WendyMcElroy.com

Much has been written about the recent Lewandowsky paper, at The Blackboard, Jo Nova, Climate Audit, and Watts Up With That, among others. This may stand as the most badly designed and executed survey ever ("a landmark of junk science" as Steve McIntyre put it).

Lewandowsky was hoping to show a linkage between climate science skepticism and a belief in (a) free markets, and (b) conspiracy theories. Even a college freshman should be able to identify the problems. First, it was a Self-Selected Opinion Poll...and these are known as SLOPs for a good reason; they're almost guaranteed to have a biased result. Second, a survey purportedly to capture the views of skeptics was posted only at anti-skeptic sites. Third, as Steve McIntyre observes, there was no control for fake responses...or for multiple responses. Multiple versions of the survey, with different questions, were circulated; and the survey results were being discussed on blogs while the survey was still underway. Some interesting "conspiracy theory" questions were omitted from the report without explanation; others were never asked (such as, "do you believe that climate skeptics are funded by a cabal of oil companies?")

On top of all that, the questions themselves were awful. I participated in an attempt to replicate the survey, and I was simply unable to answer some questions. Some make multiple assertions or embody multiple assumptions, such that one can neither agree nor disagree. For example, "The problem of acid rain is no longer a serious threat to the global ecosystem," agree or disagree. This assumes that acid rain is a problem and once was a "serious threat". There is no way to distinguish between respondents who never thought it was a problem, thought it was a problem but not a serious one, or thought it was a serious problem but has now been largely resolved.

Others leave important terms undefined. Consider "Free and unregulated markets pose important threats to sustainable development," agree or disagree. What does "sustainable development" mean? I think the free market is sustainable, because it adapts quickly to changing circumstances (such as resource availability).

Others are of the "have you stopped beating your wife" variety, where either answer is wrong. "I believe that burning fossil fuels increases atmospheric temperature to some measurable degree", agree or disagree. Now, I agree that burning fossil fuels adds CO2 to the atmosphere, and I agree that there is a "greenhouse effect." But from what I know of the magnitude of the effect and the accuracy of instrumentation used, I think the fossil-fuel contribution may be below the detectable threshold. What do I answer? Do I say "disagree", and have the surveyor conclude that I don't believe in the greenhouse effect? Or do I say "agree", and the surveyor conclude that I'm a warmist?

After all this, Lewandowsky had only 10 respondents out of 1300 respond that they thought the Apollo moon landing was faked -- and more of those respondents identified themselves as warmists than as skeptics! -- yet the paper was entitled "NASA faked the moon landing -- therefore (climate) science is a hoax: An anatomy of the motivated rejection of science." There's no way to draw conclusions from that statistically weak a sample. One might as easily have titled the paper, "NASA faked the moon landing -- therefore the world is burning up." Lewandowsky would have better advised to tout the correlation between free marketeers and skeptics*; but when you're an activist with an agenda and a weak knowledge of statistics, I guess sensationalism wins out.

P.S. The most astonishing thing was that this paper passed peer review, which I guess goes to show the quality of peer review in Lewandowsky's field.

* P.P.S. That correlation proves little, by the way. Most people who purchase chainsaws are men. That doesn't mean that most men own chainsaws.