since the data are public, they began doing their own analyses. Because there were no national figures, they gathered these data from the various police regions and then published the total national numbers of youth

80

F. van Gemert

groups. In 2008, there were 1,192 troublesome groups, 117 nuisance groups, and 70 criminal groups or a total of 1,379 youth groups. In 2009, the numbers were respectively 1,341, 327, 92, or a total number of 1,760 youth groups (RTL Nieuws 2010a). The wording “youth groups” is very neutral, and one cannot understand why the police would make an inventory of groups if they were normal. Maybe, for this reason, RTL decided to use the word that they expected to find: “jeugdbende.” They simply replaced “youth groups,” and then there were screaming headlines: “Almost 1800 jeugdbendes in The Netherlands,” and “92 criminal jeugdbendes” (RTL Nieuws 2010b). All of a sudden, there was a lot of explaining to do. From then on, national data are reported and published (Ferwerda and Van Ham 2010). But still, from a total number of 1,154 youth groups in 2010 in The Netherlands, only 6 “jeugdbendes” are reported. According to the Shortlist-method, the major cities of Amsterdam, Rotterdam, Utrecht, and the Hague have no “jeugdbendes,” no gangs. They do have respectively 7, 3, 19, and 11 criminal youth groups (Ferwerda and Van Ham 2010). “Definitions determine whether we have a large, small, or even no problem, whether more or fewer gangs and gang members exist, and which agencies will receive funds” (Spergel 1990, p. 177). Policy makers in The Netherlands are aware of this mechanism, and via the Shortlist-method, they are provided with a suitable instrument. So yes, for the police, there is consensus about a gang definition, but at the same time, it becomes clear that gang terminology is deliberately being avoided (Van Gemert 2007). By looking at depictions and definitions that were used from the 1960s until now, we see constant reference to American stereotypes. Stressing the structural elements leads to dismissing gang terminology as nonapplicable. This is the Eurogang Paradox in practice. By getting rid of the “jeugdbende” or “gang” label, policy makers have created more room to maneuver for themselves. Thus, they avoid heated discussions that set the agenda. Since this was done in a rather strict manner, the way is open for others to reintroduce the label.

Notes 1. The fact that many different depictions and definitions exist is not so unusual. In the United States, social scientists, police, and policy makers also cannot agree on a definition for gang (e.g., Ball and Curry 1995; Barrows and Huff 2009; Horrowitz 1990; Spergel 1990). 2. This overview is incomplete, as it derives principally from publications in academic journals. It does not include articles in newspapers and popular journals, and a large number of theses by students. Publications by Schuyt (1993), Hakkert et al. (1998), and Weerman (2000) do not present a definition of their own, and therefore, they are not discussed here. 3. Five characteristics of the first (tight, organized versus loose, unorganized) are tightness of the group structure, the measure of mutual solidarity, the measure of organization, the measure in which criminal actions are planned, and the measure in which changes in the group composition take place. Three characteristics of the second dichotomy (open, not hierarchical versus closed, hierarchical) are the measure of hierarchy, the measure of mutual rivalry, and the measure in which the group acts as a group (Beke et al. 2000).

5


81

References Amsterdam (2005) Aanpak overlast jeugdgroepen Amsterdam; Stand van zaken januari - juni 2005. Gemeente Amsterdam, Amsterdam Angenent H (1991) Criminaliteit van allochtone jongeren: Feiten, oorzaken en achtergronden. Intro BV, Baarn Ball RA, Curry GD (1995) The logic of definition in criminology: purposes and methods for defining ‘gangs. Criminology 33:225–245 Barrows J, Huff CR (2009) Gang and public policy: constructing and deconstructing gang databases. Criminol Public Policy 8:675–703 Beke BMWA, van Wijk AP, Ferwerda HB (2000) Jeugdcriminaliteit in groepsverband ontrafeld; Tussen rondhangen en bendevorming. SWP, Amsterdam Bervoets E (2006) Tussen Respect en Doorpakken: Een Onderzoek naar de Politiële Aanpak van Marokkaanse Jongeren in Gouda, Utrecht en Amsterdam. Elsevier, Den Haag Bovenkerk F (1989) Marokkanen, misdaad en minderhedenbeleid. In: Bovenkerk F, Brunt L (eds) De andere stad. Achter de façade van de nieuwe stedelijke vitaliteit. Centrum voor Grootstedelijk Onderzoek, Amsterdam, pp 11–21 Bovenkerk F, Husken M (2005) De culturele erfenis van Don Corleone. Justitiële Verkenningen 31:11–24 Bruinsma GJN (1991) Leeftijdgenoten en de criminaliteit van jongeren. Jeugd en Samenleving 20:15–25 Buikhuisen W (1966) Achtergronden van nozem gedrag. Van Gorcum, Assen Buiks PEJ (1983) Surinaamse jongeren op de Kruiskade; Overleven in een etnische randgroep. Van Loghum Slaterus, Deventer Buruma Y (2007) Dutch tolerance: on drugs, prostitution and euthanasia. In: Tonry M, Bijleveld C (eds) Crime and justice in The Netherlands. University of Chicago Press, Chicago, pp 73–114 Cnoop Koopmans AJ (1961) Gangvorming. Maandblad voor Berechting en Reclassering 40:279–283 De Haan W (1991) Boekbesprekingen dissertaties Werdmölder (1990) en Junger (1990). Migrantenstudies 7:39–45 de Jong JDA (2007) Kapot Moeilijk: Een Etnografisch Onderzoek naar Opvallend Delinquent Groepsgedrag van ‘Marokkaanse’ Jongens. Aksant, Amsterdam de Jong JDA (2012) Typically Moroccan? A group dynamic explanation of nuisance and criminal behavior. In: Esbensen FA, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York de Jong W, Steijlen F, Masson K (1997) Hoe doe je je ding; Antilliaanse jongeren en criminaliteit in de politieregio Rotterdam-Rijnmond. Eburon, Delft Ferwerda H, Kloosterman A (2004) Jeugdgroepen in beeld; stappenplan en randvoorwaarden voor de shortlistmethodiek. Politie en Wetenschap, Apeldoorn Ferwerda H, van Ham T (2010) Problematische Jeugdgroepen in Nederland; Omvang en aard in het najaar van 2010. Bureau Beke, Arnhem. Retrieved from http://www.beke.nl/doc/2010/ Prob_jeugdgoep_najaar_2010.pdf Fijnaut C (2010) Introduction of the New York double strategy to control organized crime in The Netherlands and the European Union. Eur J Crime, Crim Law Crim Justice 18:43–46 Fijnaut C, Bovenkerk F, Bruinsma G, van de Bunt H (1996) Georganiseerde criminaliteit in Nederland, eindrapport, Bijlage VII van Enquêtecommissie Opsporingsmethoden, Inzake Opsporing. SDU Uitgevers, Den Haag Gruter P (1997) Problematische jeugdgroepen in de regio Haaglanden anno 1997: Een hernieuwde inventarisatie onder wijkagenten en jeugdrechercheurs. Politie Haaglanden, Den Haag Gruter P, Baas M, Vegter D (1996) Problematische jeugdgroepen in de regio Haaglanden. Een inventarisatie onder wijkagenten en jeugdrechercheurs. Politie Haaglanden, Den Haag

82

F. van Gemert

Hakkert A, van Wijk A, Ferwerda H, Eijken T (1998) Groepscriminaliteit; Een terreinverkenning op basis van literatuuronderzoek en een analyse van bestaand onderzoeksmateriaal, aangevuld met enkele interviews met sleutelinformanten en jongeren die tot groepen behoren. Ministerie van Justitie (DPJS), Den Haag Hoenson L (2000) Bevolkingsonderzoek in wijkteamgebieden 2000. Politie AmsterdamAmstelland, Amsterdam Horrowitz R (1990) Sociological perspectives on gangs: conflicting definitions and concepts. In: Huff CR (ed) Gangs in America. Sage, Newbury Park, pp 37–54 Kaufman P, Verbraeck H (1986) Marokkaan en verslaafd. Een studie naar randgroepvorming, heroïnegebruik en criminalisering. Gemeente Utrecht, Utrecht Klein MW (2001) Resolving the Eurogang paradox. In: Klein MW, Kerner HJ, Maxson CL, Weitek EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Academic, Dordrecht, Boston, London, pp 7–19 Klein, MW and Maxson CL (2006) Street gang patterns and policies. New York: Oxford University Press. Klein MW, Maxson CL (1989) Street gang violence. In: Weiner NA, Wolfgang ME (eds) Violent crime, violent criminals. Sage, Newbury Park KLPD (2008) Nationaal Dreigingsbeeld 2008, Georganiseerde Criminaliteit. Korps landelijke politiediensten, Zoetermeer Kok A (2011) Holleeder, de jonge jaren. de Bezige Bij, Amsterdam Loef C (1988) Marokkaanse daders in de Amsterdamse binnenstad. Rapport bestuursinformatie Amsterdam. Gemeente Amsterdam, Amsterdam Nelen H, Huisman W (2008) Breaking the power of organized crime? The administrative approach to organized crime in Amsterdam. In: Siegel D, Nelen H (eds) Organized crime: culture, markets and policies. Springer, New York, pp 207–218 Niekerk M, Sunier T, Vermeulen H (1989) Bekende vreemden; Surinamers, Turken en Nederlanders in een naoorlogse wijk. Het Spinhuis, Amsterdam RTL Nieuws (2010a) De jeugdbendes per regio. Retrieved from http://www.rtl.nl/(/actueel/rtlnieuws/binnenland/)/components/actueel/rtlnieuws/2010/01_januari/06/verrijkingsonderdelen/jeugdbendes_cijfers.xml RTL Nieuws (2010b) 92 criminele jeugdebendes in Nederland. Retrieved from (http://www.rtl.nl/ (/actueel/rtlnieuws/binnenland/)/components/actueel/rtlnieuws/2010/01_januari/06/binnenland/92_criminele_jeugdbendes_in_Nederland.xml Piek NJM (1965) Over jeugdbendes gesproken. Maandblad voor Berechting en Reclassering 44:185–92 SAMS (2004) Minder bevreesd; Aandacht voor gevoelens van onveiligheid in de stad. Stedelijke Adviescommissie Multiculturele Stad, Rotterdam Sansone L (1992) Schitteren in de schaduw. Overlevingsstrategieën, subcultuur en etniciteit van Creoolse jongeren uit de lage klassen 1981–1990. Het Spinhuis, Amsterdam Schuyt CMJ (1993) Jeugdcriminaliteit in groepsverband. Delikt en Delinkwent 23:499–510 Solm A, Rotteveel S (2000) Problematische Marokkaanse jeugdgroepen; Bureau De Heemstraat. Regiopolitie Haaglanden, Den Haag Spergel IA (1990) Youth gangs. Continuity and change. In: Tonry M, Morris N (eds) Crime and justice. University of Chicago Press, Chicago, pp 171–275 Terpstra J (1997) Jeugdsubcultuur en de reproductie van maatschappelijke achterstand. Sociologische Gids 44:205–229 Thrasher FM (1927) The gang: 1,303 gangs in Chicago. University of Chicago Press, Chicago van Gelder P (1992) Het groepsgedrag van Marokkaanse jongens. Tussen mythe en werkelijkheid. Paper Sociaal-wetenschappelijke Studiedagen, Amsterdam, 28 en 29 April van Gelder P, Sijtsma J (1988a) Horse, coke en kansen; Sociale risico’s en kansen onder Surinaamse hardruggebruikers in Amsterdam. Inst. voor Sociale Geografie, Universiteit van Amsterdam, Amsterdam

5


83

van Gelder P, Sijtsma J (1988b) Horse, coke en kansen; Sociale risico’s en kansen onder Marokkaanse hardruggebruikers in Amsterdam. Inst. voor Sociale Geografie, Universiteit van Amsterdam, Amsterdam Van Gemert F (1998a) Crips in drievoud; Een dossieronderzoek naar drie jeugdbendes. Regioplan, Amsterdam Van Gemert F (1998b) Ieder voor zich; Kansen, cultuur en criminaliteit van Marokkaanse jongens. Het Spinhuis, Amsterdam Van Gemert F (2001) Crips in orange: gangs and groups in The Netherlands. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Boston/The Hague, pp 145–152 Van Gemert F (2002) Botsen met de buurt; Overlast en de wisselwerking tussen jeugdgroepen en de buitenwereld. Tijdschrift voor Criminologie 44:162–171 Van Gemert F (2005) Youth groups and gangs in Amsterdam: an inventory based on the Eurogang Expert Survey. In: Decker SH, Weerman F (eds) European street gangs and troublesome youth groups: findings from the Eurogang research program. Altamira Press, Walnut Creek, pp 147–169 Van Gemert F (2007) Gangs: system or madness? In: van Ginkel R, Strating A (eds) Wildness and sensation: an anthropology of sinister and sensuous realms. Spinhuis, Amsterdam, pp 141–156 Van Gemert F, Fleisher M (2002) In de greep van de groep; Een onderzoek naar een Marokkaanse problematische jeugdgroep. Regioplan, Amsterdam Van Gemert F, Fleisher M (2005) In the grip of the group: ethnography of a Moroccan street gang in The Netherlands. In: Decker SH, Weerman F (eds) European street gangs and troublesome youth groups: findings from the Eurogang research program. Altamira Press, Walnut Creek, pp 11–30 Van Gemert F, Stuifbergen J (2008) Gangs, migration and conflict: Thrasher’s theme in The Netherlands. In: van Gemert F, Peterson D, Lien IL (eds) Street gangs, migration and ethnicity. Willan, Cullompton, pp 79–96 van Helten N (1970) Bendecriminaliteit onder de Nederlandse jeugd (I en II) Tijdschrift voor Politie, 6/7, 181–193: 239–249 van Hulst H, Bos JA (1993) Pan i rèspèt. Criminaliteit van geïmmigreerde Curaçaose jongeren. OKU, Utrecht van Oosterwijk C, Gruter P, Versteegh P (1995) Haagse jeugdbendes en Amerikaanse gangs. Een vergelijkend onderzoek naar jeugdbende-criminaliteit in de regio Haaglanden. Politie Haaglanden, Den Haag van San M (1998) Stelen en steken; Delinquent gedrag van Curaçaose jongens in Nederland. Het Spinhuis, Amsterdam van Stapele S (2003) Crips.nl; 15 jaar gangcultuur in Nederland. Vassalucci, Amsterdam van Wijk AP, Beke BMWA, Versteegh PHM, van Solm AIT (2000) Inventarisatie van problematische jeugdgroepen in Haaglanden III. Politie Haaglanden, Den Haag Weerman F (2000) Samenplegen; Over criminele samenwerking en groepsvorming. Ars Aequi, Nijmegen Weerman FM, Maxson CL, Esbensen FA, Aldridge J, Medina J, van Gemert F (2009) Eurogang program manual; Background, development, and use of the Eurogang instruments in multi-site, multimethod comparative research. http://www.umsl.edu/~ccj/eurogang/Eurogang_20Manual.pdf Werdmölder H (1986) Van vriendenkring tot randgroep. Marokkaanse jongeren in een oude stadswijk. Het Wereldvenster, Houten Werdmölder H (1989) Een taboe doorbroken; Marokkaanse jongeren en criminaliteit. Intermediair 25:17–23 Werdmölder H (1990) Een generatie op drift. De geschiedenis van een Marokkaanse randgroep. Gouda Quint, Arnhem

Chapter 6

Gang Organization, Offending, and Victimization: A Cross-National Analysis David C. Pyrooz, Andrew M. Fox, Charles M. Katz, and Scott H. Decker

Identifying the influence of groups on human behavior is a central task of social scientists. For decades, researchers have sought to determine how groups affect individual behavior in various contexts, such as protest, riot, and crowd settings (Bohstedt 1994; McPhail 1991); hate, religious, and terrorist groups (Bjorgo 2005; Louis and Taylor 2002; Staub 2002); and—the focus of this study—gangs (Klein 1971, 2006; Short and Strodtbeck 1965; Thrasher 1927). Group processes unique to gangs have been touted as the mechanism rendering gangs “qualitatively different” from other criminal and delinquent groups (Klein 2006; Klein and Maxson 2006; see also Decker and Pyrooz 2011). The past two decades of empirical research have demonstrated that gang joining corresponds with an escalation of delinquent behavior for individuals (Krohn and Thornberry 2008). Less emphasis, however, has been placed on how this effect manifests conceptually and whether this effect is uniform across gangs and gang members. Part of the problem is that scholars have devoted less attention to examining characteristics of gangs as groups (Maxson and Klein 1995; McGloin and Decker 2010), opting instead to focus on characteristics unique to individuals. A critical aspect of any group, whether formal or informal, conforming or deviant, is organization—the degree to which a group effectively and efficiently coordinates and carries out activities. Gang organization has received considerable attention in the context of drug selling. Here, researchers investigated and disputed the extent to which gangs dominated the rapidly developing and increasingly violent drug markets of the 1980s and 1990s (Decker and Van Winkle 1995; Decker et al. 1998; Hagedorn 1994). Because gang organizational structure influences the function and processes of the group, there is reason to believe that this influence should be observed across gang members as well. A handful of studies have demonstrated this influence, finding that greater gang organization is associated with increased D.C. Pyrooz (*) • A.M. Fox • C.M. Katz • S.H. Decker School of Criminology and Criminal Justice, Arizona State University, Phoenix, AZ, USA e-mail: [email protected] F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_6, © Springer Science+Business Media, LLC 2012

85

86

D.C. Pyrooz et al.

delinquency and victimization of gang members (Bjerregaard 2002; Bouchard and Spindler 2010; Decker et al. 2008; Esbensen et al. 2001; Sheley et al. 1995). But as Decker et al. (2008, p. 167) observed, “gang research has provided more descriptive literature than analyses of the relationships between gang characteristics and behavior, a notable omission.” That said, it is necessary to further understand the relationship that distinguishes gangs from other types of delinquent groups. The present study extends the modest literature on gangs, organizational structure, and criminological outcomes. Using data gathered from three research projects—a juvenile arrestee sample from Arizona (ADAM), an urban school sample in Trinidad and Tobago (TTYS), and a multisite school sample in the USA. (GREAT)—this study examines the association between gang organizational structure and delinquency and victimization. The similar nature of the survey items permits an assessment of the aggregated construct and disaggregated indicators of gang organization. The key advantage in this study is that gang organizational structure can be examined comparatively across the three data sources that differ demographically and culturally, which is central to the objectives of the Eurogang research program (Klein 2005; Weerman et al. 2009). If gang organizational structure “matters” for explaining offending and victimization patterns of gang members, it should across all three research contexts.

6.1

Gangs, Groups, and Organization

Street gangs are street-oriented groups, generally comprised of youth, exhibiting persistence across time, and illegal activity constitutes a part of group identity (Klein and Maxson 2006). Thrasher’s (1927, p. 5) statement that “no two gangs are alike” elicits the variability and complexity of gangs. Organization is a key feature that differentiates gangs from one another. Gang organization is the degree to which the group effectively and efficiently coordinates and carries out activities. There are various dimensions of organization, one of which is group structure (e.g., hierarchy, composition). The organizational structure of gangs matters to the extent that it influences the social processes (e.g., contagion, collective action) responsible for making gangs qualitatively distinct from other types of criminal and delinquent groups. The degree to which gangs vary in their organizational capacities is disputed in the gang literature. It has been argued that gangs exhibit the characteristics of formal organizations, characteristics that are important for the efficient distribution of drugs, among other things. It has also been argued that despite heavy involvement in drug sales, gangs are not very well organized and thus not efficient structures for large-scale enterprises. The former view describes gangs as instrumental-rational (organized) and the latter as informal-diffuse (disorganized). These perspectives developed originally out of the debate surrounding the extent to which gangs controlled increasingly violent, rapidly expanding drug markets of the late 1980s/ early 1990s. For example, Hagedorn (1994) framed the argument around whether gangs were organized drug distributors or whether gang members were “freelance”

6

Gang Organization, Offending, and Victimization…

87

drug dealers. The research has since grown to examine areas such as the penetration of gangs into community organization (Venkatesh 1997) and the ability of gangs to organize homicide (Decker and Curry 2002). The instrumental-rational perspective, as described by Decker and Curry (2000, p. 474), holds that gangs “have a vertical structure, enforce discipline among their members, and are quite successful in defining and achieving group values.” Additional indications of an instrumental-rational gang include levels of membership, leadership roles, regularly attended meetings, coordinated drug sales, written rules and codes of conduct, expansion in legitimate business operations, and political influence (Decker et al. 1998). Examples of these descriptions can be found in the research of Decker et al. (1998), Mieczkowski (1986), Padilla (1992), Skolnick et al. (1990), Sanchez-Jankowski (1991), Taylor (1990), Venkatesh (1997), and Venkatesh and Levitt (2000). A stark example of an instrumental-rational gang can be found in research on a Chicago public housing development, where Venkatesh portrayed the “Black Kings” gang as possessing an inordinate degree of power to influence community affairs. The gang was able to negotiate with the neighborhood “council” (nominated leaders of buildings) to act as security for the housing projects, thereby bolstering drug distribution operations. In addition, the gang was organized such that each “constituent [gang] set was tied to the overall organization through trademark and fiduciary responsibilities” (Venkatesh and Levitt 2000, p. 428). Examples such as this support claims of “corporatization” (Taylor 1990)—that is, gangs oriented around economic rather than social purposes. The informal-diffuse perspective, as described by Decker and Curry (2000, p. 474), holds that gangs “are diffuse, self-interested and self-motivated aggregations of individuals, most of whom sell drugs for themselves.” Leadership is functional and situational, levels of membership are shifting, meetings are rare or informal, codes of conduct are limited to secrecy and loyalty, and, most importantly, gang members distribute drugs for individual as opposed to a collective purpose. Examples of these descriptions can be found in the research of Decker and Curry (2000, 2002), Decker et al. (2008), Decker and Van Winkle (1995), Fagan (1989), Fleisher (1995, 1998), Hagedorn (1994, 1998), Huff (1996), Klein et al. (1991), McGloin (2005), Moore (1978), and Waldorf (1993). For example, Decker et al. (1998, p. 413) found that only one—the Gangster Disciples—of four police/social service–defined most organized gangs in Chicago and San Diego could adequately fit the instrumentalrational description. Yet, their research also revealed that even in one of the most organized gangs in one of the cities with the most organized gangs, gang members rarely invested their drug profits in the gang’s treasury. As a whole, the research indicates that it was far more common for gang members to “freelance” as drug dealers, and some members filled the role of drug suppliers (Decker et al. 1998; Hagedorn 1994; Valdez and Sifaneck 2004). Evidence in favor of the informal-diffuse perspective exceeds that of the instrumental-rational perspective. It is noteworthy that the evidence supporting the instrumental-rational perspective was found (1) only in ethnographic research settings and (2) only from research carried out in large cities, such as Chicago, Detroit, Los Angeles, and New York. That said, selection bias could be facilitating

88

D.C. Pyrooz et al.

the image of organized gangs in these studies, as researchers focus on atypical gangs in areas of traditional gang cities with an extended gang history (Coughlin and Venkatesh 2003). But as Thrasher (1927) asserted, and Klein and Maxson (2006) confirmed, gangs vary. So it is perhaps more useful to conceptualize gang organization along a continuum, with informal-diffuse at the lower end and instrumentalrational at the upper end. Understanding how gang organizational structure is associated with the behavior of individuals in the group (i.e., gang member) is important for understanding the relationship between gangs and their members’ behavior.

6.2

Gang Organization, Delinquency, and Victimization

Malcolm Klein (2004, p. 14) stated that “[b]eing a gang member can be hazardous to your health.” Nowhere is Klein’s statement better demonstrated than in violence, where Decker and Pyrooz (2010) found that gang member homicide rates are approximately 100 times greater than that of the general public. The empirical literature examining gang membership and criminogenic outcomes consistently demonstrates this association (Krohn and Thornberry 2008). Less emphasis has been placed on examining how these effects differ across gangs and gang members. That is, does gang membership exert a global effect on delinquency and victimization, regardless of the type of gang someone joins? Or, does the effect of gang membership on delinquency/victimization vary according to the organizational structure of the gang? For example, it might be that gang members in instrumentalrational gangs have greater delinquency profiles than gang members in informaldiffuse gangs. In light of the robust effect of gang joining on delinquency and victimization, these are important empirical questions. Five studies have examined the relationship between gang organization and delinquency, although the stated purpose of the research differed (Bjerregaard 2002; Bouchard and Spindler 2010; Decker et al. 2008; Esbensen et al. 2001; Sheley et al. 1995). Bjerregaard (2002) and Esbensen et al. (2001) reported on this relationship indirectly, concentrating on the definitional issues surrounding gang membership. Both studies used increasingly restrictive definitional parameters—organization being one criterion1—to examine how refining categories of gang membership affected outcomes such as theft, arrest, and firearms involvement (Bjerregaard 2002) and attitudinal and behavioral characteristics (Esbensen et al. 2001). As the definitional parameters increased in restrictiveness, so too did the seriousness of gang member attitudinal profiles and delinquency patterns. These findings suggest that organization was associated with criminogenic outcomes in deleterious ways. Sheley et al. (1995) and Decker et al. (2008) carried out more direct tests of the link between gang organizational structure and gang member delinquency. Both of these studies excluded subjects without a history of gang membership, as nongang youth are devoid of the gang experience. These studies explored whether increases in gang organizational structure corresponded with increases in the offending profile of the gang and of the individual gang members. Sheley et al. surveyed 373 incarcerated youth in California, Illinois, Louisiana, and New Jersey. They found that subjects in

6


89

more structured gangs reported that their gang engaged in more drug sales, robberies, and gun carrying than unstructured gangs.2 Further, subjects in structured gangs engaged in higher levels of gun carrying than those in unstructured gangs.3 Decker et al. (2008) surveyed 241 recently arrested juveniles in Arizona correctional facilities.4 They found that current and former gang members from more organized gangs were more likely to be violently victimized, as well as engage in more violent offending and drug selling. This led Decker et al. (2008, p. 169) to conclude that the “more organized the gang, even at low levels of organization, the more likely it is that members will be involved in violent offenses, drug sales, and violent victimizations.” Bouchard and Spindler (2010) examined whether group organization influenced drug dealing and violent and property offenses among self-reported delinquent youth in the Canadian province of Quebec.5 Youth were subdivided into three categories: gang members, deviant group members, and nongroup members. Gang members and deviant group members were compared to determine if the organizational structure of the group was associated with increased delinquent activities. They found that group organization in general was associated with increases in drug dealing and violent offending, but not property offending. In addition, organization partially reduced the effect of gang membership on violent offending. These five studies lead to the same conclusion: increases in gang organizational structure correspond to increases in delinquent offending. At first glance, one might suspect that the opposite relationship might emerge. That is, informal and diffuse gangs would maintain broader offending and victimization profiles since they are an aggregation of criminally inclined individuals. More organized gangs, alternatively, are better able to control and limit the excesses of gang member behavior that could attract attention from authorities, and in turn produce more calculated, specialized offending patterns. Decker (2001) argued that organization makes gangs more efficient in accomplishing collective goals and completing discrete tasks—the majority of which are criminally oriented. Thus, increases in organization exert pressure across gang members to pursue the collective goals of the group, which could range from simply attending meetings to executing premeditated violence. From this viewpoint, it is much less surprising that gangs and gang members have broader offending and victimization profiles, considering risks internal to the gang (e.g., punishments for rule breaking, increased offending expectations) are added to risks external to the gang (e.g., rivalries, turf battles, drug dealing) that bolster the criminogenic profile of gang members (see Melde et al. 2009; Short and Strodtbeck 1965). Given the modest state of the literature and the limited attention to victimization, it is worthwhile to further investigate how organizational structure impacts the offending and victimization patterns of individual gang members.

6.3

The Current Study

Understanding the social influence of groups such as gangs on individual behavior remains of considerable importance to criminologists. Little disagreement exists that gang membership produces higher levels of offending; however, questions

90

D.C. Pyrooz et al.

remain as to how this effect varies by the gang structure to which the gang member belongs. As Decker et al. (2008, p. 157) pointed out, “understanding gang structures and gang behavior without knowing their influence on [gang member] behavior and victimization falls short of providing an explanation of the influence of such characteristics.” The existing body of literature on gang organizational structure fails to meet this critique in the following ways (1) only basic themes can be drawn from research on gang organizational structure due to variation in measurement and the lack of comparative analyses, which constrains the gang literature more broadly, (2) binary gang organization measures artificially restrict the variability to two categories instead of a continuous variable, (3) measurement properties of gang organizational structure remain unexplored, leaving questions about their dimensionality and reliability unanswered, and (4) bivariate analyses do not rule out extraneous explanations. While these critiques do not apply equally to all of the studies that have examined gang organizational structure, none are immune to at least one of these limitations. The present study extends the literature by examining the relationship between gang organizational structures and patterns of delinquency and victimization in the cross-national context. Since the structural characteristics of gang organization are measured systematically across demographically and culturally diverse research sites, it can be determined whether this relationship is robust and consistent across research studies. This approach effectively extends the modest knowledge base on gangs, organization, and criminological outcomes.

6.4 6.4.1

Methods Data

Three data sources are used for the analysis: the evaluation of the Gang Resistance Education and Training (GREAT), the Trinidad and Tobago Youth Survey (TTYS), and the Arizona Arrestee Drug Abuse Monitoring (ADAM) program. Across all three sources of data, only current gang members were included in the analysis. Gang Resistance Education and Training (GREAT). GREAT data were collected as part of an evaluation of gang resistance education administered in middle schools by police officers (Esbensen 2003). GREAT includes both cross-sectional and longitudinal components. This study used the cross-sectional component because it contains data on nearly 6,000 eighth-grade students and allowed us to maximize the pool of gang youth (N = 482). The data were drawn from students in 315 classrooms from 42 public schools in 11 US cities, representing a diverse range of US communities.6 The communities represented urban, suburban, rural, and both racially and ethnically homogenous and diverse areas. Surveys were administered in classroom settings. Participation rates varied for passive parental consent (from 98% to 100%) and active parental consent (from 53% to 75%) schools, but students were representative of the eighth graders in public schools in the communities (Esbensen et al.

6


91

2001). A more detailed discussion of the sampling procedures and completion rates is available elsewhere (Esbensen and Winfree 1998). Trinidad and Tobago Youth Survey (TTYS). The Republic of Trinidad and Tobago is a two-island nation located about 11 km off the northeastern coast of Venezuela, between the Caribbean Sea and the North Atlantic Ocean. The Trinidad and Tobago Youth Survey (TTYS) was modeled after the survey instrument developed by the Social Development Research Group at the University of Washington, which contained measures validated by the international community (Beyers et al. 2004; Harachi et al. 2003). The instrument was slightly modified for use by Trinidad and Tobago youth.7 The target population for the Trinidad and Tobago Youth Survey was defined as third and fifth form (roughly equivalent to 9th and 11th grades in the US educational system) students who attended high-risk urban public schools.8 A total of 27 schools were approached, of which 22 (81.5%) agreed to participate in data collection efforts. The school level response rate was fairly typical by international standards (Gfroere et al. 1997; Prais 2005), but high when compared to other studies conducted in some developing nations (Bulmer and Warwick 1993). Between March and June 2006, the survey instrument was administered to 2,552 students during their homeroom period, of which, 93% adequately completed the survey (N = 2,206).9 The current study uses those students who indicated they were members of a gang at the time of the survey (N = 137). See Katz and Fox (2010) for additional information on the TTYS. Arrestee Drug Abuse Monitoring (ADAM). ADAM is a project funded by the National Institute of Justice designed to monitor national trends in drug use and criminal behavior among the arrestee population, both adults and juveniles. Interview data were collected between 1999 and 2003, within 48 hrs of arrest, for all arrestees at 43 different sites that were a part of the program. In addition to the core questions, some sites administered addenda that were designed to collect more detailed information on a topic of interest. For this study, the Maricopa County site in Arizona implemented a gang addendum among the juvenile arrestees. Maricopa County includes the city of Phoenix, which is the sixth largest city in the USA. Maricopa County is diverse, particularly with regard to the Latino population. The 2000 US Census revealed that Latinos comprise 35% of Maricopa County’s population. Some of the long-standing Latino neighborhoods, known as barrios, have spawned gangs for decades (Katz and Webb 2006; Zatz 1987). The ADAM addendum asked specific questions about the gang the youth were involved with, including specific questions on their gang’s organization. Of the juveniles who were approached to participate in the interview, over 96% agreed. Juveniles were identified as gang members if they self-nominated to gang membership (N = 156).

6.4.2

Summary of Data Sources

Across all three of the studies, youth were asked if they were “currently a member of a gang?”10 This, in combination with the organization measures described below, allowed us to draw comparative linkages across gang members in the three studies.

92

D.C. Pyrooz et al.

The benefit of this approach is that we can make comparisons between school (TTYS, GREAT) and arrestee (ADAM) samples, between the USA and Trinidad and Tobago, and between younger (GREAT) and older juveniles (TTYS). In summary, the similarities in the survey instruments between the studies allow us to assess the differences in gang membership and gang organizational structure between the studies.

6.4.3

Dependent Variables

The dependent variables are all individual-level offending and victimization. The three surveys were not identical, thus the delinquency and victimization variables were constructed so that the best available items were used to create the outcome variables within that dataset. A delinquency scale was constructed for the GREAT and TTYS datasets by summating several dichotomous variables. The GREAT delinquency scale included 11 items11 related to participation in delinquent activities in the past 3 months, and the TTYS delinquency scale included 10 items related to their participation in delinquent activities in the past 12 months.12 Subjects were assigned a “1” for each type of delinquent activity in which the subject participated at least once in the designated time period. The individually summed scores ranged from 0 to the maximum in both study sites. This type of outcome is commonly referred to as a variety score. A variety score does not censor the variation like prevalence scores nor is it likely to be dominated by less serious crimes or victimizations like frequency scores (Hindelang et al. 1981). Data on the types of delinquency behaviors were not collected in the ADAM data; for this reason, only a victimization outcome variable was constructed from these data. Second, a binary victimization item was constructed across all three data sources. This study focused only on violent victimization because general victimization items were limited making it difficult to compare across studies. The GREAT victimization scale included three items related to victimization experiences in the past 3 months,13 the TTYS victimization scale included two items related to victimization experiences in the past 12 months,14 and the ADAM victimization scale included seven items related to victimization experiences in the past 30 days.15 Across all three data sources, youth that reported experiencing a victimization were coded “1” and the remaining subject were coded “0.”

6.4.4

Independent Variables

The key independent variables are all indicators of gang organization used in prior gang research (Decker et al. 2008; Peterson et al. 2001). Four questions about the structure and organization of a respondent’s gang were available in all three data sources. Respondents were asked whether their gang had leadership, regular

6


Table 6.1 Summary statistics for study variables GREAT TTYS Mean or% (SD) Mean or% (SD) Outcome variables* Delinquency 5.41 (3.32) 2.79 (2.68) Victimization 77 71 Gang organization variables Has a leadera, b, c 74 45 Regular meetingsb, c 56 56 Gang has rulesa, b 74 52 Insigniaa, c 91 45 2.87 (1.14) 1.99 (1.27) Summated organization scalea, b Control variables White 23 Blackb 32 Hispanicb 25 Otherb 19 African 42 East Indian 21 Afro/Indian 11 Other 26 Agea, b 14.12 (0.73) 15.37 (1.05) Age range 11–18 10–19 Maleb, c 63 58 N 482 137 *Note that statistical significance was not assessed for the outcome variables time for offending and victimization differed across the samples a Significant differences at p < .05 between GREAT and TTYS samples b Significant differences at p < .05 between GREAT and ADAM samples c Significant differences at p < .05 between TTYS and ADAM samples

93

ADAM Mean or% (SD)

47 31 35 53 89 2.11 (1.12)

21 15 60 5

15.32 (1.45) 9–17 87 156 because exposure

meetings, rules that members had to follow, and whether or not the gang had insignia (i.e., special colors, symbols, signs, or clothes). All these variables were measured dichotomously, that is, whether or not their gang possessed the given organizational quality. Table 6.1 shows the descriptive statistics for the four gang organizational structure variables that were available across the three sources of data. In addition, demographic characteristics—age, gender, and race/ethnicity— were entered as control variables in the multivariate models.16

6.5

Analytic Strategy

The analysis proceeds in three stages. First, the univariate statistics are presented (i.e., means, standard deviation). Second, bivariate correlations are reported for the organizational structure variables to evaluate dimensionality. All the organizational

94

D.C. Pyrooz et al.

variables are dichotomous, thus, polyserial correlation coefficients will be estimated. In order to obtain an unbiased estimate of the correlation between two dichotomous variables, tetrachoric correlations are estimated, which are a special case of polyserial correlations for dichotomous variables.17 This analysis will aid in determining the relationship between different measures of gang organization. Third, multivariate ordinary least squares and logistic regressions are estimated using the aggregated construct and disaggregated indicators of gang organization on offending and victimization. All models were estimated in STATA 10.0 using robust standard errors, which better estimate errors in the face of suspect regression assumptions such as heteroskedasticity and autocorrelation. For the GREAT and TTYS data, clustering by school is accounted for, and clustering by residential zip code is accounted for in the ADAM data.18

6.6 6.6.1

Results Summary Statistics

Table 6.1 provides summary statistics for the study variables. The respondents in each of the samples engaged in a variety of offending and experienced a high rate of victimization, which is consistent with the larger body of gang research (Krohn and Thornberry 2008). While there appears to be substantively significant differences in offending and victimization across the samples, it is important to recall that the exposure times for offending and victimization differ across samples, making comparisons inappropriate. GREAT respondents tended to be younger than respondents in the TTYS and ADAM. The racial and ethnic makeup differed across the studies, mostly due to demographic differences associated with the communities the youth were drawn from. In addition, there was a large gender difference between school and arrestee samples, where males comprised nearly 90% of ADAM respondents compared to approximately 60% of TTYS and GREAT respondents. This is likely a function of the nature of the samples, where there tends to be an inverse relationship between female composition and the criminogenic seriousness of the sample (see Klein and Maxson 2006, pp. 41–42). Turning to the gang organization variables, most gang members conveyed that their gang contained a moderate degree of organizational structure. These results differed by study, however. Youth in the GREAT study, on average, reported that their gangs had three of the organizational structure characteristics, whereas ADAM and TTYS averaged nearly an entire unit lower. Indeed, the average across all four gang organizational structure items was higher for GREAT gang youth than the other two samples, especially for the leadership and rule items. In addition, insignia was half as likely to be reported among TTYS gang youth compared to gang youth in the US samples.

6


95

Table 6.2 Bivariate correlations of gang organization variables and outcomes GREAT (N = 482) X1 X2 X3 X4 X5 Delinquency 0.06 0.07 0.10 0.12 0.08 Y1 Y2 Victimization 0.09 0.03 0.15 0.13 0.11 X1 Has a leader – X2 Regular meetings 0.58 – X3 Gang has rules 0.52 0.34 – X4 Insignia 0.34 0.18 0.44 – X5 Organization index – – – – Mean inter-item r 0.40

6.6.2

Y1 Y2 X1 X2 X3 X4 X5

TTYS (N = 137) Delinquency Victimization Has a leader Regular meetings Gang has rules Insignia Organization index Mean inter-item r

X1 0.17 0.13 – 0.05 0.18 0.27 – 0.32

X2 −0.09 0.12

Y1 Y2 X1 X2 X3 X4 X5

ADAM (N = 156) Delinquency Victimization Has a leader Regular meetings Gang has rules Insignia Organization index Mean inter-item r

X1 – 0.03 – 0.35 0.24 0.30 – 0.35

X3 0.15 0.25

X4 0.20 −0.12

0.45 0.37 –

– 0.56 –

– –

X2 – 0.29

X3 – 0.22

X4 – 0.13

– 0.48 –

– –

X5 0.16 0.11

–

X5 – 0.20

– 0.40 0.35 –

Bivariate Associations

Table 6.2 displays the bivariate associations of the gang organization variables. Because all the gang organization variables were dichotomous, we obtained tetrachoric correlation coefficients that produce more accurate estimates of the magnitude of the relationship. Our goal was to assess the internal validity of commonly used gang organization items prior to condensing these items into a factor score or summating them into an index; examining correlation coefficients is the first step in this process. If gang organization can be conceived in terms of a continuum from informal-diffused (the least organized) to instrumental-rational (the most organized), it should be reflected by consistently large tetrachoric correlation coefficients. The bivariate associations displayed in Table 6.2 do not display the degree of consistency one would expect if gang organization was indeed a latent factor. In fact, the correlation matrices are checkered with weak, modest, moderate, and strong

96

D.C. Pyrooz et al.

Table 6.3 Multivariate OLS regression predicting individual delinquency across the data sources Delinquency GREAT TTYS b b b b B (SE) B (SE) B (SE) B (SE) Gang Organization Variables Has a leader −.04 −.28 .09 .51 (.39) (.55) Regular meetings .02 .11 −.09 −.46 (.27) (.49) Gang has rules .09** .69 .14 .73 (.34) (.44) Insignia .05 .56 .10 .56 (.43) (.53) Gang organization .08** .23 .17* .35 index (.13) (.14) N 442 442 137 137 R2 .10 .09 .17 .11 All standard errors are robust standard errors for clustering. Clustering by school is accounted for in the GREAT (41 clusters) and TTYS (21 clusters) data, and by residential zip code in the ADAM (57 clusters) data Controls: Race/ethnicity, age, and gender of the respondent are controlled for in all models. Coefficients are not presented here, but are available upon request *p < .05; **p < .10

coefficients, ranging from 0.05 to 0.58. There is some semblance of bivariate consistency across the studies for insignia/rules and rules/meetings. In addition, there is relative consistency in the mean inter-item correlations across the studies. But as a whole, the mean inter-item correlations across the studies are not as large as one would expect to inspire confidence in the latent construct of gang organizational structure. This leads us to believe that certain items may be more strongly related to the organization of the gang than others, making it difficult to determine whether items are differentially related to criminological outcomes. While there is evidence that gang organization increases offending, empirically, we do not know which characteristics of gang organization are driving that relationship. Thus, we will examine the effect of these organizational characteristics on delinquency and victimization independently and cumulatively.

6.6.3

Comparative Multivariate Regression Models

Table 6.3 displays the results of multivariate OLS regression models to examine which specific gang organization variables explain the variation in delinquency. Table 6.4 shows the results of the logistic regression models predicting victimization. All standard errors are robust standard errors adjusted for clustering. This is especially

Table 6.4 Multivariate logistic regression predicting individual victimization across the data sources Victimization GREAT TTYS ADAM OR b OR b OR b OR b OR b OR b (SE) (SE) (SE) (SE) (SE) (SE) Gang organization variables Has a leader 0.78 −.25 1.49 .40 0.63 −.46 (.36) (.43) (.39) Regular meetings 1.28 .25 0.90 −.10 2.77* 1.02 (.24) (.40) (.35) Gang has rules 1.16 .15 3.35** 1.21 1.45 .37 (.21) (.69) (.37) Insignia 2.91* 1.07 0.27* −1.31 0.95 −.05 (.36) (.53) (.55) Gang organization 1.17 .16 1.01 .01 1.34* .29 index (.12) (.19) (.14) N 442 137 137 150 150 R2 .04 .11 .05 .08 .08 All standard errors are robust standard errors for clustering. Clustering by school is accounted for in the GREAT (41 clusters) and TTYS (21 clusters) data, and by residential zip code in the ADAM (57 clusters) data Controls: Race/ethnicity, age, and gender of the respondent are controlled for in all models. Coefficients are not presented here, but are available upon request *p < .05; **p < .10

6 Gang Organization, Offending, and Victimization… 97

98

D.C. Pyrooz et al.

important for the present study, where students were surveyed in schools in two of the samples, and in the third, they are the result of arrests. The standard errors are adjusted based on the presence of any clustering based on school (GREAT and TTYS) or residential zip code (ADAM). Additionally, each regression model in Tables 6.3 and 6.4 adjusts for race/ethnicity, age, and gender. Table 6.3 displays four models predicting delinquency, two from the GREAT sample and two from the TTYS sample. For the GREAT sample, only one of the four measures of gang organization was even marginally statistically significant: individuals who indicated their gang had rules reported an increase in offending variety. In the TTYS sample, none of the four common gang organization variables were significant predictors of delinquency. It should be noted that the direction of all of the gang organization variables was not positive; this suggests that having that characteristic was not consistently related to more offending. Turning to the summated gang organization scales, as opposed to the individual measures, there is modest support in both the GREAT and TTYS samples for the positive relationship between gang organization and delinquency. This finding is consistent with the larger body of research on gang organization and delinquent offending. Table 6.4 displays the logistic regression results of the common gang organization predictors on the likelihood of victimization. In the GREAT sample, insignia nearly tripled the odds of victimization among gang youth. Yet, in the TTYS sample, insignia was negatively associated with victimization. Alternatively, rules tripled the odds of victimization among Trinidadian gang youth. In the ADAM sample, only regular meetings were associated with victimization, nearly tripling the odds of its occurrence. In terms of the gang organization scale, increases in gang organization corresponded to statistically significant increases in victimization only among gang members in the ADAM sample. No relationship between the organization scale and victimization was observed in the GREAT and TTYS samples. In summary, the results of the organization-victimization analyses reveal that there is very little consistency across research contexts.

6.7

Discussion

The organizational features of youth gangs have not been the subject of a large body of research. The present study examined the relationship between gang organizational structure and delinquency and victimization using data from three research projects. The data were collected from school-based samples in the USA and Trinidad and Tobago, and a sample of juvenile arrestees in one US county. The goals of the current study included (1) examining gang organizational structure in a comparative context and (2) examining the relationship between gang organizational structure and delinquency and victimization. This study is unique in that it compares gang organizational structure across populations (i.e., school, arrestee) and settings (i.e., USA, Trinidad and Tobago). This section is guided by three key findings that merit further discussion.

