Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.

Settings

Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup

Settings


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe


Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...



Username
Password
Keep me logged in
New? Register here
Forgot your password?

Latest Posts

Archives

Climate Hustle

Why the 97 per cent consensus on climate change still gets challenged

Posted on 18 May 2015 by Andy Skuce

Here are some excerpts from an article I wrote for the magazine Corporate Knights, published on May 14, 2015. Some references and links have been added at the end.

In 2004, science historian Naomi Oreskes published a short paper in the journal Science concluding there was an overwhelming consensus in the scientific literature that global warming was caused by humans.

After the paper’s release, there was some unexpectedly hostile reaction. This prompted Oreskes and her colleague Erik Conway to go even deeper with their research, leading to the publication of the book Merchants of Doubt. It documents how a small group of scientists with links to industry were able to sow doubt about the scientific consensus and delay effective policy on DDT, tobacco, acid rain and, now, global warming.

Fast forward to two years ago: a team of volunteer researchers (myself included) associated with the website Skeptical Science decide to update and extend Oreskes’ research. Led by University of Queensland researcher John Cook, we analyzed the abstracts of about 12,000 scientific papers extracted from a large database of articles, using the search terms “global warming” and “global climate change.” The articles had been published over a 21-year period, from 1991 to 2011.

As an independent check on our results, we also sent emails to the more than 8,500 scientist authors of these articles. (These were the scientists whose e-mail addresses we were able to track down). We asked them to rate their own papers for endorsement or rejection of man-made global warming.

Both approaches yielded a very similar result: 97 per cent of the scientific literature that expresses an opinion on climate change endorses the expert consensus view that it is man-made. The results were published in May 2013 in the journal Environmental Research Letters.

We were astonished by the positive reception. Mention of the paper was tweeted by U.S. President Barack Obama, Al Gore and Elon Musk, among others. Obama later referenced it in a speech at the University of Queensland, while U.S. Secretary of State John Kerry has referred to the 97 per cent consensus in recent speeches. John Oliver based an episode of his HBO comedy show Last Week Tonight around it, a clip viewed online more than five million times.

The paper has been also been downloaded over 300,000 times – so far – which makes it a blockbuster for a science paper. ....

[Snipped sections concerning Richard Tol's criticisms of Cook et al. and Tol's history of public disputes along with discussions of the "consensus gap" and scientific consensus as a "gateway belief".]

.... Climate contrarians everywhere protest there is no scientific consensus. If that were true, they should easily be able to show there is indeed a significant body of work that challenges mainstream science. Yet they haven’t and can’t, because a robust and coherent denial of man-made global warming does not exist.

Our study describes the state of expert opinion, it does not define scientific truth nor does it tell people what to think. Climate scientists today overwhelmingly endorse the consensus view that humans are the cause of most of recent global warming. That’s a fact.

If the goal is to sell doubt and delay action on climate, it’s a fact that has to be denied.

Click here to read the full article


References and links

Publications like Corporate Knights limit the number of hyperlinks in their articles. Here are some references and links that bolster the arguments I made.

President Obama's speech at the University of Queensland.

Graham Readfearn's article on Richard Tol's attacks on Cook et al.

Richard Tol's Energy Policy article on the consensus paper, our reply and his rejoinder (paywalled).

24 Critical Errors in Tol (2014), a document in which we lay out the case against Tol's critique in more detail than Energy Policy allowed us room for.

Economist Frank Ackerman on the dispute that he and Richard Tol had.

A Statement of Support for Ackerman from many prominent economists.

A Statement from the Stockholm Environment Institute leadership in support of Ackerman (quoted from in my piece).

A Guardian article by Bob Ward on the mistakes that he and others found in one of Richard Tol's frequently-cited papers.

Richard Tol's correction in 2014 to his 2009 article, leading, with an attribution statement concerning gremlins.

A further correction in 2015.

Some critical commentary on gremlns by Andrew Gelman.

Yale Climate Change Communication study (2014) that shows the gap between public perception of the scientific consensus on climate and the actual degree of consensus.

The Scientific Consensus on Climate Change as a Gateway Belief: Experimental Evidence (van der Linden et al., 2015).

0 0

Bookmark and Share Printable Version  |  Link to this page

Comments

Comments 1 to 32:

  1. A quick glance at commentators at say WUWT would contradict your first point. A UN/liberal plot to rule the world with falsified data when clearly climate is normal seems pretty common belief.

