Measuring Gang Involvement in a Justice-Referred Sample of Youth in Treatment

• Author: Paul Boxer,Joanna Kubik,Bonita Veysey,Michael Ostermann
• Date: 01 January 2015
• Published date: 01 January 2015
• DOI: 10.1177/1541204013519828
• Subject Matter: Articles

Article

Measuring Gang Involvement

in a Justice-Referred Sample

of Youth in Treatment

Paul Boxer

1

, Bonita Veysey

2

, Michael Ostermann

2

, and Joanna Kubik

1

Abstract

Gangs are present in about 34% of all jurisdictions in the United States. Given elevations in violence

and victimization associated with gang involvement, effective means are needed for measuring

involvement among individual youth. This is especially the case among youth receiving services for

problem behavior who might benefit from targeted treatments helping them reduce involvement in

gangs. We assessed gang involvement among 421 youth referred by the justice system for intensive

home- and community-based mental health treatment. Using self-report survey and therapist-

recorded data, we identified 94 (22%) youth as gang involved. Risk factor measures provided support

for our classification, driven primarily by self-reported indicators of gang involvement as opposed to

therapist-recorded indicators.

Keywords

gang, mental health, treatment, delinquency

Street gangs havebeen present in American society for decades, and the most recent gangsurveillance

available (2010 National Youth Gang Survey) estimates that there are approximately 756,000 gang

members representing 29,400 gangswith gang presence in about 34%of all cities,suburbs, towns, and

rural counties in theUnited States (Egley & Howell, 2012). Data compiled by the National Center for

Education Statistics suggest thatabout 20%of U.S. secondary school students report gang presencein

their schools (Robers, Zhang, Truman, & Snyder, 2012). Yet the picture of what constitutes a youth

gang member,or a gang-involved youth, is fairly fuzzy. Although gang-involved youth tend toengage

in more violent and nonviolent delinquencyand have elevated profiles of personaland contextual risk

compared to youth who are not involved in gangs (see Howell & Egley, 2005 or O’Brien, Daffern,

Chu, & Thomas, 2013, for review), most are not ‘‘superpredators’’ and actually spend much of their

time on the same mundane, everyday activities as do uninvolved youth (Howell, 2012). Indeed, gang

membership or involvement has been shown to be somewhat unstable over time and composed of

1

Department of Psychology, Rutgers University, Newark, NJ, USA

2

School of Criminal Justice, Rutgers University, Newark, NJ, USA

Corresponding Author:

Paul Boxer, Department of Psychology, Rutgers University, Newark, NJ 07102, USA.

Email: pboxer@rutgers.edu

Youth Violence and JuvenileJustice

2015, Vol. 13(1) 41-59

ªThe Author(s) 2014

Reprints and permission:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/1541204013519828

yvj.sagepub.com

multiplegradations of involvement fromyouth who are central to gang leadershipto youth who merely

associate with gangmembers on occasion (Curry, Decker,& Egley, 2002; Esbensen, Winfree, He, &

Taylor, 2001; P yrooz, 2013).

Although broad-based efforts to reduce and prevent youth gang involvement have been in place

for quite some time, in recent years there has been renewed interest in topic. This interest has ranged

from the placement of preventing youth gang involvement onto the national shared research and

practice agenda of the U.S. Centers for Disease Control and the U.S. National Institute of Justice

(Simon, Ritter, & Mahendra, 2013) to new empirical studies examining the role of gang involvement

in treatment outcomes for delinquent youth (Boxer, 2011; Schram & Gaines, 2005). Considered in

tandem with elevated activity in the policy and practice sectors around keeping delinquent youth out

of detention and in their communities for intervention (Annie E. Casey Foundation, 2013), and

efforts to promote the adoption of community-based best-practice approaches for helping delinquent

youth (Henggeler & Schoenwald, 2011; Howell, 2012), there seems to be a clear need to facilitate

the assessment of gang involvement in the context of targeted intervention practices.

From the standpoint of drawing theory-guided inferences about characteristics of gang involve-

ment, self-identification of gang membership has been shown to be robust via large-sample survey

methods (Esbensen et al., 2001). Whether this method holds utility on a smaller scale—for example,

in the context of targeted intervention—is an important question to address particularly given

the enduring impacts into adulthood of gang involvement during adolescence (Krohn, Ward,

Thornberry, Lizotte, & Chu, 2011). In the context of treatment, self-reports of gang involvement

might be less reliable, given that some youths hold motivations to conceal their involvement (thus

protecting their interests to stay connected to their gang), whereas others might overstate their invol-

vement (thus projecting an image of ‘‘toughness’’ and resistance to respond to treatment efforts). In

this study, we utilized a combination of self-report and therapist-observer ratings of youth gang

involvement to explore the utility of gang classification within an outpatient sample of justice-

referred youth receiving intervention services. Given the array of risk factors shown to link to gang

involvement, and in particular the self-identification of gang membership (e.g., Howell & Egley,

2005), we sampled from a number of ecological risk domains to examine hypothesized differences

between youth classified as gang involved in comparison to youth classified as uninvolved. We

locate our results in the theoretical and applied literature on youth gangs.

Measuring Youth Gang Involvement

Law enforcement and juvenile justice agencies typically rely on a number of criteria for ascertaining

whether a youth is gang involved; for example, whether a youth has been arrested in the company of

known gang members, has been identified as a gang member by a key informant, is known to reside

in or frequent an area known to be within a particular gang’s territory, and/or has admitted to gang

involvement (National Gang Center, 2012; Schram & Gaines, 2005). Social scientists, however,

rarely have a degree of access to information about their subjects that would permit such detailed

assessment and instead must rely wholly or mostly on the latter criterion—self-identification.

Self-identification of gang membership (i.e., ‘‘Are you currently in a gang?’’) is believed to be a

robust indicator of gang membership (Esbensen et al., 2001) and has been shown to link reliably

although not perfectly to police records of gang involvement (Curry, 2000). Still, on occasion, this

self-identification question has been utilized in combination with other methods including true/false

statements or interval ratings about levels of gang involvement, teacher or counselor reports, and

peer nominations (Dishion, Nelson, & Yasui, 2005; Hermann, McWhirter, & Spisas-Hermann,

1997; Li et al., 2002) or variations on the current membership question including questions about

whether respondents were ever in a gang (e.g., Pyrooz & Decker, 2011), have friends in a gang

42 Youth Violence and Juvenile Justice 13(1)