Under a Federal grant, three California counties designed ISP's as an alternative form of supervision for high-risk probationers. These programs place offenders on small caseloads and require them to be employed, submit to random urine and alcohol testing, and perform community service. The evaluation design provided for the random assignment of offenders to either the experimental ISP or the control (routine probation) program. The study period began in January 1987 and continued through July 1988. The 1-year followup period was defined individually for each participant and began on the day of assignment to a program. The final sample in Contra Costa consisted of 170 offenders; in Ventura, 166 offenders; and in Los Angeles, 152 offenders. Detailed information on individual backgrounds, services received, and 1-year outcome measures (including recidivism and social adjustment indicators) were collected for each of the participants. At the end of the 1-year followup, approximately 25 percent of the ISP offenders in each site had no new incidents (technical violations or new arrests); approximately 40 percent had technical violations only, and approximately one-third had new arrests. Arrest rates for the experimental and control groups were virtually identical. Intensive supervision per se apparently is not more effective in deterring recidivism than traditional probation; however, it does have more punitive elements, thus offering another noninstitutional option for offenses whose severity requires a sanction more severe than regular probation. Also, the treatment elements accompanying the ISP programs did impact recidivism. ISP combined with expanded and improved treatment services may prove to be more effective in reducing recidivism than the three California programs. Appended supporting data, evaluation forms, and 102 references