Spatial Modelling of Road Traffic Accidents in Oyo State, Nigeria

  • GO Korter
  • OE Olubusoye
  • AA Salisu
Keywords: Road traffic accidents, spill over effect, spatial lag modelling, maximum likelihood estimation

Abstract

Roads often become sources of sorrow and venues of loss. Road accidents are a global scourge characteristic of our technological era, whose list of victims insidiously grows longer day by day. Most families have found themselves mourning, surrounded by indifference that is all too common as if this was a price that the society has to pay for the right to travel. The basic underlying assumption is that there is a spill over effect across the study area. The neighbourhood characteristics focused on the queen and rook contiguity based on weight matrices. The paper investigated the possible exogenous variables and built a spatial model for predicting areas with higher than expected future likelihood of accidents while controlling for spatial dependence. Removing the effect of spatial lag variable from the dependent variable (number of accidents), the area of each Local Government Authority, residential population, major road lengths and travel densities were used to predict areas with higher than expected future likelihood of accidents. The sign of the coefficient for the area is positive. This means an increase in the area of administration of local government authorities will lead to more accidents in each local government area. All other things being equal, Local Governments Areas with larger residential populations tend to have more accidents. The existence of a freeway link crossing a local government area reduces, on average approximately one accident for the period under study. Travel densities are negatively related to number of accidents, which suggests inhibiting factors in the sense that traffic generated tend to be associated with fewer crashes. Keywords: Road traffic accidents, spill over effect, spatial lag modelling, maximum likelihood estimation
Published
2014-09-16
Section
Articles

Journal Identifiers


eISSN:
print ISSN: 2315-6317