Study
Carey and Waller (2011) used a quasi-experimental design to evaluate Oregon’s 20 adult drug courts—in Benton, Clackamas, Clatsop, Crook, Douglas, Jackson, Jefferson, Josephine, Klamath, Lane, Linn, Malheur, Marion, Multnomah, Polk, Umatilla, Union, Wasco, Washington, and Yam–Hill counties—to investigate whether the adult drug courts were cost-beneficial and to identify best practices across Oregon’s Drug Court program. The overall sample comprised 13,035 participants, with an average of 244 participants per court program. Sixty-seven percent were male, 84 percent were white, 9 percent were Latino, and 5 percent were Black. The majority (75 percent) reported their drug of choice as methamphetamine, followed by marijuana (54 percent), alcohol (39 percent), cocaine (12 percent), and heroin (6 percent).
The treatment group of 5,655 drug court participants comprised all who entered into the programs between January 2000 and December 2006, with recidivism tracked through May 2010. The data was gathered from the drug court database of the Oregon Criminal Justice Commission. The comparison group of 7,380 was chosen in two stages. All drug court–eligible individuals were identified for the same period as the treatment group. Those who had previously been entered into drug courts or who were legally barred from participating were excluded after being identified within the Oregon Judicial Information Network database. The treatment group was then matched to the potential comparison sample (on aggregate within county) according to demographics and criminal history (county of residence, drug court–eligible charge, age, gender, race/ethnicity, prior drug charges, prior property charges, and prior person/violence charges). The treatment group was routed through the Oregon Drug Courts program, while the comparison group remained within the traditional criminal justice system.
Recidivism was analyzed based on rearrest data. Univariate analyses of covariance were used to compare the means of rearrest, as well as the means for time spent on probation, on parole, in jail, and in prison for both treatment and comparison group 3 years after the entry into the program (and comparison group equivalent). The analysis controlled for both demographics and criminal history, with the adjusted means being used to calculate costs. Crosstabs and chi-square analysis was used to identify differences in rearrests between the treatment and comparison group.
The study also looked at 335 aspects of adult drug court program implementation, part of the 10 Key Components of Drug Courts (National Association of Drug Court Professionals 1997), which are the operational guidelines for drug court implementation. Data on these practices were obtained through Web-based surveys completed by the drug courts. The researchers performed quantitative analysis to identify and measure practices that were conducive to lower recidivism, higher graduation rate, and greater cost savings. Many court practices were the same in all drug courts, making it impossible for this study to detect their potential positive effects. Where variation was deemed sufficient (if more than 15 percent of drug courts varied in the practice), however, these could be examined in relation to outcomes. Bivariate correlations were also used to examine the effects of demographics on the outcomes. T-tests were used to determine what court-level practices affected court-level outcomes.