<%BANNER%>

Understanding Gang Membership, Crime Perpetration, and Victimization among Jail Inmates

Permanent Link: http://ufdc.ufl.edu/UFE0024198/00001

Material Information

Title: Understanding Gang Membership, Crime Perpetration, and Victimization among Jail Inmates A Test of Self-Control and Social Disorganization Theories
Physical Description: 1 online resource (227 p.)
Language: english
Creator: Fox, Kathleen
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: crime, gang, inmates, jail, theory, victimization
Criminology, Law and Society -- Dissertations, Academic -- UF
Genre: Criminology, Law, and Society thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: While prior research has examined factors related to gang membership and the relationship between gangs and crime perpetration, research on the relationship between gangs and victimization is limited. The present study builds on prior research and examines factors predicting gang membership, crime perpetration, and crime victimization among over 2,000 jail inmates in Florida. In late 2008 and early 2009, I surveyed jail inmates about their experiences with crime, victimization, self-control, and perceptions of neighborhood disorganization. Results indicate that gang members are more likely than non-members to commit crime and be victimized by crime. Low self-control and some social disorganization factors are also associated with gang membership, crime perpetration, and victimization.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kathleen Fox.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Lane, Jodi S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024198:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024198/00001

Material Information

Title: Understanding Gang Membership, Crime Perpetration, and Victimization among Jail Inmates A Test of Self-Control and Social Disorganization Theories
Physical Description: 1 online resource (227 p.)
Language: english
Creator: Fox, Kathleen
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: crime, gang, inmates, jail, theory, victimization
Criminology, Law and Society -- Dissertations, Academic -- UF
Genre: Criminology, Law, and Society thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: While prior research has examined factors related to gang membership and the relationship between gangs and crime perpetration, research on the relationship between gangs and victimization is limited. The present study builds on prior research and examines factors predicting gang membership, crime perpetration, and crime victimization among over 2,000 jail inmates in Florida. In late 2008 and early 2009, I surveyed jail inmates about their experiences with crime, victimization, self-control, and perceptions of neighborhood disorganization. Results indicate that gang members are more likely than non-members to commit crime and be victimized by crime. Low self-control and some social disorganization factors are also associated with gang membership, crime perpetration, and victimization.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kathleen Fox.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Lane, Jodi S.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024198:00001


This item has the following downloads:


Full Text

PAGE 1

1 UNDERSTANDING GANG MEMBERSHIP, CRIME PERPETRATION AND VICTIMIZATION AMONG JAIL INMATES : A TEST OF SELF CONTROL AND SOCIAL DISORGANIZATION THEORIES By KATHLEEN A. FOX A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Kathleen A. Fox

PAGE 3

3 To Chris Talbot

PAGE 4

4 ACKNOWLEDGMENTS I am very gra teful for the help, guidance, and patience of many special people during the process of this research and degree. I am especially thankful to Chris Talbot for his love, encouragement, patience, and for believing in me every step of the way. I also owe ma ny thanks to my sister, Laura Fox, and my parents, Bob and Janet Fox, for their calming influence s I am indebted to my committee chair, Dr. Jodi Lane, for years of mentoring, advice, support, and encouragement. Dr. Lane has been particularly influentia l for me and I have learned valuable skills from her that I will certainly use throughout my career as a researcher and teacher. I truly look forward to working with Dr. Jodi Lane over the course of our careers as colleagues. I am thankful to my committe e members Drs. Ron Akers, Marv Krohn, and Richard Schneider for their time, constructive feedback, and help I could not have hoped for a better committee. I also owe a very special thanks to Dr. Ruth Steiner (Urban and Regional Planning) for t he employm ent opportunities over the years as a research assistant, and especially for her mentoring, friendship, and advice. I am very thankful for c ooperation from fourteen Florida county jails that participated in this research. I am especially grateful for th e county jail administrators from Alachua (Captain Jeff Cloutier and Program Manager Maggie Donnell) Broward (Captain Randy Smith and Lieutenant Angela Neely) Collier (Chief Scott Salley, Captain Joe Bastys, and Lieutenant Green) Duval (Lieutenant Geral d Milan and Sergeant Adriane Head) Escambia (Captain Chromiak, Felicia Bradley, and Officer White) Hillsborough (Captain David Parrish, Programs Manager Jan Bates, and Leslie Parker) Lee (Captain Eberhardt and Sergeant David Velez) Leon (Major Bennett and Officer Jared Lee) Miami Dade ( Lieutenant Richardson Public Affairs Manager Janelle Hall and Officers Johnson and Rivera ) Palm Beach (Colonel Michael Gauger Major Kneisley, Captain Sleeth, Lieutenant Tammy Bussey, and Deputy Wheeler), Pasco

PAGE 5

5 ( Capt ain Ed Beckman and Deputy Penny Nourse) Pinellas (Major Alexis Davis, Program Services Supervisor Ramona Schaefer, and Jill Casey) Polk (Chief Steve Lester, Captain Michael Alan, and Lieutenant Derwent Palmer) and Seminole (Captain Manley) Conducting this r esearch would have been impossible without the help of my phenomenal undergraduate research assistant, Kathy Zambrana I am indebted to Kathy for accompanying me to every jail to collect data which required very early mornings, long drives, overni ght stays, late hours and exhausting days To say that Kathy played an integral role in this project is an understatement I am very grateful for her help and her genuine interest in and excitement about this project. I fear I have accrued a debt to Ka thy I can never repay for all she has done for me. I appreciate Kath y Zambrana and David Hassan for translating the survey and informed consent form into Spanish and back to English I am also grateful for the research assistants who helped me enter surv ey data into SPSS so accurately and efficiently, including Jen Klein, Kathy Zambrana, Jean Rodriguez, Johnny Ramirez, Luis Prieto, Eliana Torres, and Shakira Henry. Without their help, I would have been forever entering data from those thousands of survey s. I would like to especially thank Jen for going well above and beyond my expectations by putting in much extra time (even over the holidays ) to ensure the surveys were entered. Both Kathy and Jen have been my angels and they have not only tremendousl y impacted this final product, but they have also kept me sane throughout the duration of this project. I thank Charlotte Nelson ( Alachua County Drug Court Program ) for facilitating the pilot study with her clients I am also grateful for the University of Florida undergraduate students enrolled in Vi ctimology during the summer 2008 term who reviewed the survey and provided very helpful feedback. I also owe many thanks for the f unding of this project to the UF Graduate Student Council and the Society fo r the Psychological Study of Social Issues.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES .........................................................................................................................10 LIST OF FIGURES .......................................................................................................................13 ABSTRACT ...................................................................................................................................14 CHAPTER 1 INTRODUCTION ..................................................................................................................15 2 LITERATURE REVIEW .......................................................................................................18 Demographic Predictors of Gang Membership, Crime Perpetration, and Victimization .......18 Gang Membership ...........................................................................................................18 Crime Perpetration and Victimization .............................................................................19 The Relationship between Gang Membership and Crime Perpetration .................................21 The Relationship between Gang Membership and Crime Victimization ...............................22 Theoretical Background ..........................................................................................................25 Self Control Theory .........................................................................................................26 Low self control and gang membership ...................................................................28 Low self control and crime perpetration ..................................................................29 Low self control and crime victimization ................................................................30 Social Disorganization Theory ........................................................................................32 Social disorganization and gang membership ..........................................................33 Social disorganization and crime perpetration .........................................................34 Social disorganization and crime victimization .......................................................36 Unique Contributions of the Current Study ............................................................................37 3 RESEARCH METHODOLOGY ...........................................................................................39 Research Hypotheses ..............................................................................................................39 Hypotheses Predicting Gang Membership ......................................................................39 Hypotheses Predicting Crime Perpetrati on .....................................................................39 Hypotheses Predicting Crime Victimization ...................................................................39 The Research Design ..............................................................................................................39 Research Setting ..............................................................................................................39 Sample .............................................................................................................................41 Procedure .........................................................................................................................41 Response Ra te .................................................................................................................44 Protection of Human Subjects .........................................................................................47 Pilot Study .......................................................................................................................47 Operational ization ...........................................................................................................49

PAGE 7

7 Gang membership .....................................................................................................49 Crime perpetration ....................................................................................................50 Crime victimization ..................................................................................................54 Self control ...............................................................................................................56 Perceptions of social disorganization .......................................................................58 Demographic variables .............................................................................................60 Measurement limitations ..........................................................................................61 Analytic Plan ..........................................................................................................................64 4 DESCRIPTIVE STATISTICS ................................................................................................78 Sample Demographics ............................................................................................................78 Gang Membership ..................................................................................................................79 Crime Perpetration ..................................................................................................................81 Crime Victimization ...............................................................................................................81 Self Control and Perceptions of Social Disorganization ........................................................82 Accuracy of Self Report Survey Data ....................................................................................83 Sample Representativeness .....................................................................................................85 5 RESU LTS PREDICTING GANG MEMBERSHIP ...............................................................92 Crime Perpetration and Gang Membership ............................................................................92 Hypothesis 1 Supported: Crime Perpetration Increases the Likelihood of Gang Membership .................................................................................................................93 Hypothesis 2 Supported: Low Self Control Increases the Likelihood of Gang Membership .................................................................................................................93 Hy pothesis 3 Partially Supported: Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Gang Membership ...................................94 Crime Victimization and Gang Membership ..........................................................................96 Hypothesis 4 Supported: Crime Victimization Increases the Likelihood of Gang Membership .................................................................................................................96 6 RESULTS PREDICTING CRIME PERPETRATION ........................................................104 Property Crime Perpetration .................................................................................................105 Hypothesis 5 Supported: Gang Membership Increases the Likelihood of Perpetrating Crime .....................................................................................................106 Hypothesis 6 Supported: Low Self Control Increases the Likelihood of Perpetrating Crime ..........................................................................................................................106 Hypothesis 7 Partially Supported: Perceptions of Socia lly Disorganized Neighborhoods Increase the Likelihood of Perpetrating Crime ................................107 Comparing Self Control Theory and Social Disorganization Theory Predicting Property Crime Perpetration ......................................................................................108 Personal Crime Perpetration .................................................................................................109 Hypothesis 5 Supported: Gang Membership Increases the Likelihood of Perpetrating Crime .....................................................................................................110

PAGE 8

8 Hypothesis 6 Supported: Low Self Control Increases the Likelihood of Perpetrating Crime ..........................................................................................................................110 Hypothesis 7 Partially Supported: Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Perpetrating Crime ................................111 Comparing Self Control Theory and Social Disorganization Theory Predicting Personal Crime Perpetration ......................................................................................113 Combined Crime Perpetration ..............................................................................................115 Hypothesis 5 Supported: Gang Membership Increases the Likelihood of Perpetrating Crime .....................................................................................................115 Hypothesis 6 Supported: Low Self Control Increases the Likelihood of Perpetrating Crime ..........................................................................................................................116 Hypothesis 7 Partially Supported: Pe rceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Perpetrating Crime ................................116 Comparing Self Control Theory and Social Disorganization Theory Predicting Combined Crime Per petration ...................................................................................118 7 RESULTS PREDICTING CRIME VICTIMIZATION .......................................................127 Property Crime Victimization ...............................................................................................128 Hypothesis 8 Supported: Gang Membership Increases the Likelihood of Being Victimized by Crime ..................................................................................................129 Hypothesis 9 Unsupported: Low Self Control Does Not Increase the L ikelihood of Being Victimized by Crime .......................................................................................129 Hypothesis 10 Unsupported: Perceptions of Socially Disorganized Neighborhoods Do Not Increase the Likelihood of Being Victimized by Crime ...............................130 Comparing Self Control Theory and Social Disorganization Theory Predicting Property Crime Victimization ....................................................................................131 Personal Crime Victimizatio n...............................................................................................132 Hypothesis 8 Supported: Gang Membership Increases the Likelihood of Being Victimized by Crime ..................................................................................................133 Hypothesis 9 Supported: Low Self Control Increases the Likelihood of Being Victimized by Crime ..................................................................................................133 Hypothesis 10 Partially Supported: Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Being Victimized by Crime ..................134 Comparing Self Control Theory and Social Disorganization Theory Predicting Personal Crime Victimization ....................................................................................135 Combined Crime Victimization ............................................................................................137 Hypothesis 8 Supported: Gang Membership Increases the Likelihood of Being Victimized by Crime ..................................................................................................138 Hypothesis 9 Supported: Low Self Control Increases the Likelihood of Being Victimized by Crime ..................................................................................................138 Hypothesis 10 Unsupported: Perceptions of Socially Disorganized Neighborhoods Do Not Increase the Likelihood of Being Victimized by Crime ...............................139 Comparing Self Control Theory and Social Disorganization Theory Predicting Combined Crime Victimization .................................................................................140

PAGE 9

9 8 CONCLUSIONS ..................................................................................................................150 Discussion .............................................................................................................................150 Theoretical Implications .......................................................................................................154 Theory Based Policy Implications .......................................................................................156 Limitations and Suggestions for Future Research ................................................................158 APPENDIX A MAP OF FLORIDA COUNTY JAILS CONTACTED .......................................................161 B SURVEY (ENGLISH) ..........................................................................................................162 C SURVEY (SPANISH) ..........................................................................................................185 D INSTITUTIONAL REVIEW BOARD APPROVAL ..........................................................208 E INFORMED CONSENT FORMS .......................................................................................209 English ..................................................................................................................................209 Spanish..................................................................................................................................210 F ORIGINAL AND MODIFIED SURVEY QUESTIONS AND SOURCES ........................211 LIST OF REFERENCES .............................................................................................................217 BIOGRAPHICAL SKETCH .......................................................................................................227

PAGE 10

10 LIST OF TABLES Table page 31 Types of Inmat es Excluded from Participating by Each Jail .............................................65 32 Jail Characteristics: Number, Response Rate, Housing Type, and Supervision Type ......66 33 Crime Perpetration Correlations ........................................................................................68 34 Crime Perpetration Factor Analysis (with Factor Loadings) .............................................69 35 Cri me Victimization Correlations ......................................................................................70 36 Crime Victimization Factor Analysis (with Factor Loadings) ..........................................71 37 Self Control Correl ations ...................................................................................................72 38 Self Control Factor Analysis (with Factor Loadings) .......................................................74 39 Physical Disorder Correlations, Factor Loadings, Reliability, and Descriptive Statistics .............................................................................................................................74 310 Social Disorder Correlations, Factor Loadings, Reliability, and Descriptive Statistics ....75 311 Collective Efficacy Correlations, Factor Loadings, Reliability, and Descriptive Statistics .............................................................................................................................75 312 Social Disorganization Correlations ..................................................................................76 313 Regression Model Variables ..............................................................................................77 41 Characteristics of the Full Sample, Non Gang Sample, and Gang Sample .......................86 42 Crime Perpetration Descriptive Statistics ..........................................................................88 43 Crime Victimization Descriptive Statistics ........................................................................89 44 Sex Comparison of County, Jail Population, and Jail Sample ...........................................90 45 Race Comparison of County, Jail Population, and Jail Sample .........................................90 46 Offense Types for the Jail Population and Sample ............................................................91 51 Logistic Regression Predicting Gang Membership with Property Crime Perpetration as an Independent Variable ................................................................................................98 52 Logistic Regression Predicting Gang Membership with Personal Crime Perpetration as an Independent Variable ................................................................................................99

PAGE 11

11 53 Logistic Regres sion Predicting Gang Membership with Combined Crime Perpetration as an Independent Variable .........................................................................100 54 Logistic Regression Predicting Gang Membership with Property Crime Victimization as an In dependent Variable .......................................................................101 55 Logistic Regression Predicting Gang Membership with Personal Crime Victimization as an Independent Variable .......................................................................102 56 Logistic Regression Predicting Gang Membership with Combined Crime Victimization as an Independent Variable .......................................................................103 61 Negative Binomial Regression Predicting Property Cri me Perpetration (Full Sample) ..121 62 Negative Binomial Regression Predicting Property Crime Perpetration (Gang versus Non Gang Samples) .........................................................................................................122 63 Negative Binomial Regression Predicting Personal Crime Perpetration (Full Sample) ..123 64 Negative Binomial Regression Predicting Personal Crime Perpetration (Gang ve rsus Non Gang Samples) .........................................................................................................124 65 Negative Binomial Regression Predicting Combined Crime Perpetration (Full Sample) ............................................................................................................................125 66 N egative Binomial Regression Predicting Combined Crime Perpetration (Gang versus Non Gang Samples) ..............................................................................................126 71 Negative Binomial Regression Predicting Property Crime Victimization (Full Sample) ............................................................................................................................144 72 Negative Binomial Regression Predicting Property Crime Victimization (Gang versus Non Gang Samples) ..............................................................................................145 73 Negati ve Binomial Regression Predicting Personal Crime Victimization (Full Sample) ............................................................................................................................146 74 Negative Binomial Regression Predicting Personal Crime Victimization (Gang versus Non Gang Samples) ..............................................................................................147 75 Negative Binomial Regression Combined Crime Victimization (Full Sample) ..............148 76 Negative Binomial Regression Predictin g Combined Crime Victimization (Gang versus Non Gang Samples) ..............................................................................................149 F 1 Crime Perpetration/Victimization Constructs Measured, Modified Survey Questions, and Original Survey Questions with Sou rces ..................................................................212

PAGE 12

12 F 2 Self Control Constructs Measured, Modified Survey Questions, and Grasmick et al.s Original Survey Items ..............................................................................................214

PAGE 13

13 LIST OF FIGURES Figure page 31 Conceptual model of the gang perpetration link using self control and perceptions of social disorganization. ........................................................................................................67 32 Conceptual model of the gang victimization link using self control and perceptions of social disorganization. ...................................................................................................67

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in P artial Fulfillment of the Requirements for the Degree of Doctor of Philosophy UNDERSTANDING GANG MEMBERSHIP, CRIME PERPETRATION, AND VICTIMIZATION AMONG JAIL INMATES: A TEST OF SELF CONTROL AND SOCIAL DISORGANIZATION THEORIES By Kathleen A. Fox May 2009 Chair: Jodi Lane Major: Criminology, Law and Society While prior research has examined factors related to gang membership and the relationship between gangs and crime perpetration, research on the relationship between gangs and victimization is limited. The present study builds on prior research and examines factors predicting gang membership, crime perpetration and crime victimization among over 2,000 jail inmates in Florida. In late 2008 and early 2009, I surveyed jail inmates about their experience s with crime, victimization, self control, and perceptions of neighborhood disorganization. Results indicate that gang members are more likely than non members to commit crime and be victimized by crime. Low self control and some social disorganization f actors are also associated with gang membership, crime perpetration, and victimization

PAGE 15

15 CHAPTER 1 INTRODUCTION Gangs were a hot political topic in the 1980s and 1990s. As the gang problem rapidly made headlines, the public became fearful of gangs (Lane 2002; Lane & Meeker, 2000; Lane & Meeker, 2003). Policymakers and practitioners rushed to implement policies and programs designed to attack the growing gang problem nation wide. Individual states took action by enacting legislation that prohibited gan g membership (National Youth Gang Center [NYGC], 2007a). The federal government quickly initiated minimum penalties for involvement in gang related crime (Violent Crime Control and Law Enforcement Act of 1994). Only a decade ago, President Clinton declar ed that fighting gangs was a top priority and announced a war on gangs in his 1997 State of the Union address (Clinton, 1997). Clearly, gangs were a major social problem. With the turn of the century a few years later, it appeared as if the panic over ga ngs was overshadowed by other social issues including Y2K (programming glitch threatening computer systems in 2000), the terrorist attacks of 2001, and the war in Iraq beginning in 2003. Interestingly, the problems with and concern about gangs appears to have recently renewed. For example, a federal bill, The Gang Abatement and Prevention Act of 2007, was recently passed by the Senate ( but not from th e House) and proposed allocating over $1 billion for suppression, intervention, and prevention programs ai med at reducing the threat of gangs ( U S Senator Dianne Feinstein 2007). Like other states, Florida is also experiencing an increase in the growth of gangs and gang related crime. A recent report by Florida Attorney General Bill McCollum and statewide prosecutor Bill Shepherd indicates that gangs are now active in all 67 counties within the state and there are more than 1,500 gangs (over 65,000 gang members) in Florida ( Office of the Attorney General of Florida, Bill McCollum 2008). A major consequen ce of the

PAGE 16

16 increasing presence of gangs is the gang related v iolence. For example, according to a recent content analysis of national newspaper s Florida is currently the second highest ranking state (behind California) for experiencing drive by shootings (Violence Policy Center, 2007). Evidently, gangs are reemerging as a major social problem. While much research on gangs indicates that gang members perpetrate crime, research to date has generally overlooked the extent to which gang members are victimized by crime Of the few studies that focus on the victimization of gang members findings appear to be mixed and the extent to which gang members experience victimization is unclear. Furthermore, theoretical explanations for the relationship between gang m embership and crime victimization remain unexplored Objectives : The current study examine s factors predictive of (1) gang membership, (2) crime perpetration, and (3) crime victimization using two theoretical explanations (self control and social disorga nization). This research contribute s to the extant gang crime, victimization and theoretical literature s by (1) examining the crime perpetration and victimization experiences of gang members in comparison with non gang members and (2) applying both macr oand micro level theoretical perspectives to understanding gang membership, crime perpetration, and crime victimization Specifically, jail inmates were queried about their involvement with gangs, experiences with a variety of crime perpetration and vic timization types, level of self control, neighborhood characteristics, and personal characteristics First, the following provide s a review of the literature on (1) the demographic predictors of gang membership, crime perpetration, and victimization, (2) the gang perpetration relationship, (3) the gang victimization relationship and (4) the theoretical background of self control and social disorganization Second, the research

PAGE 17

17 methodology is described Third, the results chapters present the analyses pr edicting (1) gang membership, (2) crime perpetration, and ( 3) crime victimization Fourth, the conclusio n section provides a summary of the results, theoretical implications, theory based policy implications, limitations, and suggestions for future resear ch Finally, the appendixes include a map of Florida jails contacted to participate in this research survey (in both English and Spanish), University of Florida Institutional Review Board approval, informed consents (in both English and Spanish), and a l ist of original and modified survey questions and sources

PAGE 18

18 CHAPTER 2 LITERATURE REVIEW Demographic Predictors of Gang Membership Crime Perpetration, and Victimization Gang Membership I n an attempt to understand the characteristics of gang members resea rchers have examined the sex, race, ethnicity, and age of gang members. Gang members are largely (although not entirely ) male, and this finding holds across time, place, and gang types ( NYGC 2007b). Early research on gangs focused on males (Thrasher, 19 27 ; Whyte, 1943). While more recent research also focuses on male gang members (Decker & Van Winkle, 1996; Vigil, 2002 ), other research examines both ma le and female gang members ( Esbensen & Deschenes, 1988 ; Gover, Jennings, & Tewksbury, forthcoming 2009) and female gang members exclusively (Miller 1998, 2002 ) Law enforcement agencies report that less than ten percent of gang members are female (NYGC 2007b). R esearch that examines sex differences among gang members suggests that while females are les s likely to be involved with gangs in comparison to males, nearly one third of gang members are women ( Esbensen & Winfree, 1998; Gover et al., forthcoming 2009). Much research has focused efforts exclusively on particular types of race/ethnic specific g angs, including Hispanics /Latinos ( Padilla, 1992; Vigil, 1988), Asians (Chin, Fagan, & Kelly, 1992; Tsunokai & Kposowa, 2002), and Blacks (Cureton, 2002). Therefore, determining racial differences with regard to gang involvement has been somewhat more lim ited. According to the National Youth Gang Survey, law enforcement agencies indicate that Whites are less likely to be gang members than Blacks or Hispanics (NYGC 2007b). According to th e 20012004 survey nearly 50% of the gang members were Hispanic a pproximately 35% of gang members were Black, and less than 10% were White. However, an examination of the Gang Resistance

PAGE 19

19 Education and Training program ( G.R.E.A.T. ) self report data suggests that the racial distribution of gang members is similar for Whi tes (25%), Blacks (31%), and Hispanics (25%) while very few Asians (5%) and other racial groups (15%) are gang members (Esbensen & Winfree, 1998).1 Overall, research on the rac ial composition of gang members has received limited attention and findings comparing race and ethnicity of gang membership suggest mixed results with some research indicating Whites are less likely to become involved with gangs (NYGC, 2007b) and other research showing that Whites are just as likely as Blacks and Hispanics to join ga ngs (Esbensen & Winfree, 1998) Prior research assessing the age of gang members also appear s to be somewhat mixed. While some research suggests that gang involvement occurs more often during adolescence (Lasley, 1997), other research shows that gang me mbers are primarily adults ( NYG C 2007b). More specifically, between 2001 and 2004 the National Youth Gang Survey found that law enforcement agencies reported that over 60% of gang members were over the age of eighteen. Crime Perpetration and Victimizat ion Given the close relationship between gang membership and crime ( Huff, 1998) and crime and victimization ( Lauritsen & Laub, 2007; Lauritsen, Sampson, & Laub, 1991; Lauritsen, Laub, & Sampson, 1992; Schreck, Stewart, & Osgood; 2008; described in further detail in the following section ) it follows, then, that the demographic variables predictive of gang membership are also predictive of crime perpetration and victimization. Similar to the demographic variables associated with gang membership, t he charact eristics that have received much attention with regard to crime and victimization include sex, race, ethnicity, and age differences. 1 See Esbensen, Osgood, Taylor, Peterson, & Freng (2001) for a comprehensive review of the G.R.E.A.T. program

PAGE 20

20 In terms of sex differences, prior research overwhelmingly indicates that men are substantially more likely than women to come in contact with the police (Durose, Smith, & Langan 2007) be arrested (U.S. Department of Justice, 2007) charged (Kyckelhahn & Cohen 2008), sentenced (Durose & Langan 2007), incarcerated in jail ( James, 2004 ) and incarcerated in prison (Sabol, Couture, & Harrison 2007) for criminal behavior While men are more likely to perpetrate most crimes, both men and women report being victimized by specific types of crime at higher rates. For example, men are more likely than women to be victims of agg ravated assault, robbery, and violent crimes in general (Craven, 1997; Rand, 2008) whereas women are more likely than men to be victimized by sexual assault, stalking, and intimate partner violence ( Craven, 1997; Gover, Kaukinen, & Fox, 2008; Nobles, Fox, Piquero, & Piquero, forthcoming 2009; Rand, 2008 ). Prior research indicates significant racial and ethnic differences with regard to crime perpetration and victimization. Using recent data from the National Crime Victimization Survey (NCVS), Rennison (2001) and Rand (2008) conclude d that Whites were less likely than Blacks or multiracial individuals to be victims of robbery, simple assault, and aggravated assault. Whites were also less likely than Blacks (and equally as likely as multiracial respondents ) to be victims of theft (Rand, 2008) Rand (2008) also reports that while Hispanics were more likely than non Hispanics to be victims of robbery, Hispanics were less likely to be victimized by other types of crime, including theft, simple assault, and ag gravated assault. Research examining the effect of age on crime and victimization largely find s that younger individuals are more likely than older individuals to be offenders and victims of crime (Klaus & Rennison, 2002) L ife course theory is particular ly relevant for discussing the effect of age on crime given its attention to the extent to which involvement with crime persists or desists over

PAGE 21

21 time (or, as individuals age) ( Blumstein, Cohen, Roth, & Visher, 1986; Moffitt, 1993; Piquero, Farrington, & Blumstein, 2003; Sampson & Laub, 1993 ) .2 Stemming from the work examining career criminals, Farre ll, Tseloni, Wiersema, and Pease ( 2001 ) argue that examining career victims or victimization over time is equally as important. The NCVS also r eveals that crime victims tend to be young given that victimization is highest for individuals age 16 19, then for individuals age 20 24, next for individuals age 25 34, and individuals age 35 and older continue to decrease their risk of victimization (Rand, 2008). T aken together, these works indicate that involvement with crime and victimization are reduced over time (as people age). The Relationship between Gang Membership and Crime Perpetration The link between gang membership and involvement in criminal behavior is well established. Gang members report more involvement than non gang members with many crimes, including auto theft, drive by shootings, homicide, an d drug sales (Huff, 1998). This finding is consistent even among research using a variety of methodol ogical techniques, including observational methods ( Hagedorn, 1988; Klein, 1971; Miller, 1966; Spergel, 1964; Thrasher, 1927; Vigil, 1988 ) official statistics ( Cohen, 1969; Maxson & Klein, 1990 ), interviews (Decker & Van Winkle, 19 96), and survey research ( Esbensen & Huizinga, 1993 ; Thornberry et al., 1993). While strong evidence of the relationship between gang membership and criminal behavio r is well established, few studies have examined explanations for this phenomenon. Thornberry et al (1993) describ e three explanations for heightened criminal activity among gang members, including the selection model, social facilitation model, and enhancement model. The selection 2 A complete discussion of life course theory is beyond the scope of the current study; however, comprehensive explanations of the theory are provided by Blumstein et al. ( 1986 ), Moffitt ( 1993 ), and Piquero et al. ( 2003 ), among others.

PAGE 22

22 model assumes individuals who exhibit ed delinquent or criminal behavior prior to gang membership are attracted to gangs given their common interest in deviance whereas the facilitation model posits that gang membership increases delinquency and criminal behavior given group dynamics (Thornberry et al., 1993; Thornberry, Krohn, Lizotte, Smit h, & Tobin, 2003a; Thornberry, Lizotte, Krohn, Smith, & Porter, 2003b). The enhancement model represents a combination of both the selection and facilitation models s uch that individuals engaged in higher levels of delinquency/criminal behavior are more l ikely to join gangs and are likely to exhibit increased levels of delinquency during gang membership due to the influence of the group (Thornberry et al., 1993, 2003a, 2003b) Using data from the Rochester Youth Development Study, the researchers found no significant differences between gang and non gang members in terms of delinquency before gang membership, yet rates of delinquency increased once gang members entered the gang. These findings indicated support for the facilitation model as an explanation for the gang offending link (Thornberry et al., 1993, 2003a, 2003b ). W hile examining delinquency using the Pittsburg Youth Study recent research by Gordon et al. (2004) found more evidence specifically for the selection model than Thornberry et al (1993 ), yet the strongest findings indicated support for the facilitation model. The Relationship between Gang Membership and Crime Victimization While the relationship between gangs and crime perpetration is evident, the association between gang membership and crime victimization is less clear Empirical evidence suggests that perpetrators of crime are also likely to be victimized by crime (Lauritsen et al., 1991, 1992; La uritsen, & Laub, 2007; Schreck et al., 2008). Therefore, there is reason to suspect that gang members not only perpetrate crime, but are also victimized by crime. Some scholars have recently pointed out that gang members are at increased risk for victimization given their risky lifestyle (e.g., drug use, drug sales and crime ), their risk of retaliation from rival gangs (e.g.,

PAGE 23

23 drive by shootings and assault), and their risk of violence from within their own gang (e.g., gang initiation and punishment for breaking rules) (Taylor, Peterson, Esbensen, & Freng, 2007). Crime victimization among g ang members has received very little research attention. Among the few studies that have examined this relationship, findings are mixed although the majority of the research suggests a relationship between victimization and gang membership Research empl oying qualitative methods often show support for the link between gang membership and victimization whereas the quantitative examination of this relationship is scarce. For example, qualitative interviews with active gang members reveal victimization from gang members own gang (e.g., initiation rituals) and from other gangs (e.g., injuries from fighting and from being shot) (Decker & Van Winkle, 1996) Although the researchers did not directly ask about family and neighborhood violence, they note that e nough background information allows us to infer its existence at high levels (Decker & Van Winkle, 1996, p. 171). Interestingly, approximately fifteen years after conducting these interviews, twenty six of the ninetynine interviewed gang members had die d (Decker, personal communication, December 18, 2007). Similarly, Joe and Chesney Lind (1995) interviewed forty eight male and female youth gang members and describe d victimization by family members, specifically child physical abuse and sexual assault. Other qualitative studies have highlighted the relationship between gang victimization and gender For example, i nterviews with female gang members suggest that they often use their gender to abstain from violence with rival gangs, but that male members o f their own gang therefore characterize them as weaker members and subject the female members to other forms of victimization (Miller, 1998; Molidor, 1996). Taylor (2008) best summarizes the state of the knowledge by explaining that researchers do not yet fully understand the complex relationship between g ang s and victimization among men and women

PAGE 24

24 Of the handful of studies that have recently examined the relationship between gang membership and crime victimization using quantitative methods, the majority have shown support for the gang victimization link For example, Peterson, Taylor, and Esbensen (2004) were among the first to assess crime victimization among gang members using data from the G.R.E.A.T. p rogram Elementary students were asked to reflect on the past six months and respond to three items measuring violent victimization: (1) assault without a weapon, (2) assault with a weapon, and (3) robbery. Findings suggest ed that gang members were more likely to be victimized than non gang members before, during, and after gang membership. Taylor et al. (2007) examined this relationship using a sample of nearly 6,000 eighth grade students across eleven locations. Employing the same measures of violent victimization (using a reference of 12 months inst ead of six months ), the researchers conclude d that gang members were more likely (and more frequently) victims of violence than non gang members. Similarly, using survey research from over 4,500 high school students Gover et al. (forthcoming 2009) found that gang members were more likely than non gang members to be victimized by dating violence, sexual assault, and violent victimization (injured during a physical assault). Gibson, Miller, Swatt, Jennings, and Gover ( forthcoming 2009 ) argue that these studies are limited due to the inability to determine a causal relationship between gang membership and violent victimization. Using the G.R.E.A.T. data, and the same three measures of violent victimization over a 12 month period, the researchers employ a unique statistical procedure, propensity score matching (PSM), to examine the causal relationship between gang membership and victimization. Contrary to prior research findings, PSM analysis reveals that gang members are not violently victimized more tha n non gang members.

PAGE 25

25 Overall, the scant research on the relationship between gang membership and crime victimization appears to be mixed ; however, the majority of the research shows support for the gang victimization link While the handful of studies tha t have quantitatively focused on the victimization of gang members offer original and important scientific contributions, they also have several limitations. First, these studies all sample d juveniles and therefore, the extent to whi ch adult gang members experiencing crime victimization is unknown Second, these studies used few measures of victimization (ass ault, assault with a weapon, robbery sexual assault, and dating violence ). Prior research uses measure s of a select few types of violent victimiza tion and notably omits other forms of vio lent victimization often associated with gangs (e.g., stabbing, carjacking, witness intimidation, home invasion, shooting, drive by shooting) as well as types of property c rime victimization (e.g., theft, vandalism) Third, measures employed thus far have limited the timeframe for experiencing victimization to the past six or twelve months. This limitation prevents a thorough examination of the extent to which gang members are victimized over the life course. Four th, the theoretical understanding of gang victimization is underdeveloped. The current research address es the limitations of prior research by being among the first to quantitatively assess the theoretical explanations of the gang perpetration and the gan g victimization link. Theoretical Background This study is among the first to examine gang membership, crime perpetration and gang membership using two theoretical perspectives (self control and social disorganization) T hese t wo diverse theoretical ap proaches are important to examine given that both are well established theories that have garnered much empirical support with regard to their abilities to predict crime perpetration (Pratt & Cullen, 2000, 2005) and even crime victimization (Sampson & Grov es, 1989; Schreck, 1999; Stewart, Elifson & Sterk 2004). A wide variety of both macro level and

PAGE 26

26 micro level theories have generated support for predicting crime perpetration and victimization (Vold, Bernard & Snipes, 2001), which suggests that crime an d victimization are likely function s of both individual level and societal level factors. However, a substantial amount of theory testing examines the explanatory power of a single theoretical perspective separately without controlling for or examining different theories It is important to test different theories simultaneously for theory advancement as well as practical policy implications. According to the theoretical concepts of both theories, a relationship between social disorganization and self co ntrol is not expected ( Gottfredson & Hirschi, 1990; Shaw & McKay, 1969 ) Rather, the theories are primarily examined in separate models and are only examined in simultaneous models to determine the explanatory power of one theory over another. Self Contro l Theory The concept of self control is a micro level theoretical perspective derived from sociologists Gottfredson and Hirschis (1990) general theory of crime. This individual based explanation for delinquency assumes that presented with an opportunity individuals with low er self control are more likely than those with higher self control to commit crime and ana logous behavior (i.e., smoking and drinking) Gottfredson and Hirschi (1990) argue that self control is established early in life (by age 8 to 10) and that this trait remains stable throughout the life course.3 Gottfedson and Hirschi (1990) identified six c haracteristics that comprise low self control which include impulsivity, insensitivity, risk seeking, short sightedness, non verbal tenden cies and a preference for physical activities Individuals who are impulsive are more likely to engage 3 Turner and Piqueros (2002) assessment of the stability of self control using longitud inal data revealed that, contrary to the theory, levels of self control varied somewhat over time. Consistent with the theory, offenders exhibited significantly lower levels of self control than non offenders after age eight and men demonstrated lower lev els of self control than women.

PAGE 27

27 in crime given their tendencies to act quickly without careful consideration of the potential risks associated with criminal behavior. Insensitive peo ple are likely to act criminally due to their own self interest without concerning themselves with the possible ramifications others may experience because of their actions. Individuals who actively seek risky behavior are more likely to become involved w ith crime given the risky nature of crime itself. For example, illegal activities are often dangerous and involvement with crime is associated with a certain risk of being caught. Riskseekers may enjoy the thrill of committing crime as well as the thrill of potentially being caught. People who are short sighted, who do not regularly think about and plan for the future, are more likely to commit crime given that they are more likely to consider the short term benefits of crime rather than the long term c onsequences. Individuals who are non verbal may not be able to appropriately articulate their desires and may express themselves in other, more deviant or criminal ways. Similarly, people who prefer physical activities over mental activities may be more likely to engage in physical remedies to opposition (such as involvement in assaults). Gottfredson and Hirschis (1990) self control theory has been extens ively examined and has received much empirical support (Pratt & Cullen, 2000) as well as critiques (Akers, 1991). Hirschi (2004, p. 543 ) recently revised the theory by merging concepts from social control theory (Hirschi, 1969 ) and self control theory (Gottfredson & Hirschi, 1990), and redefines self control theory as the tendency to consider the full range of potential costs of a particular act.4 While Hirschis (2004) revised version of self control was developed recently and some empirical support for the theory is beginning to develop ( Piquero & Bouffard 2007 ), future attention to this 4 Hirschi (2004) developed a scale to empirically test the redefined concept of self control, including survey items such as: Do you like or dislike school? Do you care what teachers think of you? and Does your mother k now where you are when you are away from home?

