The report of the initial evaluation of the ISP is based on the analysis of the performance of 1,528 probationers who were sentenced between October 1978 and June 1979. The probationers were selected for program participation based on their likelihood of failure, as indicated by application of the 10-item classification instrument developed for that purpose. Outcome data cover the period of assignment to a local ISP probation component through June 1980. The outcome measures used address the issue of success or failure in terms of rate of failure as a function of estimated level of risk, nature of failure, degree of seriousness and harm of all new offenses, and the sentences received for such offenses. For purposes of comparison, a comparable group of nonprogram probationers, sentenced within a 4-month period prior to implementation of the ISP, was identified in each of the participating counties. Results show that 'high risk' probationers supervised by the program succeeded at a higher rate than expected; those who did not succeed failed 8 percent less often for committing a new offense. Second offenses committed by ISP probationers were less likely to be crimes against persons and more likely to be victimless crimes. Over 40 percent of probationers supervised by ISP officers were transferred to routine departmental supervision in 12 months or less. Ninety-five percent of ISP probationers transferred to routine supervision continued to succeed regardless of their original likelihood of failure. Probationers at the extreme levels of risk were twice as likely to be in absconder status if they were under routine departmental supervision than if supervised by ISP officers. All ISP units performed at similar levels of success regardless of geographic location. Five tables are provided. Supplemental data are appended. (Author abstract modified)