6


99

First, our findings suggest that gangs might be more organized in the USA than in Trinidad. School-age youth in Trinidad and Tobago reported the lowest level of gang organizational structure when compared to both school-age youth and juvenile arrestees in the USA. While there could be numerous explanations for this finding, it might suggest that the organizational processes within gangs are unique and subject to social, structural, or cultural factors, as well as time periods. A similar finding was observed when comparing gang youth in the USA to gang youth in the Netherlands (Esbensen and Weerman 2005). Another consideration is that gangs in Trinidad and Tobago are a relatively recent phenomenon, not emerging until the turn of the century (Katz and Fox 2010), whereas many of the youth participating in the GREAT project were from communities that had long-standing gang problems. Gangs in communities with long-standing gang problems might be more organizationally structured. This has been a consistent observation about English gangs when compared with those in the USA (Pitts 2007) and may reflect the history of gangs in the USA. With the above said, juvenile gang members from our arrestee sample belonged to less organized gangs than our respondents from the GREAT sample. This is contrary to what we expected; youth engaged in more serious forms of delinquency and involved in the criminal justice system (i.e., those found in an arrestee sample) would belong to more organizationally structured gangs. This may be a function of dramatic ethnic differences across the research contexts, although group processes within gangs tend to trump demographic differences across gangs (Klein and Maxson 2006). Second, our analyses reveal mixed support for the relationship between gang organizational structure and offending and victimization. Contrary to our hypotheses and inconsistent with the extant literature, there was little to no evidence for this relationship, although nearly all of the coefficients were positive. At best, however, the results of this analysis provide only modest support that increased organizational structure is positively related to individual patterns of delinquency and victimization among gang members. Our aggregated gang organizational structure measure performed somewhat consistently with our hypotheses. Gang organizational structure was positively related to delinquency for both samples (GREAT and TTYS) and was positively related to victimization among one of three samples (ADAM). It should be noted, however, that the associations were modest and the level of variance explained was low. When all of the common organizational structure predictors were examined individually, none performed consistently across the samples. Whether a gang had leaders was unrelated to offending and victimization in all settings. Whether a gang had regular meetings, rules, and insignia mattered differently for each of the outcomes in each of the settings. This introduces a problem for interpreting the meaning of these effects. For example, if rules were consistently related to offending and victimization across research contexts, this would make sense. The presence of rules, a measure of the ability of groups to discipline their members and structure behavior more formally, is a key component of formal organizations. Membership in gangs that contain rules may introduce the need to sanction ruleviolating members for wayward behavior. The fact that rules mattered only for

100

D.C. Pyrooz et al.

victimization among gang youth in TTYS raises the question of whether certain characteristics of organizational structure might be unique to the research context. How structural and cultural factors interact to produce varying levels of gang organizational structure should be on the agenda of future empirical gang research. Third, it is necessary to better understand empirically the organizational structure of gangs. To date, researchers have rarely examined the structural factors that may be related to offending and victimization, and when they have done so, they have tended to focus on a fairly limited number of structural attributes. Future research should explore different survey items and analytic strategies to better understand the organizational structure of gangs. A limitation of this line of research more generally is that is relies on several dichotomous variables for information on gang organizational structure. This constrains the variability in the construct of gang organizational structure, reducing the likelihood of identifying statistically significant findings or identifying a latent construct of gang organizational structure. We recommend that future survey development introduce ordinal variables to measure gang organizational structure. In the current study, for example, if Likert-scaled items for leaders, rules, meetings, and insignia were available for the present study, it would have increased the variability to 5 categories for individual measures and to 20 categories for the aggregate measure. This speaks to the larger subject discussed earlier in the framework of this chapter: gang organization should operate on a continuum ranging from informal-diffuse gangs to instrumental-rational gangs rather than as a dichotomy. Extant methods make it impossible to identify this construct. Another analytic strategy to consider is employing social network methodology; however, others have noted the challenges posed for collecting social network data (McGloin and Kirk 2010; Morselli 2008; Papachristos 2006). Our findings affirm the value of more comparative research and provide support for the comparative methodological objectives of the Eurogang research program. There is limited work that compares the extent and nature of gang problems across nations—i.e., Italy and France (Blaya and Gatti 2010), Italy and Switzerland (Haymoz and Gatti 2010), the USA and the Netherlands (Esbensen and Weerman 2005), and the USA and Germany (Huizinga and Schumann 2001). More work in this area of study is warranted to better understand the generalizability of gang research and, in turn, better understand the causal pathways toward gangs, gang membership, and gang violence. These issues should be on the agenda of future gang research.

Notes 1. Bjerregaard’s (2002) study was based on a sample of 1,663 mostly black and Hispanic innercity high school students (17% “current” gang prevalence) in California, Illinois, Louisiana, and New Jersey. Youth reported whether they were in a gang, then if that gang was “just a bunch of guys” or an “organized gang.” From this, youth were placed in four categories (1) youth not in gangs, (2) youth not in gangs but hung around with “a bunch of guys,” (3) youth

6

2.

3.

4.

5.

6.

7.

8.

9.


101

in gangs that were a “bunch of gangs” (i.e., not an organized gang), and (4) youth in gangs that were organized. These four categories were then compared independent of one another. Esbensen et al.’s (2001) study was based on a sample of 5,935 middle school students (17% “ever” gang prevalence) in 11 demographically diverse, urban, rural, and suburban cities. Their method of partitioning began by self-nominated gang membership, then diverged to the delinquency of the gang, the organization of the gang, and the individual centrality within the gang. This resulted in a total of six categories: (1) youth never in gangs, (2) youth ever in gangs, (3) youth currently in gangs, (4) youth in delinquent gangs, (5) youth in organized (initiations, leaders, and symbols/colors), delinquent gangs, and (6) youth with high centrality in organized, delinquent gangs. Sheley et al. (1995) created a binary measure of organization that they labeled gang structure, as follows: the individual had to (1) self-nominate the gang as organized, (2) report that the gang was comprised of at least 50 members, and (3) indicate that the gang meets at least 3 of the criteria of organization—“an ‘official’ name, designated leadership, regular meetings, designated clothing, and a specified turf to be defended” (Sheley et al. 1995, p. 59). If the subjects’ gangs did not meet the above criteria, then they were classified as being a member of an “unstructured” gang. No statistically significant effects emerged for gang member drug sales or robbery, which were likely absorbed by including gang drug sales and robbery in the analytic model. In analyzing the effect of gang structure on gang member delinquency, Sheley et al. (1995) included the subjects’ perceptions of their gang’s drug sales, drug use, burglary, robbery, and gun carrying in the models predicting gang member drug sales, drug use, burglary, robbery, and gun carrying. This raises a question of whether gang structure could possibly exert an effect net of the egocentric measure of gang criminal activity—i.e., the statistical insignificance could be an artifact of the analytic approach. Decker et al. created a summative index drawn from seven binary indicators of gang organization, including gang leadership, regular meetings, rules, punishments for rule breaking, unique insignia (e.g., colors, signs, symbols), member responsibilities, and shared money. Two models were estimated partitioning the sample by current and former gang membership, as recommended by prior research (see Katz et al. 2005, p. 83). Bouchard and Spindler’s (2010) original sample consisted of 1,166 respondents, but they excluded youth that did not self-report to engaging in delinquency, resulting in a final sample of 523 respondents. These respondents were then classified as “gang members” (N = 44), “delinquent group members” (N = 171), and “nongroup members” (N = 308) based on their responses to whether they were (1) a member of a relatively organized gang, and if not, then (2) a member of a youth group involved in deviance of any sort. If respondents answered “no” to both of these questions, they were classified as nongroup members. Group organization was measured in a similar manner to Decker et al. (2008) and included nine binary items summated to create an organization scale. Empirical tests of the organization hypothesis were applied only to individuals that reported belonging to some type of group (N = 215). These cities include Kansas City, MO; Las Cruces, NM; Milwaukee, WI; Omaha, NE; Orlando, FL; Philadelphia, PA; Phoenix, AZ; Pocatello, ID; Providence, RI; Torrance, CA; and Will County, IL. The instrument was provided to key stakeholders employed by the Ministry of Education to seek their advice on altering the instrument so that it reflected regional language and culture (i.e., monetary units, social activities, and organizations). High-risk schools were defined as those that the Ministry of Education had identified as having a disproportionate number of students who either lived in high-crime communities or attended a school that recorded a high number of delinquent incidents. Urban was defined as any school that was located within one of five urban school districts. This eliminated schools from three school districts, including the school district of Tobago. Of the 67 public schools that were eligible for inclusion in the study, 27 were identified as being at “high risk.” All students attending school and present in their homeroom on the day the survey was administered were provided with a copy of the survey instrument. Students were informed that if they

102

D.C. Pyrooz et al.

did not wish to participate in the survey, they were to not fill out any questions and turn it in incomplete. Likewise, they were informed that if they did not want to answer a specific question, they were not to provide an answer to that question. As a consequence, our sample excludes those youth not enrolled in school, missing from school, or those youth in hospitals or committed to detention facilities. The Ministry of Education in Trinidad and Tobago stated to the researchers that the data on absenteeism and enrollment is not routinely collected and that the data that is available is not necessarily accurate. School officials estimate that about 5–10% of students are absent on a given day, leaving the participation rate at about 90%. 10. The exact language of this question varied slightly across the studies but collectively inquired into whether a respondent self-nominated to being a gang member. 11. (1) Attacked someone with a weapon, (2) hit someone, (3) used a weapon to rob someone, (4) went into (or tried to) a building to steal something, (5) stole or tried to steal a motor vehicle, (6) carried a hidden weapon, (7) purposely damaged or destroyed property, (8) spray painted, (9) stole something worth less than $50, (10) stole something worth more than $50, and (11) avoided paying for something. 12. (1) Attacked someone with a weapon, (2) attacked someone with the intention of seriously hurting them, (3) used a weapon to rob someone, (4) gone into or tried to go into a building to steal something, (5) carried a handgun, (6) taken a handgun to school, (7) stolen or tried to steal something worth more than $300 (TT) (equivalent to about $50 USD), (8) stolen or tried to steal something worth less than $300 (TT), (9) been drunk or high at school, and (10) got drunk or high. 13. (1) Hit by someone trying to hurt you, (2) someone used a weapon or force to get money or things from you, and (3) been attacked by someone with a weapon or by someone trying to seriously hurt or kill you. 14. (1)Someone threatened or injured you with a weapon such as a gun, knife, or small stick on school property, (2) you in a physical fight on school property. 15. Threatened with a gun, (2) shot at, (3) shot, (4) threatened with some other weapon, (5) injured with some other weapon, (6) jumped or beat up, (7) robbed. 16. Klein and Maxson (2006) have suggested that the age composition of a gang’s members will contribute to gang behavior. Because our measure of age was truncated in our sample—juveniles—we were unable to assess this relationship. 17. These coefficients are estimated in order to obtain an unbiased estimate of the magnitude of the relationship between the organizational variables. Briefly, there are some assumptions one must make when estimating tetrachoric correlations. This involves the hypothesized existence of continuous latent variables underlying each dichotomous variable (X*), and analyzing these rather than the measured binary variables. For each continuous latent variable, there is a latent threshold parameter (v) (Muthén 1993). The latent threshold is essentially the cut point at which the variable goes from 0 to 1 on the measured dichotomous variable. Thus, we assume X = 1 if X* > v and X = 0 if X* < v. The resulting correlation coefficient is an unbiased estimate for the latent, normally distributed, continuous variables. 18. Multicollinearity diagnostics indicated that multicollinearity was not a problem in any of the multivariate models (i.e., no variance inflation factor exceeded 2).

References Beyers JM, Toumbourou JW, Catalano RF, Arthur MW, Hawkins JD (2004) A cross-national comparison of risk and protective factors for adolescent substance use: The United States and Australia. J Adolesc Health 35:3–16 Bjerregaard B (2002) Self-definitions of gang membership and involvement in delinquent activities. Youth Soc 34:31–54 Bjorgo T (2005) Root causes of terrorism: myths, reality, and ways forward. Routledge, New York Blaya C, Gatti U (2010) Eur J Crim Policy Res 16:127–144

6


103

Bohstedt J (1994) The dynamics of riots: escalation and diffusion/contagion. In: Potegal M, Knutson JF (eds) The dynamics of aggression: biological and social processes in dyads and groups. Lawrence Erlbaum, Hillsdale Bouchard M, Spindler A (2010) Groups, gangs, and delinquency: does organization matter? J Crim Justice 38:921–933 Bulmer M, Warwick, DP (1993) Social research in developing countries: surveys and censuses in the third world. New York: Wiley Coughlin BC, Venkatesh S (2003) The urban street gang after 1970. Ann Rev Sociol 29:41–64 Decker SH (2001) The impact of organizational features on gang activities and relationships. In: Klein MW, Kerner H-J, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Norwell, pp 21–40 Decker SH, Curry GD (2000) Addressing key feature of gang membership: measuring the involvement of young members. J Crim Justice 28:473–482 Decker SH, Curry GD (2002) Gangs, gang homicides, and gang loyalty: organized crimes or disorganized criminals. J Crim Justice 30:343–352 Decker SH, Pyrooz DC (2010) Gang violence: context, culture, and country. In Small Arms Survey, 2010. Cambridge University Press, Cambridge, pp. 128–155 Decker SH, Pyrooz DC (2011) Gangs: another form of organized crime? In: Paoli L (ed) Handbook of organized crime. Oxford University Press, Oxford Decker SH, Van Winkle B (1995) “Slinging dope”: the role of gangs and gang members in drug sales. Justice Quart 11:583–604 Decker SH, Bynum T, Weisel D (1998) A tale of two cities: gangs as organized crime groups. Justice Quart 15:395–425 Decker SH, Katz CM, Webb VJ (2008) Understanding the black box of gang organization: implications for involvement in violent crime, drug sales, and violent victimization. Crime Delinq 54:153–172 Esbensen F-A (2003) Evaluation of the Gang Resistance Education and Training (GREAT) program in the United States, 1995–1999. 2nd ICPSR version. Inter-university Consortium for Political and Social Research, Ann Arbor, MI. DOI: 10.3886/ICPSR03337 Esbensen F-A, Weerman F (2005) Youth gangs and troublesome youth groups in the United States and the Netherlands: a cross-national comparison. Eur J Criminol 2:15–37 Esbensen F-A, Winfree LT (1998) Race and gender differences between gang and non-gang youths: results from a multi-site survey. Justice Quart 15:505–526 Esbensen F-A, Winfree LT Jr, He N, Taylor TJ (2001) Youth gangs and definitional issues: when is a gang a gang, and why does it matter? Crime Delinq 47:105–130 Fagan J (1989) The social organization of drug dealing and drug use among urban gangs. Criminology 27:633–670 Fleisher MS (1995) Beggars and thieves: lives of urban street criminals. University of Wisconsin Press, Madison Fleisher MS (1998) Dead end kids: gang girls and the boys they know. University of Wisconsin Press, Madison Gfroere J, Wright D, Kopstein A (1997) Prevalence of youth substance use: the impact of methodological differences between two national surveys. Drug Alcohol Depend 47:19–30 Hagedorn JM (1994) Neighborhoods, markets, and gang drug organization. J Res Crime Delinq 31:264–294 Hagedorn JM (1998) People and folks: gangs, crime and the underclass in a rustbelt city, 2nd edn. Lakeview Press, Chicago Harachi TW, Hawkins JD, Catalano RF, LaFazia AM, Smith BH, Arthur MW (2003) Evidencebased community decision making for prevention: two case studies of Communities That Care. Jpn J Sociol Criminol 28:26–37 Haymoz S, Gatti U (2010) Girl members of deviant youth groups, offending behaviour and victimisation: results from the ISRD2 in Italy and Switzerland. Eur J Crim Policy Res 16:167–182 Hindelang M, Hirschi T, Weis J (1981) Measuring delinquency. Sage, Newbury Park

104

D.C. Pyrooz et al.

Huff CR (1996) The criminal behavior of gang members and nongang at-risk youths. In: Huff CR (ed) Gangs in America, 2nd edn. Sage, Thousand Oaks, pp 75–102 Huizinga D, Schumann K (2001) Gang membership in Bremen and Denver: comparative longitudinal data. In: Klein MW, Kerner H-J, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Norwell Katz CM, Fox A (2010) Risk and protective factors associated with gang involved youth in a Caribbean nation: analysis of the Trinidad and Tobago Youth Survey. Pan-Am J Public Health/ Rev Panam Salud Publica 27:187–202 Katz CM, Webb VJ (2006) Policing gangs in America. Cambridge University Press, Cambridge Katz CM, Webb VJ, Decker SH (2005) Using the Arrestee Drug Abuse Monitoring (ADAM) program to further understand the relationship between drug use and gang membership. Justice Quart 22:58–88 Klein MW (1971) Street gangs and street workers. Prentice-Hall, Englewood Cliffs Klein MW (2004) Gang cop: the words and ways of Officer Paco Domingo. Altamira, Walnut Creek Klein MW (2005) The value of comparisons in street gang research. J Contemp Crim Justice 21:135–152 Klein MW (2006) Peer effects in naturally occurring groups: the case of street gangs. In: Dodge KA, Dishion TJ, Lansford JE (eds) Deviant peer influences in programs for youth: problems and solutions. Guilford, New York, pp 234–250 Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, New York Klein MW, Maxson CL, Cunningham LC (1991) “Crack”, street gangs, and violence. Criminology 29:623–650 Krohn MD, Thornberry TP (2008) Longitudinal perspectives on adolescent street gangs. In: Liberman AM (ed) The long view of crime: a synthesis of longitudinal research. National Institute of Justice, Washington, DC, pp 128–160 Louis WR, Taylor DM (2002) Understanding the September 11 terrorist attack on America: the role of intergroup theories of normative influence. Anal Soc Issues Public Policy 2:87–100 Maxson CL, Klein MW (1995) Investigating gang structures. J Gang Res 3:33–40 McGloin J (2005) Policy and intervention considerations of a network analysis of street gangs. Criminol Public Policy 4:607–636 McGloin J, Decker SH (2010) Theories of gang behavior and public policy. In: Barlow HD, Decker SH (eds) Criminology and public policy: putting theory to work. Temple University Press, Philadelphia, pp 150–165 McGloin JM, Kirk D (2010) Social network analysis. In: Piquero AR, Weisburd D (eds) Handbook of quantitative criminology. Springer, New York, pp 209–224 McPhail C (1991) The myth of the madding crowd. Walter de Gruyter, Hawthorne Melde C, Taylor TJ, Esbensen F-A (2009) “I got your back”: an examination of the protective function of gang membership in adolescence. Criminology 47:565–594 Mieczkowski T (1986) ‘Geeking up’ and ‘throwing down’: Heroin street life in Detroit. Criminology 24:645–666 Moore JW (1978) Homeboys: gangs, drugs, and prison in the barrios of Los Angeles. Temple University Press, Philadelphia Morselli C (2008) Inside criminal networks. Springer, New York Muthen BO (1993) Goodness of fit with categorical and other nonnormal variables. In: Bollen KA, Long JS (eds) Testing structural equation models. Sage, Newbury Park, CA, pp 205–234 Padilla F (1992) The gang as an American enterprise. Rutgers University Press, New Brunswick Papachristos AV (2006) Social network analysis and gang research: theory and methods. In: Short JF Jr, Hughes LA (eds) Studying youth gangs. AltaMira, Lanham, pp 99–116 Peterson D, Miller J, Esbensen F-A (2001) The impact of sex composition on gangs and gang member delinquency. Criminology 39:411–440 Pitts J (2007) Reluctant gangsters: youth gangs in Waltham forest. University of Bedfordshire, Bedfordshire, UK Prais SJ (2005) Two recent (2003) international surveys of schooling attainments: England’s problems. The National Institute of Economic and Social Research Discussion paper, No.258

6


105

Sanchez-Jankowski M (1991) Islands in the street: gangs and American Urban Society. University of California Press, Berkeley Sheley JF, Zhang J, Brody CJ, Wright JD (1995) Gang organization, gang criminal activity, and individual gang members’ criminal behavior. Soc Sci Quart 76:53–68 Short JF Jr, Strodtbeck FL (1965) Group process and gang delinquency. University of Chicago Press, Chicago Skolnick JH, Correl T, Navarro E, Rabb R (1990) The social structure of street drug dealing. Am J Police 9:1–41 Staub E (2002) The roots of evil: the origins of genocide and other group violence. Cambridge University Press, New York Taylor C (1990) Dangerous society. Michigan State University Press, East Lansing Thrasher FM (1927) The gang: a study of 1,313 gangs in Chicago. University of Chicago Press, Chicago Valdez A, Sifaneck SJ (2004) “Getting high and getting by”: dimensions of drug selling behaviors among American Mexican gang members in South Texas. J Res Crime Delinq 41:82–105 Venkatesh S (1997) The social organization of street gang activity in an urban ghetto. Am J Sociol 103:82–111 Venkatesh SA, Levitt SD (2000) “Are we a family or a business?” History and disjuncture in the urban American street gang. Theory Soc 29:427–462 Waldorf D (1993) When the Crips invaded San Francisco: gang migration. J Gang Res 1:11–16 Weerman FM, Maxson CL, Esbensen F-A, Aldridge J, Medina J, van Gemert F (2009) Eurogang program manual: Background, development, and use of the Eurogang instrument in multi-site, multi-method comparative research. Retrieved December 8, 2010 from http://www.umsl. edu/~ccj/eurogang/Eurogang_ 20Manual.pdf Zatz MS (1987) Chicano youth gangs and crime: the creation of moral panic. Contemp Crisis 11:129–158

Chapter 7

Betwixt and Between Street and Prison Gangs: Defining Gangs and Structures in Youth Correctional Facilities Cheryl L. Maxson

Gang activity in correctional institutions presents an enormous challenge to corrections personnel because gang issues are perceived to underlie much of the violence and institutional misconduct that occurs in these settings. Moreover, gang dynamics may interfere with programs or rehabilitative efforts if gang tensions deter some youth from participation or where programmatic activities are suspended due to fighting. Exposure to gang activity in institutions may solidify gang identities, particularly in the absence of more socially conforming peers. To the extent that these settings may foster more group organization, these groups may become more successful in accomplishing criminal goals and actualize the potential of influencing criminal activities and gang dynamics in the community (Hunt et al. 1993; McGloin 2007). Given the significance of gang dynamics in American correctional settings, it is surprising that little research has been conducted to illuminate the similarities and differences between street and prison gangs. Most research about gangs in correctional institutions focuses on adult prisons. Since gang participation is primarily a youthful phenomenon, it is very surprising that studies about gangs and gang membership in specifically youth correctional facilities are rare. I will adopt the term “institutional gangs” to denote groups that form in youth correctional settings and to distinguish them from street and from prison gangs. My primary interest in this chapter is to examine this distinction and to see whether the Eurogang framework can help guide this work. This chapter addresses both methodological and substantive issues in an attempt to apply the Eurogang research paradigm in a carceral environment. Following a brief treatment of the research literature, I describe this particular research setting and the unique policy challenges that mark this organization at this point in time. I then turn to a discussion of the way in which I conceptualize this research within the Eurogang research paradigm and argue that while it was developed to foster research in

C.L. Maxson (*) Department of Criminology, Law and Society, University of California, Irvine, CA, USA e-mail: [email protected] F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_7, © Springer Science+Business Media, LLC 2012

107

108

C.L. Maxson

community settings, this approach is flexible enough to guide research in correctional institutions, even while requiring certain accommodations. In particular, I review the elements of the Eurogang definition and suggest modifications that maintain its spirit while making it more appropriate to this setting. Next, I report data from interviews with correctional gang specialists which include the Eurogang Expert Survey which I used to generate profiles of all gangs active in these facilities. These profiles confirm that these institutional gangs fall within the terrain of the Eurogang definition but present several challenges in the application of the structural typology proposed by Maxson and Klein (1995; Klein and Maxson 2006). The expert survey data generate a depiction of gang forms that more closely resemble street gangs than the purportedly more organized prison gangs. I conclude with a consideration of the policy implications of these findings and further steps this research may take.

7.1

Research on Gangs in Institutional Settings

The focus of the empirical literature that addresses gang issues in institutional settings can be roughly placed into four topics. The area that has spawned most of the studies published in the past two decades is the relationship between gang membership (either prior to or during incarceration) and institutional misconduct or violence. Some of this work concerns misconduct in youth facilities (DeLisi et al. 2010; MacDonald 1999; Parker et al. 2008; Trulson 2007) or of youth in adult facilities (Kuanliang et al. 2008; Tasca et al. 2010) but most studies focus on adult prisons (DeLisi et al. 2004; Drury and DeLisi 2011; Gaes et al. 2002; Griffin and Hepburn 2006; Huebner 2003; Huebner et al. 2007; Krienert and Fleisher 2001). The overwhelming conclusion to be drawn from this segment of the research literature is that gang affiliation—whether street or prison—is associated with misconduct and/or violence. A second area of research concerns the relationship between prior gang membership and recidivism once individuals leave carceral environments (Benda et al. 2001; Caudill 2010; Huebner et al. 2007; Lattimore et al. 2004; Trulson et al. 2005). These studies find that gang members are more likely to recidivate than those without affiliation. Third, scholars have addressed the security and treatment responses to gang members in correctional settings and the challenges they present to both correctional officials and service-providers (Davis and Flannery 2001; DiPlacido et al. 2006; Fleisher and Decker 2001; Howell and Roush (1997), Inderbitzin (2007), Pyrooz et al. 2011; Rudell et al. 2010; Winterdyk and Rudell 2010). This work describes primarily suppressive responses with no empirical assessment of effectiveness; only DiPlacido et al. (2006) report on the evaluation of a program for gang members, implemented in a forensic mental health facility in Canada. The fourth area of research concerns organizational characteristics of gangs in carceral environments. I could find no empirical work on prison gangs that approximates the type of analyses reported on street gangs. For example, Pyrooz and his

7

Betwixt and Between Street and Prison Gangs…

109

colleagues (Chap. 6, this volume) compare four gang organizational characteristics in three different samples. One of these, a sample of incarcerated youth, was used by Decker et al. (2008) to investigate seven gang organizational features. Both studies are interested in the relationship of the level of street gang organization to delinquent offending and/or victimization in community settings. They report mixed results with the stronger relationship reported by Decker et al. (2008). Weerman and Esbensen (2005) compare six gang organizational characteristics in the Netherlands with those of the same US sample of gang members reported in Pyrooz et al. (Chap. 6, this volume). They find marked differences in the two countries, with US gangs generally bearing markers of more organization. Decker et al. (1998) compare eight gang characteristics described by members of four gangs from Chicago and San Diego, revealing marked differences between the two cities. Each of these studies utilizes the responses of individual, self-identified gang members to depict gang organization; none of these studies makes a comparison of different types of gangs. Finally, Pyrooz et al. (2011) update the previous review of prison gang literature by Fleisher and Decker (2001). Pyrooz and colleagues (2011, see Table 1) list comparisons between street and prison gangs along ten dimensions, although they do not provide specific data to support the depictions. While acknowledging some similarities, they note: It can be argued that prison gangs are more controlled, organized versions of street gangs. Prison gangs exhibit higher levels of racial and ethnic homogeneity, instrumental violence, entrepreneurial offending, covert behaviors and actions, collective drug dealing, and unqualified loyalty to the gang (Pyrooz et al. 2011, p. 13).

My goal is to expand the literature on gangs in carceral environments by describing organizational characteristics of groups in youth correctional institutions and considering whether they fit more closely with depictions of street gangs or prison gangs.

7.2

California’s Youth Correctional Facilities as a Research Context

This study is situated in a policy environment that, on the one hand, created the circumstances that made it possible to conduct the study, and on the other, may render the findings as applicable only to this time and place.2 In 2004, the Department of Corrections entered into a consent decree to settle the Farrell v Allen (also Farrell v Cate) lawsuit. This required the California Youth Authority (CYA) to file remedial plans to address all areas of deficiency identified by the court experts, including Safety and Welfare (see Krisberg 2011). In July of 2006, following a major reorganization in which the state correctional agency became the California Department of Corrections and Rehabilitation (CDCR) and CYA became the Division of Juvenile Justice (DJJ), the Safety and Welfare remedial plan was accepted by the court. The

110

C.L. Maxson

plan recognized high levels of gang activity and violence in DJJ facilities and directed DJJ to consult with nationally recognized experts to design strategies to reduce gang and racial violence. In late 2009, I was invited to conduct a study that would gather and collate data to inform the development of gang prevention and intervention programs. By the time the research team began collecting data toward the end of summer 2010, it was abundantly clear that DJJ was an agency in transition. One facility staff member boldly stated: “We’re an organization in decline.” Certainly, the numbers of youth housed in DJJ have been declining for several years, likely due to a combination of lower crime rates, higher costs, and changing attitudes toward incarcerating youth in state-level facilities. From 1999 to 2009 the population counts in DJJ decreased from 7,761 to 1,602 (Office of Research 2010). The nature of the population placed in DJJ facilities changed as well. Historically, youth who committed their crimes before the age of 18 and were found by local juvenile court judges to not be amenable to the services available on the county level were remanded to the jurisdiction of the CYA/DJJ for a period of time determined by the correctional agency, but not to exceed age 25. Youth transferred to, and convicted by, the adult court could be housed in CYA/DJJ facilities until age 25. Since 2007, only youth convicted of violent or serious felonies eligible for transfer to adult court, as well as sex crimes for which registration is required, can be committed by the juvenile court to DJJ facilities (Office of Research 2010). A later policy change requires all adult court commitments to be transferred to a CDCR facility (state prison or fire camp) at age 18. Lower numbers, budget pressures, and aging structures led to closures of half of the facilities that were operating at the turn of the century. Another facility that was included in our study closed 9 months after we completed data collection, leaving just four facilities across the state. Facility closure requires relocating youth (and some staff), inevitably causes disruption in programming, creates constraints on housing choices, and increases tension between youth as social (and gang) relationships are resolved. Currently, two facilities are located in northern California, in the city of Stockton. The recently closed facility is about 30 miles away. The southern portion of the state has two facilities, and one houses all females committed to DJJ (between 50 and 60 young women) as well as males. One of these facilities is scheduled to close by the end of 2011. A southern California facility that housed older youth and parole violators had been closed about 6 months before the onset of our data collection. We obtained population lists of youth to interview during September 2010, when about 1,200 youth were housed in five facilities. Even while these dramatic changes in the DJJ population were occurring, the Farrell lawsuit required massive changes in service delivery and the conditions of confinement. There are annual or semiannual audits of staff and youth to ensure progress. This means that the facilities are under constant scrutiny. In fact, along with the usual assurances of confidentiality and research protections, we began our interviews with the disclaimer that we were not law enforcement or social workers, lawyers, or auditors! Thus, we conducted this study during a period of rapid institutional change and staff job insecurity. Staff perspectives on

7


111

safety and violence issues most certainly reflect this and youth perceptions are affected as well.

7.3

Can the Eurogang Research Paradigm Be Applied to Correctional Settings?

Developed over a series of workshops, the Eurogang research paradigm incorporates several core tenets that distinguish it from less complex approaches (Maxson 2001). As most recently articulated in the Eurogang manual (Weerman et al. 2009), Eurogang proselytizes a multimethod, multisite research approach that encourages strict adherence to the Eurogang definition of gangs and application of consistent measures (see Chap. 1, Esbensen and Maxson, this volume). The five core research instruments span multiple levels of analysis (individual, group, program, neighborhood, and city) and are explicitly based in community settings. I wanted to explore whether this paradigm might be useful in guiding gang research in youth correctional institutions. What types of adaptation of concepts and methods might be necessary? Is the Eurogang paradigm flexible enough to be stretched into this environment or is it so rigid it will crack under the weight of its community, street gang origins? The research design adopted for this study captures several aspects of the Eurogang paradigm, but not with a perfect fit. Eurogang promotes comparative, multisite research, and in particular, aims to foster cross-national research (see Chap. 17, Klein, this volume). The research reported here is not cross-national and, further, is limited to correctional institutions within just one state, California. Comparable methods were pursued in all five facilities. All facilities are maintained by the DJJ within the CDCR, and thus are subject to statewide institutional policies, training, and administration, but local cultures, histories, and management promote different institutional practices. Moreover, geographic location and specialized services provided at particular facilities suggest that the youth populations would vary considerably, as would the gang dynamics. Therefore, this study most closely approximates a study of multiple neighborhoods or communities (i.e. individual facilities) within a particular city (i.e. state-level DJJ policy) rather than independent sites (for an example in this volume, see Chap. 3, Aldridge, Medina, and Ralphs, this volume). I expected to see variation in gang profiles within each facility and, given different youth populations and administrative procedures, that these would vary across facility. However, considering this is one state and one correctional system, I cannot generalize findings to other US states or to other countries. Thus, the Eurogang approach fostered strict adherence to methods from one site to the next, but I freely acknowledge the limitation of situating the research within just one state. Consistent with the Eurogang paradigm, the research design draws from multiple methodologies that incorporate both qualitative and quantitative research techniques. Central to this chapter is the incorporation of the Eurogang Expert Survey items into the more qualitative staff interviews I conducted with six DJJ

112

C.L. Maxson

Gang Information Coordinators (GICs). The primary role of the GICs is to assess, monitor, and investigate gang issues and problems in each facility. Except for one person that acted as a temporary replacement (but with 1 year’s prior experience as a GIC), the GICs had been employed in this position for 5–10 years. Their period of employment with DJJ ranged from 14 to 29 years. These interviews covered topics of safety and violence in institutions, as well as policies and programs best suited to address these issues. Altogether, the research team interviewed 64 staff members for an average of just under 45 min. These interviews were conducted in confidential settings and most were recorded, with the permission of the staff member. The Eurogang Expert Survey items were also incorporated into a structured interview which we conducted with a representative sample of about 350 youth residents in the five correctional facilities. Many of the questions on our youth interview instrument were similar to those included on the Eurogang Youth Survey. The study benefits from a wealth of data provided by DJJ, pertaining to individuals that we interviewed as well as the population from which we drew the sample. This type of data falls within the category of archival data, a methodological approach that has received only scant attention from Eurogang (Weerman et al. 2009). I have also been provided with institutional data and reports that fulfill many of the data requirements expressed in the City-level instrument. The research team has gathered information about programs and services from youth and staff, has been provided institutional service data, and has collected official program descriptions and assessment reports. Thus, we have approximated the spirit, if not the letter, of the Eurogang Prevention and Intervention Inventory. Finally, our field notes from site visits, informal conversations and observations, while not gathered under the protocol specified by the Eurogang Ethnographic Guidelines, nevertheless provide a rich source of qualitative data. In summary, the correctional research setting does not appear to constrain the application of the Eurogang tenets and instruments. On the contrary, the approach was quite useful in guiding the development and implementation of this study research design.

7.4

Does the Eurogang Definition Fit the Institutional Context?

A fundamental requirement of participation in Eurogang research is adherence to the Eurogang definition: A street gang (or troublesome youth group) is any durable, street-oriented youth group whose involvement in illegal activity is part of their group identity. All the definitional elements were adaptable to the groups we observed in youth correctional facilities save one: applying the “street-oriented” criterion to institutional settings where the ability to congregate or hang out in public places, as well as to engage in other routine activities, is tightly controlled by correctional officers. According to the Eurogang definitional statement (Weerman et al. 2009, p. 20):

7


113

The element of “street-oriented” implies spending a lot of group time outside home, work, and school. It is not necessary that the group always be found on a street location. Streetoriented groups may also meet in malls, in parks, in cars, and so on. We regard groups as street-oriented when they hang out predominantly in public places without the supervision of adults.

Youth in correctional facilities are rarely beyond the supervision of adults, yet our field observations in living units, recreation areas, and other semipublic spaces suggest that youth regularly hang out together, and in some very visible ways. The definitional statement used in the DJJ study to solicit gang profiles from staff replaced “street-oriented” with “visible” and defined the latter as “spending a lot of group time—often in recreational areas, day rooms, and so on.” The staff survey item capturing this visible hanging out time caused so much confusion among the respondents that I often skipped it. I was informed that routine activities, such as eating, recreation, and treatment programs, typically are structured by living unit, leaving little opportunity for members of a group assigned to different living units to spend time together socializing. School and job locations are more integrated but these are heavily supervised. When asked where members of a particular gang spend most of their time together, one staff members replied “Muslim services” [15–31]. This is hardly the street-oriented aspect of the Eurogang definition that I was trying to capture! It appears that only members of the same gang assigned to the same living unit—a circumstance that DJJ personnel strive to minimize—have much opportunity to hang out together informally.3 As an aside, territoriality also has a different meaning in this context. Institutional gangs are not free to claim living units or areas within the facility as their territory that is off-limits to others. However, it is quite common for certain corners, tables, or seats to be understood by youth to be a particular gang’s exclusive area. There was some variation in responses to this series of questions, so I include this item among other gang characteristics I explored. In addition to these instrument changes, I extended the Eurogang Expert Survey items that asked about involvement in illegal activity to include institutional rule infractions. An awkward moment arose in interviews with both staff and youth with the item that was designed by Eurogang to capture the element of illegal activity as a component of group identity: “Is doing illegal things < or breaking DJJ rules > accepted by, or ‘okay,’ for this group?” Often youth would react quizzically to this item, as if the answer was quite obvious and researchers were asking a really dumb question. One DJJ GIC reacted quite vehemently: “Don’t insult me” [12–31]. The requirement of illegal activity is a controversial dimension of gang definitions among scholars (Short 1996; Klein 1995), but it was quite clear from our youth and staff experts that from their point of view, involvement in criminal activity is an intrinsic element of gang affiliation. With these exceptions, the specific definers of gangs in the Eurogang Expert Survey and the Youth Survey translated easily into the institutional context.4 The Eurogang approach appears to be robust and resilient to the constraints of this very different setting. I now turn to the results of the Expert Survey to consider whether these institutional gangs fall within the definition of Eurogang, and related, whether these gangs evidence the structural types identified among street gangs.

114

7.5

C.L. Maxson

Are Institutional Gangs Eurogangs?

The DJJ identifies eight institutional gang affiliations, in addition to tagger groups and an “other” category.5 Virtually every individual that has a street gang affiliation also is designated with an institutional affiliation. The perception of some GICs is that institutional affiliation overwhelms street gang identity: “Usually when they come in, they don’t keep their street alliance” [11–31].6 In other views, the connection between street and institutional gang is isomorphic: when you join a street gang, you automatically affiliate with a specific institutional gang. In September 2010 when we drew our youth interview sample, 72% of the 1,153 male youth7 in DJJ had an official gang designation. Table 7.1 lists the distribution of institutional gang affiliations as they appear in DJJ records. Reflecting the racial/ ethnic mix of the DJJ population, the most common affiliations are with primarily Hispanic gangs. Forty percent of the gang members affiliate with the Southerners (Surenos) and 19% as Northerners (Norteno). The third most commonly used institutional designation is Crips (16% of the gang members), an amalgam of African American street gangs with this identity. The other institutional affiliations comprise less than 10% of the gang population: 415 (or Bays, incorporating primarily African American gang members from the San Francisco bay area), Bloods (primarily African American), Bulldogs (a Hispanic breakaway from the Northerners), Asians, Supreme White Power/Peckerwoods, taggers, and other. The gang profiles provided by the GICs included just those groups that are present in the facility to which the GIC was assigned.8 The number of profiles provided by each GIC varied from two to eight. As shown in Table 7.1, Southerners is the only institutional gang that was profiled in all five facilities. The Bulldogs and white supremacist gangs were identified by GICs in just one or two facilities. The “other” gang listed in Table 7.1 is “African American” which one GIC opted to profile instead of separate responses for Crips and Bloods. The GIC stated that occasionally Crips or Bloods

Table 7.1 Distribution of institutional gang designation among DJJ male youth and distribution of gang profiles by facility DJJ gang designation # of facilities Gang affiliation No of youth % of gang members with gang profiled Asian 15 2 3 Blood 52 6 4 Crip 135 16 4 Bulldog 20 2 2 415/Bays 76 9 3 Northerner/Norteno 155 19 3 Southerner/Sureno 331 40 5 Tagger group 11 1 0 SWP/Peckerwood/White Power 14 2 1 Other 15 2 1 Total 824 100 26

7


115

would spend time together in separate groups, but mostly, they identified as one group. When I questioned this lack of distinction between historical street gang rivals, this GIC framed institutional gang affiliation in racial terms9: GIC: Right now, my issues are black/brown, in this facility. I: And you don’t distinguish between Crips and Bloods? You think of them as one group? GIC: Well, when we have issues amongst them, then I will look it up and figure out if they’re rivals, what their issues are, where they are from …. As far as right now, we’re not having those issues. [15–31] In order to determine whether DJJ’s institutional gangs might appropriately be considered within the Eurogang definitional framework, I tallied the expert survey items that operationalize the definitional criteria of a durable (existed at least 3 months), street-oriented (members spend a lot of time together in visible places) youth (most members are 12–25 years) group (more than two members) whose involvement in illegal activity (they do illegal things or break rules together) is part of their group identity (illegal activity is accepted by, or okay, for this group). All 26 groups met each of these criteria, with the exception of street orientation or visible hanging out, for reasons I explained earlier. Interestingly, there is no variation among the gangs in duration. Every GIC reported that each gang had been in existence in the facility for more than 20 years. Typically, they responded that the gang had been there since the facility opened, or “forever.” Gangs in DJJ are clearly a chronic rather than emergent problem (Spergel and Curry 1993) and certainly meet the 3-month threshold of Eurogang. The modal age of gang members in particular groups reflects DJJ housing practices insofar as one facility is the preferred placement for younger boys. The GIC in this facility chose 16–18 years as the modal age for all six gangs profiled. With the exception of this facility, the GICs specified the older adolescent category of 16–18 years about half the time and the young adult range of 19–25 years about half the time. There was no apparent pattern of modal age within institutional gang across facility. It should be noted that the mean age of males placed in DJJ at this time was just over 19 years. The Eurogang definitional procedure notes that the attribution of “youth” can be ambiguous, with street gang membership sometimes extending into the twenties or thirties, but most members should be in their teens or early twenties (Weerman et al. 2009). Accordingly, all DJJ institutional gangs meet the Eurogang age criterion. The DJJ gang database provides a count of gang members in each facility that ranges from about 150 to just over 200 (DJJ Gang Operations Unit 2010). GICs sometimes estimated the size of each group in their facility and sometimes checked the database during the interview. The modal size category, selected for 11 of the 26 gangs, was 21–50 members; 11 gangs had 20 or fewer members (but more than two), and three gangs had 51–100 members. Just one gang was placed in the over 100 size category. The larger gangs were concentrated in facilities located in southern California and, reflecting the distribution reported in Table 7.1, three of the four large gangs were Southerners.

116

C.L. Maxson

Unfortunately, one GIC was uncooperative in responding to questions regarding illegal activity among members of the gangs he profiled. The other GICs indicated that each gang in their facility (a) did illegal things and/or broke DJJ rules together and (b) that doing so was accepted by, or okay, for each group. Since there is no evidence that the groups in the recalcitrant CIG’s facility are different from those in the rest of DJJ, it appears safe to assume that these groups fall within the final Eurogang criterion of a crime-imbued, group identity. In addition to self-identification as a member of a gang, Eurogang permits one other path to gang designation (Weerman et al. 2009). Maxson and Klein (1995) proposed a structural typology of street gangs that is useful to categorizing gangs in the US and Europe (Weitekamp 2001). The typology takes into account gang size, age range, duration, territoriality, subgroup structure, and crime versatility. If an institutional gang manifests the same combination of these characteristics as one of the five structural gang types, then it could arguably be designated as falling within the Eurogang definition. However, the situational constraints imposed by the DJJ setting make the application of some of the typology characteristics inappropriate. The size of each gang in each facility is determined somewhat by DJJ placement decisions. The possible age range is truncated by statute and justice system policy, so most or all institutional gangs would fall in the narrow age range category (defined in the typology as less than 10 years). Furthermore, all of the DJJ institutional gangs have been in existence for more than 20 years, the long-duration category. Just one structural type, the traditional gang, has long duration. Other features of the traditional gang, large size (more than 100 members), and wide age range (20–30 years) preclude its applicability to the groups in DJJ settings. While I detected variation in subgroup structure, territoriality, and crime versatility among institutional gangs, it would appear that the Maxson/Klein typology is not amenable to categorizing gangs in the DJJ, at least not without a great deal of adaptation. So, how do we answer the question of whether institutional gangs are appropriately considered under the Eurogang umbrella as gangs? On all counts save street orientation, DJJ institutional gangs appear to pass the test of the Eurogang definitional criteria. I have argued that the lack of street orientation is a function of structural constraints imposed by this particular setting rather than the proclivities of gang members, but those that argue for a strict adherence to the criteria would likely exclude institutional gangs from the Eurogang sphere (see Chap. 17, Klein, this volume). Public visibility is likely important to gang dynamics such as intimidation, reinforcement of group identity and cohesion, and to establishing conflict-laden relations between street gangs and other community residents. The degree to which the constraints against hanging out impact on group identity among these young men is not clear. It would be useful to measure and compare level of group orientation (or what Chap. 8, Hennigan and Spanovic, this volume, refer to as “entitativity” or “groupness”) in institutional gang members and street gang members. It may be that institutional gang members find other paths to sustaining group identity than street orientation. Until such research can enlighten us regarding the similarity in substance if not strict adherence to the criteria, it would appear that institutional gangs, at least in California, do not fit the criteria. Ironically, in carceral environ-

7


117

ments that are less able, or willing, to restrain the interactional patterns of residents, the Eurogang criteria may apply.

7.6

Are Institutional Gangs More Similar to Prison Gangs or Street Gangs?