PAGE 28

28 reconceptu alization is necessary. Given that very little research has focused on Hirschis (2004) redefined theory, the following focuses on the original concept of self control theory developed by Gottfredson and Hirschi (1990). The linkages between self control theory and (1) gang membership, (2) crime perpetration, and (3) crime victimization are discussed next. Low self control and gang membership With their theory of self control, Gottfredson and Hirschi (1990) sought to not only explain criminal behavior, but also delinquency and analogous behavior. B ehavior analogous to crime can include a multitude of activities, including drinking, smoking, truancy, cheating, gambling, and even accidents (Hirschi, 2004). Given the well established relationship between cri me and gang membership (Huff, 1998), Peterson Lynskey, W infree, Esbensen, and Clason (2000) extended Gottfredson and Hirschis (1990) concept of analogous behavior to include gang membership. Essentially, Peterson Lynskey et al. (2000) argued that self co ntrol theory may explain why some individuals become involved with gangs while others do not. Using data from evaluations of the G.R.E.A.T. program, the researchers found support for self control theorys ability to pr edict gang involvement. E ighth grade elementary students with lower self control were more likely to become involved with gangs than those with higher self control (Peterson Lynskey et al., 2000 ) This finding suggests important implications for theory, policy, and future research. That se lf control theory is successful in predicting gang membership is an indication that the theory truly may be a general theory, as Gottfredson and Hirschi (1990) arge, that can explain a variety of crime, deviance, and analogous behavior (including gang invo lvement) In the context of self control theory, policies designed to reduce gang membership may be most effective when targeted to parenting (given that Gottfredson & Hirschi suggest that early parenting practices shapes self control). Given that Peters on Lynskey et al.s (2000) work is the

PAGE 29

29 only known published test of self control and gang membership, additional research is needed to examine this relationship, especially while controlling for criminal behavior. While other research has not confirmed P eterson Lynskey et al.s (2000) work by directly examining the link between self control and gang membership some research indirectly supports th is relationship in at least two important ways First, in the context of Gottfredson and Hirschi (1990) s exp la n ation that self control is the product of poor parenting and little supervision prior research has established a link between poor parenting and gang membership. When compared with non gang members, more gang members come from single parent families (Esbensen & Winfree, 1998) with little parental warmth (Walker Barnes & Mason, 2001). Moreover, g ang members report that the gang is a surrogate family ( Joe & Chesney Lind, 1995) Second some of the characteristics of self control ( impulsivity, risk see king, short sightedness, etc. ) identified by Got tfredson and Hirschi (1990) have been linked to the characteristic s of gang members. For example, Peterson Lynskey et al. (2000) argue that g ang membership is risky given the dangers associated with initiati on practices within the gang and interaction with riv al gangs (see also Taylor 2008). Peterson Lynskey et al. (2000) also describe j oining a gang as a f unction of impulsivity and short sightedness, given that individuals may join spontaneously or without carefully considering the costs of gang membership. As an extension of these important findings, I argue that gang membership represents a commitment to crime on a greater level than non gang membership and, therefore, lower levels of self control may be a precursor to joining a gang. Low self control and crime perpetration Gottfredson and Hirschi (1990) presented their general theory of crime a s an explanation of crime perpetration and research has examined this relationship in a variety of ways. In one of the first and most well known empirical tests of self control theory, Grasmick, Tittle, Bursick,

PAGE 30

30 and Arneklev (1993) developed a survey to measure each of the six self control components outlined by Gottfredson and Hirschi (1990). The twenty four item scale created by Grasmick et al. (1993) continues to be one of the most commo n ways to measure self control .5 While this scale examines attitudinal measures of self control, others found support for the theory using behavioral measures as well as attitud inal measures (Turner & Piquero, 2002). Regardless of the use of attitudinal versus behavioral measures, the relationship between low self control and crime perpetration is well established (Pratt & Cullen, 2000; Tittle, Ward, & Grasmick, 2003 ) Prior r esearch testing self control theory has successfully found links between low self control and many types of crime, including property and personal crime (Longshore, 1998) and, more specifically intimate partner violence (Sellers, 1999). Low self control is also an important predictor of analogous behavior such as drinking alcohol and skipping school (Gibbs & Giever, 1995; Gibson, Schreck, & Miller, 2004). In line with the theory and prior research, the current study hypothesizes a relationship between crime perpetration and low self control. Low self control and crime victimization G iven the similarities between crime perpetrators and victims ( Lauritsen et al., 1991, 1992; Lauritsen & Laub, 2007; Schreck et al., 2008), it is plausible that crime vic tims may also exhibit low levels of self control. The six characteristics of low self control outlined by Gottfredson and Hirschi (1990) that explain criminal and analogous behavior may also explain victimization (Schreck, 1999; Stewart et al., 2004) As Schreck (1999 ) and Stewart et al (2004) point out, i ndividuals who are impulsive, shortsighted, physical, and risk seeking may be more likely to engage in fun, adventurous, or dangerous behavior without considering the consequences. This behavior may ex pose these individuals to potential offenders and may render them at risk for 5 While the work of Grasmick et al. (1993) has generated a multitude of support, it has also been subjected to criticism (Arneklev, Grasmick, & Bursik, 1999; Marcus, 2004; Piquero & Bouffard 2007).

PAGE 31

31 victimization. Similarly, individuals who exhibit non verbal tendencies and who are easily angered may react to stress and unpleasant situations physically rather than through s ocially acceptable verbal negotiations This behavior may foster confrontations with others and may in turn, result in victimization. Additionally, insensitive or self centered individuals may provoke others, escalate negative situations, and may find i t difficult to maintain close personal relationships, which Stewart et al. ( 2004 ) argue affects victimization risk given the lack of guardians. Certainly abrasive behavior could put individuals at increased risk of victimization. Examining the extent to which victims are impulsive, insensitive, risk seeking short sighted, physical, and non verbal is important for determining whether low self control is a risk factor associated with victimization. Although Gottfredson and Hirschi (1990) do not specifica lly mention crime victimization, I argue that crime victimization may be considered as a form of analogous behavior Based on prior research linking crime perpetration with victimization ( Lauritsen et al., 1991, 1992; Lauritsen, & Laub, 2007; Schreck et a l., 2008 ), it follows that victimization is related to or analogous to crime perpetration. Considering victimization as analogous behavior also appears justified given that Gottfredson and Hirschi (1990) include accidents as an analogous behavior to crime (Akers, 1991) It is not only plausible, but likely, that crime victimization may be accidental. The link between self control and victimization has recently received limited empirical attention. Using a large sample of college students, Schreck (1999) was among t he first to examine the relationship between self control and victimization among offenders F indings revealed that crime victims exhibited significantly lower levels of self control than non victims (for both personal and property crime vic timization) Similarly, Stewart et al. (2004) later confirmed these findings with a sample of female drug offenders and suggested that low self -

PAGE 32

32 control is a risk factor for criminal victimization. Both studies controlled for the influence of risky lifest yle behaviors. Given that research has recently found support for Gottfredson and Hirschis (1990) general theory of crime to explain crime victimization, and in light of the logical connection between offending, victimization, and low self control, the c urrent study propose s to examine this relationship among offenders generally and among ga ng members specifically This study is grounded in the recent contributions by Schreck (1999) and Stewart et al. (2004) and extends their work by assessing the self c ontrol of a unique sample (jail inmates) in relationship to s pecific types of crime victimization While Gottfredson and Hirschi (1990) assume all offenders exhibit low self control, prior research examining self control among offenders found variability in self control levels (Longshore, Turner, & Stein 1996). Furthermore, jail inmates represent a heterogeneous sample of offenders exhibiting wide ranges of offending intensity, duration, and severity. For example, individuals may be incarcerated in jail for a variety of reasons including f ailure to pay child support or traffic citations, illegally residing in the United States property crimes, violent crimes and drug related crimes Given the diversity of offender types in jail, this setting is ideal for examining variation in offenders self control levels. Social Disorganization Theory Social disorganization is a macro level theoretical perspective derived by sociologists at the Chicago School (Vold et al., 2001) This ecologically based explanatio n for delinquency assumes that rapid urbanization leads to a deterioration of community controls, resulting in disorganization and the replacement of traditional values with criminal values (Shaw & McKay, 1969). Then, the area not individuals, breeds crime regardless of who moves in and out of these neighborhoods. Shaw and McKay (1969 ) identified three neighborhoodlevel characteristics that contribut e to social disorganization, including low socioeconomic status (SES), ethnic

PAGE 33

33 heterogeneity, and high res idential mobility. In other words, these scholars argued that social disorganization is likely to occur within neighborhoods that are financially disadvantaged and racially mixed wherein residents frequently move in and out of the neighborhood. Sampson a nd Groves (1989) built upon the early work of Shaw and McKay and identified other measures of social disorganization, in addition to SES, ethnic heterogeneity, and residential mobility, including: community control over juveniles, local friendships, partic ipation in local organizations, family disruption, and urbanization. More recently, Sampson Raudenbush, and Earls (1997 ) suggested that social disorganization can be measured by assessing collective efficacy. Collective efficacy is informal social contr ol and is defined as a combination of social cohesion and mutual support as well as shared expectations for social control among neighbors. In other words, neighbors who work together informally control crime, delinquency, and disorder. Evidence that col lective efficacy is associated with less crime is strongly supported (see Pratt and Cullens (2005) meta analysis). The following briefly describes research that has established a link between social disorganization theory and (1) gang membership, (2) cri me perpetration, and (3) crime victimization. Social disorganization and gang membership While prior research has not yet empirically examined the relationship between social disorganization and gang membership using comprehensive measurement of the the ory, some qualitative and quantitative work suggests that social disorganized neighborhoods are a risk factor for gang membership. Among qualitative works, interviews with gang members point to specific factors associated with social disorganization, such as lack of employment opportunities, low socio economic status, high mobility, and racially heterogeneous neighborhoods (Decker & Van Winkle, 1996; Joe & Chesney Lind, 1995; Moore, 1978, 1991; Thrasher, 1927 ). These qualitative works suggest that a relat ionship exists between gang membership and socially

PAGE 34

34 disorganized neighborhoods. Although qualitative studies suggest gang membership is affected by social disorganization, research employing quantitative analyses are mixed. Some quantitative work indicat es that social disorganization and gang membership are related ( Bowker & Klein, 1983; Curry & Spergel, 1992 ; Short, 1990 ) whereas others do not ( Bjerregaard & Smith, 1993; Fagan, 1990). Given that a substantial amount of qualitative and quantitative resea rch suggests a link between social disorganization and gang membership, the current study examines this link using comprehensive measures of the theory. Social disorganization and crime perpetration Social disorganization theory is a theory designed to e xplain crime, and prior research has devoted much attention to the theorys ability to predict crime perpetration. One of the challenges to interpreting the ability for social disorganization theory to explain crime is the various ways in which the theory has been measured. As described earlier, social disorganization theory is comprised of many broad factors, including physical disorder, social disorder, collective efficacy, neighborhood poverty, neighborhood unemployment, racial heterogeneity, and resid ential mobility (Sampson & Groves, 1989; Sampson & Rauden bush, 2001; Shaw & McKay, 1969). Interpretation of prior research examining social disorganization must be put in the context of studies limitations given that many tests of social disorganization examine some factors while omitting others. While a comprehensive review of the partial tests of social disorganization theory is beyond the scope of this research (see Pratt and Cullen, 2005), research employing the most comprehensive measurement s of soc ial disorganization will be briefly reviewed. Using the British Crime Survey, Sampson and Groves (1989) examined the effect of a number of social disorganization factors (SES, ethnic heterogeneity, residential stability, family disruption, urbanization, and a number of factors related to collective efficacy) on self reported

PAGE 35

35 property and personal crime perpetration Their analysis revealed mixed support for social disorganization theorys ability to predict crime given that few of the variables were sign ificantly related to property crime (higher levels of ethnic heterogeneity and unsupervised peer groups) or personal crime perpetration (higher levels of family disruption and unsupervised peer groups and lower levels of local friendship networks) (Sampson & Groves, 1989) While Sampson and Groves (1989) examined many aspects of social disorganization, perhaps the most robust test of social disorganization theory to date was conducted by Sampson and Raudenbush ( 2001) as part of the Project on Human Develo pment in Chicago Neighborhoods Sampson and Raudenbush ( 2001) comprehensively measured social disorganization as a combination of (1) physical disorder, (2) social disorder, (3) collective efficacy, and (4) neighborhood characteristics. Physical disorder was measured by the presence of garbage on the streets, graffiti, abandoned cars, and needles and syringes used for drugs. Social disorder was measured by the presence of loitering, public alcohol consumption and intoxication, drug sales, and gang activi ties. Collective efficacy was measured by asking residents if their neighbors would take action if they saw unattended children misbehaving (shared expectations for social control) and whether their neighbors were willing to help each other (social cohesio n and mutual support). Neighborhood characteristics were assessed with measures of poverty, immigration, and residential mobility. Findings revealed that neighborhood characteristics, especially poverty, and collective efficacy were predictive of crime ( not levels of physical or social disorder). In line with the theory and prior research, the current study hypothesizes a positive relationship between social disorganization and crime perpetration (Pratt & C ullen, 2005; Sampson et al., 1997 ; Sampson & Gro ves, 1989; Sampson & Raudenbush, 2001; Shaw & McKay, 1969).

PAGE 36

36 Social disorganization and crime victimization Although the connection between socially disorganized neighborhoods and crime perpetration is well established, the relationship between social dis organization and crime victimization is less understood. While social disorganization theory has traditionally focused on explaining offending, the theory may also successfully explain crime victimization. It can be argued that socially disorganized area s that exhibit physical disorder, social disorder, a lack of collective efficacy among neighbors, and high levels of poverty and residential mobility are not only indicative of areas with high crime but that the crime experienced in these areas produce cri me victims Given that crime results in crime victimization (Karmen, 2009), it follows that social disorganization theorys ability to explain c rime may also explain victimization. The relationship between social disorganization and crime victimization ha s received very limited attention from researchers. Sampson and Groves (1989) were among the first to examine the extent to which crime victims reported social disorganization within their neighborhoods. Interestingly, their analysis revealed that more o f the social disorganization factors were related to crime victimization than crime perpetration. In terms of personal crime victimization, family disruption, urbanization, and the measures related to collective efficacy (local friendship networks, unsupe rvised peer groups, and organizational participation) were significant. However, only one of the three measures of social disorganization outlined by Shaw and McKay (1969) was significantly related to robbery victimization (ethnic heterogeneity). In term s of property crime victimization, the majority of the social disorganization factors (including the original measures by Shaw and McKay) were significant predictors. However, it is important to note that one of the variables ( SES ) operated differently fo r victims of burglary (indicating a positive relationship) and victims of auto theft and vandalism (suggesting a negative relationship). In other words, victims of burglary were more likely to have higher SES whereas

PAGE 37

37 victims of auto theft and vandalism we re likely to have lower SES. Sampson and Groves (1989) suggest that the homes of individuals with higher SES may attract more attention from burglars. Alternatively, individuals with lower SES may be more inclined to leave their vehicles and other valuab les unattended or unsecured (i.e., in locked garages or homes), which may explain the link between lower SES and more auto theft and vandalism victimization. Like Sampson and Groves (1989), Pratt and Cullen (2005) also find that some social disorganization variables exhibit both positive and negative effects. Meta analyses revealed that unemployment in particular generated both positive and inverse relationships with crime. While higher rates of unemployment as a predictor of crime perpetration is consist ent with social disorganization theory, lower rates of unemployment associated with crime may simply suggest the presence of attractive targets (like Sampson and Groves, 1989 burglary finding) Despite the ways in which these variables operate (positively or negatively), it is important to acknowledge that social disorganization is an important predictor of victimization and crime. In light of the important work by Sampson and Groves (1989) on the relationship between social disorganization and crime vict imization, the current study examines this connection using additional social disorganization and victimization measures. Unique Contributions of the Current Study This study contribute s to the understanding of gang members hip, crime perpetration, and cri me v ictimization by expanding on prior literature in at least three important ways. First, this will be among the first quantitative studies to examine gang s, crime, and victimization a mong a sample of adult offenders (not juveniles, as prior research has done). Given that gang membership and gang related victimization is not limited to adolescence, it is important to examine the experiences of adults (NYGC, 2007b) Second, this study is the first to measure a variety of types of crime perpetration and victimization Prior research examining the gang -

PAGE 38

38 victimization link has examined only a handful of victimization types (e.g., assault, assault with a weapon, robbery dating violence, and sexual assault ). The current study assess es the extent to which off enders (and, specifically, gang members) have experienced other forms of violent victimization (e.g., stabbing, carjacking, witness intimidation, home invasion, shooting, and drive by shooting) as well as property crimes (e.g., theft, vandalism). Third, t his study will be among the first to investigate the theoretical predictors of gang membership, offending, and victimization. Furthermore, the current study examine d two diverse theories: a macro level theory (social disorganization) and a micro level the ory (self control). The following Chapter details the methodology of the current study The results Chapters are presented next, which report the predictive factors for gang membership, crime perpetration, and crime victimization. Finally, a discussion of the findings is offered in the context prior literature and theoretical implications and policy implications are discussed while acknowledging the limitations of the study.

PAGE 39

39 CHAPTER 3 RESEARCH METHODOLOGY Research Hypotheses Given prior research on gan g membership, crime, victimization, social disorganization and self control, several hypotheses are empirically examined : Hypotheses Predicting Gang Membership 1. Crime perpetration increases th e likelihood of gang membership 2. Low self control increases the li kelihood of gang mem bership 3. Perceptions of s ocially disorganized neighborhoods increase th e likelihood of gang membership 4. Crime victimization increases th e likelihood of gang membership Hypotheses Predicting Crime Perpetration 5. Gang membership increases th e likelihood of perpetrating crime 6. Low self control increases the l ikelihood of perpetrating crime 7. Perceptions of s ocially disorganized neighborhoods increase the l ikelihood of perpetrating crime Hypotheses Predicting Crime Victimization 8. Gang membership i ncreases the likelihood of being victimized by crime 9. Low self control increases the likeliho od of being victimized by crime 10. Perceptions of s ocially disorganized neighborhoods increase the likelihood of being victimized by crime Overall, I expect that gang members hip will be significantly related to crime perpetration, victimization, low self control, and perceptions of soci ally disorganized neighborhoods. To clarify, the dependent and independent variables change (described below) and several hypotheses t est inter related relationships (i.e., hypotheses 1 and 5 and hypotheses 4 and 8). The Research Design Research Setting As mentioned earlier, Florida is experiencing an increase in the growth of gangs and gang related crime. Given the recent and increasi ng gang problem within the state of Florida,

PAGE 40

40 researching gang members throughout the state will provide important information about the reasons people join gangs and potential causes and consequences of gang membership (e.g., crime victimization) Given t hat the state of Florida is large (e.g., the population exceeds 17 million and the land area is almost 54 thousand square miles) (U.S. Census Bureau, 2008), it is especially important to collect data within counties located throughout the state in order to more comprehensively examine the theoretical relationship between crime victimization and gang membership. The site s for this research include jails in Florida counties that have the largest county populations .1 I decided that sampling jail inmates wou ld allow for a wider array of ages and a variety of offenders while avoiding the bulk of the temporary prison gang culture. Jails in the twenty largest counties were contacted, including: Alachua, Brevard, Broward, Collier, Duval, Escambia, Hillsborough, Lee, Leon, Manatee, Marion, Miami Dade, Orange, Palm Beach, Pasco, Pinellas, Polk, Sarasota, Seminole, and Volusia These counties are geographically dispersed across Florida (see map in Appendix A ). Permission to conduct this research was obtained from fourteen of the twenty jails which represented 70% of the target population ( Alachua, Broward, Collier, Duval, Escambia, Hillsborough, Lee, Leo n, Miami Dade Palm Beach, Pasco, Pinellas, Polk, and Seminole ) T hree facilities (Brevard, Sarasota, and Volu sia) declined to participate given that they report ed being unable to provide correctional officers as escorts and the remaining three jails ( Manatee, Marion, and Orange ) were not included in the sample given that administrators were unresponsive to reques ts to permit the research 1 Given th at gang members (NYGC, 2007b) and crime (Duhart, 2000) are more prevalent in urban areas, the sample was derived from the largest counties within the state of Florida in an attempt to obtain a higher rate of gang members, offending, and victimization.

PAGE 41

41 Sample Inmates in county jails across the state of Florida were asked to complete a survey regarding their experiences with gangs, crime, and victimization. Given that the current study aim ed to focus on gang membership, obta ining information from jail inmates was ideal since many gang members are incarcerated temporarily in jail facilitie s. Furthermore, this study aimed to compare gang members with non gang members. For analysis purposes, described below, it is important th at the comparison group (offenders) closely matched the population of interest (gang members). Jail is a convenient location where individuals who have engaged in criminal behavior are congregated. Prison inmates were deemed an inappropriate sample for this project given the tendency for inmates to join prison gangs temporarily (during their incarceration). For this reason, and others, many scholars do not include prison gangs in their definition of gang s ( Klein & Maxson, 2006; NYGC 2007b) While some prison inmates were gang members before incarceration and some inmates who joined prison gangs remain gang members upon exiting prison, the current study aimed to avoid confound ing these results by mixing street gang members with prison gang members. Juveniles in detention centers were excluded as a possible sa mple for the current research given that s ampling juvenile inmates exclusively (as prior research examining the gang victimization link has done) would unnecessarily restrict the age of respondents below age 18.2 Procedure Jail inmates were asked to report their involvement with (both perpetration and victimization) specific types of crime (theft, vandalism, assault, drive by shootings, carjacking, 2 Fl eisher & Decker (unpublished manuscript) indicate that members of street gangs are often teenagers whereas prison gangs are typically comprised of individuals in their mid twenties. Given that research on gang members within jails is limited, the ages of gang members in jail likely range between the street and prison gang age estimates. Therefore, it is important to include older individuals in the sample.

PAGE 42

42 etc.). The survey also included questions about inmates demographic characteristics, gang membership, self control, perceptions of neighborhood social and structural characteristics, fear of crime, and control balance. Each jail was visited between one and six days to administer surveys to jail inmate s. A research assistant who spoke Spanish accompanied me to each jail and assisted with data collection procedures (e.g., administering surveys, ensuring the return of pencils, collecting completed surveys, recording important information). Participants were given the choice between completing the survey on their own or having the survey questions and response options read aloud to them by the researcher(s). The survey was available in both English and Spanish (see A ppendix B and C ).3 I read aloud the English version to participants, and the research assistant (who is bilingual in both Spanish and English) read aloud the Spanish version. In an effort to avoid identifying some inmates as functionally illiterate (which may have resulted in negative stigm atization from correctional officers and/or inmates), the researchers did not ask participants or correctional officers about specific inmates inability to read. Instead, the researchers informed all participants that the survey questions would be read a loud while they respond to each question individually using a paper and pencil format. Participants were, however, given the option to complete the survey at their own pace. Given that most participants completed the survey on their own while others fo llowed along as the survey was read aloud, participants completed the survey at different times. Pencils were provided for the inmates to use to complete the survey and these were collected along with the completed surveys. In an effort to minimize the t ime and resources of the correctional facilities, 3 The survey was back translated (Bernard, 2000) from English to Spanish and back to English by two undergraduate students fluent in both languages. The survey was administered in both Spanish and English, depending on the preference of the respondent. Given the diverse population of many jails in large cities, including Spanish speaking individuals in the study was important for generating a more comprehensive and representative sample. Nine percent of respondents (n = 212) chose to complete the Spanish version.

PAGE 43

43 and to ensure proper administration of the survey, the survey was administered and collected on the same day. The jail staff and administrators did not handle the surveys. The survey was administered t o the inmates in a variety of settings, depending on the preference and procedures of individual jail facilities. In all but one facility, the research team personally explained the study to groups of inmates while they were housed in their cells and/or pods.4 Surveys were administered in a variety of jail settings (e.g., in the recreation area, dayroom, individual cells, multipurpose rooms). Inmates interested in participating typically completed the survey in the dayroom or were relocated by correction al officers into a multipurpose room. Regardless of the location, attempts were made to space the participants adequately to make sure they did not look at others survey answers. Ultimately, the research team remained as flexible as possible to accommod ate the daily routine of the inmates and staff. If arrangements to relocate inmates to locations within the jail were made, this was coordinated with jail staff prior to beginning the research (e.g., participants were moved from their cells/dayroom to ano ther room). These arrangements also included relocating inmates who wished to discontinue participation or who finished early so that they were able to leave the room (e.g., and be escorted by jail staff back to their pods). In some cases, the survey was read aloud to individuals separately, in the event that they requested individual assistance.5 Protection procedures designed to safeguard the researchers from inmates varied with each jails standard procedures. For example, one jail provided the res earchers with an electronic radio Some jails housed the researchers and participants in a room with windows near jail staff and other jails required officers to be in the room during the survey administration ; 4 Pasco jail administrators did not allow me to enter pods due to safety concerns. Instea d, I provided the administrators with a memo briefly explaining the study which officers read aloud to the inmates in order to obtain volunteers. 5 Nearly 5% of respondents (n = 110) preferred that the entire survey was read aloud as they followed along.

PAGE 44

44 however, jail staff were unable to see the i nmates survey responses Protection procedures were arranged with each jail prior to and upon entry (before data collection). Response Rate The size of the sample depend ed upon the willingness of jail inmates to voluntarily complete the survey without co mpensation. With the exception of t wo sites (Esc ambia and Pasco) the researchers personally visited each eligible pod and invit ed all inmates present to participate in the survey .6 All eligible volunteers were allowed to participate Each jail determin ed certain types of inmates were ineligible to participate in the study (see Table 3 1 for a list of the types of inmates excluded at each jail) Inmates incarcerated in units specifically for severe psychiatric disorders or communicable diseases were exc luded from participating at the request of the researchers. All jails excluded inmates housed in solitary disciplinary confinement and s ome jails excluded federal inmates and those in high security dorms. While e ach j ail housed between approximately 1,000 and 4,000 inmates on any given day the majority of inmates refuse d participation. A lthough the response rate for each pod varied (between 0 and 93%) the average rate of participation was 25% .7 Table 3 2 allows for a comparison of the county populatio n, the jail population, and the sample size. The table also presents the average response rate for each of the jails in addition to a description of the housing and supervision type. When interpreting response rates derived from research based on incarcer ated populations, an understanding of the setting is necessary. For example, the number of inmates housed within 6 J ail administrators from Escambia and Pasco agreed to allow a correctional officer to act as an escort for the duration of one day only. Escambia administrators allowed one dorm to participate on each of four floors (consisting of a North, South, East, a nd West dorm per floor), which resulted in three dorms of general population male inmates and one dorm of high security male inmates. Pasco administrators allowed three dorms to participate in the research and selected two general population male pods in addition to a high risk female pod 7 While a 25% response rate may be considered relatively low, other research targeting incarcerated individuals report similar rates (Struckman Johnson, Struckman Johnson, Rucker, Bumby, & Donaldson, 1996).

PAGE 45

45 each pod (the population) was often very large (between 60 and 100 inmates). While the researchers were able to obtain a count of the exact n umber of inmates present during the researchers visit, it was clear that this number almost never accurately represented the number of inmates who heard the researchers invitation to participate in the study. While the escort or the officer overseeing ea ch pod quieted the inmates and called their attention to the researchers, some inmates remained in the cells, beds, recreation room, or showers and other inmates choose not to remove their headphones while the announcement about the survey was being made. Given the large size of the dorms where many areas and inmates were hidden from the view of the researchers, counting the number of people believed to hear the researchers announcement was impossible. Especially given that some inmates participated in t he survey who appeared to be asleep or who were in their cells at the time of the announcement. Therefore, the response rate for each pod was calculated based on the total number of inmates present in the pod, which represented a worst case scenario of the most conservative response rate estimate. It is also important to note that the procedures of the current study easily allowed inmates to decline participation by asking those interested in participating to congregate in a specific area to then be mo ved to a location suitable to administer the survey. Thus, the path of least resistance for inmates was clearly to decline participation. Other research on incarcerated individuals has adopted alternative procedures by selecting particular inmates to p articipate relocating groups of selected inmates to a designated survey room, and then asking for their cooperation and participation (see Peterson, Braiker, and Polich, 1981). While this approach resulted in a higher response rate (Peterson et al. recei ved a 57% response rate), this approach was not used for the current study given the possibility for inmates to feel coerced into participating. I nmates who have already been relocated to a specific area may be more likely to

PAGE 46

46 participate than those who ar e invited to participate and then relocated for a variety of reasons. First, the path of least resistance is now to remain in the room and comply with the researchers request. Second, inmates may be coerced into participating because the majority (or all) of the other inmates remain in the room. Third, inmates understand that a correctional officer will be required to escort them out of the room and inmates may participate in the research in order to avoid additional contact with the officer (especial ly if the officer would not otherwise be called to the room or if the inmate did not want the officer to know s/he was not complying with the request to participate). Therefore, the current study adopted a procedure that yielded a lower response rate whil e avoiding additional potential for coercion. Although no data were systematically collected on this issue, the researchers made observations in an effort to understand the lower response rate I determined that t he response rate within each pod was affec ted by several identifiable factors, including time of day, officer rapport with inmates, and interference with daily activities For example, fewer inmates were awake in the early mornings and, therefore, conducting the survey in the mornings generated c onsistently lower response rates than administering the survey after lunch time. Furthermore, it became apparent that in some cases the escort (officer or program staff) may have a ffected the response rate based on his/her degree of rapport with the inmat es (e.g., officers /staff with less rapport with inmates generated lower response rates). Moreover, conducting the survey in pods during scheduled activities resulted in lower response rates. Inmates were unlikely to volunteer to participate if they antic ipated leaving for visitation, to meet with their attorney, for an infirmary appointment, for physical recreation etc The response rate was also reduced if the inmates believed that participation would interfere with scheduled meals, although this was o nly an issue in the event that participants left the pod to take the survey in a separate room.

PAGE 47

47 Alternatively, the response rate increased when the survey interfered with scheduled lockdowns Inmates on lockdown (confined to their cells) were more likely to participate when given the opportunity to leave the ir cells to complete the survey or stay on lockdown. Protection of Human Subjects This research was approved by the University of Floridas Institutional Review Board (Protocol # 2008U 752) (see A ppendix) An informed consent form was given to participants to keep (see A ppendix D ) Participants were asked to foll ow along as the researcher s read the consent form in both English and Spanish As outlined i n the informed c onsent, participants were inv ited to participate voluntarily and were assured that they may decline to participate, may skip any questions they do not want to answer, or stop participating at any time without penalty from the researchers or from the jail. Participants were told that their answers were anonymous, given that the researchers did not collect their names or inmate identification numbers. Respondents did not receive any rewards for their participation. Similarly, refusal to participate involve d no penalty or loss of benef it. One potential risk participants may have experience d by participating in the research is discomfort associated with the crime perpetration and victimization questions. To minimize this risk, participants may have contacted the jails counseling servi ces, if available. In an effort to help ensure anonymity of the inmates who participate d, signed consent forms were not collected By completing and turning in the survey, participants provid ed their implied consent for research participation. I expecte d that participants would be more comfortable answering the survey questions if they were not required to turn in a consent form that had their name/signature on it. Pilot Study The survey was subjected to a pilot test in an effort to (1) ensure that the s urvey questions could be easily understood by offenders and to (2) determine the length of time respondents

PAGE 48

48 needed to complete the survey. Two versions of the survey were piloted to volunteers from the Alachua County Department of Court Services Drug Cour t Program. The two versions presented identical questions and response options for all survey items and only differed in terms of the way the crime perpetration and victimization questions were displayed One version of the survey presented the response options for these questions as a set of written out statements (version A) whereas the other version displayed the response options in matrix form (version B).8 Volunteers were administered and read aloud version A (n = 5) and version B (n = 11) on separa te occasions with between one and eight individuals at a time. Volunteers were asked to follow along and individually mark their answers as the survey was read aloud to them, although they were informed that they could move forward on their own. All sixt een volunteers chose to read and answer the survey items at a faster pace than the survey was read aloud to them. Version A took between 20 and 60 minutes to complete and version B took between 30 and 35 minutes to complete. Many volunteers provided hel pful feedback and offered suggestions for simpler wording for questions and response options. When volunteers were asked what they disliked about the survey (both version A and B), the majority identified the repetitive nature of the victimization and per petration questions as problematic. In an effort to eliminate the repetitiveness of the crime perpetration and victimization questions and the follow up questions, a third version (version C) was designed. Version C is presented in the final survey and consolidates the questions presented in versions A and B without losing information by displaying columns of questions horizontally rather than vertically positioning separate questions. Version C was selected as the preferred style after college students in an upper division Victimology course at the University of Florida were asked which 8 Examples of the differences between versions A and B are available upon request. The dissertation committee reviewed survey version A in June 2008 and all four committee members expressed an interest in condensing these questions. Therefore, version B (and later, ve rsion C) was created.

PAGE 49

49 version they favored. All students in attendance during class one day (n = 25) were randomly administered either version A or B of the survey in addition to a 1 page al ternative (version C) and were asked to (1) provide feedback and suggestions about the ease with which inmates could understand each question and response option and to (2) identify the best way of asking the victimization and perpetration questions (eithe r their version A or B or version C). Half of the students were given version A and the other half were given version B. All were also given version C. Many students provided positive and helpful feedback. Of the twenty five students present, twen ty four (96%) voluntarily participated: twelve received version A and the other twelve received version B. The majority (n = 17; 71%) indicated their preference for version C rather than their version of either A or B. The remaining students (n = 7; 29% ) preferred either version A (n = 4) or version B (n = 3). Based on this feedback, I decided that version C would reduce unnecessary repetition and be easiest for inmates to understand. Operationalization This research statistically examine d several rela tionships. Therefore, conceptual model s visually display the analysis for each of the hypotheses With this big picture in mind, I first describe how these variables were operationalized and then describe t he analytical plan Figure s 31 and 3 2 visua lly represent the research questions that relate to self control and perceptions of social disorganization. Statistically testing the relationships among these variables require d multiple models. The dependent variables and independent variables change d, depending on the model (see Analytic Plan below). The following describes the way in which the variables of interest were operationalized. Gang m embership Gang membership w as operationalized by asking respondents : Are you currently or have you ever be en in a gang? Response options include d (1) I am not in a gang now and I have

PAGE 50

50 never been in a gang, (2 ) I am not in a gang now, but have been in a gang in the past and (3) I am in a gang now Given the low prevalence rate of inmates who admitted being current or former gang members (1 5%), gang membership was recoded into a dichotomous me asure whereas (1) indicated past or present g ang membership and (0) indicated no gang membership. Crime p erpetration Crime perpetration was operationalized by asking par ticipants if they had ever committed fourteen types of crimes. The perpetration items represent ed a variety of personal crimes (stabbing, carjacking, witness intimidation, home invasion, drive by shooting, robbery, physical assault, assault with a weapon, sexual assault, being threatened with a weapon, being shot at and being shot) and two property crimes (theft and vandalism). Given the focus of the current study on gangs, it was important to include gang related crimes. All fourteen items were listed as criminal gang activity by the California Street Terrorism Enforcement and Prevention (STEP) Act (California Penal Code 186.22). The perpetration survey questions were based on a modified version of questions from several sources (see Table F 1 in t he Appendix for a list of the modified items, original items, and sources for both the perpetration and victimization items ). Although response options allowed respondents to indicate whether or not they ever committed each of the items (yes and no re sponse options), analyses are based upon follow up survey questions that asked respondents to provide the number of times they committed each crime For each crime perpetration question respondents were allowed to write in the number of times the crime ha ppened (a) if they were never in a gang, (b) before they were in a gang, (c) while they were in a gang, and (d) after they were in a gang. Responses from (a), (b), (c), and (d) were combined, which represented the total number of times the crime type was committed by each respondent For example, a gang member who indicated they had been physically assaulted twice before gang membership, three times during gang membership, and once after gang

PAGE 51

51 membership would receive a total score of six. Respondents who indicated that they did not commit specific crimes were assigned a 0 for those particular items. Data indicating a range of the number of times specific items happened (e.g., 5 to 10 times ) were coded conservatively with the lowest number provided in an effort to avoid over inflation. Conservatively coding data also included coding a written response of a couple times as and a few times as Some respondents perpetrated certain items more times throughout their lives than they could recal l. A separate code (999) was assigned for responses of many, lots, countless, all my life, etc., and other codes were assigned for several (998), some (997), rarely (996), and various (995). These responses (coded 995 through 999) were coded as missing data given the inability to attribute any meaningful numerical values and given the ir infrequency.9 Analyses for the perpetration items were conducted using the count data (number of times each item happened), which allowed for a greater understanding of the extent to which jail inmates commit ted crime (in comparison with the yes/no responses) While using count data has the benefit of determining the level of participation in offending the skewness from outliers is problematic for data analysis. For example, a handful of respondents entered large numbers that substantially skewed the distribution (e.g., 400, 1,000 or 1 million ) In an attempt to eliminate the methodological issues associated with outliers, each of the perpetrati on indexes (described next) were truncated at the 99th percentile (see Nagin and Smith, 1990).10 9 Although these responses were coded as missing for the current analysis, it was important to differentiate the qualitative meaning of these written responses with different codes for future analyses. All written responses coded as 999 included: many, too many, lots, a lot, countless, all my life, all the time, dozens of, often, multiple, numerous, hundreds, thousands, plenty, and a bunch. 10 There are several ways of addressing analysis issues presented by outliers, including trunc ating at various percentiles or cut off points. Some prior research examining the gang victimization link has truncated count data at twelve (Gibson et al., forthcoming 2009; Peterson et al., 2004; Taylor et al., 2007). In order to preserve the maximum o riginal variation, individual items were not truncated; only the composite indexes were truncated at the 99th percentile. Crime perpetration responses above the 99th percentile were infrequent and were considered as outliers. Truncating responses affecte d very few cases for both the victimization and perpetration indexes. The

PAGE 52

52 Several preliminary analyses were performed to determine the most appropriate way to combine the crime perpetration items. As expected, bivariate correlation s indicate d that many of the items were significantly correlated with one another (see Table 33). A ll items were significantly and positively correlated with multiple other crime types with the exception of sexual assault perpetration which was not sign ificantly correlated with any of the other thirteen items. This finding supports research indicating t hat sexual assault is fundamentally different from other crime types (Myers & LaFree 1982 ). Correlations are also designed to detect any potential issu es resulting from multicollinearity, which can be defined as a linear functional relationship between two or more independent variables that is so strong that it can significantly affect the estimation of the coefficients of the variables (Studenmund, 2001, p. 247). An examination of the perpetration correlation coefficients indicates two potential multicollinearity issues given that robbery and vandalism show a correlation coefficient of .880 (p < .05) and being attacked without a weapon and being threa tened with a weapon indicate a correlation coefficient of 1.000 (p < .01) .11 These are potential issues of multicollinearity because, as Parker and Smith (1984) point out, high correlations do not necessarily mean that multicollinearity will be an issue. Because only two sets of the fourteen items we re highly correlated, and in light of the drawbacks associated with multicollinearity remedies ( Studenmund, 2001), these items were included in the indexes and analyses (described next). Studenmund (2001, p. 259) suggests that doing nothing with potential multicollinearity issues is often the correct decision given that deleting a particular item may cause specification bias such that a model fits because it accidentally works for the particular data set invol ved, not because it is the truth. number of affected cases for the perpetration indexes was 22 for the property crime perpetration index 24 for the personal crime perpetration index and 24 for the combined crime p erpetration index 11 This conclusion was determined using the standard cut off point of a correlation coefficient exceeding .80 (Studenmund, 2001)

PAGE 53

53 A factor analysis was then performed with the crime perpetration items in an effort to determine possibilities for index construction. A Principal Component factor analysis using Varimax rotation with Kaiser Normalizatio n12 performed with all fourteen perpetration items revealed five factors (see Table 3 4) Two items did not load on any of the factors (witness intimidation and drive by shooting) and were removed from the subsequent factor analysis (also indicating a five factor solut ion). The first factor included vandalism, robbery, and t heft; the second factor included carjacking, sho ot ing at someone home invasion, attack ing someone with a weapon, and stabbing someone ; the third factor was comprised of threatening som eone with a weapon and attacking someone without a weapon ; the fourth factor indicated that shooting someone loads independe ntly; the fifth factor indicated s exual assault loaded on a separate factor ( which wa s in line with the correlation results indicating that sexual assault is different from the other crime types ). Sexual assault perpetration (and victimization, as will be discussed below) was removed from the analyses because (1) it was not significantly correlated with any of the other crime types, ( 2) it loaded independently in the factor analysis, and (3) prior research indicates that sexual assault is a unique type of crime that is dissimilar from other crime types ( Myers & LaFree 1982). The items loading together in the factor analysis present so me methodological obstacles when determi ning the way in which indexes c ould be constructed. First, robbery ( personal crime ) load s with theft and vandalism (property crimes), and combining personal and property crimes together violates intuitive reasoning and contradicts practices of well established prior research that measures perpetration based on crime type (Lauri tsen et al., 1991; Thornberry et al., 12 Employing a rotation often allows for a simpler interpretation in comparison with unrotated solutions. Rotation recalculates the factor loadings compared with the new rotated factors (Warner, 2008). Varimax rotation is the most common and widely accepted method for factor analysis given that it reduces the number of factors that load highly together (Warne r, 2008)

PAGE 54

54 1993). Second, the differences between the factor loadings are puzzling and are not clearly indicative of well established crime types. While the second factor may appear to include serious weapon related offenses, the third factor also includes offenses for which weapons are used ( being threatened with a weapon ). Finally, as will be discussed in the following section, the factor loadings for the perpetration items are dissimilar to the factor loadings for the victimization items. Therefore, creating indexes based on the factor loadings not only violate several assumptions about the nature of crime types (e.g., property versus personal crimes), but creating dissimilar victimization and perpetration indexes would not allow for comparisons (between victimization and perpetration) or meaningful interpretations.13 Therefore, I determined that matching perpetra tion and victimization indexes would allow for a greater degree of interpretation and comparison. Three perpetration indexes were created (property crime perpetration, personal crime perpetration, and combined crime perpetration). Cronbachs alphas indic ate high reliability for each of the perpetration indexes (. 813 for property crime perpetration, .770 for personal crime perpetration, and .693 for combined crime perpetration).14 Crime v ictimization Crime victimization survey items, response options, and recoding mirrored the perpetration items. S imilar to the perpetration correlations b ivariate correlations among the victimization items indicated many of the victimization items were positively and significantly associated with one another Identical t o sexual assault perpetration sexual assault victimization wa s not 13 For example, interpreting the effect of certain variables in the model (i.e., self control, social disorganization, gang membership) predicting perpetration using the categories produced by the factor analysis would result in less meaningful fi ndings given that the perpetration items in each index are not logically grouped. 14 The reliability analysis for the property crime items represents the original untruncated count variables. Due to outliers, the reliability analysis generated negative ave rage covariances with the original untruncated count variables for the personal and combined property crime variables. Therefore, the Cronbachs alphas presented for the personal and combined perpetration indexes are based on the individual items truncate d at the 99th percentile.

