Women wearing seatbelts are 73% more likely to be killed or seriously injured in a car crash than men in seatbelts, and while it's not entirely certain why this is, it's a pretty good bet that sampling bias in crash-testing is to blame.
Specifically, until very recently all crash-test dummies were male-bodies; when woman-shaped dummies were added, they were shorter and smaller than most women, so that the researchers could capture a wider range of datapoints, from small people to big people.
Female bodies and male bodies have typical differences in fat distribution and skeletal structures, so crashes that don't injure male dummies might still present problems for female dummies — and crashes that involve women who are shaped different from the petite dummies now in use might also cause injuries that are not visible with the dummies that are a proxy for them.
All this is uncertain, but not because it's impossible to increase certainty — rather, there's been next to no research on the subject, and that research will take a long time, even if we start studying it now. The statistical picture built up with male-bodied dummies has been decades in the making, and simply diversifying the shape of dummies in tests starting from today will not provide that longitudinal picture that we need to make real improvements.
Mueller says retooling crash-test dummies can't happen overnight—it takes 20 to 30 years of bio-mechanical research and testing to build and fine-tune each model. Many of the dummies manufacturers are using now were built using data from the 1970s and 1980s, which skewed heavily male. "The creation of new dummies [physically] will be able to advance faster in the future but it will still take 10-plus years to collect enough data to relate the dummy's performance to real world injuries," she said. "You never want people to get injured, but in order to gain enough info about the real world, we have to sit patiently and wait for real-world data to come in and for real-world specimens to come in."
A Clue to the Reason for Women's Pervasive Car-Safety Problem [Sarah Holder/Citylab]