The researchers developed two sets of models, relating to whether a household was a victim for the specified property offense category and whether or not multiple victimization was experienced. Variables taken from the BCS included the dwelling, nature of the household, and community context. Small area studies data were used to provide a picture of the dwelling's neighborhood. The results showed that, in the simple victimization model, both micro and macro variables were significant across all models, both at differing levels of aggregation. Consequently, the authors will pursue a multi- level modeling approach to these issues in the future. In the multiple victimization models, however, no variables showed significance across all the models and very few were significant in any individual model. Further extensive work is required on the nature of multiple victimization, as it is believed to play a vital role in area crime prevention. 2 appendixes and 25 references