The study was a cooperative effort between a large municipal police department in California and California State University. The primary intent was to determine, by a carefully controlled experimental design, whether juvenile offenders randomly assigned to alternative treatment strategies were more or less likely to be rearrested than a comparison control group which received no diversion services. The two groups in the study were less serious offenders (all youths who normally would have been diverted under the existing program) and more serious offenders (who normally would have been referrred to probation with a request for probation supervision). Those eligible for diversion were randomly assigned to three treatment programs: counseling and release, in-house diversion counseling, and referral to community release. Recidivism rates of the groups were determined by checking police department records for rearrests at 6 and 12 months. A total of 512 subjects participated in the experiment and 210 were interviewed. Results showed that more serious offenders showed higher rearrest rates, although rearrest rates of youths assigned to alternative treatment strategies did not differ significantly. Few differences in services were received by the various disposition groups. Those receiving additional services were rearrested as often as those not receiving additional services. Many of the diversion programs served primarily low risk offenders who would not have penetrated the juvenile justice system further. The assumption that diversion reduces contaminating contact with other delinquents was found questionable, and the findings did not support the assumption that diversion reduces recidivism. Several recommendations for modifications to diversion programs were suggested: resources should be concentrated on higher risk clients, monitoring diversion procedures should be developed to assure prompt disposition and service linkage, and provision should be made for a no-service option. Tabular data, footnotes, and about 30 references accompany the study.