NCJ Number
120519
Journal
Journal of Quantitative Criminology Volume: 5 Issue: 3 Dated: (September 1989) Pages: 215-221
Date Published
1989
Length
15 pages
Annotation
The probability of recidivism among juveniles relates to age in a curvilinear manner.
Abstract
Rates of reoffending increase as a function of age up to a point of peak activity and then decrease with increasing age. Therefore, the forms of recidivism functions between cohorts of widely varying ages will be considerably different. Not applicable for the analysis of recidivism rates is the Cox proportional hazards regression model which assumes proportional hazards; applicable models will posit curved nonmonotonic hazard functions. Fits of the exponential, Weibull, and non-normal models to recidivism data collected on samples of chronic juvenile offenders are compared. Closer estimation of recidivism functions is found using the latter model. Implications for recidivism modeling development are discussed. 2 tables, 5 figures, 17 references. (Author abstract modified)