Arguments
Bulletin of the Atomic Scientists: Closing the Consensus Gap
Posted on 20 August 2013 by John Cook
The Bulletin of the Atomic Scientists is a prestigious journal, established in 1945 to warn the public about the consequences of using nuclear weapons. They've published the writings of Hans Bethe, Albert Einstein, J. Robert Oppenheimer and Mikhail Gorbachev, among others. The Bulletin is closely followed in Washington, DC, and other world capitals and uses its iconic Doomsday Clock to draw international attention to global risks and solutions. It links the work of researchers and experts with policymaking entities, with the goal of influencing public policy to protect the Earth and its inhabitants. Thus I was honoured to be invited to submit an article, Closing the consensus gap: Public support for climate policy. In this article, I discuss our paper Quantifying the Consensus and why there was a need for it - because of the two-decade long misinformation campaign against the consensus. Here's an excerpt:
Since the late 1980s, governments and policy makers have worked to develop policy to mitigate climate change. At the same time, opponents have worked to delay and prevent climate action—not just by attacking policy solutions, but also by attacking climate science. A key focus in this decades-long campaign has been to cast doubt on the scientific consensus that humans are causing global warming.
Why attack the consensus? In recent years, social scientists have started to put the pieces together. A study published in the journal Nature Climate Change in 2011, replicated by a 2013 study published in the journal Climatic Change, found that public perceptions about scientific agreement are linked to support for policy to mitigate climate change. When people think that scientists are still debating about what’s causing climate change, they’re less likely to support climate action.
Social scientists were not the first to come to this realization. Political consultant Frank Luntz advised Republicansin the 2000 presidential election to cast doubt on the consensus, arguing “should the public come to believe that the scientific issues are settled, their views about global warming will change accordingly.” More than a decade before social scientists observed the link between perceived consensus and support for climate policy, opponents of climate action understood this link and implemented communication strategies designed to erode public support for climate policy.
In fact, attacks on the consensus date back to the early 1990s. In 1991, the Western Fuels Association spent more than $500,000 on a campaign to “reposition global warming as theory (not fact).” More recently, an analysis of conservative columns published from 2007 to 2010 found that the most repeated climate myth was “there is no scientific consensus.”
These strategies have been effective. To this day, there is a significant “consensus gap” between public perception and the actual scientific consensus. A 2012 poll conducted by the Pew Research Center for the People & the Press found 43 percent of Americans thought climate scientists were still in disagreement about whether the Earth is getting warmer because of human activity.
I recommend you go read the full article (it gets a bit more up-beat further on :-)
This is why John Cook's refereed paper 'Qualifying the Consensus' is so important! It is also why it is attacked with the most spurious inane analyses.
The anti science brigade will always do damage to the rest of us. The anti vaccination mob against preventable diseases are a good example of a subset of deniers that fear an outcome that has no basis in scientific reality. Yet when there is a lethal outbreak of whooping cough these deniers line up in droves.
I look forward to all the deniers having a 'road to Damascus' conversion. The comments on denial sites seem to me to be now only the most fervent ignorati. Cherry picking has become such an art, that the Japanese would categorise it as a 'National Treasure'!
Bert
Does the J. Cook paper say that there is a 97% consensus regarding AGW or a 97% consensus regarding dangerous AGW. I'm still not clear a bout this.
Bert points out an interesting, if painful, truth: The more valuable a contribution is to the discussion, the more it helps people (particularly newcomers) realize what's going on re:CC, the more viciously it will be attacked. If the deniers keep it up, QtC might someday achieve the exalted status of the hockey stick.
I have to respectfully disagree about the deniers having any sort of public conversion. I think that be far the most likely outcome is that they will dwindle in numbers and influence (even as their claims become ever more shrill and absurd, their tactics ever more outrageous) until they're just another group that's widely recognized for being, shall we say, excessively colorful in their views. Other examples would be vaxxers, the HIV/AIDS deniers, the moon hoaxers, and the people holding out hope that Elvis is still alive.
For comparison, consider all the people who were adamant that Y2k was going to be TEOTWAWKI (Google it) and were convincing people to buy remote property stocked with tuna and sacks of beans, guns, diesel generators, etc. When Y2k was a non-event (because we fixed it, not because it wasn't a problem), these people simply disappeared or moved on to other things.
I also expect to see the highest profile deniers, like our friends in right-wing radio, suddenly pivot and blame the scientists for "not making a sufficiently compelling case." The sound of heads exploding at that point will be deafening.
