Statistics for Dummies

Greg Easterbrook:

I suspect one reason the Iraq death toll elicits so little concern is that exaggerated estimates exist. Americans can say of the exaggerated estimates, "Oh, that's way too high" and skip over thinking about the more probable numbers. The latest silly estimate comes from a new study in the British medical journal Lancet, which absurdly estimates that since March 2003 exactly 654,965 Iraqis have died as a consequence of American action. The study uses extremely loose methods of estimation, including attributing about half its total to "unknown causes." The study also commits the logical offense of multiplying a series of estimates, then treating the result as precise. White House officials have dismissed the Lancet study, and they should. It's gibberish.

Well, no. That is not it at all. The authors used a statistical method that, as they perfectly well knew, doesn't generate especially precise results. That's why when they calculated the confidence interval for their estimate it turned out to be rather wide. The 654,965 number is the middle point of the confidence interval. The true number could very easily be thousands higher or lower than that, but the true number is extremely likely to fall somewhere within the band they laid out. This isn't hard stuff and it certainly isn't gibberish.