<![CDATA[Traffic accidents are one of the major transportation safety problems that may cause casualties and material losses. Analysis of the number of accident victims is important to describe accident severity and identify influencing factors. This study aims to model the number of traffic accident victims in Lamongan Regency based on time characteristics and road conditions using Quasi Poisson regression. The data used were secondary data obtained from the Lamongan Police consisting of 1,202 traffic accident observations during 2025. The analysis stages included descriptive analysis, Poisson regression modeling, dispersion testing, and Quasi-Poisson regression modeling. The dispersion test result showed a dispersion parameter value of 0.4161, indicating underdispersion where the variance was smaller than the mean. This condition caused the standard Poisson regression model to be less appropriate because the equidispersion assumption was not fulfilled. The Quasi Poisson model produced more reliable statistical inference by adjusting the variance through a dispersion parameter. The significant variables affecting the number of traffic accident victims were month, day category, road status, and road condition. Therefore, the Quasi Poisson regression model was more suitable for modeling underdispersed traffic accident count data.]]>
Copyrights © 2026
