OBJECTIVES: While epidemiological evidence suggests homes with guns are more likely to be the site of a suicide or homicide than homes without guns, the public's perception of these risks remains unknown. This study assesses the prevalence of the belief that homes with guns are safer than homes without guns, and factors associated with this belief. METHODS: Telephone interviews were conducted with a random sample of 4138 registered voters in urban areas in the US. Multinomial logistic regression was used to assess correlates of beliefs about the safety of keeping a gun in the home. RESULTS: Twenty nine per cent of respondents believed keeping a gun in the home makes the home more safe, 40% said less safe, 23% said it depends, and 9% were unsure. The belief that a home is more safe with a gun was associated with being male, young, completing 12 years or fewer of education, having no children living at home, Republican party affiliation, and low levels of trust in the police for protection. Prior exposure to violence and fear of victimization were not associated with the outcome. CONCLUSIONS: Findings may increase understanding about the public's perception of the risk in keeping guns in the home and assist educational efforts to decrease the risk of these injuries.