UC Irvine

There are theoretical reasons to expect that crime opportunities created by the ambient population impact when and where crime occurs. Due to the temporally fluctuating nature of the ambient population, it is necessary to measure it with temporal precision over the hours of the day to assess its relationship with crime. This study uses geo-referenced location-based services data from Streetlight to create measures of the ambient population over 2-h time periods across both weekdays and weekends, and assess its relationship with five major types of crime. A proxy for guardianship potential is created based on the percent of the population that voted in the last election. The results indicate that the circulating ambient population is positively associated with all five types of crime measured over nearly all 2-h time periods. The relationship was particularly strong late at night after midnight, and was particularly strong over most time periods on weekends for robbery. A greater guardianship potential was associated with reduced robberies, burglaries and motor vehicle thefts over all time periods, and with assaults and larcenies in the daytime. The results highlight the importance of measuring the ambient population and guardianship potential, and the temporal variability in their relationships with crime.