Association for Public Policy Analysis & Management

Significance: There were approximately 850,000 arrested juveniles as of 2016, of which 61,000 were housed in correctional facilities. However, most had failed to reintegrate into the community, resulting in high recidivism rates. In this context, criminal desistance theory may be used to explore factors contributing to people’s offending behaviors, which impacts recidivism. Sampson and Laub’s theory on social bonds provides an appropriate theoretical foundation on how social bonds, such as employment, affect criminal desistance through exogenous circumstances. This study uses this theory to examine the extent to which work commitment is related to antisocial behavior among formerly incarcerated youth.

Methods: A longitudinal panel analysis with a random effects model, using robust standard errors to deal with the violation of the normal distribution of the dependent variable, was conducted to examine the association between work commitment and antisocial behavior, after controlling for covariates. Data were drawn from the Pathways to Desistance study (60, 72, 84 month follow-ups), a longitudinal project following 1,354 serious juvenile offenders from adolescence to young adulthood in Phoenix and Philadelphia. DV was antisocial behavior measured by total offending variety proportion scales ranging from 0 to 1, computed from the Self-Reported Offending scale comprised of 22 items. These proportions were converted into scores by multiplying them by 22. IV was the work commitment measured by the number of weeks worked across all community and under-the-table jobs. Covariates included demographic and family/juvenile characteristics, and substance use. Based on the theoretical framework, we hypothesized higher work commitment would predict decreases in antisocial behavior, after controlling for selected control variables. Spearman’s rho, which was robust to outliers, was used to measure the strength of the association between the two variables.

Results: The correlation results indicated that work commitment and antisocial behavior were significantly negative (T₁: ρ = -.06, p < .05; T₂: ρ = -.08, p < .05; T₃: ρ = -.09, p < .01). The random effects model results also confirmed that those who had more work commitment were significantly likely to decrease antisocial behavior across time points, after controlling for covariates (b = -.12, RSE = .02, p < .001). The study also found that individuals with more substance use (b = 1.45, RSE = .08, p < .001) and gang involvement (b = 1.03, RSE = .19, p < .001) were positively associated with criminal behavior. Those who were older earned a variety of scores of antisocial behavior that were .06 points lower than those who were younger (b = -.06, RSE = .03, p < .05).

Conclusion: Although the magnitude is small, the findings suggest it is evident employment works to reduce antisocial behavior consistent with the theory. Relevant stakeholders should commit to synthesizing their efforts to deliver more integrative policies and sophisticated ways to support and facilitate employment among formerly incarcerated young people to promote criminal desistance.