Here's my recent Comment on Verheggen et al. in Environmental Science & Technology.
Their core argument in their Reply is that all the non-climate scientists who were invited to participate in this survey because of their climate-related papers – all the psychologists, sociologists, Marxists, economists, palladium experts, etc. – would deny having done any climate-related work, and would thus not be a significant portion of the responses to the survey. This is nonsensical and a waste of everyone's time.
Background: The authors don't know who is in their data, due to inadequacies in their design. That's the key fact that frames this issue.
1) We don't know who is in this data.
2) We know vast numbers of non-climate scientists were invited.
3) We don't know the actual results of this study. That is, we don't know the results that are strictly based on qualified experts – climate scientists. They should just retract, or the journal should. We might need to deploy arms-reach retraction buttons to grease the wheels for pro-science decisions, for reasons I explain below.
The larger issue here is that there is likely nothing happening in science right now that is of lower quality than climate consensus research. It's a disaster. Much of the research doesn't meet anyone's standards for science. It's pre-scientific. Many of these studies are politically motivated junk that we couldn't possibly draw any inferences from.
The junk studies report the highest consensus figures. They're inflating the consensus, probably distorting people's estimates of certainty, given the nature of human minds and how "97%" might be processed. (There is no 97%. That was a scam, as I'll explain below.)
This is a terrible distraction, because it casts doubt on the whole premise of a climate consensus, and it risks enormous harm to the reputation and standing of science in the public mind. Skeptics see this garbage and find it satisfying and convenient for their prior commitments to climate skepticism. They may not pay attention to the valid studies that show a consensus, albeit a smaller one. It unreasonable to expect the general public to wade through all the junk and find the good stuff.
There are valid studies. There is almost certainly a consensus in climate science with respect to human-caused warming. It's not as though anyone has done a survey and reported a 40% consensus or anything like that. There are no disconfirming findings, to my knowledge. Every survey reports a consensus well north of 50%.
The junk studies started with Oreskes (2004) and mimic her methods. That study is not a study. It's a one-page paper whose methods section is a paragraph or two, that offers no detail or validation of its methods. We don't even know how these subjective ratings of abstracts were conducted, or by whom. I'm slowly getting my head around the idea that she did them all herself, apparently. That's very confusing, for someone to basically say, hey I read 900 abstracts, decided what they mean, and none of them disagree with human-caused warming. I'm not a climate scientist or anything, but here's my one-page paper. (Oreskes is a staunch environmentalist activist who in a recent book projects a collapse of civilization because of our anti-environment ways. Also note the incredible fallacy in demanding to see explicit disagreement with a proposition or hypothesis, and treating a lack of explicit disagreement as positive support for the hypothesis. There are several problems embedded in that bizarre supposition.)
Subjective rating is a social science method. No social scientist would ever submit the results of a subjective rating study where he alone did all the ratings and had an ideological conflict of interest with respect to the outcome.
This is all a horrible joke. Let me pause and note that I have never been more confused than I have been over the past year by all the fraud and all these bizarre junk studies. It's disorienting. This can't be what science is. Science is this precious, wondrous thing. It's arguably the best thing humans do. Political ideology is eating science alive. This collapse of integrity, the incredibly bold acts of fraud and scientific authorities' attempts to protect that fraud, the apparent lack of serious peer review and of even minimal methodological standards, this is all a disaster. Science can't be this. Politics is just killing us right now. Politics is acid on science. It always has been. But I think our era is more political than many other eras. I think the the influence of political ideology in academia is at a historic peak.
The Oreskes method includes searching climate science literature with plain English phrases like "global warming". All the ideologically driven junk studies that copied her method likewise searched on "global warming" or "global climate change". That was it. They take the results of these searches and include them all in their mystical counting rituals. This is what Verheggen et al. did.
In fact, Verheggen et al. did not even bother to uncheck the Social Science and Arts & Humanities boxes in their search. (This is the Web of Science index.)
I'm sorry, I know this is very negative, but this honestly isn't even undergraduate-level work. You could get a kid to do this. This is so awful that it's a disgrace that any of these studies were published in 21st-century scientific journals. Politics is the reason they are published. Politics is just devastating us. It's turning science into a scam.
These people had no idea how to search scientific literature. This is confusing. Did they have no training? They think climate scientists are going to commonly say "global warming" in their titles or something. This is bizarre. 21st-century scientific fields have their own technical terminology. We could never search scientific literature with casual English phrases. Climate scientists don't talk like that. They'll be talking about aerosol spectra, ENSO, and seasonal variation in CFCs.
The search these people did has extreme asymmetries in its results. Searches on casual English phrases like "global warming" will capture lots of non-climate scientists who would use less technical language – most especially activists and people who are framing their non-climate science work around warming.
The search was never validated. Searching scientific literature is a well-documented discipline, most notably as a core feature of meta-analyses. It's not a casual thing. You have to test your search. A basic way to test it would be to compare the results to known sets of climate science papers, for example from climate scientists' CVs. The Cook 97% scam conducted this same search. Look at how many papers they included by James Hansen, compared to Richard Lindzen. That study was based on counting the papers. Those people conducted a bizarre casual English search, got thousands of papers, had environmental activists who had profound conflicts of interest subjectively rate the abstracts, à la Oreskes, and then counted the papers, sorting them into their rating categories. They simply... counted... the papers. They thus arbitrarily gave some scientists over a dozen votes and other scientists zero votes. They treated these votes, these papers, as quanta of consensus.
