A random sample of 1,000 households in the city of Cincinnati were sent a questionnaire titled the "1992 Cincinnati Crime Survey" during the fall of 1992. Half of the surveys instructed that the survey should be completed by an adult female over the age of 21, and the other half were to be completed by an adult male 21 or older. In the event that an adult of the sex requested was not present at the address, then any adult residing at the address was asked to complete the survey. Age 21 was used, because it is the minimum legal age in Ohio to purchase a handgun. The respondents were given the following instructions: "Let's assume that you have a loaded gun in your hand or within reach. For each of the following situations, would you use the gun to shoot the criminal?" The scenarios involved an unarmed, nonthreatening burglar; an unarmed but threatening burglar; an armed and threatening burglar; and two robbers, one having a knife, who demand money on the street. Almost 70 percent of the respondents indicated a willingness to shoot an offender in at least one of the scenarios presented. The results do not suggest, however, that citizens favor shooting offenders under virtually any circumstance. The criterion that guided respondents' willingness to shoot an offender was the physical threat posed by the offender. In contrast to previous research, firearm owners and non-owners did not differ significantly in their willingness to shoot offenders or in their support for using violence for defensive purposes. The willingness to shoot an offender, however, was positively related to childhood socialization into a gun culture and to holding violent attitudes. It was also positively related to the belief that crime was increasing, perceived risk of victimization, being nonwhite, and being male. 3 tables and 57 references