The research aimed to determine the unique contribution of gang membership to juvenile delinquency, above and beyond having delinquent peers. The longitudinal sample of 808 students included 412 males and 396 females who attended 18 elementary schools serving high-crime neighborhoods of Seattle in the fall of 1985. Data were collected from the youths, their parents or adult caregivers, teachers, school records, and King County court records. Data were analyzed through multivariate analysis of variance and follow-up univariate analyses to determine differences on measures of delinquency at ages 14 and 15. Results revealed that gang members had a higher rate of offending in the past year when compared with the other groups. Structural equation modeling revealed that gang membership independently predicted both self-reported delinquency and officially recorded delinquency beyond the effects of having delinquent friends and prior delinquency. Findings provide support for the enhancement model and clearly suggest that preventing youth from joining gangs holds promise for preventing and reducing crime and drug use. However, virtually no effective strategies for preventing youth from joining gangs have been identified. A program consisting of a gang prevention curriculum and after-school recreational activities had positive effects, but the sample was quite small. In contrast, interventions seeking to redirect gang members into prosocial activities have been counterproductive. Therefore, significant effort should focus on designing, testing, and implementing effective strategies to prevent youth from joining gangs. Table, figures, appended scale-item descriptions, and 43 references (Author abstract modified)