Loading...

Studying the Role of Distraction of Men and Women in Accidents Using Structural Equation Modeling

Piri, Mehran | 2022

182 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 54947 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Kermanshah, Mohammad
  7. Abstract:
  8. Crash severity is among the fundamental safety indicators in transportation. In recent years, distracted driving has been one of the biggest concerns in road crashes. Many scholars have evaluated the effects of distraction in crash severities and injuries. This study uses the latest general estimates system (GES) crash dataset, which is a subset of the US National Automotive Sampling System (NASS) and the structural equation model (SEM) approach to explore the relationships between distraction factors and crash severity in male and female drivers. Using SEM for modeling about 22 thousand observations offers the opportunity to simultaneously examine complex relationships between variables by handling endogenous and exogenous variables. Six latent variables were conducted for the modeling purpose. Driver characteristics, weather conditions, road conditions, and lighting conditions are the four exogenous variables. The endogenous variable is the crash severity, and the mediator variable is the distraction. Three SEM models (one for men, one for women, and one for all observations) were developed to estimate the hypothesized relationships among latent variables. The results showed that distraction is inversely related to the crash severity, and the probability of severe crashes was more seen in distracted female drivers than males (-0.378 compared to -0.447). The cognitive distractions had the most significant impact among different distraction factors on the crash severity. Moreover, the most influential factor affecting severity is the driver characteristics which affects men and women's crash severities by 0.379 and 0.272, respectively. Additional insights are provided through the development and analysis of each model.
  9. Keywords:
  10. Driver Gender ; Structural Equations Modeling ; Crash Severity ; Driver Distraction ; Traffic Data

 Digital Object List

 Bookmark

No TOC