This study examined the factors relating to severity of accidents in the Northern Region during the period of
2007-2009. Since the response variable is of binary nature (i.e., has two categories: fatal or non-fatal), logistic
regression technique was used to develop the model. Using the concept of Deviance together with Wald Statistic,
the study variables were subjected to statistical testing. Only two variables were included in the model, namely,
Over Loading and Obstruction. The observed level of significance for regression coefficients for the two variables
were less than 5% suggesting that these two variables were indeed good explanatory variables. The results also
showed that the model provided a reasonable statistical fit. Analysis of odds in this study also showed that
accidents due to Obstruction are more likely to be fatal.
The results of this study could be used to develop strategies to prevent and reduce fatal accidents in the northern
region in particular. The strategies include conducting awareness programs to educate both road users and drivers
and strict enforcement of the road safety regulations since the variables overloading and obstruction are all
violations of the road safety regulations