An Exploration of Temporally Differentiated Associations Between Alcohol Outlets, Social Disorganization, and Assaults: A Multilevel Approach

A substantial body of research consistently shows that the concentration of alcohol outlets is positively associated with violent crimes, such as assaults. While recent studies have incorporated both spatial and temporal dimensions of this relationship, particularly within-day variations, how this temporally differentiated association of alcohol outlets with violent crimes varies depending on the broader context, such as social disorganization, remains underexplored. This study addresses this research gap by investigating whether the relationship between alcohol outlets and assaults in New York City from 2017 to 2019 differs by time of the day, while accounting for neighborhood structures in terms of their independent and conditional associations with assaults. Data on assaults, alcohol outlets, and neighborhood characteristics were acquired from various open sources, resulting in a dataset with 37,259 census blocks embedded within 2,090 census tracts. Using the 2015 to 2019 American Time Use Survey report, this study divided the day into four periods-morning, daytime, evening, and night-and applied a series of multilevel negative binomial regressions to capture the geographically hierarchical structure between census blocks and census tracts. Findings reveal that on-premises alcohol outlets, grocery/convenience stores, liquor/wine stores, and social disorganization indicators are positively associated with assaults throughout the day. The only exception is racial heterogeneity at night. Furthermore, cross-level interaction analyses indicate that the significance of the interaction terms differs by the time of day, the type of alcohol outlets, and social disorganization indicators. Furthermore, the relationship between alcohol outlets and assaults is weaker in more disorganized neighborhoods. These results highlight the importance of considering crime opportunity, social disorganization, and temporal dynamics in understanding the relationship between alcohol outlets and crime. Limitations and implications for future research and policy are also discussed.