wrong statistics

The Guardian reported today on a larger probability for men drivers to kill/injure a pedestrian, when compared with female drivers. Which is somewhat misleading. The figures report on the probability that the driver is male, given a serious collision occurred. But it does not directly reflect on the (very probable!) gender difference in dangerousity, i.e. the difference in the probabilities that a female driver vs. a male driver run over a pedestrian because it does not account for the probable difference in women vs. men drivers on the road at any given time, or on the mileage accumulated by each gender over a given year. Which differs by almost two…  (As mentioned in the article, a disproportionate percentage of accidents are due to vans, which again are more likely to be driven by men.)