This paper examines conspicuity-related motorcycle accidents in Malaysia. It presents a statistical analysis, whose aim was to establish a mathematical model to explain the frequency of such accidents in terms of a range of variables. In this way, it was hoped that new insights could be obtained into the effects of 'daytime running headlight' (RHL) intervention, whose purpose was to make motorcycles more conspicuous and thus reduce accidents. The explanatory variable effects investigated included: (1) the influence of time trends; (2) changes in the accident recording system; (3) the effect of fasting during the month of Ramadan; and (4) the Balik Kampong culture, a seasonal cultural-religious holiday activity unique to Malaysia's multiracial society. The database contained 4958 conspicuity-related accidents, which were later classified by week from January 1991 to December 1993. A loglinear time series intervention model and generalised linear modelling (GLIM) software were used for the analyses. The results of univariate and multivariate analysis and significance testing are tabulated. The best model to explain conspicuity-based accidents per week was identified, and it was concluded that RHL intervention had successfully reduced such accidents in Malaysia.
Abstract