RESUMO
The past decade has seen a rapid growth in the use of a spatial perspective in studies of crime. In part this growth has been driven by the availability of georeferenced data, and the tools to analyze and visualize them: geographic information systems (GIS), spatial analysis, and spatial statistics. In this paper we use exploratory spatial data analysis (ESDA) tools and Bayesian models to help better understand the spatial patterning and predictors of property crime in Seattle, Washington for 1998-2000, including a focus on built environment variables. We present results for aggregate property crime data as well as models for specific property crime types: residential burglary, nonresidential burglary, theft, auto theft, and arson. ESDA confirms the presence of spatial clustering of property crime and we seek to explain these patterns using spatial Poisson models implemented in WinBUGS. Our results indicate that built environment variables were significant predictors of property crime, especially the presence of a highway on auto theft and burglary.
RESUMO
The purpose of this study was to examine whether incorporating information regarding neighborhoods which were more distal to the child's neighborhood added any explanatory power to models of child cognitive competence. Participants included a sample of young African-American children living in an urban setting in the northeast United States. Spatial geographic methods were used to estimate the concentration of economic disadvantage, population instability, and crime in the neighborhoods surrounding the child's residence, and multilevel modeling methods were used to estimate the contribution of these factors to between-neighborhood variance in child cognitive scores. Results indicated that the conditions of distal neighborhoods were related to cognitive scores among the preschooler-age children in this sample.