Members of the diversion-referral group were channeled from normal juvenile processing to one of several community-based treatment programs. To evaluate the effectiveness of the program, a random sample of subjects was selected for intensive study over 12 months. Each subject was administered a self-reported delinquency questionnaire which listed a series of delinquent behaviors ranging from status offenses to felony acts. Subjects also rated the seriousness of various delinquent acts and completed a series of attitudinal and self-esteem items. Official records were used to create offense histories for each subject. At the beginning of the study, the three groups did not differ significantly on any of the performance measures. Six months later there were some differences among the groups, but they did not indicate dramatic change. Compared to their status at time 1, the referred group was significantly better at time 2 in terms of positive image, family ties, legal respect, self-reported drug use, and number of arrests. The dropout group consistently performed poorer at time 2 compared to the diverted group. The findings suggest that diversion can be beneficial for selected juveniles, but persons not amenable to completing diversion programs may be significantly damaged by their failure. Research should show that juveniles with particular characteristics have a high probability of benefiting from a program before they are diverted to that program. 5 tables and 36 footnotes.