[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...