Citation

Griswold DB. J. Crim. Justice 1978; 6(3): 247-252.

Copyright

(Copyright © 1978, Elsevier Publishing)

DOI

10.1016/0047-2352(78)90006-5

PMID

unavailable

Abstract

Three categories-self-reported, F.B.I., and Florida Division of Corrections-of measures designed to tap recidivism are compared. When comparisons are made using different baseline measures, it is observed that variation is greatest between rather than within categories. Further, when two log-linear regressions are fitted where only the recidivism measures are different, the results are radically different. The findings demonstrate that all recidivism measures are not necessarily equally valid or reliable and that the use of different measures can produce discrepant findings.