A Real Inconvenient Truth about the Oreskes Affair

[excerpts]

 HYPERLINK "http://www.ericdsnider.com/blog/" \o "Return to Blog" Eric D. Snider's Blog

http://www.ericdsnider.com/blog/2006/07/07/absolutely-the-last-inconvenient-truth-blog-entry-ever/

Diligent, long-suffering readers of this blog will recall that one of the strengths of Al Gore’s global warming documentary  HYPERLINK "http://www.ericdsnider.com/movies/an-inconvenient-truth" “An Inconvenient Truth” is a study he cites that supposedly demonstrates how the vast majority of scientists are in agreement about global warming, and that the debate should therefore be over. 

The study was by Nancy Oreskes. According to Gore, she randomly chose 928 global warming-related articles published in science journals between 1993 and 2003 and found that ALL of them supported the majority view — i.e., Gore’s position, that global warming is real, bad and preventable.

I said in a  HYPERLINK "http://www.ericdsnider.com/blog/2006/06/18/a-summary-of-inconvenient-truth-e-mails-and-responses/" previous blog entry that I have no choice but accept that study as legitimate. I don’t have access to all the science journals, nor the resources to duplicate the study. I said someone who did have the resources was probably already working on either refuting or supporting it.

And I was right! And it turns out Oreskes’ study — and thus Gore’s support of it in his film — was deeply flawed.

On the supposed “scientific consensus, Dr. Naomi Oreskes, of the University of California, San Diego, did not examine a “large random sample” of scientific articles. She got her search terms wrong and thought she was looking at all the articles when in fact she was looking at only 928 out of about 12,000 articles on “climate change.” Dr. Benny Peiser, of Liverpool John Moores University in England, was unable to replicate her study. He says, “As I have stressed repeatedly, the whole data set includes only 13 abstracts (~1%) that explicitly endorse what Oreskes has called the ‘consensus view.’ In fact, the vast majority of abstracts does (sic) not mention anthropogenic climate change. Moreover — and despite attempts to deny this fact — a handful of abstracts actually questions the view that human activities are the main driving force of ‘the observed warming over the last 50 years.’ 

The way it went down was,  HYPERLINK "http://www.sciencemag.org/cgi/content/full/306/5702/1686?maxtoshow=&HITS=10&hits=10&RESULTFORMAT=&fulltext=oreskes&searchid=1&FIRSTINDEX=0&resourcetype=HWCIT" Oreskes’ study was published Science Magazine on Dec. 3, 2004. In it, Oreskes said she had done an  HYPERLINK "http://www.isiwebofknowledge.com/" Institute for Scientific Information (ISI) database search for the keywords “climate change” and had come up with 928 abstracts published between 1993-2003, and that not one of them rejected the consensus position. 

When Dr. Benny Peiser did the same search, however, he came up with some 12,000 papers, not 928. Confronted with this information, Oreskes confirmed she’d screwed up: She hadn’t searched for the keywords “climate change,” as her article said, but for “global climate change.” That search brings up only 1,247 documents. (Where she got the number 928, who knows.)

Seeing that Oreskes’ study was flawed from the get-go, Peiser did his own. He used the same keywords — “global climate change” — for 1993-2003 and came up with 1,247 documents, as just mentioned. Only 1,117 of those had abstracts (you know, the paragraph that summarizes the whole paper). He analyzed those 1,117 abstracts and found that only 13 explicitly endorse the consensus view; 322 implicitly accept it but focus on other aspects; 44 focus on natural factors of global climate change; and 34 reject or doubt the Al Gore view altogether. (Oh, and 470 of the 1,117 articles include the keywords “global,” “climate” and “change” but don’t actually have anything to do with the matter at hand.)

Science Magazine ran a brief correction a few weeks later, but refused to published Peiser’s more detailed study on the grounds that the information he was presenting was already widely disseminated on the Internet. (In other words: “Yeah, we screwed up when we ran Oreskes’ article. Quit rubbing it in.”)

Peiser recounts the whole thing, including his exchanges with Science Magazine,  HYPERLINK "http://www.staff.livjm.ac.uk/spsbpeis/Scienceletter.htm" here. It’s good reading. 

Several readers also brought to my attention an article from the  HYPERLINK "http://www.opinionjournal.com/extra/?id=110008597" Wall Street Journal that argues with some more of Gore’s points. This article also mentions the Oreskes/Peiser studies. 

The inaccuracy of the Oreskes study hurts part of Gore’s case: the part where he says scientists all more or less agree with him. In truth, while there is a majority opinion (MOST scientists seem to be onboard with it), it’s far from being an overwhelming consensus.