The Lancet, blunted
Yesterday The Gantelope wrote about The Lancet’s Iraqi casualty estimate of “8,000 more people and 194,000 more people may have died because of the war”. Lots of room for error there. Hardly surprising sinceAlthough the teams relied primarily on interviews with local residents, they also asked to see at least two death certificates at the end of interviews in each area. That means that in 30 areas with 30 interviews each, a total of 900 interviews (30 x 30), they only asked to see 60 death certificates (30 clusters x 2) -- and, did they actually see those? Fred Kaplan in Slate points out,

It means that the authors are 95 percent confident that the war-caused deaths totaled some number between 8,000 and 194,000. (The number cited in plain language—98,000—is roughly at the halfway point in this absurdly vast range.)
This isn't an estimate. It's a dart board.

"The study purports to show civilian casualties 5 to 6 times higher than any other reputable source", states Shannon Love and she explains why the methodology is off. She explains cluster samples, faulty assumptions (such as "violent deaths were widespread"), self-reporting, and the study's statement, "Two-thirds of all violent deaths were reported in one cluster in the city of Falluja", among other things. One of her commenters asked,

The lead author was an opponent of the war.
The lead author submitted it to the Lancet on the express condition that it be published before the election.
Do you suppose the guy might have a bit of bias of his own?

Tim Worstall takes The Lancet to task in his article The Lancet: A Casualty of Politics

. What is being said is that we don't have enough information to be able to say anything meaningful about it. "Statistically insignificant" means "we don't know".

In effect, what has been found in this paper is nothing. Nada. Zip.

Except of course that one of the two leading medical journals in the world has published a piece of shoddy research four days before the US elections with the obvious motive of influencing them. Sad, that, and my apologies as an Englishman that it should be one of my countrymen who did such a thing.

.

It wasn’t just a Brit. The team included researchers from the Johns Hopkins Center for International Emergency, Disaster and Refugee Studies as well as doctors from Al Mustansiriya University Medical School in Baghdad. Does that mean that Johns Hopkins and The Lancet now join the Guardian and CBS?
Sad.