Study 1
Rengifo and Stemen (2009) evaluated the policy implications of Kansas’s Senate Bill (SB) 123. SB 123 was a mandatory sentencing option for judges who adjudicated nonviolent individuals with a criminal history involving drugs. The treatment group included all “SB 123 Eligible” individuals sentenced to SB 123 between Nov. 1, 2003, and Aug. 31, 2005 (n = 1,494). The comparison group included all “SB 123 eligible” individuals sentenced to prison, court services, or standard community corrections without SB 123 drug treatment (n = 4,395) from April 1, 2001, through Aug. 31, 2005. The study sample included only individuals who completed at least 18 months of community-based supervision during the study to account for differences in time at risk. The CrimeSolutions review of this study focused on the comparisons between “SB 123 Eligible” individuals in the treatment group and “SB 123 eligible” individuals in the unmatched comparison group sentenced to standard community corrections.
The treatment group was 81.6 percent white, 17.3 percent Black, 8.2 percent Hispanic and 1.1 percent other. The comparison group sentenced to prison was 77.4 percent white, 21.3 percent Black, 14.1 percent Hispanic, and 1.2 percent other. The comparison group sentenced to standard community corrections was 75.5 percent white, 22.7 percent Black, 8.9 percent Hispanic, and 1.7 percent other. The comparison group sentenced to court services was 78.2 percent white, 20.1 percent Black, 14.1 percent Hispanic, and 1.6 percent other. The treatment group, prison comparison group, standard community corrections comparison group, and court services comparison group were all mostly male (71.0 percent, 80.7 percent, 76.2 percent and 73.5 percent, respectively).
Recidivism outcomes were tracked for both treatment and comparison groups through February 2007. The main outcome measures were recidivism, reconviction for a new offense, and revocation to prison for a technical violation in the first 18 months of follow-up in the community. Data was collected from the Kansas Department of Corrections (KDOC). Numerous offender-level covariates were included in the analyses. Measures were included that looked at their criminal history and categorized the seriousness of offenses committed before their drug possession convictions (from 1 = least serious to 5 = most serious). The study measurement included their additional charges at time of drug possession convictions and their demographic attributes.
In the first evaluation component of the study, multinomial logistical regression estimated the effect of SB 123 relative to other sentences on recidivism at the 18-month follow-up date. The authors did not conduct subgroup analyses.