Weighing ‘convinced’ and ‘unconvinced’

As polls indicate that fewer Americans say they see solid evidence for global warming, and as climate change skeptics have grown emboldened in the wake of last fall’s hacked e-mail episode at the University of East Anglia, Stanford grad student Bill Anderegg pursued a few simple questions:

What makes someone who speaks or writes about the climate a climate expert?

Is there a difference in scientific expertise between those who subscribe to the view that humans are significantly driving up the globe’s thermostat and those who are skeptical or incredulous?

Anderegg, who studies biology at Stanford University under well-known climate researcher Stephen Schneider, teamed up with researchers at Stanford, the University of Toronto, and The William and Flora Hewlett Foundation to find out. His results are certain not to please all parties in the contentious climate change arena, including among those who fall into the category of being “convinced” by the scientific evidence.

The research, published in the respected “Proceedings of the National Academy of Sciences,” PNAS, argues that the differences between the two groups are striking. The primary conclusions: climate researchers who have published the most and been cited the most are nearly unanimous in their agreement with the latest conclusions of the IPCC. Researchers who are most skeptical of the IPCC’s conclusions, on the other hand, have published little and have been cited much less often in the scientific literature.

The study, analyzing the work of more than 1,300 climate researchers and entitled “Expert credibility in climate change,” is available as a PDF.

Bloggers Erupt, Not Solely Climate Skeptics

The study had scarcely gone public on June 21 before the blogosphere erupted with analyses and a fair number of criticisms. Most objections focused on its methodology but some expressed doubt that it is the sort of policy perspective the premier scientific journal of the National Academy of Sciences should be publishing in the first place. Not all the criticisms came from climate contrarians, clearly the group most targeted by the work.

For instance, Judith Curry, a climate scientist at Georgia Tech who supports the scientific evidence for human-caused warming but has called for more openness among climate scientists and engaged skeptics, criticized the PNAS study on a few fronts.

One primary flaw, she wrote in comments to a story on Collide-a-Scape, a blog by former Aububon editor Keith Kloor, is that many of the researchers grouped in the study’s database as supportive of mainstream climate science — specifically on the IPCC’s latest AR4 report — are not experts involved in the “detection” or “attribution” of human-caused warming.

“I think this is an interesting database … (b)ut I don’t think it was appropriately analyzed in the PNAS paper in context of ‘credibility,’ particularly in how the scientists were classified,” Curry wrote (her comment is the 198th listed in a highly energized comments section).

“How important is this? (W)ell, it is providing much entertaining fodder for the blogosphere,” commented Curry, who was not included in either camp that the PNAS study analyzed.

Spencer Weart, author of the highly regarded “The Discovery of Global Warming,” also expressed criticisms of the study.

Climate blogger Joe Romm at Climate Progress, meanwhile, not unexpectedly praised the “important first-of-its-kind study.”

“It could theoretically open the eyes of those in the status quo media who keep suggesting the ‘experts’ they cite that keep pushing anti-science disinformation are somehow close to being equal in number, credibility, or expertise to the broad community of climate scientists, thereby implying serious disagreements among mainstream scientists,” Romm wrote on his blog.

More reactions to the “Expert Credibility” report can be found here.

Key Study Findings … and Methodology

The PNAS paper presented a handful of key findings:

• 97-98 percent of the climate researchers most actively publishing in the field support the tenets of anthropogenic climate change as outlined by the IPCC.
• Researchers unconvinced by the evidence for anthropogenic climate change comprise only 3 percent of the top climate researchers. “Top” researchers are defined as the ones who have published the most and been cited the most by their peers.
• Among researchers unconvinced by the evidence for global warming, 80-percent had published fewer than 20 papers. Among those who are convinced, fewer than 10 percent had published fewer than 20 papers.

In a telephone interview, Anderegg said he and his colleagues had expected there to be differences in expertise between the two groups, “but we didn’t expect them to be as large.”

Why do the exercise at all? The charismatic and straight-talking Schneider, one of the study’s authors and himself a member of the National Academy of Sciences, wrote in an e-mail interview that the project was born from frustration — frustration over “claims by climate deniers dutifully reported in the media and repeated in the Congressional Record … that they are somehow as equally credible … as the mainstream and their scientific opinions deserve equal weight with the National Academy of Sciences assessment reports or the IPCC.”

In their study, Anderegg, Schneider and colleagues James Prall from the University of Toronto and Jacob Harold from the Hewlett Foundation compiled a database of 1,372 climate researchers.

‘Yes…No’ answers on climate convictions… or more nuanced?

They classified each researcher in one of two broad categories, a generalization that some commenters found overly simplistic: those convinced by the evidence for anthropogenic climate change, and those unconvinced by the evidence.

