Aims: This paper considers statistical relationships often observed between densities of bars and pubs and rates of violence as suggested by two general approaches: (i) social influence and (ii) social selection.
Methods: A stratified sample of 36 zip code areas in California was identified as having 'high', 'medium' and 'low' densities of bars and pubs. Aggregate US Census 2000 data were used to characterize population demographics of each zip code area. Telephone surveys were conducted assessing respondent demographics, drinking patterns, utilization of different places for drinking, self-report measures of hostility, norms for aggression and norms for alcohol-related aggression. Hierarchical linear models assessed the degree to which densities of bars and pubs were related to self-reports of hostility and norms for aggression, and if the individual measures of hostility and norms for aggression were related to choice of drinking venue.
Results: Respondents living in areas with greater densities of bars and pubs reported lower norms for aggression and greater norms for alcohol-related aggression. Greater peak drinking levels were related directly to greater levels of hostility and norms for both aggression and alcohol-related aggression. Self-reported hostility and norms for alcohol-related aggression were related directly to drinking at bars and pubs, parties and friends' homes. Aggressive norms were related to drinking at parties.
Conclusions: Whether bars serve to concentrate aggressive people into selected environments, whether these environments serve to increase levels of aggression, or whether both these processes reinforce each other mutually is not known. However, our findings do indicate relationships between certain exogenous measures, including alcohol outlet densities and social-psychological characteristics associated with violence. Many of these measures are also associated with the social contexts in which people drink.