This study hypothesized that power discrepancies in the marital relationship, where the husband is subordinate, serve as risk factors for husband-to-wife violence. The construct of marital power was assessed from 3 power domains operationalized by discrepancies in economic status, decision-making power, communication patterns, and communication skill. Three groups of married couples (N = 95) were compared: domestically violent (DV), maritally distressed/nonviolent (DNV), and maritally happy/nonviolent (HNV). DV couples were more likely than the 2 nonviolent groups to engage in husband demand/wife withdraw interactions. Within the DV group, husbands who had less power were more physically abusive toward their wives. Thus, violence may be compensatory behavior to make up for husbands' lack of power in other arenas of marriage. Difficulties in assessing marital power and future direction for the study of power and violence are discussed.