DESISTANCE FROM CRIME AND SUBSTANCE USE: A UNIVERSAL PROCESS OR BEHAVIOR-SPECIFIC?

Abstract

Several prominent criminologists have suggested desistance from crime is in many ways similar to desistance from substance use. While a review of this literature supports this proposition in general, most of this research has focused on desistance from either crime or substance use rather than considering change across both behaviors. Indeed, those few studies that consider both behaviors often find individuals persist in substance use despite desistance from crime. Despite this discrepancy, there has yet to be a systematic comparison between desistance from these two behaviors. This dissertation seeks to address this gap by asking (1) whether the same set of social and psychological factors that distinguish crime desisters from persisters also differentiate heavy substance use desisters from persisters and (2) to what extent individuals who are desisting from crime are also desisting from heavy substance use. In addition to addressing these two primary research questions, a set of substance specific and subgroup analyses were performed to assess whether the results differ across substance type (alcohol, marijuana, and hard drugs) or along the demographics of race and gender. These analyses were performed using data from the National Longitudinal Survey of Youth 1997 cohort. Desisters were identified using group-based trajectory modeling while multinomial logistic regression was used to examine the factors associated with desistance from each of these behaviors. The results of the analyses indicate that desistance from crime is associated with differences in social bonds and reduced levels of strain, while desistance from substance use is primarily associated with reduced levels of strain and individual personality differences. The substance specific analyses suggest different factors are associated with desistance from the use of different substances, while the race- and gender-specific analyses suggest differences across these demographics. The implications of these results for theories of desistance from crime and substance use are discussed as are the limitations of this dissertation and suggestions for future research.