This article introduces two new conceptualizations of desistance based on individuals' personal assessments of their own movement away from crime. Drawing on qualitative accounts of changes in offending, survey items indexing subjective desistance and reference group desistance are developed. A representative community sample of young adults is used to compare these new conceptualizations of desistance against more established measures derived from changes in arrest and self-reported crime. Results indicate that the prevalence and predictors of desistance vary with these alternative conceptualizations. Although relationship quality is consistently related to each desistance measure, the effects of prior crime, peer relationships, race, gender, and parental status depend on the outcome under consideration. These results show both the generality of the desistance process and the utility of comparing subjective accounts of this process alongside official and self-reported behavioral measures.
- Life course