Multinomial Logistic Regression Model to Analysis Traffic Accident on Indonesia’s Regional Data

Abstract

Traffic accident is one of the highest causes of death in Indonesia following coronary disease and tuberculosis. Traffic accident is classified into three: mild, moderate and severe. The aim of this research was to determine the significant factors causing traffic accidents in Cilacap Regency using multinomial logistic regression. The data used were secondary data from the Resort Police of Cilacap Regency. The research’s response variable was accident classification deemed as a nominal scale, while the predictor variables were day of occurrence, time of occurrence, accident type, accident location, situation, weather problem, number of vehicles involved, and number of victims of nominal, ordinal and ratio scales. The research results show that the accident type, accident location, situation, and number of victim variables significantly influence the three accident classifications.