[jefftk]

> women are much more likely than men to die or be seriously injured in a car crash

I don't think that's right, at least not as stated? Many more men die in car crashes than women: https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/...

Men tend to be in crashes that are more severe than women, though, probably because men are more likely to speed and drive drunk. If you try to look at crashes of equal severity (which is always a bit fraught since you don't have a real control group) then you do see something like what you're describing, though the NHTSA thinks this is mostly due to how robust men vs women are on average and not car design: http://www-nrd.nhtsa.dot.gov/Pubs/811766.PDF

[Sharlin]

Natural language is always ambiguous, but it is very reasonable to interpret "women are much more likely than men to die or be seriously injured in a car crash" as

    P(death_or_injury | crash AND female) > P(death_or_injury | crash AND male)

That is, women are more likely to die or be seriously injured CONDITIONAL ON being in a crash. So it is irrelevant that more men than women are injured in total.

[jefftk]

Conditional on being in a crash men are probably still more likely to be killed or injured, because the crashes men are in tend to be more severe (often due to their driving decisions).

[supercheetah]

Sure, there's still the question of survivability between men and women if those accidents are categorized based on severity with measuring things like speeds involved, whether it was a T-bone, type of road (highway, city street, etc.) and probably lots of other dimensions I'm leaving out.

[magicalist]

> Conditional on being in a crash men are probably still more likely to be killed or injured

Sounds like the "probably" here is a guess?

[lesuorac]

Young + Male is a great way to increase your car insurance [1]. I think it's safe to say car insurance's cost is reflective of risk (cost) with a more serious accident costing more than a less serious.

CDC also states that men get into fatal accidents much more often (2x) than females [2].

[1]: https://www.moneygeek.com/insurance/auto/rates-by-age-and-ge... [2]: https://www.cdc.gov/transportationsafety/teen_drivers/teendr...

[magicalist]

No, that says "motor vehicle death rate for male drivers aged 16–19 was over two times higher than the death rate for female drivers of the same age".

You're rehashing the above conditional probability discussion.

[Sharlin]

"Crashes involving male drivers often are more severe than those involving female drivers."

"The number of driver fatal crash involvements per 100 million miles driven in 2016-17 was 63 percent higher for males (2.1 per 100 million miles traveled) than for females (1.3 per 100 million miles traveled). Rates were substantially higher for males than for females ages 16-29, but were only slightly higher for ages 30 and older. The sex difference was largest among drivers ages 20-29."

https://www.iihs.org/topics/fatality-statistics/detail/males...

[magicalist]

"per 100 million miles" is also restarting this conversation from scratch.

Conditional on being in a crash is the operative clause here.

> Crashes involving male drivers often are more severe than those involving female drivers

is closer to the mark, and the next unquoted sentence "However, females are more likely than males to be killed or injured in crashes of equal severity, although sex differences in fatality risk diminish with age" seems like it may be what the article was talking about, though the cited source is from 1998, well before a number of modern car safety features were standard (even air bags weren't required until that year).

https://www.iihs.org/topics/fatality-statistics/detail/males...

[Sakos]

The key point is that in the same situation, women are more likely to be injured from what I understand.

[taeric]

It does sound somewhat dodgy, though; as it implies they throw out the fatality data. Would love to see the data.

[Sharlin]

More men than women die conditional on being in a crash, because men tend to be involved in more severe crashes. But in crashes "of similar severity" women tend to be more severely injured than men. One confounding variable is the size of the vehicle, but it may be factored into the "similar severity" calculation.

[mathieuh]

Maybe they were injured instead of killed?

[themaninthedark]

I remember hearing somewhere that a number of the men involved in "speed and drunk" crashes are going through financial issues, life insurance doesn't pay out if you commit suicide but it does if you crash into a bridge at 100mph.

[kube-system]

Life insurance does typically pay out in case of suicide after a 1-2 year cliff. It's a health issue like any other.

I doubt those instances are conscious financial decisions. Suicide methods are highly correlated to the methods people have at their disposal... and cars are for many, the most dangerous item they have.

[runnerup]

Life insurance often does but sometimes has a 2 year waiting period before it does. AD&D is a separate policy on top of life insurance, often worth an equal amount, which does not pay out in cases of suicide.

So masking a suicide can double payments to the family. AD&D also does not pay out in deaths involving toxic substances or “reckless driving”.

[watwut]

Men also drive more. Taxi drivers, truck drives, cops, all those professions are male dominated. And at least where I live, it is man drive family car to work and woman is more likely to go by public transport to work. Man is more likely to be the one to drive when family goes somewhere (and some are quite territorial about it).

[pwinnski]

In a car crash, that is, per crash, women are much more likely than men to die or be seriously injured.

Men are involved in many, many more crashes overall, but in a car crash, cars are currently designed to protect men better than women.

The indefinite article is doing a lot of work there.

[Sharlin]

Conditional on being in a crash, men are more likely to die due to being involved in more severe crashes. But ceteris paribus, women tend to be more easily injured.

[pwinnski]

Literally the first sentence in the article says, "Compared to men, women are 17 percent more likely to die in a car crash and 73 percent more likely to sustain serious injuries in a front-end collision."

[jefftk]

The article does say that, but the article is simplifying. If you click through they're citing https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/... which says 17% "given similar physical insults". But conditional on being in a crash is not the same thing as conditional on being in a crash of a given severity.