A new analysis of 1372 climate scientists who have participated in major climate science reviews or have signed statements in support or opposition to their main conclusions confirms what many researchers have said for years: Those who believe in anthropogenic climate change rank much higher on the scientific pecking order than do those who take issue with the idea.
The paper shows that "the vast majority of working [climate] research scientists are in agreement" on climate change, says climate science historian Naomi Oreskes  of the University of California, San Diego. "Those who don't agree, are, unfortunately—and this is hard to say without sounding elitist--mostly either not actually climate researchers or not very productive researchers."
But the paper, published today  in the Proceedings of the National Academy of Sciences, faces criticism on three fronts: how it divides scientists into one of two groups, whether the scientists have been chosen properly, and whether the peer review process stacks the deck in favor of the consensus view. "This is a completely unconvincing analysis," says climate expert Judith Curry of the Georgia Institute of Technology, who was included in neither group.
The co-authors, led by graduate student Bill Anderegg of Stanford University, tapped online lists of scientists  who have signed statements (like this , this , or this ) in support or opposition to the main findings of the Intergovernmental Panel on Climate Change (IPCC), namely, that the planet is warming and humans are largely responsible. They categorized the scientists as either "convinced" or "unconvinced" and then analyzed the groups for the number of papers they had published which included the word "climate" in a Google Scholar search . "Unconvinced" scientists comprised only 2% of the top 50 researchers ranked by number of climate publications and 3% of the top 100. Among scientists with 20 or more papers on climate, the so-called convinced group had an average of 172 citations for their top paper compared with 105 for the unconvinced.
The first critique of the paper is that the grouping of researchers into "unconvinced" and "convinced" fails to capture the nuances of scientific views on the subject. "By putting scientists into two categories which do not reflect the subtleties of the debate, ... this paper simply reinforces the pathological politicization of climate science in policy debate," says Roger Pielke Jr. of the University of Colorado, Boulder. His father, Roger Pielke Sr., for example, was among the most prominent and cited of the "unconvinced." But in an e-mail to ScienceInsider, the elder Pielke says that although greenhouse gas emissions are important to consider, so are land-use changes, black carbon and aerosol pollution—a position perhaps more nuanced than the convinced/unconvinced dichotomy the paper postulates.
In addition, the paper defines as "unconvinced" someone who signed a paper "arguing against any need for immediate cuts to greenhouse gas emissions." Pielke Jr. takes exception to that definition. "So you are a "climate skeptic" if you have a certain view on climate policy? Bizarre," he wrote in an e-mail.
"It would be helpful to have lukewarm [as] a third category," says Jim Prall, a computer support professional at the University of Toronto in Canada who was one of the four authors on the paper. But Prall believes that the paper makes a valuable contribution by allowing the public to measure the scientific prominence of researchers who identify with certain views.
Another area of controversy is the authors' selection of scientists to study. The paper focuses on only scientists who have either participated in the IPCC or signed public statements on the state of the science. Are those the right 1372 scientists to analyze? Scientists who are "unconvinced," for example, may feel peer pressure not to publish public statements on that view, which might dilute the strength of that view over all. Its decision to include all IPCC contributing authors  as "convinced" by that document's main conclusions is also debatable.
Finally, does peer review affect a scientist's ability to contribute to the field? John Christy of University of Alabama, Huntsville, another of the prominent "unconvinced" scientists analyzed in the study, blames the disparity between the two groups on "the tight interdependency between funding, reviewers, popularity. ... We are being "black‑listed," as best I can tell, by our colleagues."
Pat Michaels, a well-known climate skeptic, says in an e-mail that the paper's conclusions are "a self-fulfilling prophecy." He notes that "I have three manuscripts that have been out for nearly two years. There's nothing fundamentally wrong with them except they indicate that warming will be at the low end of the frequency distribution given by [a middle-of-the-road IPCC greenhouse emissions scenario]. Every time we answer reviewers, the editor then sends it out to someone else to cook up another complaint. I gave up on one on Greenland ice that indeed models 2007 with a 50 percent probability of being the largest annual loss since at least the late 18th century."
Prall agrees that the system may not be perfect, but he thinks it's good enough. "It's conceivable that some people have formed a fixed point of view," he says. "But the editors of journals, if they have formed a resistance to outside points of view, they have done so after years of seeing all the good, bad, and in-between papers. They know the field better than anyone else."