The foregoing analyses suggest that the institutional gangs in my study share important defining elements with street gangs. The research and practitioner literature ascribes a much higher level of organization to prison gangs than is typically observed among street gangs (Fleisher and Decker 2001; Pyrooz et al. 2011), but rarely addresses gangs in youth correctional facilities. The gang profiles offer the opportunity to examine organizational features of gangs in DJJ to see whether the characteristics of such groups resemble depictions of street gangs more than prison gangs. Typical indicators of higher levels of organization or structure included in the profiles are recognizable leadership, distinguishing group symbols (i.e., clothing, ways of speaking, or other physical identifiers), selective entry or initiation rites (i.e., “have to do special things to get into this group”), and rules that members are supposed to follow. Although the DJJ setting would place constraints on some aspects of gang activity, I also include subgroup structure (“distinct subgroups or cliques that especially spend time together”), territoriality (“have an area or place it calls its own”), and crime specialization (“do a lot of different kinds of illegal things together [in DJJ facilities] or, just a few kinds”). Finally, use of illegal drugs (sometimes or often) might suggest the organizational ability to access illegal goods, getting into fights with other groups (sometimes or often) suggests a level of “groupness” or defense of group, and ethnic/racial exclusivity might indicate an ability to control access to group membership.10 Since the prison gang literature is lacking an empirical assessment of gang organization, I approach this analysis by considering whether or not the organizational features of institutional gangs, as described by the GICs, resonate with my understanding of street gangs, based upon my familiarity with the research literature (see also, Klein and Maxson 2006; Chap. 6, Pyrooz et al., this volume). The distribution of these ten indicators of organization within these 26 gangs profiled by the GICs is displayed in Table 7.2. Use of illegal drugs, having distinguishing group symbols, having rules that members are expected to follow, and fighting with other groups are common indicators in these groups, surfacing in twothirds or more of these gangs. All four features are also common among US street gangs. Typical examples of rules offered often included backing up other members of their gang or other alignments, do not be a snitch, and the requirement to respond to disrespectful behavior; these are common expectations of street gang members as well. Some of the rules reflected institutional cultural norms that dictate race relations (do not drink or eat behind blacks). In a few cases, detailed expectations were listed, such as working out, getting up in the morning, rolling up your mattress a certain way, attending a roll call, and marching to cadence. One gang was described

5

19 14

4

Crime specialization (n = 21)

Use illegal drugs (n = 20) Fights other groups (n = 20)

Ethnicity/race exclusive (n = 26)

15

95 70

24 None All Hispanic gangs fight; less common in black gangs None

More common in Asians

Table 7.2 Organizational indicators in 26 institutional gangs Gang profiles Patterns by gang or race/ Organizational indicator N % ethnicity Recognized leaders (n = 26) 16 62 Somewhat more in Hispanic than black gangs Group symbols (n = 26) 21 81 No symbols in 415 gang; less common in Asians Entry criteria (n = 22) 4 18 All gangs with criteria are Hispanic Rules (n = 21) 17 81 None Subgroups (n = 22) 13 59 None Territoriality (n = 20) 5 25 None

¾ exclusive gangs in one northern facility

All gangs without symbols are in northern facilities All gangs with criteria are in northern facilities ¾ with no rules in one northern facility None All gangs with territoriality in southern facilities 3/5 with specialization in one southern facility None None

Patterns by facility location More often in northern facilities

No

No No

No

No No Yes, low for street gangs

Yes, low for street gangs

No

Levels inconsistent with street gang? No

118 C.L. Maxson

7


119

as forming “so they didn’t have to follow the rules [of the group from which they broke away]. They didn’t want structure; they’re more into lounging” [13–31]. Another gang was described as “creating rules as they go along” [15–31]. Race or ethnic exclusivity among membership, having to do special things to get into the group, doing just a few types of illegal activities, and territoriality occur in less than one-third of gangs. Institutional gangs do not appear to limit their membership by imposing stringent entry criteria or racial exclusivity. This contrasts with street gangs, where, for example, one multisite study found that 80% of US gangs members reported having to do something special to join their street gangs (Weerman and Esbensen 2005). The low rate of perceived institutional gang entry restrictions may reflect the isomorphism with which the GICs view street and institutional gang affiliation: They come here already part of it. [15–31] They get “crimed in” on the street. [13–31] Lack of territoriality is perhaps not a good indicator of institutional gang organization for the reasons I mentioned earlier, but the frequency of territorial gangs in DJJ appears low in comparison to street gangs. The reliability of the crime specialization measure may be suspect as when I asked for examples of typical illegal activities, most GICs gave three examples, whether or not they reported a lot of different kinds or just a few kinds. The most common answers were drug and/or alcohol use, obtaining contraband (like cell phones and smoking devices), and fighting. If we trust this measure, the low levels of crime specialization reported for institutional gangs would also be typical of street gang membership. The mid-range levels of subgroup structures and recognizable leaders are typical of street gangs. The subgroup question often stimulated the clarification that cliques formed among members of separate street gangs within the larger institutional gang structure. GICs stated that leaders are “representatives” or “guys of influence” and suggested these would be selected based on age, maturity, longevity, or physical size. Sometimes, they would clarify their responses to indicate that leadership is not necessarily indicative of high levels of organization. One GIC explained that “it’s not paramilitary” and for another gang, “there is no organizational chart” [13–31]. The low numbers of profiles on each gang make it difficult to identify patterns or distinctions among them. Despite high levels of symbols among institutional gangs, all three profiles of the 415/Bays denied that they had them as did two of the three Asians profiles. Crime specialization was indicated on both Asians profiles that responded to this question. Combining individual gangs by the race/ethnicity of most of their members revealed a few patterns: Hispanic gangs were somewhat more likely to have recognized leaders than black gangs (8/10 Hispanic gang profiles; 6/12 black gang profiles) and fighting with other groups was less common among black gangs than Hispanic gangs (8/8 Hispanic gang profiles; 5/9 black gang profiles). All four gang profiles that indicated entry criteria were Hispanic gangs, all located in northern California facilities. Northern facilities also more often generated profiles of gangs with recognized leaders (12/20 northern facility profiles; 2/6 south-

120

C.L. Maxson

ern facility profiles). All five gangs without group symbols were in northern facilities whereas all gangs with territoriality were in southern facilities. This offers some evidence of different gang cultures in the two halves of the state. I identified a pattern of gang characteristics in just one facility on three occasions. While this might reflect localized features of institutional gangs, it could also be the tendency of individual GICs to report certain characteristics, so I hesitate to expand on these.11 In general, the features of group organization resonate with depictions of street gangs in the research literature (see Klein and Maxson 2006; Chap. 6, Pyrooz et al., this volume). I see little evidence of the prison gang characteristics detailed by Pyrooz and his colleagues (Pyrooz et al. 2011), such as race exclusivity, hierarchical or rigid organizational structure, absolute loyalty to gang, and unqualified fidelity as a key to membership in either these gang profiles or the qualitative data from interviews with other staff members.

7.7

Discussion

In this chapter, I show that the Eurogang research paradigm is useful in guiding research on gangs in correctional institutions, even though it was developed for gathering data in communities. Most of the methodological instruments developed by Eurogang could be adapted to this new setting, and the core principles embraced by Eurogang certainly are applicable. The institutional setting required some minor changes in wording of the Expert Survey. I find that gangs in youth institutions meet all the Eurogang definitional requirements except for street orientation and that their organizational structures look much more like street gangs than what we know of prison gangs. The issue of street orientation or visible hanging out as a defining characteristic of gangs deserves increased scrutiny. Other chapters in this volume (Chap. 3, Aldridge et al.; Chap. 14, Pederson and Lingstad; Chap. 18, Maxson and Esbensen) address this concern and my work adds the increased complexity of structural constraints. Aldridge and her colleagues as well as Pederson and Lindstad allude to increased law enforcement activity or community actions as a catalyst for moving groups of gang members out of public arenas. What impact might that have on group interactions and the social processes that serve to develop group identity? The groups in DJJ are undeniably gangs; their group organizational characteristics approximate that of street gangs and their members clearly see themselves as part of a gang. Perhaps the DJJ institutional gang members do not need as much hang time to maintain their group identity. Sela Shayovitz (Chap. 12, this volume) introduces interaction on the Internet as a source of reinforcement of group identity for gang members. Might this serve as an alternative to street time? More attention should be accorded to that which binds these individuals together in a range of settings, including carceral environments. The finding that institutional gangs share organizational features with street gangs rather than prison gangs also should be investigated further. Since the literature

7


121

on prison gang characteristics lacks systematic empirical assessment, it may be that the organized nature of prison gangs has been overstated. Also, my study used a small number of expert informants to describe institutional gangs whereas the comparable street gang studies build upon many gang member reports. My confidence in my findings is increased by a preliminary comparison of the depictions provided in our youth interviews. As reported in Maxson et al. (2011), 107 institutional gang members provided gang profiles. Gang members’ aggregated organizational depictions of their gangs are nearly identical to that provided by GICs for two characteristics [has group symbols (79%) and entry criteria (19%)], are slightly lower for territoriality (17%), and are quite a bit lower for recognized gang leaders (39%), rules (64%), and subgroups (32%). We are in the process of conducting direct comparisons for particular gangs for a forthcoming paper, but the pattern of lower than expected organizational qualities in institutional gangs is supported. The finding that institutional gangs appear to approximate street gangs has important implications for intervention programs. While there is little advice we can offer regarding proven, effective interventions for street gang members (Klein and Maxson 2006), as positive evaluations emerge we should find them promising for tests among institutional gang populations. Moreover, the street gang research literature is rife with lessons learned from failed programs. We can productively draw from this evidence to help guide us in what not do with institutional gang members. There is no evidence to suggest that desistance from institutional gangs is more difficult than exiting street gangs, although the lack of prosocial peer networks creates a special challenge for institutionalized youth. Finally, it is important to recognize that these findings emerge from one carceral environment—DJJ youth correctional facilities—in one state—California—that is arguably the epicenter of gang activity in the US, and beyond. Moreover, this state is undergoing enormous change in juvenile justice policy. Increasingly, youth such as those involved in our study will be incarcerated in local, county-supported facilities such as juvenile halls and probation camps. These local facilities presumably evidence very different gang dynamics and the findings from our study may not translate well into these other settings. Ironically, because California recently has been so restrictive as to the seriousness of youth that can be incarcerated in DJJ, our findings may be more appropriate to European and US locations that tend to incarcerate only the worst of the worse juvenile offenders.

Notes 1. This study was conducted with the support of funding from the California Department of Corrections and Rehabilitation (CDCR), Agreement #5600001351. The opinions expressed herein represent those of the author and do not necessarily represent the position of the CDCR. The author is grateful for numerous contributions from the study team which includes Charlotte Bradstreet, Danny Gascon, Julie Gerlinger, Jessica Grebenkemper, Darin Haerle, Jacob KangBrown, Analicia Mejia-Mesinas, Randy Myers, Marisa Omori, Shannon Reid, and Daniel Scott.

122

C.L. Maxson

2. Portions of this section were previously drafted by the author and included in an unpublished technical report to the funding agency (Maxson et al. 2011). 3. The most aggressive youth who get into regular conflicts are placed on a special unit with more constraints on where they spend their free time, and with whom. Given the constraints imposed on patterns of interaction, and therefore, the group process that takes place among gang members, I considered whether gang members within a living unit might best be considered sets or cliques of a gang, comparable to a neighborhood-based clique. However, these living unit assignments are changed quite frequently, particularly during this period when the DJJ population is shrinking and DJJ has had to reconfigure special-needs units on an ongoing basis. 4. Respondents were asked to consider only gang activities that occurred in the specific facility, as I treated each facility as a “neighborhood.” 5. According to DJJ policy, each youth entering a facility is interviewed by a Gang Information Coordinator regarding gang affiliation within a few days of arrival. The youth completes a disclosure form describing their gang. The GIC inspects each youth for gang tattoos and records them. This information is coupled with information in the youth’s file to make a determination regarding gang status. The GICs have regular contact with youth throughout their stay in the facility, according to our interviews with personnel in the Gang Operations Unit (DJJ headquarters) and with each of the GICs. While a youth that initially denies gang membership might later be designated as such, there is no indication that DJJ ever removes gang affiliation or reclassifies an individual as an inactive, or former, gang member. 6. Although not the topic of this chapter, it should be noted that our interviews with youth present a far more complicated picture of street and institutional gang affiliation, including selfreports of street gang membership (current or former) but no institutional affiliation and vice versa. 7. There were only 59 females placed in DJJ when we drew our sample and 19% of them are designated as gang members. The GIC in the facility that houses all females and some males rarely referred to female gang activity during the staff interview. Therefore, the analyses reported here focuses exclusively on males. 8. I interviewed all six GICs that were assigned to the five DJJ facilities. I compared the responses of the two GICs assigned to the same facility. One GIC completed a separate profile on the Asians while the other included them as Crips. There were only minor differences in responses and ultimately I decided to use the second interview I conducted because I used an earlier version of the instrument in the first. The conclusions derived from the study are not influenced by this decision. 9. A different GIC profiled one other group, the Juggalos, as an institutional gang, but since this group was limited to just two members, I exclude it from this analysis. 10. There were no profiles that indicated a gang “never” used illegal drugs or got into fights with other groups like it. 11. I looked for reporting patterns among individual GICs, and for the most part, they offered varied depictions of the different gangs within their facilities.

References Benda B, Corwyn R, Toombs N (2001) Recidivism among adolescent serious offenders: prediction of entry into the correctional system for adults. Crim Justice Behav 28(5):588–613 Caudill JW (2010) Back on the swagger: institutional release and recidivism timing among gang affiliates. Youth Violence and Juvenile Justice 8(1):58–70 Davis MS, Flannery DJ (2001) The institutional treatment of gang members. Correct Manag Q 5(1):37–46 Decker SH, Bynum T, Weisel D (1998) A tale of two cities: gangs as organized crime groups. Justice Q 15:395–425

7


123

Decker SH, Katz CM, Webb VJ (2008) Understanding the black box of gang organization: implications for involvement in violent crime, drug sales, and violent victimization. Crime Delinquency 54:153–172 DeLisi M, Berg MT, Hochstetler A (2004) Gang members, career criminals and prison violence: further specification of the importation model of inmate behavior. Crim Justice Stud 17(4):369–383 DeLisi M, Drury AJ, Kosloski AE, Caudill JW, Conis PJ, Anderson CA, Vaughn MG, Beaver KM (2010) The cycle of violence behind bars: traumatization and institutional misconduct among juvenile delinquents in confinement. Youth Violence and Juvenile Justice 8(2):107–121 DiPlacido C, Simon TL, Witte TD, Gu D, Wong SC (2006) Treatment of gang members can reduce recidivism and institutional misconduct. Law Hum Behav 30:93–114 Division of Juvenile Justice, Gang Operations Unit (2010) Gang overview. Presentation to the UCI research team. Division of Juvenile Justice, California Department of Corrections and Rehabilitation, Sacramento Drury A, DeLisi M (2011) Gangkill: an exploratory empirical assessment of gang membership, homicide offending, and prison misconduct. Crime Delinquency 57(1):130–146 Fleisher MS, Decker SH (2001) An overview of the challenge of prison gangs. Correct Manag Q 5:1–9 Gaes GG, Wallace S, Gilman E, Klein-Saffran J, Suppa S (2002) The influence of prison gang affiliation on violence and other misconduct. The Prison J 82:359–385 Griffin ML, Hepburn JR (2006) The effect of gang affiliation on violent misconduct among inmates during the early years of confinement. Crim Justice Behav 33(4):419–448 Howell JC, Roush DW (1997) Youth gang problems in juvenile detention and corrections facilities. J Juvenile Justice and Delinquency Services 12:53–61 Huebner BM (2003) Administrative determinants of inmate violence: a multi-level analysis. J Crim Justice 31:185–208 Huebner BM, Varano SP, Bynum TS (2007) Gangs, guns, and drugs: recidivism among serious, young offenders. Criminology and Public Policy 6(2):187–222 Hunt G, Riegel S, Morales T, Waldorf D (1993) Changes in prison culture: prison gangs and the case of the Pepsi generation. Soc Probl 40(3):398–409 Inderbitzin M (2007) Inside a maximum-security juvenile training school: institutional attempts to redefine the American dream and ‘normalize’ incarcerated youth. Punishment & Society 9:235–251 Klein MW (1995) The American street gang: its nature, prevalence, and control. Oxford University Press, Oxford Klein MW, Maxson C (2006) Street gang patterns and policies. Oxford University Press, New York Krienert JL, Fleisher MS (2001) Gang membership as a proxy for social deficiencies: a study of Nebraska inmates. Correct Manag Q 3:47–58 Krisberg B (2011) The long and winding road: juvenile corrections reform in California. The Chief Justice Earl Warren Institute on Law and Social Policy, Berkeley Kuanliang A, Sorenson JR, Cunningham MD (2008) Juvenile inmates in an adult prison system: rates of disciplinary misconduct and violence. Crim Justice Behav 35(9):1186–1201 Lattimore PK, MacDonald JM, Piquero A, Linster RL, Visher CA (2004) Studying the characteristics of arrest frequency among paroled youthful offenders. J Res Crime Delinq 41:37–57 MacDonald J (1999) Violence and drug use in juvenile institutions. J Crim Justice 27:33–44 Maxson CL (2001) A proposal for multi-site study of European gangs and youth groups. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Dordrecht Maxson CL, Klein MW (1995) Investigating gang structures. J Gang Res 3:33–40 Maxson CL, Bradstreet C, Gascon D, Gerlinger J, Grebenkemper J, Haerle D, Kang-Brown J, Omori M, Reid S, Scott D (2011) Developing effective gang policies for California’s division of juvenile justice: preliminary findings draft report. University of California, Irvine McGloin JM (2007) The continued relevance of gang membership. Criminology and Public Policy 6(2):231–240

124

C.L. Maxson

Office of Research, Juvenile Justice Research Branch (2010) 2010 Juvenile justice outcome evaluation report. California Department of Corrections and Rehabilitation, Sacramento Parker RN, Negola T, Haapanen R, Miranda L, Asencio E (2008) Treating gang-involved offenders. In: Hoge RD, Guerra NG, Boxer P (eds) Treating the juvenile offender. Guilford, New York Pyrooz DC, Decker SH, Fleisher M (2011) From the street to the prison, from the prison to the street: understanding and responding to prison gangs. J Aggress Confl Peace Res 3(1):12–24 Rudell R, Decker SH, Egley A Jr (2006) Gang interventions in jails: a national analysis. Crim Justice Rev 31:1–14 Short JF (1996) Gangs and adolescent violence. University of Colorado, Center for the Study and Prevention of Violence, Boulder Spergel IA, Curry GD (1993) The National Youth Gang Survey: a research and development process. In: Goldstein AP, Huff CR (eds) The gang intervention handbook. Research Press, Champaign Tasca M, Griffin M, Rodriguez N (2010) The effect of importation and deprivation factors on violent misconduct: an examination of black and latino youth in prison. Youth Violence and Juvenile Justice 8(3):234–249 Trulson CR (2007) Determinants of disruption: institutional misconduct among state-committed delinquents. Youth Violence and Juvenile Justice 5(1):7–34 Trulson CR, Marquart JW, Mullings JL, Caeti TJ (2005) In between adolescence and adulthood: recidivism outcomes of a cohort of state delinquents. Youth Violence and Juvenile Justice 3(4):355–387 Weerman FM, Esbensen F-A (2005) A cross-national comparison of youth gangs: the United States and the Netherlands. In: Decker SH, Weerman FM (eds) European street gangs and troublesome youth groups. AltaMira Press, New York Weerman FM, Maxson C, Esbensen F-A, Aldridge J, Gemert F van, Medina J (2009) Eurogang program manual. Background, development and use of the Eurogang tools in multisite multimethod comparitive research. (www.umsl.edu/~ccj/eurogang/eurogangmanual.htm) Weitekamp EGM (2001) Gangs in Europe: assessments at the millennium. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the US and Europe. Kluwer, Dordrecht Winterdyk J, Ruddell R (2010) Managing prison gangs: results from a survey of U.S. prison systems. J Crim Justice 38:730–736

Part II


Chapter 8

Gang Dynamics Through the Lens of Social Identity Theory Karen Hennigan* and Marija Spanovic

The gang as a form of human association and a social problem, an object of curiosity and commentary is at once ancient and contemporary … reports suggest that collective forms of delinquency have become cause for alarm in such widely separated areas as Western Europe and the Iron Curtain countries, the Far East, in Australia, and in such rapidly changing underdeveloped countries as Ghana and Kenya. Picturesque names establish the public identity of these young people … names like “zoot-suiters” and “boppers” in the U.S., “Teddy boys” in England, “blousons noir” in France, “vitelloni” in Italy, “Halbstarke” in Germany, “bodgies” (boys) and “widgies” (girls) in Australia and New Zealand, “tsotsio” in South Africa, “mambo” boys and girls in Japan, “hooligans” in Poland and Russia, the “tap-karoschi” of Yugoslavia, and the “lui-mang” and “tai-pau” of Taiwan (Short and Strodtbeck 1965, p. 1).

This is the opening paragraph of one of the first detailed studies of group processes observed in street gangs.1 Published in 1965, Short and Strodtbeck’s work documents the ubiquitous nature of delinquent adolescent groups in many parts of the world, now going back five decades or more. Over the years, the descriptors used to depict such groups have changed, but their presence has been continuous and expanding. Efforts to understand the impact of street gangs have been continuous and expanding as well. A street gang is foremost a group. Young men and women all over the world attach importance to becoming part of such a group, gaining part of their identity from belonging to it. Relatively little academic and

*

The data used for this work was supported by the National Science Foundation under Grant No. 0550228, Karen Hennigan principal investigator and David Sloane co-principal investigator. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation (NSF). The authors wish to thank Malcolm Klein, David Sloane, Cheryl Maxson, and Finn-Aage Esbensen for their encouragement and helpful comments on this work. This chapter is based in part on a presentation given at Eurogang Workshop X in Neustadt, Germany, in July 2010.

K. Hennigan (*) • M. Spanovic Department of Psychology, University of Southern California, Los Angeles, CA, USA e-mail: [email protected] F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_8, © Springer Science+Business Media, LLC 2012

127

128

K. Hennigan and M. Spanovic

professional work on street gangs has taken advantage of the development of social psychological theories on social identity to understand how the street gang as a group influences its members. The purpose of this chapter is to use the framework of social identity theory (Tajfel 1978; Tajfel and Turner 1979, 1986), self-categorization theory (Turner 1985, 1987), and related work (Brewer 1991; Hogg 2001; Pickett et al. 2002; Postmes and Spears 1998) to enrich our understanding of group processes in adolescent street gangs. The hope is to stimulate further exploration of gang group processes through the lens of these theories that have been developed over the last several decades by social psychologists in many parts of the world, especially in Europe, the United States, and Australia. We argue that this particular framework, mostly ignored in recent gang research and applied practice, holds promise for furthering our understanding of the dynamics of these groups and may also bring new insights to the work of formulating effective prevention and intervention strategies that heretofore have had disappointingly little impact on the continuity and growth of street gangs (see Klein and Maxson 2006, Chap. 3 for a review). We begin by laying out key theoretical components that we use to derive hypotheses about groupbased mediators of gang violence. We test these hypotheses using data from a recent study of gang social identity among the youth living in Los Angeles, California (Hennigan and Sloane 2011). Social identity theory is quite distinct from other identity-based theories (e.g., Erikson 1963; Stryker 1987; Stryker and Burke 2000) because of its focus on intergroup dynamics rather than interpersonal dynamics. Several authors have reviewed these differences in detail2 which we will not repeat here. The common ground among these theories is a focus on the self in general, but the processes described are quite different and operate at different levels. Social identity theory is based on the model of a layered self-concept that includes layers defined by social characteristics that arise from identification with social groups or categories that are distinct from personal or individual characteristics. The personal layer of one’s self-concept is defined in the context of interpersonal relations. In contrast, the social layers of one’s self-concept are defined in the context of intergroup relations. Gang ethnographers use working definitions of the concept of identity or gang identity that may or may not overlap with the social identity model discussed below. For example, much of Vigil’s (1988a, 1988b) discussion on gang identity is conceptualized through an interpersonal lens from the point of view of Erikson’s (1963) work and others (Wallace and Fogelson 1965; Whiting 1980). However, he explicitly develops his discussion beyond personal identity to encompass the notion of group identity. His description of group processes highlights aspects of the group context of gang membership that overlaps with the social identity perspective, though he does not explicitly refer to this theoretical orientation. In his treatise, Vigil (1988b) emphasizes the contribution that gang membership makes on one’s self-definition. He observes that “to gain acceptance from peers, an individual will adopt behavior patterns that initially have little intrinsic meaning to him, and perhaps might even be repugnant, but nevertheless are requisites for gang membership: for example, showing that one can feel, act, and look hard and uncaring” (pp. 427–428).

8 Gang Dynamics Through the Lens of Social Identity Theory

129

“Growing up in the barrio and becoming socialized and enculturated to street peer networks and beliefs makes an individual group-oriented (emphasis added) early in life” (p. 435). “A person learns to feel what the group feels,” and he must “learn the expectations of the group” (p. 437). “The barrio group … provides [its own] norms and patterns for emotional stability, social interaction and friendship, protection and street survival” (Vigil 1988b, p. 432). Overall, “for those who become members, the gang norms help shape what a person thinks about himself and others and provides models for how to look and act” (Vigil 1988b, p. 421).

8.1

The Social Identity Perspective

Social identity theory focuses on the way people think about themselves and others in an intergroup context. By identifying oneself as a member of a particular group (like a sports team, a school band, or a street-oriented group such as a street gang), one is accepting or taking on a social identity that has powerful implications cognitively for how an individual views himself and others and behaviorally for how an individual acts within his group and toward members of other groups. Tajfel (1978) defined social identity as “that part of an individual’s self concept which derives from knowledge of his membership in a social group (or groups) together with the emotional value and significance attached to that membership” (p. 63). His definition is deliberately more specific and more limited than a general definition of identity which typically refers to an overall concept of self from an individual or interpersonal point of view. Social identities are specific to intergroup contexts as opposed to interpersonal ones. In the discussion below, social identity is defined in reference to a particular group or social category. Most individuals have multiple social identities related to multiple social categories or groups with which the person identifies, all of which may become part of the individual’s overall self-concept. Tajfel and Turner’s (1979, 1986) work was in the context of intergroup relations involving real or perceived competition or conflict between social groups and categories. Studies have documented that even the simple awareness that there are two groups or categories, including one that I belong to and another that I do not, gives rise to ingroup favoritism and often outgroup derogation in terms of stereotyping and distributing rewards (Allport 1954; Billig and Tajfel 1973; Brewer 1999; Tajfel et al. 1971; Turner 1975). Research on social identity and related theories describe the dynamics by which we commonly perceive that our own group is better and more deserving than other groups, a perception that is continually reinforced by selectively comparing one’s ingroup with an outgroup on whichever evaluative criteria come to mind. Research has documented that the criteria that typically come to mind are those by which the outgroup falls short in relation to the ingroup (and not vice versa). In short, we are motivated to make comparisons that support the perception that our ingroup is superior to an outgroup in important ways because it promotes a sense of positive distinctiveness, self-definition, and self-esteem.

130

8.1.1


Continuum of Behavior: From Individual-Level to Group-Level Behavior

From the point of view of social identity theory, an individual’s social behavior falls between two extremes defined as interpersonal or intergroup behavior. At the interpersonal end of the continuum, the behavior between two or more individuals is fully determined by their interpersonal relationships and individual characteristics, where social identity has little or no influence. At the intergroup end of the continuum, the behavior between two or more individuals is fully determined by their membership in a particular social group or category; here, personal values and characteristics are not influential. Social behavior rarely actually falls at one extreme or the other, but varies somewhere in between these two extremes, depending on the salience of group (intergroup) or individual (interpersonal) concerns. Tajfel (1978, p. 41) used soldiers at war as an example of behavior at the intergroup end of the continuum. A poem written by an American infantryman, James Lenihan, provides a poignant example of this. His poem expresses his shift from the group end of the continuum to the individual end of the continuum after he shot and killed a German soldier during World War II3. The poem begins: I shot a man yesterday And much to my surprise, The strangest thing happened to me I began to cry. He was so young, so very young And Fear was in his eyes, He had left his home in Germany And came to Holland to die. And what about his Family Were they not praying for him? Thank God they couldn’t see their son And the man that had murdered him.

Murdered is a powerful word and it conveys perspective from the individual end of the continuum. Viewed from the group end of the continuum, the soldier’s action was likely viewed as duty: It was the War And he was the enemy If I hadn’t shot him He would have shot me. I saw he was dying And I called him “Brother” But he gasped out one word And that word was “Mother.”

His “surprise” is based on switching perspectives from the group end of the continuum (it was my duty as much as it was his duty) to the individual end (I saw him dying and called him brother). A soldier’s training is meant to support “group

8 Gang Dynamics Through the Lens of Social Identity Theory Individual –Level Behavior

Group-Level Behavior

Personally responsible

Personally anonymous (not responsible)

Further personal goals

Further group goals

131

Fig. 8.1 Tajfel’s continuum of behavior

behavior” and avoid the individual end of the continuum. The notion of a continuum is a heuristic that conveys how behavior toward another person may be controlled or influenced by the social identity-relevant aspects of an interaction Fig. 8.1. One of the corollary predictions of this model is that when a given social identity is controlling (at the group end of the continuum), the individual avoids feeling personally responsible for his actions. This is accomplished by depersonalizing the outgroup individuals involved as well as depersonalizing oneself, viewing each generically as a prototype of their group (Postmes and Spears 1998; Reicher et al. 1995). In this depersonalized state, the influence exerted by the social group through its normative expectations is paramount, meaning that the group-level expectations or norms outweigh one’s own individual behavioral preferences. It is interesting to reflect on stories that speculate as to how a person who showed no signs of aggression growing up became involved in group-based violence. From a social identity point of view, even persons who would be unlikely to be involved in violent acts at an interpersonal level, may well be influenced to do so at the intergroup level. This is consistent with the finding in longitudinal studies that criminal activities heighten when the youth become involved in street gangs and recede after they leave the gang (Esbensen and Huizinga 1993; Gordon et al. 2004; Thornberry et al. 2003). There also is evidence that persons with aggressive tendencies are more likely to be drawn to join a street gang group (Thornberry et al. 2003), presumably in the same way that empathic persons are more likely to be drawn to groups involved in community service. However, the collective norms of street gangs support and expect aggressive action when the gang is challenged and individuals that identify with a street gang can be expected to increase this type of behavior above levels motivated by individual proclivities. Gang members and members of other groups are clearly not behaving near the group end of the continuum all of the time. A comment made by one respondent in a recent study illustrates this point.4 This respondent related an incident where he had helped a friend that was jumped (beat-up) on the playground at school, even though the friend was involved with a rival gang. Then he cautioned the interviewer, “please don’t tell anyone this because if my homies knew they would kill me.” This young man engaged in interpersonal behavior (at the individual end of Tajfel’s continuum) to help a friend and was concerned that the incident not be viewed in an intergroup context because if it was, there would be consequences for violating his group’s normative expectations. What factors mediate shifts in focus along this hypothetical continuum from interpersonal behavior to intergroup behavior? In short, the salience of interpersonal versus intergroup aspects in any given situation are affected externally by the particulars of the social context (including a history of intergroup conflict or

132


competition, the location, who is present, etc.), but also internally by the individual’s strength of identification with his relevant ingroup, as well as his level of ingroup cohesion or perceived “groupness.” In the next sections, we will briefly review the literature on these concepts as they relate to internal or psychological factors that affect the consistency and extremity of intergroup behavior in general and gang behavior in particular.

8.1.2

Strength of Social Identity

Most of us are involved in various groups or social activities, yet not all of these inform our social identity. We identify strongly with some groups and less so with others, and this calculus may shift over time in reaction to circumstances and events. For example, there are those for whom identification with a political party strengthens around election time but otherwise remains relatively weak or unimportant. As a result, most of the time the normative expectations associated with political activism are inconsequential with little or no routine influence on behavior. For others, this social identity may be front and center and influential much of the time. One way to understand how strongly a person actively identifies with one particular social group or another is to consider what one gains from doing so, a concept called social identity value (Sherman et al. 1999). Research supports the notion that social identities are assimilated because (1) they help one feel good about oneself and achieve positive self-esteem (Tajfel 1978; Tajfel and Turner 1979, 1986); (2) they help reduce the uncertainty about who you are and who you are not, based on a generic or prototypical understanding of the characteristics and behavioral norms of members of specific groups or categories5 (Hogg 2000, 2007); and (3) they help fulfill basic human needs (i.e., needs to both fit in socially and at the same time be distinct or stand out in some unique way; Brewer 1991). Brewer maintains that both of these seemly contradictory needs can be achieved through social identity. Being part of a group or category allows each of us to fit in or belong, and at the same time, we are motivated to view our own group, and by extension ourselves, as being unique or distinctive in some positive way. Other things being equal, the strength of identification with a particular social group predicts or explains the extent to which an individual’s behavior is consonant with the normative expectations of that particular social group or category (Hogg and Abrams 1988; Hogg and Reid 2006; Smith et al. 2007). In other words, groups with which one is weakly identified (chronically or in certain situations) will have little normative influence. Among gang members, one would expect to find more correspondence between an individual’s behavior and the gang’s behavioral norms when social identification with the gang is high. Individuals that strongly identify with a group are expected to operate toward the group end of Tajfel’s behavioral continuum much more frequently than individuals with weaker group identification.


8.2

133

Ingroup Cohesion and Intergroup Conflict

Other factors that affect how much group membership influences one’s behavior include group cohesion and intergroup conflict. Early research in the area of group dynamics (c.f., Cartwright and Zander 1968; Gerard and Miller 1967) identified key processes that support group cohesion. Group cohesion, conceptualized as an attribute of the group, has been defined as “the total field of forces which act on members to remain in the group” (Festinger et al. 1950, p. 164). A group may be more or less cohesive depending on the interconnectedness of the individuals involved. The level of group cohesion has important implications for a group’s capacity for effectively responding to intergroup challenges. Research supports the view that group cohesion facilitates concerted action and aggressive response to conflict with another group. The reverse, that intergroup conflict promotes cohesion, is also empirically supported (Brewer 1999; Dion 1979; Grant and Brown 1995; Sherif 1967). As early as 1906, Sumner observed that “intergroup conflict and in-group solidarity may form a mutually reinforcing feedback system in which hostility between groups becomes a self-sustaining cycle” (Sumner, 1906 as cited in Dion 1979). Early on, gang researchers focused on this aspect of group processes. Klein and Crawford (1967) and Short and Strodtbeck (1965) observed that street gangs were not highly cohesive groups. They observed that gang membership and gang leadership were generally relatively unstable. Maintaining a focus on group behavior was difficult. To be effective, leaders needed to direct the members focus toward threats or challenges from outside of the group. Outside threats or challenges motivated participation in violent and criminal activities directed toward defending the group’s turf or honor and at the same time played a central role in maintaining the group’s cohesion as well as a leader’s own status. Spergel (1995), Klein (1995), and Decker (1996; Decker and Van Winkle 1996) have all made similar observations. They observed that it is competition with nearby street gangs that draws local gang members together and gang cohesiveness thrives on gang-on-gang hostility. Consistent with social psychological research, conflict or threats from the outside increases a group’s cohesion. In turn, strong group cohesion supports increased hostility toward persons outside the group. For street gang members, both group cohesion and intergroup conflict can be expected to strengthen identification with the gang and thereby increase the likelihood that gang members will behave near the group end of Tajfel’s continuum.

8.3

Ingroup Entitativity or “Groupness”

A related but separate line of research also sheds light on factors that support a shift toward group behavior. This research focuses on whether or not a collection of individuals is perceived to be a group (a social entity) or merely an aggregate of people. For example, if the youth are gathered in a local park, impressions and reactions to

134


the gathering (made by onlookers and by the participants themselves) may well depend on the extent to which the gathering is perceived to be a group or just a gathering. Campbell (1958) used the term “entitativity” to denote the degree of “groupness” inferred for an aggregate of people.6 Several cues have been found to underlie the inference that a collection of people is a group. Brewer et al. (2004) suggest that two general types of cues are often used to make inferences about groupness. One type of cue has to do with essence. A collection of individuals is more likely to be perceived as a group if they have similar traits or trait-related behaviors that define an essence for the group. Another aspect that conveys an essence is the sense that there are clear boundaries as to who is or is not part of the group. A second type of cue has to do with factors related to agency. Confirmation of a group’s agency includes indications of interdependence, interaction, or coordinated efforts among the individuals involved. Group cohesion is a key attribute of this type of entitativity. The more a collection of persons gather together, interact in concert, or convey a sense of interdependence, the higher the perceived groupness of that collection of people. In a recent study, Ip et al. (2006) characterized these two approaches to making inferences about groupness as being based on “birds of a feather” (essence) or on “flocking together” (agency). And other dichotomies have been suggested.7 Whichever approach is used, the degree of groupness a person infers for their own ingroup has implications for just where on Tajfel’s continuum their behavior is likely to fall. This inference of groupness underlies the basis for intergroup behavior linked to a social identity because behavior at the group end of his continuum requires that at least a minimal degree of groupness be perceived. An individual’s strength of identification with a collection of people is predicated in part by the extent to which that individual infers some level of groupness among the individuals involved. Spears et al. (2004) argue that both “group distinctiveness and entitativity play an important role in creating social identity and putting it to use” (p. 293). Outside observers generally infer that street gangs have a high level of entitativity or group cohesion (see Lickel et al. 2000),8 but research findings on inside observations are at odds with this (Decker and Van Winkle 1996; Klein and Crawford 1967). Inside observers document relatively low levels of cohesion among traditional street gang members themselves (i.e., low ingroup entitativity). We can speculate that some of the reasons for this have to do with a street orientation common among traditional street gangs that is open to a variety of youth hanging out together. Like most adolescents, street gang youth spend most of their time hanging out, engaged in activities that are not uncommon among adolescents such as drinking, using recreational drugs, trespassing, loitering, and vandalism. Violence is rare in proportion to all gang activities (Battin et al. 1998; Decker and Van Winkle 1996; Esbensen et al. 1993; Klein 1995; Short and Strodtbeck 1965). And not all of the youth hanging out together will join in the violent gang activities. In fact, it is sometimes difficult for researchers to determine just who is or is not a member of a street gang. Ethnographers, including Vigil (1988a), have categorized gang-involved youth as regularly, peripherally, situationally, or temporarily engaged in gang activities (p. 99). Similarly, other gang researchers categorize gang-involved youth at


135

various levels of commitment to gang life, such as hardcore, associates, fringe, or wannabe (Klein 1971, 1995). The point we are making is that despite an easy consensus among observers that street gangs have high entitativity or groupness (from an outgroup perspective), there is less consensus among the youth hanging out themselves (from an ingroup perspective) just how cohesive the membership is or even how clear the boundaries of the gang are. Vigil (1988b, p. 429) offers these perspectives from two youth who were fringe members for a short time during high school. One said that he joined to “feel good because someone is behind you,” but “I never made it as a true gang member.” The other youth said that he became part of the gang to blend in to the social environment. He would “dress and act like a cholo” and “act cool and not stare at anybody” to avoid being in fights. He stated that he left the group in high school as the coping strategy of gang identification brought more problems than it solved.

8.4

Can a Social Identity Perspective Help Explain Involvement in Group-Level Behaviors?

Our review of a mix of social psychological and gang research findings suggests that two factors contribute to the influence that an ingroup such as a street gang has on an individual’s behavior. These are the strength of identification with one’s group (How integral is this group in a person’s definition of who he is?) and perceptions of the groupness (How strongly defined is the group to that person?). Both of these factors promote depersonalization (i.e., seeing oneself as a prototypical member of the ingroup which facilitates compliance with ingroup norms). We expect that ingroup identification and ingroup entitativity are among key mediators of group behavior (i.e., behavior that is consistent with the group’s normative expectations— the group end of Tajfel’s continuum). If our goal is to understand the behavior of gang members in social situations, we should find that an individual’s level of identification with his gang as well as his perception of gang’s groupness (e.g., cohesion and clarity of the gang’s boundary) help explain or predict group behavior defined as behaviors that are consistent with the gang’s normative expectations. For most street gangs, normative group behavior includes involvement in criminal and violent activities. The same principle should be true for nongang youth, but in their case, criminal and violent activities are generally not part of the normative expectations for their peer group. So no relationship between these types of behaviors and markers of group identification or group entitativity with in nongang peer groups is expected. Other variables studied by criminologists may explain an individual’s involvement in criminal activities and violence. A large body of research has focused on the deterrent value of individually held perceptions of the likelihood of getting caught and punished for delinquent or criminal activities. All things being equal, the more likely an individual believes it is that he may be caught and punished, the less likely he is to engage in criminal or violent activity (see Pratt et al. 2006 for a meta-analytic

136


review; see also Maxson et al. 2011). These deterrence-related factors should be negatively associated with participation in delinquent and criminal activities when personal interests are salient (i.e., when the person is behaving close to the individual end of Tajfel’s continuum). Because serious delinquency and violence are generally not normative behavior within nongang peer groups, we predict that individually held beliefs about the likelihood of getting caught and punished will be associated with participation in these activities and strength of identification with the peer group should have nothing to do with it. One the other hand, for ganginvolved youth, we make the opposite prediction. Criminal and violent activities are strongly normative for gang members (see Decker 1996; Klein 1995). We predict that gang members, participation in these highly normative behaviors should be more strongly related to ingroup identification and group cohesion than to individual-level concerns such as the likelihood of getting caught and punished. In short, group-level expectations are expected to trump any individual-level concerns gang members may have for getting caught, but not so for the nongang youth.

8.5

Methods

Hennigan and Sloane (2011) interviewed the youth in areas claimed by local street gangs to study their reactions to civil gang injunctions in Los Angeles, California. The youth were interviewed in areas that were roughly matched on gang prevalence and sociodemographics. Some of the areas were involved in gang injunction efforts and some were not. Over a 10-month period, interviewers recruited male youth between the ages of 14 and 21 in designated gang neighborhoods in Los Angeles, California, by going door to door initially, then by engaging the youth hanging out on the street. Informed consent was obtained for all respondents and also from a parent for those who were less than 18 years old. As intended, this sampling approach resulted in oversampling street-oriented youth in the selected gang neighborhoods. Of the 416 respondents, 97% were Hispanic, and 87% were born in the United States. Thirty-nine percent (39%) of the gang-involved youth had been held in custody overnight relative to 7% of the nongang respondents. Eighty-four percent of the gang and 59% of nongang youth had been the victim of violence.9 Early in the interview, each respondent was asked to name various types of groups in which someone his age could become involved in his neighborhood including leadership groups, competitive teams, other organized clubs or activity groups, street gangs, tagger or skater groups, or party posses or crews. The interviewer then asked each respondent to list his own important peer groups. Interviewers were instructed to work with each youth to identify the respondent’s primary unconventional (i.e., not adult organized or sponsored) informal peer group in addition to one or more conventional groups.10 The analyses here are organized solely around the primary unconventional informal peer group that each respondent identified, which included street gangs (27%),


137

other crews or posses (27%), or informal groups of friends (46%). In the middle part of the interview, respondents answered a series of questions about their group. Social identity was measured using four items from the Identity Subscale of Luhtanen and Crocker’s (1992) Scale of Collective Self-Esteem.11 The four items were highly correlated (alpha = 0.79) and were combined to form a scale score with higher scores indicating stronger identification with one’s group. Focusing on the same construct used by Klein in his work (Klein and Crawford 1967; Klein 1971), cohesion with a peer group was operationalized as the frequency of meeting or getting together. Group cohesion was measured with a single item. “In some groups, the members meet or get together frequently, but in other groups the members rarely meet or get together at all. Recently, how often do you meet or get together with members of ?”12 Perception of group boundaries was also measured with a single question. “For some groups it is really clear who is in the group and who is not. These groups have clear boundaries. In others, it is not very clear just who is in the group and who is not. These have fuzzy boundaries. How clear or fuzzy are the boundaries in ?”13 In the second half of the interview, each respondent’s involvement in criminal activities and violence was measured with a frequently used self-report protocol based on the one developed for the National Youth Study (Huizinga and Delbert 1986) that was subsequently revised and used in the Causes and Correlates studies (see Esbensen and Huizinga 1993; Loeber et al. 1998; Thornberry et al. 2003). For the analyses presented here, level of involvement in criminal activities was defined using the same list of activities employed by Thornberry et al. (2003, Appendix A), except that the index of 31 items here which includes two rather than four items for theft, sexual assault, and prostitution were omitted, and questions about tagging, intimidation, and extortion were added. The subset of six violent activities included here were the same as those used by Thornberry et al. (2003), except that sexual assault was omitted and witness intimidation was added. A variety index of criminal activities was formed by counting how many activities on the list the youth admitted doing in the last 6 months. Similarly, a variety index of violence was created by counting how many violent activities the youth did in the last 6 months. (See Thornberry and Krohn 2000 for a discussion of the validity of this measurement approach.) Views of deterrence were measured by asking the youth to estimate how many times they would be caught and punished if they were to commit various types of crimes in their neighborhood on ten different occasions. The youth made estimates for 11 different criminal activities including three violent ones.14 For each activity, the youth indicated his expectation of being caught and punished—from zero to ten times. While the level of consequences expected varied for different kinds of criminal activities, across respondents there were significant correlations among these estimates (Pearson’s values were all statistically significant ranging from 0.22 to 0.60). This indicates that the youth who estimated a higher likelihood (relative to other respondents) on one type of crime tended to do the same for other crimes, even though the absolute level of their expectations varied across crimes. For this reason, we chose to combine these estimates to form a general scale of

138


expected consequences across a variety of 11 criminal activities and for the subset of three violent activities. There was a high correspondence in the responses across these 11 items (alpha = 0.86) and the subset of aggressive or violent items (alpha = 0.72).

8.6 8.6.1

Results Means and Standard Deviations of Criminal and Violent Activities and Proposed Mediators

The means and standard deviations of the indices of criminal behavior and of the subset of violent activities self-reported by gang and nongang youth are given in Table 8.1. Involvement in these types of behaviors is significantly higher for ganginvolved youth than nongang youth. This is consistent with the supposition that these types of behaviors are normative for gang-involved youth but much less so for nongang youth Table 8.1. Further, as suggested by our review of gang-related ethnographic and empirical research, on average gang-involved youth have weaker ingroup identification and more diffuse inferences about groupness, measured here as group cohesion and clear boundaries, than the youth involved in nongang peer groups. Nonetheless, there is considerable variability within both types of groups. Similarly, the means for deterrence-related expectations associated with criminal and violent activities vary by gang status. Higher expectations for being caught and punished are expressed by the nongang than by the gang respondents. However, considerable variation in perceived deterrence is evident within both groups.

