The study used a 9-percent random sample of 1,112 pretrial interview records on males arrested for domestic violence in 1997 in Kentucky. The study examined perpetrators in three geographical areas so as to determine regional differences. The arrest data were collected during interviews conducted by Pretrial Services during the 1997 fiscal year. The interviews were completed in the jail within 12 hours of arrest. Beale Urban Influence Codes were used to classify arrestees into an urban-rural continuum. The counties in which the individuals were arrested were classified into an urban-rural continuum. Of the individuals, 67 percent were classified as urban, 11 percent were classified as urban influenced, and 22 percent were classified as rural. Data analysis focused on geographic location as the independent variable and used chi-squares for dichotomous variables and ANOVA's for continuous dependent variables. Findings showed that community context was critical in understanding domestic violence. Generally, rural males apparently presented more significant problems, given their lower employment rates, lower educational attainments, greater use of psychoactive medications, and higher arrest rates. Alcohol use was significantly prevalent across all three groups, but combined alcohol and nerve-pill drug use was more prevalent among rural domestic violence arrestees. A low percentage of males, regardless of area, had received anger management training or drug/alcohol treatment. The data suggest the need for more intervention services for domestic violence perpetrators, particularly in rural areas. 4 tables and 45 references