The groups were uncounseled batterers (63); counseled (court-mandated) batterers (58); and three groups of nonmaritally violent men who were happily married (58), unhappily married (51), or previously convicted of a violent crime (59). The study hypothesized that the uncounseled batterers would most favor the use of violence, with the exception of the maritally nonviolent, violent men. The subjects completed subsections of the Relationship Abuse Questionnaire which included 16 sociodemographic items and five items related to attitudes toward violence. A one-way ANOVA, chi-squares analyses, t tests, F tests of variance, and a Pearson product-moment correlation examined the data for significant differences. Results strongly support the hypotheses. Results suggest that the counseling of batterers reduced support for violence among maritally violent men. Results also indicate that attitude toward violence is not the sole source of male to female aggression. Arrest may contribute to attitudinal change. Study instrument, 12 tables, 17 references.