This randomized experimental prevention study (a) evaluated the effectiveness of a parent-training program in a sample of 238 divorcing mothers with sons in Grades 1-3 and (b) provided an experimental test of coercion theory. The intervention produced reductions in observed coercive parenting, prevented decay in positive parenting, and generally improved effective parenting practices in comparisons of mothers in experimental and control groups. Moreover, coercion theory was supported. Improved parenting practices correlated significantly with improvements in teacher-reported school adjustment, child-reported maladjustment, and mother-reported maladjustment. The intervention indirectly benefitted child outcomes through improved parenting practices for a model based on child report and, to a lesser extent, on teacher report. The intervention did not produce direct effects on child outcomes.