Scientists say the size of the survey was adequate for extrapolation to the entire country.
Maybe if it was truely random but it doesn't seem to have been random. Miniter weakens his argument by trying to make it to strong and saying "Game over", but he still has a point.
Researchers typically conduct surveys in 30 neighborhoods, so the Iraq study's total of 33 strengthens its conclusions.
33 beighborhoods but only 78 households where confirmation of death was attempted and only 63 here death certificates were provided. From those 63 households the study projects out to a country with more people than any US state except California. Or if you consider the projection to be from the larger set of people who provided a response, including the majority of the households where not attempt was made to confirm the deaths then Miniter is mostly right about the death certificates.
The uncertainty leads to the breadth of the so-called 95-percent confidence interval -- in other words, the 95-percent chance that the number of deaths in Iraq resulting from military activities is between 8,000 and 194,000.
Critics like the Slate writer seized on that range, says Dr. Woodruff, the government epidemiologist. "They thought, 'Well, it's just as likely to be 18,000 as 100,000.' That's not true at all," he says. "The further you get away from 100,000, the probability that the number is true gets much smaller."
The extremes of the range are less likely but they do show the lack of precision of the estimate even if you accept the sample as properly chosen, and accept the other assumptions of the study.
Households were informed about the purpose of the survey, were assured that their name would not be recorded, and told that there would be no benefits or penalties for refusing or agreeing to participate.
Those with deaths in the family might be more likely to participate.
and asked about any births, deaths, or visitors who stayed in the household for more than 2 months. Periods of visitation, and individual periods of residence since a birth or before a death, were recorded to the nearest month. Interviewers asked about any discrepancies between the 2002 and 2004 household compositions not accounted for by reported births and deaths. When deaths occurred, the date, cause, and circumstances of violent deaths were recorded.
Possibly causeing some double counting of some of the visitors. But that is probably a minor point I don't think it would make a huge difference.
A larger issue is the studies estimate of pre-war mortaility rates which are not sufficently established and could easily be much higher -
"The study, though, does have a fundamental flaw that has nothing to do with the limits imposed by wartime—and this flaw suggests that, within the study's wide range of possible casualty estimates, the real number tends more toward the lower end of the scale. In order to gauge the risk of death brought on by the war, the researchers first had to measure the risk of death in Iraq before the war. Based on their survey of how many people in the sampled households died before the war, they calculated that the mortality rate in prewar Iraq was 5 deaths per 1,000 people per year. The mortality rate after the war started—not including Fallujah—was 7.9 deaths per 1,000 people per year. In short, the risk of death in Iraq since the war is 58 percent higher (7.9 divided by 5 = 1.58) than it was before the war.
But there are two problems with this calculation. First, Daponte (who has studied Iraqi population figures for many years) questions the finding that prewar mortality was 5 deaths per 1,000. According to quite comprehensive data collected by the United Nations, Iraq's mortality rate from 1980-85 was 8.1 per 1,000. From 1985-90, the years leading up to the 1991 Gulf War, the rate declined to 6.8 per 1,000. After '91, the numbers are murkier, but clearly they went up. Whatever they were in 2002, they were almost certainly higher than 5 per 1,000. In other words, the wartime mortality rate—if it is 7.9 per 1,000—probably does not exceed the peacetime rate by as much as the Johns Hopkins team assumes."
slate.com
" Want more evidence the researchers knew their paper wasn't worthy of wrapping fish? The 100,000 figure is allegedly the excess over pre-war Iraqi mortality, which they claimed was 5.0 per 1,000 people annually. Not only is that far below the U.S. rate of 8.5 per 1,000, it's even below Saddam's own 2001 propagandist figure of 5.66!"
fumento.com
"Have a look at those confidence levels. Yup, 95%. That is, a one in twenty chance that the effect simply does not exist. Look at the relative risk ratios (leave out Falluja; I don't think anyone is really very surprised to see a higher mortality rate there): 1.1-2.3. It isn't just that it is an absurdly wide one (note, a relative risk ratio of 1 would mean no effect whatsoever) it is that if this paper was written to generally accepted statistical standards it would never have been published. With a 95% confidence level a relative risk ratio of anything less than three is regarded as statistically insignificant. Just to clarify that, by "insignificant" no one is stating that it is not important to those people who undoubtedly have been killed during the War. 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"."
techcentralstation.com
Also see
seixon.com
patterico.com
particuarly the first one.
maybe logictimes.com as well. It has its flaws, but it does make some points.
|
