Something caught my eye in the ABC/USA Today/BBC/ARD survey.

[solarbird]

From the copy of the final report hosted at USA Today:

35. Have you or an immediate family member - by which I mean someone living in this household - been physically harmed by the violence that is occurring in the country at this time?

3/5/07: Yes: 17% No: 83%

CIA Factbook population of Iraq: 26,783,383
World Book household size, Iraq: 7.7

Assuming each "yes" answer above referred only to one and exactly one person (the most optimistic assumption, and the one which minimises the casuality count), this number implies that Iraqi casualties due to violence alone has been (26,783,383 * .17)/7.7, or 591,321.

Some recent studies of developing-world ethnic civil wars have shown that in these conditions in developing countries, the usual ratio of 2:1 injured:killed is, unfortunately, not as high. In these conditions, the fatality rate is dramatically higher. Unfortunately, I don't have links handy.

The Johns Hopkins survey put a highest-confidence fatality number included premature deaths from all sources, not only direct violence. That number was around 650,000. This survey shows close to 600,000 injured and/or killed by direct violence alone.

I would assert that the Johns Hopkins survey's findings are, unfortunately, indirectly supported by the results of this survey.

Comments

[llachglin.livejournal.com]

On first glance, this arguably shows fewer casualties than the Johns Hopkins estimate. That study also estimated the cause of the additional deaths, almost 92% of which were caused by violence. That's 601,000 dead from violence alone. So unless almost all of the casualties due to violence in this new survey are deaths and not injuries, it seems to be estimating lower numbers of deaths.

On the other hand, this analysis of the new survey is a minimum whereas the Johns Hopkins number is the middle of the range, so it's possible that they're consistent after all. The low end of the John Hopkins range is around 400,000, which is consistent with an estimate with a 2:1 death to injury ratio.

That was of last July. The death rate has, if anything, probably gone up since then. So it's likely than 80,000-120,000 additional Iraqis have died in the last eight months, if not more.

There are also reasons why the John Hopkins study would be more accurate. If you look into the details of their sampling, they have a good representative sample of different parts of the country. My guess is that ABC/USA Today/BBC probably had a less rigorous sample for their survey, one that would tend to produce slightly lower casualty estimates.

In any case, the order of magnitude is the same. It's really hard anymore to believe the lowball estimates that are based upon only the cases that can be confirmed by media reports, which tend to be off by a factor of ten or more.

[jordan179.livejournal.com]

What do you recommend as the solution to the Iraqi death toll? In particular, given the insurgency's shift to the deliberate murder of Iraqi civilians, how can we stop them? Simply pulling out won't help -- it will put the killers in power, from which position they can kill a lot more people. (See Cambodia, 1975).