|
Abstract
|
PURPOSE: Most criminal career research must use data with inherent limitations. Two common areas of data-imposed limitations for recidivism and desistance researchers are short observation lengths (i.e., follow-up periods) and jurisdiction-specific criminal histories. To help researchers understand the effect of data-imposed limitations, this study explores and compares the criminal career patterns observed (1) with varying follow-up periods and (2) using individuals' jurisdictionally limited vs. national criminal history records.
METHODS: This study uses criminal history data to track a sample of about 70,000 prisoners selected to represent the more than 400,000 prisoners released in 2005 from 30 states from their first arrest to 9 years after their release. Using a multivariate analysis, this study measures desistance as the cessation of arrest activity after release from prison in 2005. First, the study examines if the correlates of desistance after release from prison are consistent across 3-, 6-, and 9-year follow-up periods. Second, the study examines if the correlates of desistance are consistent when post-release data only describe arrests that occurred in the state of release from prison compared to arrests occurring within and outside the state of release (i.e., nationwide). Given their unique offending patterns and pathways to crime, separate models are presented for males and females.
RESULTS: Those who are falsely labeled as desisters in the shorter follow-up periods or when using just in-state arrest records differ from those labeled as desisters in the more complete criminal histories.
CONCLUSIONS: This study has implications for developmental and life-course theories that rely on empirical support from limited recidivism follow-up periods and jurisdiction-specific criminal histories. By understanding the effects of data-imposed limitations of follow-up periods and geographical coverage on criminal career research, the validity of criminological theories of desistance across the life course and across place can be better measured and understood.
Language: en |