Evaluating offender rehabilitation programs

Research summary
Vol. 12 No. 2
March 2007

Question

How can we tell whether interventions with offenders reduce recidivism?

Background

There is considerable controversy concerning the effectiveness of rehabilitation programs for offenders who have committed serious crimes, particularly sexual offenders. On average, sexual offenders who attend treatment have lower recidivism rates than offenders who do not attend treatment. These findings have not been persuasive to skeptics, however, who attribute the observed differences to bias in the way the research was conducted. Reviewers agree that most of the current studies have significant limitations, but they disagree about which studies are the most informative. Consequently, a credible set of guidelines concerning research quality would help advance the debate about the effectiveness of treatment for sexual offenders and other serious offenders.

Method

A committee of international experts in research evaluation was formed with a mandate to critically review the sexual offender treatment outcome literature and to promote high quality evaluations. The committee members each wrote general recommendations for rating study quality and detailed critiques of specific studies. As well, the published literature on study quality rating scales was reviewed. The recommendations of individual committee members were integrated with the existing ratings scales to create a consensus document focused on sexual offender treatment outcome studies.

Results

The committee defined a high quality study as one in which there was high confidence that there was no more than minimal bias in the findings. In other words, readers could be confident that the differences between the treated group and the comparison group was based on the effect of treatment and not on other factors. The committee identified 20 key features that should be considered when making these ratings. These features included the description of the treatment delivered, experimenter expectations, sample size, program drop-out, composition of the comparison group, outcome variables, and data analysis. Once the key features have been rated, the reviewer makes a global judgement concerning overall study quality: strong, good, weak or rejected.

A field test found that undergraduate students could reliably use the coding scheme after a week of training. Simply reading the guidelines, however, was insufficient for resolving the long standing debate between researchers concerning the quality of sexual offender treatment outcome studies. Although experts agreed about the features of strong studies, they continue to disagree about how the various key features of the study should influence the overall rating of study quality.

Policy implications

  1. Researchers should consider the guidelines when conducting evaluations of sexual offender treatment programs. Even when the ideal procedures cannot be used, there are feasible and inexpensive ways to improve upon typical evaluation studies by considering the key features described in the guidelines. Researchers examining other types of offender programs may also be interested in adapting the guidelines to apply to their specific areas (e.g., substance abuse, domestic violence). Structured risk assessments are the most accurate and should be routinely used.
  2. Funding decisions concerning program evaluations should consider the quality of the research designs. Priority should be given to evaluations that will contribute the most to cumulative knowledge.
  3. Professionals charged with implementing rehabilitation programs for sexual offenders should understand the key requirements of program evaluation and structure their programs to facilitate evaluations that are as informative as possible.

Source

For further information

R.K. Hanson, Ph.D.
Corrections Research
Public Safety Canada
269 Laurier Avenue West,
Ottawa, Ontario K1A 0P8
Phone: 613-991-2840
Fax: 613-990-8295
E-mail: Karl.Hanson@ps-sp.gc.ca

Date modified: