NCJ Number
84533
Date Published
1981
Length
17 pages
Annotation
This study of the geographic distribution of crime shows that the census-tract analysis of crime depicts it in a highly generalized fashion, thus concealing gross disparities in both the patterns and types of offenses at the micro level (grid level).
Abstract
The primary research goal was to measure the effects of a change in the size of the units of analysis of the geographic distribution of crime within a given study region (New Castle County, Del.) on research outcomes. A second objective was to group crime types according to their patterns of spatial covariation and examine the potential value of these groupings for criminologic research. A final objective was to describe the distributional patterns of crime elicited by the groupings of offense types at different spatial scales. The two geographical units of analysis were the census tract and the administrative units of the Division of Highways for the State, which are subdivisions of census tracts and are the smallest geographic units for which crime data are maintained. Nineteen crime indicators representing 45,740 of the 168,672 offenses reported during the 18-month period studied were selected. Generalizations from the study are that (1) crime components associated with the dimensions delineated at the grid level are found throughout the study area, including the urban, suburban, and rural areas; (2) certain offense types tend to be concentrated in urban centers (retail economic crimes, drug offenses, and violent crimes); (3) all offense categories tend to show high spatial association with the distribution of major highways; and (4) the distributional pattern depicted by the mapped factor scores at the grid level pinpoints the occurrence of specific classes of offenses. Implications of the findings for theory construction and the development of public policy are discussed. Tabular data and 13 footnotes are provided.