PAGE 55

55 significantly associated with any other victimization crime type. Furthermore, no issues of multicollinearily were detected (see Table 35). A Principal Component factor analysis using V arimax rotation with Kaiser Normalization performed with all fourteen victimization items revealed seven factors. The first factor included carjacked, witness intimidation, being shot, home invasion, and being stabbed; the second factor included being thr eatened with a weapon and being attacked without a weapon; the third factor included theft, vandalism, and robbery; the fourth factor revealed that being attacked with a weapon loaded by itself, as did the fifth factor (being shot at), sixth factor (sexual ly assaulted), and seventh factor (drive by shooting) (see Table 3 6). The factor analysi s for the victimization items substantially differs from the factor analysis for the perpetration items. While the perpetration items suggested a five factor solutio n, the victimization items load ed on seven factors. Furthermore, items loaded on different factors between the two analyses, with the exception of sexual assault loading independently and robbery loading with theft and vandalism for both perpetration and victimization factor analyses Instead of creating indexes based on the unintuitive factor loadings, victimization indexes were created which mirrored the perpetration indexes using the count data which measur ed (1) property crime ( theft and vandalism), ( 2) personal crime ( being threatened with weapon, attack ed without weapon, attack ed with weapon, robbed, stab bed carjack ed witness intimidation, home invasion, drive by shooting, shot at and shot), and (3) combined crime ( combining the property and person al crime items ).15 Categorizing variables in this way (property, personal, and a combined measure) is consistent with well established prior literature (Peterson et al., 2004; Thornberry et al., 1993) and yielded acceptable Cronbachs reliability alphas ( .591 for property 15 Recall that sexual assault, was removed from the analysis of both victimization and perpetration given the unique nature of this type of crime (see Myers & LaFree, 1982), which has been the focus of unique theories (Koss, 2001 ) and unique consequences (i.e., fear) (Ferraro, 1995). Furthermore, the index reliabilities decreased with sexual assault included.

PAGE 56

56 crime victimization .693 for personal crime victimization and .725 for combined crime victimization ) .16 The count data for each index were also truncated at the 99th percentile.17 Self c ontrol Self control was operationalized by asking respondents a series of twenty three modified questions from Grasmick et al.s (1993) well established self control scale ( see Table F 2 in Appendix F for a comparison of the modified and original items ).18 Four point Likert response options for the twenty three questions range d from strongly agree (1) to strongly disagree (4) Table 3 7 shows that the bivariate correlations indicate the self control items we re significantly rel ate d to one another with few exceptions .19 Additionally, the bivariate correlat ion reveals no concerns about multicollinearity among the self control items. Given the length of the survey questions, t he following tables comprised of the self control items identify the self control constructs as well as the original survey numbers (i .e., Impulsivity36). Readers can link this information with Table F 2 in Appendix F or to the survey in Appendix B. A Principal Component factor analysis using Varimax rotation with Kaiser Normalization of the selfcontrol items revealed findings consiste nt with Grasmick et al. (1993 ). Instead of a 16 Similar to the perpetration reliability analyses, the original untruncated victimization items resulted in negative aver age covariances due to outliers. The Cronbach alphas presented for all victimization indexes are based on each item truncated at the 99th percentile. 17 As with the perpetration indexes, the number of victimization responses above the 99th percentile were infrequent and were considered as outliers. The number of affected cases for the victimization indexes was 23 for the property crime victimization index, 24 for the personal crime victimization index, and 23 for the combined victimization index. 18 A 24th item was included in Grasmick et al.s (1993, pp. 14) original index, which read I seem to have more energy and a greater need for activity than most other people my age. However, this item, originally placed in the physical activities component, reduced Grasmick et al.s index reliability analysis and therefore, the researchers recommend omitting this item from inclusion based on the possibility for respondents to interpret activity as inclusive of non physical activities. 19 One self control item repre senting physical activities (I like to get out and do things more than I like to sit around) was not significantly correlated with two other items, one representing self centeredness (I dont care so much when other people are having problems) and the other representing simple tasks (I dont like really hard jobs that push me).

PAGE 57

57 six factor solution, as Gottfredson and Hirschis (1990) theory suggests a four factor so lution was observed (four eigenvalues greater than 1.0) whereas many of the like ite ms did not load together ( items meas uring temper loaded with self centeredness, items measuring a preference for simple tasks loaded with impulsivity, etc.) Furthermore, a few i tems loaded on multiple factors (self centere dness55 and 47, impulsivity44, and simple tasks 42) and some of the i tems did not load on any of the factors (self centeredness43 and impulsivity36). S ee Table 38 for the best fitting solution Again, t he following table identif ies the self control constructs as well as the original survey numbers (i.e., Impulsivity36). Readers can link this information with Table F 2 in Appendix F or to the survey in Appendix B to view the survey questions. Similar to the current study, Grasmick et al. (1993) find a five factor solution with the same issue of different items loading to gether (also combining impulsivity with a preference for simple tasks) and several items loading on more than one factor. Like Grasmick el al. (1993, p. 17), the current study supports a unidemensional, rather than multidimensional, conceptualization of s elf control given that the identifiable factor loadings do not produce readily interpretable multidimensionality. Consistent with Grasmick et al.s (1993) conclusion, the current study also argues for unidimensionality given the largest difference in ei genvalues between the first and second factor. While a comprehensive review of the evidence supporting unidimensionality (Grasmick et al., 1993; Piquero & Rosay, 1998; Turner & Piquero, 2002) and multidimensionality (Longshore et al., 1996; Longshore, Ste in, & Turner 1998) is beyond the scope or aim of this study, much empirical and theoretical evidence exists to suggest that the six components of self control collectively measure the broad self control construct. For purposes of this analysis, the self c ontrol items were summed and divided by twenty three which resulted in

PAGE 58

58 scale scores ranging from 1 (indicating lower self control) to 4 (indicating higher self control). The self control scale ha d a strong reliability (Cronbachs alpha = .902). Perceptio ns of social disorganization Perceptions of s ocial disorganization were operationalized by asking respondents a series of questions related to the structural and social characteristics of their neighborhoods. Measuring both constructs (structural and soc ial neighborhood characteri stics) wa s critical to gaining a comprehensive assessment of perceptions of social disorganization (Sampson & Raudenbush, 2001). Given the context of survey research, social variables (such as social disorder ) were measured by a ssessing respondents perceptions of certain neighborhoodlevel conditions. Constructs measuring perceptions of social disorganization include d : (1) physical disorder, (2) social disorder, (3) collective efficacy, and (4) neighbor hood diversity characteri stics (Sampson & Raudenbush, 2001). Physical disorder was measured by asking respondents to report the extent to which each of the following was a problem in their neighborhood (outside of jail) : garbage on the streets, graffiti, abandoned cars, needles a nd syringes used for drugs, and buildings or storefronts sitting abandoned or burned out Response options include d (1) not a problem, (2) some problem, and (3) a big problem Using the same response options, social disorder wa s measured by asking respon dents to report the extent to which each of the following wa s a problem in their neighborhood (outside of jail) : kids hanging out when they should be at school (truant), p eople vandalizing other peoples property people hanging around with nothing to do (loitering) people drinking alcohol in public places people drunk in public places people who looked like they were selling drugs people using illegal drugs people who looked like they were in a gang Collective efficacy wa s measured by three questio ns using a 4point Likert scale ranging from very unlikely (1) to very likely (4). Respondents were asked to rate the likelihood that: Your neighbors would do something if they saw unattended kids misbehaving Your

PAGE 59

59 neighbors would be willing to help ea ch other You could trust your neighbors Neighborhood characteristics was measured by asking respondents four questions. The first two questions ask ed respondents: About how many of your neighbors live in poverty? and About how many of your neigh bors are unemployed? Response options included (1) none, (2) very few, (3) about half, (4) more than half, and (5) I dont know. Third, respondents answered the question About how often do your neighbors move away ? by using response options of (1) rar ely, (2) occasionally, (3) often, and (4) I dont know. Fourth respondents answered the question About how racially mixed is your neighborhood? by using response options of (1) not very mixed (almost all of the neighbors are of the same race), (2) some what mixed (most of the people are of the same race and there are some other races), (3) very mixed (there are people from many different races). All responses of I dont know for the social disorganization items were recoded as missing data. A Princi pal Component factor analysis using Varimax rotation with Kaiser Normalization for the perceptions of social disorganization variables revealed a four factor solution with the physical disorder variables loading together, the social disorder variables load ing together, the collective efficacy variables loading together, and the neighborhood diversity varia bles loading together. Two variable s (truant kids and gangs) loaded on multiple factors (both physical and social disorder) and, therefore, these items w ere deleted from the analysis. With the exception of the neighborhood diversity variables, scales were created based on the factor loadings and responses were summed and divided by the number of items to yield average scale scores for the following: phys ical disorder (ranging from 1 to 3), social disorder (ranging from 1 to 3), and collective efficacy (ranging from 1 to 4) Cronbachs alphas were .881 for the physical disorder scale .892 for the social disorder scale and .741 for the collective efficac y scale The four

PAGE 60

60 neighborhood diversity variables were operationalized individually, rather than collectively as a scale for several reasons. First, the survey items were comprised of different response ranges (e.g., two of the questions range d from 1 to 4 while the other two questions range d from 1 to 3).20 Thus, combining these into a scale would result in half of the variables weighted more heavily than the o thers Second, one of the four neighborhood diversity items (racial diversity with a factor loading of .471 ) did not successfully load with the other neighborhood diversity items ( with factor loadings of .665, .717, and .609). Third, Cronbachs alpha indicated a low reliability of 594 for the neighborhood div ersity items. Therefore, it was det ermined that measuring neighborhood diversity individually would result in the best fitting model The four neighborhood diversity items were: neighborhood poverty (ranged from 1 to 4), neighborhood unemployment (ranged from 1 to 4), residential heterogen eity (ranged from 1 to 3), and neighborhood racial diversity (ranged from 1 to 3) Tables below display correlations, factor loadings, and reliabilities for the physical disorder scale (Table 39), social disorder scale (Table 310), collective efficacy scale (Table 311) and the 7 social disorganization items (Table 312) Demographic variables Demographic variables of interest include d sex, race, ethnicity, and age Sex was meas u red by asking respondents What is your sex? with responses of (0 ) fem al e and (1) male Race was measured by asking respondents What is your race? with responses o f (1) White, (2) Black, (3) Asian, and (4) other A dummy variable dichotomize d race into W hite (1 ) and non 20 The response options yielded different ranges for the neighborhood diversity questions because they were not originally constructed to be combined into a index given that pri or research measures variables related to these separately (Sampson et al., 1997; Sampson & Groves, 1989).

PAGE 61

61 White ( 0 ) Hispanic ethnicity was assessed by aski ng respondents Are you Hispanic? with response options of ( 0) non Hispanic and ( 1) Hispanic A g e w as a continuous variable 21 Measurement l imitations Gang Membership: T he measurement of gang membership relies upon self identification (with no definition provided) rather than an official measurement. Therefore, membersh ip in different types of gangs wa s indistinguishable (e.g., street gangs, prison gangs, motorcycle gangs, etc.). Self identifying t hemselves as gang members allowed respondents to potenti ally deny their gang membership or, alternatively, this method may also have permit ted non gang members to claim gang status. However, it was clear on multiple occasions that denying gang membership was more common than falsely admitting to being in a gan g. For example, one respondent denied gang membership while answering the survey questions and at the end of the survey wrote tell the true [truth] Im in a Gang but real Gang Members never tell baby and thats code, ya dig, holla (male respondent #1,241 from Duval C ounty jail). Given some hesitation to admit to gang membership, the self report measure represents a conservative estimate of gang membership. Nevertheless, self report may be the most valid method for determining gang membership given t hat correctional institutions (and police agencies) frequently identify inmates as gang members based on self declarations (Ruddell, Decker, & Egley, 2006). Furthermore, some correctional facilities do not keep detailed records of gangs (e.g., length of g ang membership and level of gang involvement) (Fleisher & Decker, unpublished manuscript ; and Personal Communication with Florida jail administrators ). 21 Although respondents were informed that they must be at least 18 years old to participate, six juveniles (four 17 year old respondents and two 16 y ear old respondents) participated in the research and were excluded from the analyses.

PAGE 62

62 Crime Perpetration and Victimization : P articipants were asked to retrospectively recall their experien ces with crime perpetration and victimization issues which may have result ed in memory decay, especially with regard to the specific n umber of times particular items occurred. While this is a limitation of the data, some research indicates that respondents can recall salient life events with a substantial degree of accuracy (such as those examined in this study, including types of delinquency) although recalling specific details of those events is less accurate (Henry, Moffitt, Caspi, Langley, & Silva, 1994). Gang members may have experienced greater difficulty in recalling the number of times specific victimization and perpetration items happened given that they, unlike the non gang members, were asked to specify the number of times a behavior occurred be fore, during, and after gang membership. Furthermore, after gang members hip may have been mistaken as after an individual had joined the gang rather than after an individual had left the gang. However, this methodological concern does not hamper the me asurement or findings of the current analysis given that the measurement combines all responses (before, during, and after gang membership) into variables representing total count data. Self Control : Measuring self control has provoked an abundance of l iterature arguing for behavioral (Gottfredson & Hirschi, 1990) versus attitudinal measures (Grasmick et al., 1993; Turner & Piquero, 2002) Behavioral measures of self control assess the extent to which individuals act in certain ways (e.g ., bullying) whereas attitudinal measures asses the extent to which individuals agree with specific preferences (e.g., preference for physical activities). The current study employed a n attitudinal measure of self control given prior research which indicates that attitu dinal and behavioral measures are essentially measuring the same self control construct (Turner & Piquero, 2002).

PAGE 63

63 Perceptions of Social Disorganization : Measuring perceptions of social disorganization raises several potential measurement concerns First, unlike self control (which is established early in life), social disorganization may not affect gang membership crime, or victimization while individuals are incarcerated in jail, prison, or juvenile detention facilities. Given that some inmates have bee n incarcerated continuously or for the majority of their lives, neighborhood influences may have little or no effect on the dependent variables of interest Second, w hile research has long existed on the neighborhoodlevel effects of crime ( K elling & Wils on, 1982; Shaw & McKay, 1969), scholars continue to struggle with the most methodologically sound way to measure neighborhoods (see Martin, 2003, for a comprehensive review). Neighborhoods have been measured in a variety of ways, inclu ding by census tra cks (Sampson et al. 1997), by a 15minute walk radius from residents homes (Sampson & Groves, 1989), and by allowing respondents to interpret the meaning of neighborhood (Skogan & Maxfield, 1981). While each method has unique advantages and disadvantag es, the current study utilized a measure that allowed an open interpretation of the meaning of neighborhood, which allowed respondents to select the neighborhood most salient to their lives. Some research indicates that children have different perceptio ns of what their neighborhood is than adults, which confirms that the concept of neighborhood differs among individuals (Coulton, Korbin, Chan, & Su, 2001). However, Coulton et al. (2001) argue that this may not be a limitation given that the neighborho od respondents select are likely to b e most meaningful in terms of determining the settings with the most influence and impact on behavi or. In other words, if neighborhoods affect behavior (i.e., gang involvement, crime, victimization), then the most salient measurement of neighborhood factors that are important to respondents should be those with which they can identify. The third methodological issue often associated with measuring perceptions of social disorganization

PAGE 64

64 concerns respondents ability to accurately determine the extent of certain social characteristics. While it may be argued that this measure is subject to individual biases and experiences, recent research indicates that respondents are able to assess some social conditions (e.g., poverty and female headed households) with remarkable accuracy when compared with official data (correlations exceeding .80) (Coulton et al., 2001). Given the support from prior research, the current study employs a measurement of perceptions of social disorgani zation based on respondents interpretation of their neighborhood. Analytic Plan Quantitative statistical analyses were used to examine the relationship between micro level and macro level theoretical explanations of gang membership, crime perpetration, an d crime victimization among jail inmates Descriptive statistics correlations, factor analyses, and reliability ana lyses were performed and discussed in this Chapter Multivariate regression models were estimated to explore relationships between the dep endent and independent variables. L ogistic r egression models were estimated for the analyses predicting gang membership due to the dichotomous nature of the dependent variable (Long, 1997) Negative binomial regression models were estimated for analyses predicting crime perpetration and victimization given that the dependent variables are comprised of count data ( for a thorough description of negative binomial regression analysis, see Hilbe, 2007) Negative binomial regression is technique suitable for d ependent variables representing count data that are overdispersed and that have many 0 values (Gardner, Mulvey, & Shaw, 1995 ; Haynie & Armstrong, 2006; Hilbe, 2007; Long, 1997; Osgood, 2000).22 Table 3 13 outlines the independent and dependent variables us ed in each of the models. 22 Given that the dependent variable is comprised of numerical values, employing OLS regression is an unsuitable statistical technique given that the standard error may result in deflation Therefore it is likely that using OLS

PAGE 65

65 Table 3 1. Types of Inmates Excluded from Participating by Each Jail Jail Excluded Psychological Disorders Excluded Communicable Diseases Excluded Solitary Confinement Excluded Federal Inmates Excluded High Security Alachua X X X Broward X X X X Collier X X X X Duval X X X Escambia X X X Hillsborough X X X Lee X X X Leon X X X Miami Dade X X X X Palm Beach X X X X X Pasco X X X Pinellas X X X X Polk X X X Seminole X X X X would result in incorrect findings (such as a finding of statistical significance when the relationship is really non significant). Additionally, using logistic regression (and dichotomizing th e dependent variables) is also an unsuitable method for this analysis. By dichotomizing the dependent variables, important variation would be unnecessarily sacrificed

PAGE 66

66 Table 3 2. Jail Characteristics: Number, Response Rate, Housing Type, and Supervision Type Jail County Populationa Jail Populationb Jail Sample Average Response Rate c Jail Housing Typed Jail Supervision Type Alachua 240,082 866 101 31% Podular Direct Broward 1,759,591 990 196 22% Podular Direct Collier 315,839 1,061 152 29% Podular, Dormitory Remote Duval 849,159 2,518 628 29% Podular Remote Escambia 306,407 1,672 29 32% Podular Direct/ Remote Hillsborough 1,174,727 3,535 119 13% Podular Direct Lee 590,564 988 93 18 % Podular Direct/ Remote Leon 260,945 1,065 207 19% Podular Direct Miami Dade 2,387,170 1,073 259 35% Podular Direct Palm Beach 1,266,451 2,774 129 22% Dormitory, Linear, Podular Direct Pasco 462,715 1,019 42 25% Podular Direct Pinellas 917,437 2,930 1 42 28% Dormitory, Podular Direct/ Remote Polk 574,746 2,288 193 27% Podular, Dormitory Remote Seminole 409,509 1,029 123 19% Podular Remote a County size based on most recent (2007) census data. b Number of inmates on the day(s) of survey administration. c Response rates are based on the total number of inmates present in each pod during the research ( not based on the entire jail population ). The response rate for each jail represents an average of the response rates for each pod. d Podular housing incor porates separation between bed areas; Dormitory housing is often described as open bay with many b unk beds in open area; Linear housing is comprised of a row of dormitories separated by a hallway (see Beck 1999).

PAGE 67

67 Figure 31. Conceptual m odel of the gang perpetration link using self control and perceptions of social disorganization Figure 32. Conceptual m odel of the gang victimization link using self control and perceptions of social disorganization Self Control / Social Disorganization Crime Perpetration Gang Membership Self Control / Social Disorganization Crime Victimization Gang Membership

PAGE 68

68 Table 33. Crime Pe rpetration C orrelations T V RY TW A AW S SA C WI HI DB SHA SH T 1 V ** .722 1 RY ** .787 880 1 TW .001 .001 .007 1 A .001 .001 .008 ** 1. 000 1 AW ** .069 .043 .053 ** .252 ** .251 1 S 035 .036 .035 .015 .025 .139 1 SA .007 .004 .000 .001 .001 .020 .002 1 C ** .085 ** .060 ** .059 .002 .002 ** .330 ** .210 .007 1 WI .025 .029 .019 .002 .002 ** .070 ** .105 .002 ** .135 1 HI .053 .039 .050 .042 .042 ** .357 ** 165 .002 ** .436 ** .222 1 DB .002 .001 .001 .000 .000 .003 .006 .001 .002 .002 .003 1 SHA ** .067 .044 .043 .019 .019 ** .464 ** .126 .003 ** .502 ** .153 ** .531 .003 1 SH .006 ** .139 .013 .026 .025 ** .091 .035 .004 .053 .036 ** .181 .019 ** .220 1 T = Theft; V = Vandalism; RY = Robbery; TW = Threatened with a weapon; A = Attacked without a weapon; AW = Attacked with a weapon; S = Stabbed; SA = Sexually assaulted/rapped; C = Carjacked; WI = Witness intimidation; HI = Home invasion; DB = Drive by shooting; S H A = Shot at ; SH = Shot Correlations significant at the .05 level ** Correlations significant at the .01 level

PAGE 69

69 Table 34. Crime Perpetration Factor Analysis (with Factor Loadings) Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Vandalism (.970) Robbery (.970) Theft (.914) Carjack (.798) Shot at (.760) Home invasion (.738) Attack with weapon (.641) Stab (.566) Threat with weapon (.992) Attack without weapon (.992) Shot (.919) Sexua l assault (1.000)

PAGE 70

70 Table 3 5. Crime Victimization Correlations T V RY TW A AW S SA C WI HI DB S H A SH T 1 V ** .295 1 RY ** .294 ** .194 1 TW .005 .007 .008 1 A .011 .030 .020 ** .743 1 AW 046 ** .097 ** .062 .014 ** 067 1 S .026 ** .109 ** .067 .007 .028 ** .128 1 SA .000 .035 .003 .001 .001 .004 .005 1 C .007 .034 ** .081 .003 .006 ** .072 ** .302 .001 1 WI .014 .045 .048 .002 .028 ** .077 ** .233 .003 ** .63 3 1 HI ** .411 ** .093 ** .085 .003 .006 ** .056 ** .199 .003 ** .400 ** .270 1 DB .005 .007 .002 .000 .002 .005 .007 .001 .036 .002 .003 1 S H A .005 .007 .005 .000 .002 .005 .043 .001 .003 .002 .003 .001 1 SH .021 .034 ** .065 .00 3 .014 ** .081 ** .253 .003 ** .454 ** .308 ** .300 .003 .011 1 T = Theft; V = Vandalism; RY = Robbery; TW = Threatened with a weapon; A = Attacked without a weapon; AW = Attacked with a weapon; S = Stabbed; SA = Sexually assaulted/rapped; C = Carjacked; WI = Witness intimidation; HI = Home invasion; DB = Drive by shooting; SHA = Shot at; SH = Shot Correlations significant at the .05 level ** Correlations significant at the .01 level

PAGE 71

71 Table 3 6. Crime Victimization Factor Analysis (with Factor Loading s) Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7 Carjack (.876) Witness intimidation (.751) Shot (.680) Home invasion (.622) Stab (.530) Threat with weapon (.934) Attack without weapon (.932) Theft (.850) Vandalism (.593) Robbery (.587) Attack with weapon (.834) Shot at (.984) Sexual assault (.987) Drive by shooting (.997)

PAGE 72

72 72 Table 3 7. Self Control Correlations i36 i44 i51 i54 p37 p4 0 p52 t3 8 t46 t50 t53 i36 1 i44 ** .231 1 i51 ** .333 ** .313 1 i54 ** .280 ** .375 ** .428 1 p37 ** .244 ** .074 ** .073 ** .041 1 p40 ** .302 ** .176 ** .255 ** .233 ** .229 1 p52 ** .247 ** .149 ** .318 ** .328 ** .227 ** .288 1 t38 ** .410 ** .226 ** .294 ** .228 ** .123 ** .325 ** .224 1 t46 ** .390 ** .239 ** .323 ** .287 ** .077 ** .236 ** .263 ** .504 1 t50 ** .328 ** .246 ** .384 ** .312 ** .078 ** .270 ** .274 ** .496 ** .523 1 t53 ** .380 ** .249 ** .356 ** .395 ** .091 ** .285 **. 347 ** 427 ** .486 ** .480 1 r39 ** .410 ** .209 ** .401 ** .255 ** .173 ** .398 ** .297 ** .398 ** .303 ** .332 ** .305 r45 ** .325 ** .391 ** .433 ** .300 .048 ** .226 **. 224 ** .335 ** .341 ** .343 ** .291 r48 ** .326 ** .243 ** .415 ** .287 ** .109 ** .268 ** .283 ** .291 ** .312 ** .37 2 ** .338 r56 ** .302 ** .364 ** .489 ** .425 ** .105 ** .284 ** .299 ** .278 ** .280 ** .306 ** .306 c41 ** .264 ** .249 ** .291 ** .273 .050 ** .238 ** .201 ** .371 ** .358 ** .392 ** .340 c43 ** .237 ** .296 ** .328 ** .306 ** .079 ** .226 ** .248 ** .327 ** .314 ** .347 ** .306 c 47 ** .194 ** .326 ** .294 ** .273 .012 ** .174 ** .164 ** .318 ** .346 ** .313 ** .292 c55 ** .339 ** .316 ** .451 ** .397 ** .077 ** .243 ** .289 ** .391 ** .378 ** .445 ** .427 s42 ** .202 ** .212 ** .278 ** .275 ** .056 ** .229 ** .219 ** .224 ** .235 **.2 69 ** .257 s49 ** .154 .310 ** .289 ** .315 .042 ** .165 ** .174 ** .157 ** .232 ** .276 ** .274 s57 ** .175 ** .318 ** .293 ** .369 .035 ** .150 ** .220 ** .210 ** .258 ** .287 ** .290 s58 ** .192 ** .363 ** .302 ** .342 .052 ** .126 ** .160 ** .181 ** .220 ** .228 ** .255 i = impulsivity item; p = physical activity item; t = temper item; r = risk seeking item; c = self centered item; s = simple tasks item Correlations significant at the .05 level ** Correlations significant at the .01 level

PAGE 73

73 73 Table 37. (continued) r39 r45 r48 r56 r41 c43 c4 7 c55 s42 s49 s57 s58 i36 i44 i51 i54 p37 p4 0 p52 t38 t46 t50 t53 r39 1 r45 ** .464 1 r48 ** .521 ** .461 1 r56 ** .408 ** .445 ** .437 1 c41 ** .285 ** .290 ** .301 ** .306 1 c43 ** .284 ** .321 ** .309 ** .342 ** .386 1 c47 ** .209 ** .321 ** .295 ** .341 ** .367 ** .370 1 c55 ** .361 ** .450 ** .415 ** .462 ** .423 ** .456 ** .3 36 1 s42 ** .190 ** .171 ** .201 ** .238 ** .312 ** .333 ** .199 ** .287 1 s49 ** .124 ** .179 ** .195 ** .233 ** .200 ** .247 ** .265 ** .260 ** .295 1 s57 ** .173 ** .239 ** .189 ** .330 ** .256 ** .300 ** .296 ** .307 ** .284 ** .394 1 s58 ** .194 ** .299 ** .162 ** .32 9 ** .169 ** .225 ** .248 ** .282 ** .201 ** .350 ** .505 1 i = impulsivity item; p = physical activity item; t = temper item; r = risk seeking item; c = self centered item; s = simple tasks item Correlations significant at the .05 level ** Correlations significant at the .01 level

PAGE 74

74 Table 3 8. Self Control Factor Analysis (with Factor Loadings) Factor 1 Factor 2 Factor 3 Factor 4 Temper46 (.766) Temper38 (.748) Temper50 (.726) Temper53 (.626) Self Centered41 (.557) Risk Seeking45 (.733) Risk Seeking48 (.719) Risk Seeking39 (.713) Risk Seeking56 (.641) Impulsivity51 (.568) Simple Tasks57 (.754) Simple Tasks58 (.729) Simple Tasks49 (.651) Impulsivity54 (.584) Physical37 (.796) Physical52 (.624) Physical40 (.519) Table 3 9. Physical Disorder Correlations, Factor Loadings, Reliability, and Descriptive Statistics Correlations Garbage Graffiti Abandoned Cars Needles/Drugs Vandalism Abandoned Buildings Garbage 1 Graffiti **.623 1 Aba ndoned Cars **.580 **.609 1 Needles/Drugs **.534 **.545 **.572 1 Vandalism **.559 **.522 **.559 **.594 1 Abandoned Buildings **.498 **.465 **.525 **.529 ** .537 1 Factor loadings .741 .803 .795 .775 .672 .656 Scale Reliability (Cronbachs a lpha) = .881 Scale Range = 1 to 3 Scale Mean (SD) = 1.73 (.623) ** Correlations significant at the .01 level

PAGE 75

75 Table 310. Social Disorder Correlations, Factor Loadings, Reliability, and Descriptive Statistics Kids Truant Loitering Drinking Drunk Sell ing Drugs Using Drugs Gangs Kids Truant 1 Loitering ** .592 1 Drinking ** .516 ** .619 1 Drunk ** .552 ** .578 ** .769 1 Selling Drugs ** .520 ** .585 ** .593 ** .588 1 Using Drugs ** .528 ** .581 ** .591 ** .587 ** .705 1 Gangs ** .496 ** .438 ** .460 ** .499 ** .540 ** .474 1 Factor loadings Omitted .738 .815 .758 .782 .793 Omitted Scale Reliability ( Cronbachs alpha ) = .892 Scale Range = 1 to 3 Scale Mean (SD) = 2.01 (.636) ** Correlations significant at the .01 level Table 311. Coll ective Efficacy Correlations, Factor Loadings, Reliability, and Descriptive Statistics Neighbors take action Neighbors help each other Trust neighbors Neighbors take action 1 Neighbors help each other ** .470 1 Trust neighbors ** .386 ** .615 1 Fact or loadings .751 .856 .798 Scale Reliability ( Cronbachs alpha ) = .741 Scale Range = 1 to 4 Scale Mean (SD) = 2.63 (.840) ** Correlations significant at the .01 level

PAGE 76

76 Table 3 12. Social Disorganization Correlations Physical disorder Soc ial disorder Collective efficacy Poverty Unem ployed Residential mobility Racial hetero geneity Physical disorder 1 Social disorder **.696 1 Collective efficacy ** .168 ** .271 1 Poverty **.282 **.309 ** .179 1 Unem ployed **.198 **. 218 ** .176 **.449 1 Residential mobility **.147 **.164 ** .152 **.203 **.225 1 Racial hetero geneity **.056 *.044 *.046 **.093 **.076 **.173 1 Correlations significant at the .05 level ** Correlations significant at the .01 level

PAGE 77

77 Table 3 13. Regression Model Variables Dependent Variable Independent Variables Gang Membership (Tables 5 1 to 5 3) Model 1 Demographics Model 2 Demographics + Crime p erpetration Model 3 Demographics + Crime perpetration + Self control Model 4 Demographics + Crime perpetration + Social disorganization Model 5 Demographics + Crime perpetration + Self control + Social disorganization Gang Membership (Tables 5 4 to 5 6) Model 1 Demographics Model 2 Demographics + Crime victimization Model 3 Demographics + C rime victimization + Self control Model 4 Demographics + Crime victimization + Social disorganization Model 5 Demographics + Crime victimization + Self control + Social disorganization Crime Perpetration (Tables 6 1 to 6 6) Model 1 Demographics Model 2 Demographics + Gang membership Model 3 Demographics + Gang membership + Self control Model 4 Demographics + Gang membership + Social disorganization Model 5 Demographics + Gang membership + Self control + Social disorganization Crime Victimization (Ta bles 7 1 to 7 6) Model 1 Demographics Model 2 Demographics + Gang membership Model 3 Demographics + Gang membership + Self control Model 4 Demographics + Gang membership + Social disorganization Model 5 Demographics + Gang membership + Self control + Social disorganization

PAGE 78

78 CHAPTER 4 DESCRIPTIVE STATISTICS Sample Demographics The sample consisted of 2,414 Florida county jail inmates (see Table 41 for descriptive statistics) The majority of the sample (n = 1,746 ; 75%) were men The racial composit ion of the sample was diverse, with 43 % (n = 991 ) White, 41% (n = 956) Black, and 16% (n = 382) Other .1 Approximately 21% of the sample (n = 478) ind icated they were Hispanic While respondents a ges ranged from 18 to 84, the majority of the sample was y oung (mean of 32 years old median was 29 and mode was 21 years old ) The educational background of inmates ranged from 0 to 4th grade ( 1%, n = 26), 5th to 8th grade ( 6%, n = 130), 9th to 11th grade (29 %, n = 676), 12th grade or GED completion (35 %, n = 825), some college ( 21%, n = 483), college graduate ( 7%, n = 153), and graduate work (1%, n = 34) In terms of employment before incarceration, 49% (n = 1,125) of the sample reported working full time, 15% (n = 333) reported working part time, 3 % (n = 76 ) reported being seasonally employed, 9 % (n = 204) indicated being temporarily employed, and 24 % (n = 562) were unemployed or not legally employed. Respondents were able to indicate the number of days spent in jail at the time of the survey using categoric al response options including 090 days ( 60%, n = 1,383), 91180 days (18 %, n = 416), 181 to 270 days ( 9%, n = 218), 271 to 365 days ( 5 %, n = 123), and more than 365 days ( 8%, n = 180). When asked how long they expected to be in jail, many respondents adm itted that they did not know ( 50 %, n = 1,158), whereas other respondents indicated 090 days (22%, n = 507), 91180 days ( 9 %, n = 215), 181 to 270 days ( 5 %, n = 109), 271 to 365 days (5 %, n = 119), and more than 365 days ( 9%, n = 200). When asked to selec t the reason(s) why they were incarcerated, 34% (n = 779) were waiting for a trial, 19% (n = 442) were sentenced to a short jail 1 Given that only 14 respondents were Asian (less than 1% of the sample), Asians were also categorized as others.

