The evidence that we are running dangerous risks with the climate is overwhelming. In their zeal to convince the public of this fact, environmental advocates sometimes hype sensational studies and predictions that rest on weak or ambiguous logic. Every time they do, their opponents have a field day.
This week the greens have played right into that trap. On Monday, researchers at three institutions released a study purporting to use "an extensive dataset of 1,372 climate researchers" to show that almost every serious climate researcher in the world believes in the basic science of human-caused global warming, and that the few skeptics there are lack the "climate expertise and scientific prominence" of their peers. The paper, entitled "Expert credibility in climate change," was published Monday in the Proceedings of the National Academy of Sciences (PNAS). Within hours, a host of progressive and environmentalist groups were loudly touting its conclusions, and sympathetic articles appeared in USA Today, The New York Times, and Time. On Wednesday morning, the White House drove the message home with an official tweet: "Scientists agree on climate change...the few that don't? not cream of crop."
That message makes sense, intuitively: Since pretty much all arguments against the basics of climate science are silly, it stands to reason that those who peddle them are silly, too. But the PNAS paper does not simply offer a conjecture—it claims to have scientifically demonstrated its conclusions. Indeed, that is why the climate advocates have promoted it so enthusiastically and why the newspapers have given it so much attention. Alas, the study fails to substantiate its findings: It may well be that climate change skeptics are batty, but nothing in the work published this week proves they lack scholarly weight or prominence. In other words, the advocates have used bad social science to show that the science of climate change is sound.
The authors start by pulling together a list of scientists who are either "convinced" or "unconvinced" of the basics of mainstream climate science, based on whether they signed one of several climate-related documents. Scholars who contributed to reports of the Intergovernmental Panel on Climate Change, for example, qualify as convinced; those who signed open letters calling the existence or severity of human-caused climate change into question are unconvinced. (The way in which researchers were assigned to one category or the other has already been criticized by at least one scholar, who credibly claims that he'd been assigned to the wrong team.) The authors then cut these lists down to scholars who have published at least 20 climate-related papers in journals or books indexed by Google Scholar. These are the "climate experts" whose views and prominence they evaluated.
The authors establish their first point by comparing the beliefs of those scholars with the most published climate papers. They find that only 2 percent of the top 50 climate researchers, 3 percent of the top 100, and 2.5 percent of the top 200 are "unconvinced" of the basics of climate science. Thus they conclude that 97 to 98 percent of the most active researchers in the field support the tenets of anthropogenic climate change.
Those numbers are striking, but they don't tell us nearly as much about the number of skilled researchers on either team as the authors claim. One of the most frequent arguments made by climate skeptics is that they are shut out of peer-reviewed climate publications. Most analysts are rightly skeptical of this claim, and doubt that such a bias—if it did exist—could lead to such a large disparity. But it would explain the study results: If skeptics are being shut out of journals, their publication counts would go down, which would produce precisely the results shown in the PNAS paper. The same logic applies to the skeptics' claim that they are being starved of research funding for political reasons. The authors make no attempt to tease out the extent to which prejudice, rather than a disparity in expertise, can explain why so few skeptics rank among the top climate authors. Their conclusion is thus far less significant than they imply.
In any case, publication totals may not be a good indicator of credibility or expertise, since not all publications are equally important. Researchers who want to evaluate the quality of scientists' work tend, instead, to look at how often a scientist's work gets cited by others. The PNAS study used this approach next. Looking at the number of citations received by each climate expert's top four papers, the authors find that members of the "convinced" group rack up an average of 172 citations on each one, while the "unconvinced" get 105. Since, all else being equal, more citations implies better work, they conclude that believers are more expert and prominent than skeptics.
Yet all else is not equal. As the authors note, citation counts are often used by university departments to make decisions about hiring and tenure. But those situations involve comparing people who publish in the same fields. The PNAS study included scholars whose primary research is not in climate science. (A disproportionate number of those were in the climate skeptic camp.) To be fair, the authors also included papers and citations outside climate science so that accomplished scholars in all fields would be properly recognized for their talents.
The analysis of papers from diverse fields seems to have distorted the results. Take the case of Freeman Dyson, who ranks high on the list of climate skeptics. Dyson is one of the greats of Quantum Field Theory, a central branch of particle physics; he has, in recent years, become an outspoken skeptic on climate change. His top paper, on particle physics, had 749 citations when the authors checked. That's quite a few, but not as many as were received by several dozen climate scientists for their top papers. Dyson is by all indications badly wrong when it comes to climate change. But few scientists in any field would agree that his top work is less impressive than that of 40 climate researchers, most of whom are far more obscure. So what might account for the discrepancy in citations? Perhaps climate scientists tend to include a lot more citations in their papers than particle physicists do; perhaps they publish a lot more papers. A proper analysis would make sure that these sorts of things hadn't skewed the data.
All of this would be academic quibbling if it wasn't so consequential. The authors of the paper are right that the world is running dangerous risks with the climate system. They are right to be angry at those who claim that climate change is a hoax, and at those in the media who give them a platform to confuse the public. But the way to confront those skeptics is to show that they're wrong—as many dedicated climate scientists have done, again and again. Hyping this paper, instead, simply reinforces the dangerous perception that climate activists will credulously push any news that might further their case. For those who care about this issue, that's tragic.