Study 1
Hess and Turner (2021) used a quasi-experimental design to compare outcomes of the California Prison Industry Authority (CALPIA) participants (treatment group) with a matched comparison group who were eligible to participate in CALPIA but were waitlisted (i.e., waitlist comparison group). Specifically, the study compares CALPIA participants who were enrolled in the program for at least 6 months, and released between August 2014 and July 2018, with incarcerated individuals who were accepted into the program but released prior to participation. Overall, a total of 8,603 individuals who were in custody in CDCR prisons, between August 2014 and July 2018, were included in the study.
The treatment group included 2,453 individuals who participated in CALPIA, whereas the waitlist comparison group included 6,150 CALPIA-qualified individuals who were released prior to their participation in the program. There were statistically significant differences found between CALPIA program participants and the waitlist group on characteristics such as sex, ethnicity, admission age, custody years, controlling offense, prior serious offense count, current serious offense flag, current violent offense flag, risk assessment score, region, destination, and programming. To minimize group differences, propensity score matching was used.
Following matching techniques, no statistically significant differences between the groups remained. The CALPIA treatment group was predominately male (85.6 percent), 34.4 percent Hispanic, 29.8 percent Black, 29.5 percent white, 4.4 percent other, 1 percent Asian/Pacific Islander, and 0.8 percent American Indian. Admission age ranged from 14 to 55 years and older, with about 52 percent of the sample serving 0 to 4 years in prison. The most common controlling offense was assault, followed by property offense. Most individuals did not have any prior serious offenses or current violent or serious offense flags. Most individuals were assessed as low on the risk assessment (43.6 percent), followed by moderate (28.2 percent). Most individuals came from Los Angeles (30 percent), followed by North Central California (27.7 percent). Most individuals were released to parole (72.1 percent).
The waitlist comparison group was predominately male (86.4 percent), 34.9 percent Hispanic, 27.8 percent Black, 31.1 percent White, 4.4 percent other, 1 percent Asian/Pacific Islander, and 0.7 percent American Indian. Admission age ranged from 14 to 55 years and older, with about 52 percent of the sample serving 0 to 4 years in prison. The most common controlling offense was assault, followed by property offense. Most individuals did not have any prior serious offenses or current violent or serious offense flags. Most individuals were assessed as low on the risk assessment (43.3 percent), followed by moderate (27.7 percent). Finally, most individuals came from Los Angeles (29.4 percent), followed by North Central California (27.7 percent). Most individuals were released to parole (72.1 percent).
The outcome of interest in the study was recidivism, measured as rearrest, reconviction, and reincarceration. Though outcomes were measured at 1, 2, and 3 years, the CrimeSolutions review of this study focused on rearrest, reconviction, and reincarceration at 3 years.
Data from the study were provided by two sources: 1) the California Department of Corrections and Rehabilitation (CDCR) Office of Research, and 2) the California Department of Justice. The CDCR Office of Research provided the list of program participants and waitlist participants, including their demographics, work/program history, movement data, and needs assessments. CDCR also provided return-to-custody data. The California Department of Justice provided rearrest and reconviction data. No subgroup analyses were conducted.