NCJ Number
243285
Journal
Canadian Journal of Criminology and Criminal Justice Volume: 55 Issue: 2 Dated: April 2013 Pages: 263-277
Date Published
April 2013
Length
15 pages
Annotation
This article challenges the traditional use of residential population-based crime rates in criminological research and demonstrates that conventional crime rates only offer accurate information about risk in a limited number of situations.
Abstract
Crime rates provide a well-known and easy-to-understand standardized measure of the level of criminal activity of places. However, crime rates are based on several assumptions that are seldom discussed or even mentioned in most scientific analyses of crime. This article builds on a relatively simple observation: residential population is not always an appropriate measure of potential offenders and victims in an area. A typology of spatial relations between offender, victim, and criminal event is used to explore the circumstances under which residential population size is the appropriate denominator of crime rates. Spatial regression models suggest that residential population size is indicative of the number of both potential offenders and victims of domestic violence in a census tract, measures only potential targets of burglary, and was simply not representative of the population at risk of being involved in stranger assaults that took place in Montreal, Canada. This article challenges the traditional use of residential population-based crime rates in criminological research and demonstrates that conventional crime rates only offer accurate information about risk in a limited number of situations. (Published Abstract)