Examining the Continuum of Injury Severity: Pooling the GES and FARS Datasets

详细信息    查看全文

• 作者：Shamsunnahar Yasmin ; Abdul R. Pinjari ; Naveen Eluru
• 关键词：Data Pooling ; Fatality risk ; Generalized ordered logit ; FARS ; GES
• 刊名：Procedia - Social and Behavioral Sciences
• 出版年：2 December, 2013
• 年：2013
• 卷：104
• 期：Complete
• 页码：1020-1029
• 全文大小：499 K

文摘

Given the import of the consequences of motor vehicle crashes, transportation safety researchers examined the influence of exogenous variables on vehicle occupant injury severity. Our study focuses on identifying the associated risk factors of driver fatalities while recognizing that fatality is not a single state but rather is made up of multiple discrete states from dying instantly to dying within the thirty days of crash by using the data from Fatality Analysis Reporting System (FARS). The research also simultaneously examines the whole spectrum of injury severity on an eleven point ordinal severity scale - no injury to fatality characterized as instant death by using a pooled dataset from FARS and General Estimates System (GES) dataset. The data pooling exercise is done to replace the 鈥渓ess informative鈥?fatal crashes in the GES databases with the more detailed fatal crashes from FARS database. The data for the current study is sourced from the FARS and GES databases for the year 2010. The generalized ordered logit approach is considered for the empirical analyses. The important control variables that affect the both the early fatality risk and the injury severity outcome of the drivers include: driving under the influence of alcohol, medium or higher speed limits and increase in vehicle age.