Marilyn Chilvers

The New South Wales 'Compstat' Process: Its Impact on Crime

Abstract In January 1998 the New South Wales Police Service introduced a new crime control strategy, modelled on the New York “Compstat” process. The strategy involved a series of “Operation and Crime Review (OCR) Panels” in which senior police provided Local Area and Regional Commanders with information on crime trends and patterns in their local area and asked them to devise various tactics and strategies to reduce crime. At a later point in time the same commanders returned to the OCR panels and their performance in reducing crime was reviewed by senior management. This paper examines the impact of OCR Panels on the rate of recorded crime in four offence categories that fell after their introduction. The results suggest that OCR panels were temporarily effective in reducing three of these offences. Due to the suspension of OCR panels in the lead up to the Olympics it is not clear whether these effects could have been sustained.

Measuring Crime Dispersion

Existing crime statistics generally either compare crime rates within regions over time, in order to identify a worsening or improvement in crime risk, or rank regions by crime rate in a specific time period, in order to compare communities on the relative risk of victimisation. Such comparisons, however, provide no information about the relative contribution made by particular geographical areas to the observed aggregate change in the state crime rate or to the total level of crime. This paper describes the development of graphical methods and summary indices which measure the spatial dispersion of crime risk and the spatial dispersion of a change in crime risk. Measures of crime dispersion such as those described in this paper could prove useful in choosing among alternative crime control strategies and in measuring the net effects of crime control strategies or policies. Furthermore, the tendency towards a progressive concentration of crime in certain neighbourhoods may be studied.