The purpose of this study was to assess the relative cost-effectiveness of supervised probation, unsupervised probation, and community service. Data were collected from several sources: input-intake forms used by the State of Maryland, probation officers' case record files, Maryland state police rap sheets, FBI sources, and interviews with Maryland probationers. Non-violent, less serious offenders who normally received probation sentences of 12 months or less were offered randomly selected assignments to one of three treatment methods over a five-month period. Baseline data for probationers in each of the three samples were drawn from an intake form that was routinely completed for cases. An interim assessment of recidivism was made at the midpoint of the intervention for each probationer using information drawn from police records. Probationers were interviewed six and twelve months after probation ended. Demographic information on the probationers includes sex, race, age, birthplace, marital status, employment status, and education.