8.6.2

Correlations Among Criminal and Violent Activities and Proposed Explanatory Variables

The bivariate correlations between the proposed explanatory variables and criminal and violent behavior of gang and nongang respondents are given in Table 8.2. As predicted, the table shows that both group identification and cohesion were significantly related to criminal activity in general (0.32, p < 0.01, and 0.22, p < 0.05, respectively) and with violent activities in particular (0.49, p < 0.01 and 0.22, p < 0.05) among gang members but not among nongang members where none of these correlations were statistically significant (−0.04 and −0.06 for criminal activity and −0.05 and 0.03 for violence). The measure of clear boundaries was not related to these behaviors for either group. The data confirm that despite overall lower levels of group identification and cohesion among gang respondents relative to nongang respondents, these ingroup characteristics were significantly related to


139

Table 8.1 Means and standard deviations of the study variables Gang Nongang (N = 112) (N = 304) Mean (S.D.) Mean (S.D.) Criminal activity (logged) 2.18 (0.058) 1.39 (0.72)*** Violence (logged) 0.90 (0.58) 0.41 (0.47)*** Identification 3.06 (1.39) 4.19 (1.13)*** Cohesion 4.01 (1.89) 4.97 (1.27)*** Clear boundaries 3.65 (1.29) 3.96 (0.92)* Deterrence—criminal activities 3.30 (1.99) 4.96 (2.43)*** Deterrence—violence 2.82 (2.40) 4.35 (2.82)*** *The mean difference is significant at the .05 level ***The mean difference is significant at the .001 level Table 8.2 Intercorrelations of criminal activity and violence and their predictors 1a 1b 2 3 4 5a Gang (N = 112) Criminal activity (logged) (1a) 0.71** 0.32** 0.22* −0.04 −0.24* Violence (logged) (1b) 0.49** 0.22* 0.02 NA Identification (2) 0.50** 0.22* −0.21* Cohesion (3) 0.30** −0.15 Clear boundaries (4) −0.02 Deterrence—criminal activities (5a) Deterrence—violence (5b) Nongang (N = 304) Criminal activity (logged) (1a) 0.64** −0.04 −0.06 −0.08 −0.32** Violence (logged) (1b) −0.05 0.03 −0.01 NA Identification (2) 0.16** 0.21** 0.05 Cohesion (3) 0.16** 0.04 Clear boundaries (4) 0.01 Deterrence—criminal activities (5a) Deterrence—violence (5b) * Correlation is significant at the 0.05 level ** Correlation is significant at the 0.01 level

5b NA −0.21* −0.16 −0.06 0.01 0.81**

NA −0.20** 0.02 0.02 0.03 0.84**

the gang respondents’ criminal and violent behaviors. On the other hand, none of these ingroup characteristics were related to criminal or violent behaviors among nongang respondents for whom these behaviors are not socially normative. Among nongang youth, deterrence-related concerns were significantly related to both general criminal and violent activities (−0.32, p < 0.01 and −0.20, p < 0.01) and unrelated to ingroup identification or cohesion among the nongang respondents as predicted. Among gang-involved youth, there were simple bivariate correlations between deterrence-related concerns and general criminal activity (−0.24, p < 0.05) and violence (−0.21, p < 0.05). There was also a significant bivariate correlation between deterrence concerns and gang identification (−0.21, p < 0.05) but not

140


between deterrence concerns and violence (−0.16, ns). The data were then tested using structural equation modeling to take the multivariate relationships into account to determine the best fitting model.

8.6.3

Structural Equation Modeling (SEM)

SEM was used to simultaneously test the relationships between cohesion, clear boundaries, deterrence, identification, and criminal and violent activity. Models were fit separately for gang-involved and nongang respondents to determine the best fitting models for each. Goodness-of-fit was evaluated using criteria recommended by Hu and Bentler (1999).15 Gang Models. In the first model shown in Fig. 8.2, we predicted the index of general criminal activity using the proposed explanatory variables. The model was an excellent fit to the data: c2(3) = 2.94, p = 0.40, RMSEA = 0.00, CFI = 1.00.16 Cohesion was related to identification, and identification was related to criminal activity. The clear boundaries variable was associated with cohesion, but not related to identification or the criminal index. As predicted, deterrence-related concerns were not significantly related to criminal activity in this model. Figure 8.3 shows that the best fitting model predicting violence as a dependent variable yielded a similar structure, with cohesion related to identification which was related to violence. It was also an excellent fit to the data: c2(3) = 1.96, p = 0.58, RMSEA = 0.00, CFI = 1.00.17 In this model however, a weak but statistically significant relationship with deterrence concerns was also indicated. Preacher and Hayes’ (2004) SPSS macro was used to assess the extent to which identification mediated the relationship between cohesiveness and criminal activities among gang members as suggested in these models. This analysis indicated that full mediation occurred for both the criminal activity index (Sobel Z = 2.49, p < 0.05) and for violence (Sobel Z = 3.96, p < 0.001). Models wherein identification was entered as an independent variable and cohesiveness as a mediator were not significant, indicating that all of the influence gang cohesion had on these dependent variables was mediated through social identification and not vice versa. Nongang Models. The same models were tested with nongang respondents. The resulting models provided an excellent fit to the data for criminal activity in Fig. 8.4: c2(3) = 2.07, p = 0.56, RMSEA = 0.00, CFI = 1.00;18 and violence in Fig. 8.5: c2(3) = 0.87, p = 0.83, RMSEA = 0.00, CFI = 1.00.19 Among the nongang respondents, groupness variables (clear boundaries and cohesion) were related to group identification but group identification was not related to criminal or violent activities. Instead, we observed that the criminal activity index (see Fig. 8.3) and the violence index (see Fig. 8.4) were negatively related to the personal estimates of the likelihood of getting caught and punished for such activities (deterrence). Deterrence had a strong negative relationship with both dependent variables among the nongang youth (Fig. 8.4 and Fig. 8.5) and only a weak relationship with violence among the gang-involved respondents (Fig. 8.3).


141

Clear Boundary .29**

.08

Cohesiveness

.46***

–.14

Identification

.30***

Criminal Activity

–.11 –.14

Deterrence

Fig. 8.2 Gang-involved respondents only: this structural equation model tests predictors of criminal activity among gang-involved youth. Paths with single-headed arrows represent directional effects, and paths with double-headed arrows represent nondirectional correlations. The model reports standardized regression weights. Bolded paths are significant (p < .05)


.08

Cohesiveness

.03

.47***

Identification

.47***

Violence

–.08 –.17*

Violent Deterrence

Fig. 8.3 Gang-involved respondents only: this structural equation model tests predictors of violent activities among gang-involved youth. Paths with single-headed arrows represent directional effects, and paths with double-headed arrows represent nondirectional correlations. The model reports standardized regression weights. Bolded paths are significant (p < .05)

8.7

Discussion and Conclusions

We found support for the notion that groupness defined as group cohesion (frequency of getting together) and group identification (or strength of social identity) are associated with criminal and violent behavior among gang members, but not among members of other kinds of peer groups in the same neighborhoods. In the SEM models with gang-involved respondents, the association between group cohesion and

142



Cohesiveness

.19*** .12*

.04

Identification

.03

Criminal Activity

.04 –.32***

Deterrence

Fig. 8.4 Nongang respondents only: this structural equation model tests predictors of criminal activity among youth involved in nongang peer groups. Paths with single-headed arrows represent directional effects, and paths with double-headed arrows represent nondirectional correlations. The model reports standardized regression weights. Bolded paths are significant (p < .05)


Cohesiveness

.02

.19***

.12*

Identification

–.05

Violence

.01 –.19***

Violent Deterrence

Fig. 8.5 Nongang respondents only: this structural equation model tests predictors of violent activities among youth involved in nongang peer groups. Paths with single-headed arrows represent directional effects, and paths with double-headed arrows represent nondirectional correlations. The model reports standardized regression weights. Bolded paths are significant (p < .05)

both crime and violence was fully mediated by strength of social identity. These relationships were not found in the models with nongang respondents. We maintain that this disparity is due to widely different normative expectations within street gangs versus within nongang peer groups. Based on the tenets of social identity theory, we interpret criminal and violent behavior among gang members as a groupbased phenomenon, not in terms of engaging in these behaviors together at the same time (which may or may not happen), but rather in terms of one’s motivation to act. The stronger one’s identification with the gang, the stronger the individual is focused


143

on the gang’s normative expectations (i.e., behavior at the group end of Tajfel’s continuum), regardless of individual concerns. Research suggests that a person engaged in group behavior depersonalizes himself and his victims, seeing each more in terms of a generic representative or prototype of their respective group than as individuals. These findings support the value of using this theoretical framework as a model in future research and strategic thinking about gang violence reduction and gang reduction strategies. The findings support framing the motivational context for gang violence in terms of social identity theory in general and Tajfel’s continuum in particular. This study also contributes to our understanding of why the process of weighing the likelihood of personal costs for engaging in illegal activities20 does less to deter gang members than it appears to for nongang youth. Among the nongang respondents, the data suggest that the higher a respondent’s estimate of the likelihood of getting caught and punished, the less involvement in criminal and violent activities. However, among the gang-involved youth, the data suggest that identification with the gang can trump or drastically reduce personal interests such as concerns about getting caught and punished when the behaviors in question are normative and group identification is strong. Since crime and violence are normative among ganginvolved youth, personal estimates of getting caught and punished have little or no influence on their criminal and violent behaviors. Among nongang youth, however, serious crime and violence are rarely part of their groups’ normative expectations, so personal interests such as avoiding risks of punishment carry more weight. Gang researchers have observed that gang cohesion is less a function of internal factors such as affective ties or individual similarities than external factors that spawn social interaction in reaction to gang rivalries (Decker 1996; Klein and Crawford 1967). The findings for gang-involved youth in this study are consistent with this view.21 This preoccupation with group rivalries has been described as a unique characteristic of street gangs, even as one example of ways that street gangs are qualitatively different from other groups. It is our contention that this and other aspects of group dynamics in gangs may be largely predicted or explained through the lens of social identity and related theories. Below, we present examples of approaches that researchers might take toward using social identity and related theories to generate explanations that may extend our understanding of gang dynamics in ways that could be useful for gang prevention and intervention. Instead of treating social identity as a unitary concept, Leach et al. (2008) demonstrated that there may be value in examining multiple aspects of social identity. They confirmed a two-component model of group identification: self-definition and self-investment that predicted different reactions to ingroup transgressions. Specifically, when facing apparent misdeeds undertaken by group members, the self-investment component was prospectively associated with cognitively legitimizing the role of the group to the point of denying apparent incriminating evidence that the group’s actions were wrong and embracing interpretations that their group’s actions were just and warranted. Self-investment was associated with defending the group and supporting the group’s actions, regardless of the objective circumstances. Three of the four items in the measure of social identity used in the study presented here were included in the

144


self-investment component of the Leach et al. model. We suspect that it is the selfinvestment component of gang identification that is strongly related to violence and intergang rivalries. In contrast, the stronger the self-definition component of social identity, the more likely it was that individuals viewed apparent misdeeds by their group as unfortunate and even felt guilty about their own (indirect) association with it. Leach et al. reasoned that when a person identifies with his group primarily on the basis of being similar to others in the group, then the person may feel personal guilt or misgivings about the group’s misbehavior and is less likely to defend the group’s questionable actions. We wonder if this might describe the reactions of former gang members who now work with programs to discourage gang violence and help prevent the youth from joining or help the youth leave the gang. Is it possible that these former gang members have transitioned from a self-investment style of social identity to a self-definitional style and their reactions to gang violence have evolved accordingly? If so, future research could examine the possibility of creating intervention strategies that minimize the self-investment component among gang members without being concerned about a lasting self-definition component of social identification. A second line of research undertaken to test social identity-related hypotheses that may be relevant to street gang dynamics has focused on which rival outgroups are likely to command the most attention or concern. Based on the reactive distinctiveness hypothesis (see Tajfel and Turner 1979; Jetten et al. 2004), it is predicted that a group will be highly motivated to distinguish itself from other relevant groups that are minimally distinct from (i.e., are very similar to) the ingroup. The implication is that the most serious conflict for a street gang should be directed toward objectively similar groups (i.e., nearby local street gangs), because this kind of outgroup threatens the ingroup’s reputed superiority on valued characteristics (e.g., toughness and hardness). In other words, driven by competition for the superior social identity, the outgroups that would be predicted to elicit the most competition (and galvanize ingroup cohesion) are nearby street gangs seen as having similar attributes (close to equals). Striving for a superior social identity (reputation, status), gangs engage in delinquent or violent activities to distinguish themselves from similar rival outgroups. Could this motive for positive distinctiveness be met in some more socially appropriate ways?22 A third example of the way a social identity framework could be useful is suggested by research on conflicting social identities. Consider the case of conflicting social identities articulated by Lien (2005) in her interviews with a young emerging gang member in Oslo. Lien describes the internal conflict a young man felt as he committed robbery with other gang members for the first time. He explained how he had acted with steely resolve together with others in his group and assumed the role of ruthless criminal well enough to scare the victim and obtain the victim’s money. But later, at home and out of the group context, he felt remorse and feared that his family or school buddies would find out how he obtained the money. He continued his gang affiliation by compartmentalizing the activities, stealing from victims (expected in the gang context), and generously treating his friends and family to gifts (respected within his family and conventional group of friends). He told the interviewer, “I would die of shame if my girl friend and friends knew.” From a social identity perspective,


145

opposing or conflicting social identities (e.g., antisocial versus prosocial) may be unstable. Social identities are conceptualized as layers of one’s self-definition. While many people have multiple layers of social definitions of themselves, identities with opposing normative expectations such as those with antisocial versus prosocial or conventional norms may be difficult to sustain. Recent work by Brook et al. (2008) suggests that incompatibility among the normative expectations of multiple identities that are strongly held and important to the individual is related to depression, anxiety, and unhappiness. We suspect that maintaining strong opposing social identities is not tenable for most people. This suggests that gang prevention programs that find a way to assist the youth in developing strong identification23 with a peer group that has conventional normative expectations could offer some protection against the temptations of gang involvement. If group affiliation is driven by social identity value as argued by social psychologists (Sherman et al. 1999), then the affiliation that brings the greater value should dominate. In closing, we are optimistic that using the theoretical framework and empirical knowledge accumulated on the cognitive and behavioral implications of social identity and related theories can contribute new ideas to the challenge of gang reduction.

Notes 1. Here, we will use the term “street gangs,” following Malcolm Klein’s lead (Klein 1995). In this discussion, we will be referring primarily to traditional street gangs that are structured around a territory or turf that is claimed by the gang. This discussion may or may not apply to other types of gangs such as prison gangs, drug gangs, or other gangs. 2. See Brewer (2001), Hogg et al. (1995), and Stets and Burke (2000). 3. Written after he returned home, the poem was found by James Lenihan’s children after his death. They shared the poem with CNN on May 28, 2010. To view the complete poem, use the link http://articles.cnn.com/2010-05-28/us/soldier.poem_1_poem-soldier-holland?_s=PM:US. 4. Documented by an interviewer in the Hennigan and Sloane (2011) study. 5. Though clearly relevant to social identity, we have excluded from this discussion identification with social categories such as male or female, gay or straight, and broad groups based on wealth, political affiliation, country of residence or origin, etc. and focused in this article on groups of people, such as peer groups, teams, activity groups, and street gangs. 6. In this chapter, we will use the word “groupness” in place of the somewhat awkward term entitativity. 7. Other researchers have defined the mechanisms that characterize groups in different ways including dynamic and categorical (Wilder and Simon 1998), inductive or deductive (Postmes et al. 2005), and common bond versus common identity (Prentice et al. 1994). 8. Brian Lickel and his colleagues (Lickel et al. 2000) found in separate studies that college students in the United States and in Poland rated the level of entitativity or groupness of “a local street gang” as being quite high, near the top of the scale, similar to the levels rated for a rock band or a professional sports team. This is consistent with the “Westside Story” stereotype prevalent in the USA that gangs are highly cohesive groups. 9. For additional methodological details, see Hennigan and Sloane (2011). 10. Conventional peer groups were defined as youth groups sponsored or organized by adults in the context of school, community, or religious organizations. “Unconventional” peer groups

146

11.

12. 13. 14.

15. 16.

17.

18. 19. 20.

21. 22.

23.

K. Hennigan and M. Spanovic were defined as a clique of crew that was part of the street gang, a tagger or skater or party group, or some other informal group of peers (friends). The items included were the following: “Overall has very little to do with how I feel about myself. is an important reflection of who I am. is unimportant to my sense of what kind of person I am. In general, belonging to is an important part of my self-image. Six response options ranged from strongly agree to strongly disagree.” Response options used were the following: 6 = almost every day, 5 = two to five times a week, 4 = about once a week, 3 = two or three times a month, 2 = about once a month, or 1 = less than that. Five response options ranged from very fuzzy to very clear. The activities were drinking in public, shoplifting, driving while drunk or high, tagging or writing graffiti, stealing a car, breaking into a building to steal something, trespassing on private property, selling drugs in your neighborhood, hitting someone in a fight, seriously beating someone up, intimidating or challenging someone that might tell authorities about something illegal that they or a friend did. A chi-square value that is nonsignificant, an RMSEA < 0.08, and a CFI > 0.95 represents a good fit (Hu and Bentler 1999). Alternative models were also tested. Specifically, identification was entered as an independent variable and cohesion, clear boundaries, and deterrence were entered as mediators. This model provided a poor fit, c2(3) = 7.87, p = 0.05, RMSEA = 0.12, CFI = 0.90. It also seemed plausible that delinquency predicted cohesion which in turn predicted identification. This model was not a good fit for the data, c2(3) = 7.97, p = 0.05, RMSEA = 0.14, CFI = 0.68. Entering identification as an independent variable and cohesion, clear boundaries, and deterrence as mediators provided a worse fit compared to the fit of the model presented in Fig. 8.2, c2(3) = 5.27, p = 0.15, RMSEA = 0.08, CFI = 0.96. An alternative model wherein violence was entered as an independent variable and identification as a dependent variable did not fit the data well, c2(3) = 7.11, p = 0.07, RMSEA = 0.13, CFI = 0.88. Reversing the order of independent and dependent variable also resulted in a good fit, c2(2) = .22, p = 0.89, RMSEA = 0.00, CFI = 1.00. Reversing the order of independent and dependent variable also provided a good fit, c2(2) = 0.38, p = 0.83, RMSEA = 0.00, CFI = 1.00. Note that studies that include measures of the social costs of criminal offending blur the distinction made here between individual estimates of the likelihood of getting caught and punished and social identity (i.e., the social costs bleed into group identification). Presumably, the stronger one’s identification with a social group, the more likely concerns about a comember’s views of offending will be correlated with shared normative expectations. The findings are consistent though not all of the alternative definitions were tested here. This observation should not be confused with the kind of organizing that street workers used in the past to attempt to guide gang members toward more conventional activities (see Klein 1995). Clearly, this kind of approach would have to be undertaken in ways that would avoid increasing gang cohesion and avoid fueling traditional gang rivalries. With emphasis on “strong” because Brook et al. (2008) found that social identities that were not strongly held had little or no influence on psychological well-being at all.

References Allport G (1954) The nature of prejudice. Addison-Wesley, Cambridge Battin SR, Hill KG, Abbott RD, Catalano RF, Hawkins JD (1998) The contribution of gang membership to delinquency beyond delinquent friends. Criminology 36:93–115 Billig M, Tajfel H (1973) Social categorization and similarity in intergroup behavior. Eur J Soc Psychol 3:27–52 Brewer M (1991) The social self: on being the same and different at the same time. Pers Soc Psychol Bull 17:475–482


147

Brewer M (1999) The psychology of prejudice: ingroup love or outgroup hate? J Soc Issues 55:429–444 Brewer M (2001) The many faces of social identity: implications for political psychology. Polit Psychol 22:115–125 Brewer M, Hong Y, Li Q (2004) Dynamic entitativity: perceiving groups as actors. In: Judd C, Yzerbyt V, Corneille O (eds) The psychology of group perception: perceived variability, entitativity, and essentialism. Psychology Press, Philadelphia, pp 25–38 Brook AT, Garcia J, Fleming M (2008) The effects of multiple identities on psychological wellbeing. Pers Soc Psychol Bull 34:1588–1600 Campbell DT (1958) Common fate, similarity, and other indices of the status of aggregates of persons as social entities. Behav Sci 3:14–24 Cartwright D, Zander A (eds) (1968) Group dynamics, 3rd edn. Harper and Row, New York Decker SH (1996) Collective and normative features of gang violence. Justice Q 13:243–264 Decker SH, Van Winkle B (1996) Life in the gang: family, friends and violence. Cambridge University Press, New York Dion KL (1979) Intergroup conflict and intragroup cohesiveness. In: Austin W, Worchel S (eds) The social psychology of intergroup relations. Brooks/Cole, Monterey, pp 211–224 Erikson EH (1963) Childhood and society. Norton, New York Esbensen F-A, Huizinga D (1993) Gangs, drugs, and delinquency in a survey of urban youth. Criminology 31:565–589 Esbensen F-A, Huizinga D, Weiher A (1993) Gang and non-gang youth: differences in explanatory factors. J Contemp Crim Just 9:94–116 Festinger L, Schachter S, Back K (1950) Social pressures in informal groups. Harpers, New York Gerard HB, Miller N (1967) Group dynamics. Annu Rev Psychol 18:287–332 Gordon RA, Lahey BB, Kawai E, Loeber R, Stouthamer-Loeber M, Farrington DP (2004) Antisocial behavior and youth gang membership: selection and socialization. Criminology 42:55–87 Grant PR, Brown R (1995) From ethnocentrism, to collective protest: responses to relative deprivation and threats to social identity. Soc Psychol Q 58:195–211 Hennigan KM, Sloane DC (2011) Effects of civil gang injunctions on street gang-involved youth: acquiescence or defiance. Unpublished manuscript, Department of Psychology, University of Southern California, Los Angeles Hogg MA (2000) Subjective uncertainty reduction through self-categorization: a motivational theory of social identity processes. Eur Rev Soc Psychol 11:223–255 Hogg MA (2001) Social categorization, depersonalization, and group behavior. In: Hogg M, Tinsdale TS (eds) Handbook of social psychology: group processes. Blackwell, Malden, pp 57–85 Hogg MA (2007) Uncertainty-identity theory. In: Zanna MP (ed) Advances in experimental social psychology, vol 39. Academic, San Diego, pp 69–126 Hogg MA, Abrams D (1988) Social identifications: a social psychology of intergroup relations and group processes. Routledge, London Hogg MA, Reid SA (2006) Social identity, self-categorization, and the communication of group norms. Commun Theory 16:7–30 Hogg M, Terry D, White K (1995) A tale of two theories: a critical comparison of identity theory with social identity theory. Soc Psychol Q 58:255–269 Hu LT, Bentler PM (1999) Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equation Model 6:1–55 Huizinga D, Delbert ES (1986) Reassessing the reliability and validity of self-report delinquency measures. J Quant Criminol 2:293–327 Ip GW, Chiu C, Wan C (2006) Birds of a feather and birds flocking together: physical versus behavioral cues may lead to trait- versus goal-based group perception. J Pers Soc Psychol 90:368–380 Jetten J, Spears R, Postmes T (2004) Intergroup distinctiveness and differentiation: a meta-analytic integration. J Pers Soc Psychol 86:862–879

148


Klein M (1971) Street gangs and street workers. Prentice-Hall, Englewood Cliffs Klein M (1995) The American street gang: its nature, prevalence, and control. Oxford University Press, New York Klein M, Crawford LY (1967) Groups, gangs and cohesiveness. J Res Crime & Del 4:63–75 Klein M, Maxson C (2006) Street gang patterns and policies. Oxford University Press, New York Leach CW, Zomeren M, Zebel S, Vliek M, Pennekamp SF, Doosje B, Ouwerkerk JW, Spears R (2008) Group-level self-definition and self-investment: a hierarchical (multicomponent) model of in-group identification. J Pers Soc Psychol 95:144–165 Lickel B, Hamilton D, Wieczorkowska G, Lewis A, Sherman S, Uhles AN (2000) Varieties of groups and the perception of group entitativity. J Pers Soc Psychol 78:223–246 Lien I-L (2005) The role of crime acts in constituting the gang’s mentality. In: Decker S, Weerman F (eds) European street gangs and troublesome youth groups. Alta Mira Press, Lanham, pp 31–50 Loeber R, Farrington DP, Stouthamer-Loeber M, Moffitt TE, Caspi A (1998) The development of male offending: key findings from the first decade of the Pittsburgh youth study. Stud Crime and Crime Prev 7:141–172 Luhtanen R, Crocker J (1992) A collective self-esteem scale: self-evaluation of one’s social identity. Pers Soc Psychol Bull 18:302–318 Maxson C, Matsuda K, Hennigan K (2011) Deterrability among gang and nongang juvenile offenders: are gang members more (or less) deterrable than other juvenile offenders? Crime Delinquency 57:516–543 Pickett CL, Bonner B, Coleman JM (2002) Motivated self-stereotyping: heightened assimilation and differentiation needs result in increased levels of positive and negative stereotyping. J Pers Soc Psychol 82:543–562 Postmes T, Spears R (1998) Deindividuation and anti-normative behavior: a meta-analysis. Psychol Bull 123:238–259 Postmes T, Spears R, Lee T, Novak R (2005) Individuality and social influence in groups: inductive and deductive routes to group identity. J Pers Soc Psychol 89:747–763 Pratt T, Cullen F, Blevins K, Daigle L, Madensen T (2006) The empirical status of deterrence theory: a meta-analysis. In: Cullen F, Paul J, Blevins K (eds) Taking stock: the status of criminological theories. Transaction Publishers, New Brunswick, pp 367–395 Preacher KJ, Hayes AF (2004) SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behav Res Methods Instrum Comput 36:717–731 Prentice DA, Miller DT, Lightdale JR (1994) Asymmetries in attachments to groups and to their members: distinguishing between common-interest and common-bond groups. Pers Soc Psychol Bull 20:484–493 Reicher S, Spears R, Postmes T (1995) A social identity model of deindividuation phenomena. Eur Rev Soc Psychol 6:161–198 Sherif M (1967) Group conflict and cooperation: their social psychology. Routledge, London Sherif M, Sherif CW (1979) Research on intergroup relations. In: Austin WG, Worchel S (eds) The Social Psychology of Intergroup Relations. Brooks/Cole Publishing, Monterey, CA Sherman SJ, Hamilton DL, Lewis AC (1999) Perceived entitativity and the social identity value of group memberships. In: Abrams D, Hogg MA (eds) Social identity and social cognition. Blackwell, Malden, pp 80–110 Short JF, Strodtbeck FL (1965) Group process and gang delinquency. University of Chicago Press, Chicago Smith JR, Terry DJ, Hogg MA (2007) Social identity and the attitude-behavior relationship: effects of anonymity and accountability. Eur J Soc Psychol 36:239–257 Spears R, Scheepers D, Jetten J, Doosje B, Ellemers N, Postmes T (2004) Group homogeneity, entitativity and social identity: dealing with/in social structure. In: Yzerbyt V, Judd CM, Corneille O (eds) The psychology of group perception: contributions to the study of homogeneity, entitativity and essentialism. Psychology Press, New York, pp 293–316 Spergel IA (1995) The youth gang problem. Oxford University Press, New York


149

Stets JE, Burke P (2000) Identity theory and social identity theory. Soc Psychol Q 63:224–237 Stryker S (1987) Identity theory: developments and extensions. In: Yardley K, Honess T (eds) Self and identity: psychosocial perspectives. Wiley, New York Stryker S, Burke P (2000) The past, present, and future of an identity theory. Soc Psychol Q 63:284–297 Tajfel H (1978) Differentiation between social groups. Academic, London Tajfel H, Turner JC (1979) An integrative theory of intergroup conflict. In: Austin WG, Worchel S (eds) The social psychology of intergroup relations. Brooks-Cole, Monterey Tajfel H, Turner JC (1986) The social identity theory of inter-group behavior. In: Worchel S, Austin LW (eds) Psychology of intergroup relations. Nelson-Hall, Chicago Tajfel H, Flament C, Billig M, Bundy RP (1971) Social categorization and intergroup behaviour. Eur J Soc Psychol 1:149–178 Thornberry TP, Krohn MD (2000) The self-report method for measuring delinquency and crime. In: Duffee D, Crutchfield RD, Mastrofski S, Mazerolle L, McDowall D, Ostrom B (eds) CJ2000: innovations in measurement and analysis, vol 4. US Department of Justice, Washington, DC, pp 33–83 Thornberry TP, Krohn MD, Lizotte AJ, Smith CA, Tobin K (2003) Gangs and delinquency in developmental perspective. Cambridge University Press, New York Turner JC (1975) Social comparison and social identity: some prospects for intergroup behaviour. Eur J Soc Psychol 5:5–34 Turner JC (1985) Social categorization and the self-concept: a social cognitive theory of group behavior. In: Lawler EJ (ed) Advances in group processes: theory and research, vol 2. JAI Press, Greenwich, pp 77–121 Turner JC, Hogg MA, Oakes PJ, Reicher SD, Wetherell M (1987) Rediscovering the social group: a self-categorization theory. Basil Blackwell, Oxford, England Vigil JD (1988a) Barrio gangs: street life and identity in southern California. University of Texas Press, Austin Vigil JD (1988b) Group processes and street identity: adolescent Chicano gang members. Ethos 16:421–445 Wallace AFC, Fogelson R (1965) The identity struggle. In: Boszomeniji-Nagy I, Framo JL (eds) Intensive family therapy: theoretical and practical aspects. Harper and Row, New York Whiting BB (1980) Culture and social behavior: a model for the development of social behavior. Ethos 8:95–116 Wilder D, Simon AF (1998) Categorical and dynamic groups: implication for social perception and intergroup behavior. In: Sedikides C, Schopler I, Insko C (eds) Intergroup cognition and intergroup behavior. Erlbaum, Mahwah, pp 27–44

Chapter 9

Gang Membership: The Psychological Evidence Emma Alleyne and Jane L. Wood

The history of gang research stems predominantly from the fields of criminology and sociology. These perspectives have greatly informed us on the circumstances that put young people at increased risk from a societal and individual level (see Wood and Alleyne 2010, for review). For example, risk factors have included: low socioeconomic status (Spergel 1995; Rizzo 2003), parenting, and deviant peers (Thornberry et al. 2003; Sharp et al. 2006), to name a few. Historically, Short and Strodtbeck (1965) argued that the four main influences on gang youth are the environment, individual characteristics, group norms, and group process. What is astonishing is that there is an abundance of psychological literature examining group processes, and considering that the gang is, in fact, a group phenomenon (Goldstein 2002), the literature linking group psychology and gangs is scant (Wood and Alleyne 2010). Essentially, psychologically, we can assume that gangs form for the same reasons that any other group forms: peer friendship, pride, identity development, enhancement of self-esteem, excitement, the acquisition of resources, and goals that, due to low-income environments, may not be available through legitimate means (Goldstein 2002). Gangs may offer a strong psychological sense of community, a physical and psychological neighborhood, a social network, and a social support (Goldstein 1991). This chapter can be broken down into the following main sections. First, we discuss the reasons why psychology is important in gang research and outline theoretical frameworks useful for examining the psychology of gang members. We then move on to present some of our recent findings. We discuss the environmental contribution to gang membership and consider the intergroup context between gangs. We then report our findings regarding intragroup processes within the gang structure and gang hierarchy. Next, we report the contribution of antiauthority attitudes to gang membership, and lastly, we discuss how these processes translate into behavior.

E. Alleyne (*) • J.L. Wood School of Psychology, Keynes College, University of Kent, Canterbury, Kent, UK F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_9, © Springer Science+Business Media, LLC 2012

151

152

9.1

E. Alleyne and J.L. Wood

Psychology and Gang Membership

Even though the psychology of gang membership has had little attention in gang literature (Thornberry et al. 2003; Wood and Alleyne 2010), some researchers have examined how individual differences such as personality characteristics interact with environmental factors to motivate youth to join gangs. For example, research shows how youth who have psychopathic personality traits are five times more likely to become gang members than youth without such traits (Dupéré et al. 2007). These youth are less sensitive to informal social controls such as parental supervision, which is noted as contributing (albeit modestly) to gang membership (LeBlanc and Lanctot 1998). Joining a gang is even more likely if these youth live in an adverse family environment (Lacourse et al. 2006). We also know from research that gang members hold more negative attitudes to authority (Kakar 2005) such as the police (Lurigio et al. 2008; Vigil 1988). Risk factors for gang membership also function on individual differences such as lower IQ levels (Spergel 1995), learning difficulties and mental health problems (Hill et al. 1999), and low self-esteem (Dukes et al. 1997). As children grow, they may seek a status that differs from the one prescribed by the legitimate social order prescribed by parents and teachers (Anderson 1999). Young people may be tempted into gangs because they offer the potential to gain respect and status (Anderson 1999). Knox (1994) asserted that gangs exert two types of social power that attract youth: coercive power—the threat or actual use of force and violence—and the power to pay, buy, or impress, and delegate status and rank to its members. As such, gang membership may satisfy the universal needs of youth for status, identity, and companionship (Klein 1995; Vigil 1988). However, we know that people experience moral conflicts when they come across benefits requiring immoral behavior (Bandura 1990), leading them to engage in what Bandura (2002) described as moral disengagement strategies, which consist of “cognitive restructuring of inhumane conduct into benign or worthy behaviour (p. 101).” Social cognitive theory (Bandura 2002), very like neutralization theory (Sykes and Matza 1957), describes eight strategies for reducing moral conflicts. Moral justification, advantageous comparison, and euphemistic labeling refer to the cognitive reconstruction of a harmful behavior into a good one by comparing it against more inhumane and harmful ones or by giving it a sanitized label (i.e., this is business), respectively. Displacement of responsibility allows personal responsibility to be viewed as a result of societal or authority pressures (i.e., I was only following orders). When the harm is shared with others (e.g., a group), diffusion of responsibility obscures personal responsibility. Blaming the victim or circumstances also serves to obscure personal responsibility. By disregarding or distorting the consequences of a harmful behavior, guilt or shame is reduced which eliminates selfcondemnation. And dehumanization allows victims to be stripped of their human qualities and be viewed as subhuman objects devoid of feelings. We already know that youth will set aside their moral standards if by doing so they will be accepted by a chosen group (Emler and Reicher 1995). As such, it should be possible to use social cognitive processes such as moral disengagement to help explain the process

9

Gang Membership: The Psychological Evidence

153

of how youth disengage from the informal social controls they have learned in favor of the rewards gang membership offers. There is also the possibility that there are social learning aspects to gang membership. Young boys look up to gang members, mimic them, and aspire to gang membership (Hughes and Short 2005) and gang films, depicting characters rewarded for gang-like behaviors, act as a blueprint for young aspiring gang members (Przemieniecki 2005). Consequently, youth may adapt, modify, or discard their existing social controls in favor of what they perceive as the attractive or even glamorous attributes of gang membership. Accordingly, it seems that psychology has much to contribute to our understanding of gang membership (see Wood and Alleyne 2010 for a review). However, we also need to consider that individual differences do not apply solely at the individual level. No two gangs are alike, and community experiences of gang problems vary widely. Such variance is frustrating to the media and others who thrive on simplicity and sweeping generalizations (Esbensen and Tusinki 2007). The only way we can counteract the perpetuation of the myths and errors that underpin many assumptions made about gang membership is by conducting rigorous theoretically derived empirical work that includes psychological factors relevant to street gang membership. However, in order to begin to research the psychology of gang membership, it is vital that we do so within a strong theoretical framework.

9.2

Interactional Theory

Interactional theory provides us with the ability to inform our inferences about the group processes that drive the behavioral outcomes of gang members. Interactional theory (Thornberry 1987; Thornberry and Krohn 2001; Thornberry et al. 2003) has elaborated earlier criminological theories by proposing that gang membership results from a reciprocal relationship between the individual and peer groups, social structures (e.g., poor neighborhood, school, and family environments), weakened social bonds, and a learning environment that fosters and reinforces delinquency. This theory is analogous to a marriage between two theories: control theory and social learning theory. Control theory argues that people who engage in deviant behavior do so when their bonds to society weaken (Hirschi 1969). However, this theory does not acknowledge the effects of antisocial influences (e.g., delinquent peers) on gang membership, and these effects have been noted in the literature (e.g., Esbensen and Weerman 2005). On the other hand, social learning theory argues that crime is learned through: the development of beliefs that crime is acceptable in some situations, the positive reinforcement of criminal involvement (e.g., approval of friends, financial gains), and the imitation of the criminal behavior of others—especially if they are influential figures in the individual’s life (Akers 1997). One drawback with social learning theory is that it fails to specify how much individuals need to favor crime prior to engaging with like-minded delinquent peers (e.g., gang members) (Akers 1997). Unlike control theory, social learning theory, and others, which take a

154


unidirectional perspective of delinquency involving specific risk factors that cause a youth to become delinquent, interactional theory provides a more subtle developmental explanation of delinquency where societal, learning, and delinquency factors all interact and mutually influence each other across an individual’s life span (Thornberry et al. 2003). Thornberry et al. (2003) also proposed that the route to joining a gang can be forged by one of three processes. Gang membership may result from selection where gangs select and recruit members who are already delinquent (Lahey et al. 1999; Craig et al. 2002); from facilitation where gangs provide opportunities for delinquency to youth who were not delinquent beforehand (Gatti et al. 2005; Gordon et al. 2004; Thornberry et al. 1993); and from enhancement where gang members are recruited from a population of high-risk youth who, as gang members, become more delinquent (Gatti et al. 2005; Thornberry et al. 1993). Furthermore, these models introduce two different types of gang members: transient and stable. Gatti et al. (2005) showed how youth who were delinquent before becoming gang members are more likely to remain in the gang long term (i.e., stable members), while youth who were not already delinquent were more-or-less temporary members (i.e., transient members). This implies that delinquency before gang membership is an important influence on the longevity of gang membership. Past research has found that the likelihood of joining a gang increases when young people experience risk factors from all social and environmental domains, that is, individual, family, school, peer, and neighborhood factors (Howell and Egley 2005). Also, these risk factors have a cumulative effect. That is, the more risk factors a young person has or experiences in life, the more likely they are to join a gang (Hill et al. 2001; Howell and Egley 2005). Thus, Howell and Egley (2005) proposed an extension of interactional theory to include younger children when examining associated risk factors. Howell and Egley (2005) argued that examining adolescent samples alone, as most gang research does, is erroneous since the developmental trajectory toward gang membership would have already begun. By concentrating on adolescents, we are therefore missing key factors that precede the known risk factors. However, comparing adolescents who are gang members with those who are not still allows us to further isolate and define the risk factors that put young people at most risk. Past theoretical approaches have acknowledged that psychological constructs such as attitudes and beliefs bind young people together (i.e., birds of a feather flock together). For example, social learning theory argues that young people learn these antisocial attitudes and the associated behaviors from their peers (Sutherland 1937; Akers 1997). Similarly, control theory argues that young people who exhibit lawviolating beliefs and attitudes are more likely to become gang members (Esbensen et al. 1993; Hill et al. 1999). Thornberry et al. (2003) described these psychological processes as facilitators of gang membership within the framework of their interactional theory. In their research, they discuss delinquent beliefs (defined as the belief that it is acceptable to be delinquent) as causes, correlates, and consequences of delinquent behavior and gang membership. These beliefs play a dynamic role developmentally. They have been found to interact reciprocally with associations with

9


155

delinquent peers and delinquent behavior (Thornberry et al. 1994). However, there continues to be a lack of clarity about the types of psychological processes inherent in gang membership, and so these beliefs need closer examination, especially since they are more resistant to intervention (Hollin et al. 2002). Interactional theory provides an interesting way of understanding how circumstances, relationships, and behavior are intertwined within the gang structure. One of its strengths is that it takes into account that not all gang members were delinquent prior to joining a gang, and, most importantly, that once a member, delinquency significantly increases. This leaves room for incorporating group processes into the explanation and understanding of gang involvement and gang-related crime. Also, an individual’s offending history notwithstanding, interactional theory helps explain how such a person may become even more delinquent after joining a gang. Research supports this inevitable escalation in delinquency (Tita and Ridgeway 2007) while the theoretical framework dictates that any contributing factor to gang membership cannot be labeled strictly as a cause or an effect, but must be viewed as having a reciprocal relationship with other factors (Thornberry et al. 2003). However, interactional theory also has its limitations since it does not clarify why youth may leave a gang or why some youth choose not to join a gang in the first place. Neither does it provide insight into the process of leaving a gang. Nonetheless, interactional theory does provide an excellent framework for embarking on an examination of the psychology of gang membership.

9.3

Current Study

The following sections center around research conducted by the authors in the UK. Every secondary school in the 32 London Boroughs was approached for access to recruit participants (approximately 300). As a result, a sample totaling 798 high school students from five schools were surveyed on their group/gang affiliations, social and environmental characteristics, attitudes toward authority, perceptions about social status, and moral disengagement (including the aforementioned eight moral disengagement strategies). The mean age of the sample was 14.3 years (SD = 1.74, range = 12–18) with 566 boys (71%) and 231 girls (29%). A large proportion of the sample reported that both parents were born in the UK (50%), 14% reported that one parent was UK born and the other was not, and 36% reported that both parents were immigrants to the UK. For the sake of clarity, the following terminology regarding our research will be used: gang members (N = 59) were identified according to the Eurogang network’s definition (Weerman et al. 2009); peripheral youth (N = 75) were identified based on a two-cluster analysis of the remaining participants’ responses to their group’s durability, street orientation, and criminal identity; gang-involved youth (N = 134) consisted of both gang members and peripheral youth; and nongang youth (when compared to gang-involved youth, N = 664; when compared to fully fledged gang members, N = 739) were the remaining participants who did not fit any of the criteria (for further detail, see Alleyne

156


and Wood 2010). The findings from the data analyses fit into four main themes examining (1) the impact of neighborhood gangs, (2) gang structure and intragroup processes, (3) attitudes toward authority, and (4) the psychological processes specific to gang-related crime.

9.4

Neighborhood Gangs: The Psychological Effects

Interactional theory posits that a learning environment facilitative of gang behavior contributes to a young person’s decision to join a gang (Thornberry et al. 2003). Gangs tend to thrive in socially and economically disadvantaged communities (Spergel 1995; Rizzo 2003; Thornberry et al. 2003; Klein and Maxson 2006). Cloward and Ohlin (1960) attributed this relationship between low socioeconomic status and gang development to differential opportunity. That is, young people from poor neighborhoods turn to gangs because of social barriers obstructing them from legitimate opportunities outlined by middle-class norms. As a result, the presence of neighborhood gangs increases the risk that youth will join a gang (Thornberry et al. 2003). Ralphs and colleagues (2009) presented cases where nongang youth who reside in known gang neighborhoods are just as likely to be approached by police who assume that they are affiliated with local gangs (see also, Aldridge et al. 2012; Smithson et al. 2012). As a result of this negative contact, some youth felt provoked into joining a gang (Ralphs et al. 2009), and so it seems that the mere presence of neighborhood gangs coupled with the assumptions that some legitimate authorities seem to make can motivate youth to join gangs. Similarly, Pitts (2007) argued that some youth become “reluctant gangsters,” joining gangs because they fear being victimized by local gangs. Ultimately, the psychological pushes and pulls that motivate youth to join gangs are the consequences, at least in part, of the in-group/outgroup distinctions developed within gang neighborhoods (Klein and Maxson 2006). In the eyes of these youth, the gang protects them from being victimized by other local gangs (Decker and Van Winkle 1996; Pitts 2007). And police responses only serve to reinforce the gang identities youth adopt (Ralphs et al. 2009). In our first process of analyses, we identified 134 youth as having some level of gang involvement and 664 youth who showed no gang involvement. The findings (see Alleyne and Wood in press) support past literature (e.g., Thornberry et al. 2003) that neighborhood gangs do leave youth at an increased risk of joining a gang. This was exemplified in a logistic regression where out of five predictor variables (parental management, deviant peer pressure, school commitment, individual delinquency, and neighborhood gangs), individual delinquency and neighborhood gangs were significant predictors. Structural equation modeling further showed that neighborhood gangs burden appropriate parenting practices that might prevent gang involvement (e.g., parental management; see Fig. 9.1). This can result from one of the following (1) the presence of neighborhood gangs is indicative of a low-income neighborhood; therefore, parents are more likely to work exceedingly long hours leading to reduced parental supervision and (2) there is an increased likelihood that

9


157

Deviant peer pressure 0.47*

0.01 -0.08***

Neighbourhood gang

0.35*** 0.12*** -0.22***

Gang involvement

-1.19* 0.86*

-0.31

R2 = 0.05 Individual delinquency

-0.21*** Parental management

0.01**

-0.27*** 0.22*** School commitment

Fig. 9.1 Standardized estimates for the modified model predicting gang involvement. N = 798, chi-square = 0.35, df = 2, p = 0.838, CFI = 1.00, GFI = 1.00, NFI = 1.00, RMSEA = 0 (0–0.04) * p < 0.05. **p < 0.01. ***p < 0.001

parents are current/former gang members in gang neighborhoods (Klein and Maxson 2006). Young people who would normally be expected to look to their parents for guidance and support are often left to their own devices. This occurs when parents provide inadequate monitoring, supervision, and/or discipline (Thornberry et al. 2003; Klein and Maxson 2006). As a result, it seems that because many young people have to cope with adverse circumstances on their own, they become more vulnerable to the pressures from antisocial peers and that this can lead to their gang membership. Once in the gang, all other social groups then become the out-group and are viewed with negativity. In summary, and in accordance with social psychological concepts embedded in mainstream theories such as social identity theory (Tajfel and Turner 1979) and social comparison theory (Festinger 1954), once a person has identified with a particular group, in-group and out-group bias develops. Cooper and Fazio (1979) suggested that a symptom of such memberships is vicarious personalism: “the perception by members of one group that another group’s actions are aimed at and intended for them” (p. 151). So other out-group actions (e.g., the police and other gangs) may be perceived with negativity. In turn, this may facilitate group cohesion because psychologically a gang will help emphasize socio-cognitive processes for both in-group favoritism and out-group discrimination for each member (Goldstein 2002).

