1372488 - Investing the Effect of County-Level, Wealth-Related Indicator Variables on Crash Severity in Pennsylvania

Current crash injury severity estimation methods consider only roadway specific variables and exclude measures of regional economic activity or socioeconomic related data. Recent work has started investigating the impact of regional level socioeconomic measures on estimated crash injury severity, but the topic requires further investigation. This paper aims to further understanding of the relationships between crash injury severity and regional level (including traffic and roadway, demographic, and socioeconomic) variables to support continued investigation into the intersection of socioeconomics and traffic safety. Doing so can lead to a more equitable distribution and allocation of transportation safety funding. Data was collected from the Pennsylvania Department of Transportation, and the United States Census American Community Survey. Two binary logit models are estimated. The first estimates the likelihood of a crash resulting in a fatal or major injury outcome, and the second estimates the likelihood of a crash resulting in a fatal, major, or minor injury outcome. A cumulative logit model is estimated to confirm the findings from the binary logit model estimations. Overall, the results of this paper reveal that crashes occurring in counties exhibiting high levels of socioeconomic distress are more likely to result in more severe injury severity outcomes compared to crashes that occur elsewhere.