School-Based Mentoring for Aggressive Children: Effects on Children's Aggression at 6-Month Follow-up

“Early starters” are children who begin displaying conduct problems in early to middle childhood and continue on this trajectory into adulthood. Child aggression and antisocial behavior are linked with a number of negative outcomes, both in the short run and over time (Caspi, 2000). Although psychosocial interventions designed specifically for these “early starters” are generally supported by the empirical literature, a substantial proportion of children do not respond to treatment or gains are reduced at follow-up assessment.

Youth mentoring is often touted as a useful prevention tool (Dortch, 2000). Lunch Buddy (LB) mentoring is a simple yet innovative school-based intervention strategy that involves regular lunchtime visits from a college student mentor (Hughes et al., 2001; Hughes et al., 2005). Created originally as a control condition, LB mentoring was found to have unexpected benefits for aggressive, school age children in an earlier RCT. In this study, mentors were trained to enhance children’s interactions with peers, promote prosocial behaviors, and alter peer-mediated contingencies that maintain aggressive behavior. The goal of the current investigation was to examine the effects of LB mentoring on aggression outcomes at 6-months post-intervention. 

Methods

Participants

Participants were 55 2^nd-4^th-grade children (28 intervention; 27 control) from 8 elementary schools. Teachers nominated children who met a behavioral description of an aggressive child. Participants scored > 60 T on the aggressive subscale of the TRF or CBCL (Achenbach, 1991). Eligible children (n = 55; 32 boys) were randomized to conditions.

Measures

Aggression was assessed through self- and teacher-report (Dodge & Coie, 1987; Little et al. 2003). Teacher report of aggression was also assessed via the Achenbach aggression subscale (Achenbach 2001). Bullying behavior was assessed through a peer nomination inventory.

Intervention Conditions

Lunch Buddy is a stand-alone mentoring intervention that spans two academic semesters. Mentors promote positive peer interactions, discourage deviant behavior, redirect negative peer interactions, model and reinforce prosocial behavior, and challenge children’s biased perceptions toward peers. Children in the Waitlist Control condition have available to them usual and established school-based services (e.g. school counselor).

Data Analytic Plan

A repeated measures analysis of covariance will be used to examine the effect of treatment condition on measures of aggressive behavior. Analyses will control for aggression score at screening and gender. Preliminary analyses revealed a trend suggesting that the benefits of LB mentoring for aggressive children were maintained at 6-month follow-up for self-reported overt and proactive aggression.