The purpose of this paper is to present results based on different statistical measures of attitudes and to discuss theoretical implications of the different approaches used, especially with respect to questionnaire studies. Initially comparisons between average attitude scores for 13 different traffic violations and for male and female drivers will be presented. In this approach there is no reference to behaviour. Secondly, intercorrelations between measures of attitudes and self reported behaviour are presented for drivers of different gender. Thirdly, the relative influence of attitudes on behaviour in comparison with other independent variables is presented for high and low speed conditions and for males and females. The effects of four different variables on behaviour were estimated by means of regression analyses of correlations and based on the theory of planned behaviour (as presented by Ajzen). The analyses presented are made for attitudes towards speeding on both low speed roads and high speed roads. (A)
Abstract