Surveying Residential Burglaries: A Case Study of Local-Level Crime Measurement

Abstract

Our goal is to document the general issues that arise in estimating and comparing local crime rates over time and for different cities. Data are drawn from police-based residential burglary counts and the National Crime Victimization Survey to obtain interval estimates of residential burglary incidence at the local level. Ambiguity exists when making point estimates of crime that are based entirely on Uniform Crime Reports data. Sources of this ambiguity include (a) the “Hierarchy Rule,” (b) population estimates used in estimating the burglary rate, and (c) variation in the probability that a residential burglary victim reports the incident to the police. Analyses document uncertainty of burglary rates when comparing across cities and over time. We think it will be useful to begin reporting crime patterns in terms of a range of uncertainty that expresses both what is known and unknown about the numbers that are used to measure those patterns.