This is ancient history, four years late in the writing, but never mind. History is long.
I’d never really bothered much with Cook’s 97% paper, but a comment here by JonA led me to skepticalscience’s Consensus Project, where you can play with Cook’s data to your heart’s content. This leads to their data file where they define their criteria for endorsement/rejection of AGW as a prelude to rebutting various criticisms of their paper, by Roy Spencer among others.
I took a quick look at Cook et al 2013 when the paper came out and saw what a logical dog’s breakfast  it was, and assumed that someone more qualified than I would sort it out. I know plenty of people have rubbished the survey, and Richard Tol even got a reply in at Environmental Research Letters, pointing out the methodological flaws and the irrelevance of consensus to science, but I don’t think anyone has analysed the flaw which is logically prior to all the problems with coder bias etc. It’s not that sceptical papers are being undercounted or miscounted or ignored. The problem with the criteria used for categorising papers is that they entirely fail to come to grips with what sceptics are saying, with the result that the paper isn’t about scepticism at all.
Here are the definitions of the criteria used to categorise the papers, as defined in the paper itself:
Table 2. Definitions of each level of endorsement of AGW
1) Explicit endorsement with quantification – Explicitly states that humans are the primary cause of recent global warming
2) Explicit endorsement without quantification – Explicitly states humans are causing global warming or refers to anthropogenic global warming/climate change as a known fact
3) Implicit endorsement – Implies humans are causing global warming. E.g., research assumes greenhouse gas emissions cause warming without explicitly stating humans are the cause
4a) No position – Does not address or mention the cause of global warming
4b) Uncertain – Expresses position that human’s role on recent global warming is uncertain/undefined
5) Implicit rejection – Implies humans have had a minimal impact on global warming without saying so explicitly E.g., proposing a natural mechanism is the main cause of global warming
6) Explicit rejection without quantification – Explicitly minimizes or rejects that humans are causing global warming
7) Explicit rejection with quantification – Explicitly states that humans are causing less than half of global warming
While the revised definitions at SkepticalScience are as follows:
1,Explicitly endorses and quantifies AGW as 50+%
2,Explicitly endorses but does not quantify or minimise
3,Implicitly endorses AGW without minimising it
5,Implicitly minimises/rejects AGW
6,Explicitly minimises/rejects AGW but does not quantify
7,Explicitly minimises/rejects AGW as less than 50%
You can see what Cook has done here. By combining his categories 4a and 4b, he has simply got rid of those who express the “position that human’s role on recent global warming is uncertain/undefined.” Since his categories 4 a) and b) accounted for 66% of his total sample, and the disappeared category 4b) is the clearest possible definition of the standard sceptical position, by quietly counting it among the “No Positions”, he has simply abolished climate scepticism. According to the paper itself, it’s perfectly possible that a majority of papers were sceptical.
But it’s worse than that. Going back to the extended definitions in the paper, it’s possible for a sceptic to find himself in any one of the categories. Category 1 may seem a stretch, but why not, after all? It’s perfectly possible to accept that humans are the primary cause of recent global warming and say: “So what? Recent global warming is minimal, not accelerating, and probably beneficial. So why worry?” That’s the position of Matt Ridley, for example, who is regularly pilloried as a climate denier.
Personally I’m happy with criterion 2) that “humans are causing global warming” and that “anthropogenic global warming/climate change [is] a known fact,” but I’m equally at home in categories 4b) 5) 6) or 7). What this demonstrates is either: that I’m incapable of rational thought, as explicitly stated in “Recursive Fury” (Lewandowsky, Cook et al. 2013: Table 2) or that the categories are not mutually exclusive and that Cook is incapable of coding data.
There is conclusive evidence in favour of the second hypothesis.
At first sight Cook’s classification of abstracts seems to divide them by three binary criteria:
explicitly v implicitly
endorses v minimises/rejects
quantifies v does not quantify
except that categories 2) and 3) explicitly allow the possibility of minimising without rejecting. So we have at least a 2x3x2 matrix, resulting in twelve cells, while Cook’s bouillabaisse has seven and a half.
What any first year social science student would have done is code the three criteria separately, deciding for each article: whether it took a position explicitly or implicitly; whether it endorsed, minimised, or rejected; and whether it quantified or didn’t; each time with a “don’t know” category. And let the computer programme do the rest.
[That’s what I did when I took a holiday job as a coder in market research as a student, and what I continued to do up to the time I got a medal from the Market Research Society for research for the Ministry of Defence on officer recruitment advertising. Then I had a Marxist revelation of the Groucho kind and decided I didn’t want to belong to a profession which… but I digress.]
Cook’s categories are neither exhaustive nor mutually exclusive. It s perfectly possible to belong in several of them at the same time. No wonder that two thirds of the abstracts were shoved into the “don’t know/doesn’t mention” category.
Cook et al.’s 97% is nonsense, corrupt pseudo-science, passed through the gastro-intestinal tract of peer-reviewed scientific publication to emerge as one of the most hallowed turds in the climate corpus. The 97% is a fantasy figure invented by Cook and his mentor Lewandowsky, a conspiracy to label a large proportion of the human race, including all those who criticise Cook et al., as non-persons. Cook and Lewandowsky are liars and charlatans. Cook is also an idiot, a bumbling incompetent who doesn’t know his Rsquared from his elbow, promoted to a position of extraordinary influence by an academic system which exhibits all the symptoms of senile decay.
But (Cook et al. 2013) is quoted almost every time anyone wants to assert that catastrophic man-made global warming is a fact. So it seems to be working rather well.
 Proverbs 26:11 As a dog returneth to his vomit, so a fool returneth to his folly.