    "for the same reason that a butterfly flapping its wings in China can have an influence on subsequent hurricane formation in the Atlantic."

    Not remotely. Weather is chaotic but climate (weather averages over a 30 period) does not appear to be. Consider that you can get a wet cold day in summer but summer average temp is always going to be warmer than winter average temp because there is more irradiation of the surface during your hemispheric summer. Adding more non-condensing greenhouse gases to the atmosphere has same effect, but globally.

    Plenty deny it. The point of consensus is that consensus may or may not be right, but it is the only rational guide to setting policy. The consensus is that we need to reduce emissions and it appears that majority are very unwilling to do so or hate proposal that would be effective in achieving that;

    As to climate models not predicting slower warming, well what what part of  "climate models have no skill at decadal level prediction" is hard to follow? You can for instance see more of this discussed here. However, if you want to discuss this further please do so on this topic. Take very careful note of the comments policy on this site, especially the on topic/appropriate thread.

    0 0
  2. scaddenp,

    Using the comments made on sites like WUWT as a basis for what is commonly believed is like looking at the insects near an outdoor light bulb at night to determine the ratio or relative proportions of different insects that are active at night in a region, or looking at the birds at a garbage dump to determine the ratio or relative proportions of birds in a region.

    0 0
  3. Carbon Dioxide absorbs infrared heat energy. You can do 1000 experiments and get the same result 1000 times: that CO2 absorbs infrared heat energy.

    I don't care if it's Freeman Dyson, Judith Currty, or just some guy in congrss or on TV, none of these people have come up with any way to refute this very basic fact.

    Adding more CO2 to the atmosphere will cause the atmosphere to absorb more infrared heat energy. Adding sugar to water makes water sweeter. Adding black ink to white paint makes it darker. At the most fundamental level this is not so very difficult for people to understand. 

    It's far past time to stop with all this crazy denial of the basic, obvious physical reality.

    0 0
  4. It would be trivial to point out there is no overwhelming "evolution consensus" in the biological literature by counting up the number of published articles that explicitly endorse evolution in the text. Yet the argument is made that each climate paper must make an explicit claim or not be relevant to any consensus.

    Odd reasoning.

    0 0
  5. Scad, regarding your assertion that "climate...does not appear to be" chaotic, you are not in agreement with the scientific consensus.  This is what the IPCC has written:

    "In climate research and modelling, we should recognise that we are dealing with a coupled non-linear chaotic system, and therefore that the long-term prediction of future climate states is not possible."

    http://www.ipcc.ch/ipccreports/tar/wg1/505.htm

    0 2
    Moderator Response:

    [RH] Welcome to Skeptical Science!  There is an immense amount of reference material discussed here and it can be a bit difficult at first to find an answer to your questions.  That's why we recommend that Newcomers, Start Here and then learn The Big Picture

    I also recommend watching this video on why CO2 is the biggest climate control knob in Earth's history.

    Further general questions can usually be be answered by first using the Search function in the upper left of every Skeptical Science page to see if there is already a post on it (odds are, there is).  If you still have questions, use the Search function located in the upper left of every page here at Skeptical Science and post your question on the most pertinent thread.

    Remember to frame your questions in compliance with the Comments Policy and lastly, to use the Preview function below the comment box to ensure that any html tags you're using work properly.

    Your previous post was delete for being baseless gish-gallop. If you're going to post here you're going to have to be able to support your positions, and when you can't, you'll need to have the capacity to concede the point. 

    First case in point, climate, by definition, is not chaotic. If you believe this to be an incorrect assertion you need to support that with research that shows otherwise. Merely repeating the assertion without support will lead to having your posts deleted.

    (edit) Note that SkS has several articles on the issue of chaotic systems. Start here.

  6. TWFYSYWDI - I've responded on the appropriate thread

    0 0
  7. @Frightened: You don't understand what that quote means. It means that we can't predict the exact state at any particular time. It doesn't mean that we can't project long-term averages or trends with reasonable confidence.

    1 0
  8. Part of denialism is based on the argument that if we don't know everything, it means we know nothing. I think ThisWillFrighten is making that same basic argument — if any part of the climate system (which includes daily weather) is chaotic, that means it all is, and models are useless.

