Analysis of the impacts of risk factors on teenage and older driver injury severity using random-parameter ordered probit

Sina Darban Khales, Mehmet Metin Kunt, Branislav Dimitrijevic

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

The study analyzed injury severity of teenage and older drivers using 2015–2016 crash data from New Mexico. The fitness of the random-parameter ordered probit models developed for each age group was tested using likelihood ratio, comparing them to a unified model that combines both age groups, as well as comparing the random-parameter to fixed-parameter ordered probit for each age group. In both cases separate random-parameter ordered probit provided better results. It was found that vehicle type and age, lighting condition, alcohol or drug use, speeding, and seatbelt use were significant both for the teenage and older driver injury severity. The weather condition and gender were significant only in the teenage driver model, while driver inattention was significant for older drivers. The impacts of crash factors on injury severity was analyzed using marginal effects. The results indicate notable differences in the effects of contributing factors on driver injury severity between teenage and older drivers, including the sensitivity to changes in the mutual predictor parameter values.

Original languageEnglish (US)
Pages (from-to)1249-1257
Number of pages9
JournalCanadian Journal of Civil Engineering
Volume47
Issue number11
DOIs
StatePublished - 2020

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Environmental Science(all)

Keywords

  • Injury severity
  • Older drivers
  • Ordered probit model
  • Random parameter model
  • Teenage drivers

Fingerprint

Dive into the research topics of 'Analysis of the impacts of risk factors on teenage and older driver injury severity using random-parameter ordered probit'. Together they form a unique fingerprint.

Cite this