NCJ Number
190274
Journal
Reports of the National Research Institute of Police Science Volume: 40 Issue: 1 Dated: September 1999 Pages: 1-22
Date Published
September 1999
Length
22 pages
Annotation
In order to overcome the limitations of the qualitative approaches used in previous studies of crime prevention through environmental design in Japan, the current study used geographic information systems (GIS) to examine quantitatively the relationship between crime and spatial composition in Japan's urban areas.
Abstract
The crime data used in this study consisted of 176,706 official crime records for 23 "municipals" (districts within the larger metropolitan area) in Tokyo. For each offense type and modus operandi, data were aggregated for 3,080 "Cyome" districts (district smaller than municipal). The spatial data were the "NSD12,500" published by the Geographic Survey Institute of Japan and the Life-Mapple Digital Date by a private map provider. The average crime counts per Cyome district were 0.2 felonious offenses, 1.3 violent offenses, 46.5 larceny offenses, and 0.2 moral offenses. Correlation analysis showed that incident counts by modus operandi correlated with spatial compositions; whereas, counts by offense types did not. This implied that larceny criminals preferred particular spatial compositions. Larceny encompasses burglaries of all types, vehicle thefts, and other thefts of various types. Factor analysis found five latent factors for the spatial composite indicators: residential density, orderly streets, high-rises, middle-height buildings, and the size of street blocks. Residential burglary while residents were absent had strong positive correlation with residential density and negative correlation with orderly streets. This implied that spatial composition that contained many small-size dwellings blocks increased the risk for residential burglary; whereas, broad and straight streets decreased the risk. Office and shop burglaries correlated with high-rise and middle-height buildings. Vehicle thefts and other types of theft correlated with residential density, high-rises, and middle-height buildings. 10 tables and 9 references