...finally!!! Some official casualty figures from Iraq!

[jimmmy]

I find nothing to GD with in this last post XT. In fact, I will offer a third good source with the Lancet and the IBC: Human Rights Watch [no neocon friend of the war] has been outspoken in saying the Lancet numbers are an “inflation” and put the civilian casualty number at 16,000-some last fall.

[Sam_Stone]

The Lancet report is not a statement of how many people were killed in the war. It is an estimate of how many more deaths were there in Iraq during the war and post-war period than during a comparable pre-war period. If someone dies because they cannot get to the hospital because of insurgent activity, that counts as an excess death - it is likely that person would have lived had the war not occured. That said, the Lancet investigators did report a high proportion of deaths were due to violence (as reported on death certificates).

The problem I have with the Lancet report is that its error bars were so large that its conclusion was essentially meaningless. Didn’t it have a ‘low’ figure of 8,000 or something? And a high figure of 300,000? I haven’t read the full text of the study, but when I heard that the 95% confidence interval was that big I figured that they did that so that they could come up with a suitably scary number to publish just before the U.S. election.

I’m happy to be corrected if that’s not the case.

[XT]

It was 8000 and 194,000 if I remember correctly Sam…and they said the most likely figure from their data was in the 100,000 ballpark. But you are right…giving such a wide range doesn’t exacly instill confidence either.

-XT

[hildea]

@xtisme

One of my problems with the Lancet study is that they used mortality figures from pre-invasion Iraq that may or may not have been very accurate…there is no way to tell. However, from what I’ve read they were most likely low…and some say VERY low.

The Lancet study got their pre-invasion mortality figures from the same interviews as the post-invasion figures:

@The Economist

In each household, the interviewers (all Iraqis fluent in English as well as Arabic) asked about births and deaths that had occurred since January 1st 2002 among people who had lived in the house for more than two months. They also recorded the sexes and ages of people now living in the house. If a death was reported, they recorded the date, cause and circumstances. Their deductions about the number of deaths caused by the war were then made by comparing the aggregate death rates before and after March 18th 2003.

(link)
They did comment that this method means that there’s a danger of underreporting pre-war infant deaths, as they might be viewed as less significant than other deaths, and thus might be forgotten or not mentioned if it’s a long time since it happened. However, IIRC, the study comments that the pre-invasion death rate they got this way is remarkably similar to death rates from other sources. (The Lancet website is taking its sweet time sending me that password reminder, so I can’t look up the exact text right now.)

It’s also worth noting that the Lancet study asked about deaths among people who had been living in the interviewed households when they died, so Iraqi soldiers killed in combat are not likely to be included.

@Sam Stone

The problem I have with the Lancet report is that its error bars were so large that its conclusion was essentially meaningless. Didn’t it have a ‘low’ figure of 8,000 or something? And a high figure of 300,000? I haven’t read the full text of the study, but when I heard that the 95% confidence interval was that big I figured that they did that so that they could come up with a suitably scary number to publish just before the U.S. election.

If they wanted a huge number of deaths for political purposes, it would have been much smarter to use another confidence interval to make the study seem more certain. Using a confidence interval of, say 80% or so would have narrowed down the possible range considerably. (According to the Economist, there’s a 90% chance of the correct number being more than 40.000 dead, so another 10% would have done wonders.) They give their most likely number around 94.000 - that’s big and scary enough, and that wouldn’t have changed if they’d used a narrower confidence interval.

@xtisme

It was 8000 and 194,000 if I remember correctly Sam…and they said the most likely figure from their data was in the 100,000 ballpark. But you are right…giving such a wide range doesn’t exacly instill confidence either.

I see it the opposite way: A study which describes likely sources for error and gives an wide estimate of error which results in the study being less useful for bombastic poilitical statements inspires more confidence in me than a study which claims to list exact numbers of people killed and wounded in a six month period in a war-torn country. (When the Iraqi Health ministry lists i.e. number of wounded as 12.657, and not “approx. 12.660”, they imply pretty strongly that the correct number is exactly 12.657, and not 12.656 or 12.658.)

Sure, there’s a lot of insecurity around the Lancet study – as the study says itself. Running a similar study with a bigger sample would be a good thing. Unfortunately, it would also be a dangerous, difficult, and expensive thing.

[SentientMeat]

Just to draw a parallel, does anyone think that the number of actual, identified bodies whose cause of death was confirmed by post-mortem as “drowning on or around Boxing Day 2004” is anything like an accurate representation of the number of people who would be alive today had the tsunami not invaded those coastlines?

[newcrasher]

@Banquet Bear

…well, as official as leaked figures can get, of course…

I commented in this thread that it was hard to criticize the Lancet Study on Iraqi deaths when there were no official death rate figures.

Well, turns out I was wrong. There are offical figures, but they are only available to Iraqi Cabinet Ministers. Here is the breakdown on Iraqi civilian deaths due to Coalition Forces, Iraqi Forces, and the Insurgents-for the period 1st July 2004 to 1st January 2005:

3,274 civilians killed in total
2,041 by coalition and Iraqi security forces
1,233 by insurgents
12,657 civilians wounded in total
8,542 by coalition and Iraqi security forces
4,115 by insurgents
cite

So out of all civilian deaths attributed to either security forces or the insurgents, 60% were caused by the “good guys.”

US Ambassador John Negroponte stated in an interview shortly before the figures were obtained:

…so the points for debate:

Why are these figures not released to the public? Is Ambassador Negroponte so “out of the loop” that he doesn’t know who is causing the greatest amounts of civilian deaths? Or is he simply spinning the truth, or dear I say it, “not-lying”? And what, if any, are the implications of these figures?

It would have been nice for you to stick around to debate the retraction.

I sense the glee as you were typing about those killed by coalition forces. I guess once the info became innacurate, their deaths no longer pleased you so.

[Banquet_Bear]

@newcrasher

It would have been nice for you to stick around to debate the retraction.

I sense the glee as you were typing about those killed by coalition forces. I guess once the info became innacurate, their deaths no longer pleased you so.

…glee? What glee do you see, praytell?

I’ve been trying to find out the civilian death rate in Iraq since the war began. For the first time ever-we finally got some official figures-albiet from a leak.

Those figures showed that the Coalition troops were responsible for more civilian deaths than the Iraqi insurgents. The retraction stated that those “civilian” deaths also included “terrorists” killed by the Coalition troops…meaning…what, exactly? That no civilians have been killed by the Coalition? That lots have? It disturbs me that you aren’t interested in the truth newcrasher…why don’t you want to know? How hard can it be to tell a civilian death from a terrorist one? Why aren’t these figures released? How many Iraqi’s, newcrasher died of a heart attack last week? How many were murdered? How long do you think it will be before the Iraqi Government starts to record and publish the death figures again?

If you want to accuse me of taking delight in the deaths of Iraqis, feel free to take it to the pit-I find the accusation disgusting.