NCJ Number
201304
Date Published
June 2002
Length
85 pages
Annotation
This study examined the effectiveness of the redesign of the Multnomah County’s adult community supervision program, which began using a system of graduated supervision based on risk assessment in 1997.
Abstract
In an effort to increase public safety without increasing expenditures, Multnomah County (Oregon) redesigned their adult community supervision program, implementing differing levels of supervision based on offender risk level. The Oregon Case Management System includes an Initial Risk Assessment Instrument and a Risk Assessment Instrument that are used to classify those under community supervision as either a high-, medium-, or low-risk level. Corresponding levels of supervision are then employed, thus managing parolees more effectively by directing the bulk of resources toward the medium- and high-risk offenders. This study employed a quasi-experimental design using non-randomized comparison groups to determine the effectiveness of the redesign. The groups consisted of adult offenders admitted to community supervision during 1995, 1998, and 2000. The 1995 cohort experienced the traditional type of community supervision, while the 1998 and 2000 cohorts received the new version of graduated supervision levels based on risk level. Comparisons involving multivariate logistic regression analysis revealed that overall the redesign accomplished its objective of assigning offenders to differing levels of supervision, thereby reallocating Department resources toward the medium- and high-risk offenders. However, the objective risk assessment system, described as the backbone of the redesign, was discovered to contain serious flaws. The override rates were found to be much higher than the acceptable standard, with most overrides serving to bump the offender into a lower risk category. As such, the supervision level and assessment of risk are based on subjective rather than objective measures. Recommendations include an evaluation of the risk assessment system to determine the reason for the high rate of overrides. Future research should focus on how the type and intensity of supervision impacts offender behavior. Appendix
Date Published: June 1, 2002