@BillyJoe #2:
Exactly how do you distinguish between "dangerous AGW" and "AGW"?
@BillyJoe #2 - You seem to be suggesting AGW isn't dangerous. Perhaps you would like to investigage that notion elsewhere on this very site:
Most Used Climate Myth No. 3 - It Isn't Bad
The groups that push the concept of "consensus" as some kind of proof for the reality of AGW are actually the ones that are disregarding scientific process. Science does not work by consensus, you don't get 1000 scientists in a room and if they all agree about something, then it is true. Scientific process is Hypothesis, Experimentation, Results and Conclusion. Just focus on the data for AGW, not the number of people that believe in AGW. When one does that, you are moving from the realm of science to the realm of faith.
Actually, Professor Richrd Tol pointed out quantitatively what is meant by Dangerous Climate Change - it is 2C above pre-industrial global average temperature, a figure internationally agreed, by others including President Obama, and signed off by him in Copenhagen.
There should be no doubt then about what the President meant.
joeygoze, it was deniers who first started claiming there was no "consensus" among climate scientists, therefore no reason to take the science seriously. Now deniers are trying to move the goalposts, saying that consensus isn't important. I am old enough to remember the same tactics about science showing that smoking causes cancer. Different topic, but same strategies.
joeygoze - Correct, science doesn't work by consensus, rather consensus is the result of the majority of scientists being convinced by evidence. It's an outcome, not a starting point.
Public policy, on the other hand, is entirely driven by public consensus - and understanding and trusting the experts is a significant part of public opinion. Currently there is a large gap between the public understanding of the consensus in science and what that consensus actually is (fueled in no small part by lobbyists, disinformation, false 'balance' in mass media presentations, and a considerable pile of money).
Claiming a lack of consensus is a delaying tactic that has been used by the tobacco industry, over the ozone hole and DDT, mercury pollution, on and on, and now by those potentially impacted by climate action. Far better that the public should understand the expert opinions, and develop political will and action along factual lines rather than lobbying slogans.
joeygoze - Do you feel it desirable or appropriate for the public to be misinformed regarding what is an important policy issue affecting everyone? Because currently, they are.
This is a its core a public policy issue, not a climate science issue. Public policy as an input incorporates public perceptions of expert opinion - perceptions which (due to a great deal of disinformation) are currently quite distorted. Your position would appear to be allowing those misconceptions to persist, letting disinformation go unchecked - I would have to disagree.
KR - The public policy must be driven from the science and proper scientific process. Not a bunch of people agreeing with each other.
I would have to understand what you define as "misinformation". For example, is a scientific paper that is peer reviewed but has conclusions that you disagree with automatically misinformation?
Would a paper such as this: (http://www.worldscientific.com/doi/abs/10.1142/S0217979213500732) which looks at solar radiation and CFCs as a major driver of global warming be considered "misinformation" to you?
joeygoze - You appear to be ignoring the majority of my last comment. The public perception of expert opinion is incorrect, distorted by mis/dis-information. And that public perception drives public policy.
The science is just fine, with papers finding acceptance or dropping into irrelevance based on the data. The public perception of agreement on the science, however, is incorrect, and that leads to poorly considered policies. Nobody is arguing about the science in this particular discussion - that's a red herring in regards to public perceptions and policies. I have to consider diverging to various papers as wholly off-topic on your part.
Again - you appear to be arguing to allow public misperceptions to continue, misperceptions fueled in part by extensive lobbying. And I disagree.
joeygoze @6, given the concerns you have raised, you no doubt have examples from the scientific literature of scientists arguing that a concensus of climate scientists agree with AGW, and that therefore AGW is true. Absent such examples, you are clearly raising a straw man. So, could you please list even one such article.
@John Haetz #4
"Exactly how do you distinguish between "dangerous AGW" and "AGW"?"
AGW could just be an interesting fact.
Dangerous AGW requires action.
If even Andrew Watts and Christopher Monckton agree that AGW is happening, then concluding that 97% of climate scientists agree that AGW is happening doesn't say much. If these same 97% agree that action is required to mitigate its effects, then that is saying something meaningful.
Nobody here seems to want to address my question...
What exactly does J Cook's paper conclude ?
@Bay Bunny #5
"You seem to be suggesting AGW isn't dangerous."
Nope.
I'd just like to clarify what the J Cook paper actually says about the consensus
(I'm not sure what is causing the formatting errors in my post???)
