Wednesday, October 11, 2006
More October Surprises!
The Associated Press reports on the latest Public Health propaganda crafted by the partisan duo of Johns Hopkins Bloomberg School of Public Health and the British Medical Journal, The Lancet:
Deaths are occurring in Iraq now at a rate more than three times that from before the invasion of March 2003," Dr. Gilbert Burnham, lead author of the study, said in a statement.
The study by Burnham, of the Johns Hopkins Bloomberg School of Public Health, and others is to be published Thursday on the Web site of The Lancet, a medical journal.
For those who don’t recall, these are the same folks who, using an equivalent “methodology,” suggested that over 100,000 Iraqi Civilians were killed in the initial Coalition Invasion. These claims at the time of their earlier report were widely discredited, and their methods rightly derided.
Here’s how their survey is described by AP:
For Burnham's study, researchers gathered data from a sample of 1,849 Iraqi households with a total of 12,801 residents from late May to early July. That sample was used to extrapolate the total figure. The estimate deals with deaths up to July.
The survey participants attributed about 31 percent of violent deaths to coalition forces.
You don't have to be an "expert" in social scientific "method" to recognize crap when you see it.
Much like polls in general, anything based on anecdotal evidence is going to be hopelessly biased and potentially orders of magnitude from reality. Even if we take these researchers at their word that they "checked 92%" of death records, how did they ensure they didn't double count? Did they keep a copy and reconcile no dupes? In a tribal community, many "families" would claim the same family member as "one of their own."
I remember clearly when the earlier report came out from these researchers. Then, it was clear that any “insurgent” who managed to die away from the location of combat would almost surely be counted as a “civilian” casualty, as the Al Qaeda in
The same objection applies to this new report. Death certificates are no doubt completed in many cases without adequate or sufficient information to know whether those deaths were the result of combat or not, or whether the corpse was that of a combatant.
And more on the sample: how did they statistically ensure that their neighborhoods, streets, blocks, cities were a good sample? Methinks the answers sought dictated the scope for questioning.
Surveys often breed a response whereby the survey subjects steer their answers towards subtle survey or survey taker biases. They get social "credit" and approval from providing information.
One last point on methods. In Social Science in particular, statistical extrapolations are notoriously unreliable. Results need to be checked for validity (call it the sniff test).
As this post suggests, the idea that more than 700 (out of 770) deaths go unreported daily -- in an atmosphere where reporters and their sources are rewarded for high body counts -- is unbelievable on its face.
That these boobs from Johns Hopkins retain zero credulity for the magnitude of disconnect between physical evidence and what anecdotal "data" they've gathered says far more about their own biases, rather than denial on the part of us skeptics.
Forget the politics, that’s almost beside the point, however partisan the timing. As with the earlier report that came out just before US Presidential elections in November 2004, this one has been carefully timed to align with a
Republicans no doubt have their share of helpful partisans, too, but even more, they have the realities of continual reminders of emerging and growing threats to US National Security, and thus are less reliant upon manufactured news and scandals.
Unless you believe, as some do, that the evil genius Karl Rove is behind such events as Al Qaeda attacks and North Korean nuclear provocations.
My colleagues at Milblogs are all over this, of course:
Links to this post:
Subscribe to Posts [Atom]