NCJ Number
221968
Journal
Evaluation Review Volume: 27 Issue: 3 Dated: June 2003 Pages: 267-289
Date Published
June 2008
Length
23 pages
Annotation
This review of the history of the British Home Office's use of randomized experiments in testing the effectiveness of criminal justice projects shows that the decision about whether or not to use this research methodology rested as much on fashion, personality, and politics as on objective scientific criteria.
Abstract
The Home Office's Research Development and Statistics Directorate has conducted or funded hundreds of criminological research projects since its creation in 1957 as the Home Office Research Unit. Only four of them have involved the research methodology of random allocation of participants to experimental and control groups; however, these four research efforts were critically important for the development of criminal justice research in the United Kingdom. The four random-allocation research projects began between 1965 and 1973, and the findings were published between 1972 and 1978. One was an evaluation of the relative effectiveness of a therapeutic community and a traditional paternalistic regime in a residential school for delinquent juveniles. The second randomized experiment examined the effect of social work in prison; and the third randomized-allocation research project was a major evaluation of the relative effectiveness of intensive and standard probation programs. The fourth randomized-allocation research effort tested the effect of intensive prison welfare work on inmates with short-term sentences. Reasons why these four research projects marked both the beginning and the end of random-allocation experiments by the Home Office had little to do with their design or results. This article explains why the medico-psychological model, which has viewed treatment as the way to reduce crime and the randomized experiment as the highest form of evaluative research, was undermined by a political climate hostile to offender treatment and criminal justice administrators with no background in scientific research. 21 references