Youth gangs and violence have received substantial scholarly and public attention during the past two decades. While most of the extant research on youth gang members has focused on their offending behaviors, few quantitative studies have been conducted to examine the link between gang membership and violent victimization. The current study uses data from a multi-site study of youth to explore potential factors related to this increased risk. These findings suggest that gang members are more likely to experience violent victimization, as well as greater frequency of victimization, than do non-gang members. Furthermore, gang membership remains a significant correlate of the annual prevalence of victimization net other individual, family, peer, school, and situational factors. The relationship, however, is complex and dependent upon the type of victimization examined.

Active parental consent in survey research poses ethical and practical concerns. One common argument against the requirement of active consent procedures is its effect on participation rates. There is additional concern that higher risk groups may be underrepresented in the final sample. Empirical support of differential attrition, however, is lacking. In the current multisite longitudinal study, passive consent procedures were approved for the collection of pretest data. For subsequent years of data collection, active parental consent procedures were required. In this article, we use the pretest data to examine demographic, attitudinal, and behavioral differences between those students for whom active consent was provided and those for whom active consent was either denied or for whom no response was received. The results indicate that active consent procedures produce deleterious effects on participation rates and lead to an underrepresentation of at-risk youth in the sample.

Research Summary: This manuscript presents results from the National Evaluation of the Gang Resistance Education and Training (G.R.E.A.T.) program, a school‐based prevention program targeting middle‐school students. A longitudinal quasi‐experimental research design was conducted from 1995 through 1999. Beneficial program effects emerged gradually over time so that there was, on average, more pro‐social change in the attitudes of G.R.E.A.T. students than the non‐G.R.E.A.T. students four years following program exposure. Policy Implications: Two specific policy recommendations stem from this research. First, law enforcement officers can be effective providers of school‐based prevention programs. Second, to better assess program effectiveness, evaluations should include design features that allow for assessment of long‐term or delayed program effects.

P rior research has established that there is a cumulative effect of risk factors on both youth violence and gang membership and that risk factors in multiple domains increase the probability of youth violence and gang involvement. In this article we expand upon this risk factor approach to the study of youth violence by addressing two questions concerning youth violence: (1) What are the effects of cumulative risk, including risk in multiple domains, on youth violence and gang membership and to what extent are the patterns similar or different for youth violence and gang membership?(2) To what extent do risk factors exert independent effects when other factors are controlled in multivariate analyses, and are the risk factors for youth violence similar to or different from those for gang membership? We utilise survey data from a sample of 5,395 8th grade students in 11 cities across the United States to examine these issues.