The Intensive Supervision Parole (ISP) program, established by the Texas Board of Pardons and Paroles, was created to help alleviate prison crowding and reduce system costs in Dallas and Houston prisons. The primary goals of the ISP program were 1) to serve as an intermediate sanction option between routine parole and prison, and 2) to reduce commitments to the Texas Department of Corrections (DOC). The program identified parolees under community supervision who had the highest probability of returning to prison and provided more surveillance, control, and contacts with parole officers than traditional supervision.
The ISP program targeted parolees who possessed a serious prior criminal record, displayed poor parole performance, and were likely to incur a formal violation (thus, if a violation did occur, a return to prison was highly probable).
ISP relied heavily on specific deterrence and incapacitation and less on rehabilitation. Deterrence theory suggests that individuals assess the consequences of their actions, both positive and negative, and commit crimes only when it is in their self-interest to do so. Such deterrence-based programs, through increased monitoring and threats of detection and sanctions, attempt to change individuals’ perceptions of the costs of crime and influence their choices regarding participation in crime (Petersilia and Turner 1990).
The ISP Program mandated increased contacts and supervision. Parolees were required to participate in 10 personal contacts per month (one in the home, one in the office, four in any location, and four by phone); maintain employment; and participate in other activities as deemed necessary by the parole officers, including drug testing. Parolees not working or attending school full time were required to verify job search efforts, including enrolling in job preparation courses, applying for positions, and contacting the Texas Employment Commission.
The ISP program also included case management and intermediate sanctions. Case conferences served as the primary mechanism to decide on intermediate sanctions. Parole officers met with other staff and the parolee to discuss the parolee’s behavior and to reach a joint decision about which intermediate sanctions were appropriate such as curfews, house arrest, electronic monitoring, or placement in a halfway house.
The ISP program was designed to last 9 to12 months, and parolees could only be removed from the program if they fulfilled program requirements. If ISP parolees was revoked to prison, then they were released back to ISP if paroled again.
The evaluation of the ISP program by Turner and Petersilia (1992) found that ISP participants in Houston were significantly more likely to have technical violations than those in the control group, which indicated that ISP had a negative effect on technical violations
Study 1 (Dallas, Texas)
In Dallas, Turner and Petersilia (1992a) found there were no statistically significant differences in reconviction rates between Intensive Supervision Parole (ISP) participants and the control group after 1 year.
There were no statistically significant differences between the groups in Dallas rearrest rate.
Paid Employment Rate
There were no statistically significant differences between the groups in Dallas in paid employment rate.
There were no statistically significant differences between groups in Dallas for technical violations.
Study 2 (Houston, Texas)
In Houston, Turner and Petersilia (1992b) found there were no statistically significant differences in reconviction rate between ISP participants and the control group after 1 year.
There were no statistically significant differences between the groups in Houston in rearrest rate.
Paid Employment Rate
There were no statistically significant differences between the groups in Houston in paid employment rate.
ISP parolees in Houston were statistically significantly more likely to have technical violations than those in the control group (81 percent versus 33 percent, respectively), which indicated that ISP had a negative effect on technical violations.
Turner and Petersilia (1992a) used a randomized experimental design to evaluate the impact of the Intensive Supervision Parole (ISP) program in Dallas, Texas on parolees’ behavior and system costs. The implementation of ISP was assessed to determine whether parolees who participated in the program had fewer arrests and higher employment rates than those parolees who did not participate in the program and received routine parole.
Parolees were deemed eligible for the study if they had both a Salient Factor score of 0–7 (high risk of post-release recidivism) when released from prison and a Risk/Needs Reassessment score of 0–18 (high risk of post-release recidivism) when reclassified 6 months or later after release). Additional criteria excluded those who had been released from prison before January 1986, who had pending felonies, who were currently in jail or prison, who had less than 6 months remaining on parole before discharge, and whose time in community supervision was deemed too short for the ISP program.
The ISP study began in August 1987 and continued through 1988. Once ISP supervisors in Dallas determined a parolee was eligible for ISP, they referred the case for random assignment to the experimental group (ISP) or the control group (routine parolees). There were 221 parolees in Dallas assigned to one of the two study conditions. In Dallas, 91 parolees were assigned to the control group and 130 to ISP. Information was not provided regarding any differences between groups pre-random assignment.
Data for this study was drawn from three sources. First, the Background Assessment Form, which was completed immediately after random assignment, provided information about the parolee’s prior record, demographics, current offense, drug use, risk of recidivism, treatment history, and need for treatment. Second, the Data Collection Form (DCF), completed 6 months after assignment, provided information on program services (number of contacts, counseling sessions, and drug tests) and technical violations, new arrests, employment, and restitution. Third, the DCF was completed again at 12 months and collected identical information for the 7 to 12 months after assignment. The DCF also included a status calendar with the date the parolee was placed on and removed from ISP or routine parole, dates of entry into and release from jail, residential placement, and prison.
Analysis for this study included t-tests and chi-squared tests. Recidivism was measured based upon the number of technical violations or arrests, the seriousness of the violations or arrests, the average total number of arrests, reconviction rate, and the annualized arrest rate (controlling for time not incarcerated). Employment was measured based upon information drawn from the DCFs, regarding parolees who participated in work training programs or secured paid employment. Each parolee was followed for 12 months, beginning on the day of assignment to the experimental or control group.
In the same randomized experimental design, Turner and Petersilia (1992b) also evaluated the impact of the ISP program in Houston, Texas. Once ISP supervisors in Houston determined a parolee was eligible for ISP, they referred the case for random assignment to the experimental group (ISP) or the control group (routine parolees). There were 458 parolees in Houston assigned to one of the two study conditions. In Houston, 219 parolees were assigned to the control group and 239 to ISP. Information was not provided regarding any differences between groups pre-random assignment.
ISP participants in both Dallas and Houston shared similar sociodemographics, prior criminal records, risk-of-recidivism scores, and employment characteristics; however, Dallas parolees had slightly more serious records of prior criminality. Across both sites, more than 90 percent of participants were male, and 60 percent were black, 29 percent were white, and 11 percent were Hispanic. The average age of participants was 31 years at the time of study involvement. About 85 percent served at least two prison terms, averaged eight to nine prior arrests, and six to eight prior convictions. Nearly all parolees were classified as being at high or intensive risk for recidivism. Prison sentences for parolees were about 10 years, and most were released to parole supervision after 22 months. Typically, parolees assigned to this study had already been on parole for 11 months. No within-site statistical differences between ISP and control groups were found for these characteristics.
The same data on technical violations, new arrests, employment, and restitution was collected. The same statistical analysis techniques were employed.
To determine cost savings for Intensive Supervision Parole (ISP), Turner and Petersilia (1992) estimated the total criminal justice dollars spent on each participant during the 1-year follow-up period, including both corrections and court costs. This was accomplished by estimating the costs of each type of sanction or service used by study participants and using information from the status calendar.
Analysis indicated that ISP, coupled with intermediate sanctions, cost more than routine parole supervision and did not produce a subsequent reduction in imprisonment costs. The overall results indicated that the costs of supervising an ISP participant were 1.7 times that of supervising an individual on routine parole.