PAGE 79

79 term, 25% (n = 577) were in jail for a probation/parole violation, 1% (n = 21) for escaping while on bail, 1% (n = 13) were awa iting transfer to a mental health facility, 6% (n = 140) were awaiting transfer to prison, 2% (n = 56) were awaiting transfer to another jail, 2% (n = 40) were in contempt of court, nearly 1% (n = 11) were released from prison, 1% (n = 19) were court witne sses, and 28% (n = 649) were incarcerated for other unlisted reasons. Respondents were asked to identify the type s of crimes for which they were incarcerated, and 25% (n = 582) reported being in jail for a property crime, 23 % (n = 526) for a personal crim e, 28% (n = 642 ) for a drug crime, 32% (n = 733) for an other unlisted type of crime, and 6% (n = 141 ) reported being incarcerated for no crime.2 Table 4 1 describes the full sample, non gang sample, and the gang sample in terms of sex, race, ethnicity, ag e, education, employment, number of days in jail at the time of the survey number of days expected to spend in jail, reason for incarceration, o ffense type charged with or convicted of Interestingly, a chi square test of association indicated significan t differences between gang and non gang members were only observed for sex, race, and ethnicity whereas all other variables were substantively similar between gang and non gang members ( age, education, employment, number of days in jail, number of days expected to spend in jail, reason for incarcera tion, offense type charged with/ convicted of ). Gang Membership Fifteen percent (n = 370) of the sample admitted being current (6%, n = 145) or former (9%, n = 225) gang members. Consistent with prior research, g ang members are primarily young (mean age of 28 median age was 25, and mode was 22 year s old) and male (85%, n = 315) (Decker & Van Winkle, 1996). The racial composition of gang members was mixed, with 37% (n = 137) White, 36% (n = 133) Black, 27% Other (n = 100) and 29% (n = 107) Hispanic. 2 The percents reported for reasons for being in jail and crime type do not total 100% given that respondents were allowed to select all applicable categories.

PAGE 80

80 Examining the full sample shows that an equal percent of Whites and Blacks (13%) admitted gang membership while more Hispanics indicated they were gang members (22%). When gang members were asked the age they first joined the gang, ages ranged from four to thirty one however the most common age range for joining the gang was between ten and eighteen years old (mean = 14). Gang members reported leaving the gang between ages ten and forty with the most common age ran ge between sixteen and twenty five Of the gang members, 41% (n = 140) reported still being in the gang at the time of the survey. The survey asked a variety of gang related questions, which will be used for other analyses beyond the scope of this study. However, general characteristics of gang membership may be of interest and are presented briefly here. Gang members reported that their gang had initiation/joining rites (68%, n = 252) and many gang members acknowledged that they had been jumped or beat en into the gang (54%, n = 201). Most of the gang members indicated that their gang had leaders (73%, n = 270) and, surprisingly, a substantial number respondents indicated that they were gang leaders (35%, n = 131). Consistent with research on gangs, th e majority of gang members indicated their gang had a name (82%, n = 302), symbols or colors (76%, n = 282), hand signs (76%, n = 283), and that they had a nickname within the gang (74%, n = 274) (see Decker & Van Winkle, 1996). Sixty two percent (n = 228) of gang members had joined the gang before entering jail. Many gang members revealed that their gang was both inside and outside the jail (48%, n = 178) whereas only 2% (n = 7) reported their gang was inside the jail only and 15% (n = 55) reported their gang was outside the jail only.3 When asked about their future plans to remain in the gang once released from jail, responses were mixed with 18% (n = 68) acknowledging they would stay in the gang, 5% (n = 20) who would like to get out of the gang, 8% (n = 31) who would like to 3 Many ex gang members did not report whether their gang was inside or outside the jail and, instead, indicated that they were no longer gang members.

PAGE 81

81 get out of the gang but believed they could not, 10% (n = 37) who believed they would get out of the gang, and 9% (n = 33) who were unsure. Crime Perpetration While many inmates denied perpetrating each of the fourteen crime type s a substantial number of inmates admit ted committing crimes (see Table 4 2) Results indicated that a substantial number of inmates admitted to perpetrating property crimes including theft ( 44 %, n = 1,007) and vandalism (33%, n = 755). Inmates also adm itted committing personal crimes including robbery (18 %, n = 421), threatening someone with a weapon ( 30%, n = 681 ), assault without a weapon (42 %, n = 947), assault with a weapon ( 23%, n = 535), sexual assault/rape ( 2%, n = 44), stabbing ( 11%, n = 258), c arjacking ( 6 %, n = 141), witness intimidation ( 5%, n = 116), home invasion ( 13 %, n = 296), drive by shooting ( 10%, n = 223 ), shooting at someone but not hitting them ( 16%, n = 382), and shooting someone ( 10%, n = 227 ). Of the inmates who admitted offendin g, the most common response with the exception of attacking someone without a weapon was that it occurred on a single occasion.4 In fact, respondents typically indicated that they committed each offense type on very few occasions, which is in line wit h what prior survey research on incarcerated samples determined (Peterson et al., 1981). Crime Victimization Results indicated that more inmates reported being victimized by crime in comparison to perpetrating crime (see Table 4 3). Respondents reported being victimized by property crimes including theft ( 49%, n = 1,106) and vandalism ( 38%, n = 861). Inmates also admitted to being victimized by personal crimes including robbery ( 28%, n = 647 ), being threaten ed with a weapon ( 51%, n = 1,138), being assau lted without a weapon ( 50 %, n = 1,133), being assaulted with a 4 Of the inmates who reported attacking someone without a weapon, the most common response was that it occurred three times.

PAGE 82

82 weapon (37 %, n = 847), being sexual ly assault ed /rape d ( 14%, n = 316), being stabb ed ( 21 %, n = 494), being carjack ed ( 4%, n = 106), being a victim of witness intimidation ( 4%, n = 102 ), home inv asion ( 11%, n = 270), drive by shooting ( 20%, n = 477 ), being shot at but not hit ( 30%, n = 688), and being shot ( 11%, n = 249 ). Of the inmates who admitted being victimized the most common response with the exception of theft was that it occurred on a single occasion.5 Like perpetration, respondents typically indicated that they were victimized by each offense type on very few occasions, which is in line with what prior survey research on incarcerated samples determined (Peterson et al., 2004; Taylo r et al., 2007). Self Control and Perceptions of Social Disorganization While Gottfredson and Hirschi (1990) hypothesized that offenders will exhibit low self control, the current study finds a considerable amount of variation among self control levels of jail inmates. Self control ranged from 1 (low self control) to 4 ( high self control) and the mean was 2.72, which indicated a medium high level among the sample. Perceptions of social disorganization included three scales and four individual items. T he scales represented physical disorder and social disorder (both ranged from 1 to 3 with higher values represen ting higher levels of disorder) and collective efficacy (ranged from 1 to 4 with higher values representing higher levels of collective efficacy ). The four neighborhood diversity items were: neighborhood poverty (ranged from 1 to 4 with higher values indicating more poverty ), neighborhood unemployment (ranged from 1 to 4 with higher values indicating more unemployment ), residential heterogeneity (ranged from 1 to 3 with higher values indicating more heterogeneous ), and neighborhood racial diversity (ranged from 1 to 3 with higher values indicating more racial diversity ). Results show a mean of 1.73 for physical disorder, 2.01 for social disorder, 2.63 for 5 Of the inmates who reported being victims of theft, the most common response was that it occurred twice.

PAGE 83

83 collective efficacy, 2.46 for neighborhood poverty 2.61 for neighborhood unemployment, .174 for residential heterogeneity and 1.99 for neighborhood racial diversity Accuracy of Self Report Survey Data While using self report survey data is a c ommon method for ascertaining the extent to which individuals have perpetrated crime (i.e., National Survey on Drug Use and Heath and the Monitoring the Future Survey) or been victimized by crime (i.e., National Crime Victimization Survey and the National Violence Against Women Survey), it is important to address the potential for respondents to be untruthful therefore biasing the results Given the problems associated with official statistics (i.e., unreported crime, little detail surrounding the offense ), self report surveys have generated a wealth of information beyond the capacity of official data ( Maxfield & Babbie, 2008). However, the validity of self report data relies upon respondents willingness to provide truthful responses. When asked about personal (and perhaps incriminating) experiences with crime perpetration and victimization, respondents may fail to provide accurate information by underreporting or over reporting for a variety of reasons. Respondents may accidentally underreport their ex periences due to memory decay, telescoping, or misunderstanding the events or legal definitions ( Levine, 1976; Singer, 1979). Furthermore, respondents may purposefully underreport their involvement in crime or victimization due to embarrassment or fear of stigmatization, punishment, or even legal ramifications based on their admissions of criminal offending and/or victimization (Harrison, 1995) The issue of o ver reporting crime or victimization may also plague self report survey data for many of the same accidental reasons (telescoping, misunderstanding events, etc.) and purposeful reasons (over reporting seriousness of crime for sympathy or for a feeling of importance or power) ( Levine, 1976) Although self report data has been criticized some research has documented the accuracy of self report data (Harrison, 1995). Certain techniques employed by researchers have

PAGE 84

84 generated more success in soliciting more accurate survey data, such as offering anonymity and asking respondents to privately respond to su rvey questions rather than responding verbally to interviewers (Harisson, 1995 ; Turner, Lessler & Devore, 1992 ). That the current study is based on jail inmates self reported survey data on crime and victimization presents additional considerations regar ding the accuracy of responses. Jail inmates have been officially accused on some level of having committed crime (and many have been found guilty and sentenced for criminal behavior). Some may even be classified as career criminals who have repeatedly recidivated (and been incarcerated). Therefore, the sample is based on offenders who may not necessarily be characterized by society as truthful. While this may be true, using self report survey data from known offenders may be one of the most valid a nd accurate methods of obtaining information on crime perpetration and victimization for several reasons. First, prior research suggests that incarcerated individuals provide accurate data about their criminal histories ( Peterson et al., 1981). Second, the current study employed methodological procedures known to generate more accuracy such as promising anonymity by refusing to collect identifiers (name, inmate number, etc.) and asking inmates to indicate responses privately rather than verbally (Turner et al., 1992). Third, there were no reward s or punishment s for inmates to falsify their answers by indicating more socially desirable responses. For example, inmate names were not collected, jail staff were not permitted to handle any of the survey material s, and respondents completed the surveys in private while being adequately spaced apart from one another. Therefore, possibilities for inmates to be publically embarrassed about or proud of their perpetration or victimization experiences were reduced give n these procedures. Finally, the results suggest that the many of the inmates admitted perpetrating and being victimized by crime (reducing concerns about underreporting) and the vast majority of

PAGE 85

85 those admitting crime perpetration and/or victimization ind icated that each even t occurred relatively few times (reducing concerns about over reporting) .6 In fact, only approximately a quarter of the sample (27%, n = 634) reported never being victimized by any of the fourteen crime types, which means that 1,780 o f the inmates admitted being a victim of at least one crime. Similarly, 40% of the sample (n = 944) reported never perpetrating any of the fourteen types of crime, meaning that 1,470 inmates admitted that they had committed at least one of the crimes Wh ile the current study is undoubtedly plagued by threats to response accuracy that are inevitable for all types of self report survey research (memory decay, social desirability, etc.), the results can be reasonably considered as trustworthy and accurate as any other self report data. Sample Representativeness Given that th e sample is based on volunteers, it is important to examine the representativeness of the sample in comparison with the population of jail inmates. Chi square tests of association were used to determine significant sex and race differences between the sample and population ( Table s 44 and 4 5) .7 Results indicated many significant sex and race differences between the jail sample and population ; t herefore, the results should be interprete d with caution. Table 4 6 compares offense types for the sample an d population within each jail. While o nly six of the fourteen jails provided offense data for their population, it is important to note that there are many similarities between the populat ion s and sample s 6 Recall that respondents most often indicated they perpetrated each crime type one time (with the exception of attacking someone without a weapon which was most commonly perpetrated three times). Similarly, respondents most often indicated that they were victimized by each crime type once (with the exception of theft which was most commonly experienced twice). 7 Chi square analyses were not calculated for offense types given that many inmates wer e incarcerated for multiple offense types (i.e., property and personal crimes). Given that race was dichotomized for analyses purposes, race was also dichotomized here (White versus non White).

PAGE 86

86 Table 4 1. Characteristics of the Full Sample Non Gang Sample, and Gang Sample Full Sample (n = 2,414) Non Gang Sample (n = 2,044) Gang Sample (n = 370) Sex Male *** 1,746 (75%) 1,492 (73%) 315 (85%) Female 668 (25%) 552 (27%) 55 (15%) Race White 991 (43%) 899 (44%) 137 (37%) Black 956 (41%) 858 (42%) 133 (36%) Other 382 (16%) 287 (14%) 100 (27%) Ethnicity Hispanic *** 478 (21%) 388 (19%) 107 (29%) Age Range 18 84 18 84 18 59 Mean 32 33 28 Median 29 30 25 Mode 21 21 22 (SD) (11.22) (11.44) (8.98) Education 0 to 4 th grade 26 (1%) 23 (1%) 1 (0%) 5 th to 8 th grade 130 (6%) 104 (5%) 24 (7%) 9 th to 11 th grade 676 (29%) 559 (28%) 114 (33%) 12 th grade / GED 825 (35%) 695 (35%) 129 (37%) S ome college 483 (21%) 423 (22%) 58 (17%) C ollege graduate 153 (7%) 135 (7%) 18 (5%) G raduate work 34 (1%) 30 (2%) 4 (1%) Employment Full time 1,125 (49%) 951 (49%) 170 (50%) Part time 333 (15%) 290 (15%) 41 (12%) Seasonally 76 (3%) 67 (3%) 9 (3%) Temporarily 204 (9%) 181 (9%) 22 (6%) Unemployed / Not legally 562 (24%) 461 (24%) 101 (29%)

PAGE 87

87 Table 4.1. (continued) Full Sample (n = 2,414) Non Gang Sample (n = 2,044) Gang Sample (n = 370) Number of days in jail at the time of the s urvey 0 90 days 1,383 (60%) 1,200 (61%) 176 (50%) 91 180 days 416 (18%) 333 (17%) 83 (24%) 181 to 270 days 218 (9%) 178 (9%) 40 (11%) 271 to 365 days 123 (5%) 106 (5%) 17 (5%) 365 or more days 180 (8%) 145 (8%) 35 (10%) Number of days exp ected to spend in jail Unsure 1,158 (50%) 976 (50%) 181 (52%) 0 90 days 507 (22%) 446 (23%) 57 (16%) 91 180 days 215 (9%) 183 (9%) 31 (9%) 181 to 270 days 109 (5%) 89 (5%) 19 (6%) 271 to 365 days 119 (5%) 103 (5%) 16 (5%) 365 or more da ys 200 (9%) 157 (8%) 43 (12%) Reason for incarceration Waiting for trial 779 (34%) 647 (33%) 128 (37%) Sentenced to jail term 442 (19%) 373 (19%) 66 (19%) P robation/parole violation 577 (25%) 489 (25%) 87 (25%) E scaping while on bail 21 (1% ) 18 (1%) 3 (1%) Awaiting transfer to mental health facility 13 (1%) 12 (1%) 1 (0%) A waiting transfer to prison 140 (6%) 109 (6%) 30 (9%) A waiting transfer to other jail 56 (2%) 40 (2%) 16 (5%) C ontempt of court 40 (2%) 34 (2%) 5 (1%) R ele ased from prison 11 (1%) 10 (1%) 1 (0%) C ourt witnesses 19 (1%) 17 (1%) 2 (1%) Other reasons 649 (28%) 539 (28%) 108 (31%) Offense type charged with or convicted of Property crime 582 (25%) 484 (25%) 95 (27%) Personal crime 526 (23%) 414 (21 %) 111 (32%) Drug crime 642 (28%) 530 (27%) 110 (31%) Other crime 733 (32%) 624 (32%) 107 (31%) None 141 (6%) 124 (6%) 15 (4%) p < .05, ** p < .01, *** p < .001

PAGE 88

88 Table 4 2. Crime Perpetration Descriptive Statistics Perpetration Crime Type Min imum Maximu m Mean Std. Dev. Property Crimes Theft 0 2,000 6.87 59.63 Vandalism 0 2,000 2.79 42.80 Personal Crimes Robbery 0 2,000 2.50 47.51 Threatened with weapon 0 1,000,015 437.59 20,847.21 Attacked without weapon 0 1,001,100 44 8.10 20,975.85 Attacked with weapon 0 250 1.54 9.84 Sexual assault/rape 0 100 .094 2.16 Stab 0 87 .32 2.31 Carjack 0 105 .28 2.77 Witness intimidation 0 100 .19 2.49 Home invasion 0 101 .47 3.21 Drive by shooting 0 1,000,000,000 429,18 5.00 20,716,769.75 Shot at 0 152 .81 5.81 Shot 0 304 .52 7.05 Crime Perpetration Indexes a Property Crime Perpetration 0 120 5.69 16.85 Personal Crime Perpetration 0 225 9.75 28.83 Combined Crime Perpetration 0 349 16.41 45.95 a Trunca ted at the 99th percentile

PAGE 89

89 Table 4 3. Crime Victimization Descriptive Statistics Victimization Crime Type Minimum Maximu m Mean Std. Dev. Property Crimes Theft 0 300 2.38 10.89 Vandalism 0 100 1.02 3.11 Personal Crimes Robbery 0 100 .7 0 3.25 Threatened with weapon 0 100010 446.74 21093.30 Attacked without weapon 0 1200 3.61 33.99 Attacked with weapon 0 200 1.30 5.79 Sexual assault/rape 0 2500 1.46 51.96 Stab 0 22 .40 1.25 Carjack 0 20 .07 .57 Witness intimidation 0 30 .11 .92 Home invasion 0 50 .23 1.65 Drive by shooting 0 1,000,000 431.02 20747.95 Shot at 0 1,000,000,000 433146.7 20801758.78 Shot 0 50 .21 1.49 Crime Victimization Indexes a Property Crime Victimization 0 27 2.71 4.52 Personal C rime Victimization 0 107 8.14 15.42 Combined Crime Victimization 0 131 11.20 19.51 a Truncated at the 99th percentile

PAGE 90

90 Table 44. Sex Comparison of County, J ail Population, and Jail Sample Jail Male Female Jail Population and Sample 2 Difference County Jail Population Jail Sample County Jail Population Jail Sample Alachua 49% 88% 53% 51% 12% 47% 81.82*** Broward 49% 53% 73% 51% 47% 27% 26.41*** Collier 51% 87% 84% 49% 13% 16% .89 Duval 48% 89% 90% 52% 11% 10% .49 Esca mbia 50% 87% 100% 50% 13% 0% .28 Hillsborough 49% 88% 76% 51% 12% 24% 16.23*** Lee 49% 73% 68% 51% 27% 32% 1.17 Leon 48% 88% 68% 52% 12% 32% 52.91*** Miami Dade 48% 80% 70% 52% 20% 30% 12.35*** Palm Beach 49% 90% 54% 51% 10% 46% 153.94*** Pasco 48% 7 8% 62% 52% 22% 38% 5.98* Pinellas 48% 86% 66% 52% 14% 34% 44.85*** Polk 49% 85% 60% 51% 15% 40% 78.5*** Seminole 49% 87% 81% 51% 13% 19% 3.01 p < .05, ** p < .01, *** p < .001 Chi square examines significant differences between the jail population a nd the jail sample. Table 45. Race Comparison of County, Jail Population, an d Jail Sample Jail White Non White Jail Population and Sample 2 Difference County Jail Population Jail Sample County Jail Population Jail Sample Alachua 73% 33% 41% 27% 67% 59% 2.32 Broward 70% 38% 40% 30% 62% 60% .23 Collier 92% 84% 55% 8% 16% 45% 61.66*** Duval 64 % 33% 35% 36% 67% 65% .93 Escambia 71 % 42% 24% 29% 58% 76% 1.68 Hillsborough 78 % 34% 47% 22% 66% 53% 8.69** Lee 90% 73% 53% 10% 27% 47% 17.07*** Leo n 6 6% 29% 32% 34% 71% 68% .69 Miami Dade 77 % 39% 30% 23% 61% 70% .698** Palm Beach 80% 50% 56% 20% 50% 44% .90 Pasco 9 3% 83% 69% 7% 17% 31% 5.45* Pinellas 85 % 54% 55% 15% 46% 45% .05 Polk 8 3% 63% 59% 17% 37% 41% 1.17 Seminole 83 % 59% 46% 17% 41% 54% 7.2** p < .05, ** p < .01, *** p < .001 Chi square examines significant differences between the jail population and the jail sample.

PAGE 91

91 Table 46. Offense Types for the Jail Population and Sample Jail Property Crime Personal Crime Drug Crime Other Crime Population Sample Population Sample Population Sample Population Sample Alachua 26% 10% 28% 47% Broward 18% 12% 44% 32% Collier 17% 16% 20% 18% 13% 23% 50% 46% Duval 25% 21% 28% 31% 22% 22% 25% 29% Escambia 21% 29% 25% 32% Hi llsborough 23% 21% 19% 9% 25% 33% 33% 31% Lee 29% 6% 36% 30% Leon 28% 34% 33% 26% 10% 20% 29% 30% Miami Dade 22% 34% 21% 27% Palm Beach 33% 31% 22% 30% Pasco 21% 24% 19% 20% 60% 51% 0% 24% Pinellas 31% 14% 42% 26% Pol k 19% 29% 29% 10% 10% 27% 42% 40% Seminole 26% 18% 35% 36% p < .05, ** p < .01, *** p < .001 Chi square comparisons cannot be made between the population and the sample offense types due to differences in calculations. Fo r example, the populat ion data categorizes inmates into single categories (totaling 100%) whereas the sample data allowed inmates to report perpetrating multiple crime types (totaling over 100%). Alachua county jail offense data is unavailable given that the information is not housed electronically and the request would take more time than the staff can provide (Personal communication, Maggie Donnell, January 13, 2009). Broward county jail offense data is unavailable (Personal communication, Classification Manager Darren Siege r, March 9, 2009). Escambia county jail did not provide information regarding their facility in time for the publication of this research Lee county jail offense data is unavailable in the database (Personal communication Sergeant David Velez, January 1 4, 2009). Miami Dade county jail offense data is only kept in terms of classification (i.e., misdemeanors, felonies, etc.) rather than types of charges (Personal communication, Janelle Hall, February 3, 2009). Palm Beach county jail offense data is colle cted, however this request would require nearly a full day and the staffing for this request is unavailable (Personal communication, Lieutenant Brenton, January 29, 2009. Pinellas county jail offense data is not collected (Personal communication, Ramona S chaefer, January 9, 2009). Seminole county jail offense data is only kept in terms of classification (i.e., misdemeanors, felonies, etc.) rather than types of charges (Personal communication, Captain Manley, January 29, 2009).

PAGE 92

92 CHAPTER 5 RESULTS PREDICTIN G GANG MEMBERSHIP Logistic regression models were performed for analyses predicting gang membership given the dichotomous nature of the dependent variable (Long, 1997) Similar to ordinary least squares step wise analyses, each model presented in this Ch apter progressively incorporated additional independent variables. Six tables each comprise d the same five models with the exception of one independent variable change (the crime perpetration/victimization variable). The independent variables a lternating within the tables were property crime perpetration (Table 51), personal crime perpetration (Table 5 2), combined crime perpetration (Table 5 3), property crime victimization (Table 5 4), personal crime victimization (Table 5 5), and combined crime victimization (Table 5 6). For each of the tables, Model 1 included demographic variables only, Model 2 included a crime variable ( altering among property crime perpetration, personal crime perpetration, combined crime perpetration, property crime victimizatio n, personal crime victimization, and combined crime victimization), Model 3 included the measure o f self control, Model 4 replaced self control with social disorganization variables, and Model 5 is the full model that include d all variables from Models 1 t hrough 4 (examining both self control and social disorganization). The following section presents the results from analyses predicting gang membership using the crime perpetration variables as independent variables and the second section presents the find ings using the crime victimization variables as predictor variables Crime Perpetration and Gang Membership Table 5 1 presents the results from the regression models predicting gang membership using property crime perpetration as an independent variable i n all but the first model (given that the first model presents only the demographic variables) When the demographic variables we re

PAGE 93

93 examined (Model 1), sex, ethnicity, and age were significantly associated with gang membership. Males, Hi spanics, and younger inmates were significantly more likely to admit belonging to a gang than females, non Hispanics, and older inmates. Mode l 1 also suggest ed there were no significant racial differences among gang members. When compared to the subsequent models (descri bed next), the model with only the demographic variables explain ed the least amount of variance (nearly 7% of the variance). Models 2 through 5 examine d the effect of property crime perpetration, which tested Hypothesis 1. Hypothesis 1 Supported: Crime P e rpetration I ncreases the L ikelihood of Gang M embership Findings from Model 2 predict ed gang membership using the demographic variables and property crime perpetration as independent variables. The demographic variables remain ed unchanged between Models 1 and 2 and the findings indicate d th at property crime perpetration wa s significantly and positively related to gang membership. In support of Hypothesis 1, this finding suggest ed i nmates who admitte d committing property crimes were significantly more lik ely to report belonging to a gang. Furthermore, adding property crime perpetration slightly increased the explained variance to 8% (from 7%). Hypothesis 2 Supported: Low Self Control Increases the Likelihood of Gang M embership While none of the demographi c variables changed from Model 1 to Model 2, including self control in Model 3 revealed that all demographic variables were significant (including race). Mo del 3 suggested that non Whites we re more likely than Whites to be gang members. The results also provide d su pport for Hypothesis 2 given that low self control wa s associated with gang membership. This means that gang members were more likely to have lower self control than non gang members. Interestingly, adding self con trol to the model nearly doub led the explained variance (to 13%).

PAGE 94

94 Hypothesis 3 Partially Supported: Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Gang Members hip When self control was replaced with social disorganization variables, race was again not si gnificant; however, two of the social disorganization measures were predictive of gang membership including social disorder and collective efficac y (see Model 4). Consistent with social disorganization theory inmates who perceived more social disorder a nd l es s collective efficacy within their neighborhoods we re significantly more likely to be gang members. This finding partially supported Hypothesis 3 given that two social disorganization variables were significantly related to gang membership (social d isorder and collective efficacy) while many were not significant (including physical disorder and all of the neighborhood characteristic variables) For gang membership, less variance was explained using social disorganization theory than with self contro l theory (11.6% versus 13.1%, respectively). Given that the theoretical variables of self cont rol and social disorganization we re significantly related to gang membership when examined individually, it is of interest to examine both theories in the same mo del to determine whether one theory eliminates the significance of the other. Interestingly, when both theoretical perspectives we re examined in the same model (Model 5) the same self control and so cial disorganization variables we re significantly predic tive of gang membership (w hile the demographic variables we re unchanged). In other words, both theories are correct in their abilities to predict gang membership. In fact, the R2 was the highest when both theories we re included in the model (explaining n early 16% of the variance). While the findings we re supportive of both theories, it is important to point out that Gottfredson and Hirschi (1990) hypothesize d that low self control is not only predictive of crime and analogous behavior, but that it is the only predictor (meaning all other theoretical explanations should cease to be significant). Clearly, social disorganization theory is also

PAGE 95

95 important when predicting gang membership. While this does not negate Gottfredson and Hirschis (1990) theory give n that self control was significant, these findings do call into question the magnitude of the predictive ability of self control. In other words, the current study indicated that self control is important, but that it is not the only important theoretica l explanation for gang membership. The next two tables examine d personal crime perpetration (Table 5 2) and combined crime perpetration (Table 5 3) as independent variables predictive of gang membership. Results from each of the five models in the two ta bles we re substantively identical to the findings presented in Table 51 and described above. More specifically, inmates who admit ted committing property crimes, personal crimes and crime in general (combined property and personal crimes) we re significan tly more likely to be gang members. Again, this supports Hypothesis 1, which suggested that crime perpetration was predictive of gang membership. Furthermore, self control and perceptions of social disorganization (social disorder and col lective efficacy specifically) we re significantly predictive of gang membership while controlling for personal crime perpetration and the combined crime perpetration variables. As with the models employing property crime perpetration, these findings support ed Hypotheses 2 ( given that low self control was predictive of gang membership) and offer ed partial support for Hypothesis 3 (given that social di sorder and collective efficacy were predictive of gang membership while physical disorder, poverty, unemployment, residentia l mob ility and racial heterogeneity we re not). Similar to the explained variance for the models with property crime perpetration, the R2 was higher for self control in comparison to social disorganization. Finally, when both theoretical perspectives we re examined in the same model, both self control and socia l disorganization we re indicative o f gang membership, which revealed that both theories

PAGE 96

96 we re important for explaining gang membership. Consistent with this finding, the explained variance was highest when both theories we re included in the model. Crime Victimization and Gang Membership Similar to the three tables alternating crime perpetration (Tables 5 1 through 53 above ), the following three tables present models predicting gang membership by alter nat ing the crime victimization independent variables, including property crime victimization (Table 5 4), personal crime victimization (Table 5 5), and combined crime victimization (Table 5 6). Interestingly, findings from the models examining crime victimization (see Tables 5 4 through 56) we re identical to each other In other words, property crime victimization be haved in the same ways as personal crime victimization and combined crime victimization with respect to the variables included in the models Hypothesis 4 Supported: Crime Victimization Increases the Likelihood of Gang Membership Not only do the three t ables using victimization variables reveal substantively identical findings to each other but these findings also mirror ed the findings from the three crime perpetration analyses (see Tables 5 1 through 53). In terms of the demographic variables m ales, Hispanics, and young er inmates we re more likely to belong to gang s than females, non Hispanics, and older inmates In support of Hypothesis 4, crime victimization ( property persona l and combined ) was predictive of gang membership, meaning that crime victims we re significantly more likely to be gang members than inmates who d id not report victimization. Furthermore, inmates with lowe r self c ontrol, and inmates who perceived more social disorder and less collective efficacy within their neighborhoods we re significantly more likely to belong to a gang. These findings contribute d additional support f or Hypotheses 2 and 3, meaning that low self control and percept ions of social disorganization we re important for predicting gang

PAGE 97

97 membership. Similar to the models using crime perpetration, self control explained more of the varia nce than social disorganization; however, the models with both self co ntrol and social disorganization had the highest explained variance (R2 = 15% with propert y crime victimization, 18% with personal crime vic timization, and nearly 18% with combined crime victimization).1 1 When only the most robust social disorganization variable w as included into the full model (instead of all seven social disorganization variables), self control and social disorganization were statistically significant predictors of gang membership when property perpetration, personal perpetration, combined perpet ration, property victimization, personal victimization, and combined victimization were included separately as independent variables (tables not presented). The most robust social disorganization variables were social disorder and collective efficacy; mod els were run separately for both variables.

PAGE 98

98 Table 5 1. Logistic Regression Predicting Gan g Membership with Property Crime Perpetration as an Independent Variable Model 1: Demographics Model 2: Property Perpetration Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables b AOR b AOR b AOR b AOR b AOR Male ***.654 (.1 61) 1.923 *** .611 (.164) 1.842 ***.625 (.166) 1.869 *** .600 (.167) 1.823 ***.636 (.170) 1.889 White .115 (.125) .892 .200 (.128) .819 *** .354 (.132) .702 .092 (.132) .912 .264 (.137) .768 Hispanic *** .459 (.137) 1.582 *** .465 (.140) 1.591 *** .511 (.143) 1.666 *** .479 (.146) 1.615 *** .521 (.149) 1.684 Age *** .042 (.006) .959 *** .040 (.006) .961 *** .035 (.007) .965 *** .036 (.007) .965 *** .032 (.007) .969 Property Perpetration ***.013 (.003) 1.013 *** .009 (.003) 1.010 *** .012 (.003) 1.012 ** .009 (.003) 1.009 Self Control *** 1.046 (.129) .351 *** .980 (.136) .375 Physical Disorder .049 (.134) 1.051 .053 (.137) 1.054 Social Disorder ** .400 (.138) 1.492 ** .357 (.138) 1.429 Collective Efficacy *** .260 (.080) .771 .176 (.082) .838 Poverty .005 (.055) .995 .011 (.057) .989 Unemployed .102 (.060) 1.107 .080 (.061) 1.083 Residential Mobility .074 (.064) .929 .081 (.065) .922 Racial Heterogeneity .116 (.074) 1.123 .125 (.075) 1.133 Nagelkerke R 2 .069 .081 .131 .116 .156 N 2,250 2,207 2,206 2,163 2,162 *p < .05, **p < .01, ***p < .001; AOR = Adjusted Odds Ratio; B = Logistic R egression Coefficients; Standard Errors in Parentheses

PAGE 99

99 Table 5 2. Logistic Regression Predictin g Gang Membership with Personal Crime Perpetration as an Independent Variable Model 1: Demographics Model 2: Personal Perpetration Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables b AOR B AOR b AOR b AOR b AOR Male ***.654 (.161) 1.923 *** .528 (.163) 1.696 ***.550 (.166) 1.732 ** .525 (.167) 1.691 ***.563 (.169) 1.756 White .115 (.125) .892 .139 (.128) .870 .283 (.132) .754 .036 (.132) .964 .192 (.136) .825 Hispanic *** .459 (.137) 1.582 *** .492 (.140) 1.635 *** .551 (.143) 1.734 *** .507 (.145) 1.659 *** .563 (.148) 1.756 Age *** .042 (.006) .959 *** .038 (.007) .963 *** .033 (.007) .967 *** .034 (.007) .967 *** .030 (.007) .970 Personal Perpetration ***.013 (.002) 1.013 *** .012 (.002) 1.012 *** .013 (.002) 1.013 ** .011 (.002) 1.011 Self Control *** .980 (.129) .375 *** .912 (.135) .402 Physical Disorder .061 (.135) 1.063 .074 (.137) 1.077 Social Disorder ** .369 (.138) 1.447 .316 (.138) 1.372 Collective Efficacy ** .240 (.081) .786 .164 (.083) .848 Poverty .006 (.055) .994 .013 (.057) .987 Unemployed .108 (.060) 1.114 .087 (.061) 1.090/ Residential Mobility .067 (.064) .936 .075 (.065) .928 Racial Heterogeneity .118 (.074) 1.125 .122 (.075) 1.130 N agelkerke R 2 .069 .117 .160 .149 .182 N 2,250 2,243 2,242 2,198 2,197 *p < .05, **p < .01, ***p < .001; AOR = Adjusted Odds Ratio; B = Logistic Regression Coefficients; Standard Errors in Parentheses

PAGE 100

100 Table 5 3. Logistic Regression Predicting G ang Membership with Combined Crime Perpetration as an Independent Variable Model 1: Demographics Model 2: Combined Perpetration Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables b AOR b AOR b AOR b AOR b AOR Male ***.654 (.1 61) 1.923 *** .565 (.163) 1.759 ***.582 (.165) 1.789 *** .561 (.167) 1.753 ***.595 (.169) 1.814 White .115 (.125) .892 .152 (.127) .859 .299 (.131) .742 .042 (.131) .959 .203 (.135) .816 Hispanic *** .459 (.137) 1.582 *** .502 (.139) 1.652 *** .559 (. 142) 1.749 *** .520 (.144) 1.683 *** .574 (.147) 1.775 Age *** .042 (.006) .959 *** .039 (.007) .962 *** .035 (.007) .966 *** .035 (.007) .965 *** .031 (.007) .969 Combined Perpetration ***.007 (.001) 1.007 *** .006 (.001) 1.006 *** .007 (.001) 1.007 *** 006 (.001) 1.006 Self Control *** .998 (.128) .368 *** .928 (.135) .395 Physical Disorder .076 (.134) 1.079 .087 (.136) 1.091 Social Disorder ** .361 (.137) 1.435 .311 (.138) 1.365 Collective Efficacy ** .252 (.080) .777 174 (.082) .840 Poverty .002 (.055) .998 .011 (.056) .990 Unemployed .110 (.060) 1.116 .088 (.061) 1.092 Residential Mobility .076 (.064) .927 .084 (.065) .920 Racial Heterogeneity .113 (.074) 1.120 .117 (.075) 1.124 Nage lkerke R 2 .069 .104 .149 .137 .172 N 2,250 2,243 2,242 2,198 2,197 *p < .05, **p < .01, ***p < .001; AOR = Adjusted Odds Ratio; B = Logistic Regression Coefficients; Standard Errors in Parentheses

PAGE 101

101 Table 54. Logistic Regression Predicting Gan g Membership with Property Crime Victimization as an Independent Variable Model 1: Demographics Model 2: Property Victimization Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables b AOR b AOR b AOR b AOR b AOR Male ***.654 ( .161) 1.923 *** .687 (.165) 1.987 ***.701 (.167) 2.016 *** .671 (.168) 1.957 ***.709 (.171) 2.031 White .115 (.125) .892 .155 (.127) .857 .317 (.131) .728 .032 (.131) .969 .212 (.136) .809 Hispanic *** .459 (.137) 1.582 *** .501 (.139) 1.650 ***.557 (.143) 1.746 *** .529 (.145) 1.697 *** .581 (.148) 1.787 Age *** .042 (.006) .959 *** .043 (.007) .958 *** .038 (.007) .963 *** .038 (.007) .963 *** .034 (.007) .967 Property Victimization ***.042 (.012) 1.043 *** .041 (.012) 1.042 ** .038 (.012) 1.039 ** .039 (.013) 1.040 Self Control *** 1.083 (.128) .339 *** 1.010 (.135) .364 Physical Disorder .102 (.134) 1.108 .116 (.137) 1.123 Social Disorder ** .385 (.138) 1.470 .330 (.138) 1.391 Collective Efficacy *** .271 (.081) .762 .176 (.083) .838 Poverty .004 (.055) 1.004 .004 (.057) .996 Unemployed .093 (.060) 1.097 .071 (.062) 1.074 Residential Mobility .089 (.064) .915 .096 (.065) .908 Racial Heterogeneity .104 (.074) 1.109 .107 (.075 ) 1.113 Nagelkerke R 2 .069 .078 .134 .115 .158 N 2,250 2,189 2,188 2,145 2,144 *p < .05, **p < .01, ***p < .001; AOR = Adjusted Odds Ratio; B = Logistic Regression Coefficients; Standard Errors in Parentheses

PAGE 102

102 Table 5 5. Logistic Regression Predicting Gang Membership with Personal Crime Victimization as an Independent Variable Model 1: Demographics Model 2: Personal Victimization Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables b AOR b AOR b AOR b AOR b AOR Male ***.6 54 (.161) 1.923 *** .568 (.164) 1.764 ***.586 (.166) 1.797 *** .563 (.167) 1.756 ***.601 (.170) 1.823 White .115 (.125) .892 .148 (.128) .863 .307 (.132) .736 .045 (.132) .956 .221 (.136) .802 Hispanic *** .459 (.137) 1.582 *** .488 (.140) 1.629 *** 550 (.143) 1.733 *** .508 (.145) 1.661 *** .563 (.148) 1.756 Age *** .042 (.006) .959 *** .040 (.007) .960 *** .036 (.007) .965 *** .036 (.007) .965 *** .032 (.007) .968 Personal Victimization ***.025 (.003) 1.025 *** .023 (.003) 1.024 *** .024 (.003) 1.0 24 *** .023 (.003) 1.023 Self Control *** 1.046 (.128) .351 *** .975 (.134) .377 Physical Disorder .041 (.135) 1.042 .058 (.138) 1.059 Social Disorder ** .378 (.138) 1.460 .320 (.138) 1.377 Collective Efficacy *** .274 (.081) .760 .189 (.083) .827 Poverty .005 (.056) .995 .013 (.057) .987 Unemployed .098 (.060) 1.102 .078 (.062) 1.081 Residential Mobility .067 (.064) .935 .075 (.065) .928 Racial Heterogeneity .122 (.074) 1.129 .126 (.075) 1. 134 Nagelkerke R 2 .069 .113 .162 .145 .184 N 2,250 2,247 2,246 2,202 2,201 *p < .05, **p < .01, ***p < .001; AOR = Adjusted Odds Ratio; B = Logistic Regression Coefficients; Standard Errors in Parentheses

PAGE 103

103 Table 5 6. Logistic Regression Predic ting Gang Membership with Combined Crime Victimization as an Independent Variable Model 1: Demographics Model 2: Combined Victimization Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables b AOR b AOR b AOR b AOR b AOR Male **.654 (.161) 1.923 *** .588 (.163) 1.800 ***.607 (.166) 1.835 *** .582 (.167) 1.789 ***.619 (.169) 1.857 White .115 (.125) .892 .149 (.127) .862 .308 (.131) .735 .043 (.131) .958 .218 (.136) .804 Hispanic *** .459 (.137) 1.582 *** .498 (.140) 1.645 *** .558 (.143) 1.748 *** .520 (.145) 1.682 *** .575 (.148) 1.777 Age *** .042 (.006) .959 *** .041 (.007) .959 *** .037 (.007) .964 *** .037 (.007) .964 *** .033 (.007) .9 67 Combined Victimization ***.018 (.003) 1.018 *** .017 (.003) 1.017 *** .017 (.003) 1.017 *** .016 (.003) 1.017 Self Control *** 1.049 (.128) .350 *** .978 (.134) .376 Physical Disorder .042 (.135) 1.043 .057 (.137) 1.059 Social Disorder ** .376 (.137) 1.457 .320 (.138) 1.377 Collective Efficacy *** .280 (. 081) .756 .196 (.082) .822 Poverty .002 (.056) .998 .011 (.057) .989 Unemployed .095 (.060) 1.099 .075 (.062) 1.078 Residential Mobility .072 (.064) .931 .080 (.065) .923 Racial Heterogeneity .111 (.074) 1.117 .115 (.075 ) 1.122 Nagelkerke R 2 .069 .106 .157 .139 .179 N 2,250 2,248 2,247 2,203 2,202 *p < .05, **p < .01, ***p < .001; AOR = Adjusted Odds Ratio; B = Logistic Regression Coefficients; Standard Errors in Parentheses

PAGE 104

104 CHAPTER 6 RESULTS PREDICTING C RIME PERPETRATION Regression a nalyses were estimated for each of the crime perpetration dependent variables (property crime, personal crime, and combined crime perpetration) for the full sample and for the gang and non gang samples separately Negative binomial regression was employed given that the dependent variables (crime perpetration) represented count data (number of times the crime types were committed). Results from negative binomial regressions are interpreted the same way as ordinary least squares ( Hilbe, 2007). The following sections present the results from the negative binomial regressions for each of the dependent variables, beginning with property crime perpetration, then personal crime perpetration, and finally combined crime perpetration.1 E ach of the three sect ions presents two tables (totaling six tables), one table provides results for the full sample and the other table d isplays results for the gang versus the non gang samples. While self control theory suggests that differences should n ot be observed between gang and non gang members, these samples were split and analyzed separately in order to test differences between the groups. Each table included several models, similar to the tables presented in Chapter 5 predicting gang membersh ip. For the tables presenting results for the full sample, Model 1 include d demographic variables only, Model 2 included gang membership, Model 3 included the measure of selfcontrol, Model 4 replaced self control with social disorganization variables, an d Model 5 included all variables from Models 1 through 4 (examining both self control and social disorganization). For the tables comparing the gang versus non gang models the same models are presented, however gang membership was removed as a n independe nt variable ( given that 1 Unlike the logistic regression models presented in Chapter 5, R2 is not discussed with regard to negative binomial regression (McCullagh & Nelder, 1989). The R2 statistic is not reported with SPSS (the statistical program used for these analyses) or with other statistical programs (e.g., SAS).

