Tim Lambert (deltoidblog AT gmail.com) is a computer scientist at the University of New South Wales.
Gilbert Burnham has just given a talk at MIT on the Lancet studies on deaths in Iraq. You can watch the video here.
Some of things he mentioned:
- USAID (which has expertise in cluster sampling) was told to look for holes in the study, but couldn't find any.
- They will soon release the data (with identifying material removed) to other researchers.
- John Howard's idiotic comment "it's not based on anything other than a house-to-house survey" got a laugh.
- The IBC made vociferous attacks on the studies because they want to defend their methods, and Les Roberts suggests that IBC are trying to stop the donations from drying up.
Hat tip: Stephen Soldz, who attended and asked questions.
Comments
Pretty straightforward presentation. He gave some evidence that suggests MSB, if it exists, is small. Explained why public release has been delayed (an explanation I didn't quite buy). Was pretty gentle to both IBC and Pederson.
Posted by: Robert | March 1, 2007 7:46 AM
And then there's Roger Alvin Pielke Jr....As usual, he's spinning the science and contorting himself into bizarre shapes, like a child playing at shadow puppets.
New research has found a link between climate change and hurricanes.
http://sciencepolicy.colorado.edu/prometheus/archives/sciencepolitics/001131spinningscience.html
Of course, to Roger it's just misrepresentation by--get this--not only the University of Wisconsin, but also the National Science Foundation.
Just ridiculous. When is finally going to give it all up and join the Cato Institute?
Posted by: Thom | March 1, 2007 8:20 AM
Excellent lecture... One answer he gave troubled me. Asked the reasons why modern wars are much more deadly towards civilians than soldiers, he gave as the probable cause the use of air power. However, he also earlier alluded to the Congo civil war where over two millions of people died, the vast majority civilians, yet neither side used air power in a major way. In Iraq, the insurgents don't have aircraft, but are the biggest killers. Prof Burnham's own data gives gunshots as the major cuase of violent death, not aircraft. But if its not aircraft, then why are civilians dying in wars in such large numbers? For example, in World War I, only 14% of the 17 million dead were civilian, in the Congo it was 90% of two million. No, there more to this than Prof Burnham knows.
Posted by: Toby Joyce | March 2, 2007 2:08 PM
Toby, the figures for WWI seem to be roughly 50% civilian. One obvious difference from Iraq was that much of the fighting in WWI consisted entirely of large formations of soldiers concentrated on attacking one another. And after the initial troop movements, these formations settled into frontlines that hardly moved.
No, there's more to this than Prof Burnham knows.
Trivially true, yet tendentious.
Posted by: Vance Maverick | March 13, 2007 2:17 AM