U.S. flag

An official website of the United States government, Department of Justice.

An Assessment of Substance Abuse Treatment Programs in Florida's Prisons Using a Random Assignment Experimental Design

NCJ Number
Date Published
December 2015
108 pages
Using a three-pronged approach, this study examined the impact of substance abuse treatment for inmates in Florida's prisons.
First, the study assessed the results of an evaluation of the effectiveness of substance abuse treatment programming in Florida prisons. The evaluation used a randomized experimental design that involved data collection for in-prison substance abuse treatment programming from January 2006 through December 2008. Second, the study examined the impact of multiple methods of prison-based substance abuse treatment and the role that duration and recency of treatment had in determining recidivism and post-prison employment outcomes for approximately 11,000 released inmates. Third, multiple statistical techniques were used to determine whether and to what extent evaluation outcome studies of correctional substance abuse treatment programs are influenced by the type of research design and statistical methods used. A major finding of this study is that aftercare and transitional substance abuse programming were significant in increasing employment and reducing recidivism after release. The study also found that for inmates who completed in-prison substance abuse treatment programs, those who did so closer to their release date were less likely to recidivate. Regarding outcome differences based on evaluation methods used, some methods yielded similarities in the direction of the effect of the treatment on recidivism; however, other methods showed results in the opposite direction, indicating that evaluation outcomes are dependent on the evaluation method used. Random assignment yielded the most support for the effectiveness of prison-based treatment in reducing recidivism. Due to the non-significance of the majority of evaluation results across evaluation methods, no valid conclusions were offered about similarities across evaluation statistical methods used. Policy implications and future research directions are discussed. 21 tables, 13 appendixes, and 88 references

Date Published: December 1, 2015