PAGE 105

105 the separate models we re split by gang membership). Therefore, for the gang versus n on gang tables, Model 1 included demographics only, Model 2 added self control, Model 3 replaced self control with social disor ganization, and Mode l 4 included all variables of interest (including self control and social disorganization). The following sections describe the findings from the full sample with the results from the split samples (gang members versus non gang members). Property Crime Pe rpetration Table 6 1 displays the five models predicting property crime perpetration for the full sample. Results from Model 1 (Table 6 1) indicate d all demographic variables we re statistically significant Specifically, males, Whites, non Hispanics, and younger inmates we re significantly more likely to perpetrate property crimes than women, non Whites, Hispanics, a nd older inmates Because gang membership was a statistically significant predictor of property crime perpetration, it was important to examine the relationships between the dependent and independent variables for gang members compared with non gang members. Comparing gang with non gang members allows for a more comprehensive understanding of the differences between groups. Table 6 2 presents results from the split models predicting property crime perpetration for the gang versus non gang samples. Results indicated some interesting similarities and differences between groups. When only demographic variables were examined, Model 1 (Table 6 2) suggested that ethnicity and age were significant for gang members while sex, race, and ethnicity were significant for non gang members. Specifically, non Hispanics and younger gang members were significantly more likely to commit property crimes while m ales, Whites, and non Hispanic non gang members were more likely to commit property offenses. Comparing the sp l it models to the full model (where all demographic variables were significant) suggests (1) that non gang members were driving the significant s ex and race findings for the full sample, (2) that gang members were driving the significant age findings for the full sample,

PAGE 106

106 and (3) that both gang and non gang members were driving the significance of ethnicity for the full sample. Hypothesis 5 Supporte d: Gang Membership Increases the Likelihood of Perpetrating Crime When gang membership was entered into Model 2 ( Table 6 1) the statistically significant nature of the relationships between the demographic variables and property crime perpetration we re un changed for the full sample In support of Hypothesis 5 results indicate d that gang members we re significantly more likely to commit property offenses in comparison with n on gang members. This finding was similar to that presented in Chapter 5 predictin g gang membership, such that gang membership and property crime perpetration were significantly related. Hypothesis 6 Supported: Low S elf Control Increases the Likelihood of Perpetrating Crim e Next, variables meas uring the theories of interest we re ent ered separately into the following two models Results from Model 3 ( Table 6 1) reveal ed a negative and statistically significant relationship between self control and property crime perpetration, which indicated i nmates with lower self control we re signi ficantly more likely to commit property crimes. This finding was consistent with Hypothesis 6 and also supports Gottfredson and Hirschis (1990) general theory of crime. With the exception of age no longer reaching significance the other variables in th e model appear ed to be unaffected by the addition of self control (e.g., they remain ed statistically significant) According to Gottfredson and Hirschi (1990), self control is the cause of crime; therefore, the theorists argue its presence in the model sh ould render other predictor variables (i.e., gang membership) non significant. In other words, the theorists would argue that self control, not gang membership, was predictive of property crime perpetration. However, similar to the work of some prior res earch (Cauffman, Steinberg, & Piquero, 2005),

PAGE 107

107 results show ed that t he presence of self control did not eliminate the relationship between gang membership and property crime offending. Table 6 2 Model 2 added self control to the split analyses, which was significant for both gang and non gang members (without altering the significance of the demographic variables). Clearly, both the gang and non gang samples influenced the significance of self control within the full sample. Hypothesis 7 Partially Supporte d: Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Perpetrating Cri me Table 6 1 Model 4 replaced self control with social disorganization variables. A ll demographic variables were significant and gang membership remain ed unaf fected by the addition of the social disorganization variables. All but one of the social disorganization variables ( residential mobility) we re significantly associated with property crime perpetration. Of the statistically significant social disorganiza tion variables, all but two (physical disorder and neighborhood unemployment ) operate d in accordance with theoretical expectations. Consistent with social disorganization theory, i nmates who reported less collective efficacy more social disorder, more ne ighborhood poverty, and more racial heterogeneity we re significantly more likely to commit property crimes. However, p roperty crime pe rpetration was negatively associated with physical disorder and neighborhood unemployment meaning that less physical dis order and less neighborhood unemployment we re predictive of property crime. While two of the social disorganization variables operate d differently from theoretical expectations, the findings reveal ed overall su pport for Hypothesis 7. Table 6 2 Model 3 f or the split samples substituted self control with social disorganization and findings revealed that several of the social disorganization variables were significant while the demographic variables were unchanged for both the gang and non gang samples. Am ong gang members, physical disorder, social disorder, and neighborhood unemployment were

PAGE 108

108 significant. In addition to these variables, collective efficacy, poverty, and racial heterogeneity were also significant for non gang members. Similar to the models predicting gang membership in Chapter 5, physical disorder and neighborhood unemployment were negatively associated with property crime perpetration for both gang and non gang members. This means that inmates (gang and non gang members) who perceived les s physical disorder and less employment within the ir neighborhoods were more likely to commit property crimes than inmates who perceived more physical disorder and higher unemployment. All other social disorganization variables relate d to property crime perpetration in expected directions according to social disorganization theory. Comparing Self Control Theory and Social Disorganization Theory Predicting Property Crime Perpetration Given that self control and social disorganization were predictive of prop erty crime perpetration in separate models, it was important to examine the theories in the same model to determine the ex tent to which one theory eliminated the effect of the other theory. Table 6 1 M odel 5 combined all variables (including both self con t rol and social disorganization) and the f indings indicate d that demographic variables remain ed statistically associated with property crime perpetration (with the exception of age which no longer reached significance ) Additio nally, gang membership conti nued to be positively related to property crime offending (again supporting Hypothesis 5, mentioned above) The theoretical variables that indicat ed significance in Table 6 1 Models 3 (self control only) and 4 (social disorganization only) continue d to be significant in the final model. More specifically, both self control and social disorganization variables we re predictive of property crime perpetration. This means that both theories we re important for explaining property crime perpetration. This also means that

PAGE 109

109 examining t he two theories together revealed important relationship s between property crime offending and self control and property crime offending and social disorganization Table 6 2 Model 4 examined all variables (including both self co ntrol and social disorganization) among gang and non gang members Findings were identical to Models 2 (self control only) and 3 (social disorganization only) for gang and non gang members with the exception of poverty which no longer reached significance for non gang members and racial heterogeneity which no longer reached significance for gang members. While both the gang and non gang samples influenced the significance of many social disorganization variables within the full sample (physical disorder, social disorder, and neighborhood unemployment), the non gang sample was responsible for the significance of collective efficacy, poverty, and racial heterogeneity within the full sample. Overall, self control and many of the so cial disorganization variab les we re predictive of property crime perpetration among both gang and non gang members.2 Personal Crime Perpetration Table 6 3 shows results from the negative binomial regression predicting personal crime perpetration for the full sample. The following d escribes the findings presented in Table 63 and comparisons are made between the findings from the full sample predicting personal crime perpetration and (1) t he findings from the split models (Table 6 4) comparing gang versus non gang members and (2) the findings from the models predicting property crime perpetration (Tables 6 1 and 62) 2 Models were reanalyzed using the most robust social disorganization variable (social disorder and unemployment, separately). When only the most robust social disorgani zation variable (social disorder) was included into the full model (instead of all seven social disorganization variables), self control was a statistically significant predictor of property crime perpetration for both gang and non gang members, yet social disorder was only significant for gang members. When only unemployment was included in the full model, self control was a statistically significant predictor of property crime perpetration for both gang and non gang members, yet unemployment was only si gnificant for non gang members (tables not presented).

PAGE 110

110 Table 6 3 Model 1 shows that sex, ethnicity, and age were significantly related to personal crime perpetration for the full sample meaning that males, non Hispanics, and younger inmates were significantly more likely to commit personal crimes than females, Hispanics, and older inmates. Table 6 4 Model 1 indicated substantively identical results between the samples: sex, ethnicity, and age were significantly related to personal crime perpetration for both gang and non gang members. This mirrored the findings for the full sample presented in Table 6 3. More specifically, males, non Hispanics, and younger gang and non gang members were more likely to report committing p ersonal crimes than females, Hispanics, and older inmates. Hypothesis 5 Supported: Gang Membership Increases the Likelihood of Perpetrating Crime When gang membership was added to the model, the demographic variables were unchanged and findings reveal ed that gang members were significantly more likely than non gang members to perpetrate personal crimes (see Table 6 3 Model 2). This finding, which was consistent to the findings predicting property c rime perpetration, supported Hypothesis 5 Hypothesis 6 Supported: Low Self Control Increases the Likelihood of Perpetrating Crime Adding self control to the model revealed that all variables were statistically significant for the full sample Unlike Models 1 and 2 (Table 6 3) race became significant such t hat non Whites were more likely than Whites to report committing personal crimes. In support of Hypothesis 6, inmates with lower self control were more likely to commit personal crimes than inmates with higher self control This finding was identical to that for property crime perpetration, where as the effects of self control we re significant, yet no t powerful enough to render other variables ( i.e., gang membership) as non significant. More specifically, both gang membership and low self control we re pre dictive of personal crime perpetration.

PAGE 111

111 Table 6 4 Model 2 revealed that low self control was predictive of personal crime perpetration for both gang and non gang members. However, adding self control to the model changed two of the demographic variables for non gang members given that race became significant and ethnicity ceased to reach significance. More specifically, non White non gang members were significantly more likely to report personal crime perpetration than Whites. Examining gang and non ga ng members separately revealed the driving force behind the significance of factors in the models for the full sample. Non gang members drove the significant race findings for the full sample and gang members drove the significant ethnicity findings. Ove rall, the split models suggested that among gang members, males, non Hispanics, younger inmates, and those with lower self contro l were more likely to commit personal crimes and among non gang members, males, n on Whites, younger inmates, those with low er s elf control were more likely to perpetrate personal crimes. Hypothe s is 7 Partially Supported : Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Perpetrating Crime Table 6 3 Model 4 replaced self control with social disorganiza tion variables. While race became no longer significant (as Models 1 and 2 show ed ) s everal of the social disorganization variables were significantly related to personal crime perpetration, including physica l disorder, social disorder, collective efficac y, and racial heterogeneity Consistent with social disorganization theory, less collective efficacy and more social disorder were predict ive of personal crime offending. However, two of the social disorganization variables (physical disorder and racial heterogeneity) were associated with personal crime perpetration in ways that were not consistent with social disorganization theory. As with property crime perpetration, the physical and social disorder variables relate d to personal crime perpetration in opposite ways such that less physical disorder was associated with personal crime perpetration. Furthermore,

PAGE 112

112 i nmates who perceived less neighborhood racial heterogeneity we re more likely to commit personal crimes. Table 6 4 Model 3 revealed many of the so cial disorganization variables were significant for both gang and non gang members. Among gang members, more social disorder, less collective efficacy, more neighborhood poverty, less neighborhood unemployment, less residential mobility, and more racial h eterogeneity were predictive of personal crime perpetration. Among non gang members, less collective efficacy, more unemployment, and less racial heterogeneity were significant. The social disorganization variables behaved in surprisingly different ways among gang and non gang members. While some social disorganization variables were significant for gang members only (social disorder, poverty, and residential mobility), two of the three variables that share d significance between gang and non gang members operated in opposite ways (unemployment and racial heterogeneity). Racial heterogeneity was positively associated with personal crime perpetration for gang members and negatively associated for non gang members. In other words, gang members who perceive d more racial diversity were more likely to commit personal crimes where as non gang members who perceived less racial diversity were more likely to commit personal crimes. Among gang members, less neighborhood unemployment was predictive of personal crime offending whereas more unemployment was related to personal crime perpetration for non gang members. Comparing the findings from the split models ( Table 6 4 Model 3) to the full sample ( Table 6 3 Model 4) revealed interesting differences. While physica l disorder was significantly associated to personal crime perpetration for the full sample, this was not significant for either the gang or non gang samples. Social disorder was associated with personal crime perpetration for the full sample and the gang sample, suggesting that the gang sample was the driving force

PAGE 113

113 behind the findings for the full sample. Collective efficacy yielded consistent findings from analyses with all three samples given that it was negatively related to personal crime perpetration for the full sample, gang sample, and non gang sample. Alternatively, racial heterogeneity was negatively related to personal crime perpetration for the full sample and the non gang sample and positive for the gang sample. Clearly, the non gang sample f indings for racial heterogeneity were responsible for the full sample results. Comparing Self Control Theory and Social Disorganization Theory Predicting Personal Crime Perpetration The final model predicting personal crime perpetration include d all of t he variables of interest and consisted of both self control and social disorganization ( Table 6 3 Model 5) While m any of the variables significantly associated with personal crime perpetration in earlier models remain ed unchanged in the combined model s ome changes were observed Similar to Model 3, race became significant again in Model 5. Furthermore, physical and social disorder were no longer significant when self control was included in the model. Overall results from Model 5 indicate d that males non Whites, younger inmates, gang members, a nd those with low er self control we re significantly more likely to report perpetrating personal crime than females, Whites, and older inmates Furthermore, perceptions of less collective efficacy and less raci al heterogeneity we re also significantly associated with personal crime perpetration for the full sample Model 4 in Table 64 (including both self control and social disorganization variables) showed some changes from Models 2 and 3. Among gang members, males, non Hispanics, younger inmates, lower self control, less physical disorder, more social disorder, more neighborhood poverty, less unemployment, and more racial heterogeneity were significantly related to personal crime perpetration. Among non gang members, males, non Whites, younger

PAGE 114

114 in mates, lower self control, less collective efficacy, more unemployment, and less racial diversity were predictive of personal crime perpetration. While there appeared to be no significant changes for non gang members between Models 3 and 4, differences for gang members between these models involved physical disorder becoming significant and collective efficacy and residential mobility no longer reaching significance. Comparing the gang and non gang sample findings fr om the personal crime perpetration models to the property crime perpetration models revealed many similarities and few differences for the theoretical variables of interest. More similarities were observed between gang and non gang members predicting property crime perpetration in comparison with personal crime perpetration. For example, low self control, less physical disorder, more social disorder, less unemployment, and more racial heterogeneity were significantly related to property crime perpetration for both gang and non gang members. The only difference between gang and non gang members related to property crime perpetration involved a negative relationship with collective efficacy for non gang members (which was not significant for gang members) Comparing these findings with the personal crime perpetration gang and non gang samples reveal ed more similarities among gang members than non gang members. Factors predictive of property and personal offending for gang members included low self control, less physical disorder, more social disorder, less unemployment, and more racial heterogeneity. In fact, the only variable that was different between these models was the significance of poverty (for personal crime perpetration). Among non gang members, more differences were observed between the property and personal crime models. While low self control and less collective efficacy were predictive of both property and personal crime perpetration for non gang members, less physical disorder and more soci al disorder were only predictive of property crime.

PAGE 115

115 Furthermore, two of the social disorganization variables reacted differently among the property and personal crime perpetration models for non gang members. Less unemployment and more racial heterogenei ty were predictive of property crime perpetration for non gang members whereas more unemployment and less racial heterogeneity were significant for personal crime perpetration. Chapter 8 (described next) presents a discussion of these peculiar differences in the context of the other findings predicting combined crime perpetration.3 Combined Crime Perpetration The models examining combined crime perpetration were comprised of a general measure combining all of the personal and property crime items. Table 65 displays regression results predicting combined perpetration for the full sample and Table 6 6 presented the results predicting combined crime perpetration for the gang and non gang samples. For the full sample, t he combined measure of crime perpetrat ion was significantly related with ea ch of the demographic variables (Table 6 5 Model 1) More specifically, men Whites, non Hispanics, and younger inmates we re more likely to report committing crime than wo men, non Whites, Hispanics, and younger inmates Model 1 showed many similarities among gang and non gang members. Among gang members males, non Hispanics, and younger inmates were significantly more likely to report perpetrating crime. In addition to these variables, race was also significan t for n on gang members (Whites we re more likely than non Whites to commit crime). Hypothesis 5 Supported: Gang Membership Increases the Likelihood of Perpetrating Crime When gang membership was added to the combined crime perpetration model for the full sample all variables reach ed significance ( Table 6 5 Model 2). Results indicate d that, in 3 When only the most robust social disorganization variable (racial heterogeneity) was included into the full model (instead of all seven social disorganization variables), self control and racial hete rogeneity were statistically significant predictor s of personal crime perpetration for both gang and non gang members (table s not presented).

PAGE 116

116 addition to the demographic variables, gang members hip significantly predict ed combined crime perpetration which supported Hypothesis 5. This finding was also consisten t with the results examining property crime and personal crime perpetration separately. Hypothesis 6 Supported: Low Self Control Increases the Likelihood of Perpetrating Crime Consistent with Hypothesis 6 and earlier results for both property and personal crime, combined crime perpetration was significantly related with self control ( Table 6 5 Model 3) More specifically, inmates with lower self control report ed a significantly higher involvement with crime Table 6 6 Model 2 added self control and indica ted identical findings for both gang and non gang members. In addition to sex (male), ethnicity (non Hispanic), and age (younger), lower self control was also significant for gang and non gang members (as well as the full sample) In support of Hypothesi s 6, low er self control was predictive of combined crime perpetration for both gang and non gang members. Hypothesis 7 Partially Supported : Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Perpetrating Crime Table 6 5 Model 4 replaced self control with social disor ganization variables and revealed findings similar to models predicting property crime and personal crime perpetration. Like Models 1 and 2 (Table 6 5) a l l of the demographic variables we re statistically significa nt as well as gang membership. Similar to the property and personal crime models, several of the social disorganization variables we re significant, including physical disorder, social disorder, collective efficacy, neighborhood po verty and neighborhood unemployment While social disorder and poverty exhibit ed positive relationships with crime, physical disorder collect ive efficacy, and unemployment we re negatively related with crime perpetration. In other words inmates who perceived less neighborhood physical disord er, more social disorder, less collective efficacy, more neighborhood poverty and less neighborhood unemployment we re more likely to

PAGE 117

117 report committing crime. As with earlier models, some of these relationships contradict ed social disorgani zation theory, given that the theory hypothesizes that more physical disorder and unemployment predict crime. Comparing the results from the social disorganization variables in Model 4 for the models predicting property, personal, and combined crime perpe tration suggest many similarities and few differences. Poverty and unemployment were significant for the property and combined models for the ful l sample, but not for the model predicting personal cr ime perpetration. This suggested that the prop erty crim e perpetration models we re driving the findings regarding poverty and unemployment for the combined model. Also racial heterogeneity was significantly and negatively related to personal crime perpetration whereas this variable was positively related to p roperty crime perpetration. Given the stark differences between mo dels, racial heterogeneity failed to reach significance in the f ull model. Other than these few inconsistencies among models (poverty unemployment, and racial heterogeneity), factors most consistently predictive of property, personal, and combined crime perpetration for the full sample include perceptions of less physical disorder, more social disorder, and less collective efficacy. Table 6 6 Model 3 examined social disorganization among g ang and non gang members separately and indicated partial support for Hypothesis 7 given that many of the social disorganization variables predicted combined crime perpetration for both gang and non gang members. Several social disorganization variables w ere significant for both gang and non gang members, including more social disorder, less collective efficacy, more poverty, and less unemployment. Gang and non gang members differed with regard to two of the social disorganization variables given that les s physical disorder was significant for non gang members only and more racial heterogeneity was significant for gang members only. In other words, gang

PAGE 118

118 members who reported higher social disorder, poverty, and racial diversity and lower collective efficac y and neighborhood unemployment were significantly more likely to admit perpetrating crime. Non gang members who perceived less physical disorder, collective efficacy, and neighborhood unemployment and more social disorder and poverty were more likely to report committing crime. Comparing Self Control Theory and Social Disorganization Theory Predicting Combined Crime Perpetration Table 6 5 Model 5 examined combined crime perp e tration for the full sample with all independent variables of interest (including both self control and social disorganization). Results indicate d that like personal and property crime models discussed earlier, both self control and social disorganization factors significantly predicted combined crime perpetration More s pecifically in addition to sex (males) ethnicity (non Hispanic s ) and age (younger) gang membership, lower self control, less physical disorder, less collective efficacy and less neighborhood unemployment we re signi ficantly associated with crime perpetration. Co mparing these findings to the findings from models predicting property and personal crime perpetration for the full sample separately, results indicate d many similarities and few differences. All three dependent variables share d two important predictors including lower self control and less collective efficacy. Property crime perpetration drove the findings for combined crime perpetration in terms of physical disorder and unemployment. Given that racial heterogeneity was negative for personal crime perp etration and positive for property crime perpetration, combined crime perpetration indicated non significance for this variable. When all variables were entered into the split model ( Table 6 6 Model 4), many theory based differences between gang and non gang members were observed. While low self control and less physical disorder were predictive of combined crime perpetration for both gang

PAGE 119

119 members and non gang members, the majority of the social disorganization variables operated differently for gang memb ers compared with non gang members. Among gang members, more social disorder, more neighborhood poverty, less unemployment, and more racial diversity were significantly related to combined crime perpetration. Among non gang members, less collective effic acy and less racial diversity were predictive of offending. Comparing the full model for the full sample ( Table 6 5 Model 5) to the split samples (gang versus non gang) from the property, personal, and combined crime perpetration analyses revealed intere sting similarities and differences. Notably, low self control was significantly related to each of the crime perpetration variables (property, personal, and combined) for both gang members and non gang members. Results from the social disorganization var iables were less consistent between models and samples (gang versus non gang). Among gang members, less physical disorder was predictive of property, personal, and combined crime perpetration while this variable was predictive of property and combined cri me perpetration for non gang members. Among gang members, more social disorder was associated with property, personal, and combined crime perpetration whereas this variable was predictive of property perpetration only among non gang members. Less collect ive efficacy was predictive of all three perpetration models for non gang members only. More neighborhood poverty was significant for personal and combined crime perpetration for gang members only. Unemployment was significant for both gang and non gang members for all perpetration models with the exception of combined crime perpetration for non gang members. However, this was negative for gang members across all models and negative for non gang members for property crime perpetration and positive for pe rsonal crime perpetration. Finally, racial heterogeneity was positively related to p roperty, personal, and combined crime perpetration for gang members and for property perpetration

PAGE 120

120 among non gang members and negatively related to personal and combined pe rpetration for non gang members.4 Overall, hypotheses regarding crime perpetration were supported and the models predicting crime perpetration reveal ed several noteworthy findings. First, gang members were significantly more likely than non gang members t o report committing property, personal, and combined crime. Second, gang and non gang members with lower self control were significantly more likely to perpetrate property, personal, and combined crime. Third, many social disorganization factors were pre dictive of property, personal, and combined crime perpetration, although these factors differed somewhat for gang members and non gang members. For both gang members and non gang members, physical disorder, social disorder, unemployment, and racial hetero geneity played important roles in the perpetration of crime. Social disorder was important for gang members particularly while collective efficacy was important for non gang members only. Finally, results revealed that both self control theory and social disorganization theory not only predicted relationships with crime perpetration among gang and non gang members when examined separately, but these significant findings held when both theories were examined within the same model. This means that both the ories are important for explaining crime perpetration (property, personal, and combined) among gang and non gang members. 4 When only the most robust social disorganization variable (racial heterogeneity ) was included into the full mode l (instead of all seven social disorganization variables), self control and racial heterogeneity were statistically significant predictor s of combined crime perpetration for both gang and non gang members (tables not presented).

PAGE 121

121 Table 6 1. Negative Binomial Regression Predicting Property Crime Perpetration ( Full Sample ) Property Crime Perpetration (Full Sampl e) Model 1: Demographics Model 2: Gang Membership Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables Male ***.347 (.054) *** .331 (.055) ***.271 (.056) *** .349 (.056) ***.306 (.057) White *** .423 (.047) *** .482 (.047) *** .2 07 (.051) *** .419 (.050) ** .150 (.054) Hispanic *** .461 (.061) *** .457 (.061) *** .516 (.062) *** .488 (.063) *** .532 (.064) Age *** .010 (.002) ** .005 (.002) .001 (.002) .005 (.002) .001 (.002) Gang Membership ***.707 (.0647) *** .487 (.067) *** .697 (.068) *** .500 (.069) Self Control *** .790 (.052) *** .761 (.054) Physical Disorder *** .498 (.056) *** .481 (.056) Social Disorder *** .306 (.056) *** .234 (.057) Collective Efficacy *** .209 (.033) *** .156 (.033) Poverty ** .062 (.023) .053 (.023) Unemployed *** .117 (.025) *** .109 (.025) Residential Mobility .040 (.027) .043 (.028) Racial Heterogeneity ***.110 (.032) ** .101 (.032) N 2,210 2,205 2,204 2,161 2,160 *p < .05, **p < .01, ***p < .001 Standa rd Errors in Parentheses

PAGE 122

122 Table 62. Negative Binomial Regression Predicting Property Crime Perpetration (Gang versus Non Gang Samples) Property Crime Perpetration (Gang versus Non Gang Samples) Model 1: Demographics Model 2: Self Control Model 3: Soc ial Disorganization Model 4: All Variables Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Male .050 (.165) *** .393 (.059) .071 (.164) *** .324 (.060) .032 (.169) *** .443 (.061) .028 (.167) *** .386 (.062) White .220 (.131) *** .538 (.052 ) .046 (.136) *** .270 (.056) .126 (.137) *** .470 (.055 ) .075 (.143) *** .208 (.0594) Hispanic *** .77 9 (.140) *** .388 (.069) *** .617 (.144) *** .488 (.070) *** .763 (.149) *** .414 (.072) *** .659 (.149) *** .475 (.073) Age ** .019 (.007) .003 (.002) .015 (.007) .001 (.002) ** .021 (.007) .001 (.002) .018 (.008) .002 (.002) Self Control *** .952 (.143) *** .751 (.056) *** .971 (.151) *** .701 (.058) Physical Disorder .289 (.1 40) *** .604 (.062) .345 (.139) *** .559 (.062) Social Disorder *** .473 (.136) *** .282 (.062) ** .435 (.138) ** .195 (.063) Collective Efficacy .039 (.080) *** .276 (.037) .007 (.081) *** .226 (.037) Poverty .087 (.059) .060 (.026) .091 ( .059) .047 (.026) Unemployed .149 (.067) *** .119 (.027) ** .202 (.066) *** .102 (.027) Residential Mobility .100 (.073) .014 (.030) .119 (.074) .015 (.031) Racial Heterogeneity .142 (.076) **.107 (.036) *.160 (.076) *.091 (.036) N 334 1 ,871 334 1,870 329 1,832 329 1,831 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 123

123 Table 63. Negative Binomial Regression Predicting Personal Crime Perpetration (Full Sample) Personal Crime Perpetration (Full Sample) Model 1: Demogra phics Model 2: Gang Membership Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables Male ***.898 (.054) *** .813 (.055) .835 (.055) *** .779 (.056) ***.799 (.056) White .023 (.047) .015 (.047) .120 (.048) .040 (.049) .119 ( .051) Hispanic *** .245 (.057) *** .244 (.058) ** .184 (.058) .118 (.060) .093 (.061) Age *** .041 (.002) *** .033 (.002) *** .029 (.002) *** .029 (.003) *** .026 (.003) Gang Membership ***1.145 (.061) *** .982 (.063) *** 1.109 (.064) *** 1.005 (.065) Self Control *** .601 (.045) *** .597 (.049) Physical Disorder .109 (.053) .056 (.053) Social Disorder ** .151 (.051) .067 (.052) Collective Efficacy *** .199 (.032) *** .131 (.032) Poverty .012 (.021) .026 (.021) Unemploye d .022 (.023) .020 (.023) Residential Mobility .045 (.025 .016 (.025) Racial Heterogeneity ** .088 (.029) *** .122 (.029) N 2,248 2,242 2,241 2,197 2,196 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 124

124 Table 64. Negati ve Binomial Regression Predicting Personal Crime Perpetration (Gang versus Non Gang Samples) Personal Crime Perpetration (Gang versus Non Gang Samples) Model 1: Demographics Model 2: Self Control Model 3: Social Disorganization Model 4: All Variables Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Male ***.650 (.159) *** .820 (.059) ***.577 (.159) *** .854 (.060) *** .676 (.161) *** .779 (.060) ***.614 (.161) *** .813 (.061) White .128 (. 124) .016 (.051) .017 (.126) ** .160 (.053) .068 (.126) .000 (.054) .186 (.128) *** .182 (.057) Hispanic *** .496 (.131) ** .172 (.065) *** .510 (.132) .092 (.066) *** .531 (.139) .037 (.068) *** .615 (.138) .009 (.069) Age *** .031 (.007) *** .03 3 (.003) *** .028 (.007) *** .029 (.003) *** .034 (.008) *** .029 (.003) *** .033 (.008) *** .027 (.003) Self Control *** .657 (.109) *** .608 (.050) *** .662 (.120) *** .647 (.055) Physical Disorder .212 (.125) .088 (.060) .314 (.126) .0 05 (.060) Social Disorder ** .310 (.116) .096 (.058) ** .374 (.119) .021 (.059) Collective Efficacy *** .284 (.079) *** .183 (.035) .130 (.084) *** .118 (.035) Poverty .137 (.058) .006 (.023) .123 (.058) .033 (.023) Unemployed *** .343 (.064) **.072 (.025) *** .368 (.064) ** .075 (.025) Residential Mobility .153 (.068) .026 (.027) .069 (.071) .008 (.027) Racial Heterogeneity *** .322 (.071) *** .162 (.031) *** .305 (.072) *** .200 (.031) N 342 1,900 342 1,899 337 1,860 337 1,859 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 125

125 Table 65. Negative Binomial Regression Predicting Combined Crime Perpetration (Full Sample) Combined Crime Perpetration (Full Sample) Model 1: Demographics Model 2: Gang Member ship Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables Male ***.691 (.051) *** .630 (.052) ***.617 (.052) *** .617 (.053) ***.602 (.053) White *** .157 (.045) *** .183 (.045) .023 (.047) *** .175 (.048) .004 (.050) Hispanic ** .324 (.056) *** .318 (.056) *** .269 (.057) *** .231 (.058) *** .200 (.059) Age *** .026 (.002) *** .019 (.002) *** .015 (.002) *** .016 (.002) *** .012 (.002) Gang Membership ***.938 (.061) *** .771 (.062) *** .912 (.063) *** .784 (.064) Self Control *** .597 (.045) *** .574 (.049) Physical Disorder *** .252 (.051) *** .200 (.050) Social Disorder ** .148 (.050) .063 (.051) Collective Efficacy *** .225 (.030) *** .176 (.031) Poverty *.050 (.020) .017 (.021) Unemployed *** 073 (.022) ** .068 (.022) Residential Mobility .013 (.024) .040 (.025) Racial Heterogeneity .009 (.029) .019 (.029) N 2,248 2,242 2,241 2,197 2,196 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 126

126 Table 6 6. Negative Binomial Regression Predicting Combined Crime Perpetration (Gang versus Non Gang Samples) Combined Crime Perpetration (Gang versus Non Gang Samples) Model 1: Demographics Model 2: Self Control Model 3: Social Disorganization Model 4: All Variables Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Male ***.523 (.157) *** .644 (.056) ***.472 (.157) *** .632 (.056) ***.513 (.159) *** .621 (.057) **.460 (.159) *** .602 (.058) White .137 (.124) *** .187 ( .049) .011 (.127) .021 (.051) .062 (.127) *** .170 (.052) .187 (.130) .011 (.055) Hispanic *** .630 (.130) *** .249 (.063) *** .609 (.131) ** .193 (.064) *** .732 (.135) .121 (.066) *** .756 (.135) .085 (.067) Age *** .027 (.007) *** .018 (.002) ** .023 (.007) *** .014 (.002) *** .031 (.008) *** .013 (.002) *** .030 (.008) *** .010 (.002) Self Control *** .707 (.115) *** .583 (.050) *** .789 (.124) *** .570 (.054) Physical Disorder .206 (.127) *** .312 (.057) ** .331 (.128) *** .236 ( .056) Social Disorder .278 (.119) .130 (.056) ** .351 (.121) .024 (.057) Collective Efficacy .158 (.074) *** .254 (.034) .026 (.078) *** .209 (.034) Poverty *.137 (.058) *.050 (.022) .116 (.058) .020 (.022) Unemployed *** .299 (. 064) .053 (.024) *** .338 (.063) .044 (.024) Residential Mobility .062 (.068) .020 (.027) .015 (.070) .038 (.027) Racial Heterogeneity *** .343 (.071) .050 (.031) *** .324 (.072) ** .081 (.031) N 342 1,900 342 1,899 337 1,860 337 1,859 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 127

127 CHAPTER 7 RESULTS PREDICTING C RIME VICTIMIZATION This Chapter present s the results from the negative binomial regressions for each of the victimization dependent variables, beginning with propert y crime victimization then personal crime victimization and finally combined crime victimization Identical to the organization of the previous Chapter focusing on crime perpetration, e ach of the three sect ions present s two tables (totaling six tables), one table illustrates results for the full sample and the other table displays results for the gang versus the non gang samples. Similar to the examination of crime perpetration, gang and non gang members were analyzed separately in order to test differ ences between the groups. The tables presented in this Chapter also include several models, which mirror the format of the models presented in the previous Chapter For the table s presenting results for the full sample, analyses include d Model 1 ( demog raphic variables only ) Model 2 ( gang membership ) Model 3 ( self control ) Model 4 ( social disorganization ) and Model 5 ( both self control and social disorganization). For the tables comparing the gang versus non gang models, the same models are presente d although gang membership was removed as an independent variable (given that the separate models are split by gang membership). These tables include d Model 1 ( demographics ) Model 2 ( self control ) Model 3 ( social disorganization ) and Model 4 (self cont rol and social disorganization). Findings from the models are illus trated and these findings are compared with their perpetration counterpart (e.g., property crime victimization was compared with property crime perpetration) as well as the other victimiza tion models (e.g., property crime victimization was compared with personal and combined crime victimization).

