NCJ Number
173790
Journal
Criminal Justice and Behavior Volume: 25 Issue: 2 Dated: June 1998 Pages: 226-239
Date Published
1998
Length
14 pages
Annotation
Results of this study show that it is possible to use test data and basic demographic information obtained in comprehensive psychological assessment of at-risk juveniles (ages 12-15) to make predictions of adult criminal status 10 years later.
Abstract
A total of 121 juvenile offenders assigned to a regional assessment center in Medicine Hat, Alberta, Canada, between 1979 and 1984 were administered a test battery. The tests included the WISC-R, an intelligence test for children and adolescents; the MMPI, an objective inventory of psychopathology; and the Rorschach test, a psychodiagnostic test in which participants provide verbal descriptions of a standard series of 10 printed inkblots. The offenders included 84 males and 37 females ages 12 to 15; among them were 48 Native Americans. In 1992 after a mean elapsed time of 9.9 years, all 121 subjects were followed up and classified as either guilty (n=61) or not guilty (n=60) of a serious offense as an adult. Stepwise discriminant function analysis was used to find the best subset of variables with which to distinguish between the adults with a serious criminal record and those without. In order of importance, the four significant predictors selected from among 20 candidate variables were the WISC-R comprehension subtest, gender, North American Native status, and DQ+ from the Rorschach. A correct classification rate of 77.8 percent (as opposed to a base rate of 50.5 percent) was achieved. 2 tables, 3 notes, and 36 references