Journal of Quantitative Criminology, Vol. 14, No. 3, 1998 The Distribution of Household Property Crimes Denise R. Osborn1,3 and Andromachi Tseloni 2 Previous studies have established that repeat victimizations occur more fre- quently than would be expected if households within a particular area were vic- timized randomly. This implies that characteristics of the household affect the victimization rate. Even controlling for these characteristics, we find that a Pois- son model does not capture the distribution of victimizations because repeat victimizations are more concentrated than it would indicate. This leads us to adopt the negative binomial generalization of the Poisson model. Our analysis uses sociodemographic attributes of the household and community-level charac- teristics to predict victimizations, with the victimization data being the observed number of property crime victimizations from the 1992 British Crime Survey. The negative binomial generalization is found to be highly statistically significant and the crime concentration it implies becomes much more marked as the predicted number of victimizations increases. 1. INTRODUCTION Research on repeat victimization has two important and related strands. First, victims in one time period are more likely to be victimized in a subsequent period than are prior nonvictims. Second, research examining cross-section data has shown that victimization tends to be concentrated on a relatively small group of individuals. Relevant papers considering these issues include those by Hindelang et al. (1978), Fienberg (1980), Reiss (1980), Gottfredson (1984), Ellingworth et al. (1995), and Trickett et al. (1995b). Two recent studies in this journal, namely, Lauritsen and Davis- Quinet (1995) and Osborn et al. (1996), have underlined the importance of repeat crime victimization and shown that empirical models can use infor- mation on prior victimization to deliver predictions of subsequent risk for 1School of Economic Studies, University of Manchester, Manchester M139PL, U.K. 2Department of Criminology and Criminal Justice, University of Maryland, 2220 LeFrak Hall, College Park, Maryland 20742. 3To whom correspondence should be addressed. KEY WORDS: property crime; repeat victimization; heterogeneity; state depen- dence; Poisson model; negative binomial model. 307 0748-4518/98/0900-0307$15.00/0 © 1998 Plenum Publishing Corporation