Predictors of parent training efficacy for child externalizing behavior problems – a meta-analytic review


Patrick J. McGrath, Psychology Department, Dalhousie University, Halifax, NS, Canada, B3H 4J1; Tel: (902) 494-3581 Fax: (902) 494-6585; Email: Patrick.McGrath@DAL.CA


Background:  The differential effectiveness of parent training has led researchers to examine a variety of child, parent, and familial variables that may predict treatment response. Studies have identified a diverse set of child, parent psychological/behavioral and demographic variables that are associated with treatment outcome and dropout.

Method:  The parent training literature was examined to isolate child, parent, and family variables that predict response to parent training for child externalizing behavior problems. A literature review was conducted spanning articles published from 1980 to 2004 of indicated prevention (children with symptoms) and treatment (children with diagnosis) studies. Meta-analyses were conducted to determine standardized effect sizes associated with the identified predictors.

Results:  Many of the predictors of treatment response examined in this meta-analysis resulted in moderate standardized effect sizes when study results were subjected to meta-analytic procedures (i.e., low education/occupation, more severe child behavior problems pretreatment, maternal psychopathology). Only low family income resulted in a large standardized effect size. Predictors of drop-out resulted in standardized effect sizes in the small or insubstantial range.

Conclusions:  Response to parent training is often influenced by variables not directly involving the child, with socioeconomic status and maternal mental health being particularly salient factors.