Outcome Evaluation: Department of Corrections Risk Intervention Services
Background
In 2016, four of the Department of Corrections facilities began offering Risk Intervention Services-- an innovative suite of services including behavioral programming curricula, educational courses, and workforce development. Risk and other assessments at prison intake determine who is eligible to participate in RIS and identify which services should be administered.
Methods and Objective
This study explores the impact of RIS participation for individuals who received behavioral programming services between 2016 and 2019. RIS participation data for 731 individuals was matched into Vermont Crime Information Center data to examine RIS participation’s impact on recidivism. Researchers considered RIS’s impact using the statutory definition of recidivism, as well as academic conceptualizations of recidivism.
Key Findings
Using three different definitions/measures of recidivism, the study found RIS participants to have a recidivism rate of 14% (modified statutory definition), 27% (running recidivism rate), and 23% (three-year recidivism rate. These recidivism rates are lower than the 89.96% recidivism rate for high risk/high need Diversion participants found in a 2019 study.
Using the three-year recidivism rate for any new conviction, analysis found no statistical correlation between an individual’s risk score and subsequent conviction. This suggests RIS is having a positive impact on participants recidivism rate.
Notably, RIS participants with a domestic violence conviction (90) had a three-year conviction rate of 21%. This was lower than the rate of recidivism in a 2011 study which found a recidivism rate of 37% for those with prior history of domestic violence conviction.
Limitations
During the study period (2016-2019), the DOC was not consistently or uniformly collecting data on all Risk Intervention Services. As such, this study was limited to exploring the impact of behavioral programming services on recidivism and did not investigate the impact of corrections education or workforce development services. Further, data on those who were eligible but declined and those whose sentences were too short to participate was not available. Therefore, the study was not able to investigate if the program was serving marginalized people equitably.
Recommendations
The DOC Risk Intervention Services model appears promising.
The DOC should consider creating an administrative control group of people who did not receive services when the assessment tools were validated. Because these individuals did not receive the service, they can provide a benchmark recidivism rate for comparing RIS recidivism rates.
Future studies should use a three-year recidivism rate and focus only on new convictions rather than furlough violations and readmissions. This will make it easier to compare rates across years and programs and give a more holistic understanding of a person's behavior after programming is complete.