The problems in increasing road traffic accidents and their consequences to death, injury and disablity pose a growing concern, especially in developing countries including Korea.
This study reviewed some statistical analysis techniques on road traffic accidents data and investigated their accident severities by personal or environmental factors using Loglinear model in a local area.
This investigation was based on the 489 cases, obtained from Kwangju city during the period from January to May in 1992.
1. Prediction of roda traffic accident severities suing multiple linear regression was explained well by size of population and number of cars. Small number of papers published with principal component analysis, factor analysis and discriminant analysis dealing with quancitative variables were not appropriate as analytical techniqus for the data set mainly with categorical data. Therefore. more data is required for appropriate replenishment and analysis for the techniques.
2. the degree of severities of road traffic accident was not found different by personal and environmental factors, drunken driving, driving over 50㎞/h, the novice driver and size of car. But the accidents caused by pedestrians´ crossing without care were found significant.
|