Unveiling the relationship between land use types and the temporal signals of crime: An empirical decomposition approach

Whilst some land uses are highly criminogenic, others remain largely free of crime. This patterning is a reflection of the types and timing of daily activities that take place in a given land use and the opportunities that this presents for crime. While the criminology literature has developed a rigorous understanding of geographic component of crime, relatively less emphasis has been placed on the temporal dimension. Here, we address this through applying a technique to examine micro-temporal variations in crime at places. This technique adopts a factor approach to model hourly counts of crime across seven land use types (commercial, residential, parkland, agricultural, medical/hospital, industrial and education) to unveil the number and distribution of crime signals across a 24-hour period along with how these signals mix across each land use type. Results reveal clear and distinct differences between crime type and land use, highlighting the diurnal nature of crime patterns and speak to the literature on risky places and risky times. The utility of our approach lies in its capacity to delineate common temporal rhythms and how these rhythms are shared across different land use types.