View Record Details -- Report
An Exploratory Spatial Data Approach to Identify the Context of Unemployment-Crime Linkages, Final Report
United States Department of Justice, National Institute of Justice
Place of Production:
Previous research on the link between unemployment and crime has focused mainly on global explanations without a corresponding analysis of the local patterns that may influence the U-C connection. The current research explores how changes in local conditions in terms of unemployment can lead to changes in crime rates, specifically property crime rates. The methodology involved the integration of Exploratory Spatial Data Analysis (ESDA) with multilevel modeling techniques in order to closely examine the role of resource deprivation in moderating the motivational and opportunity links between unemployment and crime. Data included unemployment rates from the U.S. Department of Labor for the years 1995 through 2000 and information on crime rates from the Geospatial and Statistical Data Center of the University of Virginia. Census data were also used to examine median family income, percentage of families living in poverty, and percentage of African-American residents. Results of the statistical data analysis indicated weak relationships between the average values of unemployment and crime rates during the study period. The assessment of how local patterns influenced the U-C link revealed relationships between the U-C coefficient for larceny, motor vehicle theft, and index crimes but failed to show a relationship between local patterns of burglary and robbery and other crimes. Negative relationships were discovered between the unemployment coefficient and resource deprivation for burglary, larceny, motor vehicle theft, and index crimes; these negative relationships were especially significant in the affluent areas around Washington, DC. Future research should focus on the U-C relationships across various community contexts for different types of violent and nonviolent crimes.
This publication is related to the following dataset(s):