PAGE 128

128 Property Crime Victimization Table 7 1 presents r esults predicting property crime victimization for the full sample. Table 7 1 Model 1 indi cated Whites no n Hispanics, and older inmates we re significantly more likely to be victimized by property crime than non Whites, Hispanics, and younger inmates. Comparing property crime victimization between gang versus non gang members yielded substanti al differences (see Table 7 2). Table 7 2 Model 1 revealed that none of the demographic variables were significant for gang members while all variables with the exception of sex were significant for non gang members. In other words, there were no sex, ra ce, ethnicity, or age differences among property crime victims who were gang members. Alternatively, non gang members who were White, non Hispanic, and older were more likely to be victimized by property crimes. The demographic variables remained unchang ed among all four models predicting property crime victimization for both gang and non gang members. When comparing Model 1 (Table 6 1) predicting property crime per petration to Model 1 (Table 7 1) predicting property crime victimization several differe nces we re noteworthy in terms of sex and age While non Hispanics and Whites we re more likely to perpetrate and be victimized by property crime, males were more likely to perpetrate property crime. Furthermore, younger inmates were more likely to perpetr ate property crime whereas older inmates were more likely to be victimized by property crimes. Comparing the Model 1 findings from the split models predicting property crime perpetration (Table 6 2) to the split models predicting property crime victimiz ation (7 2) indicated additional differences across models. Among gang members, non Hispanics and younger inmates were significantly more likely to perpetrate property crimes while none of the demographic variables were predictive of property crime victim ization. Among non gang members, males, Whites, and non Hispanics were significantly more likely to perpetrate property

PAGE 129

129 crimes while the property crime victimization models indicated age differences (younger inmates victimized more) in addition to the rac e (Whites) and ethnicity (non Hispanics) differences and no sex differences. Hypothesis 8 Supported: Gang Membership Increases the Likelihood of Being Victimized b y Crime D emographic variables remain ed unchanged when gang membership was added to the proper ty crime victimization model (Table 7 1 Model 2). Like the property perpetration model, f indings suggest ed that gang membership was also positively and significantly related to property crime victimization. More specifically gang members we re more likely than non gang members to report being victimized by property cr ime. This finding supported H ypothesis 8 and was in line with prior research (Peterson e t al., 2004; Taylor et al., 2007). Hypothesis 9 Unsupported : Low Self Control Does Not Increase the Li kelihood of Being Victimized b y Crime Unlike t he property perpetration analyse s, self control was not predictive of property crime victimization for the full sample ( Table 7 1 Model 3). Results suggest ed that inmates we re victimized by property crimes re gardless of their level of self control. This finding does not support H ypothesis 9 or the work of prior research (Schreck, 1999; Stewart et al., 2004). However, examining the gang versus non gang models sheds light on this finding for the full sample. Table 7 2 Model 2 examined the effects of self control on property crime victimization among gang and non gang members. While lower self control was predictive of property crime victimization for non gang members, higher self control was significant for g ang members. In other words, gang members with higher self control were more likely to report property crime victimization. This finding was inconsistent with self control theory (Gottfredson & Hrischi, 1990) and the recent work of prior research (Schrec k, 1999; Stewart et al., 2004). Given these conflicting findings between the gang and non gang samples, the full model indicated that self -

PAGE 130

130 control was not significant. Comparing these findings with the property crime perpetration models indicated differe nces given that low self control was predictive of property crime perpetration (for both gang and non gang members) and property crime victimization for non gang members while higher self control was a significant predictor of property crime victimization among gang members. Hypothesis 10 Unsupported: Perceptions of Socially Disorganized Neighborhoods Do Not Increase the Likelihood of Being Victimized by Crime Simila rly, social disorganization did not appear to predict property crime victimization for the f ull sample ( Table 7 1 Model 4). O nly o ne of the seven social disorganization variables reached significance: racial heterogeneity. Inmates who perceived more neighborhood racial diversity we re significantly more likely to report being victimized by prope rty crime. T his finding render ed Hypothesis 10 largely unsupported. Comparing these findings with the property crime perpetration results from the full sample yield ed substantial differences. Recall that physical disorder, social disorder collective ef ficacy neighborhood poverty, neighborhood unemployment, and racial heterogeneity were predictive of perpetrating property crimes whereas only racial heterogeneity was associated with property crime victimization. When social disorganization variables we re examined for property crime victimization among gang and non gang members ( Table 7 2 Model 3), higher racial heterogeneity was significant for both gang and non gang members while higher social disorder was significant for gang members only. These find ings differed substantially from the split models predicting property crime perpetration among gang and non gang members, which found many more social disorganization variables to be important. While only more racial heter ogeneity and social disorder were predictive of property crime victimization for gang members, less physical disorder, more social disorder, and le ss neighborhood unemployment were predictive of property

PAGE 131

131 crime perpetration among gang members. Among non gang members, only more racial hete rogeneity was significantly related to property crime victimization whereas less physical disorder, more social disorder, less collective efficacy, more poverty, less neighborhood unemployment, and more racial heterogeneity were predictive of property crim e perpetration among non gang members. This means that social disorganization theory did not successfully explain property crime victimization. Comparing Self Control Theory and Social Disorganization Theory Predicting Property Crime Victimization When se lf cont rol and social disorganization we re examined within the same an alysis for the full sample ( Table 7 1 Model 5), the findings we re consistent with Models 3 and 4. The demographic variables we re unchanged and while gang membership continued to be pred ictive of property crime victimization, self control and so cial disorganization variables we re not ( again, with the exception of racial heterogeneity). Table 7 2 Model 4 revealed that self control and social disorganization operated differently for gang m embers compared to non gang members when examining property crime victimization. Again, findings revealed that while self control was significantly related to property crime victimization for both gang and non gang members, it operated in different ways. Low er self control was predictive of property crime victimization among non gang members whereas higher self control was associated with property crime victimization for gang members. Very few of the social disorganization variables were related to prope rty crime victimization among gang and non gang members. While more racial diversity was a significant predictor for both groups, only more social disorder was significant among gang members. These findings differed when compared to the gang versus non g ang member models predicting property crime perpetration. For example, low self control and most of the social disorganization variables were significant for both gang and non gang members when

PAGE 132

132 examining property perpetration. Overall, these findings ind icated that neither self control theory nor social disorganization theory successfully explained property crime victimization among gang or non gang members.1 Personal Crime Victimization Table 7 3 presents the regression models for personal crime victimization for t he full sample and Table 7 4 displays models for the gang versus non gang samples Table 7 3 Model 1 revealed tha t sex, race and age we re significantly associated with personal crime victimization for the full sample Males, Whites, and youn ger inmates we re significantly more likely to report personal crime victimization than females, non Whites and older inmates. These findings remain ed consistent across all five models, with the exception of Model 2. All demographic variables we re signif icant when gang membership was added to the model (e.g., ethnicity bec a me signi ficant such that non Hispanics we re more likely than Hispanics to report personal crime victimization). Table 7 4 Model 1 examined the relationship between personal crime vict imization and the demographic variables among gang and non gang members. Findings were consistent for both groups across all models and revealed that sex and ethnicity were significant for gang members while sex, race, and age were significant for non gan g m embers. Among gang members, males and non Hispanics were more likely to report personal crime victimizat ion. Among non gang members males Whites, and younger inmates were more likely to be victimized by personal crimes. Compared to the property cri me victimization models for gang and non gang members, some differences were observed. While sex and ethnicity were significant among gang members 1 When only the most robust social disorganization variable ( racial heterogeneity ) was included into the full model (instead of all seven social disorganization variables), self control and racial heterogeneity were statistically significant predictor s of property crime victimizatio n for both gang and non gang members (tables not presented).

PAGE 133

133 for the personal crime victimization, recall that none of the demographic variables reached significance for the property crime victimization model. For the personal crime victimization model for non gang members, sex, race, and age were significant while race, ethnicity, and age were predictive for the property crime victimization model. Furthermore, older no n gang members were more likely to be property crime victims whereas younger non gang members were more likely to be personal crime victims. Comparing the personal crime victimization results for gang and non gang members to the split models predicting pe rsonal crime perpetration also revealed interesting similarities and differences. Both models indicate d that sex and ethnicity were important predictors for gang members whereas age was also significant for the personal crime perpetration model. Furtherm ore, both models indicated that sex and age were significant for non gang members whereas ethnicity was significant for personal crime perpetration and race was significant for personal crime victimization. Hypothesis 8 Supported: Gang Membership Increases the Likelihood of Being Victimized b y Crime Cons istent with Hypothesis 8 and the results from property crime victimization (as well as all of the crime perpetration analyses), gang membership was predictive of personal crime victimization for the full sam ple (Table 7 3 Model 2) In other words, gang members we re significantly more likely to be victims of personal crimes than non gang members. As with the other victimization and perpetrat ion models, this finding remained consistent when the theoretical va riables of interest are added to the analyses. Hypothesis 9 Supported: Low Self Control Increases the Likelihood of Being Victimized b y Crime Inmates with low er self control we re significantly more likely than inmates with higher self control to report be ing victims of personal crime (see Table 7 3 Model 3). Th is finding supported Hypothesis 9 and was similar to the property, personal and combined crime

PAGE 134

134 perpetration model s. Yet, this finding was different from the property crime victimization results gi ven that self control was not significantly predictive of property crime victimization for the full sample (but was positive for gang members and negative for non gang members) Lower self control was also significantly related to personal crime victimiza tion for both gang and non gang members ( Table 7 4 Model 2), which was consistent with the personal crime perpetration results for gang versus non gang members. However, this was partially supported by the findings predicting property crime victimization given that lower self control was significant for non gang members whereas higher self control was significant for gang members. Hypothesis 10 Partially Supported : Perceptions of Socially Disorganized Neighborhoods Increase the Likelihood of Being Victimiz ed by Crime Table 7 3 Model 4 replaced self control with social disorganization and findings indicated only one of the seven social disorganization variables was significantly related to personal crime victimiza tion: social disorder I nmates who perceived more social disorder within their neighborhood we re significantly more likely to report personal crime victimization. The model predicting property crime vict imization also revealed only one significant social disorganization variable (racial heterogenei ty). Models predicting crime perpetration reveal ed that more of the social disorganization variables explain ed perpetration in comparison with victimization. Property crime perpetration was predicted by perceptions of l ess physical disorder, more social disorder, less collective efficacy, more poverty, less unemployment, and more racial heterogeneity. Furthermore, personal crime perpetration was associated with all of these variables except poverty and unemployment (but less heterogeneity rather than mor e) The combined crime perpetration model was related to each of the variables predictive of property crime perpetration with the exception of racial heterogeneity.

PAGE 135

135 Table 7 4 Model 3 predicting personal crime victimization for the split samples indicated that perceptions of more social disorder and more neighborhood unemployment were significant for non gang members while perceptions of more neighborhood poverty and less neighborhood unemployment were significant for gang members. Compared to the split m odels predicting property crime victimization, interesting differences were observed. For example, more social disorder and racial heterogeneity were related to property crime victimization among gang members whereas more poverty and less unemployment wer e related to personal crime victimization for gang members Among non gang members, more racial heterogeneity was significantly predictive of property crime victimization whereas more social disorder and neighborhood unemployment were associated with pers onal crime victimization among non gang members. Comparing the personal crime victimization split model results for social disorganization to the personal crime perpetration split models indicated even more differences. While poverty and unemployment wer e significant for personal crime victimization for gang members, more social disorder, less collective efficacy, more poverty, less unemployment, less residential mobility, and more racial heterogeneity were significant for personal crime perpetration amon g gang members. Additionally, whereas more social disorder and unemployment were significant for personal crime victimization among non gang members, less collective efficacy, more unemployment, and less racial heterogeneity were predictive of personal cr ime perpetration for non gang members. Comparing Self Control Theory and Social Disorganization Theory Predicting Personal Crime Victimization The final model predicting persona l crime victimization for the full sample ( Table 7 3 Model 5) examined all vari ables of interest, including self control and social disorganization. Results indicate d no substantial changes from Models 3 and 4 given that se lf control and social

PAGE 136

136 disorder we re predictive of personal crime victimization. When compared to the property crime victimization mod el, some important differences we re noted. Low self control was predictive of personal crime victimization but was not significant for property crime victimization for the full sample Furthermore, while the only social disorganiza tion variable predictive of personal crime victimization was social disorder, the only variable predictive of property crime victimization was racial heterogeneity. Table 7 4 Model 4 examined the full model for gang and non gang member and revealed that bo th self control and few social disorganization variables were predictive of personal crime victimization for both gang and non gang members. Both gang and non gang members with low self control were significantly more likely to be victimized by personal c rime. This finding was supported by the gang and non gang analysis predicting property crime victimization and personal crime perpetration. Among gang members, perceptions of more neighborhood poverty and less neighborhood unemployment were significant f or personal crime victimization while higher social disorder and higher neighborhood unemployment were significant for non gang members. In terms of the social disorganization variables, some important differences were observed between gang and non gang m embers when personal crime victimization was compared with property crime victimization. For example, property crime victimization was significantly associated with more social disorder for gang members and more racial heterogeneity for both gang and non gang members. Furthermore, personal crime perpetration was significantly related to many more social disorganization variables than the personal crime victimization split models. More specifically, gang members who reported less physical disorder, more s ocial disorder, more poverty, less unemployment, and more racial heterogeneity were significantly more likely to be perpetrators of personal crime. Non gang members who

PAGE 137

137 reported less collective efficacy, more unemployment, and less racial heterogeneity we re significantly more likely to perpetrate personal crimes. Overall, finding revealed that self control theory successfully explained personal crime victimization, given that lower self control was associated with more personal crime victimization. Again, social disorganization theory generated some support. However, few social disorganization variables were significantly related to personal crime victimization.2 Combined Crime Victimization The remaining two tables examine regression models predictin g combined crime victimization for the full sample (Table 7 5) and for the gang and non gang samples (Table 7 6). Among the full sample (Table 7 5), all demographic variables we re predictive of combined crime victimization (Model 1). More specifically, m ales, Whites, non Hispanics, and younger inmates we re more likely to report crime victimization than females, non Whites, Hispanics, and older inmates. T hese findings we re similar to the full sample model predicting personal crime victimization (w ith the exception of ethnicity which did not reach significance for personal crime victimization) The full sample model predicting property crime victimization suggested similar findings with regard to race and ethnicity but opposite results regarding sex and a ge. No sex differences were observed predicting property crime victimization. Furthermore, age was positively associated with property crime victimization and negatively related to personal and combined crime victimization. When the findings from the co mbin ed crime victimization model we re compared to the combined cri me perpetration model, results we re identical such that males, 2 When only the most robust social disorganization variable ( unemployment ) was included into the full model (instead of all seven social disorganization variables), self control and unemployment were statistically significant predictor s of p ersonal crime victimization for both gang and non gang members However, unemployment was negatively associated for gang members and positively associated for non gang members (tables not presented).

PAGE 138

138 Whites, non Hispanics, and younger inmates we re more likely to perpetrate and be victimized by crime overall Table 7 6 dis plays results predicting combined crime victimization for the gang versus non gang samples. Model 1 rev ealed that sex and ethnicity were significant for gang members while sex, race, and ethnicity were significant for non gang members. Among gang members males and non Hispanics were significantly more likely to report combined crime victimization than females and Hispanics. Among non gang members, males, Whites, and non Hispanics were more likely to be victimized by crime than female s, non Whites, and H ispanics. Recall that t he full sample model (Table 7 5 Model 1) showed that all demographic variables were significant. Examining the gang versus non gang samples indicated that the non gang sample was driving the full sample model findings. Hypothesis 8 Supported: Gang Membership Increases the Likelihood of Being Victimized b y Crime Table 7 5 Model 2 reveal ed that gang members hip was positively and significantly associated with combined crime victimization. In other words, gang members we re significant ly more likely to report combined crime victimization than non gang members. This finding supported Hypotheses 8 and was consistent with the other victimization models (property and personal crime victimization) as well as the perpetration models (propert y, personal, and combined crime perpetration). Hypothesis 9 Supported: Low Self Control Increases the Likelihood of Being Victimized b y Crime When self control was added to the combined victimization model for the full sample, results indicate d that low self control was a significant predictor of victimization ( Table 7 5 Model 3) This finding supported Hypothesis 9 and was consistent with the findings from the personal crime victimization model for the full sample and all three perpetration crime models

PAGE 139

139 (property, personal, and combined) However self control was not significant for the full model predicting property crime victimization. Table 7 6 Model 2 (adding self control) revealed no change for the demographic variables among gang members. Amo ng non gang members, ethnicity was no longer significant when self control was added to the model. Self control was predictive of combined crime victimization for non gang members only whereas self control was not significantly related to crime victimizat ion for gang members. T his suggests that t he non gang sample was driving the findings for the full sample, given that self control was significant for both of these models (and not the gang member sample). Hypothesis 10 Unsupported : Perceptions of Sociall y Disorganized Neighborhoods Do Not Increase the Likelihood of Being Victimized b y Crime S ocial disorder was the only significant social disorganization variable predictive of combined crime victimization for the full sample ( Table 7 5 Model 4) A percep tion of more social disorder was related to more crime victimization among the full sample. T his finding was identical to the personal crime victimization model, which suggested that the personal crime victimization model was the driving force behind this finding for the combined victimization model. The property crime victimization model revealed that racial heterogeneity was the only significant social disorganization variable When compared to the perpetration models, severa l similarities and differen ces we re noted ; however social disorganization was more successful in explaining crime perpetration than crime victimization The property crime perpetration model for the full sample revealed that many more social disorganizat ion variables were significa nt ( including lower physical disorder, more social disorder, less collective efficacy, more poverty, less unemployment, and more racial heterogeneity ) The personal crime perpetration model for the full sample indicated that less physical disorder, more s ocial disorder, less collective efficacy, and less racial heterogeneity were significant. The combined crime perpetration model revealed

PAGE 140

140 that less physical disorder, more social disorder, less collective efficacy, more poverty, and less unemployment were important predictors. Table 7 6 Model 3 examin ed social disorganization among the split models and indicated no changes in the demographic variables from the previous models for gang and non gang members with the exception of ethnicity which no longer re ached significance among non gang members Among gang members, perceptions of more neighborhood poverty and less neighborhood unemployment were significantly predictive of combined crime victimization. Among non gang members, more social disorder and mor e neighborhood unemployment were related to combined crime victimization. Only social disorder was significant for the full model, which suggested that the non gang sample was driving these findings. Comparing Self Control Theory and Social Disorganizatio n Theory Predicting Combined Crime Victimization Whe n both theories we re examined in the same model for the full sample ( Table 7 5 Model 5), findings reveal ed that lower self control was predictive of combined crime victimization whereas none of the social disorganization variables reach ed significance. This finding differed somewhat from the other victimization models. For example, low self control and high social disorder were significant for personal crime victimization whereas more racial heterogeneit y was significant for property crime victimization (and self control was not significant) among the full sample When comparing the findings from Table 7 5 Model 5 predicting combined crime victimization for the full sample to the combined crime perpetrat ion for the full sample, some similarities and differences we re noteworthy. Self control was significant for both combined perpetra tion and victimization. Among the social d isorganization variables, none were predictive of combined crime victimization wh ereas many were significant

PAGE 141

141 for the combined crime perpetration model ( less physical disorder, less collective efficacy, and less neighborhood unemployment ) Table 7 6 Model 4 examined the effects of both self control and social disorganization on combined crime victimization among the gang versus non gang samples. Self control was significant for non gang members only, which indicated that non gang members with lower self control were significantly more likely than those with higher self control to be vic timized. Clearly, the non gang member sample was the driving force behind the full sample findings given that both indicate d self control was significant (whereas the gang sample did not). Among gang members, perceptions of more neighborhood poverty, les s neighborhood unemployment, a nd more racial heterogeneity were significantly related to combined crime victimization Among non gang members, perceptions of more neighborhood unemployment was significantly associated with combined crime victimization. N one of the social disorganization variables were significant for the full sample given the opposite findings for unemployment between gang (negative association) and non gang members (positive association). While poverty was significant among gang members only, the non gang member sample overpowered this finding for the full sample model.3 Overall, many of the hypotheses regarding crime victimization were supported. Several important findings characterized the models predicting crime victimization. First, gang members were significantly more likely than non gang members to be victims of personal, property, and combined crime. This finding was consistent among the models predicting crime perpetration as well (property crime, personal crime, and combined c rime perpetration). Second self control 3 When on ly the most robust social disorganization variable (social disorder) was included into the full model (instead of all seven social disorganization variables), self control was a statistically significant predictor of combined crime victimization for non ga ng members only and unemployment was significant for both gang members (negative relationship) and non gang members (positive relationship) (tables not presented).

PAGE 142

142 was significantly related to some of the crime victimization models, indicating some limited support for hypotheses predicting low self controls association with victimization. However, low self control was not p redictive of all of the crime victimization models, as the crime perpetration models indicated. Self control was significantly related to personal crime perpetration (for the full sample, gang sample, and non gang sample). Self control was not predictive of property crime victimization, in light of the contradictory findings from the gang (showing self control as significant and positive) versus non gang samples (showing self control as significant and negative). For the combined crime victimization mode l, self control was significant for the full sample and the non gang sample, but not for the gang sample. Third, less support was observed for social disorganization theorys ability to predict crime victimization in comparison with crime perpetration. F or example, only one social disorganization variable was predictive of property crime victimization among the full sample and the non gang sample (racial heterogeneity) while one additional variable was predictive of property crime victimization among gan g members (social disorder). Social disorder was predictive of personal crime victimization among the full sample and the non gang sample. Poverty was significantly associated with personal crime victimization for gang members only, while unemployment wa s significant (but negative) for gang members and significant (but positive) for non gang members, thus rendering the full model not significant. When self control and social disorganization were tested together for the combined crime victimization model, the full sample indicated no support for social disorganization, due to inconsistencies among the gang and non gang samples. For example, poverty and racial heterogeneity were significantly predictive of combined crime victimization for gang members only and unemployment was significant (yet negative) for gang members and (positive) for non gang members. Therefore, more limited support was generated

PAGE 143

143 for selfcontrol and social disorganization theories when predicting crime victimization compared to crime perpetration

PAGE 144

144 Table 71. Negative Binomial Regression Predicting Property Crime Victimization (Full Sample) Property Crime Victimization (Full Sample) Model 1: Demographics Model 2: Gang Membership Model 3: Self Control Model 4: Social Disorganizati on Model 5: All Variables Male .059 (.057) .074 (.058) .072 (.058) .066 (.059) .064 (.059) White ***.268 (.051) *** .291 (.052) *** .279 (.053) ***.297 (.054) *** .284 (.055) Hispanic *** .286 (.066) *** .307 (.066) *** .304 (.067) *** .296 (.068) *** .293 (.068) Age *** .008 (.002) *** .010 (.002) *** .010 (.002) *** .010 (.002) *** .010 (.002) Gang Membership ***.364 (.071) ***.352 (.072) *** .341 (.0726) *** .331 (.073) Self Control .058 (.054) .058 (.056) Physical Disorder .100 (.06 1) .100 (.061) Social Disorder .089 (.059) .086 (.059) Collective Efficacy .025 (.035) .030 (.036) Poverty .003 (.024) .005 (.024) Unemployed .025 (.026) .024 (.026) Residential Mobility .017 (.028) .019 (.028) Racial Heterogen eity **.087 (.031) **.085 (.031) N 2,193 2,188 2,187 2,144 2,143 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 145

145 Table 7 2. Negative Binomial Regression Predicting Property Crime Victimization (Gang versus Non Gang Samples) Propert y Crime Victimization (Gang versus Non Gang Samples) Model 1: Demographics Model 2: Self Control Model 3: Social Disorganization Model 4: All Variables Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample N on Gang Sample Male .068 (.177) .091 (.061) .127 (.179) .085 (.062) .119 (.182) .079 (.062) .170 (.184) .070 (.062) White .122 (.135) ***.363 (.056) .059 (.138) ***.340 (.057) .165 (.141) ***.369 (.059) .099 (.145) ***.340 (.060) Hispanic .2 68 (.144) *** .325 (.075) .249 (.145) *** .319 (.075) .236 (.150) *** .319 (.077) .215 (.151) *** .310 (.077) Age .002 (.007) *** .011 (.002) .003 (.007) *** .011 (.003) .001 (.008) *** .010 (.003) .002 (.008) *** .011 (.003) Self Control *.326 (. 142) ** .118 (.059) *.320 (.147) .134 (.061) Physical Disorder .280 (.146) .067 (.067) .258 (.147) .061 (.067) Social Disorder *.312 (.137) .039 (.066) *.297 (.136) .026 (.067) Collective Efficacy .067 (.085) .029 (.039) .081 (.0 85) .045 (.039) Poverty .073 (.064) .003 (.026) .070 (.063) .000 (.026) Unemployed .126 (.071) .047 (.028) .111 (.071) .047 (.028) Residential Mobility .016 (.072) .028 (.030) .017 (.072) .031 (.030) Racial Heterogeneity *.174 (.078) *.075 (.034) *.181 (.078) *.073 (.034) N 335 1,853 335 1,852 330 1,814 330 1,813 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 146

146 Table 7 3. Negative Binomial Regression Predicting Personal Crime Victimization (Full Sample) Perso nal Crime Victimization (Full Sample) Model 1: Demographics Model 2: Gang Membership Model 3: Self Control Model 4: Social Disorganization Model 5: All Variables Male ***.420 (.052) *** .357 (.053) ***.370 (.053) *** .358 (.054) ***.371 (.054) White ** .121 (.047) ** .143 (.047) *.098 (.048) *** .196 (.049) ** .149 (.050) Hispanic .084 (.057) .112 (.057) .093 (.057) .060 (.059) .048 (.059) Age *** .012 (.002) *** .008 (.002) ** .006 (.002) ** .006 (.002) .005 (.002) Gang Membership ***.753 (.062) *** .690 (.063) *** .735 (.063) *** .683 (.064) Self Control *** .230 (.048) *** .214 (.050) Physical Disorder .020 (.053) .023 (.053) Social Disorder **.142 (.051) .121 (.051) Collective Efficacy .017 (.032) .003 (.032) Poverty .002 (.022) .003 (.022) Unemployed .043 (.023) .043 (.023) Residential Mobility .043 (.024) .043 (.024) Racial Heterogeneity .006 (.028) .006 (.028) N 2,252 2,246 2,245 2,201 2,200 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 147

147 Table 7 4. Negative Binomial Regression Predicting Personal Crime Victimization (Gang versus Non Gang Samples) Personal Crime Victimization (Gang versus Non Gang Samples) Model 1: Demographics Model 2: Self Control Model 3: Social Dis organization Model 4: All Variables Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sample Non Gang Sample Male **.476 (.159) *** .344 (.056) **.475 (.158) *** .356 (.057) **.467 (.162) *** .343 (.058) **.470 (.16 1) *** .357 (.058) White .011 (.126) **.160 (.051) .045 (.127) *.111 (.052) .073 (.129) *** .232 (.054) .138 (.132) *** .181 (.055) Hispanic .292 (.133) .077 (.064) .298 (.133) .051 (.064) .355 (.141) .038 (.067) ** .385 (.141) .016 (.067) Age .004 (.007) *** .008 (.002) .002 (.007) ** .007 (.002) .002 (.007) ** .007 (.002) .001 (.007) .006 (.002) Self Control .253 (.118) *** .228 (.053) .322 (.127) *** .213 (.055) Physical Disorder .097 (.126) .059 (.059) .120 (.12 7) .067 (.059) Social Disorder .055 (.115) ** .172 (.057) .065 (.117) .142 (.058) Collective Efficacy .169 (.088) .004 (.034) .145 (.008) .015 (.034) Poverty *** .181 (.056) .027 (.024) *** .190 (.057) .026 (.024) Unemployed *** .250 (.063) *** .084 (.025) *** .271 (.063) *** .086 (.025) Residential Mobility .086 (.065) .035 (.026) .058 (.066) .038 (.026) Racial Heterogeneity .110 (.072) .019 (.031) .128 (.073) .022 (.031) N 342 1,904 342 1,903 337 1,864 337 1,863 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 148

148 Table 7 5. Negative Binomial Regression Combined Crime Victimization (Full Sample) Combined Crime Victimization (Full Sample) Model 1: Demographics Model 2: Gang Membership Model 3: Se lf Control Model 4: Social Disorganization Model 5: All Variables Male ***.309 (.051) *** .253 (.052) ***.265 (.052) *** .254 (.053) ***.268 (.053) White ** .141 (.046) *** .169 (.046) ** .126 (.047) *** .221 (.048) *** .175 (.049) Hispanic .138 (.057) .161 (.057) ** .151 (.057) .123 (.059) .118 (.059) Age *** .007 (.002) .003 (.002) .002 (.002) .003 (.002) .002 (.002) Gang Membership ***.659 (.062) *** .603 (.063) *** .637 (.063) *** .589 (.064) Self Control *** .226 (.047) *** .218 (. 049) Physical Disorder .024 (.053) .027 (.053) Social Disorder *.116 (.051) .095 (.051) Collective Efficacy .006 (.031) .016 (.031) Poverty .000 (.022) .002 (.022) Unemployed .042 (.023) .043 (.023) Residential Mobility .034 (.024) .034 (.024) Racial Heterogeneity .027 (.028) .026 (.028) N 2,252 2,246 2,245 2,201 2,200 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 149

149 Table 76. Negative Binomial Regression Predicting Combined Crime Victimization (Gan g versus Non Gang Samples) Combined Crime Victimization (Gang versus Non Gang Samples) Model 1: Demographics Model 2: Self Control Model 3: Social Disorganization Model 4: All Variables Gang Sample Non Gang Sample Gang Sample Non Gang Sample Gang Sam ple Non Gang Sample Gang Sample Non Gang Sample Male **.451 (.157) *** .234 (.055) **.442 (.157) *** .249 (.055) **.470 (.160) *** .233 (.056) **.457 (.159) *** .251 (.056) White .054 (.124) *** .199 (.050) .074 (.125) **.152 (.051) .122 (.128) *** .270 ( .053) .162 (.130) *** .217 (.054) Hispanic .329 (.131) .130 (.063) ** .334 (.131) .115 (.063) ** .399 (.136) .101 (.066) ** .415 (.136) .088 (.066) Age .004 (.007) .003 (.002) .003 (.007) .002 (.002) .004 (.007) .00 3 (.002) .003 (.007) 002 (.002) Self Control .150 (.119) *** .238 (.051) .219 (.125) *** .231 (.053) Physical Disorder .096 (.127) .059 (.058) .118 (.128) .069 (.058) Social Disorder .033 (.116) *.137 (.057) .033 (.117) .105 (.058) Collective Efficacy 0 80 (.080) .003 (.034) .070 (.080) .020 (.034) Poverty *** .181 (.056) .027 (.024) *** .184 (.057) .024 (.024) Unemployed *** .221 (.064) **.078 (.025) *** .236 (.064) *** .081 (.025) Residential Mobility .034 (.063) .034 (.026) .018 (.064) .037 (.026) Racial Heterogeneity .134 (.071) .014 (.030) .142 (.072) .012 (.030) N 342 1,904 342 1,903 337 1,864 337 1,863 *p < .05, **p < .01, ***p < .001 Standard Errors in Parentheses

PAGE 150

150 CHAPTER 8 CONCLUSIONS Discussion This research examined the relationships between gang membership, crime perpetration and victimization using two theoretical explanations (self control and social disorganization). Using self report survey data from jail inmates, results suggest ed several important a nd statistically sig nificant relationships. Given the number of models presented, the following highlights the key findings and discusses the findings in the context of prior research. First, the demographic variables revealed interesting relationship s wi th regard to gang membership, crime perpetration, and victimization. Gang members were primarily male (85%) which was consistent with some prior re search (NYGC, 2007b), but suggested a lower rate of female gang members than other research has found (Esben sen & Winfree, 1998; Gover et al., forthcoming 2009) Sex differences were observed in the majority of the models and findings indicated that, generally, men (both gang and non gang members) were significantly more likely than women to be both perpetrator s ( Durose & Langan, 2007 ; James, 2004; Kyckelhahn & Cohen, 2008; Sabol et al., 2007; U.S. Department of Justice, 2007 ) and victims of crime ( Craven, 1997; Rand, 2008 ). The findings suggest ed some racial differences among gang members, perpetrators and vi ctims, although the findings we re less straightforward than examining sex differences. The racial composition of gang members indicated that Whites (37%), Blacks (36%), and Hispanics (29%) were almost equally repres ented. This finding contradicted some r esearch which indicat ed that non Whites we re more likely to be gang members ( NYGC, 2007b) and supported other research that suggests similar participation in gangs by Whites, Blacks, and Hispanics ( Esbensen & Winfree, 1998). In terms of race and involveme nt with crime ( perpetration and victimization ) Whites were more likely than non Whites to perpetrate property

PAGE 151

151 and combined crime and were more likely to be victimized by property, personal, and combined crime. These finding s were inconsistent with prior research (Rand, 2008; Rennison, 2001). However, when self control was controlled for, non Whites were more likely than Whites to perpetrate personal crimes. Results regarding ethnicity revealed that gang members were more likely to be Hispanic but that n on Hispanics were more likely to perpetrate property, personal, and combined crime and more likely to be victims of property and overall crime. While only one of the five models predicting personal crime victimization indicated that non Hispanics were mor e likely to be victims, the majority of the models did not show significant relationships regarding ethnicity For age, younger inmates were generally more likely to be gang members, property, personal and combined crime offenders as well as personal cri me victims, which was supported by prior research ( Klaus & Rennison, 2002 ; Rand, 2008 ). Yet, contrary to prior research, older inmates were more likely to be property crime victims. G ang members we re significantly more likely than non gang members to be perpetrators of crime (property, personal, and combined), which was supported by the work of prior research ( Cohen, 1969; Decker & Van Winkle, 1996 ; Esbensen & Huizinga, 1993; Hagedorn, 1988; K lein, 1971; Maxson & Klein, 2006; Miller, 1966 ; Thornberry et al ., 1993; Thrasher, 1927; Vigil, 1988). G ang members we re also more likely than non gang members to report being victimized by crime (property, personal, and combined), a finding which was supported by most prior research ( Gover et al., 2009; Peterson e t a l., 2004; Taylor et al., 2007) and contradicted by only one study (Gibson et al., forthcoming 2009) Given that gang members we re not only more deeply entrenched in c ommitting crime, but that they were also more likely to be victimized by crime has import ant implications for research and policy While research has explored the time ordering of the gang perpetration link, suggesting overall support for the facilitation model

PAGE 152

152 (Thornberry et al., 1993), prior research has not yet established a solid understa nding of the ways in which victimization affect gang membership. Recent work by Peterson et al. (2004) showed support for an enhancement model, whereas gang members we re more likely than non gang members to experience victimization before membership and m ore likely to be victimized at higher rates during gang membership. Especially in light of the research that suggests gang members may join gangs for protection (Peterson et al., 2004), it is important to disentangle the gang victimization link. It is al so important to examine this link using a variety of crime victimization measures, give n that the current study revealed gang members we re victimized by a variety of crime types. Examining the effects of self control revealed interesting findings that su pport ed the theorys ability to explain offending (Gottfredson & Hirschi, 1990) and some types of victimization (Schreck, 1999; Stewart et al., 2004) Findings revealed that lower s elf control was predictive of gang membership, crime perpetration (propert y, personal, and combined) and personal crime victimization but not property victimization More specifically, gang members, offenders, and victims of personal crimes were more likely to have lower self control than non gang members and individuals who do not report crime perpetration and victimization. This finding was in line with prior research ( Schreck 1999; Stewart et al., 2004) and suggests important theoreti cal advancements (described in the next section) Findings regarding social disorgani zation were less straightforward. Across models examining the full sample many of the social disorganization variables were predictive of gang membership, crime perpetration, and crime victimization. High levels of social disorder and collective efficac y perceptions were associated with gang membership. Crime perpetration (property, personal, and combined ) was related to lower levels of perceived physical disorder

PAGE 153

153 collective efficacy, and higher levels of perceived social disorder. Property crime perp etration was also associated with higher perceptions of racial diversity whereas less racial diversity was predictive of personal crime perpetration. Higher levels of perceived social disorder were predictive of the combined and personal victimization mod els whereas higher racial heterogeneity was significantly related to property crime victimization. It is important to note that some of the social disorganization variables behave d in ways that contradicted social disorganization theory. Pratt and Cullen (2005) also point out that some social disorganization va riables, such as unemployment, we re significant but in ways that counter the theory. Similarly, Sampson and Groves (1989) found that ethnic heterogeneity was positively (rather than negatively) ass ociated with personal crime victimization. For example, social disorganization theory suggests that higher levels of physical disorder and racial heterogeneity are related to higher crime rates (Sampson & Raudenbush, 2001). That lower levels of physical disorder and racial diversity were related to crime perpetration in the current study is curious. Turning to a speculative perspective, it is possible that inmates who reported involvement in crime per ceived neighborhood physical disorder as unproblematic for a variety of reasons. T he measurement of disorder may have been interpreted by respondents in multiple ways. For example, t he survey questions asked respondents to determine how much of a problem they believed each type of physical disorder was in their neighborhood. This method of assessing perceptions of disorder combined both the presence of the disorder with the respondents perceptions of the disorder type (see Skogan & Maxfield, 1981). Therefore, a response of not a problem to any of the disorder items may indicate at least two possibilities: (1) respondents believe d the disorder type wa s not prevalent and not problematic, (2) respondents believe d the disorde r type wa s prevalent but not problematic Offenders may have

PAGE 154

154 perceived even high levels of physical disorder in their neighborhood as normal rather than abnormal. Alternatively, offenders may have been reluctant to identify their neighborhoods physical disorder as problematic due to their personal contribution to th e disorder. The f inding that less racial diversity within inmates neighborhoods was associated with offending may also be attributed to its measurement. As described as a measurement limitation (Chapter 3), inmates were asked to self interpret the spatial distribution of their neighborhood. It is possible that while respondents lived in racially diverse areas, they considered only a portion of that area (the portion as racially similar to themselves) as their neighborhood. Due to the potential problems associated with the measures of social disorder these results sho uld be interpreted with caution. Returning to a discussion about gang membership, the current study examined self control and social disorganization between gang and non gang members in an effort to determine differences among the groups. Results revealed that self control and social disorganization were related to crime perpetration (property, personal, and combined) for both gang and non gang members. Interestingly, entering both self control and social diso rganization into the models did not eliminate the explanatory influence of either theory. Therefore, both self cont rol and social disorganization we re predictive of offending for both gang and non gang members. Models predicting crime victimization reveal ed some differences with regard to the theoretical variables for gang and non gang members. Self control was not predictive of property or combined victimization for gang members whereas personal crime victimization was significantly associated with lower self control. Alternatively, low self control was predictive of victimization (property, personal, and combined) for non gang members. Theoretical Implications The current study found general support for both theoretical perspectives, although m ore support was generated for self control theory in comparison to social disorganization theory.

