Women are much more likely to be injured in car crashes, probably because crash-test dummies are mostly male-shaped

Originally published at: https://boingboing.net/2019/07/23/in-every-dreamhome-a-heartache.html


Using human cadavers is often a much more informative way to test automotive safety then crash test dummies, but sometimes it’s hard to get the people who write laws governing treatment of human bodies to understand that.

A typical argument might go “why do you need to desecrate a child’s body instead of just using a dummy that behaves the same way?” To which the answer is “the only way to ensure our dummies break the same way as a child’s body is to break a bunch of children’s bodies first.”


Color me shocked; the assumptive ‘I-am-the-center-of-the universe’ personal bias of White men, (which automatically assumes that they and their typical accompanying physical traits are “the norm, the default”) strikes, yet again.


I have never worn a seat belt that did not keep creeping up until it caught me across the throat. The headrests at the lowest position are too high for me.

Half of people in cars are female but they are designed as if they were only used by men.


This isn’t news to any petite woman who’s driven a modern vehicle. Your seatbelt height adjuster might go lower nowadays, but if you’ve got boobs your shoulder belt might still try to creep up over them and onto your neck. That firm, forward-jutting head restraint that forces a short driver (or one with a ponytail) to crane her neck down? It was designed for a man who not only sits farther back and more reclined, but also has more neck muscle to prevent whiplash in a rear-end collision. You are considered to be “out of position” if you sit up straighter or move the seat forward.

It doesn’t take a ton of creativity to figure out how certain things that are a daily annoyance for many women could become dangerous in a collision, and that’s before you even consider the physiological differences that don’t create such obvious, constant reminders that your car was not built for your comfort or safety.


Big problem in medicine too. Not just physical traits, but the more they look the more they realize medications don’t work the same among different genetic populations.


I wish White men designed seats in commercial airplanes.


Yep. Modern medicine is another reason that it’s been harder to get younger cadavers than it was in the past. Plus you can’t use cadavers that are too old and frail, or use someone who died in a car crash because that would lead to bad data.


Have the crash test dummies been adjusted to take account for other changes? We often hear that the general population is getting more obese, presumably that will also have an impact on crash safety.
Is it better to say that the crash test dummies are only testing a shrinking portion of the market with more and more of the population not being catered for.
Are crash test dummies the same size and shape in Japan as they are in US?


The simple answer here is, clearly, that women should not be allowed in cars.

1 Like

AFAIK that’s only true for certain analysis, too - a dummy can have sensors in things like bones and other locations where placing similar sensors in a cadaver might weaken the bone or so on.

I’m actually surprised there hasn’t been more research in this area though, but I assume it’s because the cost-per-dummy starts to become astronomical if you start outfitting them with exotic components, novel materials, and appropriate sensors, so the existing set were likely deemed “good enough” as a cost savings measure.


Headrests have a tendency to hit me at the largest diameter of my skull, which means they push my head forward in a way that causes headaches, and I always worry that in an accident my neck would snap from the impact. (I’m an average height woman.)


Or maybe we just create special egg-shaped carrying cases for them? Women could be safely ensconced in these protective pods at special loading centres, transported to the place of their (or their male guardian’s) choosing, and then unloaded and unwrapped like precious daffodils.


The published paper is worth a read, the discussion on the gender differences is on page five and is very accessible.

The gender impact in the model analysis is huge compared to the other factors such as BMI, height and speed. Though there are some details dropped from the various summaries.

The paper points out that “the greatest sex-related effect on injury risk occurred in the
lower extremities”. In particular ankle injuries are 3.8 times more likely, leg injuries are 2.3 times more likely. They have controlled for height, bmi and age, “this suggests that sex itself has an effect on
injury tolerance”.

Which suggests that even with the inevitable improvements to lower extremity safety the gender injury gap is likely to remain. Though ankle injuries may be impacted by footwear choices.

Also interesting is that the rate of skull fractures and severe brain injuries are half as likely for women than men, even taking into account speed differences.


Well worth checking out this book on this and related “it’s a man’s world” issues…
Criado-Perez, Caroline Invisible Women: Exposing Data Bias in a World Designed for Men (Chatto & Windus, 2019)


That sounds like a cost of calibration thing. I don’t work with bodies, but I do work with modelling human behaviour (as a transport planner/transport policy wonk), and while the algorithms to describe how people move through a subway network are generally reliable, it’s the starting conditions and the detailed relationships (eg, the cross-elacticities between the costs of car vs transit) where costs of collecting data becomes a factor.

In other words, you can model this stuff really well and cheaply, but it’ll cost to provide real-world sample data as an input.

(ETA - which is no excuse not to collect population-representative data, and if women aren’t a majority in that sample then you live somewhere that the WHO doesn’t have any data for!).


The dummy family is structured around 19 family members: 8 adults and 11 children, the smallest being an infant weighing 3 kg.

According to Volvo, there are family members of different configurations and for different purposes, making a grand total of more than 100 dummies. The family is so large because it is necessary to include both children and adults of different ages and sizes to cover different collision scenarios.


Not a problem for just petite women. I’m as tall as most men and have to adjust the shoulder strap numerous times during any ride.
So boobs are an issue. So are hips- that more extreme curve on most women means the lap belt creeps up too. It’s supposed to sit around the pelvis, but ends up at the waist.


But whether you use a dummy or a cadaver in the actual crash test, you still need to break a lot of cadavers in order to get the data you need to design a dummy that approximates a human body.


:rofl: That would be one stupid mistake right there.

1 Like