Criminal Justice Evaluation Techniques - Methods Other Than Random Assignment (From Handbook of Criminal Justice Evaluation, P 259-302, 1980, Malcolm W Klein and Katherine S Teilmann, ed. - See NCJ-73970)

• NCJ Number: 73979
• Author(s): G A Grizzle; A D Witte
• Date Published: 1980
• Length: 44 pages
• Sale Source: Sage Publications, Inc
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

An overview of contrasting paradigms is presented, and accommodations that can be made among them to obtain complete evaluations of criminal justice programs are suggested.

Abstract

Six evaluation techniques are described: process evaluation, multivariate statistical techniques, time-series analysis, cost-effectiveness analysis, benefit-cost analysis, and multiattribute utility analysis. All of them extend and strengthen traditional criminal justice evaluations utilizing experimental or nonequivalent control group designs. Process evaluation analysis provides detailed information on the nature and evolution of the program evaluated and enables evaluators to monitor ongoing programs, especially developing or unstable ones. Multivariate statistical techniques and time-series analysis may be used either to strengthen traditional evaluative designs or to evaluate programs when no control or comparison group is available, such as is often the case in correctional programs. However, to be used effectively, these two techniques require advanced technical and statistical skills on the evaluator's part. Cost-effectiveness analysis provides a cost per unit of impact by dividing various program output measures into the costs incurred to obtain them. A recommended method is increased utilization of this relatively simple and easily understandable technique. Benefit-cost analysis and multiattribute utility analysis convert various program outputs to a common unit of measure and sum them up to obtain an overall measure of program effectiveness. Benefit-cost analysis is relatively difficult to use in many criminal justice settings and requires advanced technical skills. Multiattribute utility analysis, on the other hand, is an easy technique and especially useful when program effects have been measured by an experimental design and group values are clear. The six techniques are not mutually exclusive but complementary. A combination of all six techniques provides a truly complete evaluation of a criminal justice program. However, such a complete program evaluation will rarely be possible. Extensive endnotes and a list of 130 references are appended.