PAGE 155

155 With one exception, all of the hypotheses regarding self control were confirmed. More specifically, low self control was related to gang membership, propert y crime perpetration, personal crime perpetration, combined crime perpetration, personal crime victimization, and combined crime victimization. However, low self control was not predictive of property crime victimization. Social disorganization produced less straightforward findings for several reasons. First, unlike self control, social disorganization was measured with multiple variables which has the benefit of detecting specific theoretical factors of most importance and the drawback of allowing for other (less important) variables to be non significant which inevitably renders the final conclusion as offering some support for the theory. Second, while not all of the social disorganization variables were significantly predicted gang membership, cri me perpetratio n, and crime victimization, variables that were significant change d among the models. In other words, a specific component of social disorganization did not emerge as being most important. Instead, each of the seven theoretical variables we re significant on at least one occasion. Third, some of the social disorganization variables were statistically significant but in opposite ways which only added to the potential confusion about the interpretation of social disorganization theory. For ex ample, perceptions of physical disorder were negatively related with property and personal crime perpetration. Additionally, perceived racial heterogeneity was positively related with property crime perpetration and victimization yet negatively predictive of personal crime perpetration. Perceptions of unemployment were negatively related to property crime perpetra tion, which contradicts social disorganization theory s tenets yet is consistent with findings from other research on social disorganization th eory (Pratt & Cullen, 2005). While not all of the self control and social disorganization hypotheses were confirmed the current study found general support for both theories when examined separately and jointly.

PAGE 156

156 When testing the theories together both yielded statistically significant findings In other words, it can be concluded that both self control and social disorganization are important for explaining gang membership, crime perpetration, and crime victimization. In terms of theory testing, e xamin ing theories in this way can clearly provide great detail about the main effects of the theory when examined individually and it can also identify weaknesses with the theory in the event that a different theory renders it unimportant. The findings from the current study support the advancement of theory to explain crime victimization in addition to crime perpetration. With few exceptions (i.e., routine activities theory), most theories are not designed to explain crime victimization. Yet similarities between crime perpetration and crime victimization ( Lauritsen et al., 1991, 1992; Lauritsen, & Laub, 2007; Schreck et al., 2008) suggest that criminological theories may successfully explain committing and experiencing crime. Explaining criminal behavior is important in order to prevent or reduce crime (and, subsequently, victimization). Similarly, it is just as important for criminological theories to explain crime victimization in order to identify factors that increase ones risk of experiencing crime and, therefore, experiencing the physical, financial, and psychological effects of crime (Karmen, 2009). Theory Based Policy Implications The findings from the current study emphasize the importance of policies that target individuals with low self control and disorganized neighborhoods Based on the findings of this study, policies and programs may be targeted to individuals or groups who are most atrisk of gang membership or crime victimization. Given that many of the social disorganization variables w ere predictive of gang membership, crime perpetration, and crime victimization, programs focused on disorganized neighborhoods may reduce gang membership, crime, and victimization ( Sherman, Gottfredson, MacKenzie, Eck, Reuter, & Bushway 1998) An

PAGE 157

157 extensi ve amount of research has examined the effects of implementing community based programs to reduce crime. In one of the most comprehensive and methodologically robust evaluations of programs designed to prevent crime, an assessment funded by the National I nstitute of Justice generated no evidence of any community based programs that succe ssfully prevent crime (Sherman et al. 1998). In fact, some community based programs were specifically identified by Sherman et al. (1998) as ineffective at reducing crime including police organized neighborhood watch programs and community mobilization in high crime low income areas. While much research suggests that community based efforts to reduce crime are ineffective several promising programs targeting communit ies have received some success. For example, Sherman et al. (1998) identify several community based programs that have been successful at reducing crime, including: monitoring gang members by community workers, community based after school programs, and B ig Brothers/Big Sisters of America. In terms of community based policies directed toward reducing gang crime specifically, the Spergel Model has generated much support (Fearn, Decker, & Curry, 2006). The Spergel Model was developed by Irving Spergel and is also known as the Comprehensive Community Wide Approach to Gang Prevention, Intervention, and Suppression Program (Klein & Maxson, 2006). Five components comprise the Spergel Model, including (1) mobilizing community members (groups and individuals) to organize programs focused on gangs, (2) outreach efforts designed to help gang members become connected to school, criminal justice agencies, etc., (3) reintegrating gang members into the community by assisting with job training, employment, etc., (4) fac ilitating policy changes in an effort to encourage positive treatment of gang members by public and private agencies, and (5) promoting official agencies designed to suppress gang membership and

PAGE 158

158 gang crime (Klein & Maxson, 2006) A preliminary evaluation of the Spergel Model by Spergel and Grossman (1997) revealed overall support for the program in terms of reduced violent crime (but see Klein & Maxsons, 2006, discussion of the limitations of this model). Given that s elf control predicted gang membership, crime perpetration, and personal crime victimization p rogramming designed to teach youth to exhibit high self control may be effec tive in reducing gang membership, crime, and personal crime victimization Since self control is established during youth ( by ages 8 to 10), teaching children and new parents to instill high self control would be a valuable use of resources. School curriculums, after school programs, mentoring programs, and involvement in extracurricular activities (e.g., sports, music, art, dance) may encourage youth to adopt high self control. Evaluations of several school based programs indicate d a reduction in crime and delinquency, including implementing campaigns (e.g., anti bullying), communicating and reinforcing rules, and teaching l ife skills such as stress management and self control (Sherman et al., 1998). Given that parenting practices are primarily responsible for children s self control (Gottfredson & Hirschi, 1990) it may be especially important to focus efforts on teaching pa rents proper parenting skills (e.g., supervision, punishment, warmth). Sherman et al. (1998) identifies several family based programs that have been successful in reducing delinquency including frequent visits by nurses for parents with infants, weekly pr eschool and home visits by teachers, and family therapy (Sherman et al., 1998). Limitations and Suggestions for Future Research While the current study is a step forward in furthering our theoretical underst anding of the crime and victimization among gang members some limitations should be noted. In addition to the measurement issues noted in Chapter 3, several other methodological limitations suggest that the findings should be interpreted with caution. G iven the nature of survey research, the sample

PAGE 159

159 wa s comprised of self selected volunteers with a lower response rate which may limit the generalizability of the findings. It is possible that inmates were influenced to or from participating after learning that the survey focuses on the sensitive issues o f gangs, victimization, and offending. Furthermore the current study does not include a non incarcerated comparison group. However, it was determined that incarcerated offenders (jail inmates) would best serve as a comparison group for incarcerated gang members. While many of the social disorganization variables were predictive of gang membership, crime perpetration, and victimization, some of the variables behaved in ways that contradicts social disorganization theory. Specifically, physical disorder was negatively associated with crime perpetration rather than positively associated, as social disorganization theory suggests. In other words, less (rather than more) physical disorder was predictive of crime perpetration. Future research could disenta ngle the peculiar effects of physical disorder by examining each of the physical disorder items individually rather than collectively as a scale Furthermore, future research may gain a deeper understanding of the nature of jail/prison gangs by asking gang members whether they joined while incarcerated. The current study is unable to determine whether gang members had joined gangs before or during incarceration, which may limit the extent to which some theories may account for gang membership (i.e., soc ial disorganization). While the sample was primarily male (75%), female inmates were over represented in the sample compared to the female populations for some jails (see Table 4 4). Given that men and women differ with regard to their involvement with g angs (Thrasher, 1927; NYGC, 2007b), crime perpetration ( Durose & Langan, 2007; James, 2004; Kyckelhahn & Cohen, 2008; Sabol et al., 2007; U.S. Department of Justice, 2007 ), and victimization ( Craven, 1997; Gover et al., 2008; Nobles et al., forthcoming 2009; Rand, 2008), it is important for future

PAGE 160

160 research to examine the relationships among gangs, crime perpetration, victimization, self control, and social disorganization separately among men and women. Additionally the current study represents a cross sec tional design, which may render some causal time ordering among variables problematic For example, it is impossible to determine changes over time with regard to variables such as self control and perceptions of neighborhood disorganization. Therefore, the current study is unable to address the causal order between these variables and others. However, the cr osssectional survey questions are equipped to examine the time order between (1) gang membership and crime victimization and (2) gang membership an d crime perpetration.1 While some prior research has examined the temporal ordering between gang membership and crime perpetration ( Gordon et al., 2004; Thornberry et al., 1993), the temporal ordering between gang membership and crime victimization is les s understood. Given the recent literature that suggests gang membership is associated with crime victimization (Gover et al., 2009; Peterson e t al., 2004; Taylor et al., 2007) in addition to the findings from the current study it is important for future research to determine the extent to which victimization occurs before, during, and/or after gang membership. Examining when victimization and gang members hip occurs may reveal important reasons for joining a gang (e.g., victimization facilitated versus v ictimization enhanced gang membership). This knowledge may provide researchers and practitioners with a clearer understanding of the dynamics of the lives of gang members as well as potential starting points for reducing gang membership. 1 Survey questions allow ed gang members to indicate when (and how many times) each crime occ urred (as victimization and as perpetration experiences) before, during, and/or after their gang membership

PAGE 161

161 APPENDIX A MAP OF FLORIDA COUNTY JAILS CONTACTED Administrators from c ounties with a single strike through were unresponsive to requests to participate and were not included in the sa mple. Administrators from c ounties with a double strike through declined to participate and were not included in the sample. Leon Alachua Duval Hillsborough Miami Dade Orange Escambia Marion Polk Manatee Sarasota Broward Palm Beach Collier Lee Pinellas Pasco Volusia Seminole Brevard

PAGE 162

162 APPENDIX B SURVEY (ENGLISH) SURVEY University of Florida Department of Sociology and Crimin ology

PAGE 163

163 Please circle one answer for each of the following questions. When answering these questions, think about your neighborhood (outside of this jail). Before you came to this jail, in your opinion, how much of a problem in your neighborhood was No t a problem Some problem A big problem 1. Garbage on the streets? 1 2 3 2. Graffiti? 1 2 3 3. Abandoned cars? 1 2 3 4. Needles and syringes used for drugs? 1 2 3 5. Kids hanging out when they should be at school (truant)? 1 2 3 6. People vandal izing other peoples property? 1 2 3 7. People hanging around with nothing to do (loitering)? 1 2 3 8. People drinking alcohol in public places? 1 2 3 9. People drunk in public places? 1 2 3 10. People who look ed like they were selling drugs? 1 2 3 11. People using illegal drugs? 1 2 3 12 People who look ed like they were in a gang? 1 2 3 13. Buildings or storefronts sitting abandoned or burned out? 1 2 3 Please circle one answer for each of the following questions. When answering these q uestions, think about your neighborhood (outside of this jail). Before entering this jail, in your opinion, generally how likely was it that Very Unlikely Somewhat Unlikely Somewhat Likely Very Likely 1 4 Your neighbors would do something if they saw unattended kids misbehaving? 1 2 3 4 15 Your neighbors would be willing to help each other? 1 2 3 4 16 You could trust your neighbors? 1 2 3 4

PAGE 164

164 Please circle one answer for each of the following questions. When answering these questions, thin k about your neighborhood (outside of this jail). 17. About how many of your neighbors live in poverty? 1 None 2 Very few 3 About half 4 More than half 5 I dont know 18. About how many of your neighbors are unemployed? 1 None 2 Very few 3 About half 4 More than half 5 I dont know 19. About how often do your neighbors move away? 1 Rarely 2 Occasionally 3 Often 4 I dont know 20. About how racially mixed is your neighborhood? 1 Not very mixed (almost all of the neighbors are of the sam e race) 2 Somewhat mixed (most of the people are of the same race and there are some other races) 3 Very mixed (there are people from many different races) 4 I dont know 21. Are you currently or have you ever been in a gang? (circle one and complete t he blanks) 1 I am not in a gang now and I have never been in a gang

PAGE 165

165 2 I am not in a gang now, but I have been in a gang in the past. I was a gang member between the ages of ______________ and _______________ 3 I am in a gang now and have been since I was __________ years old If you answered no on the last question (if you are not in a gang now and have never been in a gang) you may skip to question # 36. If you are or have been a gang member, please circle one a nswer for each of the following I am not a gang member Yes No 22 Does your gang have initiation/joining rites? 1 2 3 23 Have you be en jumped or beaten in to a gang? 1 2 3 24 Does your gang have l eaders? 1 2 3 25 Are you a gang leader ? 1 2 3 26 Does your gang have a name? 1 2 3 27. Do you have a moniker or nickname within the gang ? 1 2 3 28 Does your gang have symbols or colors? 1 2 3 29 Does your gang have hand signs? 1 2 3 30 Were you in a gang before entering this jail ? 1 2 3 31 Are you now a me mber of the same gang you belonged to before entering this jail ? 1 2 3 32 When you get out of jail do you plan to stay in the gang? (circle one) 1 I am not in a gang 2 I plan to stay in the gang 3 I would like to get out of the gang 4 I will ge t out of the gang 5 I would like to get out of the gang but cant 6 I dont know

PAGE 166

166 33 Is your gang i nside this jail, outside this jail or both inside and outside? (circle one) 1 I am not in a gang 2 Inside the jail only 3 Outside the jail only 4 Both inside and outside the jail 34 Why did you first join a gang? (circle ALL that apply) 1 I have never been in a gang 2 Friends were gang members 3 Family were gang members 4 Protection 5 Respect 6 Money 7 For fun 8 Other 35 After you joined a gang, what was good about it ? (circle ALL that apply) 1 I have never been in a gang 2 Friends 3 Family acceptance 4 Protection 5 Respect 6 Money 7 For fun 8 Other

PAGE 167

167 Please circle one answer for each of the following questions. Plea se rate yourself on the following items Strongly Agree Agree Disagree Strongly Disagree 36 I often act on the spur of the moment without think ing 1 2 3 4 37 I like to get out and do things more than I like to sit around. 1 2 3 4 38 Often, wh en Im angry at people I feel more like hurting them than telling them why I am angry. 1 2 3 4 39 Sometimes I will take a risk just for fun. 1 2 3 4 40 If I had a choice, I would almost always rather do something physical than something mental. 1 2 3 4 Strongly Agree Agree Disagree Strongly Disagree 41 If things I do upset people, its their problem not mine. 1 2 3 4 42 The things in life that are easiest to do are the most fun 1 2 3 4 43 I often look out for myself first, even if it makes it hard for other people 1 2 3 4 44 I dont think much about the future. 1 2 3 4 45 I like to do things that might get me in trouble. 1 2 3 4 46 I get mad easily. 1 2 3 4 47 I dont care so much when other people are having problems. 1 2 3 4 48 I like to test myself by taking risks every once in a while 1 2 3 4 49 I often try to avoid things that will be hard 1 2 3 4

PAGE 168

168 50 When Im really angry, other people better stay away from me. 1 2 3 4 51 I often do whatever is fun now even at the cost of some distant goal. 1 2 3 4 52 I almost always feel better when I am on the move than when I am sitting and thinking. 1 2 3 4 53 When I have trouble with someone, its hard for me to talk calmly about it without gett ing upset. 1 2 3 4 54 Im more concerned with what happens to me in the short run than in the long run. 1 2 3 4 55 I will try to get the things I want even when I know it makes other people upset 1 2 3 4 56 Excitement and adventure are more im portant to me than security. 1 2 3 4 57 I dont like really hard jobs that push me 1 2 3 4 58 When things get hard I tend to quit. 1 2 3 4 The following questions ask about your experience s as the VICTIM of crime. Remember, all of your answers are anonymous and no one can link your answers to you. 59. Has someone ever stolen money or property from you without using force? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A membe r from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 169

169 60. Has someone ever used a weapon or force to steal money or property from you? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it h appened _______ times 5 I dont know 61. Has someone ever damaged or vandalized your property? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply ) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 62. Have you ever been threatened with a weapon? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I wa s in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 170

170 63. Have you ever been attacked without a weapon? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 Afte r I was in a gang it happened _______ times 5 I dont know 64. Have you ever been attacked with a weapon? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 65. Have you ever been sexually assaulted or raped ? (A) (B) (C) Circle one How many time s and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 171

171 66. Have yo u ever been stabbed? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I w as in a gang it happened _______ times 5 I dont know 67. Have you ever been the victim of a carjacking? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 68. Have you ever been threatened by someone so you did not act as a witness in court? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ time s 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 172

172 69. Have you ever been the victim of a home invasion ? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened ____ ___ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 70. Have you ever been the victim of a drive -by shooting (shot or shot at )? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 71. Have you ever been shot at but not hit (not military -related) ? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another g ang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 173

173 72. Have you ever been shot (not military -related) ? (A) (B) (C) Circle one How many times and when did it happen? Who did this to you? (answer al l that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know The following questions ask about your experience s with COMMITTING crime s Rem ember, all of your answers are anonymous and no one can link your answers to you. 73. Have you ever stolen money or property from someone without using force? (A) (B) (C) Circle one How many times and when did it happen? Who did you d o this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 74 Have you ever used a weapon or force to steal money or property from someone? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 174

174 75 Have you ever damaged or vandalized someones property? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answ er all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 76 Have you ever threatened someone with a weapon? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ tim es 5 I dont know 77 Have you ever attacked someone without a weapon? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 175

175 78 Have you ever attacked someone with a weapon? (A) (B) (C) Circle one How many times and when did it happen? Wh o did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gan g it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 79 Have you ever sexually assaulted or raped someone? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I w as in a gang it happened _______ times 5 I dont know 80 Have you ever stabbed someone? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that appl y) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 176

176 81 Have you ever carjacked someone? (A) (B) (C) Circle one How many times and when did it happen ? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was i n a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 82. Have you ever threatened someo ne that you did not want to act as a witness in court? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 83. Have you ever committed a home invasion ? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 177

177 84 Have you ever participated in a drive -by shooting? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened __ _____ times 5 I dont know 85 Have you ever shot at someone but not hit them (not military -related) ? (A) (B) (C) Circle one How many times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a gang member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 Whi le I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know 86 Have you ever shot someone (not military-related) ? (A) (B) (C) Circle one How man y times and when did it happen? Who did you do this to you? (answer all that apply) (circle all that apply) 1 No 1 It never happened 1 It never happened 2 Yes 2 I was never in a gang and it happened _______ times 2 It was not a ga ng member 3 Before I was in a gang it happened _______ times 3 A member from my gang 4 While I was in a gang it happened _______ times 4 A member from another gang 5 After I was in a gang it happened _______ times 5 I dont know

PAGE 178

178 87. How afraid are you that your involvement with crime will cause your family to be victimized? 1 Very afraid 2 Afraid 3 Somewhat afraid 4 Not afraid How much control do you think each of the following has OVER YOU ? No control Some control Lots o f control Total control Doesnt apply to me 88 Job/employment 1 2 3 4 5 89 Relationships with significant others 1 2 3 4 5 90 Other people (neighbors) 1 2 3 4 5 91 Society as a whole 1 2 3 4 5 9 2 Recreational or fun activities 1 2 3 4 5 How much control do you think YOU have OVER each of the following? No control Some control Lots of control Total control Doesnt apply to me 93 Job/employment 1 2 3 4 5 94 Relationships with significant others 1 2 3 4 5 95 Other people (neighbors) 1 2 3 4 5 96 Society as a whole 1 2 3 4 5 9 7 Recreational or fun activities 1 2 3 4 5

PAGE 179

179 Please circle one answer for each question. While outside of this jail, h ow likely is it that in the future you will actually Very Unlikely So mewhat Unlikely Somewhat Likely Very Likely 98. Have your money or property stolen from you without force? 1 2 3 4 99. Have your money or property stolen from you with force or by using a weapon? 1 2 3 4 100. Have your property damaged or vandal ized? 1 2 3 4 101. Be threatened with a weapon? 1 2 3 4 102. Be attacked without a weapon? 1 2 3 4 103. Be attacked with a weapon? 1 2 3 4 104. Be sexually assaulted or raped ? 1 2 3 4 105. Be stabbed? 1 2 3 4 106 Be the victim of a carjacking? 1 2 3 4 107. Be threatened by someone who did not want you to act as a witness in court ? 1 2 3 4 108. Be the victim of a home invasion? 1 2 3 4 109 Be shot at, but not hit ? 1 2 3 4 110. Be shot? 1 2 3 4 111. Have your propert y damaged by gang graffiti or tagging ? 1 2 3 4 112. Have someone break into your home while you are away? 1 2 3 4 113. Have a gang member commit a home invasion robbery against you ? 1 2 3 4 114. Be a victim of a drive by or random gang related sh ooting? 1 2 3 4 115. Be attacked or assaulted by a gang member ? 1 2 3 4 116. Be harassed by gang members ? 1 2 3 4 117. Be killed? 1 2 3 4

PAGE 180

180 Please circle one answer for each of the following questions. While outside of this jail, h ow perso nally afraid are you of the following crimes Not Afraid Somewhat Afraid Afraid Very Afraid 118 Having your money or property stolen from you without force? 1 2 3 4 119 Having your money or property stolen from you with force or by using a weapon? 1 2 3 4 120 Having your property damaged or vandalized? 1 2 3 4 121 Being threatened with a weapon? 1 2 3 4 122 Being attacked without a weapon? 1 2 3 4 123 Being attacked with a weapon? 1 2 3 4 124 Being sexually assaulted or ra ped ? 1 2 3 4 125 Being stabbed? 1 2 3 4 126 Being the victim of a carjacking? 1 2 3 4 127 Being threatened by someone who did not want you to act as a witness in court ? 1 2 3 4 128. Being the victim of a home invasion? 1 2 3 4 129 Be ing shot at, but not hit ? 1 2 3 4 130. Being shot? 1 2 3 4 131 Having your property damaged by gang graffiti or tagging ? 1 2 3 4 132 Having someone break into your home while you are away ? 1 2 3 4 133 Having a gang member commit a home invas ion robbery against you ? 1 2 3 4 134 Being a victim of a drive by or random gang related shooting ? 1 2 3 4 135 Being attacked or assaulted by a gang member ? 1 2 3 4 136 Being harassed by gang members ? 1 2 3 4 137. Being killed? 1 2 3 4

PAGE 181

181 138. What is your sex? (circle one) 1 Male 2 Female 139 What is your race? (circle one) 1 White 2 African American or Black 3 Asian 4 Other 140 Are you Hispanic? (circle one) 1 No 2 Yes 141 How old are you? (please specify) ______________ years old 142 How many children do you have? (please specify) _________ 143. As of today, about how long have you been in this jail? (please specify) 1 0 to 90 days (1 3 months) 2 91 to 180 days (3 6 months) 3 181 to 270 days (6 9 months) 4 271 to 365 days (9 12 months) 5 More than 365 days (over 1 year) 144. What was the last grade of school you finished? (circle one) 1 Grades 0 through 4 2 Grades 5 through 8 3 Grades 9 through 11 (some high school) 4 Grade 12 (high school graduate/GED completion) 5 Some college 6 College graduate

PAGE 182

182 7 Graduate work 145. What type of family did you MOSTLY live with while you were growing up? (circle one) 1 I lived with my two biological parents 2 I lived with a single parent 3 I lived with a parent and a step-parent 4 I lived with adoptive parents 5 I lived with other relatives (grandparents, aunt/uncle, siblings, etc.) 6 I lived with other people 146. How long is your jail sentence? 1 0 to 90 days (1 3 months) 2 91 to 180 days (3 6 months) 3 181 to 270 days (6 9 months) 4 271 to 365 days (9 12 months) 5 More than 365 days (over 1 year) 6 I dont know 147. Why are you in jail? (circle ALL that apply) 1 Waiting for my trial 2 Sentenced to a short jail term 3 Probation/parole violation 4 Escaped while on bail 5 Awaiting transfer to mental health facility 6 Awaiting transfer to prison 7 Awaiting transfer to another jail 8 Held in contempt of court 9 Released from prison 10 Court witness

PAGE 183

183 11 Other 148. What type(s) of crime(s) are you NOW charged with or convicted of? (circle ALL that apply) 1 Property crime (burglary, theft, arson, shoplifting, vandalism, etc.) 2 Personal crime (assault, robbery, sex crimes, homicide, etc.) 3 Drug cr ime (sales, possession, etc.) 4 Other 5 None 149 Before you entered this jail, how often were you working? (circle one) 1 Employed full -time 2 Employed part -time 3 Seasonally employed 4 Temporarily employed 5 Unemployed/Not legally employed 1 50 Before you entered this jail, what was your current relationship status? (circle ALL that apply) 1 Not currently dating 2 Sometimes dating 3 Steady/exclusively dating 4 Married 5 Divorced 6 Other 151 How often did you choose not to walk alone in your neighborhood during the day because you were afraid of being victimized? 1 Never 2 Sometimes 3 Often

PAGE 184

184 4 Always 152 How often did you choose not to walk alone in your neighborhood during the night because you were afraid of being victimized? 1 Never 2 Sometimes 3 Often 4 Always 153 Before you entered this jail, how much money did you typically make in a year? (circle one) 1 Under $5,000 2 $5,000 $9,999 3 $10,000 $14,999 4 $15,000 $24,999 5 $25,000 $34,999 6 $35,000 $49,999 7 $50,000 $74,999 8 $75,000 and over 9 I dont know This is the end of the survey. Thank you for taking the time to respond to these questions!

PAGE 185

185 APPENDIX C SURVEY (SPANISH) ENCUESTA Universi dad de la Florida Department o de Sociol oga y Criminologa

PAGE 186

186 Por favor cir c ule una respuesta para cada una de las siguientes preguntas.. Cuando conteste estas preguntas, piense en su vecindario (fuera de esta crcel ). Antes de que usted entrara a esta crcel, en su opinin, qu tan to problema en su vecindario era Ningno problema Algo de problema Un gran problema 1. B a sura en las call es? 1 2 3 2. Graffiti? 1 2 3 3. Carros a bandon a d o s? 1 2 3 4. Agujas y jeringas usadas para drogas? 1 2 3 5. Nios en la calle cuando debieran estar en la escuela? 1 2 3 6. Personas vandalizando la propiedad de otras personas? 1 2 3 7. Personas en las calles sin nada que hacer u vagabundeando ? 1 2 3 8. Personas tomando alcohol en lugares pblicos? 1 2 3 9. Personas borrachas en lugares pblicos? 1 2 3 10. Personas que parece n que estaban vendiendo drogas? 1 2 3 11. Personas usando drogas ilegales? 1 2 3 12 Personas que parecen que estaban en una pandilla? 1 2 3 13. Edificios abandonados o derrumbndose ? 1 2 3 P or favor cir c ule una respuesta para cada una de las siguientes preguntas.. Cuando conteste estas preguntas, piense en su vecindario (fuera de esta crcel ). Antes de que usted entrara a esta crcel, en su opinin, generalmente qu tan probable era que Mu y Imp robable Algo Improbable Algo Probable Muy Probable 1 4 Sus vecinos haran algo si vieran a un nio sin supervisin portarse mal? 1 2 3 4 15 Sus vecinos estaran dispuestos a ayudarse el uno al otro? 1 2 3 4 16. Usted puede confiar en sus vecinos? 1 2 3 4

PAGE 187

187 Por favor cir c ule una respuesta para cada una de las siguientes preguntas. Cuando conteste estas preguntas, piense en su vecindario (fuera de esta crcel ). 17. Cuntos de sus vecinos viven en la pobreza? 1 Ninguno 2 Muy pocos 3 Como la mitad 4 Ms de la mitad 5 Yo no se 18. Cuntos de sus vecinos estn desempleados? 1 Ninguno 2 Muy pocos 3 Como la mitad 4 Ms de la mitad 5 Yo no se 19. Con que frequencia se mudan de las viviendas sus vecinos de su vicindad? 1 Raram ente 2 Ocasionalmente 3 A menudo 4 Yo no se 20. Qu tan variado racialmente es su vecindario? 1 No muy variado (casi todos los vecinos son de la misma raza) 2 Algo variado (la mayora de las personas son de la misma raza y hay algunas otras razas) 3 Muy variado (haban personas de muchas razas diferentes) 4 Yo no se 21. Est usted actualmente o ha estado en una pandilla? (circule una y llene los espacios en blanco) 1 Yo no estoy en una pandilla ahora y nunca he estado en una pandilla

PAGE 188

188 2 Yo no he estoy en una pandilla ahora, pero yo he estado en una pandilla en el pasado. Yo fui miembro de una pandilla entre la edad de ______________ y _______________ 3 Yo estoy en una pandilla ahora y he sido miembro desde que tena __________ aos de edad Si usted con t est no a la ltima pregunta (si usted no est en una pandilla ahora y nunca ha estado en una pandilla)usted puede pasar a la pregunta # 36. Si usted es o ha sido miembro de una pandilla, por favor circule una respuesta para cada una de las siguientes Yo no soy Miembro de una pandilla Si No 22 Tiene su pandilla ritos para hacerse miembro o de iniciacin ? 1 2 3 23. Ha sido usted golpeado para pertenecer a una pandilla como iniciacin ? 1 2 3 24 Tiene lderes su pandilla ? 1 2 3 25 Es usted lder de una pandilla? 1 2 3 26 Tiene un nombre su pandilla? 1 2 3 27. Tiene usted un apodo o un alias dentro de la pandilla? 1 2 3 28 Tiene su pandilla smbolos o colores? 1 2 3 29 Tiene su pandilla un lenguaje de sea s manuales? 1 2 3 30 Estuvo usted en una pandilla antes de entrar a esta c rcel ? 1 2 3 31 Es usted ahora miembro de la misma pandilla a la que perteneca ant e s de entrar a esta crc el? 1 2 3 32. Cuando usted salga de la crcel, planea seguir en la pandilla? (circule one) 1 Yo no estoy en una pandilla 2 Yo tengo plan de seguir en la pandilla 3 Yo quisiera salir de la pandilla 4 Yo voy a salir de la pandilla 5 Yo quisiera salir de la pandilla pero no puedo

PAGE 189

189 6 Yo no se 33. Est su pan dilla dentro de esta crcel, afuera de esta crcel o tanto adentro como afuera? (circule una) 1 Yo no estoy en una pandilla 2 Dentro de la crcel solamente 3 Fuera de la crcel solamente 4 D entro y afuera de la crce l 34 Porqu se uni usted al pri ncipio a la pandilla? (circule TODO lo que se aplique) 1 Yo nunca he estado en una pandilla 2 Amigos eran miembros de la pandilla 3 Familiares eran miembros de la pandilla 4 Proteccin 5 Respeto 6 Dinero 7 Por Diversin 8 Otros 35 Despus de que usted se uni a una pandilla, qu era lo bueno sobre ello? (circule TODO lo que se aplique) 1 Yo nuca he estado en una pandilla 2 Amigos 3 Aceptacin Familiar 4 Proteccin 5 Respeto 6 Dinero 7 Por Diversin 8 Otr os

PAGE 190

190 Por favor cirule un a respuesta para cada una de las siguientes preguntas. P or favor califquese usted en los siguientes asuntos Muy de Acuerdo De Acuer do En Desa cuerdo Muy en Desa cuerdo 36. A menudo yo acto al calor del momento sin pensar. 1 2 3 4 37. A mi me gus ta salir y hacer cosas ms que estar quieto 1 2 3 4 38 A menudo, cuando estoy enojado con las personas me dan deseos de hacerles dao en vez de decirles porque yo estoy enojado. 1 2 3 4 39 Algunas veces yo corro riesgos solamente por diversin. 1 2 3 4 40 Si pudiera elegir, yo casi siempre hara algo fsico en vez de algo mental. 1 2 3 4 41. Si las cosas que hago molestan a las personas, es su problema no el m o 1 2 3 4 42. Las cosas en la vida que son ms fciles de hacer son las ms dive rtidas. 1 2 3 4 43 A menudo yo busco por lo mo primero, an si hace las cosas m s difciles para otras personas. 1 2 3 4 44. Yo no pienso mucho sobre el futuro. 1 2 3 4 45 Me gusta hacer cosas que pudieran meterme en problemas. 1 2 3 4 46 Me enojo fcilmente 1 2 3 4 47. No me importa mucho cuando otras personas estn teniendo problemas. 1 2 3 4 48. Me gusta probarme a mi mismo tomando riesgos de vez en cuando. 1 2 3 4 49. A menudo trato de evitar las cosas difciles. 1 2 3 4 50 C uando estoy realmente enojado, es mejor que las otrar personas est n lejos de mi 1 2 3 4

PAGE 191

191 51 A menudo hago cualquier cosa que es divertida ahora, an a costo de una meta distante. 1 2 3 4 52 Casi siempre me siento mejor cuando estoy en accin que cuando estoy sentado y pensando. 1 2 3 4 53 Cuando tengo problemas con alguien, me es difcil hablar calmadamente sin enojarme. 1 2 3 4 54. Estoy ms preocupado con lo que me pasa a corto plazo que a largo plazo. 1 2 3 4 55 Yo tratar de hacer la s cosas que yo quiero aunque sepa que molesta a otras personas. 1 2 3 4 56 La emocin y la aventura son ms importantes para mi que la seguridad. 1 2 3 4 57 No me gustan los trabajos difciles que me presionan. 1 2 3 4 58 Cuando las cosas se po nen difciles yo tiendo a renunciar. 1 2 3 4 Las siguientes preguntas se refieren a su experiencias como VICTIMA de un delito. Recuerde, todas su respuestas son annimas y nadie puede conectar su respuestas a usted. 59. Alguna vez alguien te ha robado dinero o alguna propiedad s i n us ar f ue r za ? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 N unca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 192

192 60. Alguna vez alguien ha usado un arma o fuerza para robarte dinero o alguna propiedad? (A) (B) (C) Circule Uno Cuntas v eces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pa ndilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 61. Alguna vez alguien ha vandalizado o daado su propiedad? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pan dilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 62. Alguna vez ha sido usted amenazado con un arma? (A) (B) (C) Circule Uno Cuntas veces y cundo suc edi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Ante s de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 193

193 63 Ha sido atacado alguna vez usted sin un arma? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ vec es 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 64. Ha sido usted atacado alguna vez con un arma? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 65. Ha sido alguna vez usted a saltado sexualmente o violado? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 S i 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 194

194 66. Ha sido usted alguna vez acuchillado? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 67. Ha sido usted alguna vez la vctima de un robo d e carro mientras que usted estaba dentro del carro? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 68. Ha sido usted alguna vez amenazado por alguien para que usted no sirviera como testigo en una corte? (A) (B) (C) Cir cule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No e ra miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 195

195 69. Ha sido usted alguna vez la vctima de una invaci n a su hogar ? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circul e todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 70. Ha sido usted alguna vez la vctima de un tiroteo abierto des de un carro ? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y p as ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estu ve en una pandilla pas _____ veces 5 Yo no se 71. Alguna vez a usted le han disparado pero sin darle (no relacionado con el ejrcito )? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 196

196 72. Alguna vez ha sido usted baleado (no rela cionado con el ejrcito )? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? Quin le hizo esto a usted? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se Las siguientes preguntas son sobre su experiencia COMETIENDO delitos. Recuerde, todas sus respuestas son annimas y nadie puede conectar sus respuestas con usted. 73. Alguna vez usted ha robado dinero o propiedad de alguien sin usar fuerza? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 74 Alguna vez ha usado usted un arma o fuerza para robar dinero o propiedad de alguien? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___vec es 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 197

197 75 Alguna vez ha usted daado o vandalizado la propiedad de alguien? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se ap lique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 76 Alguna vez ha usted amenazado a alguien con un arma? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 77 Alguna vez usted ha atacado a alguien sin un arma? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi p andilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 198

198 78 Alguna vez h a usted atacado a al guien con un arma? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___vec es 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 79 Alguna vez ha usted asaltado sexualmente o violado a alguien? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 80 Alguna vez ha usted acuchillado a alguien? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 199

199 81. Alguna vez ha usted robado un carro con una persona dentro del carro? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique ) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de m i pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 82. Alguna vez usted ha amenado a alguien que usted no deseaba que sirviera como testigo en una corte? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 83. Alguna vez ha usted invadido un hogar? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ ve ces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 200

200 84. Alguna vez ha usted participado en un tiroteo abie rto des de un carro ? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra p andilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 85 Alguna vez a usted disparado a alguien sin darle (norelacionado con el ej rcito )? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nunca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se 86 Alguna vez ha usted baleado a alguien (no relacionado con el ej rcito )? (A) (B) (C) Circule Uno Cuntas veces y cundo sucedi? A quin tu le hicisteis eso? (Conteste todo lo que se aplique) (Circule todo lo que aplica) 1 No 1 Nu nca pas 1 Nunca pas 2 Si 2 Yo nunca estuve en una pandilla y pas ___veces 2 No era miembro de una pandilla 3 Antes de que yo estuviera en una pandilla pas ____ veces 3 Un miembro de mi pandilla 4 Mientras estuve en una pandilla pas _____ veces 4 Un miembro de otra pandilla 5 Despus de que estuve en una pandilla pas _____ veces 5 Yo no se

PAGE 201

201 87. Qu tan temeroso est usted de que su involucramiento en delitos resulte en que su familia sea victimizada? 1 Muy temeros o 2 Temeroso 3 Algo temereso 4 Nada de temor Cunto contro piensa usted que lo siguiente tiene SOBRE USTED ? N ingn control Algn control Mucho control Control total No se aplica a mi 88 Trabajo /empl e o 1 2 3 4 5 89 Rela ciones con persona s significantes 1 2 3 4 5 90 Otr as personas (veci n o s) 1 2 3 4 5 91 Socie dad como un todo 1 2 3 4 5 9 2 Actividades recreacionales o divertidas 1 2 3 4 5 Cunto control usted piensa que USTEd tiene SOBRE cada uno de lo siguiente ? N ingn co ntrol Algn control Mucho control Control total No se aplica a mi 93 Trabajo /empl e o 1 2 3 4 5 94 Rela ciones con personas significantes 1 2 3 4 5 95 Ot ras personas (v e c i n os) 1 2 3 4 5 96 Soci dad como un todo 1 2 3 4 5 9 7 Actividades rec reacionales o divertidas 1 2 3 4 5 Por favor circule una respuesta para cada pregunta. Fuera de esta crcel, qu tan probable es que en el futuro usted Muy Improbable Algo Improbable Algo Probable Muy Probable 98. Sea robado su dinero o propieda d s in fuerza? 1 2 3 4

