John Cook’s 97% Consensus Claim Is About To Go ‘Pear-Shaped’

  • Date: 10/05/14
  • Anthony Watts, Watts

Dr. Richard Tol writes about a new revelation coming from an analysis of Cook’s climate publications volunteer raters, conducted by Brandon Schollenberger:

My comment on Cook’s consensus paper has at last been accepted. It was rejected by three journals — twice by Environmental Research Letters and once by two other journals for being out of scope. Fifth time lucky.

As these things go, my comment is out of date before it is published.

One of my main concerns was the partial release of data. The data that was available suggests that all sorts of weird things were going on, but without the full data it was hard to pinpoint what went on. Cook’s resistance to release the data, abetted by the editor, the publisher and the University of Queensland, suggested that he may have something to hide.

Brandon Shollenberger has now found part of the missing data.

Unfortunately, time stamps are still missing. These would allow us to check whether fatigue may have affected the raters, and whether all raters were indeed human.

Rater IDs are available now. I hope Shollenberger will release the data in good time. For now, we have to do with his tests and graphs.

His comment of May 10, 1:16 am shows that individual raters systematically differed in their assessment of the literature. This is illustrated by this figure; the circles are aligned if the raters are the same.

This undermines Cook’s paper. Theirs was not a survey of the literature. Rather, it was a survey of the raters.

Analysis of raters in Cook’s 97% paper by Schollenberger

Full story