e-Claire

A Post Millennial Consideration of Our Interconnection
by a simple tootsie from The Countryâ„¢...




If comments are closed, please email me: Claire AT e-biscuit DOT com




MAIN PAGE HERE





Dept. of Secret Messages

Quote meon an estimate et non interruptus stadium. Sic tempus fugit esperanto hiccup estrogen. Glorious baklava cheesecake ex librus hup hey yo ho ho ad infinitum. Non sequitur as usual, condominium facile et geranium incognito. Hoo-Ah! Betcha didn't know that!

Stuff by the Month

Most Recent Stuff

Syndicate

This page has been viewed 3061054 times

Referrers

Powered by ExpressionEngine

Remember that 100,000 Iraqis killed number? [updated]

I really thought better of Lancet Bunk worthy of FatAss Moore, War Profiteer.

via Jihad Watch: To come up with that figure, lead author Les Roberts of the Johns Hopkins Bloomberg School of Public Health in Baltimore and fellow researchers sent Iraqis to interview 998 families in 33 neighborhoods across the country. ...Because the sample size was so small, the range for deaths was [wide]: 8,000 to 194,000. So Roberts and company just split the difference. ...the researchers didn't feel bound by anything official like death certificates. Interviews were fine. "In the Iraqi culture it was unlikely for respondents to fabricate deaths," they wrote.

That goes beyond a SWAG into JMSU* territory. Do da name Baghdad Bob ring a bell. *Just Makin' Shit Up . UPDATE: I heard from Tim Lambert about some controversy about this Lancet study. Hell, it's turned into a Tech Central shootin' war. Piles of links and further argument here. While I am not a statistician, nor do I play one on blogs, this li'l ole redneck stater did learn in grad school how to read a study. Three major issues are a statistically significant sampling [n], clear correlation between results and conclusions drawn and reproducible results. For a country of 25 million or so, the sample size was 998 families in 33 "neighborhoods." [4e-05] Not so hot. The numerical range for their conclusion was "somewhere between 8,000 and 194,000." Not so definitive. Their conclusion of 100,000 deaths is a shot in a pretty big middle. Reproducible results will have to wait for the next study but the fact that there are so many voices weighing in with such enthusiasm to debate the methodology and the conclusions indicates that the study itself was not so very black and white proven in the first place. I will leave the details to the experts but I think I smell a bit o' bias...

Posted by Claire on 11/15 at 10:38 AM
  1. Well someone is making stuff up, but it’s Michael Fumento, the author of the critique you posted.  The researchers did check death certificates.  Why did Fumento say that they didn’t?  More here.

    Posted by Tim Lambert  on  11/15/04  at  08:51 PM

<< Back to main

Ponderables




moon phases
 




Fighting Fusileers -- Donate ! !

image

Site Meter