The MoonLanding paper is finally here. Eight months after Lewandowsky was so sure he had a “peer reviewed” conclusion that he announced his results in The Guardian and The Telegraph , the paper has finally been published.
Lewandowsky et al claimed to show skeptics are nutters who believe any rabid conspiracy like the “moon-landing was faked”. Their novel method for discovering the views of skeptics involved surveying sites frequented by those who hate skeptics.
The survey questions included conspiracies likely to appeal to a small percentage of conservative or free market thinkers, and largely left out conspiracies that would appeal more to supporters of bigger government (like the idea that the rise of “climate denial” was a big-oil funded conspiracy). It studied big-government conspiracies and ignored big-corporate ones. There are gullible conspiracists who also believe in global warming, but there was no danger this survey would find them. The survey bias was so obvious, even alarmist commenters said they feared few “denialists” would take it. The results that were headlined in newspapers were based on a tiny sample of ten respondents to an anonymous online survey. Not surprisingly Lewandowsky’s university (UWA) received many complaints about ethics, methods, and the dismal quality of the data, and bloggers had a field day shredding the paper.
In response, the Australian Research Council awarded Lewandowsky et al another $338,000. Just where do their priorities lie?
The paper was delayed. The typesetting oddly took 8 months, and includes a new key point. To answer the rabid critics, Lewandowsky needed to show that that many real skeptics did fill out the survey. The evidence for that apparently relies on Cook’s site (the ambush-labelled “Skepticalscience”). Lewandowky et al now effectively claims skeptics really were reading Cook’s site and lots of them did the survey there.
How 78,000 equals zero
Lewandowsky et al go out on a limb to say skeptics may have made 78,000 visits that month and could have seen that survey link (if only there had been one there):
Prevalence of “skeptics” among blog visitors All of the blogs that carried the link to the survey broadly endorsed the scientific consensus on climate change (see Table S1). As evidenced by the comment streams, however, their readership was broad and encompassed a wide range of view on climate change. To illustrate, a content analysis of 1067 comments from unique visitors to http://www.skepticalscience.com, conducted by the proprietor of the blog, revealed that around 20% (N = 222) held clearly “skeptical” views, with the remainder (N = 845) endorsing the scientic consensus. At the time the research was conducted (September 2010), http://www.skepticalscience.com received 390,000 monthly visits. Extrapolating from the content analysis of the comments, this translates into up to 78,000 visits from “skeptics” at the time when the survey was open (although it cannot be ascertained how many of the visitors actually saw the link.)
For comparison, a survey of the U. S. public in June 2010 pegged the proportion of “skeptics” in the population at 18% (Leiserowitz, Maibach, Roser-Renouf, & Smith, 2011). Comparable surveys in other countries (e. g., Australia; Leviston & Walker, 2010) yielded similar estimates for the same time period. The proportion of “skeptics” who comment at http://www.skepticalscience.com is thus roughly commensurate with their proportion in the population at large.
But, as Barry Woods and DHG both discovered and Geoff Chambers pursued relentlessly, it appears no link was ever posted on SkepticalScience. Steve McIntyre points out the total number of skeptics doing the survey on Cook’s site can thus be ascertained — and it is exactly zero. It’s hard to believe any scientist would think they would get away with this. The paper will surely have to be withdrawn. (But will the government funding be withdrawn? The Australian Research Council (ARC) needs to answer some very awkward questions.)
Here’s a study in dishonesty
FOI documents obtained by Simon Turnill show that Cook didn’t post the link at all, he tweeted it (how many people were on his twitter list in 2010?). You might think this is a minor change, but not at all. Cook’s tweet probably didn’t get the “390,000 visits” his site might have got that month. Instead the records show he got five retweets.
More importantly, above all else, Cook has been untruthful with his readers and with skeptics all along.
Like a five year old explaining where the cookies went, his story keeps changing as Woods and especially Chambers pin him down. Back in September last year, Lewandowsky said eight sites hosted the survey and Cook’s site was one, but strangely no one could find a link on SkepticalScience. Cook then said he hosted it in 2011. But it wasn’t there either. Cook apologized, and said it was 2010. When Chambers pressed on for evidence, Cook and the moderators stopped the questions and insisted it go private. Cook emailed Chambers, saying he couldn’t add much more. Chambers wanted the comments from the discussion of the survey at Cook’s site because the comments on other sites where the survey were hosted revealed the thoughts of the people most likely to have done the survey. Strangely the post and all the comments had been deleted. [Read the whole exchange at Climate Audit].
It’s odd for a whole post to disappear, but even odder when the post is then quoted in a peer reviewed study. Cook repeated that he had provided a link to the survey, but Chambers could find no record of it in the Wayback Machine either. Cook then suggested that he had “forensic evidence” — an email from Lewandowsky asking Cook to post the link, and Cook’s reply that he posted it the same day. What Cook wasn’t saying but must have known (since he checked his email records), was that his emails forensically showed that Cook tweeted the link.
The tweet by Cook, that Barry Woods and DHG both found, is online here on August 27, 2010. The Wayback Machine recorded Skepticalscience on August 30, 2010 — and there was no post. Nor was there a post about a survey in the following week either. So it appears unequivocal, as much as anyone bar John Cook can tell, that SkepticalScience did not host the survey.
McIntyre asked Cook to explain:
Not to put too fine a point on it, it appears to me that you lied, when you asserted that your correspondence with Lewandowsky on 28 August 2010 was “forensic evidence” that showed that you had posted a link at Skeptical Science to the Lewandowsky survey on that day. I use the word “lie” because you had clearly examined the 28 August 2010 correspondence at the time of your email to Chambers and knew that this correspondence did not show that you had posted a link at Skeptical Science to the Lewandowsky survey on 28 August 2010 or any other day.
Before I make any public statements about this matter, I am offering you an opportunity to rebut the belief that the statement bolded above was a lie.
Cook did not respond.
If Cook did post that survey long enough for anyone to comment on it, he would probably have a record of the comments made even after deleting the post. Where are they?
The “smart” thing for Cook to have done at this point was to make sure the published paper was corrected to “seven sites” and not eight, and did not rely on the link at his site which no one can find. He and Lewandowsky both must have been aware that skeptics knew. It speaks volumes that Lewandowsky and Cook cited a post on Climate Audit (Steve McIntryre’s site) about this point in their second “conspiracy paper” called Recursive Fury, so they knew that they were wrong, and that skeptics knew that too. Despite this, they still went ahead and published the revamped edition of the Moon Landing paper with the misleading information in it.
To the skeptics who suggest that Lewandowsky or Cook planned the first paper to fish for comments to use in a second paper, I say not a chance. They just aren’t that competent. They struggle from one gaffe to the next.
As Skiphil notes, Lewandowsky expects skeptics to keep two year old emails from unknown assistants at a university they may never have heard of, but he doesn’t expect his own co-authors to keep posts up displaying supposedly “scientific” surveys with results that are used in his papers.
Their incompetence will hurt the reputation of Psychological Science and Frontiers, if they do not take quick action, as well as the University of Western Australia, the School of Psychology, the University of Queensland and the ARC.
The strategy here appears to be double or nothing.