I need to rephrase my last post. I din't do a very good job with it.

Tony Blair said that the "Figures from the Iraqi Ministry of Health, which are a survey from the hospitals there, are in our view the most accurate survey there is", and the figures from the Iraqi Ministry of Health supports the Lancet survey's conclusion that most of the civilian deaths are the result of violence from coalition forces, rather than the actual numbers of civilian deaths.

The Lancet survey was based on 4300 people, not fewer than 100, GUEST. Get a grip yourself.

The Lancet people also left the region around Fallujah out of their final report. I put that number back in for the purpose of this discussion.

4300 seems like a pretty normal sample size to me. Researchers use sample sizes like that all the time. Many aspects of your daily life are determined by sample sizes like that one.