The sample used in the project research consisted of over 3,600 felony probationers directly sentenced to either Intensive Supervision (IS) or Restitution Center Probation (RDP); these programs had both been legislatively designated as diversion programs during the sample period. The survey of probation case files was taken over 65 items: 10 identification and 7 case file items, 18 variables on offender social history, 5 on alcohol and drug abuse, 14 criminal history variables, and 11 instant offense items. Intake, assessment, and discharge information was merged with the case file data. Nine variables -- employment status, alcohol abuse, illicit drug use, total number of prior arrests, number of prior felony convictions, instant offense type, attitude, and prior probation/parole revocations met the criteria for the discriminant function. The complete risk assessment instrument consists of the risk assessment scale and the outcome table which interprets the risk assessment score. The continuum of scores allows for flexibility in incarceration alternatives. A random sample of parolees has been identified as a validation sample for the risk prediction model that will be piloted in order to identify and correct potential problems in the instrument, assess the face validity and political resistance to the instrument, and inform management for policy decisions and procedural guidelines. 5 tables, 5 appendixes, and 20 references