Data analysisDescriptive statistics were used to describe the demographic characteristics of all the respondents. Abivariate-correlation analysis was then carried out toevaluate any correlation between the three dimensions ofperceived causes of motorcycle accidents and the drivingviolations carried out while riding a motorcycle. This wasfollowed by an independent ttest to evaluate any significant differences between motorcyclists who were involvedin accidents and motorcyclists who were accident free, interms of these dimensions and behaviours. For the prediction of a motorcycle accident, univariate logistic regressions were first performed to examine the effect ofdifferent demographic variables, alignment with thedifferent dimensions of perceived accident causes, and theextent of driving-violation behaviours. Variables reachingp>.05 were then eliminated. The remaining variables(that is, withp<.05) were then entered into a multivariatebackward stepwise logistic regression to determine theircontribution to predicting active involvement in accidentsamongst these motorcyclists. All statistical analyses wereperformed using the IBM SPSS programme version 20.0 forWindows (IBM SPSS Inc., Chicago, IL, USA); the significancelevel was set at p<.05.
การแปล กรุณารอสักครู่..
