Dead baby in a warzone

[pozorvlak]

There was a story on the radio this morning about a NATO official claiming that Kabul was safer for children than London, New York or Glasgow. This was followed by the usual predictable stream of denunciations and cries of what an appalling thing this was to say. But nobody directly addressed the truth of the statement. And I'd really like to know

Comments

Homicide Stats in Scotland

[ext_254110]

I found the data set for the homicide rate in Scotland for 2008-9 (http://www.scotland.gov.uk/Topics/Statistics/Browse/Crime-Justice/Datasets/HomData). I haven't crunched the numbers yet but this might be a start.

Re: Homicide Stats in Scotland

[pozorvlak]

Thanks!

[oedipamaas49]

Going on national figures, mortality under 5 is 23.54% in Afghanistan, 0.6% in the UK. Couldn't find any by-city figures, but the regional differences would have to be pretty impressive to overcome a ~4000% gap.

Agree with you about the journalism, though. No idea how much of the problem is culture, how much is just that they are throwing these articles together in not much more time than we spend grumbling about them on LJ.

[pozorvlak]

Surprisingly, no: it actually brings the gap up to 7800%. While Afghan rates of post-natal death are much, much worse than UK rates of post-natal death, almost all UK mortality under 5 is due to infant mortality.

The trouble is that this wasn't a statistical claim, as such. What parameters should be included or not-included? Disease, car-related, murder/infanticide, or simply violent deaths on the streets.

Yes, excellent point. On the other hand, I've been going through the UN stats site oedipamass49 linked to, and I don't think there's a way of comparing like with like that makes the UK come out worse...

[pozorvlak]

Good site! Subtracting infant mortality actually makes the gap worse: 78.4 non-infant deaths under 5 per thousand versus 1. On the other hand, we may now have enough data to answer my question... there are 5.1 million children under 5 in Afghanistan, and StC claim 1000 died as a direct result of the conflict last year, giving 0.2 deaths per thousand per year from violence. Chris's stats give us a figure of 5 deaths per million due to homicide in the under-16s for the same year, if I'm reading the table right.

So, even under the most charitable interpretation of Sedwill's claims, he's wrong by a factor of 40.