The groups included 47 children who reported observing marital violence that included the threat or use of knives or guns (observed), 57 children who did not report witnessing marital violence involving knives or guns but whose mothers reported that violence involving knives or guns had occurred in a recent marital dispute (occurred), and 51 children who did not report witnessing marital violence involving knives or guns and whose mothers also reported no violence involving knives or guns (neither). Children in the observed group displayed higher levels of behavior problems than children in the neither group, but they did not differ from children in the occurred group. Children in the occurred group also displayed higher levels of behavior problems than children in the neither group. Results suggested that the occurrence of interparent weapon violence, rather than children's observation of it, increased the risk of behavior problems and that the occurrence of weapon violence related to behavior problems after accounting for other aspects of interparent violence. 49 references and 5 tables