Figure 5

Collisions involving vulnerable road users: maps of the collisions and performance of the models. Maps are showing the reported numbers of vulnerable road users killed or seriously injured by a car or taxi, normalized by population. Scatter plots show the corresponding fit of the model with 2 independent covariates (see Tab. S1). Panel A refers to pedestrians, while panel B refers to cyclists. Colours correspond to those used in the legend of Fig. 3. Of the 24 cities under study, the 10 cities with the lowest vulnerable road users’ safety are British cities. Regression results showed adjusted \(R^{2}=0.55\) in panel A and adjusted \(R^{2}=0.36\) in panel B