9.5

Gang Structure and Intragroup Processes

The structure and hierarchy of a gang is a topic of debate (Spergel 1995). Researchers have struggled to identify and label the different levels of youth involvement in gangs. This is probably because “no two gangs are just alike” (Thrasher 1927, p. 36). However, an overall consensus suggests that there are at least two levels of

158


membership: leaders and core members (gang members) and those with a tangential role (peripheral youth) (Patrick 1973; Klein and Maxson 1989; Spergel 1995). However, importantly, it is not necessary to be a full gang member in order to experience the effects of gang membership (Curry et al. 2002). In our second round of data analyses, 59 high school students were identified as gang members, 75 were identified as peripheral youth whereby their level of involvement could be exemplified by how they describe their group (i.e., limited knowledge regarding the criminal capacity of the group), and 664 were identified as nongang youth (Alleyne and Wood 2010). Coupled with the hierarchical structure of a gang is an intragroup process that informs the role of each member. For example, participants’ responses on the perceived importance of social status (range = 15–80; South and Wood 2006) showed that both gang (M = 58.93) and peripheral (M = 57.31) youth value status more than their nongang counterparts (M = 54.03; F(2, 793) = 5.26, p < 0.01, partial h2 = 0.01). This suggests that peripheral youth, who are generally younger than gang members (Decker and Van Winkle 1996; Alleyne and Wood 2010), may become involved in a gang to acquire the status they see afforded to core gang members. In other words, status, just as Klein (1995) suggested, seems to be one of the motives for joining a gang. It can be argued that peripheral youth are either trying to move up the ranks of gang membership, are associated merely via friendships, or they are youth who are in the process of desistance (Curry et al. 2002). For those that are motivated to join, they may feel a need to prove themselves to the gang by mimicking what they perceive as acceptable gang behavior (Hughes and Short 2005; Przemieniecki 2005) which may be why peripheral youth (M = 4.26) are more violent than nongang youth (M = 3.87; F(2, 792) = 3.18, p < 0.05; partial h2 = 0.01; range = 3–15; Alleyne and Wood 2010). On the other hand, gang members (M = 4.18) were not more violent than peripheral youth or nongang youth. This finding is contrary to previous work that has found core members to be more violent than peripheral and nongang youth both in street gangs (Curry et al. 2002; Esbensen et al. 2001; Klein 1995) and prison gangs (Gaes et al. 2002). The reason for this inconsistency may be because in this local context, gang members’ status in the gang means they can delegate some acts of violence to more junior (i.e., peripheral youth), which indicates that the gang’s violent activity is shared among all members. Short and Strodtbeck (1965) noted that gang leaders do not necessarily have to maintain their status through aggressive, dominance-seeking behavior, but rather coordinated and nurturant manipulating of their peripheral and fringe associates. Our findings also show that peripheral youth (M = 10.54) displace responsibility (as measured in the Bandura et al. (1996) moral disengagement scale) more than nongang youth (M = 9.57; F(2, 793) = 3.05, p < 0.05, partial h2 = 0.01; range = 4–20). This is in accordance with the concepts of moral disengagement; peripheral youth displace more responsibility for their own behavior (including violent behavior), arguably onto authority figures. This seems to suggest that they hold others (e.g., legitimate authorities such as the police, schools, and parents) responsible for their own actions. Equally, peripheral youth may even displace responsibility onto higher-ranking gang members (Alleyne and Wood 2010). Roles within the gang may be explicitly expressed

9


159

via titles (i.e., lieutenants, soldiers, etc.) or implicitly via reputation (Spergel 1995). Peripheral youth, trying to earn a core place in the gang, probably obey instructions from those higher up the gang structure. However, because they are obeying orders, they feel less responsible for their actions. In other words, “I have to do this or I won’t be accepted.” Interestingly, gang members (M = 10.17), as assessed using the moral disengagement scale, did not displace responsibility more than nongang youth. This suggests that core members, unlike peripheral members, are as inclined to take personal responsibility as nongang youth. We cannot yet say with certainty why this is the case, but it might be that even though gang members and nongang youth take equal responsibility for their actions, they do so for different behaviors. For example, gang members may take responsibility for peripheral youths’ actions because they give out the orders and are thus more responsible for the gang’s behavior per se. Lastly, it can be argued that the individual identity of these young people merges with the gang identity. Gang members (M = 9.12) used euphemisms more than did nongang youth (M = 8.24; F(2, 793) = 3.71, p < 0.05, partial h2 = 0.01; range = 4–20), and gang members (M = 10.85) attributed more blame to their victims than did nongang youth (M = 10.85; F(2, 793) = 4.28, p < 0.05, partial h2; range = 4–20). Interestingly, peripheral youth (M = 8.92) did not use euphemisms more than nongang youth or less than gang members. Neither did they (M = 10.14) blame victims more than nongang youth, nor did they blame victims less than gang members. These findings are indicative of a developmental process whereby youth who become more gang-involved gradually adopt gang norms in terms of language and attitudes toward those they victimize. For example, in accordance with moral disengagement strategies, it is acceptable for gang members to blame situations or victims but not in-group members (Alleyne and Wood 2010). These intragroup processes reflect one of the fundamental aspects of interactional theory: bidirectional causality (Thornberry et al. 2003). This reciprocity highlights the importance of individual differences in gang research because the causal direction for these psychological processes is likely to vary from one gang member to the next.

9.5.1

The Influence of Antiauthority Attitudes in Gang Membership

As previously discussed, when a young person joins a group, they adopt a new social identity, which is accompanied by an adoption of attitudes and beliefs that favor the in-group (Tajfel and Turner 1979; Turner 1982). Research shows how gang members hold more negative attitudes toward authority than nongang youth (Kakar 2005; Lurigio et al. 2008), and when primed in their gang identities, their antiauthority attitudes increase (Khoo and Oakes 2000). This process may also be fostered, as previously discussed, by living in gang neighborhoods and having negative contact with authority figures (Ralphs et al. 2009). This negative predisposition can manifest itself in two ways (1) gang members interpret/perceive encounters

160


with authority figures as negative, which prevents a constructive and productive relationship between these youth and authority figures, and (2) gang members attract unwanted attention from authority by committing crime, which in turn, reinforces their negative relations (Rosenbaum et al. 2005). This shows that attitudes toward authority are important in terms of the risk they pose for gang membership. It also shows that the interaction between authority figures and young people is important especially since we also know that positive attitudes toward authority can act as protective factors against gang membership (Thornberry 1996). Our recent findings show that even if gang members and nongang members experience similar levels of deviant peer pressure, if youth are subject to deviant peer pressure and they also hold antiauthority attitudes, then they are more likely to become gang members (Alleyne 2010). We also found this to be the case for youth who were inclined to attribute blame to victims or circumstances—the more they also held antiauthority attitudes, the more likely they were to become gang members. These findings suggest that antiauthority attitudes play a very important role in the motivation to become gang members. Quite why this is so is not wholly clear at this stage of research. Arguably, authority figures could be seen as an out-group by gang members (Emler and Reicher 1995; Tarry and Emler 2007), and as a result, gang identity would include antiauthority attitudes that facilitate and justify the gang’s offending behavior (Alleyne 2010; Vigil 1988). Also, Flexon et al. (2009) found that young people’s views of police officers are significantly shaped by peer experiences of the police and not necessarily by direct personal experience. Therefore, some gang members may never have had direct contact with authority figures but still perceive them negatively. Thus, authority figures need to be aware of the “trickle down” effect of their actions and, more importantly, that even youth they have not directly encountered may hold antiauthority views. Furthermore, if young people learn these negative attitudes toward authority from their peers (or even parents), joining a gang could be perceived as their “behavioural expression of alienation from formal authority” (Emler and Reicher 2005, p. 221). Theoretically speaking, a facilitation effect could account for this relationship (Thornberry 1987; Thornberry and Krohn 2001). That is, before engaging with these deviant peers (who could arguably be gang peers), young people may not actually hold antiauthority attitudes or may even hold proauthority attitudes. Given the importance of antiauthority attitudes to gang membership, it seems that if more positive attitudes to authority can be reinforced and maintained, young people will be less likely to join gangs and commit crime (Thornberry 1996).

9.5.2

The Behavioral Outcomes of Gang Membership

Gang literature is explicit that the main consequence of gang involvement is the increase in criminal behavior over and beyond any other types of delinquent youth (Tita and Ridgeway 2007). Gang research in the UK has found that typical gangrelated crimes include: robbery (i.e., street robbery of mobile phones, wallets, etc.),

9


161

drug supply, weapons possession (Bennett and Holloway 2004; Bullock and Tilley 2002, 2008), and also making threats and property damage including graffiti (Alleyne 2010). Since some gangs tend to be territorial (Klein and Maxson 2006; Tita and Ridgeway 2007), graffiti is a typical manifestation of gangs “stamping ownership” on their territory (Spergel 1995). Spergel (1995) also argued that to protect their territory, gangs also use intimidation, and this has been exemplified in gang members’ use of threats over and above nongang youth (Alleyne 2010). Therefore, gangs mark their territory with graffiti and threaten the use of violence in order to protect it. Territorial conflict has been linked with gang violence (Spergel 1995); therefore, such threats have potentially violent consequences. From a theoretical standpoint, it was previously mentioned that a learning environment facilitative of gang behavior contributes to a young person’s decision to join a gang (Thornberry et al. 2003). It has been found that the presence of neighborhood gangs predicts gang-related crime (Alleyne 2010). This was exemplified in a standard multiple regression, which showed that of seven predictor variables (antiauthority attitudes, perceived importance of social status, moral disengagement, parental management, deviant peer pressure, individual delinquency, and neighborhood gangs), individual delinquency and neighborhood gangs were the most significant predictors of gang-related crime (an interaction variable between group crime [a continuous variable measured using the Eurogang questionnaire] and gang membership). This finding could add support to existing literature whereby young people are at a greater risk of joining a gang, and they are more likely to get involved in gang-related crime if they are already delinquent and live in neighborhoods with an existing gang presence (Thornberry et al. 2003). However, due to the cross-sectional nature of the data, we cannot say for certain. There are several explanations as to why this may occur. For example, territorial proximity has been found to instigate gang conflict (Spergel 1995). Therefore, different gangs in relative proximity to each other may compete with each other for resources (e.g., drugs, weapons, etc.) or territories, or worse, engage in intergang violence. Alternatively, neighborhoods peppered with gangs may make youth fearful of victimization (Pitts 2007), and these findings suggest that a motivation for joining a gang is for protection. Recent research has examined the relationships between psychological processes and gang-related crime. For example, Alleyne (2010) found that moral disengagement as a whole predicted gang-related crime, but only when antiauthority attitudes were high. Consequently, it seems that youth may disengage their moral standards if they develop antiauthority attitudes. This suggests that antiauthority attitudes may be acting as a cognitive strategy employed to enable youth to engage in gang-related crime. Furthermore, Alleyne (2010) also found that it was only when antiauthority attitudes were high, the perceived importance of social status predicted gang-related crime. That is, it is particularly important to a gang member’s reputation and criminal activity that they hold antiauthority attitudes. Part of the appeal of gang involvement is the message that the gang represents, a message of survival from adversity, social disadvantage, and marginalization (Emler and Reicher 2005; Hagedorn 2005). And in many cases, authority figures are seen as the culprits that enforce this marginalization (Ralphs et al. 2009); thus, the appropriate response would be to

162


break the law. In addition to Alleyne and Wood’s (2010) finding that peripheral youth are more violent and value status more than nongang youth, this research suggests that status is earned by the criminal activity young people engage in. Consequently, gangs recruit members who share the same value system that again, supports continuing delinquency.

9.6

Theoretical Development and Future Directions

Because individual differences are clearly important in gang membership, as outlined above, we need to consider psychological factors more when developing theoretical explanations of gang membership. For instance, while it is inherently useful to understand why youth might join gangs, comprehensive explanations of gang membership would also explain why other youth might not join gangs or indeed, the reasons why they might become gang members and then leave the gang. The unified theory of gang involvement (Wood and Alleyne 2010; see Fig. 9.2) considers these possibilities. The model shows how youth may take a pathway into criminal activity, which does not necessarily include gang membership. Criminal activity may occur independently of, or simultaneously to, joining a gang. However, gang membership is likely to occur for reasons over and above those underlying involvement in criminal activity. Gang membership offers additional protection, possibly from threats stemming from competing criminal out-groups (e.g., rival gangs); it provides social support and offers elevated status, the chance to acquire power, and opportunities for excitement. Gang membership may also bring with it sets of rules or new social controls that members are expected to abide by, perhaps in the form of adopting similar attitudes, beliefs, and cognitions—in this way, providing a form of familial environment. As a gang member, the youth is exposed to further opportunities for criminal learning, and he or she will become even more involved in criminal activity. Of course, hand in hand with these new opportunities for “personal enhancement” come additional chances of victimization, and these may lead gang member youth to desire a gang-free life.

9.6.1

Desistance

As mentioned previously, interactional theory’s main weakness is its neglect of the processes surrounding gang desistance. Thornberry and colleagues (2003) formulated a model of causal processes whereby neighborhood, family, school, and peer influences all play a role in the cumulative risk of gang involvement. Wood and Alleyne (2010) developed this further by arguing that perhaps desistance from gang involvement occurs when prosocial opportunities and institutions are restored (i.e., employment, stable relationships, etc.). Of course, these opportunities may be adversely affected if the youth has been caught and prosecuted for criminal acts. In this case, the youth’s criminal inclination will either dissipate (from fear of further

9


Individual Characteristics Psychopathy Hyperactivity Anxiety IQ Mental health problems

Social factors Social control formal/informal Family bonds School failure/success

163

Environment Dis/organized neighbourhood Family structure

Social cognition Perception of gangs Perceived opportunities Strain Perceived hostility Fear of victimization Attitudes to authority

Selection of peers Shared values Pro/anti social attitude development

Opportunities for new social controls E.g.. employment Marriage Parenthood

Reinforcement of new social controls Relationship/ job

No crime

Criminal activity Reinforcement of offending

Gang Membership Protection Social support Acquisition of Status Acquisition of power Excitement New social controls

Opportunity for criminal learning Social cognitive development i.e. Moral disengagement pro-aggressive cognitive schemas

Breakdown of new social controls Lose job Relationship breakdown

Desistance Stop offending Relinquish gang membership

Fig. 9.2 Wood and Alleyne’s (2010) unified theory of gang involvement

legal repercussions) or strengthen (from the obstruction of legitimate opportunities that prosecution creates). If the newly acquired social controls are reinforced (e.g., opportunities to advance in employment), the youth’s resolve to desist from crime may strengthen, and desistance will continue. If, however, they break down (e.g., employment is lost or a relationship breaks up), then the youth may return to his/her previous lifestyle (i.e., criminal involvement and/or gang membership).

164


Although Wood and Alleyne’s (2010) unified theory is in the very early stages of development, it has the potential to expand research findings regarding gang membership and delinquency at both a psychological and a criminological level. Because it includes concepts of noninvolvement in crime and gangs and concepts of desistance, it allows us to make meaningful comparisons. And, as Klein (2006) so rightly observes, comparisons are all too rare in the gang literature. We can make comparisons between gang members, between abstaining and remaining gang members, and between gang and nongang members. It is also possible to make comparisons between neighborhoods by examining the individual characteristics, social factors, and social cognitions of youth living in organized and disorganized neighborhoods. We do not suggest that this model is a panacea to all the gaps in the literature, but it is a starting point, and it can be developed and expanded as findings based on its concepts shed light on old and new ideas. Most importantly, it presents the integration of gangrelated concepts into a coherent structure that combines criminological and psychological ideas. For example, the socio-cognitive processes that feed into peer selection and attitude development can be explicated within the theoretical framework of the unified theory. This structure provides guidance for future empirical study.

9.7

Overall Discussion

The purpose of this chapter was to discuss the role that psychology has and can play in explaining the inter- and intragroup processes within the gang structure and hierarchy. The main premise is that the impact of psychological processes within and between groups dictates the behavioral outcomes of the group members. For example, the extensive role of antiauthority attitudes in gang membership and gangrelated crime appears to be seamlessly integrated into gang culture. The findings of past research (Emler and Reicher 2005; Alleyne and Wood 2010) show that these attitudes provide young people with a common purpose, an outlet for their frustration, and a platform to express themselves (i.e., delinquent behavior; Emler and Reicher 2005). There is still more research that needs to be done concerning the group dynamics in and around the gang culture. This is an area on which psychology can shed some light considering its depth of empirical research examining groups. For example, research examining the extent to which group processes distinguish transient from stable membership could perhaps pay dividends in the development of intervention programs. Also, “early thinking held that individual prejudice was the primary source of inter-group conflict” (Goldstein 2002, p. 124). Therefore, in order to reduce intergroup conflict, there needs to be intergroup contact within preconceived situational conditions; this resulted in the intergroup contact theory (Allport 1954; see Pettigrew 2008, for review). Essentially, Allport (1954) argued that intergroup contact can reduce intergroup conflict if four conditions were met (1) equal group status within the contact situation, (2) shared common goals, (3) attainment of shared goals via intergroup cooperation, and (4) support from relevant institutional

9


165

groups (e.g., authorities). As this chapter has shown, if we are to consider these important elements of intergroup contact, they need to consider the impact that status and authority-related attitudes may have on intergroup relations. Spergel (1995) argued that the impact of gangs can be broken down into four key components: individual, group, behavior, and context. Previous intervention strategies have been designed to address some, but not all, of these components. Intergroup contact theory and the unified theory of gang involvement could be used to develop further empirical work on gang membership, which may, in turn, contribute to the development of intervention programs to reduce gang involvement. Ultimately, a comprehensive, multidisciplinary approach to the study of gangs can only enlighten us and perhaps inspire us to combat the effect gangs have at an individual and societal level more effectively.

References Akers RL (1997) Criminological theories: introduction and evaluation, 2nd edn. Roxbury, Los Angeles Aldridge J, Medina J, Ralphs R (2012) Counting gangs: conceptual and validity problems with the Eurogang definition. In: Esbensen F-A, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Alleyne E (2010) Gang membership: behavioural, social and psychological characteristics. Unpublished doctoral dissertation,University of Kent, Canterbury, UK Alleyne E, Wood JL (2010) Gang involvement: psychological and behavioral characteristics of gang members, peripheral youth and non-gang youth. Aggress Behav 36:423–436 Alleyne E, Wood JL (in press) Gang involvement: social and environmental factors. Crime & Delinquency Allport GW (1954) The nature of prejudice. Addison-Wesley, Reading Anderson E (1999) Code of the street: decency, violence and the moral life of the inner city. Norton and Company, New York Bandura A (1990) Selective activation and disengagement of moral control. J Soc Issues 46: 27–46 Bandura A (2002) Selective moral disengagement in the exercise of moral agency. J Moral Educ 31:101–119 Bandura A, Barbaranelli C, Caprara GV, Pastorelli C (1996) Mechanisms of moral disengagement in the exercise of moral agency. J Pers Soc Psychol 71:364–374 Bennett T, Holloway K (2004) Gang membership, drugs and crime in the UK. Br J Criminol 44:305–323 Bullock K, Tilley N (2002) Shootings, gangs and violent incidents in Manchester: developing a crime reduction strategy, vol 13, Crime reduction research series. Home Office, London Bullock K, Tilley N (2008) Understanding and tackling gang violence. Crime Prev Community Safety 10:36–47 Cloward R, Ohlin L (1960) Delinquency and opportunity. Free Press, New York Cooper J, Fazio RH (1979) The formation and persistence of attitudes that support intergroup conflict. In: Austin WG, Worchel S (eds) The social psychology of intergroup relations. Wadsworth, Belmont, pp 149–159 Craig WM, Vitaro F, Gagnon C, Tremblay RE (2002) The road to gang membership: characteristics of male gang and non-gang members from ages 10 to 14. Soc Dev 11:53–68 Curry GD, Decker SH, Egley A (2002) Gang involvement and delinquency in a middle school population. Just Quart 19:275–292

166


Decker SH, Van Winkle B (1996) Life in the gang: family, friends, and violence. Cambridge University Press, New York Dukes RL, Martinez RO, Stein JA (1997) Precursors and consequences of membership in youth gangs. Youth Soc 29:139–165 Dupéré V, Lacourse É, Willms JD, Vitaro F, Tremblay RE (2007) Affiliation to youth gangs during adolescence: the interaction between childhood psychopathic tendencies and neighborhood disadvantage. J Abnorm Child Psychol 35:1035–1045 Emler N, Reicher S (1995) Adolescence and delinquency. Blackwell, Oxford Emler N, Reicher S (2005) Delinquency: cause or consequence of social exclusion? In: Abrams D, Marques J, Hogg M (eds) The social psychology of inclusion and exclusion. Psychology Press, Philadelphia, pp 211–241 Esbensen F-A, Tuisinki K (2007) Youth gangs in the print media. J Crim Just Pop Cul 14:21–28 Esbensen F-A, Weerman FM (2005) Youth gangs and troublesome youth groups in the United States and the Netherlands: a cross-national comparison. Eur J Criminol 2:5–37 Esbensen F-A, Huizinga D, Weiher AW (1993) Gang and non-gang youth: differences in explanatory factors. J Contemp Crim Just 9:94–116 Esbensen F-A, Winfree LT Jr, He N, Taylor TJ (2001) Youth gangs and definitional issues: when is a gang a gang, and why does it matter? Crime Delinq 47:105–130 Festinger L (1954) A theory of social comparison processes. Hum Relat 7:117–140 Flexon JL, Lurigio AJ, Greenleaf RG (2009) Exploring the dimensions of trust in the police among Chicago juveniles. J Crim Just 37:180–189 Gaes GG, Wallace S, Gilman E, Klein-Saffran J, Suppa S (2002) The influence of prison gang affiliation on violence and other prison misconduct. Prison J 82:359–385 Gatti E, Tremblay RE, Vitaro F, McDuff P (2005) Youth gangs, delinquency and drug use: a test of the selection, facilitation, and enhancement hypotheses. J Child Psychol Psyc 46:1178–1190 Goldstein AP (1991) Delinquent gangs: a psychological perspective. Research Press, Champaign Goldstein AP (2002) The psychology of group aggression. John Wiley, West Sussex Gordon RA, Lahey BB, Kawai E, Loeber R, Stouthamer-Loeber M, Farrington DP (2004) Antisocial behavior and youth gang membership: selection and socialization. Criminol 42:55–87 Hagedorn JM (2005) The global impact of gangs. J Contemp Crim Just 21:153–169 Hill KG, Howell JC, Hawkins JD, Battin-Pearson SR (1999) Childhood risk factors for adolescent gang membership: results from the seattle social development project. J Res Crime Delinq 36:300–322 Hill GH, Lui C, Hawkins JD (2001) Early precursors of gang membership: a study of Seattle youth, Juvenile Justice Bulletin. US Department of Justice, Office of Justice Program, OJJDP, Washington Hirschi T (1969) Causes of delinquency. University of California Press, Berkeley/Los Angeles Hollin CR, Browne D, Palmer EJ (2002) Delinquency and young offenders. Blackwell, Oxford Howell JC, Egley A (2005) Moving risk factors into developmental theories of gang membership. Youth Violence Juvenile Justice 3:334–354 Hughes LA, Short JF Jr (2005) Disputes involving youth street gang members: micro-social contexts. Criminol 43:43–76 Kakar S (2005) Gang membership, delinquent friends and criminal family members: determining the connections. J Gang Res 13:41–52 Khoo ACE, Oakes PJ (2000) The variability of the delinquent self: anti-authority attitudes and endorsement of neutralization techniques among incarcerated delinquents in Singapore. Asian J Soc Psychol 3:125–132 Klein MW (1995) The American street gang. Oxford University Press, New York Klein MW (2006) The value of comparisons in street gang research. In: Short JF, Hughes LA (eds) Studying youth gangs. Altamira Press, Oxford, pp 129–143 Klein MW, Maxson CL (1989) Street gang violence. In: Weiner NA, Wolfgang NW (eds) Violent crime violent criminals. Sage, Newbury Park, pp 198–237

9


167

Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, New York Knox GW (1994) An introduction to gangs. Wyndham Hall Press, Bristol Lacourse E, Nagin DS, Vitaro F, Côté S, Arseneault L, Tremblay RE (2006) Prediction of earlyonset deviant peer group affiliation: a 12-year longitudinal study. Arch Gen Psychiatry 63: 562–568 Lahey BB, Gordon RA, Loeber R, Stouthamer-Loeber M, Farrington DP (1999) Boys who join gangs: a prospective study of predictors of first gang entry. J Abnorm Child Psychol 27: 261–276 LeBlanc M, Lanctot N (1998) Social and psychological characteristics of gang members according to the gang structure and its subcultural and ethnic makeup. J Gang Res 5:15–28 Lurigio AJ, Flexon JL, Greenleaf RG (2008) Antecedents to gang membership: attachments, beliefs, and street encounters with the police. J Gang Res 4:15–33 Patrick J (1973) A Glasgow gang observed. Eyre Methuen, London Pettigrew TF (2008) Future directions for intergroup contact theory and research. Int J Intercult Rel 32:187–199 Pitts J (2007) Reluctant gangsters: the changing shape of youth crime. Willan Publishing, Devon Przemieniecki CJ (2005) Gang behavior and movies: do hollywood gang films influence violent gang behavior? J Gang Res 12:41–71 Ralphs R, Medina J, Aldridge J (2009) Who needs enemies with friends like these? The importance of place for young people living in known gang areas. J Youth Stud 12:483–500 Rizzo M (2003) Why do children join gangs? J Gang Res 11:65–74 Rosenbaum DP, Schuck AM, Costello SK, Hawkins DF, Ring MK (2005) Attitudes toward police: the effects of direct and vicarious experience. Police Quart 8:343–365 Sharp C, Aldridge J, Medina J (2006) Delinquent youth groups and offending behaviour: findings from the 2004 Offending, Crime and Justice Survey (Home Office Online Report 14/06). Retrieved from Home Office website: http://tna.europarchive.org/20100413151441/http:// www.homeoffice.gov.uk/rds/pdfs06/rdsolr1406.pdf Short JF Jr, Strodtbeck FL (1965) Group process and gang delinquency. University of Chicago, Chicago Smithson H, Monchuk L, Armitage R (2012) Gang member: who says? Definitional and structural issues. In: Esbensen F-A, Maxson CL (eds) Youth Gangs in international perspective: tales from the Eurogang program of research. Springer, New York South R, Wood J (2006) Bullying in prisons: the importance of perceived social status, prisonization and moral disengagement. Aggress Behav 32:490–501 Spergel IA (1995) The youth gang problem. Oxford University press, New York Sutherland EH (1937) The professional thief. University of Chicago Press, US Sykes G, Matza D (1957) Techniques of neutralization: a theory of delinquency. Am Sociol Rev 22:664–670 Tajfel H, Turner JC (1979) An integrative theory of social conflict. In: Austin W, Worchel S (eds) The social psychology of intergroup relations. Brooks/Cole, Monterey, pp 33–48 Tarry H, Emler N (2007) Attitudes, values and moral reasoning as predictors of delinquency. Br J Dev Psychol 25:169–183 Thornberry TP (1987) Toward an interactional theory of delinquency. Criminol 25:863–891 Thornberry TP (1996) Empirical support for interactional theory: a review of the literature. In: Hawkins JD (ed) Delinquency and crime: current theories. Cambridge University Press, New York, pp 198–235 Thornberry TP, Krohn D (2001) The development of delinquency: an interactional perspective. In: White SO (ed) Handbook of youth and justice. Plenum, New York, pp 289–305 Thornberry TP, Krohn MD, Lizotte AJ, Chard-Wierschem D (1993) The role of juvenile gangs in facilitating delinquent behavior. J Res Crime Delinq 30:55–87 Thornberry TP, Lizotte AJ, Krohn MD, Jang SJ (1994) Delinquent peers, beliefs, and delinquent behavior: a longitudinal test of interactional theory. Criminol 32:47–83

168


Thornberry TP, Krohn MD, Lizotte AJ, Smith C, Tobin K (2003) Gangs and delinquency in developmental perspective. Cambridge University Press, Cambridge Thrasher F (1927/1963) The gang: a study of 1,313 gangs in Chicago. University of Chicago Press, Chicago Tita G, Ridgeway G (2007) The impact of gang formation on local patterns of crime. J Res Crime Delinq 44:208–237 Turner JC (1982) Towards a cognitive redefinition of the social group. In: Tajfel H (ed) Social identity and intergroup relations. Cambridge University Press, Cambridge, pp 15–40 Vigil JD (1988) Barrio gangs: street life and identity in Southern California. University of Texas Press, Austin Weerman FM, Maxson CL, Esbensen F, Aldridge J, Medina J, van Gemert F (2009) Eurogang program manual background, development, and use of the Eurogang instruments in multi-site, multi-method comparative research. Retrieved from the Eurogang Network website: http:// www.umsl.edu/~ccj/eurogang/Eurogang_20Manual.pdf Wood J, Alleyne E (2010) Street gang theory and research: where are we now and where do we go from here? Aggress Violent Beh 15:100–111

Chapter 10

The Onset of (Euro)Gang Membership as a Turning Point in the Life Course Chris Melde* and Finn-Aage Esbensen

Gangs and their members are notorious for their involvement in acts of crime and delinquency. It is this reality that makes them both interesting and worthy of attention. Without this connection between such groups and deviance, gangs would rarely be a topic of discussion. The fact of the matter is, however, that gangs have been attributed a causal role in the genesis of crime. That is, gangs, it is said, produce criminal and delinquent behaviors that would not otherwise exist but for these groups. This led Egley et al. (2006) to conclude that “there is something unique [emphasis added] about gang membership itself that increases youths’ participation in serious and violent crime” (p. 224). This statement led us to ask the seemingly obvious question: What is that unique something which causes individual gang members to participate in crime and delinquency? As we will discuss below, answering this seemingly simple question was not so straightforward. After all, there is no universal definition for a gang member, so identifying who is gang involved is problematic (see Matsuda et al. 2012). Complicating things even more is the fact that individuals self-select into gangs on a nonrandom basis, meaning those youth who decide to join such groups likely have distinctive personal histories that set them apart from their peers. This chapter builds on the work of Melde and Esbensen (2011), who used Laub and Sampson’s (2003) age-graded theory of informal social control to explain the *

This is a revised version of a paper presented at the 2010 Eurogang Workshop in Neustadt/ Weinstrasse, Germany. This research was funded by the National Institute of Justice, Office of Justice Programs, US Department of Justice, Award No. 2003-JN-FX-0003. Points of view in this document are those of the authors and do not necessarily represent the official position of the US Department of Justice. Direct correspondence to Chris Melde, School of Criminal Justice, Michigan State University, 560 Baker Hall, East Lansing, MI 48824.

C. Melde (*) School of Criminal Justice, Michigan State University, East Lansing, MI, USA e-mail: [email protected] F.-A. Esbensen Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_10, © Springer Science+Business Media, LLC 2012

169

C. Melde and F.-A. Esbensen

170

processes emanating from gang membership that partially explain why involvement in such groups appears to increase involvement in crime and delinquency. In particular, we conceptualize the onset of gang membership as a turning point that impacts at least three aspects of the lives of youth: their relationships with peers, school, and family; their routine activities; and the way they think about themselves and their notion of right and wrong. In sum, being part of a gang impacts a number of important aspects of an adolescent’s life, and together, these changes can have an insidious impact on deviant behavior. Before proceeding to our analysis, we provide a brief overview of research on gangs and delinquency, Laub and Sampson’s (2003) life-course perspective, how gang membership can be viewed as a turning point in the lives of youth, and why assessing the impact of gang membership on delinquent involvement is difficult.

10.1

Gangs and Delinquency

A number of studies have examined the influence of gang membership on delinquency with the aim of explaining why gang members routinely report involvement in a disproportionately large amount of crime and delinquency. This literature has focused on the three general explanatory frameworks provided by Thornberry et al. (1993): the selection, facilitation, and enhancement models. The selection model is consistent with theories that explain criminal behavior as the product of relatively stable differences in criminal propensity between individuals (Glueck and Glueck 1950; Gottfredson and Hirschi 1990; Hirschi 1969). That is, bad kids, who are going to do bad things regardless of circumstance, join gangs and get in trouble. According to this view, gang membership has no effect on involvement in delinquency, as a common set of factors (i.e., criminal propensity) explains both delinquency and gang involvement. The facilitation model is consistent with social learning and opportunity perspectives whereby gang membership is believed to influence attitudes, norms, and routine activities associated with delinquent behavior, which in turn increases criminal involvement. Finally, the enhancement model blends the selection and facilitation models and suggests that gang members are likely more antisocial than nongang youth even before gang involvement, but the gang context further exacerbates these differences (Thornberry et al. 1993). Overall, evidence supports the enhancement model concerning the influence of gang membership on delinquency. As Thornberry and colleagues (2003) stated, “the young men and women who join gangs have multiple deficits in many developmental domains and being a member of a street gang further impedes their prosocial development” (p. 86). That is, while gang members typically have a history of deviant and/or delinquent behavior before joining a gang, involvement in such activities increases in frequency and severity after the onset of membership in such groups. To date, findings consistent with the enhancement model have been reported in American panel studies (Battin et al. 1998; Esbensen and Huizinga 1993; Gordon et al. 2004; Melde and Esbensen 2011; Peterson et al. 2004; Thornberry et al. 2003) as well as in Norwegian (Bendixen et al. 2006) and Canadian samples (Gatti et al. 2005).

10 The Onset of (Euro)Gang Membership as a Turning Point in the Life Course

171

Given the weight of evidence concerning the impact of gang membership on changes in antisocial behavior, even among youth already evincing antisocial tendencies, the next step for researchers is to explain the reasons for this observed change. To do so, Melde and Esbensen (2011) utilized Laub and Sampson’s (2003) age-graded theory of informal social control because of the explicit focus the theory had on both relatively stable differences among youth as well as mechanisms for change. That is, “a fundamental theme of [the] age-graded theory of informal social control and crime was that while individual traits and childhood experiences are important for understanding behavioral stability, experiences in adolescence and adulthood can redirect criminal trajectories in either a more positive or a more negative manner” (Laub and Sampson 2003, p. 6). In sum, Laub and Sampson (2003) provide a theory that is consistent with the evidence provided in previous work on gangs, in that antisocial youth who get involved with gangs appear to be steered on a more delinquent path than would be expected without such an experience. Using a sample of school-aged youth, Melde and Esbensen (2011) found concepts related to Laub and Sampson’s (2003) life-course theory helped to partially explain the impact of gang membership on delinquent involvement among self-identified gang members. Given our reliance on Melde and Esbensen (2011) as a guiding framework for the current study, we provide a brief overview of Laub and Sampson’s (2003) theory and how the experience of gang membership can be viewed as a particularly important turning point in the life course.

10.2

Laub and Sampson’s Life-Course Perspective

Laub and Sampson’s (2003) theory of crime and delinquency views the life course as a normatively patterned sequence of social transitions that are enacted over time (Sampson and Laub 1995). Their view of the life course is based on Elder’s (1985) conceptualization of trajectories and transitions whereby trajectories “refer to longterm patterns and sequences of behavior,” while transitions are “specific life events (e.g., first job or onset of crime) that are embedded in trajectories” and are usually rather abrupt in nature (Sampson and Laub 1995, p. 144). For the most part, the lives of youth follow a standard pattern with regard to behavioral trajectories and transitions. For instance, most youth are raised by loving parents who enroll them in schools at the appropriate age, and the student progresses from grade to grade through mid-adolescence where they begin to gradually spend more time away from the watchful eyes of parents and other authority figures (e.g., school personnel) and in the company of similar-aged peers (Warr 2002). Such a pattern is associated with a slight increase in deviant behavior as youth transition from high levels of monitoring by parents and school officials to more activities outside the purview of authority figures. Long-term patterns of behavior (i.e., trajectories), however, can be significantly altered by off-age transitions, such as dropping out of school, teenage parenthood, or experiencing traumatic life events. Such events, which Laub and Sampson (2003) refer to as “turning points,” have the potential to change long-term patterns of

172


behavior if they alter levels of informal social control, routine activities, and the decisions people make (i.e., human agency). Insofar as gang membership can be viewed as a turning point in the life course, we would expect that onset of gang membership would be associated with changes in these areas in a manner that frees youth to commit both more and more serious delinquent acts. Melde and Esbensen (2011) examined the applicability of the turning point framework to the study of gang membership. Results of their study suggested that the onset of gang membership was associated with changes in measures of informal social control, routine activities, anger identity, and beliefs about the appropriateness of crime and delinquency. In particular, data used in Melde and Esbensen’s study (2011) suggested that gang membership was associated with a reduction in associations with prosocial peers and commitment to school, both of which are positive sources of informal social control. Further, as youth became involved with gang youth, they reported socializing more with delinquent peers and in settings outside the watchful eyes of adults (e.g., at parties where alcohol and/or drugs were available), thus shaping gang youths’ routine activities. These youth also reported being more prone to violence (i.e., a greater anger identity), believed that violence was acceptable in more situations, and anticipated feeling less guilt related to their own involvement in acts of crime and delinquency, and thus, the identity and beliefs of these youth were systematically altered. Together, these effects of gang membership accounted for a 29% increase in the frequency of involvement in delinquency. All of these changes are consistent with the notion of a turning point offered by Laub and Sampson (2003).

10.3

Assessing the Impact of Gang Membership on Delinquency: Problems and Prospects

Laub and Sampson’s (2003) age-graded theory of informal social control appears to be a promising theoretical framework from which to explain the nexus of gang membership and delinquency. As others in this volume (see Matsuda et al. 2012) have alluded to, however, studies of gang membership are problematic in nature. First and foremost, defining and identifying gang members is notoriously problematic and controversial. While Melde and Esbensen (2011) relied upon self-reports of involvement with gang youth, alternative definitions of gang membership produce different samples of youth, with only slight overlap. Even though Laub and Sampson’s (2003) turning point framework appears consistent with Melde and Esbensen’s (2011) operationalization of gang membership, there is potential that this framework is not consistent with alternative definitions of gang membership, such as the Eurogang funneling approach to the identification of gang-involved youth.1 In particular, youth identified by the Eurogang funneling approach may not view themselves as “gang members,” and thus, changes in their peer group structure may not influence levels of informal social control, routine activities, and (perhaps especially) how one views oneself (i.e., identity) and their behavior (i.e., morality) in a


173

manner consistent with those youth who self-identify gang involvement. On the other hand, if the Eurogang approach identifies the essential features of gangs, as it portends, then one should expect to find results similar to Melde and Esbensen (2011), suggesting that onset of gang membership is consistent with Laub and Sampson’s (2003) turning point framework. Therefore, the research question posed and tested in the current study is: Do youth who become involved with gang members, as defined through the Eurogang funneling approach, experience a similar change in informal social control, routine activities, and human agency consistent with Laub and Sampson’s (2003) discussion of turning points? A second problem in identifying a causal connection between gang membership and involvement in crime and delinquency stems from our inability to control for preexisting differences between gang and nongang youth, which raises the possibility of confounding.2 Given the preponderance of evidence consistent with the enhancement model of gang member delinquency, it is likely that youth who become involved in gangs are more antisocial before joining a gang, and thus, we would expect them to be involved in a disproportionate amount of crime and delinquency regardless of their eventual involvement in a gang. This reality makes the creation of a comparison group problematic, as some youth are simply not as likely to be involved in crime and delinquency. Like the adage, “you cannot compare apples to oranges,” it might be unfair to compare gang joiners to nongang youth because gang-involved youth may have further increased their offending behavior relative to their peers even if they failed to join a gang. In statistical terms, the nongang portion of a given sample may not provide the correct counterfactual for gang youth (see Rosenbaum and Rubin 1983, for an overview of the counterfactual framework). To help remedy this problem, a set of statistical tools—commonly referred to as propensity score methods—are utilized to screen the comparison sample for youth who mirror our gang sample, and a special weighting system is applied to the data to create what appear to be equal groups for comparison (see Caliendo and Koepenig 2008 for a practical guide to propensity scores). Given these procedures, the weighted comparison group can more adequately be utilized to serve as a counterfactual reference group, and hence allow us to determine what would have happened to these youth but for their involvement in the gang. That is, the weighted sample serves as a pseudorandom sample, closer to those derived from experimental methods.

10.4

Data

The data utilized in this study are part of an evaluation of a school-based law-related education program. As such, a purposive sample of schools was selected for inclusion in the evaluation; only schools offering the program were eligible to participate. The following summarizes our efforts to select study sites: 1. More than 250 schools were identified as offering the program at one point in time and thus were contacted to determine current implementation status.

174


2. Eighteen schools met the evaluation criteria (i.e., confirmation that the program was actually being taught in its entirety, teaching of a sufficient number of classes to allow for matching of treatment and comparison groups while also being costeffective in terms of travel to the school for data collection, a willingness to withhold the program from some classes, and agreement to adhere to the evaluation design). 3. Three schools declined the opportunity to participate. 4. Fifteen schools in nine cities in four states agreed to the evaluation design and participated in the outcome evaluation. The selection of schools was purposive, and the final sample of 15 schools (nine in Arizona, one in New Mexico, two in Massachusetts, and three in South Carolina) reflects the fact that program adoption was more pronounced in Arizona. Classrooms were selected based upon the grade in which the program was taught (ranging from sixth to ninth grade). Random assignment was not possible, as only some teachers were trained to teach the program. Classrooms were matched by grade and subject such that if the curriculum was taught in two sixth-grade social studies classes, the comparison classrooms were the remaining grade-level social studies classes that did not receive the program. In total, there were 48 classrooms that received the curriculum and 49 comparison classrooms.3 All students in the selected classrooms (N = 2,353) were asked to participate in the evaluation. Due to the nature of the study, active parental consent was required before students could participate in the evaluation. Consent letters were sent home with students and collected by teachers. Our collaborative efforts with the teachers resulted in a 72% active consent rate (N = 1,686): 12% of parents refused their child’s participation (N = 290), and 16% of students failed to return consent forms (N = 377). Because our analysis requires two waves of data, individuals had to participate in both waves of data collection to be included in the following analyses (N = 1,450). The students participating in the evaluation resemble all students in their schools; that is, the sample demographics are similar to the school-level demographics. In fact, in several instances, the students represent all or most of the students at grade level. The sample, however, is not representative of students across the nation, as schools were purposively selected to meet the needs of the program evaluation. Pretest data (time 1) were collected prior to the delivery of the evaluated curriculum during the 2004–2005 school year, and posttests (time 2) were administered directly following the completion of the program; approximately 6 months after the pretest. Both waves of survey data were collected using group-administered selfreport methods whereby subjects answered questions individually as they were read out loud by members of the research team. The approximate time needed for completion of the survey was 40–45 min. Unless noted otherwise, scales used in this study were derived from the Denver Youth Survey (Huizinga et al. 1991) and represent the mean score across items. All items and scales remained consistent throughout both waves of data collection.


10.5

175

Measures

10.5.1

Gang Membership

The current analysis examines the impact of gang onset on processes associated with a turning point, as described by Sampson and Laub (2005). Therefore, only those respondents who were not involved in a gang at time 1 (i.e., either self-identified or Eurogang-defined) but who became a Eurogang-defined gang member at time 2 were included in the analysis as a gang member. The comparison group included those individuals who were not gang involved at either time 1 or time 2 (i.e., either self-identified or Eurogang-defined) to avoid possible “treatment” effects on those in the nontreatment category. In total, 63 (4.9%) respondents became Eurogang-defined gang members between time 1 and time 2, while 1,224 (95.1%) were not classified as gang involved at either wave of data collection (Table 10.1).

10.5.2

Demographic Characteristics

The demographic characteristics used in the present analyses are sex, race/ethnicity, and age. For analysis purposes, the sex variable was dummy coded, with male equal to one and female equal to zero. Forty-five percent of our survey respondents were Table 10.1 Sample descriptive statistics (N = 1,287) Variable N Percent (%) Sex Male 582 45 Female 705 55 Race/ethnicity White 436 34 Black 141 11 Hispanic 541 42 Other 171 13 Age (T1) Gang membership 63 4 Delinquency Prosocial peers School commitment Guilt Neutralization Parental monitoring Negative peer commitment Unstructured socializing Anger identity

Mean

SD

12.21

0.98

1.39 3.10 3.77 2.63 3.38 4.02 1.95 1.25 3.00

2.57 0.86 0.74 0.44 1.08 0.91 0.97 0.87 1.09


176

male, while females represented 55% of the total. Study participants were classified into one of four racial/ethnic groups: non-Hispanic/White, Black/African-American, Hispanic/Latino, and other, with those reporting to be non-Hispanic/White serving as the reference group. Due to the concentration of participating schools in the Southwestern United States, the largest racial/ethnic group was Hispanic (42%), followed by non-Hispanic/White (34%), other (13%), and African-American (11%). Finally, age was left in its original metric, with the mean age for the sample being 12.21 (SD = 0.98) at the time of the pretest (time 1).

10.5.3

Delinquency

The delinquency index was created using the frequency score representing the number of times respondents engaged in the described behavior in the past 3 months. Available answers ranged in magnitude from zero to four, with zero equal to never, one equal to one time, two equal to two to five times, three equal to six to ten times, and four equal to more than ten times. Questions used in the creation of the index were specifically created for this study, as well as adapted from those used as part of the Denver Youth Survey (Huizinga et al. 1991) and the National Youth Survey (Elliott et al. 1985). More specifically, the measure was created from ten items that ranged in seriousness from “stolen or tried to steal something worth less than $50,” to “used a weapon or force to get money or things from people.” All ten items were summed to create an overall frequency score.

