| Criminologists, planners, and architects search for ways to predict criminals' preferences for committing an offense in one location over another. Criminologists associate crime with socio-demographic factors such as income, racial composition, youth concentration and level of education. Architects and planners on the other hand, relate crime to environmental design factors such as lighting, security hardware, or orientation of entrances, just to name a few. Recently, some work using space syntax has demonstrated statistical relationships between properties of spatial layouts and the occurrence of certain types of crimes. Space syntax, a group of theories developed at the University College London, suggests a fundamental relationship between the configuration of urban or built space and the way that space functions.;In this study, three Space Syntax techniques are applied to quantitatively describe street accessibility and visibility, and to examine them in relationship to the occurrence of crime, controlling for such factors as neighborhood socio-demographic status. The study explores geographical patterns of four types of offense behavior: breaking and entering, larceny, vehicle theft and robbery. Crime data, at an address level, is based on a 12 month period for the city of Ypsilanti Michigan (USA). After mapping crime locations using GIS, axial maps (measuring accessibility) were prepared using Spatialist (developed by Peponis et al. at Georgia Tech), and visibility maps were prepared using Syntax 2D (developed by Turner et al. at University of Michigan) and Depthmap (developed by Turner at University College London).;Results of negative binomial distribution analysis demonstrate strong associations between aspects of street network accessibility (connectivity and integration), street network surveillability (visibility), and socio-demographic characteristics (homeownership and racial composition) with different types of crimes. The findings of the study suggest that Space Syntax can be a valuable tool in predicting targets for crime in residential neighborhoods. |
