This study examined the frequency of arrest during the first 3 years following prison release for two cohorts of male youth released by the California Youth Authority in 1981 to 1982 and 1986 to 1987.
During the mid-1980’s research on career criminals swelled, with researchers studying offense frequency as well as specialization and offense mix. Recently, attention has been turned to developing analytical methods that effectively model criminal careers. Scant attention, however, has been devoted to the way in which criminal behavior patterns are related to individual characteristics. In order to probe this issue, the authors used negative binomial models to examine a variety of variables previously linked to the frequency of offending. The sample consisted of 3,586 male parolees who were released from the California Youth Authority during two periods; 1,928 were released from 1981 to 1982 and 1,658 were released from 1986 to 1987. The dependant variable was the number of arrests accumulated during the first 3 years following release from prison. Results indicate that individual and geographic characteristics are useful in the prediction of average arrest frequencies. Furthermore, those variables that were predictive of arrest frequencies were also predictive of offense variation. The findings also indicate that there are possibly important distinctions to be made among offending populations that would increase predictability of arrest frequencies and offense variation. One important limitation of the research is that the sample was derived from a select sample of California Youth Authority parolees and may not be generalizable. Future studies should focus on time and age cohorts in the modeling of criminal behavior. Tables, figures, notes, references