Tom Curtis - actually I am making the opposite argument, that the consensus argument is a worthless conversation and the paper that claimes the 97% consensus is suspect in anycase. The abstract from the paper (http://iopscience.iop.org/1748-9326/8/2/024024/article) states, "We find that 66.4% of abstracts expressed no position on AGW, 32.6% endorsed AGW, 0.7% rejected AGW and 0.3% were uncertain about the cause of global warming. Among abstracts expressing a position on AGW, 97.1% endorsed the consensus position that humans are causing global warming." So out of ~12,000 papers that were looked at, 66.4% take NO position (~7,970 papers). Only ~32% of the papers (~3,840 papers) that took a position happened to take a positive position that manmade CO2 is driving climate change. In looking at the set of 12,000 papers, that looks to fall short of what I would think someone would define as a consensus. Also, papers that took NO position, what were the results. Did they suggest something else was driving climate changes aside from CO2, and if so, what? This type of study is not Black and White and we as much as the other side should be just as critical of things if the science looks dodgy.
KR - You are certainly fine to disagree but stop crying about public misperceptions because of some lobbying effort from the other side. Maybe our messaging must be better? (-snip-)
[DB] Sloganeering snipped.
Billy Joe - no, you are making a non-argument by not acknowledging a problem in the study that out of a set of 12,000 papers that deal with climate change, only 33% support the claim of CO2 driving climate change. You prefer to cherry pick data that out of the 33% that took a position, most support the concept that man made CO2 is driving climate change and that is where the 97% consensus comes from.
Oh my, someone's discovered the word 'cherry picking'! The nonsense is familiar but I haven't seen it mixed with 'cherry picking' before.
In fact, of course, everyone but the very few denialists knows that AGW is a fact, so there's no point in expressing support for it, any more than articles on evolution need to point out that the earth is not 6000 years old. On the other hand, if you have evidence against AGW, that's new and important, and why in the world wouldn't you put it in your summary? If anything, dropping the 67% that didn't take a position underestimated the consensus.
And if that weren't sufficient demolition, let's recall that scientists were invited to rate their own papers' support for AGW, and once again the consensus was over 97%.
joeygoze,
The 12,000 papers that were assessed were simply the set of all scientific papers published during the period of question that survived a set of easy-to-apply filters to make the manual task easier. They do not represent all papers that might be relevant to the question at hand, nor do they represent only papers that might be relevant to the question at hand. They are merely a (hopefully unbiased) sample to estimate the level of consensus in relation to the question at hand in the whole of the scientific literature.
As such, any calculations of endorsement that involve all of the papers in the sample, rather than just those papers that were deemed to be relevant to the question at hand on manual inspection, are invalid. After all, with sufficient resources, it would be possible to assess every single scientific paper ever published, guaranteeing that all relevant papers were assessed, but also guaranteeing that the percentages of endorsement and rejection out of the whole sample would be approximately zero due to the large number of irrelevant papers involved. In spite of that, we have a fair degree of confidence that if all of the irrelevant papers were excluded from the calculation, the levels of endorsement and rejection should be quite similar to that found by Cook et al due to the scale of the survey conducted by Cook et al and there being no demonstration of bias that I've seen to date in the filtering algorithm used to narrow down the sample size.
All of this has been discussed previously. If you feel you have some new insight to bring to the table then please do so, but it's annoying to have the exact same comment being made over and over again by an unending stream of different people who apparently think nobody else has considered this issue and can't be bothered reading up on it. But keep in mind two things:
1. The classification of "relevant to the question at hand" was obviously very conservative, in that the author's own ratings of many papers that Cook et al classified as "No position" based on the abstracts were actually claimed to take a position by their authors based on the entire paper. This means that the true value of "papers out of the 12,000 that support the claim" is actually much higher than 33%.
2. The authors' own ratings gave very similar results despite large numbers of "No position" papers being reclassified by their own authors — so even though significant numbers of papers were added to the pool of relevant papers, the ratio of endorse:reject barely changed.
joeygoze #11 You might want to read this post about the Lu paper you reference.
Misinformation? Dunno. Fairly sloppy work? Yep
"For example, is a scientific paper that is peer reviewed but has conclusions that you disagree with automatically misinformation?"
It isn't disagrement with the conclusions that makes something misinformation joeygoze. It is the use of cherry-picked data, logical fallacies and playing on the ignorance of your audience to falsely produce a 'conclusion' that can't be identified as spurious by your audience.
