Research on offender rehabilitation has indicated that for programs to be effective they must target high-risk offenders, address offenders’ criminogenic needs, and match program characteristics to offenders’ learning styles and cognitive abilities. The latter requirement is known as the principle of responsivity. Probation officers or community supervision officers can engage the principle of responsivity by offering programs or imposing treatment referrals that address their clients’ learning styles, level of motivation, and intellectual capabilities. The CMC model is used by about one-fourth of probation and parole agencies nationwide. The model involves a structured, interview-based assessment which classifies an offender into one of five strategy groups as the basis for their differential supervision. The CMC categorizes offenders based on their stability, motivation, number and types of needs, suggestibility, and antisocial values. The authors review prior evaluations of the CMC model and then describe their own evaluation methods, which involved a sample of 1,017 adult offenders who entered felony probation in a large urban county between March 15, 1991, and February 18, 1992. The offender sample was assigned to either a treatment or control group. The treatment group included 581 offenders who were classified and differentially supervised according to the CMC model. The control group of 436 offenders were classified but not subjected to differential supervision. Measures of case supervision included: quality of the problem statement, quality of the force-field analysis, appropriateness of the supervision priority, validity of case classification, adequacy of supervision objectives, adequacy of the offender action plan, adequacy of the officer action plan, timeliness of supervision plan, and compliance with CMC strategy. Results of statistical analyses revealed that differential supervision was not delivered to a large portion of the offenders for whom it was intended. Another striking finding indicated that the integrity of case supervision was only weakly correlated with case outcomes. Overall, these findings suggest that training in CMC heightened officers’ understanding of offenders’ motivations and needs, leading officers to view offender misconduct in a more lenient light, which ultimately produces the appearance of favorable offender outcomes. Findings hold implications for the design of evaluation studies. Tables, notes, references