/* Google Analytics Code asynchronous */

Monday, September 24, 2007

Cooked or not the error of the estimate may be wide

Putting one rumor to rest: there is no consideration given to the placement of an entry wound on a murder suspect's head in the tabulation of sectarian violence, ...SA, at TPM, reviewing the FOIA release of The Methodology

I'm not sure I'm copacetic with that.

Back in the spring of 2005, I compiled casualty figures using some of the sources that were supplying them from Iraq. Back then, I was interested in segregating what might be called a "crime rate" (i.e. violence related to ordinary criminal activity) from the nascent terror campaign (my worksheet says, "suicide", "car bomb", "roadside bomb"). [roundup of sources, here].

As the methodology from the U.S. Army notes, they rely on the Iraqi Health Ministry for reporting.

From what I recall, the doctors in the hospitals follow a process of elimination / categorization and make guesses after that. In some cases, they have information on the circumstances of the deaths, including family members' witness. In others, they have nothing. I am unaware of any coalition doctors or U.S. field personnel also in the morgues making independent assessments whatsoever (therefore, the Army appears to have a 'methodology' for something it is not doing or overseeing, which, of course, is not unheard of in the annals of Army bureaucracy).

Which raises the question, so long as "measured violence" may be different than "actual violence", ought not perhaps public conclusions about effectiveness be constrained by the statistical fog of war, too?