The Effects of 20 Mph Zones on Road Casualties in London: An Application of Doubly Robust Methods

This paper aims to evaluate the impacts of 20 mph zones on reducing road casualties in London by accounting for both confounding factors and the selection of proper control groups. A total of 234 treated zones and 2844 potential control zones are included in the data sample. This paper employed the doubly robust method which provides two opportunities for obtaining unbiased causal effects given the fact that we rarely know the true relations among potential outcomes, treatment assignment, and confounding factors. The doubly robust method is compared with another three methods: inverse probability weighting, regression adjustment and propensity score matching. The results show that the 20 mph zones consistently have significant impact on reducing casualties in both absolute number and percentages, especially for KSI and pedestrian-related casualties. It also suggests that the main factors affecting the decisions on 20 mph zones are the historical records of casualties and socio-economic characteristics.