The first group included those researchers who had signed statements broadly agreeing with or directly endorsing the primary tenants of the IPCC’s Fourth Assessment Report, published in 2007. That is, that it is “very likely” that greenhouse gases generated by human activity were responsible for “most” of the “unequivocal” warming of the Earth’s average global temperature during the second half of the 20th century. Researchers in this group were compiled primarily from the list of contributors to the Fourth Assessment Report Working Group I, and from four prominent scientific statements endorsing the IPCC.

The list of researchers defined as unconvinced by the evidence for anthropogenic climate change was compiled in part from 12 of the most prominent public statements criticizing the IPCC’s conclusions.

Names Named … But Not as Part of PNAS Report

The lists that researchers relied on for both groups are not included in the PNAS posting itself, but are compiled on a website maintained by Prall. The lists can be found here.

Prall’s blog, “Green Herring,” is devoted primarily to building lists of climate scientists and petition signers.

Once they had the lists compiled, Prall, Anderegg and their colleagues then mined Google Scholar to find out how many articles including the word “climate” each of the 1,372 researchers had published.

Finally, they determined how many times each of the researchers’ top-four cited articles had been cited by their peers, as number of citations is considered a valuable indicator of a researcher’s standing among scientific peers.

To qualify as a climate researcher for the analysis, individuals had to have published at least 20 papers. That threshold brought the database of 1,372 researchers down to 908.

Anderegg and his colleagues wrote that they recognize that the number of times a researcher’s paper is cited can be skewed, either by the researcher citing his or her own paper in subsequent writings, or by “clique-citation” in which colleagues supportive of one another’s work cite each other.

But they said they were able to minimize those patterns by analyzing a large number of researchers, and by focusing their analysis only on the researchers’ four most-cited papers.

“Bottom line: those who deny (anthropogenic global warming) are much less productive and much less cited as climate scientists than those who accept AGW,” Schneider wrote.

The study should help journalists who cover climate change, he said, reflecting his continuing concern that media reports too often risk accuracy in the name of a misleading journalistic “balance.”

“Science is not a democracy, but evidence-based and difficult judgments on how much confidence to assign to issues like the reality of (anthropogenic global warming) can only be competently assessed by those who understand the complex details,” Schneider wrote. “Others’ opinions are simply irrelevant.”

Pielke, Jr.: Study is ‘Red Meat’ for Those Politicizing

Roger Pielke, Jr., a professor of environmental studies at the University of Colorado at Boulder and a frequent commentator on climate issues, criticized the study as an oversimplification.

“This paper continues a pathological tradition of reducing debate over climate change to black and white — believers and non-believers,” Pielke wrote in a recent e-mail interview. “It also contributes to the well-worn but wrong notion that political action can be sped along by ridding the world of incorrect scientific beliefs about climate change.

“The issue of climate change is far more complex than represented by this paper, and the distribution of views among experts is far more nuanced than can ever be represented by two categories.”

Pielke wrote that he took particular issue with the study’s assertion that there is “substantial and growing public doubt” about the science surrounding climate change.

He pointed to an article by Ted Nordhaus and Michael Shellenberger last November, in which they wrote:

“Public opinion about global warming, it turns out, has been remarkably stable for the better part of two decades, despite the recent decline in expressed public confidence in climate science. Roughly two-thirds of Americans have consistently told pollsters that global warming is occurring. By about the same majority, most Americans agree that global warming is at least in part human-caused, with this majority roughly equally divided between those believing that warming is entirely caused by humans and those who believe it to be a combination of human and natural causes. And about the same two-thirds majority has consistently supported government action to reduce greenhouse gas emissions since 1989.”

Those conclusions are consistent with a number of other reputable polling results, notwithstanding poll findings of recent declines in concern over climate change.

Pielke called the PNAS study “‘red meat’ for those who would like to politicize the science of climate change (on all sides of the debate) through oversimplification and a conflation of views on science and views on politics.”

“It will do nothing to move policy debate forward,” Pielke wrote.

“Public opinion on climate change has been largely stable for decades — both on understandings of the science and support for action,” he continued. “Policy will move forward when this ample support is treated as a resource to be worked with, rather than seeking to create and enforce a uniformity of opinion.”

Pielke wrote more about the study at his blog.

Boykoff: ‘Productive Contribution’ … but Concerns

Meanwhile, Max Boykoff, an assistant professor of environmental studies at the University of Colorado at Boulder, called the study a “productive contribution.”

But he said he too was wary of assigning blanket labels like “skeptics” to those who have expressed doubts about human-caused climate change. In a recent e-mail interview, he said he wanted to see analyses that would examine the motivations behind such stances. “Skepticism can be derived from evidence as well as ideology,” he wrote.

Boykoff said further that he is unconvinced that the number of citations can be accurately used in the study to help define a researcher’s expertise — in part because of the tendency of colleagues to cite one another.