Another thought-provoking article on the Johns Hopkins University Iraqi death toll figures (as reported in the Lancet) turned up in my Slate feed today. This time written by Fred Kaplan.
Kaplan gets much more into a detailed criticism of the statistical techniques used. In particular, he argues that the sampling methodology is problematic: "In statistically proper random surveys, each household has the same probability of being chosen. Yet in the Lancet survey, if a household wasn't on or near a main road, it had zero chance of being chosen. And "cluster samples" cannot be seen as representative of the entire population unless they are chosen randomly."
Combine this with the criticisms offered by Hitchens discussed here yesterday and there is little doubt that the death toll figure arrived at is inaccurate at best.
(via Slate [and for those following our affair, there will be an Autumn wedding])
Previously: Hitchens on the Iraqi death toll.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment