Daran does an effective job demolishing the Slate response to the Lancet survey. Then states:
But the true drawdropping irony in all of this is that Adam, apparently with a straight face, should cite this claptrap in a post exhorting the rest of us to higher standards of evidence!
I happen to believe that what occurred supports the arguments I made in that post.
I stated that unless two parties entering a debate, or one party presenting a thesis, begin by setting a standard to judge their conclusions by, the information being spewed out would essentially be without meaning.
The link to the Slate article was a foolish idea I did without much thought–I merely wanted to demonstrate that reasonable disagreement could occur on one specific subject, that being the number of casualties in Iraq, and wanted to use the debate over the Lancet article as an example. I remembered the Slate article from back when I first read it, and linked to it. I had no intention of getting into a debate on that particular subject.
Well, if I was going to link to it, then I had to be prepared to defend it–and Ampersand began by setting the standard at the level of statistical theory. Unless I could understand the statistical methodology behind the Lancet study as it was used under any circumstance, then I had little foundation to criticize it.
In response, I read a fair amount on the basic statistics surrounding the subject, as well as the article itself and some peer-reviewed criticism. I admitted that the Slate criticism had, as far as I could tell, no basis in mathematical analysis. Ampersand and I debated further, and I could see that I would need to learn a lot more before I could match his knowledge on the subject.
So I took down the Slate article from the original post, and used instead Amp and my debate as an example of disagreement on the subject.
This is precisely the kind of exchange that I was talking about in the post. It isn’t about being right–science, and all knowledge, is about trial and error. The important thing is setting methodological standards that help you judge when you are in error.
Ampersand didn’t just argue “you are wrong, Slate is politically motivated”, he pointed out to me that the Lancet survey used a fair amount of standard statistical procedures, and stated that I would need to demonstrate how the study had been in error within that framework if I wanted to refute their results.
Setting out to do this, it didn’t take long to see that I simply hadn’t the knowledge. But I made an attempt to the best of my ability, and Ampersand’s criticism of that second attempt will help serve as the basis for further research.
I’m not entirely certain why it is that Daran and Brutus have reacted with such outrage to my call for sterner methodological scrutiny. All I can say is that, once applied, I find it to make debates fulfilling in a way that they are not when standards are set low and things devolve into ad-hominem.