Methods for Evaluating the Association Between Alcohol Outlet Density and Violent Crime

Area of Science:

• Environmental Criminology
• Urban Planning
• Public Health Research

Background:

• Understanding the relationship between alcohol outlet density and violent crime is crucial for public health and urban planning.
• Previous research has yielded conflicting findings, potentially due to methodological variations in measuring these factors.

Purpose of the Study:

• To compare different methods for calculating alcohol outlet density and violent crime.
• To determine which measurement methods provide the most accurate and reliable associations between outlets and crime in an urban setting.

Main Methods:

• Utilized violent crime data (homicides, assaults, rapes, robberies) and alcohol outlet data from Baltimore, Maryland (2016).
• Calculated density and crime rates at the census block (CB) level and aggregated to the census tract level.
• Employed negative binomial and linear regression models, guided by choropleth maps, R-squared, AIC, and RMSE to assess model fit and error.

Main Results:

• Count-based methods for alcohol outlet density and violent crime often failed to detect associations or indicated an inverse relationship.
• Proximity, mean distance, and spatial access methods consistently identified a significant association between alcohol outlets and violent crime.

Conclusions:

• Proximity, mean distance, and spatial access methods demonstrated superior model fit and lower error rates for analyzing outlet-crime relationships in urban areas.
• Measurement methodology significantly impacts study inferences; spatial access methods may offer advantages over proximity and mean distance.
• Discrepancies in the literature may stem from errors in how alcohol outlet density is measured.