Community structural predictors of spatially aggregated motor vehicle theft rates: Do they replicate?

Community-level motor vehicle theft (MVT) is not spatially random but is influenced by the structural composition of the community. Work to date did not provide a clear picture of the structural correlates of community-level MVT rates for two reasons. Cross-sectional studies had been limited to a single point in time (one wave design). In addition, studies had not adequately controlled for MVT rates in adjoining communities (spatially autocorrelated rates). The current study addressed these limitations. Drawing on structural correlates highlighted by factorial ecology and past work on motor vehicle theft, it anticipated cross-sectional connections between status, stability, age composition, and racial heterogeneity. It sought to learn if these connections persisted at two points in time spanning a decade. Census block group data from a midwestern city were merged with geocoded vehicle theft data, and a comprehensive spatial lag variable was constructed (Land & Deane, 1992). At both points in time, communities with higher MVT rates had lower socioeconomic status, and were surrounded by other communities with higher MVT rates. Community processes driving the connection between status and vehicle theft were suggested. The strong spatial dependency of MVT rates suggests attributes, events, or longer-term trends located in a section of a city may be affecting the communities located there. Issues for prevention were addressed.