Interactions between the built and socio-economic environment and driver demographics: spatial econometric models of car crashes in the Columbus Metropolitan Area

• Published in: International Journal of Urban Sciences Vol. 22; no. 1; pp. 17 - 37
• Main Authors: Lee, Dongkwan, Guldmann, Jean-Michel, von Rabenau, Burkhard
• Format: Journal Article
• Language: English
• Published: Abingdon Routledge 02.01.2018; Taylor & Francis Ltd; 도시과학연구원
• Subjects: Age; Age groups; Aggregate data; at-fault drivers; built environment; crashes; Destruction of property; Econometrics; gender; Injuries; Public lands; Public safety; spatial regression; Traffic; Transportation policy; 학제간연구; United States > US; Ohio
• ISSN: 1226-5934; 2161-6779
• DOI: 10.1080/12265934.2017.1369452

This research analyzes car crashes resulting from the interactions between (1) the characteristics of the built and socio-economic environment where the crashes take place and (2) the gender and age of the driver at fault. Crashes are classified in terms of seriousness (fatalities/injuries, property damages only) and driver demographics. Data are drawn for the Central Ohio Region over 2006-2011 from the multiple files of the crash database of the Ohio Department of Public Safety. These data are aggregated over Traffic Analysis Zones (TAZ). Additional data include socio-economic, land-use, public transit, road network, and other locational/physical factors, also specified at the TAZ level. Regression analysis is used to explain the numbers of crashes in each of 12 groups. Three age groups are considered: young (15-24), adult (25-64), and older (65+). Spatial autocorrelation effects are tested and corrected by estimating spatial econometric models. The implications of the results for transportation safety policy are discussed.