Effortful control, negative emotionality, and juvenile recidivism: an empirical test of DeLisi and Vaughn’s temperament-based theory of antisocial behavior

Abstract

Recently, DeLisi and Vaughn articulated a temperament-based theory of antisocial behavior which they expressed as the first within criminology to use temperament explicitly, and as the exclusive explanatory construct of both antisocial behavior and negative interactions with the criminal justice system. We provide an initial empirical test of the theory’s two main constructs of effortful control (EC) and negative emotionality (NE) with respect to juvenile offending using a sample of 27,712 adjudicated youth. Cox regressions reveal youth with lower levels of EC and those with higher NE re-offend faster, controlling for demographics plus many prominent risk factors. Furthermore, the approximately 5% of youth at the extreme ends of both low EC and high NE evidence higher recidivism rates and faster time to failure. The findings show strong initial support for temperament theory, and we discuss relevancy for future research, prevention and intervention.

Keywords:

• Juvenile offenders
• temperament theory
• recidivism
• negative emotionality
• effortful control

Notes

1. We focus on youth who completed a community-based service for several reasons. First, the official measure of recidivism for FDJJ is an adjudication/adjudication withheld or adult conviction for an offense that occurred within 12 months of completing a FDJJ service. Additionally, it is arguable that examining re-offending from placement into service is an evaluation of that service and must account for dosage (intensity and duration of each service delivered) as well as fidelity measures to model protocols. Using completions aligns our research with FDJJ definitions and allows for a more salient examination of temperament theory as we measure risk after services have been completed.

2. Of those not receiving a Full C-PACT, 64% were diversion youth, 32% were probation supervision youth, while the remaining 4% split between day treatment, Redirections (i.e. community-based intensive family therapy programs), and aftercare youth. Youth who do not receive a Full C-PACT Assessment are all assessed as low or moderate risk on the C-PACT Pre-Screen. Accordingly, these youth do not require a Full C-PACT as per FDJJ policy (though obviously since our sample was 55% low and moderate risk, these youth may receive a Full C-PACT). All youth being placed into day treatment and Redirections services are supposed to receive a Full C-PACT (hence the low percentage of those youth that did not), regardless of risk to re-offend level.

3. As some youth were, or turned, 18 years of age or older during the follow-up period, we included both juvenile and adult official offending in our recidivism measure.

4. Temperament theory argues that conduct problems in school are indicative of lower EC. We included graduates and youth who had earned a GED diploma, as well as youth expelled or who have dropped out of school in the measure to avoid limiting the issues of generalizability that would come from only having measures of conduct for enrolled youth in our analysis. We believe that those youth who have successfully completed education and those who have dropped out or been expelled are most likely at the tail ends of a school conduct continuum.

5. We conducted all analyses using a slightly modified antisocial peer measure that dichotomized youth that had exclusively antisocial peers or gang member associates (with no pro-social friends). Results were substantively similar, with the modified measure (not shown) significantly predicting re-offending across estimated models.