OK, so even if I accept the "standard practice" of throwing out papers that do not take a position and look at 3,870 papers in which 97% of them support human driven AGW, let's look at that set more deeply from Cook et al.
http://www.populartechnology.net/2013/05/97-study-falsely-classifies-scientists.html#Update2
Many scientists, when spoken to about their papers and how they were classified in the Cook et al. paper claim that their paper was not classified correctly. One scientist makes the same point that I made earlier, "Science is not a democracy, even if the majority of scientists think one thing (and it translates to more papers saying so), they aren't necessarily correct. Moreover, as you can see from the above example, the analysis (in Cook et al.) itself is faulty, namely, it doesn't even quantify correctly the number of scientists or the number of papers which endorse or diminish the importance of AGW." (quote from Dr. Shariv, Astrophysics, Univ. of Jerusalem)
Phil Cohen - On your statement, " if you have evidence against AGW, that's new and important, and why in the world wouldn't you put it in your summary?" Maybe because Cook et al. recruited a bunch of die hard AGW believers to help conduct the study and any work that contradicts the faith may have been ignored? I would rather see a study to proove consensus would be a survey study sent to scientists to answer straight forward questions thus you eliminate the entire difficult process of analyzing the words of a paper and subjectively determine what the authors meant.
joeygoze,
Mr. poptech's argument has been addressed.
www.theguardian.com/environment/climate-consensus-97-per-cent/2013/may/28/global-warming-consensus-climate-denialism-characteristics
Remember there were 12,000 papers surveyed. What percentage are then under dispute by those carefully-selected skeptical authors?
NewYorkJ - So can you point me to the following data? From the article in your link.
"We invited the scientific authors to categorize their own papers, so if they responded, their 'correct' classifications of the full papers are included in our database. As illustrated in the graphic below, we found the same 97% consensus in both the abstracts-only and author self-rating methods." I see the graph below the text but it is expressed in %s, want to see the number of papers in which authors also gave a response and agree with the classification, is that available?
joeygoze,
See "Table 5. Comparison of our abstract rating to self-rating for papers that received self-ratings."
Useful would be to read the full study, the supplementary material, the FAQ, and examine the full list of papers and authors yourself, and as a learning exercise, doing your own ratings. Most of the criticism tends to be from individuals who do not have a good understanding of the study, or hope their target audience doesn't.
joeygoze #24,
That information is available, and in fact has already been published here, at SkS, along with the following histogram:
joeygoze,
In case you don't find it in there (although you really should read the post to which I linked -- you seem to be suffering from a lot of misconceptions based on "yeah, but what if..." thinking), the self rating data is here.
Sphaerica @26, there is one instance in the self rated papers of a paper rated by abstract as explicitly endorsing with quantification (1) being self rated as explicitly rejecting with quantification (7), ie, a difference of -6, which is not shown on the graph. That represents the same percentage of papers as that shown for differences of -4 and 4. It is, therefore, not consequential, but should be shown for accuracy.
It is very amusing to see Joeygoze trying to suggest that the data has been hidden, when it is freely available for download.
I cannot personally understand why people seem so intent on attacking the 97% figure as unrealistic when clearly it is the 3% figure that really deserves scrutiny, since it really is not 3% at all, but over a dozen fractions of a percentage point combined without any real reason for doing so.
Take for example the Lu study blaming CFCs by blatant curve fitting. Do you imagine for one moment that any survey of climate scientists asked the question "Are you convinced that CFC'c are responsible for climate change?" (although that would, technically constitute AGW, since there are no naturally occuring CFCs) that over 99% of those expressing an opinion would disagree? Wouldn't that hold equally true of theories about solar fluctuation, cosmic rays, or the intervention of the Flying Spaghetti Monster?
Calling it 3% vastly inflates the credibility of AGW opponents, and hides what they really are ... cranks at the fringes of scientific thought.
Old Mole - Quite true, quite true.
It's just that "We are that particular tiny fraction of the 3%", although accurate, isn't a very catchy slogan for denialists. It might make a good T-shirt, though.
Newyork J - I think the fact that 6 skeptical papers got into the "support AGW" column is reason enough to suspect a methodology problem and this unfortunately weakens the conclusion of the study.
Tom Curtis - I was not suggesting data was hidden, was honestly asking for someone to point me to the data
I am amazed at your ability to cherry pick red herrings joeygoze. It is a shame you missed my Freudian slip in calling John Cook's paper 'Qualifying the Consensus' instead of Quantifying the Consensus. The irony is by quantifying the consensus one can then qualify the consensus! This point has completely eluded your obvious talents. Bert