The objective of this article is to report the results of an ecological study into the geography of rape in Stockholm, Sweden, using small area data. In order to test the importance of factors indicating opportunity, accessibility and anonymity to the understanding of the geography of rape, a two-stage modelling approach is implemented. First, the overall risk factors associated with the occurrence of rape are identified using a standard Poisson regression, then a local analysis using profile regression is performed. Findings from the whole-map analysis show that accessibility,

opportunity and anonymity are all, to different degrees, important in explaining the overall

geography of rape - examples of these risk factors are the presence of subway stations or whether

a

basområde is close to the city centre. The local analysis reveals two groupings of high risk of

rape areas associated with a variety of risk factors:

city centre areas with a concentration of alcohol

outlets, high residential population turnover and high counts of robbery; and

poor suburban areas

with schools and large female residential populations where subway stations are located and where

people express a high fear of crime. The article concludes by reflecting upon the importance of these results for future research as well as indicating the implications of these results for policy