Previous research suggests an association between partner violence and child behavior problems. However, methodological shortcomings have precluded the formation of directional conclusions. These limitations include failure to control for the effects of child physical abuse and general life stress, employment of nonrepresentative samples from battered women's shelters, and reliance on a single contemporaneous reporter, usually the mother, for information on both independent and dependent measures. This study used prospective, longitudinal data (N = 155) and multiple informants to examine the relation between maternal reports of partner violence in the home and teacher- and youth-report ratings of concurrent and prospective child behavior problems. Hierarchical multiple regression analyses were used to control for the effects of child physical abuse, child physical neglect, socioeconomic status, child cognitive ability, and life stress. The contribution of partner violence to child behavior problems was confirmed for boys' (n = 81) externalizing problems and girls' (n = 74) internalizing problems. Child developmental status at the time of exposure further influenced these relations. For boys, behavior problems in middle childhood were most strongly related to contemporaneous partner violence, whereas behavior problems among both boys and girls at age 16 were most strongly related to partner violence exposure during the preschool years.