The CVS used a stratified cluster design with a sample size of 7,530 households, of which 6,315 completed personal interviews. The main survey objective was to estimate the volune and nature of crime in Queensland over the 12-month period prior to the CVS. A secondary objective was to obtain information on crime types, location, and time of occurrence and on fear of crime variables. An empirical model was applied to CVS data, and fear of crime was measured from answers to a question about feelings of safety when respondents walked alone in their area after dark. Results suggested that factors such as gender, age, poverty, educational level, labor force status, level of incivility in the area, perceived level of crime in the area, and neighborhood cohesion had an impact on fear of crime. The value of the empirical model in formulating crime prevention strategies is discussed. 22 references and 7 tables