Exploratory Spatial Analysis of Relationships between Crimes and Socioeconomic Factors in St. Louis, Missouri

St. Louis, Missouri as the major metropolitan area in the Midwest, has faced persistent challenges related to violent crimes and property crimes. Existing research suggests that crime and socioeconomic status influence each other; however, empirical studies specifically focused on the St. Louis area remain limited. This study utilized data on crime patterns across the St. Louis area to explore the complex interrelationships among five types of crime—assault, auto theft, burglary, homicide, and robbery—and various socioeconomic characteristics, including housing conditions, poverty levels, transportation access, educational attainment, and employment rates. An exploratory regression analysis was conducted to identify the independent variables that would construct the best-fitting model. Both Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) results indicate that, whether considered collectively or individually, the five crime types do not fundamentally alter the overall relationship between crime and the selected socioeconomic factors. However, the socioeconomic factors affecting crime statistics varied by crime type, with each type being associated with different combinations of independent variables. Additionally, Multiscale Geographically Weighted Regression (MGWR) results reveal that model performance varies spatially across all crime types, with local R² values being higher on the east side of St. Louis city and gradually decreasing toward the west side of St. Louis county. Moreover, in zip codes to the north of downtown St. Louis, which are perceived as less safe than average, socioeconomic indicators are relatively poor. This suggests that policies should be formulated based on the spatial distribution of crime.