WITHDRAWN: Does Family Support Reduce Youth Crime? the Impact of the Pathways to Prevention Project on Behavior in the Primary School and Early Adolescent Years

Introduction: For children, little is known about the medium to long-term effects of family support that is routinely delivered in disadvantaged communities by non-government organizations. Pathways-to-Prevention (2002 to 2011) was a comprehensive early prevention initiative with family support at its core. It was a partnership between national community agency Mission Australia, Griffith University, and 7 primary schools in a socially disadvantaged area of Brisbane, Australia. We report the effects of Pathways on parent efficacy and classroom behavior (ages 5-11, Grades 1-7), and on involvement in the youth justice system (10-16 years).

Methods: Classroom behavior was measured annually by teachers for all children enrolled in preschool and Grades 1-7, using the Rowe Behavioral Rating Inventory. The Parent Efficacy and Empowerment Measure (PEEM) is a validated scale developed for this project that was administered each year to willing parents. Adjudicated youth justice outcomes for the 615 preschool children in 2002-3 were obtained from the Queensland Department of Justice. 123 of these children whose parent(s) received Pathways support between Grades 1 and 7 were matched one-on-one with 123 non-Pathways children on: classroom behavior at the end of preschool; age; gender; ethnicity; and child reported level of adversity. Changes in classroom behavior at primary school were compared across intervention and control groups using Bayesian multilevel modeling and vague priors, with covariates including baseline scores on the dependent variable.

Results: Pathways family support was strongly related to improved PEEM (E.S.=4.2; 95% CrI=1.2-7.5). Improvement in child behavior was negatively related to their initial PEEM, especially when moderate (E.S.=-1.9; 95% CrI=-5.2-0.6), but was positively related to moderate levels of Pathways family support (E.S.=3.4; 95% CrI=0.3-6.8). For the preschool cohort, 6% had a youth justice record by age 17, and preliminary analyses suggest improved classroom behavior related to a lower risk of youth justice involvement.

Conclusions: Routinely delivered family support in disadvantaged contexts can greatly improve parent efficacy, and at moderate levels can improve child classroom behavior. Ongoing analyses via Bayesian multilevel modeling are evaluating the validity of preliminary analyses that suggest these Pathways effects flowed through to lower rates of involvement in the youth justice system. These analyses are important for policy since family support is a ‘generic’ response to needs of deprived communities that may provide, at scale, a cost-effective alternative to criminal justice responses.