Thursday, January 05, 2006

Fabricating Iraq's death toll

Remember the ridiculous 100,000 civilian deaths figure touted by the Left? Richard Miniter debunks the study it was derived from in this Washington Times column.


The claim that 100,000 civilians were killed in Iraq derives from a study done by an international research team led by Les Roberts of Johns Hopkins University's Bloomberg School of Public Health, which included researchers from Columbia University and Al-Mustansiriya University in Iraq. The research was done in Iraq in September 2004 and was published in the online edition of the Lancet on October 29, so that it would appear days before the November 2004 presidential election.


Aside from the obvious political bias of those conducting the study, problems with the methodology include:
  • Fogginess on the definition of "family"
  • Non-random sample selection
  • Unrealistic response rate, indicating that the most vocal Iraqis were represented disproportionately
  • Cluster sampling in Fallujah

    • Two-thirds of the deaths reported in the Roberts study were located in one cluster in Fallujah. Journalist Michael Fumento comments: "That's it, game over, report worthless." Fallujah, the scene of heavy fighting for more than two years, is simply not representative of Iraq as a whole.

      Oh yes, and remember the conclusion to the report?

      Roberts and his co-authors conclude that "there were 98,000 extradeaths(95%CI 8,000?194,000) during the postwar period." This means that, as Fred Kaplan writes, "that the authors are 95 percent confident that the war-caused deaths totaled some number between 8,000 and 194,000."

      I'm 100% certain that the chances of this study being bunk total some number between 100% and 100%.

      0 Comments:

      Post a Comment

      Links to this post:

      Create a Link

      << Home

      Blogarama - The Blog Directory