Men's Domestic Violence and Other Forms of Deviant Behavior, Executive Summary

The initial survey sample consisted of 1,725 youths (918 males) interviewed in 1976, with follow-up interviews conducted in 1977, 1978, 1979, 1980, 1983, and 1986. Data for the current research were obtained from 175 male participants who reported being married or cohabiting with a female partner at Wave VI (1983). Men's violence toward their female partners was measured at Wave VI with eight items from the physical violence subscale of the Conflict Tactics Scales. Men's general deviance was measured at Wave VI and at earlier waves with items that described illegal or socially proscribed behavior. Deviance was operationalized in several ways in the research, so that the persistence and frequency/seriousness of various forms of deviance (violent and nonviolent deviant acts) could be considered. To assess specific pathways by which early deviance might be linked to domestic violence, measures of marital dissatisfaction, peer deviant behaviors, and peer approval of deviance were created. Of the 175 married or cohabiting men in the sample, 38 percent reported having engaged in domestic violence in the year prior to the Wave VI assessment. As expected, both the persistence of deviant activity and the frequency/seriousness of deviant activity predicted later domestic violence; however, in logistic regression analyses, neither of these two measures of deviance contributed uniquely in the prediction of domestic violence after accounting for the other. Thus, the information contained in these two measures of deviance apparently was redundant regarding the prediction of domestic violence. As expected, the persistence and the frequency/seriousness of violent and nonviolent deviance predicted later domestic violence. The persistence of violence emerged as an important aspect of youth deviance in the prediction of domestic violence in the sample. In a multivariate model, the persistence of violence during adolescence/young adulthood was linked directly to domestic violence, and the association between these two variables was particularly mediated by relationship dissatisfaction. Consistent with prior research, findings suggest that youth violence increases the likelihood of affiliation with deviant peers as well as peers who approve of deviance; however, the findings suggest that deviant peer affiliation and peer approval of deviance do not mediate the link between earlier violence and domestic violence. These findings suggest that programs designed to prevent or reduce adolescent antisocial behavior, especially violent behavior, may help prevent the processes by which domestic violence emerges.