Regression analyses revealed that experiencing violence from an intimate partner accounted for the most variance in predicting who would inflict physical violence against an intimate partner. In addition, the infliction of physical violence against an intimate partner accounted for the most variance in predicting who would receive physical violence from an intimate partner. Other analyses indicated that parent-child violence experienced during childhood accounted for the most variance in explaining the risk of committing child abuse in both females and miles; the infliction of intimate partner violence added to the prediction of child abuse risk only in females. Furthermore, after the effects of violent experiences were removed, alcohol problems contributed significantly, although very modestly, to the prediction of who expressed intimate partner physical violence, a male's being physically injured by an intimate partner, and the risk of child physical abuse. Tables and 60 references (Author abstract modified)