10.5.4

Prosocial Peers

Associating with prosocial peers was measured with an eight-item scale in which students were asked, “During the last year, how many of your current friends have done the following?” Responses were based on a five-point scale ranging from “none of them” to “all of them.” Items in the scale included, “Have been thought of as good students,” “Have gotten along well with teachers and adults at school,” and “Have been involved in school activities or school athletics” (a = 0.84) [for a more complete description of how each of these indicators fits within the turning point framework, see Melde and Esbensen 2011].

10.5.5

School Commitment

Commitment to school was measured with a seven-item scale. Students were asked whether they agreed or disagreed with a number of statements regarding school


177

activities, including “I try hard in school,” “In general, I like school,” and “Grades are very important to me.” Responses were based on a five-point Likert scale ranging from “strongly disagree” to “strongly agree” (a = 0.79).

10.5.6

Guilt

To measure the respondents’ anticipated guilt4 related to participation in delinquent activities, we used a 13-item guilt scale. The stimulus for the measure stated, “How guilty or how bad would you feel if you” and was followed by statements ranging in severity from “skipped school without an excuse” to “used a weapon or force to get money or things from people.” Responses were based on a three-point scale ranging from “not very guilty/bad” to “very guilty/bad” (a = 0.93).

10.5.7

Neutralization

To measure the use of techniques of neutralization (Sykes and Matza 1957), we relied upon three questions related to the use of force, including: “It’s okay to beat up someone if they hit you first,” “It’s okay to beat up someone if you have to stand up for or protect your rights,” and “It’s okay to beat up someone if they are threatening to hurt your friends or family.” Answers were collected using a five-point Likert scale, with one equal to “strongly disagree” and five equal to “strongly agree” (a = 0.78).

10.5.8

Parental Monitoring

Parental monitoring was measured using a four-item scale scored on a five-point Likert scale (1 = strongly disagree and 5 = strongly agree). Statements included items such as “My parents know who I am with if I am not at home” and “My parents know where I am when I am not at home or at school” (a = 0.72).

10.5.9

Negative Peer Commitment

Commitment to deviant peers consists of three items measured on a five-point scale ranging from “not at all likely” to “very likely.” The three questions included in the scale asked respondents, “If your group of friends was getting you in trouble (at home/at school/with the police), how likely is it that you would still hang out with them?” (a = 0.80).


178

10.5.10

Unstructured Socializing

To measure the extent to which respondents were involved in unstructured socializing, including activities outside the purview of adult supervision, a two-question index (Esbensen et al. 2001) was utilized. Questions in the index asked respondents, “Do you ever spend time hanging around with your current friends, not doing anything in particular, where no adults are present” and “Do you ever spend time getting together with your current friends where drugs and alcohol are available?” Respondents indicated either “yes” (= 1) or “no” (= 0) for each question, and the index was created based on the sum of responses across items.

10.5.11

Anger Identity

The survey instrument included the four-item scale developed by Grasmick et al. (1993) to measure the anger/temper component of the self-control construct, which is consistent with the operationalization of anger identity in Giordano, Schroeder, and Cernkovich (2007). Items included in the scale included: “I lose my temper pretty easily” and “When I’m really angry, other people better stay away from me.” Responses ranged from one to five, with one equal to “strongly disagree” and five equal to “strongly agree” (a = 0.74).

10.6

Analysis Strategy

Gang membership is a status that is influenced to some degree by the self-selection of individuals with antisocial tendencies into these groups (DeLisi et al. 2009; Haviland et al. 2007; Krohn and Thornberry 2008; Melde and Esbensen 2011), making proper controls for confounding necessary (Caliendo and Kopeinig 2008). The current study utilizes two mechanisms to reduce the probability that the noted association between gang membership and delinquency is the result of confounding. First, to examine the direct effect of gang membership on delinquency and those processes related to “turning points” (Sampson and Laub 2005), we use propensity score matching using the psmatch2 module (Leuven and Sianesi 2003) available in Stata 10.0 (Stata 2007). The propensity score was derived using probit regression and represents the probability of becoming gang involved at time 2 given the observed characteristics at time 1. To assure that outliers (i.e., bad matches) were excluded from the analysis, an Epanechnikov kernel matching procedure was used, as this allows one to constrain matches to a specified difference in treatment propensity. All reported analyses are based on a bandwidth of 0.05, while statistical significance of the ATT (average treatment effect on the treated) was determined based on bootstrap (50 replications) standard errors, as suggested by Caliendo and


179

Kopeinig 2008; see Melde and Esbensen 2011 for further discussion of the analysis procedure). Because the current analysis is also interested in the degree to which those factors associated with “turning points” (Sampson and Laub 2005) can explain the association between gang membership and delinquency, we utilize the Preacher and Hayes (2008) technique for assessing direct and indirect effects in multiple mediator models.5 This method of assessing mediation with multiple mediators has a number of advantages, including (1) the ability to determine the unique effect of individual mediators, conditional on other mediators and covariates in the model; (2) the ability to assess the relative magnitude of specific indirect effects in the model; and (3) the ability to include multiple mediators as well as covariate controls in a single model reduces the possibility of omitted variables bias.

10.7

Results

To determine if the onset of gang membership is associated with an increase in delinquent involvement, a propensity score model comparing those who reported involvement in a gang for the first time at time 2 (N = 63) with nongang youth (N = 1,224; i.e., youth who did not report gang involvement at time 1 or time 2) was conducted (probit regression results; chi2 = 123.23, p < 0.001; pseudo R2 = 0.24; available upon request). Results of the initial test of selection bias revealed a number of preexisting differences between those who joined a gang and those who failed to do so.6 Specifically, 23 of the 34 time 1 variables used to match the time 2 gangonset group to the nongang youth were significantly different7 and suggested that the gang joiners were more at risk for delinquency and gang involvement before onset of gang membership. This suggests that failure to properly account for selection bias related to joining a gang could lead to inaccurate conclusions. After matching, however, no significant differences between the two groups remained (i.e., all standardized biases were below 20). Given the success of our propensity score model in removing the initial bias between the two groups, the unconfoundedness assumption (Blundell and Costa Dias 2009; Caliendo and Kopeinig 2008) has been met, and the estimated outcomes can more confidently be attributed to having become gang involved. Thus, we turn next to a discussion of the results of our propensity score analyses. The propensity score analysis presented in Table 10.2 contains three separate parameters of interest, all of which are mean differences between gang and nongang youth. The first set of analyses represents raw mean comparisons based on the observed (i.e., unweighted) differences between nongang youth and those who joined a gang at time 2. The second set of analyses reports the sample means derived from the weighted sample such that the difference score represents the average treatment effect on the treated (ATT). This comparison, therefore, is used to determine if gang joining (“the treated”) produced statistically significant differences in outcome (i.e., delinquency and variables associated with turning points) after

Table 10.2 The effect of onset of Eurogang-defined gang membership at time 2 Unmatcheda Matched ATTb Gang onset Nongang Difference Gang onset Nongang Difference Time 2 (N = 63) (N = 1,224) (N = 63) (N = 1,224) 5.98 2.05 3.93* Delinquency 5.98 1.19 4.79* Prosocial peers 2.46 3.14 −0.68* 2.46 2.88 −0.42* School commitment 3.08 3.82 −0.74* 3.08 3.49 −0.41* Guilt 2.14 2.66 −0.52* 2.14 2.49 −0.35* * Neutralization 4.20 3.34 0.86 4.20 3.67 0.53* Parental monitoring 3.65 4.04 −0.39* 3.65 3.89 −0.24* Negative peer commitment 2.96 1.90 1.06* 2.96 2.38 0.58* Unstructured socializing 1.34 0.61 0.73* 1.34 0.85 0.49* * Anger identity 3.51 2.98 0.53 3.51 3.15 0.36* Propensity score analyses were done using kernel matching with a bandwidth of 0.05. Bootstrap standard errors (50 replications) of the ATT were calculate statistical significance * p < 0.05 (t test) a Difference scores represent the raw mean difference between nongang youth and those who reported onset of gang involvement at time 2 b The ATT is based on the differences in mean outcome for propensity-score-matched individuals using kernel matching

3.68 −0.69 −0.53 −0.29 0.74 −0.36 0.51 0.57 0.58 used to

ATE

180 C. Melde and F.-A. Esbensen


181

accounting for preexisting differences in treatment likelihood.8 The final column is the average treatment effect (ATE), which represents the normal difference in expected outcomes after involvement in a gang. In the current context, the ATE represents the average estimated effect of gang involvement if respondents were randomly assigned to become involved with gangs. The first research question to be examined is whether or not gang membership impacts involvement in delinquency above and beyond the influence of preexisting differences between gang and nongang youth. As Table 10.2 demonstrates, even after controlling for preexisting differences in the likelihood of gang involvement, gangs produced a statistically significant effect on delinquency for those youth who reported onset of gang involvement at time 2 (ATT = 3.93, p < 0.05). As expected, there appears to be a direct effect of gang involvement on contemporaneous delinquent involvement even after compensating for preexisting differences between gang and nongang youth. The second research question to be addressed is whether onset of gang involvement is consistent with the notion of a “turning point,” as discussed by Sampson and Laub (2005). That is, does onset of gang involvement influence factors associated with turning points in a theoretically consistent manner? Results provided in Table 10.2 suggest that onset of gang membership impacts factors related to Sampson and Laub’s (2005) discussion of turning points. After controlling for propensity to join a gang, those who became involved in a gang at time 2 reported significantly different mean values at time 2 on all measures associated with turning points in the theoretically expected direction. Specifically, onset of gang involvement was associated with fewer associations with prosocial peers (ATT = −0.42, p < 0.05), lower school commitment (ATT = −0.41, p < 0.05), and less anticipated guilt for involvement in delinquency (ATT = −0.35, p < 0.05), as well as greater acceptance of neutralizations (ATT = 0.53, p < 0.05), less parental monitoring (ATT = −.24, p < 0.05), higher negative peer commitment (ATT = .58, p < 0.05), more unstructured socializing (ATT = .49, p < 0.05), and a greater anger identity (ATT = .36, p < 0.05). Overall, onset of gang membership is consistent with Sampson and Laub’s (2005) definition of a turning point. Given these findings, we next examine whether or not these factors associated with gang involvement as a turning point can explain the higher levels of reported delinquency at time 2. To examine the potential mediating processes associated with the influence of gang membership on delinquency, we utilize the Preacher and Hayes (2008) method of multiple mediation analysis using SPSS (16.0). Table 10.3 presents results from this analysis, which included covariate controls for our demographic variables (e.g., sex, age, race/ethnicity), the estimated propensity score for gang involvement (Coffman 2011), and all time 1 values of our proposed mediators to control for potential confounding. That is, all coefficients represent the unique effect of the predictor on outcome, above and beyond any other mediators or covariates in the model (Preacher and Hayes 2008). The effects of the covariate controls are not presented, but are available upon request. Model A in Table 10.3 displays the effect of onset of gang membership on the proposed mediators. In comparing these results with those in Table 10.2, a similar

182


Table 10.3 The total, direct, and indirect effects of onset of gang membership and factors associated with a turning point on delinquency b SE t Model A (a paths): the effect of onset of gang membership on factors associated with a turning point 0.09 −4.31 Prosocial peers −0.40* School commitment −0.38* 0.08 −4.64 Guilt −0.32* 0.05 −6.24 Neutralization 0.52* 0.12 4.14 Parental monitoring −0.20 0.11 −1.84 Negative peer commitment 0.56* 0.12 4.79 Unstructured socializing 0.48* 0.10 5.90 Anger identity 0.37* 0.13 2.79 b SE t exp(b) Model B (b paths): the effect of factors associated with a turning point on delinquency Prosocial peers −0.06* 0.02 −2.31 0.94 School commitment −0.04 0.03 −1.62 0.95 Guilt −0.28* 0.05 −6.23 0.76 Neutralization 0.10* 0.02 5.22 1.11 Parental monitoring 0.05* 0.02 2.14 1.05 0.02 2.61 1.05 Negative peer commitment 0.05* Unstructured socializing 0.15* 0.02 3.66 1.16 Anger identity 0.07* 0.02 3.86 1.07 Model C (c path): total effect of onset of gang membership on delinquency Onset of gang membership 0.85* 0.08 10.11 2.34 Model D (c’ path): direct effect of onset of gang membership on delinquency Onset of gang membership 0.55* 0.08 7.12 1.73 Model E (ab paths): the indirect effects of onset of gang membership on delinquency Total indirect effects 0.30* 0.04 6.26 1.35 Prosocial peers 0.02 0.01 1.90 1.02 School commitment 0.02 0.01 1.20 1.02 Guilt 0.09* 0.03 3.45 1.10 Neutralization 0.05* 0.01 3.46 1.05 Parental monitoring − 0.01 0.01 −1.30 0.99 Negative peer commitment 0.03* 0.02 1.99 1.03 Unstructured socializing 0.07* 0.02 2.99 1.07 Anger identity 0.03* 0.01 2.58 1.03 Estimates are based on the Preacher and Hayes (2008) method of effect decomposition, including bootstrap standard errors (1,000 replications) for indirect effects and covariate controls for gang propensity and time 1 constructs. The dependent variable represents the natural log of delinquency frequency plus one * p < 0.05; R2 = 0.38; (N = 1,287)

pattern emerges, although with slightly reduced effect sizes given the controls for all other variables in the model. With the exception of parental monitoring, which was not systematically effected by gang onset, all variables associated with the turning point framework are significant and in the expected direction.


183

Model B in Table 10.3 exhibits the effects of the mediators on the natural log of delinquency frequency at time 2 for the entire sample, net all other mediators and covariate controls. With the exception of school commitment, all proposed mediators are significantly associated with delinquency, and in the expected direction. Specifically, a one-unit increase in prosocial peers or guilt is associated with a 6% (exp(b) = 0.94; p < 0.05) and 24% (exp(b) = 0.76; p < 0.05) reduction in delinquency frequency, respectively. Increases in the use of techniques of neutralization (exp(b) = 1.11; p < 0.05), parental monitoring (exp(b) = 1.05; p < 0.05), negative peer commitment (exp(b) = 1.05; p < 0.05), unstructured socializing (exp(b) = 1.16; p < 0.05), and anger identity (exp(b) = 1.07; p < 0.05) are also systematically associated with increased delinquency frequency. Overall, the onset of gang membership between time 1 and time 2 is associated with a 134% increase in delinquency frequency at time 2 (see Model C in Table 10.3 (exp(b) = 2.34; p < 0.05)). After controlling for the mediating pathways associated with the turning point framework, however, the direct effect is reduced to a 73% increase in delinquency that can be attributed to gang onset alone (see Model D in Table 10.3 (exp(b) = 1.73; p < 0.05)). This reduced effect can be attributed to a significant total indirect effect of gang membership on delinquency through the mediating pathways (see Model E in Table 10.3 (exp(b) = 1.35; p < 0.05)). In particular, five of the hypothesized mediating pathways are significant mediators of the effect of gang membership on delinquency, including guilt (exp(b) = 1.09; p < 0.05), neutralizations (exp(b) = 1.05; p < 0.05), negative peer commitment (exp(b) = 1.03; p < 0.05), unstructured socializing (exp(b) = 1.07; p < 0.05), and anger identity (exp(b) = 1.03; p < 0.05). Therefore, while gang membership was associated with a reduction in associating with prosocial peers and school commitment, changes in these two variables do not appear to explain the relationship between gang membership and delinquency.

10.8

Discussion and Conclusion

The purpose of the current analysis was to determine the applicability of Laub and Sampson’s (2003) age-graded theory of informal social control to the onset of gang membership. Given extant research on the apparent enhancement effect of gang membership on delinquency, the explicit focus of Laub and Sampson’s (2003) theory on both stable individual differences in criminal propensity and mechanisms for change appeared consistent with empirical findings. While the work of Melde and Esbensen (2011) supported the use of Laub and Sampson’s (2003) framework, complications related to gang research more generally make replication of these findings with alternative definitions of gang involvement necessary. Specifically, both defining and identifying gang membership are problematic and controversial. Melde and Esbensen (2011) found support for using the turning point framework for the onset of self-reported involvement with gangs, in that youth identified their group of friends as a gang. The current study built upon this work by using the Eurogang

184


definition of gang membership and replicated the utility of the turning point framework for understanding the effects of gang membership. As discussed in Matsuda et al. (2012), there is a lack of concordance among different operational definitions of gang membership. In that research, the authors reported little overlap among gang-involved youth using three different measures of gang membership. Importantly, however, they found that gang youth, regardless of how measured, were substantively different from nongang youth but similar to each other. In this chapter, we found further evidence for the robustness of gang membership, regardless of definition. Our analyses allow us to draw the following conclusions (1) after controlling for preexisting differences between nongang and gang youth, onset of gang membership exerts an independent effect on delinquency; (2) gang joining is associated with significant changes in informal social control, routine activities, and attitudes and identity consistent with Laub and Sampson’s (2003) turning point framework; and (3) factors associated with the turning point framework partially mediate the effect of onset of gang membership on delinquency. These findings further highlight the group processes discussed by Klein and Maxson (2006) as the key differentiating factor between youth groups and youth gangs. Another important finding of the current study is the nonnegligible role of selfselection into and out of gangs. Consistent with the enhancement model, current results suggest that at-risk youth are more likely to join gangs than more prosocial youth. Given our reliance on nonexperimental data to study the consequences of gang membership, the failure to control for such differences could lead to biased estimates of the effect of gang membership. Future research should examine the potential for confounding when estimating effects associated with gang membership. An advantage of using Sampson and Laub’s (2005) life-course perspective of criminal development is the policy relevance of the findings. In the current study, in particular, this framework underscores the necessity of primary prevention and suggests potential avenues for intervention in the lives of gang-involved youth. Regarding primary prevention, current results suggest that gang involvement not only has immediate effects on delinquent involvement but also negatively effects sources of social capital that can impact long-term developmental trajectories. Prevention of gang involvement, therefore, can have long-term benefits above and beyond the immediate impact on criminal involvement. For youth involved in gangs, however, current results imply that merely getting youth out of gangs may not be enough to curtail the immediate or long-term consequences of their involvement; intervention programs should focus also on the restoration of social bonds with conventional others. That is, gang interventions may be most effective if they provide youth with avenues to build on sources of social capital, such as improved relationships with prosocial peer networks, school officials, and their families. Interventions should also work on identity reorientation, given the importance of identity in adolescent development (Giordano et al. 2007), and the likely impact gang membership has on individual self-concepts (see Hennigan and Spanovic 2012). In conclusion, while prior research has suggested that the social environment associated with youth gang membership impacts attitudes, emotions, and ultimately


185

delinquency, the current study documented this dynamic process by including information on youth before and during membership in such groups. Results suggest that gang membership impacts informal social control, routine activities, and identity in a manner consistent with Sampson and Laub’s (2005) turning point framework. Future research should further explore gang membership from a life-course perspective and include more refined measures of important social bonds and the social-psychological processes associated with gang membership and delinquency.

Notes 1. The Eurogang definition (a youth gang, or troublesome youth group, is a durable, streetoriented youth group whose involvement in illegal activity is part of their group identity) contains four key definitional elements: stability or durability, street orientation, involvement in illegal activity, and group identity. To capture these elements, respondents were first asked if they had a group of friends with whom they did things. Next, they were asked about the age of the group, how long it had existed, whether they spent a lot of time hanging out in public places, and finally, whether it was okay for members of the group to do illegal things and if they in fact do illegal things together. In this manner, the researcher can construct or classify gang membership based upon the defining characteristics of the peer group rather than rely upon the self-nomination technique. 2. Confounding stems from omitted variables bias wherein a statistically significant relationship between two variables is the product of a third, unmeasured variable. A classic example of confounding is the relationship between carrying matches and lung cancer. While the relationship between these two variables is robust, a third variable, smoking, is responsible for their association. In the current context, there is a possibility that the association between gang membership and increased delinquency is the result of preexisting differences between youth who join a gang and those that fail to do so. 3. While there was the potential for the curriculum under evaluation to effect study results, process evaluation results suggest that this is an unlikely scenario. A process evaluation concluded that the program was not taught with sufficient fidelity for assessment of program impact (Esbensen 2009; Melde et al. 2006). Outcome analyses revealed no statistically significant differences between the two groups. 4. Evidence suggests that anticipated emotions, particularly guilt, play a central role in the decision-making process. Baumeister and colleagues (2007, p. 193) describe this process by stating that “Guilt can exert a strong effect on behavior even if people rarely feel guilty, simply because people learn what will make them feel guilty and then change their behavior so as to avoid guilt.” Anticipated guilt, therefore, is a better predictor of behavior than current guilt levels. 5. The dependent variable represents delinquency frequency and, as expected, is overdispersed (i.e., the variance is greater than the mean), which could lead to biased estimates if analyzed in an untransformed state using ordinary least squares regression techniques (Osgood 2000). The dependent variable in analyses using the Preacher and Hayes (2008) method represents the natural log of delinquency frequency plus one (skewness = 0.984). 6. While the inclusion of such a high number of highly correlated variables in a probit regression presents problems in interpreting the coefficients, collinearity among the covariates is not a problem in propensity score analysis as the predicted probability of treatment (i.e., propensity) score is the only value of interest, not unbiased coefficient estimates (Harding 2003). Given the difficulty in interpreting the derived coefficients due to collinearity issues, detailed results are not presented but are available upon request.

186


7. Covariate balance was examined using the standardized bias statistic (Rosenbaum and Rubin 1985). Absolute values above 20 represent imbalance (i.e., bias) between the treatment and control groups. 8. Comparison of the ATT score and the raw mean differences between gang and nongang youth is an indication of the extent to which preexisting differences influence the outcomes of interest. The greater the discrepancy between raw mean differences and ATT, the greater the influence of preexisting differences in gang and nongang youth on the outcome of interest.

References Battin SR, Hill KG, Abbott RD, Catalano RF, Hawkins JD (1998) The contribution of gang membership to delinquency beyond delinquent friends. Criminology 36:93–115 Baumeister RF, Vohs KD, DeWall CN, Zhang L (2007) How emotion shapes behavior: feedback, anticipation, and reflection, rather than direct causation. Pers Soc Psychol Rev 11:167–203 Bendixen M, Endresen IM, Olweus D (2006) Joining and leaving gangs: selection and facilitation effects on self-reported antisocial behaviour in early adolescence. Eur J Criminol 3:85–114 Blundell R, Costa Dias M (2009) Alternative approaches to evaluation in empirical microeconomics. J Hum Resour 44:555–640 Caliendo M, Kopeinig S (2008) Some practical guidance for the implementation of propensity score matching. J Econ Surv 22:31–72 Coffman D (2011) Estimating causal effects in mediation analysis using propensity scores. Struct Eq Model 18:357–369 DeLisi M, Barnes JC, Beaver KM, Gibson CL (2009) Delinquent gangs and adolescent victimization revisited: a propensity score matching approach. Crim Just Behav 36:808–823 Egley AJ, Maxson CL, Miller J, Klein MW (eds) (2006) The modern gang reader, 3rd edn. Roxbury Publishing, Los Angeles Elder GH (1985) Perspectives on the life course. In: Elder GH (ed) Life course dynamics. Cornell University Press, Ithaca, pp 23–49 Elliott DS, Huizinga D, Ageton SS (1985) Explaining delinquency and drug use. Sage Publications, Beverly Hills Esbensen F-A (2009) Final report: evaluation of the teens, crime and the community and community works program. National Institute of Justice, Washington, DC Esbensen F-A, Huizinga D (1993) Gangs, drugs, and delinquency in a survey of urban youth. Criminology 31:565–589 Esbensen F-A, Osgood DW, Taylor TJ, Peterson D, Freng A (2001) How great is G.R.E.A.T.?: results from a longitudinal quasi-experimental design. Criminol Pub Pol’y 1:87–118 Gatti U, Tremblay RE, Vitaro F, McDuff P (2005) Youth gangs, delinquency and drug use: a test of the selection, facilitation, and enhancement hypotheses. J Child Psychol Psychiatry 46: 1178–1190 Giordano PC, Schroeder RD, Cernkovich SA (2007) Emotions and crime over the life course: a neo-meadian perspective on criminal continuity and change. Am J Sociol 112:1603–1661 Glueck S, Glueck E (1950) Unraveling juvenile delinquency. The Commonwealth Fund, New York Gordon RA, Lahey BB, Kawai E, Loeber R, Stouthamer-Loeber M, Farrington DP (2004) Antisocial behavior and youth gang membership: selection and socialization. Criminology 42:55–88 Gottfredson MR, Hirschi T (1990) A general theory of crime. Stanford University Press, Stanford Grasmick HG, Tittle CR, Bursik RJ, Arneklev BJ (1993) Testing the core empirical implications of Gottfredson and Hirschi’s general theory of crime. J Res Crime Del 30:5–29 Harding DJ (2003) Counterfactual models of neighborhood effects: the effect of neighborhood poverty on dropping out and teenage pregnancy. Am J Sociol 109:676–719


187

Haviland A, Nagin DS, Rosenbaum PR (2007) Combining propensity score matching and groupbased trajectory analysis in an observational study. Psychol Methods 12:247–267 Hennigan K, Spanovic M (2012) Gang dynamics through the lens of social identity theory. In: Esbensen F-A, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Hirschi T (1969) Causes of delinquency. University of California Press, Berkeley Huizinga D, Esbensen F-A, Weiher AW (1991) Are there multiple routes to delinquency? J Crim Law Criminol 82:83–118 Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, New York Krohn MD, Thornberry TP (2008) Longitudinal perspectives on adolescent street gangs. In: Lieberman A (ed) The long view of crime: a synthesis of longitudinal research. Springer, New York, pp 128–160 Laub JH, Sampson RJ (2003) Shared beginnings, divergent lives: delinquent boys to age 70. Harvard University Press, Cambridge Leuven E, Sianesi B (2003) PSMATCH2: stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. Retrieved from http://ideas.repec.org/c/boc/bocode/s432001.html Matsuda K, Esbensen F-A, Carson D (2012) Putting the “Gang” in “Eurogang”: characteristics of delinquent youth groups by different definitional approaches. In: Esbensen F-A, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Melde C, Esbensen F-A (2011) Gang membership as a turning point in the life course. Criminology 49:513–552 Melde C, Esbensen F-A, Tusinski KT (2006) Addressing program fidelity using onsite observations and program provider descriptions of program delivery. Eval Rev 30:714–740 Osgood DW (2000) Poisson-based regression analysis of aggregate crime rates. J Quant Criminol 16:21–43 Peterson D, Taylor TJ, Esbensen F-A (2004) Gang membership and violent victimization. Justice Quarterly 21:794–815 Preacher KJ, Hayes AF (2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 40:879–891 Rosenbaum PR, Rubin DB (1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70:41–55 Rosenbaum P, Rubin DB (1985) Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. Am Stat 39:33–38 Sampson RJ, Laub JH (1995) Understanding variability in lives through time. Stud Crime Crime Prev 4:143–158 Sampson RJ, Laub JH (2005) A life-course view of the development of crime. Ann Am Acad Polit Soc Sci 602:12–45 Stata (2007) Stata statistical software: release 10. StataCorp, TX, College Station Sykes G, Matza D (1957) Techniques of neutralizations: a theory of delinquency. Am Sociol Rev 22:664–670 Thornberry TP, Krohn MD, Lizotte AJ, Chard-Wierschem D (1993) The role of juvenile gangs in facilitating delinquent behavior. J Res Crime Delinq 30:75–85 Thornberry TP, Krohn MD, Lizotte AJ, Smith CA, Tobin K (2003) Gangs and delinquency in developmental perspective. Cambridge University Press, New York Warr M (2002) Companions in crime: the social aspects of criminal conduct. Cambridge University Press, Cambridge

Chapter 11

The Sex Composition of Groups and Youths’ Delinquency: A Comparison of Gang and Nongang Peer Groups* Dana Peterson and Dena C. Carson

Much research demonstrates that delinquency is higher among gang members than among nongang members (e.g., Battin et al. 1998; Esbensen and Huizinga 1993; Esbensen et al. 2010; Thornberry et al. 2003). Violence, in particular, has been identified as a unique and organizing feature of gangs. In fact, some scholars, most notably Malcolm Klein, have argued that gangs are “qualitatively different” from other peer groups, and it is often the violence associated with gang membership, as well as group identity and cohesion derived from crime and violence, that is highlighted as a key distinguishing characteristic (e.g., Decker 1996; Klein 1995). Other factors differentiating gangs from nongang groups include organizational characteristics such as the presence of a leader, hierarchy, and specific rules (Bouchard and Spindler 2010). Specific features of gangs, such as degree of organization or sex composition, appear to structure delinquent activities within gangs (Decker et al. 2008; Peterson et al. 2001). Do such features structure delinquent activities within nongang peer groups? Some research indicates that group structural characteristics, such as racial composition, influence delinquency of peer group members (e.g., Haynie and Payne 2006). Similarly, studies of gendered processes in youths’ peer groups also suggest that group sex composition influences the behavior of group members (e.g., Giordano 1978; Warr 1996). * This project was supported by Award No. 2006-JV-FX-0011 awarded by the National Institute of Justice, Office of Justice Programs, US Department of Justice. This research was made possible, in part, by the support of the School District of Philadelphia and six other school districts. The opinions, findings, and conclusions expressed in this chapter are those of the authors and do not necessarily reflect the views of the Department of Justice or those of the School District of Philadelphia. D. Peterson (*) School of Criminal Justice, University at Albany, Albany, NY, USA e-mail: [email protected] D.C. Carson Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_11, © Springer Science+Business Media, LLC 2012

189

190

D. Peterson and D.C. Carson

From these various lines of research, we would therefore expect delinquency and violence to be lower among members of nongang groups than among gang members; but do gender dynamics within groups operate in the same way to produce the same patterns of delinquency among nongang as among gang youths? Or is there something about youth gangs that produces different patterns of delinquency among their members, potentially making them “qualitatively different?” In this chapter, we examine these questions set within the context of various literatures that are briefly reviewed below: How group sex composition may relate to gender dynamics within the group, how group sex composition may affect members’ behavior such as delinquency, how delinquency varies by gang status, and how gangs may or may not provide a unique context that differentiates them from nongang peer groups in terms of the effect of group sex composition on members’ behavior.

11.1

Gender Dynamics by Groups’ Sex Composition

As reviewed in Peterson et al. (2001), research in the sociology of organizations and groups has demonstrated that sex composition shapes interactional dynamics within groups (e.g., Blalock 1967; Blau 1977; Gutek 1985; Kanter 1977a, b). Prior research in the gang arena also indicates that organizational factors such as the sex composition of gangs influence the experiences and behaviors of gang members (Joe-Laidler and Hunt 1997; Miller 2001; Miller and Brunson 2000; Nurge 1998; Peterson et al. 2001; Weerman 2012; see too Bowker et al. 1980). Theoretical perspectives about how groups’ demographic composition affects both their normative features and their members’ experiences and behaviors have been categorized into two types: generic and institutional. Generic approaches suggest that any group in the numeric minority in a larger group will have similar experiences based on their proportion of the larger group (e.g., Blau 1977; Kanter 1977a, b). Thus, for example, females in majority-male groups would have similar experiences as would males in majority-female groups. The experiences of these tokens are of marginalization and exclusion, though different scholars offer different hypotheses regarding the reasons for this marginalization. In addition, according to these perspectives, gender relations and token members’ experiences improve as the sexes equal each other in number within groups. By contrast, institutional approaches suggest that the social position, either in larger society or in the larger group, of the members in the minority also plays a role. Blalock’s (1967) minority group–threat hypothesis, for example, argues that when minority tokens’ numbers increase, members of the dominant group exercise power and control to increase social and behavioral distance and maintain their position. Females, who occupy a “lower” social status than do males, would have very different experiences as tokens in a male-dominated group than would males who are tokens in a majority-female group. And, when females’ numbers in a group increase, males would engage in suppressive actions to maintain their own power and dominance in the group.

11 The Sex Composition of Groups and Youths’ Delinquency…

191

In organizational studies in which females are minorities or tokens, the assertions of the generic perspective appear to be upheld: Female tokens experience social isolation and role entrapment, but as their proportion of the larger group increases, gender relations and the experiences of females improve (e.g., Spangler et al. 1978; Yoder et al. 1983). Studies of males as token minorities, however, fail to support some of the generic perspective’s tenets, finding that male tokens do not have the same negative experiences as do female tokens (e.g., Fairhurst and Snavely 1983b; Heikes 1991; Yoder and Sinnett 1985). Correspondingly, much of the organizational research offers support to the propositions of the institutional over the generic perspective, finding that it is not numbers, alone, that affect the experiences of group members, but rather a combination of the number of each sex and factors such as group member status, gender stereotypes, and the group’s goals and activities (Fairhurst and Snavely 1983a, b; Gutek 1985; Konrad et al. 1992; South et al. 1982; Williams 1992; Yoder 1991, 1994).

11.2

Gangs, Sex Composition, and Delinquency

Applying these two perspectives to sex composition of gangs and its relationship to delinquency of members requires additional understanding of the gang as an organizational context. Groups or organizations have internal power structures that are maintained, in part, by members’ engagement in and differential access to particular activities that confer status. While some of the same characteristics, such as assertiveness or courage, might be valued in both organizational and gang contexts, the accepted mechanisms by which these are demonstrated are likely to differ. The importance of delinquency and violence in the gang context, for example, was highlighted earlier in this chapter. Commission of various delinquent and violent acts provides an important mechanism by which status is obtained and conferred in gangs. Involvement in delinquency is, therefore, a valued and desirable activity among gang members, but one that, from an organizational perspective, may be the subject of tension between the sexes. The generic perspective argues, for example, that tokens are visible to majority group members and that the majority exaggerates tokens’ difference and distorts tokens’ attributes to coincide with preexisting stereotypes. Accordingly, token gang females may be seen by male gang members as “weak” and unable to handle certain tasks of gang membership, such as criminal activity. As females’ numbers increase, the proportion of males’ interactions with females increases; females may therefore no longer be viewed differently, and there may be more normative acceptance of girls’ potential contributions to gang work, rather than exclusion and marginalization. The institutional perspective would offer different predictions regarding the gang’s sex composition and delinquency of members. Since delinquency is a statusenhancing activity and males occupy positions of power within the gang (as within larger society), males in gangs with relatively equal numbers of females would react to their numerical “threat” by excluding females from delinquency participation.

192


By contrast, in gangs with token numbers of females, males would perceive little threat to their dominance and care little about females’ activities, as females’ participation even in status-enhancing activity would not change the power dynamics within the group. It is useful to briefly recap more specific hypotheses that were outlined in Peterson et al. (2001) regarding how delinquency may vary by group sex composition, according to each of the two perspectives and drawing on additional gang research. Two kinds of hypotheses can be drawn: Those that predict within-sex differences across gang types (e.g., comparing females in different gang types to each other) and those that predict between-sex differences within gang types. Our focus is on hypotheses related to the two most common gang types, sex-balanced gangs (with a relatively equal number of females and males) and majority-male gangs, because these are the two we can most clearly analyze in the present study, but we offer tentative hypotheses regarding other group types as appropriate. Within-sex hypotheses drawn from generic perspective would predict that males in sex-balanced gangs will have lower rates of delinquency than males in majoritymale gangs (see Table 11.1 for a brief summary of the major competing hypotheses). This would be due to the influence of females in sex-balanced gangs; that is, females’ greater numbers and tendency toward lower levels of delinquency would influence the “culture” of the group, resulting in lower rates of delinquency for males as well. If we were to hypothesize about males in majority-female gangs1 (which we are not able to examine in the current study), the generic perspective might suggest these males would have the lowest delinquency rates because they are tokens in the group and are also influenced by the culture of the dominant group (females), which would tend toward lower levels of delinquency (according to research comparing delinquency of the sexes). The generic perspective would also suggest that females in sex-balanced gangs will be more delinquent than females in majority-male or in majority-female gangs, due to a combination of factors. Females in majority-male gangs are likely to be less delinquent than those in sex-balanced gangs because, as tokens, they would be marginalized and excluded by males, who view crime commission as “man’s work,” not suitable for females. Meanwhile, females in majority-female gangs would be less delinquent than females in sex-balanced gangs not only because they have limited exposure to and influence of (more deviant) males (e.g., Giordano 1978; Morash 1986; Warr 1996) but because they are more likely to describe their gang involvement in social terms, highlighting aspects such as the gang “being like a family” and providing “sisterhood” among members, rather than describing their gangs as context for acquisition of material goods or gang fighting (e.g., Joe and Chesney-Lind 1995; Miller 2001). The institutional perspective provides the opposite predictions (see Table 11.1): Males in sex-balanced gangs are likely to be more delinquent than their same-sex peers in majority-male gangs, engaging more frequently in status-enhancing activities to maintain their dominant position in the face of “threat” posed by a nearly equal number of females in their gangs, a threat not felt by males in majority-male groups. Likewise, females in sex-balanced gangs are likely to be less delinquent


193

Table 11.1 Summary of competing hypotheses from two organizational perspectives Generic Institutional Within-sex hypotheses Males in sex-balanced gangs will be less Males in sex-balanced gangs will be more delinquent than males in majority-male delinquent than males in majority-male gangs gangs Females in sex-balanced gangs will be more Females in sex-balanced gangs will be less delinquent than females in majority-male delinquent than females in majority-male gangs gangs Females in sex-balanced gangs will be more Females in sex-balanced gangs will be less delinquent than females in majority-female delinquent than females in majority-female groups gangs (tentative) Between-sex hypotheses Females will be significantly less delinquent Females will be significantly less delinquent than males in majority-male gangs than males in sex-balanced gangs Females and males in sex-balanced gangs will Females and males in majority-male gangs will have more similar levels of delinquency have more similar levels of delinquency

than females in majority-male gangs; in the former, girls’ delinquency would be suppressed by males, while in the latter, girls’ activities are relatively unfettered. From an institutional perspective, the suppression of status-enhancing activities of females in sex-balanced gangs may even result in their delinquency rates being lower than those of females in majority-female gangs, although this prediction would be tempered by the findings of research on girls in all- or majority-female gangs that was cited in the previous paragraph. We recognize that the above across-sex (intergroup) hypotheses may be a bit of a reach from the propositions of two organizational perspectives, but believe that, coupled with findings from research on peer groups and gangs, they are reasonable assertions. Likely, the most appropriate hypotheses to be drawn from the two perspectives, however, concern intragroup relations between the sexes, that is, betweensex comparisons within group type. Table 11.1 also presents the major intragroup hypotheses drawn from each of the two competing perspectives. Comparing females to males within gang types, generic perspectives would predict that females will have significantly lower delinquency rates than males in majority-male gangs, while females and males in sex-balanced gangs will be similar to each other (i.e., the gap in delinquency rates between the sexes will be far larger in majority-male gangs than in sex-balanced gangs). An institutional perspective would predict the opposite: Females in sex-balanced gangs will have significantly lower rates of delinquency than males in those gangs, while females and males in majority-male gangs will be more similar to each other (i.e., the female-male gap in delinquency rates will be far larger in sex-balanced than in majority-male gangs). Consistent with more general findings from organizational sociology research, the limited findings to date on gangs as groups appear most in line with an institutional perspective. Females in majority-male gangs have the highest frequencies of both property and violent offending, followed by females in sex-balanced gangs, and, lastly, females in majority- or all-female gangs (Peterson et al. 2001; Weerman

194


2012). Comparing females and males within gang type, more gender similarity in delinquency frequency was seen in majority-male gangs than in sex-balanced gangs, in which females’ offending was significantly lower than males’ (Peterson et al. 2001). Explanations consistent with institutional theories emerge from extant research. In sex-balanced gangs, those with a sizeable proportion of “lower-status” female members, females appear to be excluded by males from many serious forms of violence that represent status-enhancing activities within the gang (Miller 2001; Miller and Brunson 2000; Peterson et al. 2001; Weerman 2012; see too Bowker et al. 1980). According to an institutional perspective, greater numbers of females may be seen by males as a threat to the perceived male-dominated world of gangs; thus, females are kept from engaging in activities that confer status in these settings. Meanwhile, females in majority-male gangs, because of their fewer numbers, are not seen as a threat to male power structure within the gang; rather, they are seen as “one of the guys” and granted more freedom to participate in delinquent activities (Miller 2001; Miller and Brunson 2000).2

11.3

Groups, Sex, Sex Composition, and Delinquency

We know that, in general, gang members and males engage in greater levels of delinquency than do nongang members and females. Although comparatively little of gang members’ time is spent in law-violating activity (Fleisher 1998; Klein 1995), they do commit more than their fair share of delinquent acts (Esbensen et al. 2010; Esbensen and Huizinga 1993; Thornberry et al. 2003). And while gang girls’ levels of property and violent crime are lower than gang boys’, their delinquency is higher than nongang girls’ and even nongang boys’ delinquency (Deschenes and Esbensen 1999; Esbensen et al. 1999; Esbensen and Huizinga 1993; Miller 2001; Weerman 2012). The recent research reviewed above shows that it is not just sex or gang status that is important to understanding individuals’ delinquency but also the sex composition of the gang (Miller and Brunson 2000; Miller 2001; Peterson et al. 2001; Weerman 2012). Is this the case also for nongang groups? Does sex composition of nongang groups also structure the behavior of nongang youths, and if so, does it do so in the same ways as in gangs? Giordano (1978) and Warr (1996) document greater delinquency in groups consisting of both sexes than in other group types and, in particular, higher delinquency of females in mixed-sex than in other groups. These authors suggest that learning processes within the group are responsible: Females in mixedsex groups learn delinquent behavior from males in the group. In these lines of inquiry, however, we are largely missing studies that can demonstrate just how gangs differ from other peer groups. That is, while studies of gang members often include comparisons to nongang youth, most studies of gangs do not compare these groups to nongang groups. For example, Hughes and Short (2005) reexamined field notes collected in the 1960s to illuminate microprocesses that produce violence in gangs. Their study is not able, however, to describe whether and


195

how these processes differ in nongang groups. We are therefore left, for the most part, to theorize how gangs may be qualitatively different. Prior work has supported the perspective that male gang members perceive status and power as resources to be protected. Since delinquency is among the activities that confer these resources, males may restrict the participation of females in such activities when females’ presence in the group nearly equals that of males and poses a threat to males’ dominance in the group. To the extent that gender imbalances are present in larger society and, ostensibly, in nongang peer groups, we might expect that females would experience suppression of their activities in these nongang groups as well. Gangs, however, may arguably have a generally different, and more masculine, quality than most nongang groups. Some scholars, for example, assert that gangs and crime are arenas in which masculinities are played out or performed by males who have little access to conventional, or hegemonic, masculinity (e.g., Anderson 1999; Connell 1995; Messerschmidt 1993). We might therefore expect greater suppression of females’ activities within sex-balanced gangs than within sex-balanced nongang groups. That is, it may be that an institutional perspective is suitable for gangs, but a generic explanation is better suited for nongang peer groups. In this chapter, we build on the work of organizational sociology and social psychology, as well as prior work of Peterson et al. (2001). We apply these perspectives to the study of individual delinquency among gang and nongang peer groups. The key research question of interest is as follows: To what extent are the patterns of delinquency among members of groups of differing sex composition similar or different for gang and nongang peer groups? That is, are the effects of gender dynamics similar in gang and nongang groups, are they amplified in the gang context, or are they completely different?

11.4

Current Study

The data used in this study come from the National Evaluation of G.R.E.A.T. (Gang Resistance Education and Training), a longitudinal panel study (2006–2012) of a school-based gang prevention program taking place in seven cities across the USA (see Esbensen et al. 2011, for more detail about the evaluation design). The G.R.E.A.T. program is an American gang prevention program taught by law enforcement officers and targeted at middle school youth (i.e., 11- and 12-year-olds). Cities were chosen to participate in the evaluation based on the existence of an established G.R.E.A.T. program, geographic and demographic diversity, and presence of gang activity. The final seven sites represent a wide range of cities from the east to the west coast and include the following: Albuquerque, New Mexico; Chicago, Illinois; Greeley, Colorado; Nashville, Tennessee; Philadelphia, Pennsylvania; Portland, Oregon; and a city in the Dallas/Fort Worth, Texas area. All students in the selected classrooms were eligible to participate in the evaluation. A total of 4,905 students were enrolled in the 195 participating classrooms in the 31 middle schools at the beginning of the data collection process. After a thorough

196


active parental consent process, 89% (N = 4,372) of youth returned consent forms, and 78% (N = 3,820) were given permission by a parent or guardian to participate in the evaluation (11% of parents declined) (see Esbensen et al. 2008 for a detailed description of procedures and factors related to active consent). In 2006, students completed pretest and posttest (waves 1 and 2) self-report surveys with completion rates of 98.3% and 94.6%, respectively.3 The sample has also completed four annual follow-up surveys (waves 3, 4, 5, and 6), with rates of 87%, 83%, 75%, and 73%, respectively. These completion rates are very good, especially given the mobility of the sample. While the evaluation began in 31 schools, by wave 4, the research team surveyed students in 219 different schools. The present study analyzes data from wave 4, collected during the 2008–2009 school year. This wave was selected so that students in this sample would be of comparable age (13–14 years old) to the sample utilized in Peterson and associates’ (2001) prior study of gang sex composition.4 The final analysis sample was derived using listwise deletion of data due to attrition and missing information on the key variables. Deletion of individuals who did not complete wave 4 (N = 661) led to a sample size of 3,159. Missing data on the key variables led to the exclusion of an additional 865 cases. This produced a final analysis sample size of 2,294 or 60% of the original sample. Missing data analysis demonstrated that youth lost due to listwise deletion had significantly higher levels of property and violent offending than the analysis sample. In terms of demographics, there was a significantly higher number of female youth in the analysis sample, so males were more likely to be excluded.

