Summary and Conclusions

Abstract

The purpose of this study was to investigate the usefulness of various statistical survival time models for prediction of the length of time that a prison releasee will remain free after release from incarceration. In particular, we were interested in the following three issues. First, we wished to compare the results from parametric statistical models with those from nonparametric models. Second, we wished to examine the importance of including individual characteristics as explanatory variables in such models. Third, we wished to examine the usefulness of so-called split models, in which not every individual would eventually return to prison.