Rethinking Rehabilitation: Why Can't We Reform Our Criminals? By David Farabee AEI Press, 2005, $20

 These key findings and recommendations are also available here as an Adobe Acrobat PDF.

Findings

• Over 2,000,000 people are incarcerated in state, federal, or local correctional facilities in the United States, with more than 600,000 of them returning to their home communities each year.
• Nearly seven in ten of these released offenders will be rearrested within three years. More offenders are expected to be released in the wake of the Supreme Court’s recent ruling on federal sentencing guidelines.
• This trend has led to a renewed interest in expanding “social programs” for offenders, based on the assumption that drug use, poor education, limited vocational skills, etc. cause crimes.
• Drawing on relevant research as well on his own experiences evaluating correctional programs, Farabee finds underlying sources of bias in offender rehabilitation research. He also concludes that significant structural changes are needed, not minor policy tweaks, and that the effectiveness of intuitively appealing prison-based social programs—including the use of workbooks, videos, and group meetings—has been grossly overstated.

Recommendations

• Use an alternative approach based in part on the “Broken Windows” theory of crime prevention (but targeted at the individual level). That is, return to the basic principles of behavior and do a better job of detecting crimes and swiftly applying sanctions.
• De-emphasize prison as a sanction for non-violent re-offenses and increase the use of intermediate sanctions.
• Use prison programs to serve as institutional management tools, not rehabilitation.
• Mandate randomized, controlled evaluations for all publicly funded offender programs.
• Establish evaluation contracts with independent agencies (rather than the agency operating the program).
• Increase the use of indeterminate community supervision, requiring three consecutive years without a new offense or violation.
• Reduce parole caseloads to 15 to 1 and increase the use of new tracking technologies.

# # #