Development of Latent Class Model to Analyze the Severity of Road Accidents in Iran

Accidents in Iran pose a significant public health burden, requiring effective prevention and management strategies. Latent variable modelling provides a promising approach to understanding the complex factors that contribute to accidents in Iran. This paper reviews the current state of knowledge of latent variable modelling in Iran's accidents. The paper provides a synthesis of the literature on the measurement and modelling of latent variables in accidents, including human factors, environmental factors, and organizational factors, and also reviews the common statistical techniques used in the analysis of latent variables, such as structural equation modelling, latent class analysis, and factor analysis. The strengths and limitations of these approaches are discussed. The review shows that although some studies have applied latent variable modelling to accidents in Iran, the use of these techniques is still relatively limited in comparison to other countries. Overall, It has been argued that latent variable modelling can offer valuable insights into the underlying mechanisms of accidents in Iran and guide more effective strategies for prevention and management. This study used data from accidents resulting in the death or injury of passengers gathered on Iranian roadways in 2015, and the latent class model was fitted to this data using the NLOGIT6 software. And the findings have been reported in this article. The findings suggest that elements such as the season and day of the accident, as well as weather conditions, have an impact on the severity of accidents.