Respondents resided in a retirement community in New Jersey, a retirement community in Florida, and two age-integrated Florida communities. The primary data collection instrument was a structured, closed-ended interview conducted in respondents' homes. Frequency of alcohol drinking in the past year was measured on a nine-point scale from abstinence to daily drinking. Contact with the criminal justice system was an index based on whether or not the respondent reported a ticket for a moving traffic violation or having been arrested, been convicted, been placed on probation, or been incarcerated during the past year. Crime was measured by self-reports of assault, shoplifting, public intoxication, drunk driving, and illegal gambling since becoming 60 years old. The relationship between drinking and crime was strongest for problem drinking. Study conclusions are tentative and must be interpreted with an awareness that they are based on analyses of cross-sectional data, that there are few problem drinkers in the study, and that the number reporting contact with the legal system is small. 7 tables.