NCJ Number
85701
Date Published
1982
Length
16 pages
Annotation
Research needs in correctional classification include improvement in the utility of offender data, empirical research results to provide linkage between research design features and design criteria, and the development of a classification model.
Abstract
Recently, empirically derived and theoretically informed classification research has produced advances in taxonomic efforts. Reduction of attribute space, typology construction, algorithmic procedures, and the use of computers have produced breakthroughs. A number of researchers are now working with automated classification methods using computers. The issue for criminal justice is to produce data of sufficient quality to make the new methods useful. There have been many offender classification systems and numerous offender typologies produced by academics, but there has been no systematic evaluation of them. Parole prediction studies have been questioned by those who argue that unless classification systems can predict differential program impacts, there is little reason to classify. Perhaps longitudinal studies which avoid the ambiguities of the use of the term recidivism and carefully specify what is meant by success and failure in realistic terms can be performed. Research to validate classification instruments, methods, and procedures also remains to be done. Twenty-six references are listed. (Author summary modified)