Poster #229 - Risk Variation Among Current, Former, and Non-Members of Youth Gangs: More Validity Evidence For Self-Nominations

• In Event: 3-159 - Poster Session 12
  In Poster Session: PS 12 Section - Social Relationships

Integrative Statement

Youth gang membership is associated with elevated problem behavior and is a serious developmental risk factor (Howell, 2016; Thornberry et al., 2003). Despite a well-elaborated literature on the correlates and consequences of youth gang involvement, studies examining precisely what it means to be “in” a gang are few and far between. At present, the most reliable and valid indicator of gang involvement is considered to be simple and straightforward self-nomination. However, this method of assessment is subject to all the usual concerns regarding single-item measurement of complex constructs as well as the method bias of adolescents reporting on their own behavior. Recently, some gang researchers have made some key advances in delineating the utility of this approach to gauging gang involvement. For example, Decker and colleagues observed significant and robust differences on a variety of demographic and theoretical gang-related constructs between self-nominated gang members, former members, and nonmembers. Melde and Esbensen (2013) showed that the odds of violent behavior were related in predictable ways over time to whether youth reported current gang involvement, and whether they subsequently left their gangs.
The present study extends this line of inquiry to consider whether current, former, or non-involvement in gangs might be linked differentially to different types of personal and contextual indicators of risk, including static/historical factors (e.g., temperament) and dynamic factors (e.g., parental monitoring). Participants in this study were 421 youth (ages 11-17; 69% male; 66% nonwhite) who were referred for intensive home and community intervention services by justice authorities. Youth and their caregivers completed a battery of survey measures during the treatment intake process (time 1), and were followed through the close of treatment (time 2) and then 12 months post-discharge (time 3). In addition to responding to a variety of personal and contextual risk and problem behavior measures at time 1, youth indicated whether they were currently in a gang, and if not whether they were ever in a gang. Per self-nominations, the sample included 21 current gang members (5%), 35 former gang members (8.3%) and 365 (86.7%) nonmembers. Treatment success was coded at time 2 from case files, and recidivism was recorded at time 3.
Analyses showed that current gang members exhibited reliably higher levels of problem behavior (including violence) than did former and nonmembers (p < .001), and both current and former gang members were generally higher on the dynamic risk factors of parental monitoring and violent victimization than were nonmembers. There were no differences by gang status on the static risk variables of neighborhood disorder and trait impulsivity. Current gang members were significantly less likely (p < .001) to complete treatment successfully (38% success rate) compared to former (71%) and nonmembers (79%). There were no significant differences in re-arrest rates by gang status (p = .434), though practical differences were observed (current = 43% re-arrested; former = 29% re-arrested; nonmembers = 30% re-arrested). Findings have implications for the broader and better assessment of the developmental dynamics of gang involvement, as well as interventions for gang-involved youth.

Author

• Paul Boxer, Rutgers University
  Presenting Author