That is something I hope we never have to deal with again. The sheer dimwittedness of this stuff scares the hell out of me. I'm not sure that we can have a civilization if people can do that and plant a 97% meme all over the world. If people can do what they did, and get the President of the United States to tout their "study", holy cow. This seems incredibly dangerous. This is like watering your crops with Gatorade. We can't do too much stuff like this and expect to have a reliable food supply, technology, hospitals, smartphones, a stable civilization. We're not going to have nice things.
Anyway, for good consensus research:
Bray and von Storch are outstanding. Their studies are real, and bear the customary marks of scientific effort.
The AMS studies are excellent, and also feature valid scientific methods. You can read their latest, July, 2014 study here.
Note that the American Academy for the Advancement of Science (AAAS) only cited the junk studies in their What We Know missive. That was incredible. (I'm going to use the word incredible a lot this year.) They cherry-picked the studies that gave them the inflated, false 97% figure they apparently wanted. I couldn't believe it. They completely ignored virtually the entire body of research on the consensus. They ignored the two sources above, which are of massively higher quality than the bizarre "studies" they cited, and much more up to date. They ignore every sub-97% study. Instead, they cited Oreskes, the one-pager from 2004, as one of their three sources, along with Cook et al's bizarre counting study (which is also a fraud case, with three separate fraud acts, three different lies about their methods...unbelievable.) I doubt the AAAS membership are aware of this, would have paid a lot of attention to it. I assume someone's going to have to retract that report if they want to be considered a scientific body in the future. The report is incredibly misleading. It's simply lying to the public, inflating the consensus by citing ten-year-old one-page studies that no scientific body could ever cite. That's unconscionable, and again this behavior is killing us. AAAS is sabotaging the reputation and trustworthiness of science as a whole. I would not assume that these scandals, these politically-driven frauds, will have no impact on the public's respect for science. They know we're lying to them. Someday we might really need them to believe us.
AAAS is not some two-bit organization known for scams. They're much more serious than IOP. They're an august scientific body. When I think of august scientific bodies, I think of AAAS and the National Academy of Sciences. We're running out of august and honest scientific bodies. They're falling like dominoes to political ideology and fraud. We need to have a home, a place to go for truth, integrity, and sober science. We need a place where the average applied IQ is somewhat north of 80.
We can't run out of scientific bodies. If science has no home, no reliable non-fraudulent, non-political institutions, I think that could seriously weaken our civilization. We're nearing a point where science will be broadly associated with fraud that particularly serves left-wing political agendas. We're nearing a point where the rational knower would be well-advised to ignore what contemporary scientists say, because employing such a heuristic would lead to the most accurate set of beliefs about the world, that one's ratio of true to false beliefs would be maximized by ignoring scientists. That scenario entirely possible as an epistemic reality given our current course. Another ten or twenty years of this, and we're there. The "deniers" could end up being the rational knowers, the pro-science among us.
Note that there are some very large cases emerging right now, where the scientific consensus was wrong. There's a massive new study in Annals of Oncology by Wang, et al. that reports that second-hand smoke does not significantly increase lung cancer risk (in women.) A review in PLOS One by Yang et al. finds no link between second-hand smoke in breast cancer. The second-hand smoking case is emerging as something that might never have been well-researched. That whole issue seemed political, where people were using the political process to coerce property owners (bars, restaurants, etc.) into enacting their preferred comfort and lifestyle settings, and using the machinery of the state to enforce various prejudices against smokers, who are now corralled into holding pens hundreds of feet from building entrances.
And the Washington Post reports that the US government is poised to withdraw its warnings about cholesterol. On both second-hand smoke and cholesterol, we were sold a consensus, we were sold "Science says X"-style propositions, which are generally barbaric and take no account of where science is in our lifetimes, in our era – which is an arbitrary era – and whether the methods employed in said arbitrary era are capable of giving us a workable grasp of reality on any arbitrary issue Y, and that our scientifically sourced grasp of reality will not be mediated or shaped by political processes, the media, and the ideological and financial biases of scientists. Scientists are wrong all the time. We have to be. In a sense, it's almost part of the job, bias notwithstanding. We need to be smarter about how we understand science.
In any case, there is indeed a consensus in climate science. It's probably in the 80s though, maybe as low as the 60s for some questions. It's not very meaningful to speak of "the" consensus, since there are a number of different propositions one could pose to climate scientists, and the two sources above do a great job of posing a range of relevant propositions. What you do with that consensus up to you. There will be all sorts of philosophical, ethical, and political factors that people will reasonably apply, and there will be a host of different perspectives along those dimensions. But I don't think climate change skepticism per se is justified. I would love to be wrong.
There are new arguments and insights in my Comment below, so I encourage you to read it. To my knowledge, no one has before made the argument about mitigation studies, for example.