Two measures of criminal behavior were used -- arrests for new crimes and revocations -- to compare offenders who participated in IPS and those who did not. The offenders included in this study were drawn from four groups: those sentenced to standard probation, those sentenced directly to IPS, those reinstated to IPS due to probation violation, and those sentenced to prison and released. Offenders sentenced to IPS were riskier in terms of the severity of their offense and the likelihood that they would commit future criminal acts. About half the offenders in one county, and one-third of the offenders in the other county, were arrested for new crimes within three years of their initial sentencing. GAO concluded that the Arizona IPS program was effective in controlling criminal behavior because fewer offenders under IPS supervision than offenders under standard probation supervision committed new crimes. However, once offenders finished the IPS program and moved to standard probation, arrests among the group increased, leading to the conclusion that a sentence to IPS did not ensure public safety. 11 tables, 8 figures, and 4 appendixes