U.S. flag

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

The Georgia Cognitive Skills Experiment: Outcome Evaluation, Phase I

NCJ Number
Date Published
March 2001
93 pages

This report presents outcome results for Phase I of an experimental study of the impact of the Georgia Cognitive Skills Program on offender attitudes, in which experimental participants completed the program by July 1998; participants and comparison groups were followed through the duration of the four-month program plus 12 additional months.


The author presents outcome results for the first phase of an experimental study to determine the impacts of the Georgia Cognitive Skills Program on offender attitudes. Findings of the study address the effects of cognitive skills programming on the following parolee outcomes: cognitive skills and offender thinking patterns as measured by the Colorado Offender Attitude Survey; returns to prison; rearrests and revocations; technical violations; and employment. The program itself is structured into 35 lessons that cover seven key components: problem-solving; creative thinking; social skills; management of emotions; negotiation skills; values enhancement; and critical reasoning. In this document, the author reviews the evaluation methodology, data collection, sample characteristics of participants and control-group members, measures and indicators of parole performance, data analysis, and results. Results for Phase I, as discussed in this document, showed modest but not significant differences between those randomly assigned to the program and comparison groups. The author notes that findings were consistent with other large-scale cognitive interventions which are cited by the author; the emerging pattern across evaluations of cognitive programs find the largest effects for smaller programs; and findings were also consistent with research conducted in other settings. Findings indicated that program completion substantially reduces offender recidivism, and a “risk effect” was not found to suggest that the program is more suitable to medium- and high-risk offenders than low-risk offenders, as the results were beneficial to both groups.

Date Published: March 1, 2001