The Past is Prologue

Cook et al., searched the ISI web of science between 1991 and 2011 for articles which were returned following a keyword search on "global warming" and "global climate change".  A team of reviewers then categorized the ABSTRACTS.  Back in 2008, Brian organized a similar exercise for a smaller number of abstracts, about 650 published between 2002 and 2006 which will be designated below as the prequel.  A team of six reviewers then categorized the ABSTRACTS with two reviewers looking at each abstract.

As a first step it is useful to compare the categories which were independently settled on by both teams.  Until about a week ago, neither team was aware of the others work, so these were indeed independent exercises..

Prequel
Explicit Endorsement of AGW	Explicit Endorsement of AGW
Implicit Endorsement	Implicit Endorsement
No Position	Discussion
Implicit Rejection	Methods
Explicit Rejection	Paleoclimate
	Unrelated
	Undecided
	Implicit Rejection
	Explicit Rejection

The primary difference is that the Prequel Team had more categories in the middle to deal with the predictable muddle.

An important point that has been missed is that MOST, essentially the vast majority of the papers, tossed up by such a search are biologically, ecologically or economically oriented.  Relatively few are what the Bunnies would call WGI material.  Most are WGII and WGIII.  Go ahead try it in Google Scholar.  Such papers, like WGII and WGII one way or another take the conclusions of WGI (that there is human caused climate change) as their starting point and we had to figure out how to deal with that

The prequel of the prequel was a small training set selected from the ~650 from which rating instructions were developed and then each member of the team rated about 200 abstracts.  The results were

Prequel	#
Explicit Endorsement of AGW	244	21%
Implicit Endorsement	407	34%
Discussion	222	19%
Methods	78	7%
Paleoclimate	107	9%
Unrelated	43	4%
Undecided	33	3%
Implicit Rejection	21	2%
Explicit Rejection	28	2%

Cross-Tabs tomorrow, but in almost all the cases the two reviewers were close to each other lumping the five neutral ratings as one.  There were only 13 problem papers where there was significant disagreement (support/reject) between the two reviewers, at which point everyone ran out of steam before trying to to reconcile them by actually looking at the papers.

Why bring this up now.  Well there have been a bunch of surveys, this being an unpublished one, and they all are within a stone's throw of each other, the "doubting" literature is really thin.

Congratulations to John, Dana, and the rest of the authors and helpers from Skeptical Science.  The amount of work they did is astounding.  Eli knows.  Brian is still in the recovery room.