Injury severity analysis of commercially-licensed drivers in single-vehicle crashes: Accounting for unobserved heterogeneity and age group differences

Mohamed Osman, Sabyasachee Mishra, Rajesh Paleti

Research output: Contribution to journalArticlepeer-review

46 Scopus citations

Abstract

This study analyzes the injury severity of commercially-licensed drivers involved in single-vehicle crashes. Considering the discrete ordinal nature of injury severity data, the ordered response modeling framework was adopted. The moderating effect of driver's age on all other factors was examined by segmenting the parameters by driver's age group. Additional effects of the different drivers’ age groups are taken into consideration through interaction terms. Unobserved heterogeneity of the different covariates was investigated using the Mixed Generalized Ordered Response Probit (MGORP) model. The empirical analysis was conducted using four years of the Highway Safety Information System (HSIS) data that included 6247 commercially-licensed drivers involved in single-vehicle crashes in the state of Minnesota. The MGORP model elasticity effects indicate that key factors that increase the likelihood of severe crashes for commercially-licensed drivers across all age groups include: lack of seatbelt usage, collision with a fixed object, speeding, vehicle age of 11 years or more, wind, night time, weekday, and female drivers. Also, the effects of several covariates were found to vary across different age groups.

Original languageEnglish (US)
Pages (from-to)289-300
Number of pages12
JournalAccident Analysis and Prevention
Volume118
DOIs
StatePublished - Sep 2018

All Science Journal Classification (ASJC) codes

  • Human Factors and Ergonomics
  • Safety, Risk, Reliability and Quality
  • Public Health, Environmental and Occupational Health
  • Law

Fingerprint

Dive into the research topics of 'Injury severity analysis of commercially-licensed drivers in single-vehicle crashes: Accounting for unobserved heterogeneity and age group differences'. Together they form a unique fingerprint.

Cite this