Corridor Safety Modeling Via Segmentation : A Selectivity Bias Perspective to Crash Count Data

This research shows a methodology to address a long- standing question in traffic safety relating to the evaluation of safety risk and the benefits associated with safety interventions. Safety interventions are associated with selectivity bias problems. By proposing a joint model of heterogeneity and selection bias using a discrete- count approach via multiple scales, the methods in this research may provide guidance on incorporating bias from site selection into design contexts associated with high levels of unobserved heterogeneity. This in turn can provide focused guidance on the measurement of factors that do not currently exist in highway databases.