The influence of urban built environment on residential burglary in China: Testing the encounter and enclosure hypotheses

Abstract

It is widely recognized that crimes are clustered in space. Features of the built environment are among various factors associated with crime clustering patterns. There are two competing perspectives on this association: the “encounter” and “enclosure” hypotheses. This study examines the effect of the built environment on the spatial distribution of residential burglary in Wuhan, China, aiming to test these two competing perspectives. First, we review existing studies analyzing the influence of the built environment on residential burglary, especially those based on the two hypotheses. We then conduct an exploratory data analysis to observe the spatial and temporal distribution patterns of residential burglaries. Next, we use a multivariate spatial regression model to evaluate the influence of various built environment characteristics on residential burglaries, after controlling for socio-demographic variables, guardianship, and the time-of-day effect. The results suggest a double peak in the hourly distribution of burglaries: 6:00 a.m.–8:59 a.m. and 5:00 p.m.–7:59 p.m. Burglaries are positively and significantly autocorrelated in space and mainly concentrated in the north-west and central areas of Wuhan. Two transport facilities features (intersection density and the presence of subway stations), three building features (multiple stories, high housing prices, and older houses), and land-use mix are each positively correlated with burglary. However, the burglary count is negatively correlated with the ratio of the residential land. Overall, neighborhoods with high permeability are vulnerable to residential burglaries. Thus, the principle of access control promoted by the enclosure hypothesis appears effective in preventing residential crime. The results’ implications for crime prevention and urban design practice are also discussed.