|
Abstract
|
[Note: The full text of this article is no longer available because the publisher's website is no longer online. (October 2022)]
Road accident in Ghana is known to be among the nine major causes of deaths, WHO (2011). It is reported that there is an average of 1909 people who are killed by road accidents annually. The most dangerous part of it is that about 60% of these people who die through road accidents are within the ages of 16 and 45 years who represent the bigger part of labour force of the country. The ultimate goal of this paper is to use negative binomial regression as an approximation to Poisson regression to fit a model to the secondary data which was obtained from the Building and Road Research Institute of the Council for Scientific and Industrial Research on the number of people killed by road accidents in Ghana from 2001-2010, given the ages of those who were killed as a risk factor in a given time period (in years). The descriptive data analysis of the data was performed using the Microsoft excel and Minitab 14 statistical software, while the model fitting and model selection were done with R statistical package which applies Akaike Information Criterion (AIC) for the best model selection. Maximum likelihood estimation was employed by the R statistical package for the parameter estimates for the model.
It was observed from the results that in general the number of people killed in road accidents gradually increases with time (as the years go by, the number of people killed by road accidents increases). The age of a person involved in road accident could significantly determine as to whether one would be killed in road accident or not. The results also showed that people in the age group of 26-35 were significantly killed most in road accidents in Ghana. This was followed by 16-25 years group and those in the ages of 36-45 were slightly smaller than 16-25 years but those above 60 years were least killed in road accidents.
Keywords: Ghana; road traffic accidents; fatality; Negative Binomial Regression; Poisson Regression.
Language: en |