Toggle Main Menu Toggle Search

Open Access padlockePrints

Analysing Motorcycle Injuries using A Latent Class Binomial Logit Model

Lookup NU author(s):

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.


Abstract

The unobserved heterogeneity in accident data has widely been discussed as one of a major issue in road safety studies. Biased outcomes and predictions from the data analysis leading to misinterpretations due to this unobserved heterogeneity has been not given careful attention when analyzing such data. This study aims to identify the influencing factors for motorcycle accident injuries. The dataset used for the study consists of 1061 motorcycle accident injuries from 2010 to 2015 in Tabanan Regency, Bali. A latent class binomial logit model was estimated with specific attention to unobserved heterogeneity issues by classifying homogeneous attributes of two different accident data classes. This study found that male and older motorists, head on collisions and motorcycle at fault significantly influencing fatal motorcycle accident injuries. In addition, accident between/among motor vehicles, day time, right angle collisions and wrong directions significantly associated with serious motorcycle accident injuries. Based on the contributing factors identified in this study, some countermeasures for reducing the motorcycle accident injuries were proposed.


Publication metadata

Author(s): Wedagama DMP, Dissanayake D

Publication type: Article

Publication status: Unpublished

Journal: Journal of the Eastern Asia Society of Transportation Studies

Year: 2019


Share