Article

2005-01-27 Fifth Column

Researchers Who Rushed Into Print a Study of Iraqi Civilian Deaths Now Wonder Why It Was Ignored

Archived material is restricted to Rantburg regulars and members. If you need access email fred.pruitt=at=gmail.com with your nick to be added to the members list. There is no charge to join Rantburg as a member.

Posted by anonymous2u 2005-01-27 1:11:22 PM|| || Front Page|| [2 views since 2007-05-07] Top

#1 Eh?

It wasn't ignored -- it was turned into the A-#1 talking point of the perpetually indignant.

Posted by Robert Crawford 2005-01-27 1:46:55 PM|| [http://www.kloognome.com/] 2005-01-27 1:46:55 PM|| Front Page Top

#2 Idiots. You'd think that there would be a modicum of itelligence, but no, you get drivel like this.

"That's a classical sample size," says Michael J. Toole, head of the Center for International Health at the Burnet Institute, an Australian research organization. Researchers typically conduct surveys in 30 neighborhoods, so the Iraq study's total of 33 strengthens its conclusions. "I just don't see any evidence of significant exaggeration," he says.

As any Statistics 101 student knows a random sample is only valid if a random distribution can be reliably assumed. In a country like Iraq that is poppycock of the worst kind!

This is statistical analysis on the order of the man putting one hand in freezing water and one hand in boiling water and saying on average I feel fine. One skewed site could totally corrupt the extrapolations.

Posted by AlanC 2005-01-27 2:09:11 PM|| 2005-01-27 2:09:11 PM|| Front Page Top

#3 "... The Lancet, a formerlyprestigious British medical journal ..."

It's not quite as prestigious as it once was. Hope the incinerated credibility was worth whatever it achieved.

Posted by Bulldog 2005-01-27 5:59:53 PM|| 2005-01-27 5:59:53 PM|| Front Page Top

#4 Why would a Medical Journey print politics?

Posted by TMH 2005-01-27 6:27:06 PM|| 2005-01-27 6:27:06 PM|| Front Page Top

#5 Sorry. That should be Journal not journey.

Posted by TMH 2005-01-27 6:27:46 PM|| 2005-01-27 6:27:46 PM|| Front Page Top

#6 Yeah, I read that "study." It was a statistically inept and inaccurate load of CRAP.

Don't have a copy of it, but if memory serves they took the word of people - rather than looking at actual death certificates - and the guestimate (which is what it was) was anywhere from 8,000 to 200,000. So they just decided on 100,000. And this piece of flatulance was supposed to be peer reviewed? By whom? The DU?

What The Lancet did - for political purposes not even in its own country - was to make every other article they publish suspect. Their articles are often used in court cases to prove a particular medical point. I know if I'm involved in one in the future I intend to suggest to the attorneys they use this very flawed and obviously rigged article to question any Lancet article used by the other side.

If I were a genuine researcher who regularly published in The Lancet, I'd be livid at what they did.

Posted by Barbara Skolaut 2005-01-27 7:47:10 PM|| 2005-01-27 7:47:10 PM|| Front Page Top

#7 There is a discussion of this going on at Michael Totten's blog. A fool named Factcheck insisting that this is a perfectly reliable scientific study.

To refresh your memory Barbara the "sample" included 33 neighborhoods, 30 homes per neighborhood for a total of 8,000 people interviewed. They verified with a death certificate ~6% of the deaths, but, only tried to verify ~8%. There was a total of 21 violent war realted deaths from which they extrapolated 100,000. The 8,000 to 200,000 range was the 95% confidence factor.

If I tried this crap in a stat class (oh that was a loooooonnng time ago) I wouldn't have gotten an F; I would have been laughed out of class!

Posted by AlanC 2005-01-27 8:46:14 PM|| 2005-01-27 8:46:14 PM|| Front Page Top

14:45 True German Ally
14:45 True German Ally
00:56 Unagum Ulomoper7151
00:56 Unagum Ulomoper7151
00:54 Gravise Spolutle2771
00:54 Gravise Spolutle2771
00:02 Denver Reader 303
23:58 Barbara Skolaut
23:52 AJackson
23:52 Seafarious
23:47 wadikitty
23:44 2xstandard
23:43 phil_b
23:39 RWV
23:25 RWV
23:22 Mike Sylwester
23:04 crazyhorse
22:58 Mike Kozlowski
22:58 Dishman
22:55 Alaska Paul
22:54 Dishman
22:53 Frank G
22:47 Jonathan
22:38 Jarhead