11.4.1

Measures

This study uses both demographic and offending variables to examine the differences (within- and across-sex) in delinquency between members of different types of gang and nongang peer groups. Demographics. Three demographic variables, sex, race/ethnicity, and age, were included to examine the differences between gang and nongang females and males across group types. In terms of the sex distribution in the full sample, females comprised 54.1% of the sample. Race/ethnicity was dummy-coded into White (27.5%), Black (17.2%), Hispanic (39.5%), and other race/ethnicity (15.8%). The other category included biracial youth, Native American youth, and Asian youth. The majority of the youth were 13 years of age or younger, with a mean age of 13.52. Gang membership. Gang membership is measured in a manner that is consistent with the Eurogang Program of Research definition of youth street gangs: “A street gang is any durable, street-oriented youth group whose involvement in illegal activity is part of its group identity” (Klein and Maxson 2006, p. 4). Gang membership was determined using the Eurogang funneling approach. In this approach, youth are asked specific questions about different characteristics and behaviors of their primary peer group. These seven questions address the different aspects of the


197

Table 11.2 Questions in the Eurogang definition funneling approacha Question Na %a 1. Do you have a group of friends that you spend time with doing things 2,265 97.6 together or just hanging out? (no, yes) 2. About how many people, other than you, belong to this group? 2,165 92.4 (1, 2–5, 6–10, 11–20, 21–50, 51–100, more than 100) 3. Which one of the following categories best describes the ages of most 2,108 90.0 of the people in your group? ( 5 euro 28.6 3.9 7.33 7.2 % buying stolen goods 28.6 6.4 4.47 48.2 % stealing bike or moped 11.4 1.3 8.77 34.9 % stealing a car 2.9 0.3 9.67 10.8 % burglary 2.9 0.4 7.25 7.2 % stealing in other ways 8.6 1.6 5.38 18.1 % hitting or fighting 28.6 14.3 2.00 55.4 % wounding someone 22.9 4.8 4.77 32.5 % robbing someone 2.9 0.3 9.67 8.4

279

Boys Non-gang (N = 915) 0.86 9.4 12.2 9.5 3.0 12.0 6.1 1.2 0.8 2.3 21.7 6.8 0.8

Ratio 4.15 5.31 4.44 3.17 2.40 4.02 5.72 9.00 9.00 7.87 2.55 4.78 10.50

gangs with a majority of males. About one in three are in sex-balanced groups, and almost no boys (only two) are in female-dominated gangs. In Table 16.3, the level of delinquency among gang and non-gang girls and boys is presented. The first row shows the average delinquency score (the number, or variety, of offense types) of gang and non-gang girls and gang and non-gang boys. The next 12 rows show percentage of youths who were involved in the specific offense types during the school year up to the moment of data collection, ranging from acts of vandalism to minor and serious forms of property and violent crime. The table shows that gang girls are much more involved in delinquent behavior than non-gang girls, and also than non-gang boys. Their average delinquency score is four times that of non-gang girls (and more than three times that of non-gang boys). The last year prevalence for all offense types also is much higher among gang girls than it is for non-gang girls and boys. The difference is most marked for destroying things, the more serious forms of property crime, and for robbery. For most offenses, gang boys are more involved than gang girls (but interesting exceptions are graffiti, and shoplifting, that seem to be more common among gang and non-gang girls). The gang/non-gang ratios for boys and girls, however, are not very different in their general magnitude (with a few exceptions, e.g., for shoplifting more expensive things). Table 16.4 shows the average delinquency scores for girls and boys in gangs of different sex composition. Because there are so few gang members in groups with a female majority, these types are combined with the sex-balanced gangs in Table 16.4. The table shows that girls in gangs with a male majority are slightly more delinquent on average; however, the difference is not statistically significant. For boys, the pattern is different. Boys who are in all-male gangs appear to be less delinquent on average than other boys. Boys who are in gangs with a minority of girls appear to be the most delinquent on average; boys with a substantial number of girls in the

280

F.M. Weerman Table 16.4 Average delinquency scores among girls and boys in differentially composed gangs Girls Boys Composition (N = 32) (N = 82) All-male gangs 2.40 Majority (>2/3) male 2.85 4.30 Sex-balanced or predominantly female gangs 2.00 3.55 F-value and significance 1.57 n.s. 4.30* n.s. not significant. *p < 0.05

Table 16.5 Risk factors for gang membership among girls and boys: results from a logistic regression analysis Girls Boys B (s.e.) B (s.e.) Constant −4.218 (.325)** −3.414 (.213)** Bond with parents −.201 (.202)n.s. .067 (.152)n.s. n.s. Bond with school −.361 (.215) −.065 (.146)n.s. Level of self-control −.528 (.245)* −.719 (.173)** Conventional morality −.966 (.246)** −.308 (.177)n.s. n.s. Time spent with friends −.151 (.209) .395 (.156)* n.s. Perceived peer delinquency .236 (.216) .912 (.170)** Nagelkerke’s R2 .278 .333 * ** n.s. not significant. p < 0.05, p < 0.001

gang are slightly less delinquent but more delinquent than boys in all-male gangs. So, while for girls in gangs, the presence of girls has no effect or seems to temper delinquent behavior, for boys, the presence of girls, in particular when they are a minority in the group, seems to stimulate offending. In Table 16.5, we turn to the correlates of (or risk factors for) involvement in gangs. A logistic regression analysis (because the dependent is a dichotomous variable) was conducted to see which of the included variables were independently associated with membership in troublesome youth groups according to the Eurogang definition. The analysis was conducted separately for girls and boys to scrutinize similarities and differences; the independent variables are standardized to enable comparison. Table 16.5 reveals some interesting differences between girls and boys regarding the multivariate effects of the various potential risk factors for gang membership.14 For girls, the level of morality has the strongest effect of all variables on being involved in a gang, while the effect of morality for boys is not significant. For boys, on the other hand, the strongest effect is found for having highly delinquent friends (as perceived by the respondent), while for girls, peer delinquency has no statistically significant effect. Another peer-related variable, time with peers, also has a significant effect on gang involvement for boys, but not for girls. Only a few effects are found that are similar for both sexes: relatively low levels of self-control have a

16

Are the Correlates and Effects of Gang Membership Sex-Specific…

281

Table 16.6 Multivariate effects of gang membership on delinquent behavior among girls and boys: negative binomial regression analysis Girls Boys B (s.e.) B (s.e.) Intercept −.976 (.075)*** −.433 (.056)*** Bond with parents −.056 (.071)n.s. −.058 (.056)n.s. * Bond with school −.155 (.070) −.175 (.055)** *** Level of self-control −.342 (.079) −.224 (.062)*** Conventional morality −.145 (.082) n.s. −.124 (.066) n.s. Time spent with friends .253 (.070)*** .264 (.056)*** *** Perceived peer delinquency .694 (.079) .576 (.066)*** n.s. Being in a gang .447 (.234) .475 (.150)** Log likelihood −788.58 −1,226.14 AIC 1,593.16 2,468.29 BIC 1,630.95 2,507.53 n.s. not significant. *p < 0.05, **p < 0.01, ***p < 0.001

significant effect on gang involvement for boys and girls and the bonds with parents and school have no significant effect on gang membership for either sex. Table 16.6 presents the results of an analysis in which the effect of gang involvement on delinquent behavior is estimated for both sexes, net of the other correlates for gang membership and delinquency. A negative binomial regression analysis was employed to account for the skewed distribution of the dependent variable; all independent variables are standardized to facilitate comparison. This type of regression analysis is suitable because the delinquency variety score can be regarded as a count variable that fits the negative binomial distribution much better than the normal distribution that is the underlying assumption of the traditional OLS regression analysis. Table 16.6 reveals that there is quite some similarity between boys and girls regarding the effects of the included risk factors on delinquent behavior. For boys, as well as for girls, bond with parents and conventional morality have no significant effect, while bond with school, level of self-control, time spent with peers, and perceived peer delinquency do have significant effects on the total variety of offending. Of the included control variables, peer delinquency seem to have the strongest independent effect on delinquency,15 followed by level of self-control and time spent with peers. The last row, printed in bold, shows that for both sexes, being involved in a gang (according to the Eurogang definition) has an independent and substantial effect. It may be noted that the estimate for girls is not significant16; however, the difference in the magnitude of effects between boys and girls is not significant. The formula 100x (eb-1) can be used to estimate the percent change in delinquency for being a member of a gang compared to not being in a gang. This estimate is 56.4% for girls and 60.8% for boys; these are considerable effects that are almost equal between the sexes.

282

16.6

F.M. Weerman

Conclusion and Discussion

Since the beginning of the 1990s, criminological attention toward girls in gangs has increased. Nevertheless, studies on gangs and girls outside the United States are still scarce. For example, only a few studies in Europe quantified how often girls are involved in gangs and whether this is associated with delinquent behavior (Bradshaw 2005; Sharp et al. 2006; Haymoz and Gatti 2010), a number of American but no European studies have investigated whether risk factors for gang membership differ between boys and girls (Bjerregaard and Smith 1993; Thornberry 1998; Esbensen and Deschenes 1998), and also only American studies investigated whether girls’ and boys’ gang membership is associated to delinquency independent from important correlates of delinquency (Battin et al. 1998b; Deschenes and Esbensen 1999). In this chapter, results are presented from a Dutch study in which 1,830, mainly lower educated, adolescents are surveyed. The questionnaire contained core questions from the Eurogang youth survey, together with scales covering many potential risk factors for gang membership and delinquency. These data made it possible to investigate the representation of girls in Dutch troublesome youth groups that meet the Eurogang definition, the sex distribution of these gangs, the association of girls’ and boys’ gang involvement with delinquent behavior, risk factors for gang membership among girls and boys, and the independent effect of gang involvement for girls and boys. To begin with, the findings reveal that 4.2% of the investigated girls were members of a troublesome youth group according to the Eurogang definition. This is a substantial proportion, despite the fact that it is about half the proportion of boys (8.3%). It is quite comparable to gang membership among girls in Italy and Switzerland (Haymoz and Gatti 2010) and somewhat lower than found in the UK where girls seem to be almost equally involved in gangs or delinquent youth groups as boys (Bradshaw 2005; Sharp et al. 2006). A large part of the Dutch gang girls are either in sex-balanced or male-dominated groups; only a few are involved in a gang that is all or mostly female. For Dutch boys in gangs, the majority is in an all-male or male-majority groups; one in three is in sex-balanced gangs. This is quite comparable to what is found in Italy and Switzerland (Haymoz and Gatti 2010) and in the United States (Curry et al. 1994; Peterson et al. 2001). Like elsewhere (e.g., Bjerregaard and Smith 1993; Esbensen and Winfree 1998; Haymoz and Gatti 2010), Dutch girls in gangs are much more involved in delinquent behavior than non-gang girls and also non-gang boys. Their average delinquency score is more than four times higher than girls who are not involved in gangs, and their last year prevalence figures are elevated for all 12 investigated offense types, including serious property crimes and robbery. Among girls, gangs and delinquency seem to be associated with each other as much as among boys. Further, it seemed that the sex composition of gangs is related to delinquent behavior. Girls in gangs with relatively more boys were on average more delinquent than girls in more sex-balanced gangs. This is in line with earlier findings by Peterson et al. (2001) and with the findings for property offenses reported by Peterson and Carson (2012).17 However, the difference in the current study was not statisti-

16


283

cally significant. This may be due to a smaller sample size, but it may also be that girls in male-dominated gangs are less urged to be “one of the guys” (Miller 2001) in the Netherlands than in the United States. The presence of girls in the gang did not seem to temper the delinquent behavior of boys in the gang; boys in a gang with some girls had a significantly higher average delinquency scores than boys in allmale gangs and boys in mixed gangs. This is again in line with the earlier findings by Peterson et al. (2001). It departs from the more recent finding by Peterson and Carson (2012) that boys in sex-balanced gangs are most involved in delinquency, although it does resemble their findings for males in non-gang groups. Future research, including qualitative studies, is needed to obtain more insight into why sex composition of gangs is associated with delinquent behavior. The generic and institutional explanations provided by Peterson and Carson (2012) offer a good point of departure, but as those authors acknowledge, those perspectives do not seem to fit all of the current findings. For girls in gangs with a high proportion of boys, it may be general learning processes that account for the increase in their delinquency levels. For boys in gangs with a minority of girls, elevated levels of offending might be related to attempts to impress members of the other sex with violent or illegal behavior. The results regarding sex differences in the risk factors for gang membership were quite surprising. It appeared that for girls, low morality, low self-control, and (to a lesser extent) a weak bond with school were the most important correlates of gang involvement; for boys, peer-related factors (time with peers and peer delinquency) were the most important correlates, together with low self-control. The finding that peer-related factors are more important for boys’ gang membership is in line with the findings of Thornberry (1998) but not with the results of Bjerregaard and Smith (1993) and Esbensen and Deschenes (1998), and no study thus far has found that unconventional morality is more important for gang involvement among girls than among boys. This adds to the inconclusive results of existing studies on sex differences in gang risk factors; clearly, more research is needed, including qualitative studies, to shed more light on this issue. Are there really sex-specific risk factors for gang membership, and is it possible to trace distinct paths toward involvement in gangs for boys and girls? Another issue to investigate is whether reasons for joining gangs are different for girls as compared to boys. Earlier qualitative (e.g., Fleisher 1998) and quantitative (Esbensen et al. 1998) studies suggest that relational reasons (belonging somewhere, the gangs as a family) were more important for girls to join a gang than for boys; it would be interesting to see whether this is also the case in European gangs. The findings about the effect of gang membership on delinquency controlled for other risk factors were remarkably similar for boys and girls. For girls as well as for boys, perceived peer delinquency, level of self-control, and time spent with friends had considerable effects, but net of these effects, gang membership had a significant and substantial effect on total delinquency. For both sexes, involvement in a troublesome youth group that fits the Eurogang definition is estimated to increase delinquent behavior with more than 50%. These findings are in line with the studies that report an effect of gang membership for girls and boys above and beyond having delinquent peers (Battin et al. 1998b) and above and beyond many other correlates

284

F.M. Weerman

of delinquency (Deschenes and Esbensen 1999). Being in a gang appears to have its own effect on behavior, independent from a wide array of risk factors, for girls as well as for boys. It is interesting to note that the findings about the correlates of delinquency depart from the results deriving from the prior analysis of risk factors for gang membership. While the effects on delinquency are very similar among the sexes, there are differences in the risk factors for gang membership. For girls, morality seems to have an important effect on gang membership, but not on delinquency, while time spent with peers is not related significantly with gang membership while it is with delinquent behavior. These different results implicate that involvement in gangs and involvement in delinquency are distinct phenomena that require separate analyses and social responses to these distinct problems. As usual, this study has its limitations. An important one is the cross-sectional nature of the analysis, which has the implication that the associations and effects cannot be interpreted as causal. To get more insight into which factors really lead to more gang involvement and delinquent behavior, longitudinal data are needed (like those collected by Thornberry et al. 2003). Future research should look more into the various processes that may cause girls and boys to become involved in a gang and be influenced by them. Another limitation is that there is a lot of variation among the groups that meet the criteria of the Eurogang definition; variation that has not been incorporated in the current analyses. Future research could delve more deeply into the organizational and cultural characteristics of gangs, the risk factors that lead to membership, and the effects they have on delinquent behavior of girls and boys that are involved. A final limitation is that the current study did not include analyses investigating whether the gang effect is moderated by individual characteristics of gang girls and boys. It might be the case that some youths are more susceptible to group pressures within gangs and that other youths are relatively resistant. It is also possible that these characteristics are sex-specific. A potential line of future research may be to investigate which girls (and boys) are the most or least sensitive to the effects of gang membership and which processes are involved in it. Despite these limitations, this study has provided further evidence that a substantial number of girls are involved in gangs and that it is still worthwhile to study the associations between sex, gang involvement, and delinquent behavior. The findings show remarkable similarities with the many studies that have been conducted on gangs and girls in the United States and the ones that already have taken place in Europe. Across samples, countries, and definitions used, girls in gangs appear to be relatively more involved in delinquency than female and male non-gang members, and this relation is independent from several risk factors. Like elsewhere, it also appeared that boys and girls differed in the association between the sex composition of their gangs and delinquent behavior. However, the number of non-US studies on this issue is still limited, and it remains important to investigate whether these results can be replicated in other countries and cultures. Furthermore, several questions and uncertainties are still pending, in particular, with regard to the question whether there are sex-specific risk factors for gang involvement. In short, there is ample reason to further increase the number of studies on girls and gangs in Europe and elsewhere.

16


285

Notes 1. In some of the earlier reports on gangs resulting from the NSCR School project, data were used from the first wave in which an earlier version of the Eurogang questionnaire was employed (Esbensen and Weerman 2005; Weerman 2005). The second wave data with the current Eurogang measure were used in a report on gang violence (Klein et al. 2006). 2. In the Netherlands as a whole, 60% of the youth attends this type of school. 3. A substantial number of respondents from Surinamese (11%), Turkish (7%), and Moroccan (6%) origin were represented in the sample together with respondents from a wide range of other ethnic backgrounds. 4. Strictly speaking, alpha scores are not applicable because this is not a Likert-type scale; it consists of dichotomized items with non-normally distributed responses. 5. “I have nice parents,” “I would rather have other parents (reverse coded),” “I don’t like being with my parents (reverse coded),” “I feel fine when I am with my parents,” “I can notice that my parents love me,” “My parents are nice to me,” “My parents know what I like,” and “My parents tell me when I have done a good job.” 6. “I enjoy going to school,” “I get bored at school (reverse coded),” “I have a nice school,” “I would rather attend another school (reverse coded),” “At school I feel like at home,” “I pay attention during lessons,” “I do as much as I can for school,” and “I work hard to get good grades.” 7. “I often do things without thinking first,” “I make fun if I can, even if it leads to trouble,” “I say what I think immediately,” “I often do what I feel like immediately,” “I like to do exciting and adventurous things,” “I like to try out scary things,” “I love doing dangerous things,” “I think it’s stupid to do things for fun when you might get hurt (reverse coded),” “People better stay away from me when I am angry,” “I would rather hit someone than talk when I am angry with someone,” “I can talk out arguments calmly (reverse coded),” and “I get angry easily.” 8. “It’s all right to do something that is illegal now and then, as long as you don’t get caught,” “It’s all right to lie, if that brings you a lot of money,” “Breaking and entering at rich people’s houses is not so bad,” and “It’s all right to steal if you need money.” 9. “After school, are you usually at home or away from home?,” “How often do you spend time with your friends after school?,” “How long do you spend with your friends during weekdays?,” “How often do you meet your friends on weekends?,” and “How much time do you spend with your friends during weekend days?” 10. How many of your friends have “destroyed things on purpose,” “hit somebody so hard he or she gets wounded/hurt,” “steal small things from shops worth less than 5 Euros (like candy, pencils),” “steal things worth more than 5 Euros (like CDs, make-up, bags, jackets, or money),” “break or enter to steal something,” and “rob someone?” 11. The analyses were also done with a direct, network measure of (school) friends’ delinquency instead of a perceptual one; the results were largely similar to those presented here. I decided to use the perceptual measure in the current analyses (reported in this chapter) because this would be the most restrictive test for an independent effect of gang membership. 12. The same categorization as in Peterson et al. (2001) was used. Not all respondents in a gang provided information about how many girls and boys were in their group; therefore, the n’s are slightly lower than in Table 16.1. 13. Because of the relatively small numbers, I provide indications instead of exact percentages here. 14. It is important to note that these analyses are based on cross-sectional data, so we cannot conclude with certainty that the statistical effects on gang membership can be interpreted with causes. 15. However, it is important to keep in mind that this effect is probably overestimated because youths may have a tendency to project their own behavior on their friends. Analyses in which perceived peer delinquency is replaced by a direct, social network measure of peer delinquency results in more modest but still significant effects. 16. However, it is marginally significant (p < 0.10) and in the expected direction. 17. For property offenses, they find that girls in sex-balanced gangs have the highest levels.

286

F.M. Weerman

References Battin SR, Hill KG, Abbott RD (1998a) The contribution of gang membership beyond delinquent friends. Criminology 36:93–115 Battin SR, Thornberry TP, Hawkins JD and Krohn MD (1998b) Gang membership, delinquent peers, and delinquent behavior. Retrieved from the office of juvenile justice and delinquent prevention website: http://www.ojjdp.gov/jjbulletin/9810_2/intro.html Bendixen M, Endresen IM, Olweus D (2003) Variety and frequency scales of antisocial involvement: which one is better? Leg Crim Psychol 8:135–150 Bjerregaard B, Smith C (1993) Gender differences in gang participation, delinquency, and substance use. J Quant Criminol 4:329–355 Bradshaw P (2005) Terrors and young teams: youth gangs and delinquency in Edinburgh. In: Decker SH, Weerman FM (eds) European street gangs and troublesome youth groups. Altamira, Lanham, pp 193–218 Campbell A (1984) The girls in the gang. Basil Blackwell, Cambridge Curry GD, Ball RA, Fox RJ (1994) Gang crime and law enforcement recordkeeping. ACE-Federal Reporters, Washington, DC Decker SH, Weerman FM (2005) European street gangs and troublesome youth groups. Altamira Press, Walnut Creek Deschenes EP, Esbensen F-A (1999) Violence and gangs: gender differences in perceptions and behavior. J Quant Criminol 15:63–96 Elliott DS, Huizinga D (1989) Improving self-reported measures of delinquency. In: Klein MW (ed) Cross-national research in self-reported crime and delinquency. Kluwer, Dordrecht Esbensen F-A, Deschenes EP (1998) A multisite examination of youth gang membership: does gender matter? Criminology 36:799–828 Esbensen F-A, Huizinga D (1993) Gangs, drugs, and delinquency in a survey of urban youth. Criminology 31:565–589 Esbensen F-A, Weerman FM (2005) A cross-national comparison of youth gangs and troublesome youth groups in the United States and the Netherlands. Eur J Criminol 2:5–37 Esbensen F-A, Winfree LT (1998) Race and gender differences between gang and nongang youths: results from a multisite survey. Justice Q 15:505–526 Esbensen F-A, Deschenes EP, Winfree LT (1999) Differences between gang girls and gang boys: results from a multisite survey. Youth Soc 31:27–53 Esbensen F-A, Peterson D, Taylor TJ, Freng A (2010) Youth violence: sex and race differences in offending, victimization, and gang membership. Temple University Press, Philadelphia Fleisher M (1998) Dead end kid: gang girls and the boys they know. University of Wisconsin Press, Madison Grasmick HG, Tittle CR, Bursik RJ, Arneklev BJ (1993) Testing the core empirical implications of Gottfredson and Hirschi’s general theory of crime. J Res Crime Delinq 30:5–29 Haymoz S, Gatti U (2010) Girls members of deviant youth groups, offending behaviour and victimization: results from the ISRD2 in Italy and Switzerland. Eur J Crim Pol Res 16:167–182 Klein MW, Kerner H-J, Maxson CL, Weitekamp E (2001) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Press, Amsterdam Klein MW, Weerman FM, Thornberry TP (2006) Street gang violence in Europe. Eur J Criminol 3:413–437 Maxson CL, Whitlock ML (2002) Joining the gang: gender differences in risk factors for gang membership. In: Huff CR (ed) Gangs in America III. Sage, Thousand Oaks Miller WB (1975) Violence by youth gangs and youth groups as a crime problem in major American cities. GPO, Washington, DC Miller J (2001) One of the guys: girls, gangs and gender. Oxford University Press, New York Moore J (1991) Going down to the barrio: homeboys and homegirls in change. Temple University Press, Philadelphia

16


287

Peterson D, Carson DC (2012) The sex composition of groups and youths’ delinquency: a comparison of gang and non-gang peer groups. In: Esbensen F-A, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Peterson D, Miller J, Esbensen F-A (2001) The impact of sex composition on gangs and gang member delinquency. Criminology 39:411–439 Sharp C, Aldridge J, Medina J (2006) Delinquent youth groups and offending behaviour: findings from the 2004 offending, crime and justice survey. Home Office, London Thornberry TP (1998) Membership in youth gangs and involvement in serious and violent offending. In: Loeber R, Farrington DP (eds) Serious and violent juvenile offenders: risk factors and successful interventions. Sage, Thousand Oaks Thornberry TP, Krohn MD, Lizotte AJ, Smith CA, Tobin K (2003) Gangs and delinquency in developmental perspective. Cambridge University Press, New York Trasher FM (1927) The gang: a study of 1,313 gangs in Chicago. University of Chicago Press, Chicago Weerman F (2005) Identification and self-identification: using a survey to study gangs in the Netherlands. In: Decker SH, Weerman FM (eds) European street gangs and troublesome youth groups. Alta Mira Press, Lanham Weerman FM, Smeenk WH (2005) Peer similarity in delinquency for different types of friends: a comparison using two measurement methods. Criminology 43:499–524 Weerman FM, Maxson CL, Esbensen F, Aldridge J, Medina J and van Gemert F (2009) Eurogang program manual background, development, and use of the Eurogang instruments in multi-site, multi-method comparative research. Retrieved from the Eurogang network website: http:// www.umsl.edu/~ccj/eurogang/ Eurogang_20Manual.pdf Van Gemert F, Peterson D, Lien I-L (2008) Street gangs, igration and ethnicity. Willan, Cullompton

Part IV

Future Directions

Chapter 17

The Next Decade of Eurogang Program Research Malcolm W. Klein

Since its inception in the late 1990s, the Eurogang Program has accomplished much to further the cause of a comparative street gang science. It has grown in numbers of countries from one dozen to two, from a dozen participants to over 200, from confusion about gang definition to firm and codified agreements, and from no common research methods to agreement on use of, and dissemination of, five data collection instruments. Yet all this is but a foundation for the possible—and, I hope—likely future for the enterprise. In this chapter, I will cover four topics that should engage Eurogang Program scholars as well as the public officials who should be engaged in street gang policies. I will first address the question of why there has been so little comparative street gang research. By “comparative,” I refer explicitly to cross-national research. Second, I will suggest some directions that the Eurogang Program should go in order to make use of its accomplishments to date. Third, I will suggest some procedures that will further these directions. Finally, I will offer a few thoughts on the relationship between Eurogang research and European street gang policies and practices.

17.1

Why Has There Been So Little Comparative Street Gang Research?

In the published materials from the very first Eurogang workshop held well over a decade ago, Maxson articulated the program’s ultimate goal: “Fight the compulsion to ‘do your own thing’ and instead organize a systematic approach to gang research in multiple settings” (2001, p. 303). That is far more easily said than done. The program’s intent is to foster comparative research, meaning two or more sites, not

M.W. Klein (*) Department of Psychology, University of Southern California, Los Angeles, CA, USA F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_17, © Springer Science+Business Media, LLC 2012

291

292

M.W. Klein

one site, with reference to data in others. The emphasis is on cross-national comparisons. Imagine, for example, research on street gangs in Frankfurt and Manchester, or Bosnia-Herzegovina and Genoa, or Oslo and Stockholm, or neoNazi gangs evolving in Moscow and Jerusalem. A few such comparisons are already in the Eurogang publications: van Gemert (2001), Weerman and Esbensen (2005), and Huizinga and Schumann (2001). Two others have emerged to date from the use of the Eurogang definition of street gangs in the International Self-Report Delinquency Program (ISRD II) comparing French and Italian gangs (Blaya and Gatti 2010) and Swiss and Italian gangs (Haymoz and Gatti 2010). The current volume includes an additional study comparing US and Caribbean gangs (Pyrooz et al. 2012, in this volume). These examples comprise a good beginning, but far more is needed and anticipated. If indeed more is to come, collaborators will have to deal with a series of complexities such as those listed here1: • Researchers must reduce their reliance on stereotypical gang depictions, such as those often offered by the media, the police, and the “traditional” street gangs (Klein and Maxson 2006) described in earlier American works. • Language and translation problems must be faced in the derivation and implementation of survey instruments. • Because official (government) definitions and statistics vary widely across nations, far more reliance must be placed on self-report and quantifiable observational crime data. • Attention must be paid to the comparability of research sites (blocks, neighborhoods, cities) for purposes of statistical control and contextual interactions with street gangs. • It is absolutely critical that identical (not roughly similar) methods and instruments be used in the sites being compared. • Ethnic, cultural, and national origin differences must be assessed and appreciated across research sites. In many potential European research sites, for example, it is not sufficient merely to document street gangs under the broad category of Muslim immigrants. • Sample sources need to be similar, whether drawn from informant suggestions, school or household surveys, social agency rosters, or police, court, or correctional data files. • Because gangs themselves are complex phenomena and draw forth highly varied perspectives from those who are concerned with them, reliance on any one method of study can be very misleading. Youth surveys yield different pictures from street observations and ethnographies, while surveys of “experts” and data collection from official files yield other, often conflicting, conclusions. Triangulation by the use of several methods, as urged and provided for by the Eurogang program, is highly recommended. • Collection of “contextual” data can illuminate gang issues. For example, demographic data and city descriptors as well as gang descriptors have proven most helpful. Typologizing gang locations and gang structures avoids inappropriate generalizations and burrows in on gang/environment interactions (see Klein and Maxson 2006, Chaps. 5 and 6).

17 The Next Decade of Eurogang Program Research

293

• One of the nagging problems throughout the years of the Eurogang program has been resistance to its interests and procedures from two sources. Both have largely been overcome, but not without some serious tension and—frankly—suspicions. Both sources of conflict must still be faced with good spirit and intellectual honesty. The first is the tendency, often politically based, to deny that there are street gangs “in our city” (see the extended discussions in Klein et al. 2001). The second is the fear among some scholars that to do research on gangs is to reify them and create a “moral panic” about their existence. Certainly, there is room for caution in response to such concerns. But as one of my students noted colorfully, “If you hide your head in the sand, you’re gonna be bitten in the butt!” • Finally (in this listing, at least) is the issue of obtaining funding for comparative, especially cross-national, gang research. The costs of street gang research can range from low to levels well beyond the capacities of most individuals and institutions. Add to those costs the solutions to problems such as those listed above and the complexity become manifest. Comparative, multi-method research is a large enterprise, requiring both researcher dedication and sponsor appreciation of its value. This combination has not yet been achieved in the gang world. The above list of problems, issues, and needs may seem prohibitive to a program of cross-national street gang research. And indeed, I think it has been. This, alone, may explain why so little comparative research has been undertaken. But also, such research requires scholars with an intellectual commitment to the values of collaboration and the ambition to strike out beyond one’s own, convenient research home, to cross borders to unfamiliar locations and researchers. It is not necessarily comfortable out there.

17.2

Where We Ought to Go

Based on the above, it becomes clear that our comparative enterprise is not for the naïve nor for the faint of heart. Yet solutions to complexities are available, and the challenges are invigorating. Two old but somewhat unresolved issues strike me as needing continuing explication as we resolve the research complexities. Both are pivotal to advancing the science of street gangs: definitions and group processes. Fortunately, both are highlighted in many chapters of this volume. Five comments seem particularly in order on street gang definitions, especially in view of the extensive work undertaken by members of the Eurogang Program. First, clarity is needed not only on what is a street gang but also how we define gang members and gang-related crime. As for defining street gangs, this volume echoes the approach explicated in all the previous Eurogang writings, and I will return to this in later pages. As to gang members, the issue has not been resolved. How youth identify themselves as members or nonmembers varies greatly, and often does not correlate well with the designations given by outsiders—other youth, parents, police, social service providers, the media, and certainly researchers. Clearly, gang membership

294

M.W. Klein

resides in the eyes of the beholders, and the beholders are often in disagreement. Since gangs are informal groups, the criteria for membership will necessarily be ambiguous. Further, most gang researchers consider membership to be fluid rather than static: gang members move in and out, more or less committed to the groups over time and location. For research purposes, it has proven quite adequate to determine one’s gang status by self-admissions (“I am, I am not”) or by response to the seven operationalization items in the Eurogang protocols. But for policy purposes, let the definer beware! Gang-related crime is perhaps a less difficult definitional issue, as three major variations have emerged. The most comprehensive and most common of these is that any crime committed by or against a gang member (however defined) is a gangrelated crime.2 A more restricted approach accepts as gang-related only those offenses that are motivated by gang membership. Common examples are gang graffiti, offenses designed to increase one’s status in the gang, and offenses required by the gang or its leaders. Personally motivated offenses (e.g., theft, robbery, minor drug sales for one’s own profit) are not included. US research suggests that this motive definition yields about half as many “gang-related” offenses as the definition based solely on member status (Maxson and Klein 1990, 1996). A third, even more restrictive, definition of gang-related crime is any offense demonstrably committed with the intent of benefiting the gang rather than the individual. Common examples are “drive-by” shootings or other retaliations against rival gangs, robberies, or drug sales in which the proceeds are shared with or turned over to the gang, witness intimidation in gang trials, and intragang assaults which sanction members for violation of gang norms. Because both the motive definition and the gang benefit definition involve often difficult inferences, the simpler and more productive member definition is usually employed. It may even be logically more appropriate since so much research has shown that gang joining itself leads to a major increase in members’ offense levels (Battin et al. 1998; Thornberry et al. 2003). A second comment on street gang definitions, especially using the Eurogang approach, is pivotal to establishing a body of knowledge specific to street gangs. The Eurogang definition of a street gang—“…any durable, street-oriented youth group whose involvement in illegal activity is part of its group identity”—was designed among other things to distinguish the street gang from other groups often referred to as “gangs.” This permits the development of a science of street gangs, specifically, by ruling out many other (and far more numerous) youth groups as well as prison gangs, motorcycle gangs, terrorist gangs, adult crime syndicates and drug cartels, and so on. Thus, we can exclude, but leave for separate study, the global groups discussed by Hagedorn (2008); the arrest-defined networks described by Sarnecki (2001); the soccer hooligans dramatically pictured by Buford (1991); the social movement groups described by Brotherton (2007), by Barrios (2007), and by Feixa et al. (2008); and the three variations on youth groups described by Sullivan (1989) but not labeled “gangs” by him. The less we lump all deviant groups together, the better we can understand the properties of the street gang by itself.


295

Still, we must not so reify the street gang that we overlook the variability in its nature and in the nature of its membership. The Maxson/Klein gang structure typology (2006) emphasizes this variability, but those authors’ data also reveal considerable structural change over time as one kind of gang structure may morph into another. Good illustrations are provided by Salagaev in Russia (2001) and by Lien in Norway (2008). There is also the clear fact that individual members’ status in their gang changes over time.3 Many join and then desist more than once; many participate for a while and then disappear, often to reappear and participate again. Some will affiliate with a particular clique within a larger gang but foreswear membership in the larger group. These kinds of fluidity seem almost inherent in the nature of street gangs, despite the implication of a static structure implied by the Eurogang definition. Two nagging additional themes often present themselves in the Eurogang research and clearly present definitional quandaries. The first is the place of females in street gangs, and second is the ethnic character of gang membership. Attention to these remains warranted. While substantial female gang membership has been noted in recent American research, gender has not gained research attention in European studies. This latter note is probably due more to the perspectives of the researchers than to the absence of girls in European gangs. An overview of street gangs in 30 countries by Gatti et al. (2011) finds female membership at about 30%. Such a figure demands more research attention in the future concerning the roles of female gang members. The same is true about the role of ethnicity or national origin of European street gang members. It is not a question of the existence of nonindigenous gang members—this is well demonstrated in many studies—but of the “so what” implicit in this fact. Prominent American research suggests that group processes trump race and ethnicity in street gang function and structure (see, e.g., Jankowski 1991; Klein and Maxson 2006; Vigil 2002). But in Europe, the variety of ethnic and national backgrounds is far greater. It is not yet clear how important these orientations may be, nor how significantly their impact differs from that of indigenous gang membership; “white” gangs seem more numerous in Europe than in the USA. All the above points about street gang definitions—especially given the Eurogang Program’s consensus definition—illustrate for me the importance of “sticking to our guns” about this definition. It is not an individual arbitrary definition such as those usually cited in the gang literature. It results from some rather intensive discussions started in the very first Eurogang workshop in Schmitten, Germany, in 1998. This, in turn, led to a “working group” on definitions in Oslo in 1999, continued thereafter in Leuven, Belgium, and Egmond aan Zee, Holland. The original working group of eleven researchers has expanded to include members from the USA, Russia, Finland, Sweden, Italy, Belgium, Holland, and Norway, and eventually as well to the continuing discourse among other working group leaders and the steering committee of the Eurogang program. The final consensus Eurogang Program definition of street gangs has now been applied to research in over 30 countries, with remarkable success. It has, in Europe, proven even more satisfactory than the robust self-reported

296

M.W. Klein

definition (“are you a member”—“have you ever been a member”) in use in most previous studies (Gatti et al., 2011). This is in part due to the greater ambiguity of the term “gang” in many European settings, yielding a higher percentage of gang admissions. I emphasize all this because, even among some Eurogang researchers, there has been an understandable tendency to “play” with the Eurogang definition to fit local or national exigencies or to include youth groups not fully covered by the definition (see Aldridge et al. 2007) and Pedersen and Lindstad 2012, in this volume). I find it important for the science of street gangs, for building a reliable and valid knowledge base about gangs, and for positioning ourselves to advise and affect public policy to remain “pure” to our consensus Eurogang definition. If there are local or national practices that hamper the use of the definition in a gang study, then go ahead and do the study, but not under the umbrella of the Eurogang Program. If there is a desire to study groups that do not fall under the definition—and there are far more of these than “our” gangs—then go ahead and do so, but not under the umbrella of the Eurogang program. One should neither alter the definition to fit the group nor transform the description of the group to fit the definition. But meanwhile, most components of the Eurogang surveys and ethnographic guidelines can still be used in non-Eurogang research settings to study non-Eurogang defined groups. We do not have to throw out the baby with the bathwater. With this said, I want to offer a few comments about group processes and street gangs. One of the most remarkable failings among street gang researchers has been their negligence in noting that gangs are groups, not mere aggregations of individuals. This point is nicely made by Hennigan and Spanovic (2012) in this volume: gang membership provides a social identity that goes well beyond individual essences. The study of street gangs must account for these group identities, for alternate group structures, for cohesiveness and group norms and leadership, and for the “oppositional culture” of gangs described by Moore and Vigil (1989). There are several points I find especially intriguing in connection with gang identities: • The risk factors for joining street gangs are far fewer than those for merely engaging in antisocial activity—they are group specific. When we have isolated these (see Hennigan and Maxson, in press and Klein and Maxson 2006, Chap. 4), we will have a major handle on gang prevention. • The “reality” of gang membership is often very different among the members from that same “reality” as viewed by outsiders such as police, parents, social workers, neighborhood residents, and researchers. What the group means affects each of these perspectives with often conflicting implications for understanding and control. • While both scholars and practitioners often specify gang differences according to ethnic backgrounds or neighborhood (community) characteristics, my own experience and a good deal of research suggest that group processes trump ethnicity and neighborhood. It is this predominance of group factors over others that allows us to suggest the generic existence of street gangs as a class of phenomena, distinguishable from other classes.


297

• As a corollary to the above, I have become convinced by a broad array of gang research, both in the USA and Europe, that street gangs are qualitatively different from most other youth groups. If we stroll through an urban area at lunchtime, we may see many collectivities and groups—football or soccer teams in the park, office workers around a lunch table, a telephone work crew outside an office building, a school group visiting a museum with its teacher, and drivers at a rail station taxi stand conversing while waiting for their next customers. We note these groups but think little about them. Yet if we see a dozen oddly dressed youth with tattoos or shaved heads hanging aimlessly on a corner, we may pay special attention. A street gang, we wonder? They seem different from other groups; we are prepared to be wary of them. • Street gang members endorse some norms that most groups do not. They often express exaggerated in-group/out-group sentiments. They express rhetoric (if not true commitment) to special loyalty, to violence, and to resistance to adult efforts to intervene with them. In short, we come to see street gangs not just as collectivities subject to normal group processes but as youth groups qualitatively different from others. They are in a class almost of their own. Before intervening with them, we had better understand their special nature lest we do more harm than good in our sincere efforts to affect them.

17.3

How Should We Get There?

The Eurogang Program has moved from an exciting new enterprise to a relatively mature and stable international endeavor. It has a large membership, a continuing strategy of multinational workshops and research publications, and a well-tested set of research protocols. But to address many of the concerns raised in this chapter and to progress to a more sophisticated level of contribution, I would like to suggest more emphasis on several developments. First, while there was mention in the first section of this chapter of several cross-national gang comparisons, these were retrospective analyses. The program now needs to foster prospective research comparisons, wherein collaborating teams of researchers plan new research using comparable sampling designs and instruments. I entertain a fantasy (which includes abundant funding, of course) something like this: A research team in Frankfurt selects a school district that includes several gang territories (probably predominantly Turkish in makeup). The Frankfurt police already have intelligence on these gangs in the district.4 The team undertakes a school survey (primary and secondary levels), while ethnographers initiate gang member contacts through several social service agencies. Both the survey instrument and the ethnographic guidelines are those available from the Eurogang Program. At the same time, a research team in Stockholm undertakes the same approach in a school district with several active street gangs including at least one with a Turkish background. Again, the police have intelligence on these groups.5

298

M.W. Klein

A third team undertakes a similar study as a follow-up to already existing work in Amsterdam undertaken by van Gemert and Fleisher (2005), a Hollander and an American working together. Finally, a comparative study is launched in an American city—same design, same instruments—by one or more of the young scholars in attendance at the ninth Eurogang Workshop in Los Angeles in 2008.

In my fantasy, all four (I would settle for three) projects run smoothly and conterminously, a model of prospective gang research procedures and analyzes. An adjunct to the fantasy is my second suggestion, implicit in the above, that henceforth each Eurogang project be a multimethod project. The instruments for this have already been devised. I would like to see survey researchers forced to entertain the advantages of concurrent ethnographic or observational procedures; I would like to see ethnographers forced to undertake standardized survey procedures. The ambiguities of street gangs all but require triangulation of research methods. Granted, this is easy to say and not so easy to do. It will require practice by tolerant researchers. A third suggestion, raising a host of problems which nonetheless have been overcome in other gang research, is to initiate some long-term longitudinal or panel studies of gang and non-gang youth, again of course, using comparable designs and methods across sites. The model seen in the Rochester, Denver, and Pittsburgh coordinated studies is proof of how valuable such an approach can be (see, e.g., Thornberry et al. 2003). My fourth and final suggestion in this section is for Eurogang researchers to align themselves with and undertake evaluations of street gang prevention and intervention programs. This kind of exposure to gang issues produced a whole era of new and extensive gang knowledge in the twentieth century in the USA. The major advances provided by combining programs and evaluations constitute what now is much of the knowledge base upon which modern gang research—often unknowingly—is supported.6 One reason, among many, is that gang intervention programs greatly enhance researchers’ access to gang areas and gang members. Further, I know of no street gang researchers who have been harmed, physically or intellectually, by direct exposure to street gang practitioners!

17.4

A Few Thoughts on Policies and Practices

In addition to data advances noted above, two policy-related gang programs of major import have recently been evaluated in the USA, the G.R.E.A.T model7 and the Spergel model.8 Both were (and are) multisite applications of theoretically relevant practices with reported mixed results as to their effectiveness (Klein and Maxson 2006). My interest here is twofold. First, both programs have yielded reams of data about a wide variety of gang members and gang issues. Second, both illustrate the potential for incorporating street gang knowledge into public policies and practices.


299

It is not a requirement of gang research (or gang researchers, for that matter) to be directly relevant to policy. But many researchers wish to bridge the gap and should not be discouraged from doing so. I myself have been somewhat resistant over the years, but without clear success (Klein 2009a). Thus, I have a recommendation for the Eurogang Program’s Steering Committee and active members. I suggest that the next (if possible) Eurogang workshop be organized around the question of collaboration between street gang researchers and policy makers (not just practitioners). Past workshops have included a few such people on rare occasions (from ministries of justice and police departments), but there has been no consistent attempt to foster a serious dialog (but see the contrast reported in Klein 2009b). This is unfortunate, since one of the initial motivations for the program was to avoid in Europe the remarkable failure of American communities over many decades to guide gang control policy by reference to the increasingly available knowledge base about street gangs (see Klein and Maxson 2006, Chaps. 7 and 8). By and large, policy has not been guided by data, but rather by conventional wisdoms of unproven value. Similarly, many gang researchers have been ill-tuned to the songs of practitioner groups and makers of public policy about gangs. The advantage we would have over most other attempts to bridge the research/ policy gap is that the Eurogang program explicitly endorses and (hopefully) facilitates comparative and cross-national endeavors which force greater attention to both gang and policy differences. This makes the task more difficult but also more enlightening about the barriers to collaboration and the alternative routes to collaboration. What more could one ask?

Notes 1. The following items are explicated more fully in the Eurogang Program Manual (Weerman et al., 2009) and in Klein (2011). 2. A few exceptions are often noted, such as domestic disputes and child abuse. 3. To say nothing of some members’ switching their affiliation to another, or even rival, gang. 4. I have seen some of these police files. 5. I have seen some of these materials as well. 6. See Cloward and Ohlin, 1960; Miller, 1967; New York City Youth Board, 1960; Yablonsky 1962; Short and Strodtbeck, 1965; Spergel, 1966; Moore, 1978, 1991; Klein, 1971, 1995. 7. Gang Resistance Education and Training, evaluation of police-directed prevention programs. 8. Labeled for its founder, but known technically as the Comprehensive Community-Wide Approach to Gang Prevention, Intervention, and Suppression Program.

References Aldridge et al. (2007). A Cross-national multimethod study about the reciprocal relationships between deligment youth groups and social exclusion: lessons for policy and pratice. Proposal submitted. The European commission Manchester, University of Manchester school of Law.