PAGE 202

202 99. Sea robado su dinero o propiedad con fuerza o usando un arma? 1 2 3 4 100. Sea su propiedad daada o vandalizada? 1 2 3 4 101. Sea amenazado con un arma? 1 2 3 4 102. Sea atacado sin u n arma? 1 2 3 4 103. Sea atacado con un arma? 1 2 3 4 104. Sea asaltado sexualmente o violado? 1 2 3 4 105. Sea acuchillado ? 1 2 3 4 106. Sea vctima de robo de carro mientras que usted esta dentro del carro? 1 2 3 4 107. Sea amenazado por alguien que no qui era que usted sirva como testigo en una corte? 1 2 3 4 108. Sea vctima de una invasin de hogar? 1 2 3 4 109 Le disparen pero no le den? 1 2 3 4 110. Sea baleado ? 1 2 3 4 111. Sea su propiedad daada por por graffiti de una pandilla o pa ra sealarlo a usted ? 1 2 3 4 112. Alguien entre a su casa mientras usted no est? 1 2 3 4 113. Que un miembro de una pandilla inv ada su casa para robarle ? 1 2 3 4 114. Sea vctima de un a balacera casual u tiroteo abierto desde un carro rel acionada con una pandilla? 1 2 3 4 115. Sea atacado o asaltado por um miembro de una pandilla? 1 2 3 4 116. Sea acosado por miembros de una pandilla? 1 2 3 4 117. Sea asesinado? 1 2 3 4

PAGE 203

203 Por favor circule una respuesta para cada una de l as siguientes preguntas. Fuera de esta crcel, personalmente que tanto temor tiene usted de los siguientes delitos N ada de Temor Algo de Temor Temor Mucho Temor 118 Que su dinero o propiedad sean robados sin fuerza? 1 2 3 4 119 Que su dinero o propiedad sean robado con fuerza o usando un arma? 1 2 3 4 120 Que su propiedad sea daada o vandalizada? 1 2 3 4 121 Ser amenazado con un arma? 1 2 3 4 122 Ser atacado sin un arma? 1 2 3 4 123. Ser atacado con un arma? 1 2 3 4 124 Ser asaltado sexualmente o violado? 1 2 3 4 125 Ser acuchillado ? 1 2 3 4 126 Ser vctima de un robo de carro mientras estar en el carro? 1 2 3 4 127 Ser amenado por alguien que no quiera que usted sirva de testigo en una corte? 1 2 3 4 128. Ser vctima de una invasin de hogar? 1 2 3 4 129 Ser atacado a balazos pero fallaron? 1 2 3 4 130. Ser baleado ? 1 2 3 4 131 Que su propiedad sea daada por el graffiti de una pandilla o para sealarlo a usted ? 1 2 3 4 1 32. Que alguien entre a su casa mientras usted no est? 1 2 3 4 133 Que um miembro de una pandilla cometa una invasin de hogar contra usted? 1 2 3 4 134 Ser vcitma de una balacera casual u tiroteo abierto desde un carro relacionada con una pandilla? 1 2 3 4 135 Ser atacado o asaltado por un miembro de una pandilla? 1 2 3 4

PAGE 204

204 136 Ser acosado por miembros de una pandilla? 1 2 3 4 137. Ser asesinado? 1 2 3 4 138. Cul es su sexo? (circule uno) 1 Masculino 2 Femenino 13 9 Cul es su raza? (circule una) 1 Blanco 2 Afro -Americano o Negro 3 Asitico 4 Otra 140 Es usted Hispano? (circule uno) 1 No 2 Si 141 Cul es su edad? (por favor especifique) ______________ aos 142 Cuntos hijos tiene usted? (por fa vor especifique) _________ 143. Al da de hoy, cunto tiempo ha estado usted en esta crcel? (por favor especifique) 1 0 to 90 das (1 3 meses) 2 91 to 180 das (3 6 meses) 3 181 to 270 das (6 9 meses) 4 271 to 365 das ( 9 12 meses) 5 Ms de 365 das (ms de 1 ao) 144. Cal fue el ltimo grado de la escuela que usted termin? (circule uno) 1 Grados 0 hasta 4 2 Grados 5 hasta 8 3 Grados 9 hasta 11 (algo de secundaria)

PAGE 205

205 4 Grado 12 (graduado de secundaria/graduado por madurez o GED) 5 Algo de Universidad 6 Graduado Universitario 7 Trabajo Graduado 145. En qu tipo de familia vivi usted LA MAYOR PARTE DEL TIEMPO mientras creca? (circule una) 1 Yo viv con mis dos padres biolgicos 2 Yo viv con uno de mis pa dres soltero 3 Yo viv con uno de mis padres y un padadrastro o madrastra 4 Yo viv con padres adoptivos 5 Yo viv con otros parientes (abuelos, ta/to, hermanos, etc.) 6 Yo viv con otras personas 146. Qu tan largo es tu sentencia? 1 0 to 90 d as (1 3 meses) 2 91 to 180 das (3 6 meses) 3 181 to 270 das (6 9 meses) 4 271 to 365 das (9 12 meses) 5 Ms de 365 das (ms de 1 ao) 6 Yo no se 147. Porqu est usted en la crcel? (circule TODOS lo que se aplican) 1 Esperando por mi juicio 2 Sentenciado a un trmino corto de crcel 3 Probacin/violacion de parole 4 Escapar mientras estaba bajo fianza 5 Esperando transferencia a una institucin de salud mental 6 Esperando transferencia a la prisin

PAGE 206

206 7 Esperando transferencia a otra crcel 8 Detenido por desacato a la corte 9 Liberado de la prisin 10 Testigo de la corte 11 Otro 148. De qu tipo(s) de delito (s) est usted AHORA acusado o convicto? (Circule TODOS los que se aplican) 1 Crimen contra la propiedad (hurto, incendio, robo en tiendas, vandalismo, etc.) 2 Crimen contra las Personas (asalto, robo, crimenes sexuales, homicidio, etc.) 3 Crimenes relacionados con Drogas (venta, posesin, etc.) 4 Otros 5 Ninguno 149 Antes de que usted entrara a esta cr cel, qu tan a menudo estaba trabajando? (circule uno) 1 Empleado a tiempo completo 2 Empleado a tiempo parcial 3 Empleado por pocas o estaciones 4 Empleado temporalmente 5 Desempleado/No empleado legalmente 150 Antes de qu e usted entrara a esta crcel, cul era su estado civil? (circule TODOS los que se apliquen) 1 Soltero (sin novio/novia) 2 De vez en cuando con novio/ novia 3 Noviazgo serio/exclusivo 4 Casado 5 Divorciado 6 Otro

PAGE 207

207 151 Cun a menudo usted prefiri no caminar solo en su vecindario durante el da porque usted tuvo temor de ser victimizado? 1 Nunca 2 Algunas veces 3 A menudo 4 Siempre 152 Qu tan a menudo usted prefiri no caminar solo en su vecindario durante la noche porque usted tuvo temor de ser victimizado? 1 Nunca 2 Algunas veces 3 A menudo 4 Siempre 1 53 Antes de que usted entrara a esta crcel, cuanto dinero haca tpicamente en un ao? (circule uno) 1 Menos de $5,000 2 $5,000 $9,999 3 $10,000 $14,999 4 $15,000 $24,999 5 $25,000 $34,999 6 $35,000 $49,999 7 $50,000 $74,999 8 $75,000 y ms 9 Yo no se Este es el fin de la encuesta. Gracias por tom ar tiempo para responder a estas!

PAGE 208

208 APPENDIX D INSTITUTIONAL REVIEW BOARD APPROVAL

PAGE 209

209 APPENDIX E INFORMED CONSENT FORMS English Inf ormed Consent Title: Understanding Offending and Victimization Experiences among Offenders Please read this consent document carefully before you decide to participate in this study. Purpose of the Study: To understand jail inmates experiences with ga ngs, crime, and being a victim. What You will be Asked to do in the Research Study: If you decide to be in this study, you will be asked to answer questions about your experiences with crime, victimization, and gangs. The researchers conducting this stud y are not associated with the jail and your individual answers to the survey questions will not be shared with the jail staff. PLEASE DO NOT PUT YOUR NAME ON THE SURVEY AND DO NOT GIVE THE SURVEY TO ANYONE OTHER THAN THE RESEARCHERS. Time Required: Betwe en 20 minutes and 1 hours, depending on your pace. Confidentiality: All of your answers will be anonymous. No one will be able to link your answers to you since we will not know your name. Your answers will be coded with numbers and these codes cannot be traced to you. The results of the study will present patterns of how everyone answered. It will not focus on any one persons answers. Voluntary Participation and Right to Withdraw From the Study: There are no benefits or rewards for participating in this study. This study will in no way affect how you are treated in jail. One potential risk that you may experience by participating in this research is that some of the questions might make you feel uncomfortable or may be upsetting to you. To min imize this risk, you may talk with the jails counseling services, if available. Also, you do not have to answer any questions that you do not want to answer, and you can stop participating at any time. No one will be upset or angry if you decide not to participate or if you stop participating at any time for any reason. Whom to Contact if you Have Questions About the Study: Kate Fox or Dr. Jodi Lane, Department of Sociology and Criminology, 3219 Turlington Hall, PO Box 117330, Gainesville, Florida 326117330; Te lephone : (352) 3920265; Email: katefox@ufl.edu. Whom to Contact About Your Rights as a Research Participant in the Study: UFIRB Office, Box 112250, University of Florida, Gainesville, Florida 326112250; Phone: (352) 3920433. Agreement: By completing and turning in the survey you will consent to participate in this study. This informed consent description is yours to keep. THANK YOU FOR YOUR T IME!

PAGE 210

210 Spanish Consentimiento Informado Ttulo: Comprendiendo Experiencias de Delitos y Victimiza cin entre los Delincuentes Por favor lea este documento de consentimiento cuidadosamente antes de que usted decida participar en este estudio. Propsito del Estudio: Entender experiencias de los encarcelados con pandillas, delitos y ser vctima. Qu s e le preguntar a usted en este estudio?: Si usted decide participar en este estudio, se le pedir que responda preguntas sobre su experiencia con delitos, victimizacin y pandillas. Las personas realizando este estudio no estn asociadas con la crcel y sus respuestas a las preguntas de la encuesta no sern compartidas con el personal de la crcel. POR FAVOR NO ESCRIBA SU NOMBRE EN LA ENCUESTA Y NO DE LA ENCUESTA A NADIE MAS QUE A LOS QUE ESTAN HACIENDO EL ESTUDIO. Tiempo Requerido: Entre 20 minutos y 1 hora y media, dependiendo de su paso. Confidencialidad: Todas sus respuestas sern annimas. Nadie podr conectar sus repuestas a usted, ya que nosotros no conoceremos su nombre. Sus respuestas sern codificadas con nmeros y estos cdigos no podrn se r conectados con usted. Los resultados del estudio presentarn patrones de como cada persona contest. No se enfocarn en las respuestas de una persona. Participacin Voluntaria y Derecho a retirarse del Estudio: No hay beneficios o recompensas por pa rticipar en este estudio. Este estudio no afectar de ninguna manera como es usted tratado en la crcel. Un riesgo potencial que usted pudiera experimentar al participar en este estudio es que algunas preguntas le hagan sentirse incmodo o que le moleste n. Para minimizar este riesgo, pueden contactar ha los servicios de consejera de la crcel, si estn disponibles. Tambin, usted no tiene que contestar ninguna pregunta que no desee responder, y usted puede dejar de participar en cualquier momento. Nad ie se molestar o enojar si usted decide no participar o si para su participacin en cualquier momento por cualquier razn. A Quien Contactar si usted Tiene Preguntas sobre el Estudio: Kate Fox o Dr. Jodi Lane, Departmento de Sociologa y Criminologa, 3219 Turlington Hall, PO Box 117330, Gainesville, Florida 326117330; Telfono: (352) 392 0265; Email: katefox@ufl.edu. A Quien Contactar Respecto A Sus Derechos Como Participante en este Estudio: UFIRB Office, Box 112250, University of Florida, Gaine sville, Florida 326112250; Telfono: (352) 3920433. Acuerdo: Al completar y devolver esta encuesta usted consiente en participar en este estudio. Esta descripcin de su consentimiento informado es suya y la puede conservar con usted. GRACIAS POR SU TI EMPO!

PAGE 211

211 APPENDIX F ORIGINAL AND MODIFIED SURVEY QUESTIONS AND SOURCES

PAGE 212

212 Table F 1. Crime Perpetration/Victimization Constructs Measured, Modified Survey Questions and Original Survey Questions with Sources Construct Modified survey item Original survey it em and source Theft Has someone ever stolen money or property from you without using force? Was something stolen or an attempt made to steal something that belonged to you or another household member? (National Crime Victimization Survey) Robbery Ha s someone ever used a weapon or force to steal money or property from you? Have you h ad someone use a weapon or force to get money or property from you ? (Taylor et al., 2008) Vandalism Has someone ever damaged or vandalized your property? Has anyo ne intentionally damaged or destroyed property owned by you or someone else in your household? (National Crime Victimization Survey) Threatened with weapon Have you ever been threatened with a weapon? Has anyone attacked or threatened you in any of the se ways with any weapon, for instance a gun or a knife? (National Crime Victimization Survey) Physical assault Have you ever been attacked without a weapon? Have you been hit by someone trying to hurt you ? (Taylor et al., 2008) Assault with weapon Have you ever been attacked with a weapon? Have you been attacked by someone with a weapon or by someone trying to seriously hurt or kill you? (Taylor et al., 2008) Sexual assault Have you ever been sexually assaulted or raped? Has anyone attacked or threatened you in any of these waysany rape, attempted rape or other type of sexual attack? (National Crime Victimization Survey) Stabbed Have you ever been stabbed? Have you ever been stabbed? (Aguilar & Nightingale, 1996)

PAGE 213

213 Table F 1. (contin ued) Construct Modified survey item Original survey item and source Witness intimidation Have you ever been threatened by someone so you did not act as a witness in court? Have you or (your family) anyone else in your household experienced any haras sment or intimidation from the offender(s), or their family or friends since this incident occurred? (British Crime Survey, 1998) Home invasion Have you ever been the victim of a home invasion? How afraid are you of having a gang member commit a home i nvasion robbery against you? (Lane et al., SOCP) Drive by shooting Have you ever been the victim of a drive by shooting (shot or shot at )? How afraid are you of being a victim of a drive by or random gang related shooting? (Lane et al., SOCP) Shot a t Have you ever been shot at but not hit (not military related)? Have you ever been shot? (Aguilar & Nightingale, 1996) Shot Have you ever been shot (not military related)? Have you ever been shot? (Aguilar & Nightingale, 1996)

PAGE 214

214 Table F 2. Self Control Constructs Measured, Modified Survey Questions and Grasmick et al.s Original Survey Items C onstruct (with survey question number) Modified survey item Original Grasmick et al. (1993) survey item Impulsivity36 I often act on the spur of the moment w ithout thinking. I often act on the spur of the moment without stopping to think. Physical Activity 37 I like to get out and do things more than I like to sit around. I like to get out and do things more than I like to read or contemplate ideas. Temper 3 8 Often, when Im angry at people I feel more like hurting them than telling them why I am angry. Often, when Im angry at people I feel more like hurting them than talking to them about why I am angry. Risk Seeking 39 Sometimes I will take a risk just f or fun. Sometimes I will take a risk just for the fun of it. Physical Activity 40 If I had a choice, I would almost always rather do something physical than something mental. Same as modified. Self Centered 41 If things I do upset people, its their prob lem not mine. Same as modified. Simple Task 42 The things in life that are easiest to do are the most fun. The things in life that are easiest to do bring me the most pleasure. Self Centered 43 I often look out for myself first, even if it makes it hard for other people. I try to look out for myself first, even if it means making things difficult for other people. Impulsivity 44 I dont think much about the future. I dont devote much thought and effort to preparing for the future. Risk Seeking 45 I li ke to do things that might get me in trouble. I sometimes find it exciting to do things for which I might get in trouble.

PAGE 215

215 Table F 2. (c ontinued) Construct (with survey question number) Modified survey item Original Grasmick et al. (1993) survey item Te mper 46 I get mad easily. I lose my temper pretty easily. Self Centered 47 I dont care so much when other people are having problems. Im not very sympathetic to other people when they are having problems. Risk Seeking 48 I like to test myself by taking risks every once in a while. I like to test myself every now and then by doing something a little risky. Simple Task 49 I often try to avoid things that will be hard. I frequently try to avoid projects that I know will be difficult. Temper50 When Im r eally angry, other people better stay away from me. Same as modified. Impulsivity51 I often do whatever is fun now, even at the cost of some distant goal. I often do whatever brings me pleasure here and now, even at the cost of some distant goal. Physi cal Activity52 I almost always feel better when I am on the move than when I am sitting and thinking. Same as modified. Temper53 When I have trouble with someone, its hard for me to talk calmly about it without getting upset. When I have a serious disag reement with someone, its usually hard for me to talk calmly about it without getting upset. Impulsivity54 Im more concerned with what happens to me in the short run than in the long run. Same as modified. Self Centered55 I will try to get the things I want even when I know it makes other people upset. I will try to get the things I want even when I know its causing problems for other people.

PAGE 216

216 Table F 2. (c ontinued) Construct (with survey question number) Modified survey item Original Grasmick et a l. (1993) survey item Risk Seeking56 Excitement and adventure are more important to me than security. Same as modified. Simple Task57 I dont like really hard jobs that push me. I dislike really hard tasks that stretch my abilities to the limit. Simp le Task58 When things get hard, I tend to quit. When things get complicated, I tend to quit or withdraw.

PAGE 217

217 LIST OF REFERENCES Akers, R. L. (1991). Self control as a general theory of crime. Journal of Quantitative Criminology, 7(2), 201211. Arneklev, B J., Grasmick, H. G., & Bursik, R. J. (1999). Evaluating the dimensionality and invariance of low self control. Journal of Quantitative Criminology 15, 307331. Beck, A. R. (1999). Deciding on a new jail design. Justice Concepts, Inc. Bernard, H. R. (2000). Social research methods. Newbury Park, CA: Sage Publications. Bjerregaard, B., & Smith, C. A. (1993). Gender differences in gang participation, delinquency, and substance use. Journal of Quantitative Criminology, 9, 329355. Blumstein, A., Cohen, J., Roth, J., & Visher, C.A. (1986). Criminal careers and career criminals Washington, D.C.: National Academy Press. Bowker, L. H., & Klein, M. W. (1983). The etiology of female juvenile delinquency and gang membership: A test of psychological and soc ial structural explanations. Adolescence, 18, 739751. California Penal Code 186.22. California Street Terrorism Enforcement and Prevention Act. Cauffman, E., Steinberg, L., & Piquero, A. R. (2005). Psychological, neuropsychological and physiological c orrelates of serious antisocial behavior in adolescence: The role of self control. Criminology, 43(1), 133175. Chin, K. L., Fagan, J., & Kelly, R. J. (1992). Patterns of Chinese gang extortion. Justice Quarterly, 9, 625646. Clinton, W. J. (1997). State of the Union Address. Cohen, A. K. ( 1969 ). Forward and Overview. In C. R. Huffs (Ed.) Gangs in America pp 721. Newbury Park, CA: Sage Publications. Coulton, C. J., Korbin, J., Chan, T., & Su, M. (2001). Mapping residents perceptions of neighborhood boundaries: A methodological note. American Journal of Community Psychology, 29(2), 371383. Craven, D. (1997). Sex differences in violent victimization, 1994. Bureau of Justice Statistics Special Report. U.S. Department of Justice. NCJ 164508. Cureton, S. R. (2002). Introducing Hoover: Ill ride for you, gangsta. In C. R. Huffs (Ed.) Gangs in America III. Thousand Oaks, CA: Sage Publications.

PAGE 218

218 Curry, G. D., & Spergel, I. A. (1992). Gang involvement and delinquency among Hispanic and African American ad olescent males. Journal of Research in Crime and Delinquency, 29, 273291. Decker, S. H., & Van Winkle, B. (1996). Life in the Gang: Family, Friends and Violence. New York: Cambridge University Press. Duhart, D. T. (2000). Urban, suburban, and rural vi ctimization, 19931998. Bureau of Justice Statistics Special Report. U.S. Department of Justice. NCJ 182031. Durose, M. R., & Langan, P. A. (2007). Felony sentences in state courts, 2004. Bureau of Justice Statistics Bulletin. U.S. Department of Justice. NCJ 215646. Durose, M. R., Smith, E. L., & Langan, P. A. (2007). Contacts between police and the public, 2005. Bureau of Justice Statistics Special Report. U.S. Department of Justice. NCJ 215243. E sbensen, F. A., & Deschenes, E. P. (1998). A multisite ex amination of youth gang membership: Does gender matter? Criminology 36(4), 799827. Esbensen F. A., & Huizinga, D. ( 1993). Gangs, drugs, and delinquency in a survey of urban youth. Criminology, 31(4), 565589. Esbensen, F. A., & Winfree, L. T. (1998). Race and gender differences between gang and non gang youths: Results from a multi site survey. Justice Quarterly, 15(3), 505526. Esbensen, F. A., Osgood, D. W., Taylor, T. J., Peterson, D. & Freng, A. (2001). How great is G.R.E.A.T.? Results from a longitudinal quasi experimental design. Criminology and Public Policy, 1(1), 87118. Fagan, J. (1990). Social processes of delinquency and drug use among urban gangs. In C. R. Huff (Ed.), Gangs in America. Newbury Park, CA: Sage Publications. Farrell, G., Ts eloni, A., Wiersema, B., & Pease, K. (2001). Victim careers and career victims?: Toward a research agenda. Crime Prevention Studies, 12, 241254. Fearn, N. E., Decker, S. H., & Curry, G. D. (2006). Public policy responses to gangs: Evaluating the outco mes. In A. Egley Jr., C. L. Maxson, J. Miller, & M. W. Kleins (Eds.) The Modern Gang Reader, 3rd edition. Los Angeles, CA: Roxbury Publishing Company. Ferraro, K.F. (1996) Womens fear of victimization: Shadow of sexual assault? Social Forces 75(2), 667690. Fleisher, M., & Decker, S. H. (unpublished manuscript). Gangs behind bars: Prevalence, conduct, and response.

PAGE 219

219 Gardner, W., Mulvey, E. P., & Shaw, E. C. (1995). Regression analysis of counts and rates: Poisson, overdispersed Poisson, and negative bi nomial models. Psychological Bulletin, 118(3), 392404. Gibbs, J. J. & Giever, D. M. (1995). Self control and its manifestations among university students: An empirical test of Gottfredson and Hirschi's general theory. Justice Quarterly, 12(2), 231255. Gibson, C., Miller, J. M., Swatt, M., Jennings, W. G., & Gover, A. R. ( forthcoming, 2009). Using propensity score matching to understand the relationship between gang membership and violent victimization: A research note. Justice Quarterly. Gibson, C., Sc hreck, C. J., & Miller, M. (2004). Binge drinking and negative alcohol related behaviors: A test of self control theory. Journal of Criminal Justice, 32(5), 411420. Gordon, R. A., Lahey, B. B., Kawai, E., Loeber, R., Stouthamer Loeber, M, & Farrington, D P. (2004). Antisocial behavior and youth gang membership: Selection and socialization. Criminology, 42(1), 55 87. Gottfredson, M.R. & Hirschi, T. (1990). A general theory of crime Stanford, CA: Stanford University Press. Gover, A. R., Jennings, W. G., & Tewksbury, R. (forthcoming, 2009). Adolescent male and female gang members experiences with violent victimization, dating violence, and sexual assault. American Journal of Criminal Justice. Gover, A. R., Kaukinen, C., & Fox, K. A. (2008). The relatio nship between violence in the family of origin and dating violence among college students. Journal of Interpersonal Violence, 23(12), 16671693. Grasmick, H.G., Tittle, C.R., Bursick, Jr., R.J., & Arneklev, B.J. (1993). Testing the core empirical implicat ions of Gottfredson and Hirschis general theory of crime. Journal of Research in Crime and Delinquency, 30, 529. Hagedorn, J. ( 1988). Gangs in America. Newbury Park, CA: Sage Publications. Harrison, L. D. (1995). The validity of self reported data on drug use. Journal of Drug Issues, 25(1), 91111. Henry, B., Moffitt, T. E., Caspi, A., Langley, J., & Silva, P. A. (1994). On the remembrance of things past: A longitudinal evaluation of the retrospective method. Psychological Assessment, 6 92101. Hi lbe, J. M. (2007). Negative binomial regression. New York, NY: Cambridge University Press. Hirschi, T. (1969). Causes of delinquency. New Brunswick, NJ: Transaction Publishers.

PAGE 220

220 Hirschi, T. (2004). Self control and crime. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self regulation: Research, theory, and applications. New York: Guilford Press. Huff, C. R. (1998). Criminal behavior of gang members and at risk youths. National Institute of Justice Research Preview. James, D. J. (2004). Profile of j ail inmates, 2002. Bureau of Justice Statistics Special Report. U.S. Department of Justice. NCJ 201932. Joe, K. A., & Chesney Lind, M. (1995). Just every mothers angel: An analysis of gender and ethnic variations in youth gang membership. Gender & Societ y, 9(4), 408431. Karmen, A. (2009). Crime victims: An introduction to victimology. Thomson Wadsworth Publishing. Kelling, G. L., & Wilson, J. Q. (1982, March). Broken windows. The Atlantic. Klaus, P., & Rennison, M. (2002). Age patterns in violent vi ctimization, 19762000. Bureau of Justice Statistics Crime Data Brief. U.S. Department of Justice. NCJ 190104. Klein, M. W. ( 1971 ). Street Gangs and Street Workers. Englewood Cliffs, NJ: Prentice Hall. Koss, M. P. (2001). Evolutionary models of why men r ape: Acknowledging the complexities. Trauma, Violence, & Abuse, 1, 182190. Kyckelhahn, T., & Cohen, T. H. (2008). Felony defendants in large urban counties, 2004. Bureau of Justice Statistics Bulletin. U.S. Department of Justice. NCJ 221152. Lane, J. ( 2002). Fear of gang crime: A qualitative examination of the four perspectives. Journal of Research in Crime and Delinquency, 39(4), 437471. Lane, J., & Meeker, J. W. (2000). Subcultural diversity and the fear of crime and gangs. Crime and Delinquency, 46(4), 497521. Lane, J., & Meeker, J. W. (2003a). Fear of gang crime: A look at three theoretical models. Law & Society Review, 37, 425456. Lasley, J. R. (1997). Age, social context, and street gang membership: Are youth gangs becoming adult gangs? I n G. L. Mays (Ed.) Gangs and gang behavior pp 99124. Chicago, IL: Nelson Hall Publishers. Lauritsen, J. L., & Laub, J. H. (2007). Understanding the link between victimization and offending: New reflections on an old idea. Crime Prevention Studies, 22, 5575.

PAGE 221

221 Lauritsen, J. L., Laub, J. H., & Sampson, R. J. (1992). Conventional and delinquent activities: Implications for the prevention of violent victimization among adolescents. Violence and Victims, 7, 91108. Lauritsen, J. L., Sampson, R. J., & Laub, J. H. (1991). The link between offending and victimization among adolescents. Criminology, 29(2), 265292. Levine, J. P. (1976). The potential for crime overreporting in criminal victimization surveys. Criminology, 14(3), 307330. Long, J. S. (1997). Regr ession models for categorical and limited dependent variables. Thousand Oaks, CA: Sage Publications. Longshore, D. (1998). Self control and criminal opportunity: A prospective test of the general theory of crime. Social Problems, 45(1), 102113. Longshor e, D., Stein, J. A., & Turner, S. (1998). Reliability and validity of a self control measure: Rejoinder. Criminology, 36(1), 175182. Longshore, D., Turner, S., & Stein, J. A. (1996). Self control in a criminal sample: An examination of construct validity Criminology, 34(2), 209228. Marcus, B. (2004). Self control in the general theory of crime. Theoretical Criminology, 8(1), 3355. Martin, D. G. (2003). Enacting neighborhood. Urban Geography, 24(5), 361385. Maxfield, M. G. & Babbie, E. (2008). Resea rch methods for criminal justice and criminology, fifth edition. Thomson Wadsworth Publishing. Maxson, C. L., & Klein, M. W. ( 1990). Street gang violence: Twice as great or half as great? In C. R. Huffs (Ed.) Gangs in America pp 77100. Newbury Park CA : Sage Publications McCullagh, P., & Nelder, J. A. (1989). Generalized linear m odels second edition. London: Chapman and Hall. Miller, J. (1998). Gender and victimization risk among young women in gangs. Journal of Research in Crime and Delinquency, 35( 4), 429453. Miller, J ( 2002) The girls in the gang: W hat weve learned from two decades of research. Pp. 175197 in C. R Huff (E d.). Gangs in America III. Thousand Oaks CA : Sage Publications Miller, W. B. ( 1966). Violent crimes by city gangs. Annals of the American Academy of Political and Social Science, 364, 96112.

PAGE 222

222 Moffitt. T. E. (1993). Life course persistent and adolescence limited antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674701. Molidor, C. E. (1996). Femal e gang members: A profile of aggression and victimization. Social Work, 41(3), 251257. Moore, J. W. (1978). Homeboys. Philadelphia, PA: Temple University Press. Moore, J. W. (1991). Going down to the barrio: Homeboys and homegirls in change. Philadelp hia, PA: Temple University Press. Myers, M. A., & LaFree, G. D. (1982). Sexual assault and its prosecution: A comparison with other crimes. Journal of Criminal Law & Criminology, 73(3), 12821305. Nagin, D. S., & Smith, D. A. (1990). Participation in and frequency of delinquent behavior: A test for structural differences. Journal of Quantitative Criminology, 6(4), 335356. National Youth Gang Center (NYGC). (2007a). Analysis of gang related legislation. Institute for Intergovernmental Research. Nationa l Youth Gang Center (NYGC). (2007b). National Youth Gang Survey Analysis Nobles, M. R., Fox, K. A., Piquero, N. L., & Piquero, A. R. (forthcoming, 2009). Career dimensions of stalking victimization and perpetration. Justice Quarterly. Office of the Attor ney General of Florida, Bill McCollum. (2008, January 15). Statewide grand jury makes significant anti gang recommendations in first interim report. Osgood, D. W. (2000). Poisson based regression analysis of aggregate crime rates. Journal of Quantitativ e Criminology, 16(1), 2143. Padilla, F. M. (1992). The gang as an American enterprise. New Brunswick, NJ: Rutgers University Press. Parker, R. N., & Smith, M. D. (1984). High correlations or multicollinearity, and what to do about either: Reply to Light. Social Forces, 62(3), 804807. Peterson, D., Taylor, T. J., & Esbensen, F. (2004). Gang membership and violent victimization. Justice Quarterly, 21 (4), 793815. Peterson, M. A., Braiker, H. B., & Polich, S. M. (1981). Who commits crimes a survey of pris on inmates. Cambridge, MA: Oelgeschlager, Gunn, and Hain Publishers, Inc. Peterson Lynskey, D., Winfree, L. T. Jr., Esbensen, F. A., & Clason, D. L. (2000). Linking gender, minority group status, and family matters to self control theory: An analysis of

PAGE 223

223 key self control concepts in a youth gang context. Juvenile and Family Court Journal, 51(3), 119. Piquero, A. R., & Bouffard, J. A. (2007). Something old, something new: A preliminary investigation of Hirschis refined self control. Justice Quarterly, 24, 127. Piquero, A. R., & Rosay, A. B. (1998). The reliability and validity of Grasmick et al.s self control scale: A comment on Longshore et al. Criminology, 36(1), 157173. Piquero, A. R., Farrington, D. P., & Blumstein, A. (2003). The criminal care er paradigm. In M. Tonry (Ed.), Crime and justice: A review of research, 30, 359506. Chicago, IL: University of Chicago Press. Pratt, T. C., & Cullen, F. T. (2000). The empirical status of Gottfredson and Hirschis general theory of crime: A meta analysis Criminology, 38(3), 931964. Pratt, T. C., & Cullen, F. T. (2005). Assessing macrolevel predictors and theories of crime: A meta analysis. Crime and Justice, 32, 373450. Rand, M. R. (2008). Criminal victimization, 2007. Bureau of Justice Statistics Bul letin. U.S. Department of Justice. NCJ 224930. Rennison, C. (2001). Violent victimization and race, 19931998. Bureau of Justice Statistics Special Report. U.S. Department of Justice. NCJ 176354. Ruddell, R., Decker, S. H., & Egley, A. (2006). Gang interve ntions in jails: A national analysis. Criminal Justice Review, 31 (1), 1 14. Sabol, W. J., Couture, H., & Harrison, P. M. (2007). Prisoners in 2006. Bureau of Justice Statistics. U.S. Department of Justice. NCJ 219416. Sampson, R. J., & Groves, W. B. (1989) Community structure and crime: Testing social disorganization theory. American Journal of Sociology, 94(4), 774802. Sampson, R. J., & Laub, J. H. (1993). Crime in the Making. Cambridge, MA: Harvard University Press. Sampson, R. J., & Raudenbush, S. W. ( 2001). Disorder in urban neighborhoods Does it lead to crime? National Institute of Justice Research in Brief. Sampson, R. J., Raudenbush, S. W., & Earls, F. ( 1997 ). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918924. Schreck, C. J. (1999). Criminal victimization and low self control: An extension and test of a general theory of crime. Justice Quarterly, 16, 633654.

PAGE 224

224 Schreck, C. J., Stewart, E. A., & Osgood, D. W. (2008). A reappraisal of the overlap of vio lent offenders and victims. Criminology, 46(4), 871905. Sellers, C. S. (1999). Self control and intimate violence: An examination of the scope and specification of the general theory of crime. Criminology, 37(2), 375404. Shaw, C. R., & McKay, H. D. (1969). Juvenile Delinquency and Urban Areas Chicago: The University of Chicago Press. Sherman, L. W., Gottfredson, D. C., MacKenzie, D. L., Eck, J., Reuter, P., & Bushway S. D. ( 1998). Preventing crime: What works, what doesnt, whats promising. A report t o the United States Congress. Prepared for the National Institute of Justice by the Department of Criminology and Criminal Justice University of Maryland. Short, J. F. ( 1990 ). New wine in old bottles? Change and continuity in American gangs. In C. R. Hu ff (Ed.), Gangs in America. Newbury Park, CA: Sage Publications. Singer, S. I. (1979). A comment on alleged overreporting. Criminology, 16(1), 99103. Skogan, W. G., & Maxfield, M. G. (1981). Coping with crime: Individual and neighborhood reactions. Sage Publications. Spergel, I. A. ( 1964). Slumtown, Racketville, Haulberg. Chicago, IL: University of Chicago Press. Spergel, I. A., & Grossman, S. F. (1997). The Little Village Project: A community approach to the gang problem. Social Work, 42(5), 456470. Stewart, E. A., Elifson, K. W., & Sterk, C. E. (2004). Integrating the General Theory of Crime into an exploration of violent victimization among female offenders. Justice Quarterly, 21, 159181. Struckman Johnson, C., Struckman Johnson, D., Rucker, L., Bumby, K., & Donaldson, S. ( 1996). Sexual coercion reported by men and women in prison. The Journal of Sex Research, 33(1), 6776. Studenmund, A. H. (2001). Using econometrics: A practical guide, fourth edition. Addison Wesley Lognman, Inc. Taylor, T. J (2008). The boulevard aint safe for your kids: Youth gang membership and violent victimization. Journal of Contemporary Criminal Justice, 24(2), 125136. Taylor, T. J., Peterson, D., Esbensen, F., & Freng, A. (2007). Gang membership as a risk factor for adolescent violent victimization. Journal of Research in Crime and Delinquency, 44(4), 351380.

PAGE 225

225 Thornberry, T. P., Krohn, M. D., Lizotte, A. J., & ChardWierschem, D. (1993). The role of juvenile gangs in facilitating delinquent behavior. Journal of Research in Crime and Delinquency, 30(1), 55 87. Thornberry, T. P., Krohn, M. D., Lizotte, A. J., Smith, C. A., & Tobin, K. (2003a). Gangs and delinquency in developmental perspective New York, NY: Cambridge University Press. Thornberry, T. P., Lizotte, A. J., Krohn, M. D., Smith, C. A., & Porter, P. K. (2003b). Causes and consequences of delinquency: Findings from the Rochester youth development study. In T. P. Thornberry and M. D. Krohn (Eds.), Taking stock of delinquency: An overview of findings from contemporary longitudinal studies. New York, NY: Kluwer Academic/Plenum Publishers. Thrasher, F. (1927). The gang. Chicago, IL: University of Chicago Press. Tittle, C. R., Ward, D. A., & Grasmick, H. G. (2003). Self control and crime/deviance: Cognitive vs. behavioral measures. Journal of Quantitative Criminology, 19(4), 333365. Tsunokai, G. T., & Kposowa, A. J. (2002). Asian gangs in the United States: The current state of the research literature. Crime, Law and Social Change, 37(1), 3750. Turner, C. F., Lessler, J. T., & Devore J. W. ( 1992). Effects of mode of administration and wording on reporting of drug use. In C.F. Turner, J.T. Lessler and J.C. Gfroerer s (Eds.) Survey Measurement of Drug Use: Methodological Studies Rockville, MD : U.S. Departm ent of Health and Human Services, National Institute on Drug Abuse. Turner, M. G., & Piquero, A. R. (2002). The stability of self control. Journal of Criminal Justice, 30(6), 457471. U. S. Census Bureau. (2008). State and county quick facts Florida. U S. Department of Justice (2007). Uniform Crime Report Crime in the United States. U S Senator Dianne Feinstein. (2007, June 14). Senate judiciary committee overwhelmingly approves Feinstein Hatch comprehensive gang legislation. Vigil, J. D. ( 1988). Bario gangs: Street life and identity in Southern California. Austin, TX: University of Texas Press. Vigil, J. D. (2002). A rainbow of gangs: Street cultures in the megacity. Austin, TX: University of Texas Press. Violence Policy Center. (2007). Drive b y America. Washington DC. Violent Crime Control and Law Enforcement Act. ( 1994). Public Law 103 322,

PAGE 226

226 Vold, G. B., Bernard, T. J., & Snipes, J. B. (2001). Theoretical criminology, fifth edition. New York, NY: Oxford University Press. Walker Bar nes, C. J., & Mason, C. A. (2001). Ethnic differences in the effect of parenting on gang involvement and gang delinquency: A longitudinal, hierarchical linear modeling perspective. Child Development, 72(6), 18141831. Warner, R. M. (2008). Applied statist ics: From bivariate through multivariate techniques. Thousand Oaks, CA: Sage Publications. Whyte, W. F. (1943). Street corner society. Chicago, IL: University of Chicago Press.

PAGE 227

227 BIOGRAPHICAL SKETCH Kathleen A Fox earned her Bachelor of Science degree i n S ociology from the University of Utah (2004) and her Master of Arts (2006) and Doctor of Philosophy (2009) in Criminology, Law and Society from the University of Florida. This fall 2009 she will begin her academic career as an Assistant Professor at Sam Houston State University in Texas.