In recent years, it has increasingly been recognized that due to the uncertain geographic context problem caused by daily human mobility, the residential population is too static to serve as a valid measure of the population at risk for criminal victimization. Various alternative measures have been suggested instead. Guided by the routine activity approach, this study furthers the concept of crime risk population and its measurement across space and time. Using exceptionally comprehensive data sets on population mobility and on theft from the person in a large city in China, we select the best indicator of the risk population from the following four candidates: residential population, subway ridership, taxi ridership, and mobile phone users. Controlling for the potentially confounding effects of offender and guardian presence, we show that on both weekdays and weekends, the best indicators of risk population vary over the course of the day. In the morning, residential population outperforms other measures. In the afternoon and evening, taxi ridership and phone users are better indicators. Although the mobile phone user base forms an arguably more representative measure of ambient population, during some periods taxi ridership is superior because it provides a better indicator of outdoor (as opposed to indoor) activities. In terms of practical applications to security policy and law enforcement, these findings can help identify crime hot spots by calculating accurate crime risks.

Negative binomial (NB) regression model has been used to analyze crime in previous studies. The disadvantage of the NB model is that it cannot deal with spatial effects. Therefore, spatial regression models, such as the geographically weighted Poisson regression (GWPR) model, were introduced to address spatial heterogeneity in crime analysis. However, GWPR could not account for overdispersion, which is commonly observed in crime data. The geographically weighted negative binomial model (GWNBR) was adopted to address spatial heterogeneity and overdispersion simultaneously in crime analysis, based on a 3-year data set collected from ZG city, China, in this study. The count of residential burglaries was used as the dependent variable to calibrate the above models, and the results revealed that the GWPR and GWNBR models performed better than NB for reducing spatial dependency in the model residuals. GWNBR outperformed GWPR for incorporating overdispersion. Therefore, GWNBR was proven to be a promising tool for crime modeling.

Research on journey-to-crime distance has revealed the importance of both the characteristics of the offender as well as those of target communities. However, the effect of the home community has so far been ignored. Besides, almost all journey-to-crime studies were done in Western societies, and little is known about how the distinct features of communities in major Chinese cities shape residential burglars’ travel patterns. To fill this gap, we apply a cross-classified multilevel regression model on data of 3763 burglary trips in ZG City, one of the bustling metropolises in China. This allows us to gain insight into how residential burglars’ journey-to-crime distances are shaped by their individual-level characteristics as well as those of their home and target communities. Results show that the characteristics of the home community have larger effects than those of target communities, while individual-level features are most influential. Older burglars travel over longer distances to commit their burglaries than the younger ones. Offenders who commit their burglaries in groups tend to travel further than solo offenders. Burglars who live in communities with a higher average rent, a denser road network and a higher percentage of local residents commit their burglaries at shorter distances. Communities with a denser road network attract burglars from a longer distance, whereas those with a higher percentage of local residents attract them from shorter by.

The relationship between burglary and socio-demographic factors has long been a hot topic in crime research. Spatial dependence and spatial heterogeneity are two issues to be addressed in modeling geographic data. When these two issues arise at the same time, it is difficult to model them simultaneously. A cross-comparison of three models is presented in this study to identify which spatial effect should be addressed first in crime analysis. The negative binominal model (NB), Bayesian hierarchical model (BHM) and the geographically weighted Poisson regression model (GWPR) were implemented based on a three-year residential burglary data set from ZG, China. The modeling result shows that both BHM and GWPR outperform NB as they capture either of the spatial effects. Compared to the NB model, the mean absolute deviation (MAD) of BHM and GWPR was decreased by 83.71% and 49.39%, the mean squared error (MSE) of BHM and GWPR was decreased by 97.88% and 77.15%, and the ${\mathrm{R}}_{\mathrm{d}}^2$ of BHM and GWPR was improved by 26.7% and 19.1%, respectively. In comparison with BHM and GWPR, BHM fits the data better with lower MAD, MSE and higher ${\mathrm{R}}_{\mathrm{d}}^2$ . The empirical analysis indicates that the percentage of renter population, percentage of people from other provinces, bus line density, and bus stop density have a significantly positive impact on the number of residential burglaries. The percentage of residents with a bachelor degree or higher, on the other hand, is negatively associated with the number of residential burglaries.

With the rapid development of China’s economy, the demand for labor in the coastal cities continues to grow. Due to restrictions imposed by China’s household registration system, a large number of floating populations have subsequently appeared. The relationship between floating populations and crime, however, is not well understood. This paper investigates the impact of a floating population on residential burglary on a fine spatial scale. The floating population was divided into the floating population from other provinces (FPFOP) and the floating population from the same province as ZG city (FPFSP), because of the high heterogeneity. Univariate spatial patterns in residential burglary and the floating population in ZG were explored using Moran’s I and LISA (local indicators of spatial association) models. Furthermore, a geographically weighted Poisson regression model, which addressed the spatial effects in the data, was employed to explore the relationship between the floating population and residential burglary. The results revealed that the impact of the floating population on residential burglary is complex. The floating population from the same province did not have a significant impact on residential burglary in most parts of the city, while the floating population from other provinces had a significantly positive impact on residential burglary in most of the study areas and the magnitude of this impact varied across the study area.