300

M.W. Klein

Barrios L (2007) Gangs and spirituality of liberation. In: Hagedorn JM (ed) Gangs in the global city. University of Illinois Press, Urbana Battin SR, Hill KG, Abbot RD, Catalano RF, Hawkins JD (1998) The contribution of gang membership to delinquency beyond delinquent friends. Criminol 36:93–115 Blaya C, Gatti U (2010) Deviant youth groups in Italy and France: prevalence and characteristics. Eur J Criml Policy Res 16:127–144 Brotherton DC (2007) Toward the gang as a social movement. In: Hagedorn JM (ed) Gangs in the global city. University of Illinois Press, Urbana Buford B (1991) Among the thugs: the experience, and the seduction of crowd violence. Norton, New York Cloward RA, Ohlin LE (1960) Delinquency and opportunity: a theory of delinquent gangs. The Free, New York Feixa C, Canelles N, Porzio L, Recio C, Giliberti L (2008) Latin Kings in Barcelona. In: van Gemert F, Peterson D, Lien IL (eds) Street gangs, migration, and ethnicity. Willan Publishing, Portland Gatti U, Haymoz S and Schadee H (2011) Deviant youth groups in 30 Countries: results from the second international self-report delinquency study. Int Crim Just Rev Hagedorn JM (2008) A world of gangs: armed young men and gangsta culture. University of Minnesota Press, Minneapolis Haymoz S, Gatti U (2010) Girl members of deviant youth groups, offending behavior and victimization: results from the ISRD 2 in Italy and Switzerland. Eur J Crim Policy Res 16:167–182 Hennigan KM and Maxson CL (in press) New directions in street gang prevention for youth: the Los Angeles experience. In S Sides (Ed) Post-Ghetto: reimaging South Los Angeles. Berkeley, University of California Hennigan KM, Spanovic M (2012) Gang dynamics through the lens of social identity theory. In: Esbensen FA, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Huizinga D, Schumann KF (2001) Gang membership in Bremen and Denver: comparative longitudinal data. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Academic Publishers, Dordrecht Jankowski MS (1991) Islands in the street: gangs and American urban society. University of California Press, Berkeley Klein MW (1971) Street gangs and street workers. Prentice-Hall, Englewood Cliffs Klein MW (1995) The American street gang: its nature, prevalence, and control. Oxford University Press, Oxford Klein MW (2009a) Bootlegging: a career caught between fantasy and reality (Volmer Award Address). Criminol Public Pol 8:1–12 Klein MW (2009b) The street gangs of Euroburg: a story of research. iUniverse, Bloomington Klein MW (2011) Who can you believe? Complexities of international street gang research. Int Crim Just Rev Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, Oxford Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) (2001) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Academic Publishers, Dordrecht Lien IL (2008) “Nemesis” and the Achilles heel of Pakistani gangs in Norway. In: van Gemert F, Peterson D, Lien IL (eds) Street gangs, migration, and ethnicity. Willan Publishing, Portland Maxson CL (2001) A proposal for multi-site study of European gangs and youth groups. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Academic Publishers, Dordrecht Maxson CL, Klein MW (1990) Street gang violence: twice as great, or half as great? In: Huff CR (ed) Gangs in America. Sage Publications, Newbury Park Maxson CL, Klein MW (1996) Defining gang homicide: an updated look at member and motive approaches. In: Huff CR (ed) Gangs in America, 2nd edn. Sage Publications, Thousand Oaks


301

Miller WB (1967) Theft behavior in city gangs. In: Klein MW (ed) Juvenile gangs in context: theory, research, and action. Prentice-Hall, Englewood Cliffs Moore JW (1978) Homeboys: gangs, drugs, and prison in the barrios of Los Angeles. Temple University, Philadelphia Moore JW (1991) Going down to the barrio: homeboys and homegirls in change. Temple University, Philadelphia Moore JW, Vigil JD (1989) Chicano gangs: group norms and individual factors related to adult criminality. Aztlan 18:31–42 New York City Youth Board (1960) Reaching the fighting gang. New York City Youth Board, New York Pedersen ML, Lindstad J (2012) The Danish gang joining project: methodological issues and preliminary results. In: Esbensen FA, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Pyrooz DC, Fox AM, Katz C, Decker SH (2012) Investigating gang organization: dimensionality and validity across multiple data sources. In: Esbensen FA, Maxson CL (eds) Youth gangs in international perspective: tales from the Eurogang program of research. Springer, New York Salagaev A (2001) Evolution of delinquent gangs in Russia. In: Klein MW, Kerner HJ, Maxson CL (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Academic Publishers, Dordrecht Sarnecki J (2001) Delinquent networks: youth co-offending in Stockholm. Cambridge University Press, Cambridge Short JF, Strodtbeck FL (1965) Group process and gang delinquency. University of Chicago Press, Chicago Spergel IA (1966) Street gang work: theory and practice. Addison-Wesley Publishing Co., Reading Sullivan ML (1989) Getting paid: youth crime and work in the inner city. Cornell University Press, Ithaca Thornberry TP, Krohn MD, Lizotte AJ, Smith CA, Tobin K (2003) Gangs and delinquency in developmental perspective. Cambridge University Press, Cambridge van Gemert F (2001) Crips in orange: gangs and groups in the Netherlands. In: Klein MW, Kerner HJ, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer Academic Publishers, Dordrecht van Gemert F, Fleisher MS (2005) In the grip of the group. In: Decker SH, Weerman FM (eds) European street gangs and troublesome youth groups. Altamira, Lanham Vigil JD (2002) A rainbow of gangs: street cultures in the mega-city. University of Texas Press, Austin Weerman FM, Esbensen FA (2005) A cross-national comparison of youth gangs: the United States and the Netherlands. In: Decker SH, Weerman FM (eds) European street gangs and troublesome youth groups. Altamira, Lanham Weerman FM, Maxson CL, Esbensen FA, Aldridge J, Medina J and van Gemert F (2009) Eurogang program manual. Available at www.umsl.edu/~ccj/eurogang/euroganghome.htm Yablonsky L (1962) The vistent gang. The macmillion press, 1963

Chapter 18

The Intersection of Gang Definition and Group Process: Concluding Observations Cheryl L. Maxson and Finn-Aage Esbensen

The themes of gang definitions and group processes comprised the substantive agenda of the tenth Eurogang workshop in June 2010, and the chapters in this volume reflect major advances in our understanding of both issues. Why focus on these particular themes among all the important and compelling aspects of the multimethod, comparative study of street gangs? And, why at this particular time? Engaging in the editing process of these chapters has caused us to reflect back on our original intention in selecting these themes for the workshop and to offer a broader context in which to consider their contributions. We follow that discussion with an assessment of the ways in which we believe these chapters advance our knowledge of street gang dynamics and the methods we use to understand them. While we have separated the chapters into two substantive sections on definitional issues and group processes, we find these chapters—and those appearing in the third section of miscellaneous gang depictions—cross over this conceptual divide and speak directly to one another. The chapters in the definitional section capture the debates about the content of the Eurogang and other definitions of street gangs, particularly the necessary and sufficient elements of definition. This work considers the application of definitions in research and in practice. The challenge of replication of core findings about street gangs across different methods and locations compels several of the authors to reconsider definitional alternatives.

C.L. Maxson (*) Department of Criminology, Law and Society, University of California, Irvine, CA, USA e-mail: [email protected] F.-A. Esbensen Department of Criminology and Criminal Justice, University of Missouri-St. Louis, St. Louis, MO, USA e-mail: [email protected] F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3_18, © Springer Science+Business Media, LLC 2012

303

304

C.L. Maxson and F.-A. Esbensen

The chapters that we placed in the group process section engage the important issues of group identity, shared marginalization experiences, social bonding, group interaction, and the development of normative expectations. These dynamics provide a fulcrum for chapter authors to consider the ways in which these processes contribute to the reinforcement of oppositional norms and to the facilitation of criminal activity within street gangs. Group identity and the normative orientation toward crime are the elements of the Eurogang definition most directly challenged by work in the first section and represent a clear intersection between the two themes. Another is the way in which aspects of group interaction and gang organization—or lack thereof—becomes a defining element in practice. In the remainder of this chapter, we illustrate these intersections with the work compiled in this volume.

18.1

Why Visit the Theme of Definition Now

As we note in the introductory chapter, definitional issues emerged in the earliest discussions of the Eurogang participants. It became clear early on in our group process that there was a compelling need to develop broad agreement on a definition of the central object of study. This need stemmed from our methodological ambitions, political or representational concerns, and conceptual orientation. Methodologically, we needed to come to a consensus about a definition of street gangs that would allow multiple investigators, working in various locations and utilizing different methods, to have confidence that they were studying the same apples and not oranges. Although the straightforward, self-report item: “Are you a member of a street gang?” has served US researchers well,1 our work required an alternative that did not specifically employ the “g word.” This was partially attributable to sensitivity to the possibility of moral panic and overreaction arising from a visible assessment of the prevalence and nature of street gangs in Europe. We were conversant with instances of overreaction to gangs as they emerged in nontraditional settings in the USA (Huff 1989; Zatz and Portillos 2000), and this potential was evident from our European colleagues’ reactions at the first Eurogang workshop. Moreover, one motivation to foster systematic gang research in Europe arose from what Klein (2001) first dubbed “The Eurogang Paradox.” He argued that some European policymakers, practitioners, and researchers failed to recognize street gang issues in their locales based on misperceptions of US street gangs. These common stereotypes ascribed levels of organization, structure, cohesion, and violence that in fact do not typify American street gangs. We needed an approach that did not require subjects to share a common understanding of a term, “gang,” or its closest approximation in various languages and cultures. Moreover, we were sensitive to the potential that the much more ubiquitous lawviolating, delinquent groups that characterize adolescent cultures might be mislabeled as street gangs and therefore subject to inappropriate law enforcement suppression. Finally, our desire to include ethnography and other qualitative techniques among the panoply of methodological approaches embraced by the Eurogang

18 The Intersection of Gang Definition and Group Process…

305

research design required that we articulate a gang definition that could be utilized for ground-level observation in a comparable manner to that adopted by a survey researcher. The Eurogang definition was an elegant solution to these issues! Researchers could avoid use of “gang” and deconstruct the definitional elements to measure dimensions of the youth’s primary friendship group, and via Eurogang guidelines, determine whether or not the group is a gang. By extension, membership in such groups may signal gang membership, although presumably, some youth do not consider themselves members even though their group of friends is a gang. As Klein and Maxson (2006, p. 4) note: The components—durable, street oriented, youth group, identity with illegal activity—are definers of street gangs. They are the minimal necessary and sufficient elements to recognize a street gang. Many other characteristics are common descriptors, not definers. One thinks of leadership, cohesiveness, ethnicity, gender and distinctive argot, clothing, tattoos, or hand signs, for instance. These are variables that help us to capture variations across gangs, but they are not necessary definers of a street gang…

We raise this distinction between definers and descriptors in the introductory chapter, and it is worth noting that group processes are captured among definers (i.e., street orientation and illegal activity as a component of group identity) and descriptors (e.g., leadership and cohesiveness). We intentionally avoided group structure and organization as defining elements, and this is a point to which we will return in a later discussion. Some of the definitional elements were the subject of intense debate over the series of workshops in which it was developed and refined through pilot testing of the instruments. The debate represented the range of views regarding what each of us distilled from extant gang research to be most important about street gangs and the desire to achieve a consensus on a fundamental conceptualization of the intrinsic nature of gangs. Some argued that the element of street orientation and public visibility of group activity, hanging out, was critical to distinguishing street gangs from other group forms, as well as the dynamic that sets the stage for negative responses to the group and development of oppositional norms within the group. There has been a long debate about the inclusion of crime as a defining element of street gangs (Klein 1995; Short 1996), but we concluded that it was an orientation toward the group’s involvement in crime that was essential: gang members view their group as involved in crime, often violence, they expect it, and it is a critical element of the group identity (Decker 1996). Other components of the Eurogang definition such as group (size of at least three), age (youth to young adult), and duration (a few months) were less subject to debate, although the de Jong chapter in this volume reveals that even spontaneously formed groups may bear markers of gang activity and response. We faced challenges as to the validity of the Eurogang definition from the outset. This is perhaps understandable as new researchers joined the group without having weathered the long and intense debates required to reach the consensus. Moreover, as the salience of gang issues has grown in more places and more research has been conducted, researchers’ understandings of what constitutes a gang might change. Eurogang has been successful in fostering more research on gangs in Europe, opening up

306


new empirical avenues to fuel the discussions. For instance, Aldridge and colleagues in this volume draw examples from their Eurogang-guided ethnographic study in an English city, as well as from the utilization of Eurogang survey items in the 2004 Offending Crime and Justice Survey conducted in England and Wales and the 2008 Madrid Youth Survey to formulate their challenge to the Eurogang definition. Finally, despite our earliest efforts to avoid American centrism in the development of Eurogang, it must be noted that US gang researchers represented a large contingent in the first several workshops. There is always the possibility that the Eurogang definition might not provide a good fit with Euro-gangs (as we come to learn more about them from new research) or with European researchers. While the Eurogang’s OGs argue for preserving the original content of the definition—at least until empirically invalidated—there has been constant scrutiny and discussion. With a cadre of new researchers and new studies becoming available, it seemed to be a good time to revisit the debate. In particular, recent research which has implemented the Eurogang definition in youth and expert surveys as well as with qualitative approaches allows us to examine the deconstructed components of youth groups. The chapters by Pederson and Lindstad, Aldridge and colleagues, de Jong, and Smithson and colleagues articulate new challenges to building a European consensus on defining street gangs. As we develop new stances regarding the intrinsic elements of street gangs, it will be important to consider the implications for theory and gang practice. If street orientation (defined broadly by Eurogang) and crime identity are not defining elements, then the theories we use to explain the emergence of gangs and their persistence, as well as the policies and programs we develop to encourage desistance, will need to reflect these new understandings. But the definitional issues are not merely limited to the content of the Eurogang definition; we are also challenged to consider the meaning and implication of the application of different definitional approaches. This new research also affords the opportunity to examine the categorization of youth as gang members from three different standpoints. As we note in the introduction, youth can self-identify themselves as a street gang member, they can identify their primary group of friends as a street gang, or their primary group can be designated as a street gang due to measures on the Eurogang definitional criteria. We now have studies that capture data articulating all three definitional approaches so we can begin to examine the issue of convergence, as Matsuda and her colleagues illustrate in this volume. Even with perfect measurement, we expect some differences, particularly where individual self-identification may differ from whether the primary group of friends is a gang, whether by the youth’s nomination or by the Eurogang construction. Despite the common use of individual self-identification, the most direct test of the Eurogang construction is the youth’s nomination of their primary group as a gang. Ultimately, these definitional debates and empirical assessments become most interesting when centered upon the question of whether gang definitions matter, and if so, how. Clearly, a lack of convergence in definitional approaches would generate differences in who we called gang members, and this is vividly illustrated by the study by Matsuda, Esbensen, and Carson. We might expect these differences to result in different understandings of patterns of gang membership and crime.


307

However, the Matsuda analysis reported here and a very different investigation of the impact of different police designations of gang and nongang homicides by Maxson and Klein (1996) suggest that these definitional differences matter more for numbers, and less for the patterns and our understanding of what really constitutes a gang and gang crime. Fundamentally, we need to know whether gang knowledge built largely on the self-identification approach and in the USA will hold up elsewhere and be resilient to these different definitional approaches. We place a high value on studies that permit the replication of findings with different methods and settings, and several of the chapters in this volume accomplish this. The studies by Melde and Esbensen, by Peterson and Carson, and by Weerman provide evidence that core findings about gang membership can be replicated with the application of the Eurogang definition.

18.2

Why Visit the Theme of Group Process Now

There has been woefully little progress in our understanding of the group dynamics in street gangs since the pioneering mid-twentieth-century work by W. Miller (1973), Spergel (1966), Short and Strodtbeck (1965), and Klein (1971). Notable exceptions to this generalization can be found in the early work of Vigil (1988), Hagedorn (1988), Decker and van Winkle (1996), Fleisher (1998), and Miller (2001). However, given Klein’s (1995; this volume) contention that these processes are what make street gangs qualitatively different from other youth groups, it is striking that the field has not advanced further in this arena. In recent years, the field of gang research has favored survey techniques over ethnographic methods, which necessarily has meant that attention to group characteristics has been limited to that which individuals report rather than that which is observed by a researcher. These survey studies, often generating large data sets and often capturing individual changes over time, have generated new knowledge on a variety of important topics such as individual gang/crime relationships, risk factors and reasons for joining, and patterns of victimization as well as offending. These surveys have captured youths’ views on the groups to which they belong, including participation in group delinquent and noncriminal activity as well as sex and racial/ethnic composition. Researchers have used survey approaches to ask youth and other gang experts about gang structures and organization as witnessed by several chapters in this volume (Pyrooz, Fox, Katz and Decker; Maxson; Hennigan and Spanovic; Pedersen and Lindstad), but these fail to capture the dynamics within and between groups at the group level. This was one of the reasons that it was so important to us to get the ethnographers on board with the Eurogang program. There has been a failure to build on earlier ethnographic work to more recent understandings about what makes these groups so special—so qualitatively different—from other youth groups. Qualitative research could help us better to understand how these group processes solidify gang/crime relationships, promote the seduction of gangs and their persistence, and generate resistance to intervention efforts. For example, how do group

308


processes like the formation of group identity, status derived from belonging, normative expectations of violence, and group conflict promote crime involvement, collective action, and the willingness to join or move out of gangs? In addition to the survey researchers noted above, these are the kinds of issues which ethnographic accounts such as those offered by de Jong, van Gemert, Sela-Shayovitz, and Smithson and colleagues attempt to address in this volume. The purpose in making group processes a theme of the Eurogang X workshop was both to highlight new research conducted within the Eurogang framework and independent from it and to foster more attention on gang group processes. The discourse on definitions has been interesting in and of itself, but a very useful byproduct of the debate has been the steady attention to dynamics of street orientation, group visibility (with its links to reinforcement of oppositional values and negative interactions with the community), and crime orientation.

18.3

The Intersection of New Knowledge About Gang Definitions and Group Processes

In the preceding chapter, Klein strikes a cautious note regarding the designation of youth as gang members by outsiders rather than relying on youths’ perceptions of their own status. He issues this caution even as he urges Eurogang to “stick to our guns” and retain the Eurogang definition which permits the application of the necessary and sufficient elements of researchers’ conceptions of street gang by the operationalizations agreed to by Eurogang participants. In other words, researchers can designate youth as gang members if their group of friends falls within the parameters specified by the Eurogang definition. Several of the chapters in this volume provide grist for the mill of the definitional debates. Van Gemert’s review of the history of defining gangs in the Netherlands argues that police practitioners have relied on a misperception of structure or level of organization to inappropriately exclude groups that should be designated as gangs (jeugdbendes, in Dutch). He argues that by emphasizing structural elements seen as characterizing American street gangs (i.e., tight, organized, and hierarchical), Dutch police have defined their way out of a street gang problem. Notwithstanding the dangers of falling into the trap of the Eurogang Paradox, the Pyrooz et al. chapter in this volume uses multinational data to examine, and ultimately challenge, the construct of gang organization. The Dutch practitioner approach would typify the instrumentalrational end of the organizational continuum identified by Pyrooz and his colleagues: vertical structure, rules and discipline, levels of membership, regular meetings, and stable leadership. Maxson’s discussion of perceptions of prison gang structures also places them on the high end of the organized scale. At the other end of the continuum, the informal-diffuse organizational structure characterizes gangs as having limited codes of conduct, informal interpersonal relationships, shifting leadership, and members who act in self-interested ways rather than for group goals (see also Decker and Curry 2000). These researchers detect inconsistent and relatively modest correlations


309

among four common measures of organizational structure in three different samples (two from the USA and one from the Caribbean), which they observe does not inspire confidence in a latent structure of gang organizational structure. Thus, the Dutch gang designation procedure described by van Gemert employs a definitional construct (organizational structure) of questionable reliability and validity. The ambiguity about the impact of organizational structure on gang behavior is evidenced by the Pyrooz et al. findings: a summated index of the four features of gang organization is marginally associated with delinquency, but not victimization, in two of the samples, whereas individual items are not. Individual items are associated with victimization in all three samples, but they are different items except in one case, where the direction is reversed. The gang organization scale is related to victimization, but only in one US sample. This study epitomizes the value of multinational, comparative research even while it reminds us of the difficulty in identifying patterns of gang characteristics that withstand the test of place. In an interesting contrast to van Gemert, Smithson, Monchuk, and Armitage are primarily concerned with the overlabeling of youth groups as gangs by police practitioners in England. They argue that police designate groups as gangs in two of their four study neighborhoods that appear to fall on the informal-diffuse end of the organizational continuum hypothesized by Pyrooz et al. Echoing Klein’s warning, Smithson and colleagues find that the youth they studied do not identify as gang members, and in fact, eschew the label. But the gang label carries with it attribution of coherence and identity, and the researchers suggest that in raising the labeled groups’ status, the group becomes more attractive to youth. Youths’ use of public space appears to be part of the catalyst for this labeling process, which relates to the Aldridge et al. challenge to street orientation as a defining element that we discuss shortly. Smithson et al. appear to be more comfortable with the designation of the more organized, “family firms” as gang. Indeed, the work by van Gemert and by the Smithson group touches on the controversial issue of the validity of organizational structure as a defining element of street gangs. Pyrooz et al. (this volume, p. 86) argue that organizational structure means much more: “The organizational structure of gangs matters to the extent that it influences the social processes (e.g., contagion, collective action) responsible for making gangs qualitatively distinct from other types of criminal and delinquent groups.” This provides an example of how definitional debates intersect with the theme of group process. Matsuda and her colleagues attack the definitional debate directly by applying to their youth survey sample three commonly used gang definitions: self-identification, the youth’s attribution of the main friendship group as a street gang, and the friendship group meets the Eurogang definitional criteria. The researchers argue that this is not just a game of numbers: past studies reveal that as definitions change, the characteristics of gang members and gangs change as well. This body of research suggests that how we understand gangs—their core characteristics and dynamics— is shaped by the way we define them. The Matsuda study clearly shows that the numbers do change. They find very little convergence in the three approaches; just 9% of youth that met at least one of the three conditions (12% of the overall sample) is designated a gang member by all three and two-thirds of the youth by just one

310


approach. The Eurogang definition netted the largest number of youth. We might have expected the most overlap between the Eurogang and youth attribution of group as gang approach since each addresses the group rather than individual participation, but there is roughly the same convergence between each of the pairs. However, in contrast with the extant literature, Matsuda et al. find that the different youth captured by these definitional approaches share common ground on the risk factors, attitudes, and behaviors we have come to associate with gang membership. Moreover, all three approaches identify youth that are notably distinct from nongang youth. This work suggests that gang designation matters particularly in terms of numbers— and less so in terms of characterization of gang involvement and gangs. The Matsuda chapter implies that designations of gang membership are fairly robust, and this is confirmed by work presented in other chapters. Melde and Esbensen replicate earlier work on the processes emanating from gang membership by utilizing a “turning points” framework. They investigate changes in relationships, routine activities, attitudes, and emotions associated with gang membership and clarify the impact of gang participation on crime. In this chapter, they use the Eurogang approach to designate gang members, whereas their earlier study used the friendship-group-is-a-gang designation by the youth. These findings confirm and strengthen the observation that gang membership results in deep and deleterious impacts on youths that join them—regardless of the way in which researchers operationalize gang membership. Peterson and Carson also attempt to replicate earlier work on the impact of sex composition of the gang on delinquency by employing the Eurogang definition instead of self-nomination. In general, they replicate the earlier study in a different sample and find that gendered processes within groups vary considerably in gang and nongang groups. Accordingly, they reinforce the depiction of gangs, along with their internal group processes, as wholly distinct from the dynamics within other youth groups. They single out violence as a critical dimension of gang group process. Specifically, they highlight the group orientation toward violence and the gendered dynamic of male preservation of dominance by controlling access to status-conferring involvement in violence. Sex composition of gangs influences how normative expectations and values are expressed in patterns of criminal activity. This chapter also provides a wonderful illustration of the intersection between definitional issues and group process. It would be interesting to know more about the distinctions between different gang designations in practice, in particular the way in which they might be explained by variations in self- and group identity. Matsuda et al. suggest that different gang structures might lend themselves more or less to certain variations, but it would also be useful to explore the group processes that provoke one designation over the other. A strong group identification and cohesion may promote the group-is-a-gang definition, whereas marginal/fringe gang status may engender denial of individual gang membership even though the primary group is considered to be a gang. These issues could profitably be pursued with qualitative research techniques as exemplified by the Aldridge et al. challenge to the inclusion of the more process-oriented elements in the Eurogang definition.


311

Aldridge and her colleagues argue against the inclusion of two elements of Eurogang definition: street orientation and illegal activity as part of its group identity. They correctly observe that determining group identity is not an easy task for researchers, and Hennigan and Spanovic share their concern that group identity has not been sufficiently conceptualized by gang researchers. In their qualitative study of the gang landscape of a city in England, Aldridge et al. identified groups that spend most of their time together in private residences rather than in public spaces. They note that such groups meet other Eurogang criteria of gang designation and conclude that street orientation should be a descriptor rather than definer of gangs. This suggestion deserves further investigation in more locations to determine whether electronic modes of communication or other shifts in youth culture have dramatically changed this foundational element of gang group process. Researchers have argued that it is this public face of gangs—large gatherings of unoccupied males hanging out on streets and public spaces, intimidating neighbors and passersby—that supports a cycle of negative response by authorities and the development of oppositional norms that reinforce gang identity (Moore and Vigil 1989). If this is not generally the case among street gangs in Europe, we need to understand why not (weather; public transportation; informal social controls; youth culture; policing) and especially more about the alternative processes that foster negative norms and values. Maxson notes that this type of visibility is achieved by institutional gangs even where their freedom to congregate is tightly constrained. Her work suggests that this means of projecting “groupness” (see Hennigan and Spanovic) is a critical feature of gang dynamics. Pedersen and Lindstad provide another illustration of a research site that struggled with the street orientation component of the Eurogang definition. In order to increase the numbers in their gang category for analysis, these researchers adopted a policy of designating Copenhagen youth that endorsed five of the six Eurogang definitional criteria as gang members. Ten percent of the student sample met the established criteria, but an additional group of 64 students (3%) met five. All belonged to groups with a crime orientation, but most did not report a lot of time spent in public places. The authors argued that it was appropriate to include these youth in the gang category because severe Danish weather made it less likely that youth would congregate outside. Furthermore, there is a concerted effort by Danish social workers to promote access to youth clubs. Similar to findings by Matsuda and her colleagues, Pedersen and Lindstad report that less than one-third of gang members consider their group of friends to be a gang. The researchers attribute this to the Danish perception that gangs connote more organized groups, perhaps similar to the Dutch situation described by van Gemert. Pedersen and Lindstad interpret their finding that gang members have more friends in gangs and spend more time with them as establishing a pathway to more organized gangs. Their work illustrates the value of the Eurogang deconstruction of definitional elements in an environment where direct gang attribution may be limited by laden terminology (i.e., preconceptions of gangs as highly organized groups). One alternative to street orientation might be the power of the marginalization processes described in this volume by Sela-Shayovitz. In her study of a neo-Nazi gang comprised of former immigrants from the USSR, Sela-Shayovitz finds that migration

312


and globalization create the conditions of marginalization and alienation that spawn the formation of group identity and reinforce the adherence to negative norms and values. She identifies the Internet as critical to the shaping and reinforcement of deviant normative codes. The Internet was not only a major source of information about neo-Nazi ideology and culture but also served to connect these youth to the members of other German and Russian neo-Nazi gangs, providing the opportunity for further socialization and the solidification of group identity. Elleyne and Wood also highlight marginalization processes in discussing their study of high school students in London. For these researchers, conflicts with authorities and antiauthority attitudes are critical to reinforcing group identity. They find such attitudes are directly related to gang crime. Moreover, they note that group identity trumps individual identity: “it can be argued that the individual identity of these young people merges with the gang identity” (p. 159). The Hennigan and Spanovic chapter speaks directly to this issue of the distinction between individual and group identity, a point to which we will return shortly. The de Jong chapter also illustrates the process of negative social reinforcement in his description of a spontaneous group on a 1-day crime spree. While not meeting the Eurogang definitional element of duration (although de Jong argues that durable social networks spawn such spontaneous crime groups), he speculates that similar group processes, such as the strengthening of group identity through oppositional experiences, reinforcement of delinquent norms, and peer pressure to adapt behavior to group values, can explain individual involvement in crime and nuisance activity. But Aldridge et al. also challenge the component of group identity that includes an orientation toward criminal activity that is a definitional element of Eurogang. These scholars encountered groups in their study that fall within the scope of the Eurogang definition that they feel should not be called gangs. The types of groups to which they refer are pot smokers that hang out in a park, illegal ravers, and clubbers. It is interesting to note that the criminal activity involved in all three types revolves around the consumption of illegal substances. The use of drugs and not other criminal activity also typifies other groups in Spain which the researchers are concerned about designating as “gang.” Notwithstanding the legitimate operational arguments proffered by Aldridge and colleagues, this strikes us as throwing out the baby with the bathtub, as Klein (this volume, p. 298) exhorts us specifically not to do: “One should neither alter the definition to fit the group, nor transform the description of the group to fit the definition.” Drug consumption by youths and the means by which they might be captured by the operationalization (“Is doing illegal things accepted by or okay for your group?”) of a dimension of this definitional element represents a legitimate challenge to Eurogang participants. We are not sure that Aldridge et al.’s (this volume, p. 48) preferred alternative of replacing involvement in crime as part of the group identity with “group has a reputation of a willingness to resort to violence or its threat, even if rarely enacted” is a profitable direction, but it certainly bears further discussion and empirical assessment. Clearly, the Swedish gangs described by Rostami and Leinfelt fit squarely within this parameter of group identity. They refer to “violence capital” as an intimidation technique employed by gang members and provide a fascinating description of the use of crime as a business trademark among Swedish gangs.


313

The issue of group identity provides the central focus of the chapter by Hennigan and Spanovic. These researchers investigate the relationships between group identification, cohesiveness, and group boundaries to criminal activity reported by Los Angeles youth. It is important to note that their measure of group identity is not imbued with orientation toward crime or street visibility; their goal is to understand how aspects of group processes related to criminal activity may differentiate gang from nongang youth. Hennigan and Spanovic find that the strength of identification with the primary friendship group is strongly related to criminal activity and violence for gang members, but not for other youth. Gang members have weaker ingroup identification and less cohesiveness and describe less clear boundaries to their groups. Cohesiveness increases identification for both gang and nongang youth, but identification mediates the relationship between cohesiveness and crime among gang members. Hennigan and Spanovic include deterrence measures in their analysis and show that expectations of detection and punishment are associated with crime involvement by nongang but not gang youth. This study provides empirical evidence of different group processes among gang and nongang youth. Gangs are qualitatively different, and group identity matters as a possible explanation for the higher rates of crime and violence among gang members noted by so much gang research.

18.4

Conclusion

The work gathered in this volume advances our understanding of gang definitions and group processes in several directions. Even while aspects of the gang definition adopted by Eurogang continue to provoke discussion, researchers have employed it to study gangs in a range of locations and generated replication of core findings in previous research. The group process components of the definition emerge as more problematic than other elements, further reinforcing the need to expand research efforts in this area. We argue that group processes distinguish gangs from other delinquent youth groups, and the work reported here takes a small step in advancing our understanding of these dynamics. A better understanding of group processes in gangs may well be a key to producing more effective efforts to prevent and reduce gang involvement. Klein (this volume) exhorts Eurogang participants to engage more directly with gang practitioners. He jokes that he knows no researchers that have been harmed by exposure to gang practitioners. Some of the authors draw practice implications from their research, but only the Rostami and Leinfelt chapter highlights a gang program. Klein suggests that the next Eurogang workshop focus on collaboration between researchers and gang practitioners and policy makers. The 2011 Eurogang workshop (EG XI) in Denmark will do just that. It is scheduled to coincide with a conference of gang practitioners, and members of each group are attending both types of sessions. Moreover, planning is already underway for Eurogang XII to coincide with the Stockholm SGIP practitioner conference, again allowing for the type of interaction which might promote collaborative projects. We hope that sufficient progress will be made to generate a future Eurogang volume on the intersection between gang research and practice.

314


Finally, we note that the impediments to conducting comparative, multimethod gang research detailed by Klein are expressed in the contents of the current volume. The Pyrooz et al. chapter is the only work included that provides cross-national comparisons, and none of the data sets they use were developed under the Eurogang aegis. As we continue our work to build a consensus gang definition, to explore the fundamental group processes that seem so important to making gangs distinctive, and to investigate the other aspects of gangs that intrigue us as gang scholars, it is critical to advocate for studies that truly implement the Eurogang research paradigm. That is our purpose and that is our hope.

Note 1. In an earlier example of a comparative effort, Huizinga and Schumann (2001) used the youth’s identification of their friendship group as a gang (Denver) or bande (Hamburg).

References Decker SH (1996) Collective and normative features of gang violence. Justice Q 13:243–264 Decker SH, Curry GD (2000) Addressing key feature of gang membership: measuring the involvement of young members. J Crim Just 28:473–482 Decker SH, van Winkle B (1996) Life in the gang: family, friends, and violence. Cambridge University Press, Cambridge Fleisher M (1998) Dead end kids: gang girls and the boys they know. University of Wisconsin Press, Madison Hagedorn JM (1988) People and folks: gangs, crime, and the underclass in a rustbelt city. Lake View, Chicago Huizinga D, Schumann KF (2001) Gang membership in bremen and Denver: comparative longitudinal data. In: Klein MW, Kerner H-J, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Norwell Huff RC (1989) Youth gangs and public policy. Crime Delinq 35:524–537 Klein MW (1971) Street gangs and street workers. Prentice-Hall, Englewood Cliffs Klein MW (1995) The American street gang: its nature, prevalence, and control. Oxford University Press, Oxford Klein MW (2001) Resolving the Eurogang paradox. In: Klein MW, Kerner H-J, Maxson CL, Weitekamp EGM (eds) The Eurogang paradox: street gangs and youth groups in the U.S. and Europe. Kluwer, Norwell Klein MW, Maxson CL (2006) Street gang patterns and policies. Oxford University Press, Oxford Maxson CL, Klein MW (1996) Defining gang homicide: an updated look at member and motive approaches. In: Huff CR (ed) Gangs in America, 2nd edn. Sage Publications, Thousand Oaks Miller J (2001) One of the guys: girls, gangs and gender. Oxford University Press, New York Miller W (1973) Race, sex, and gangs: the Molls. Society 11:32–35 Moore JW, Vigil JD (1989) Chicano gangs: group norms and individual factors related to adult criminality. Aztlan 18:31–42 Short JF (1996) Gangs and adolescent violence. University of Colorado, Center for the Study and Prevention of Violence, Boulder


315

Short JF, Strodtbeck FL (1965) Group process and gang delinquency. University of Chicago Press, Chicago Spergel IA (1966) Street gang work: theory and practice. Addison-Wesley Publishing Co., Reading Vigil JD (1988) Barrio gangs: street life and identity in southern California. University of Texas Press, Austin Zatz MS, Portillos EI (2000) Voices from the barrio: Chicano/a gangs, families, and communities. Criminology 38:369–401

Index

A Arizona Arrestee Drug Abuse Monitoring (ADAM), 86, 90–95, 97–99 Attitudes, 10, 19, 21, 23, 25–27, 30, 31, 42, 110, 151, 152, 154–156, 159–165, 170, 184, 208, 213, 233, 272, 310, 312 and delinquency, 170

B Bikers, 240 Blalock, H.M., 190, 207 Blau, P.M., 190

C Cohesion, 31, 57, 75, 117, 132–135, 137–144, 157, 189, 212, 248, 265, 306, 310 Comparative, 1–13, 35, 54, 57, 86, 90, 91, 96–98, 100, 111, 194, 291–293, 298, 299, 305, 316 Comparative research, 2, 100, 291, 293, 299, Convergence, 306, 309, 310 Crime and violence among neighborhood peer groups, 143

D Definition, 1, 17, 35, 53–67, 69, 88, 108, 128, 155, 169, 196, 212, 226, 240, 252, 272, 291, 303 Delinquency, 7, 18, 40, 67, 86, 136, 153, 169, 189–209, 213, 243, 259, 271–285, 292, 309

Delinquent behavior, 85, 154, 155, 164, 169, 170, 194, 221, 225–229, 231–234, 271, 272, 274–277, 279–284 Deterrence, 136–140, 146, 313 Drug use, 43, 46, 91, 101, 275

E Effect of gang involvement on delinquency, 88, 274, 284 Emotions, 10, 184, 185, 214, 310 Enforcement, 2, 6, 12, 17, 20, 30, 72, 111, 121, 195, 251–268, 304 England and Wales, 37, 47, 48, 68, 306 Ethnic youth, 24, 28, 71, 72, 94, 200 Ethnography, 3, 4, 8, 74, 226, 304 EU, 81 Eurogang definition, 4, 19, 35–50, 54, 74, 108, 175, 197, 215, 226, 243, 274, 294, 306, 309 future, 313 history, 13 paradox, 9, 66, 69–80, 304, 308 progress, 313 Eurogang youth survey, 37, 46, 112, 243, 277, 284

F Female gang members and delinquency, 277 Friends are gangs (or gang friends), 27, 28

F.-A. Esbensen and C.L. Maxson (eds.), Youth Gangs in International Perspective: Results from the Eurogang Program of Research, DOI 10.1007/978-1-4614-1659-3, © Springer Science+Business Media, LLC 2012

317

318 G Gang(s) attitudes, 10, 19, 25, 26, 88, 152, 159–160, 170, 184, 208, 216 behavior, 12, 90, 102, 132, 156, 158, 161, 211, 220, 309 correlates, 12, 29, 273–287 crime and violence, 142, 143, 189, 313 definition, 3, 6, 17–19, 23, 25–28, 30, 66, 70–74, 80, 114, 295, 297, 305–316 depictions, 12–13, 70, 75, 294, 305 girls in, 273, 275, 276, 281, 282, 284–285 insignia, 93 interventions, 184 involvement, 30, 50, 55–57, 60, 67, 145, 155–157, 160–163, 165, 170, 173, 179–181, 183, 184, 192, 277, 282–286, 310, 313 joining, 10, 12, 85, 88, 156, 179, 184, 239–249, 283, 296 leadership, 101, 133 meetings, 4, 7, 87, 92–93, 99 organization, 9, 10, 85–102, 109, 117, 118, 304, 308, 309 rivalry, 55 rules, 92–93, 95–99, 101, 162, 240, 255 structure, 55, 62–65, 90, 101, 118, 151, 155–162, 164, 294, 297, 307, 308, 310 and group process, 151, 156–162, 164 sex composition, 194 Gang Resistance Education and Training (GREAT), 20, 32, 86, 90 – 99, 103, 195, 209, 299 Gender issues and group dynamics, 11, 190–191, 195, 205–209, 310 and group process, 189, 208, 310 and sex composition, 190–191 Globalization, 11, 211–221, 312 GREAT.See Gang Resistance Education and Training (GREAT) Group dynamics, 128, 133, 143, 164, 227–229, 231–234, 309 Group identification, 132, 135, 136, 138–141, 143, 310, 313 Group identity, 5, 6, 9, 11, 13, 31, 36, 40–46, 48, 54, 56, 74, 86, 113–117, 121, 128, 185, 189, 196, 215, 226, 228, 232–234, 249, 256, 296, 304 305, 309, 310–313 in Spain, 312

Index Groupness, 10, 117, 132–135, 138, 140, 141, 145, 313 Group norms, 10, 44, 135, 151, 228, 232, 298 Group processes, 1–3, 10–11, 43, 44, 85, 99, 122, 127, 128, 133, 151, 153, 155, 164, 184, 211–221, 226, 228, 231, 233, 293, 295–297, 303–314 and delinquency, 12, 194–195, 205–207, 271–285

H Holistic, 12, 251–268

I Immigration, 72, 212 Informal-diffuse, 86–88, 95, 100, 308, 309 Institutional gangs, 108–109, 113–121, 311 Instrumental-rational (with respect to gang types and/or organization), 86–88, 95, 100, 308 Instrument development, 4, 5 Interactional theory, 153–156, 159, 162 Internet, 11, 45, 47, 121, 211, 213, 214, 217–221, 235, 314

J Jeugdbende, 7, 69, 73, 74, 78–80, 276, 278, 308

K Kanter, R.M., 190

L Labeling, 11, 62, 67, 152, 225, 227, 309 Life course, 10, 169–185 Logistic regression, 28–29, 94, 96–98, 156, 280

M Measurement, 1, 31, 48, 49, 54, 90, 137, 207, 208, 249, 306 Migrant youth, 78, 215, 217 Multivariate regression, 96–98

N Negative Binomial regression, 281 Neo-Nazi gangs, 213–215, 219, 220, 292, 312

Index Non-gang peer groups, 11, 135, 136, 138, 142, 189–208 Non-gang youth groups, 37 NSCR Schoolproject, 275, 285

O Organizational theory generic perspectives, 193 institutional perspectives, 205, 208 Organized crime, 53, 69, 70, 74, 239, 240, 242, 243, 245, 253–255, 266

P PANTHER. See Preventive Analysis about Network Targets for a Holistic Enforcement Response (PANTHER) Peer pressure, 156, 160, 161, 228, 231–234, 272, 312 Policy implications, 108 Practitioner’s perspective, 59–60 Preventive Analysis about Network Targets for a Holistic Enforcement Response (PANTHER), 12, 262–268 Prison gangs, definitions, 10, 107–122 Problem-oriented policing, 260, 263, 265 Psychology, 151–153, 155, 164, 195

R Recruitment, 12, 58, 72, 239, 242, 256, 257, 259, 260, 263, 265, 268 Reputation for Violence, 49, 50 Research collaboration, 2, 260, 293, 299, 313 Research methods, 3, 291, 298 Risk factors, 18–20, 24, 25, 27, 30, 74, 151, 152, 154, 212, 242, 244, 246, 248, 272, 274, 275, 277, 280–284, 296, 306, 310 gang membership, 152, 275, 277, 280, 282–284

S Self-nomination, 8, 18, 19, 26–28, 31, 48, 185, 197, 209, 310

319 Sex composition, 209, 272, 274, 282–284, 310 and delinquency, 11, 189–209, 272, 279, 282, 284 of gangs, 11, 190, 191, 272, 274, 278, 279, 282, 283, 284, 310 of peer groups, 11, 189–209 Sex differences, 23, 192, 200, 204, 207, 283 Social group processes, 11, 211–221 Social identity, 44, 157, 159, 228, 229, 296 Social identity theory, 10, 127–146 Social networking, 39, 50, 213 Street culture, 230–232, 234 Street gangs, 3, 17, 37–40, 55, 74, 86, 107, 127, 153, 170, 196, 212, 240, 251, 291, 303 definition, 295 Street orientation, 7, 9, 39, 50, 115, 117, 120, 134, 155, 185, 249, 305, 306, 308, 309, 311 Structural elements, 78, 80, 308 Structures, 10, 21, 30, 63, 64, 69, 86, 90, 107–122, 153, 191, 265, 296 Sweden, 1, 3, 12, 240, 251–261, 268, 295

T Trinidad and Tobago, 1, 9, 10, 86, 90–92, 98, 99, 102 Turning Point, 10, 169–186, 310

U UK, 40, 41, 44, 46, 50, 53–58, 60, 62, 64, 66, 67, 155, 160, 273, 282

V Validity, 9, 19, 25, 26, 30, 31, 35–50, 95, 137, 241, 305, 309 Variety scores, 92, 281

Y Young people’s perspective, 60–62 Youth corrections, 110–111 Youth survey, 2, 3, 8, 12, 36, 37, 43–47, 54, 90, 91, 112, 114, 174, 176, 243, 292, 306, 309

Youth Gangs in International Perspective: Results from the Eurogang Program of Research

Gangs

The Amazon from an International Law Perspective

Effective Negotiation: From Research to Results

Social Capital: An International Research Program

Gangs and Delinquency in Developmental Perspective (Cambridge Studies in Criminology)

Gangs and Delinquency in Developmental Perspective (Cambridge Studies in Criminology)

X-15 Research Results

Gangs of Antares

The lobster gangs of Maine

Kuzin L.T.: Research Program

Peacebuilding: Women in International Perspective

Prostate Cancer (Recent Results in Cancer Research)

Pancreatic Cancer (Recent Results in Cancer Research)

Pancreatic Cancer (Recent Results in Cancer Research)

Angiogenesis Inhibition (Recent Results in Cancer Research)

Responding To Gangs: Evaluation And Research

Action Research: From Practice to Writing in an International Action Research Development Program (Utrecht Publications in General and Comparative Literature)

Gangs of Antares

The Problems of Disadvantaged Youth: An Economic Perspective (National Bureau of Economic Research Conference Report)

The Origins of Agriculture: An International Perspective

Encyclopedia of Gangs

Gangs and Immigrant Youth (Criminal Justice: Recent Scholarship)

Uncertainty in International Law: A Kelsenian perspective (Routledge Research in International Law)

The Loom of youth

Media and Youth: A Developmental Perspective

ELF from the programmer's perspective

The Pace of Youth

The Fountains of Youth

The fountains of youth

Youth Gangs in International Perspective: Results from the Eurogang Program of Research

Gangs

The Amazon from an International Law Perspective

Effective Negotiation: From Research to Results

Social Capital: An International Research Program

Gangs and Delinquency in Developmental Perspective (Cambridge Studies in Criminology)

Gangs and Delinquency in Developmental Perspective (Cambridge Studies in Criminology)

X-15 Research Results

Gangs of Antares

The lobster gangs of Maine

Kuzin L.T.: Research Program

Peacebuilding: Women in International Perspective

Prostate Cancer (Recent Results in Cancer Research)

Pancreatic Cancer (Recent Results in Cancer Research)

Pancreatic Cancer (Recent Results in Cancer Research)

Angiogenesis Inhibition (Recent Results in Cancer Research)

Responding To Gangs: Evaluation And Research

Action Research: From Practice to Writing in an International Action Research Development Program (Utrecht Publications in General and Comparative Literature)

Gangs of Antares

The Problems of Disadvantaged Youth: An Economic Perspective (National Bureau of Economic Research Conference Report)

The Origins of Agriculture: An International Perspective

Encyclopedia of Gangs

Gangs and Immigrant Youth (Criminal Justice: Recent Scholarship)

Uncertainty in International Law: A Kelsenian perspective (Routledge Research in International Law)

The Loom of youth

Media and Youth: A Developmental Perspective

ELF from the programmer's perspective

The Pace of Youth

The Fountains of Youth

The fountains of youth

Recommend Documents