Hacker News Reader

Comment by cafard | Hacker News Reader

Eton has been around since before the Reformation, so it has had the time to build up a list of murderers, etc.

Sure, but the article gives four examples that took place over the course of thirty years, two of which were within twelve months of each other.

Eton appears to be a 6 year school with a capacity of 1390. So, at most, 230 at year, so at a murder (or murderer, really) rate of 4/30=0.13 per year, that implies a murder rate of (0.13/230)*100000 = 56 murders per hundred thousand. The UK homicide rate (which includes non-murder homicide) is 1.1 per hundred thousand.

56 per hundred k is higher than _any country on earth_, so this is pretty impressive.

(The absolute numbers are small enough that it could all be fluke, of course. Also I have some vague qualms about comparing murderer rate to murder rate, but I think it _mostly_ works, as most murderers only do one).

Thanks for doing the math!

Murders have been declining, right? I wonder how much of this is due to comparing historical murders with current murder rates?

For the UK they're declining from a peak in the late 90s, but they were kinda flat before that, and in in any case you're talking about a difference of under 1% either way. There was never a time that the UK's murder rate was anything like 56/100k, at least not in the last century (before the late 19th century, useful stats aren't really available, but some estimates do show _very_ high murder rates in the 19th century).

"56 per hundred k is higher than _any country on earth_, "

Plenty of locations have rates around that. That's why it's important to compare similar sized areas.

An elite school having a similar rate as the bad parts of Mexico would still be shocking. But, Mexico is counting deaths not killers so victim rate is a better point of comparison.

Some US cities have similar rates. Even then, it's not the same as comparing to other schools. And yeah, murder rate vs murderer prevalence would would different things. I don't think there's a lot of data on the latter.

Whoa, I just double checked and 3 US cities are over 50 victims per 100,000.

The 0.13 murderers per year were Old Etonians, so need to be divided by that population. Assuming alumni live on average another 50 years, the murderer rate per 100,000 would be 1.13, compared with the homicide rate of 1.17 for England and Wales in 2020. [1]

[1] https://www.ons.gov.uk/peoplepopulationandcommunity/crimeand...

Murder rate counts victims not killers, one of these guys killed 9 people. Also, only including people who got caught very likely under reports the numbers.

But murder is mostly committed by young men. And 3 out of 4 of the Etonian murderers were below 30 at the time of the murder.

So this correction is not as big as you think.

The murderer rate for young alumni of Eton would be higher, but so would the rate for young people generally. In fact, the correction would be even bigger as Eton is a boys-only school and the homicide rate for the wider population was for both men and women.

This is the right calculation. It baffles me that there are so many comments assuming the homicides are committed by current students.

Correct it further for the fact that convicted murderers are overwhelmingly male (93% according to your same source) and old Etonians begin to look downright peaceful. But that wouldn't make a good article.

The average of one murder per murderer I'd generally agree with, but one of the Eton murderers killed 9 before committing suicide. So you are right to have qualms. But it is an upper bound and so still useful.

That said, one correction. El Salvador had a murder rate in 2017 of 61.7/100,000. This is comparable to Eton.

Let's take your analysis a step farther. If we have 230 * 30 people, each of whom has 1.1 chances in 100,000 of committing murder, the number of expected murderers is 0.0759. And the distribution of number of murders is a Poisson distribution. That means that the probability of k murders is λ^k e^(-λ) / k!.

Therefore our estimates are:

    0 murderers = 0.9269088928142737
    1 murderers = 0.07035238496460337
    2 murderers = 0.0026698730094066973
    3 murderers = 6.754778713798945e-05

Add those up and the expected probability of 4 or more murderers is only 1.3014245782150269e-06.

Therefore, even with a small absolute number, we can be very sure that the true murder rate for graduates of Eton are significantly higher than the UK population.

However Eton does attract people internationally. And the international murder rate in 2017 was 6.1/100,000 per https://worldpopulationreview.com/country-rankings/murder-ra.... Using that as a murderer rate, the odds against Eton producing 4 murderers in 30 years improve...to about 1/1000. Which means that Eton's murders are still likely to not be chance.

the true murder rate should be higher than the general population because its a boys school and men are carrying out more murders than women. but I assume its still higher once you account for that.

Good point.

But looking at https://data.worldbank.org/indicator/VC.IHR.PSRC.P5?location... it looks like the homicide rate already was single gender. Even the peak rate in 2002 (probably skewed upwards by Harold Shipman) is well below the current world average. And UK rates have been low for many decades. 2002 is the peak of a rise in homicides starting around 1960, and it had been low since at least the 1800s.

I dug in a bit farther. Only about 5% of Eton's students are international. So let's lose them, and also lose the older murderer and the international murderer. Even using the highest murder rate from 2002, there is less than a 1% chance that they'd have had 2 or more murderers in a 30 year period.

Eton is all male and males are far more likely to be killers than females. Still a high murder rate for sure.

To remove the problem of small numbers, you could use a Poisson distribution to perform a statistical test on the number of murderer rate. The actual murderer count (4 murderers) as compared to the expected murderer rate (0.07 murderers) gives a p-value of about 1e-6.

That said, this doesn’t account for the look-elsewhere effect, so it probably should be scaled by the population size. However, even after scaling by the ~32k public schools in the UK, it still gives a p-value of 0.03, so it’s statistically unlikely that there would exist a single public school among all 32k that would have this high a rate of murderers.

The list is not exhaustive.

Reply on news.ycombinator.com