Credbility gap

It's only been a few days, but already the Lancet study of excess deaths in Iraq has faded from the headlines. Even NPR seems to have decided that further analysis is not worthy of interrupting this week's pledge drive pleas. Which is a pity, because this is the sort of thing that should decide elections.

Almost as depressing as the media's offensive diminution of the story's import was the bizarre juxtaposition of George W. Bush's reaction to the study with his support of embattled House Speaker Dennis Hastert. Both the media and the presidential handling of the issue betray a dismal level of respect for science in America.

On the same day that Bush dismissed as "not very credible" the findings of a team of respected experts in statistical analysis of epidemiology that the occupation of Iraq has been responsible for at least 400,000 and possibly as many as 940,000 deaths, the commander in chief described a man who turned a blind eye to sexual predation by a fellow member of his Republican caucus as "very credible."

Ah. As Douglas Adams might have said, "a use of the word 'credible' with which I was not previously familiar."

Let's face it. After much gnashing of teeth, feigned surprise and outrage, the verdict is in. The Lancet study may not be perfect, but it has withstood scrutiny from the toughest critics. My fellow science bloggers, who are more learned in the ways of reviewing such things, have just about unanimously offered their support to the basic thrust of the study's conclusion, specifically that the "coalition" occupation of Iraq since 2003 has made things worse for the average Iraqi. Much much worse. Tim Lambert, as usual, is worth reading. One of the best analyses that I have found elsewhere comes from Daniel Davies at the Guardian.

But I don't think you need to be an expert to understand the study, which is relatively easy to read. The fact is, even if the lowest end of their estimate (just a tad under 400,000) is off by a factor of 2, we're still talking about several times more excess deaths that the Bush administration is willing to admit to. Not that I'm suggesting it is off by 100 percent. Anything but. As Davies notes:

This is the question to always keep at the front of your mind when arguments are being slung around (and it is the general question one should always be thinking of when people talk statistics). How Would One Get This Sample, If The Facts Were Not This Way? There is really only one answer - that the study was fraudulent.

So is Bush charging the authors of the study -- namely Gilbert Burnham, Riyadh Lafta, Shannon Doocy, Les Roberts -- with fraud?

That a man who promotes the teaching of intelligent design in the classroom and the notion that the jury is still out on anthropogenic climate change would dismiss a solid piece of science as "not credible" suggests a profound lack of respect for anything remotely approaching the scientific method.

I can find no record of any member of the White House press corp pressing Bush for a justification for his dismisal of the Lancet study. No one has asked him why he didn't think it was credible. I recognize that this is not news. We all know about the contempt Bush has for those who hold ideas that contradict his own. What I find less easy to digest is the media's apparent lack of interest in the implications of the study.

I'm sorry folks, but simply reporting what the president says isn't enough. Especially when what he says is so unsupportable. Haven't the last five years taught us anything about taking presidents at their word? Isn't this study a critical finding with respect to American foreign policy?

Which brings me to the subject of my next post, a subject that was briefly all the rage at ScienceBlogs. Until the Lancet study, that is: the state of science journalism. It may take a little while for me to gather my thoughts on that one, though, as I'm one of the few bloggers here who actually does engage in science journalism for a living.