This paper aims to evaluate the impacts of speed limit enforcement cameras on reducing roadaccidents in the UK. The propensity score matching (PSM) method is employed to control forselection bias and selecting proper reference groups. A naïve before and after approach and theempirical Bayes (EB) method are compared with the PSM method. We observe 771 sites and 4787sites for the treatment and the potential reference groups respectively for a period of 9 years. Both thePSM and the EB methods show similar results that there are significant reductions in accidentsnumber at all severities at speed camera sites. It is suggested that the propensity score can be used asthe criteria for selecting the reference group in before-after control studies.
展开▼