    This is also, I think, the basis for disputing the "consensus" argument. For example, if any questions can be raised about methodology, it means the study was imperfect, and therefore cannot be trusted (or "can be dismissed", which is the same thing).

    Another example: Are you =certain= the "true number" for consensus isn't 96%? or 94%? or even 90%? If it isn't =precisely= 97%, that proves the study is flawed, and we shouldn't draw any conclusions from it.

    This "reasoning" is insidious — even if all doubt could be removed that there is, in fact, a 97% consensus on this matter, that means 3% are unsure — if anyone is unconvinced, there must be legitimate question about the matter, and we shouldn't move forward.

    I don't know how to combat this problem. It's funnny, because the same people who make these arguments are willing to, for instance, play poker, or bet on football games, or even cross the street when there could be cars around. The "if we don't know everything, we can't know anything" ploy is really a rationalization, not an argument that is sincerely held. It is an excuse for ignoring what is really a convincing reality.

    2 0
  9. Like KR (and MARodger), I have also responded to TWFYSYWDI on the appropriate thread.

    1 0
  10. For some reason, Tom's response on the apropriate thread cannot be "upvoted". I'd like to stress the valueable, insightful details of Tom's comment, especially its last paragraph (about the meaning of TWFYSYWDI web name), so I've upvoted Tom's link above here.

    0 0
  11. Assessment of the hypothesis that rapid, irrversible climate change is under way would be more credible if the point was made that it is the operation of technological systems using fossil fuels that is producing the damaging greenhouse gas emissions. Saying climate change is man made does not help rational consideration of the evidence. People made unwise decisions but it was the operation of the systems that has caused the CO2 atmospheric concentration level to increase rapidly.

    0 0
  12. denisaf@11,

    Your point narrows the issue unnecessarily. The man made increase of CO2 is the result of not only "the operation of technological systems using fossil fuels" but broader human activities such as:

    - fossil fuels burning in a large sense, not only to operate technological systems, but e.g. burning coal/gas/petrol for heating,

    - land use changes,

    - cement production

    One broad definition that encompasses all such activiteis is: permanently (on human timescale) and irresponsively changing composition of the atmosphere by adding to it carbon that belongs to other reservoirs.

    0 0
  13. Moderator - Thank you for your kind response and your helpful suggestions.  But now I am utterly confused.  Perhaps you can clear it right up.

    I quoted directly from an official IPCC report which unambigously states that "In climate research ... we are dealing with a ... chaotic system..."  I gave the source of this report so you and your readers may check for themselves:

    http://www.ipcc.ch/ipccreports/tar/wg1/505.htm

    You responded with an assertion for which you offered no supporting evidence, as far as I can detect.  You wrote "climate, by definition, is not chaotic."

    Now we must choose.  Either we must believe the IPCC, or we must believe your contradictory unsupported assertion.

    I don't wish to jump to the wrong conclusion.  Please help me with this difficult choice.

    0 0
    Moderator Response:

    [PS] The points you made have been answered in the appropriate thread. See the pointers from posters above. As noted in comments policy, moderation complaints are always off topic. Please continue to discuss the main point but in the appropriate thread.

    However, given your chosen pseudonym, I must say that if your intentions are simply snark and trolling rather than engaging with science then please find another forum for your entertainment.

  14. Do we really have to bother with this nonsense? I understand that mods give everyone the benefit of the doubt but, after a numberof years of this, I have learned to identify the signs showing that one has nothing of value to contribute.

    TWFYSYWDI can't be bothered to go the right thread after several commenters indicate they have responded there, and are kind enough to include links. He/she uses one of the most basic tools of dishonest rethoric (quote out of context). And on top of it all coats his latest comment with snark when it is obvious he/she lacks the most basic understanding of the issue at hand.

    It's not like any learning will happen here. No matter how the exchange proceeds, the individual will portray it elsewhere with the kind of faithfulness given to the IPCC quote. Can't we just dispense from the hassle?

    0 0
  15. BBHY comment coveers the basics of the role that carbon dioxide plays in the argument.

    Tom Curtis has dealt with the role of chaos theory in highlighting the differences between weather and climate.

    Climate is predictable. Tendencies in the weather that are central to all weather forecasting is also valid. It is based on weather records from each locality and comparing the situation of today with the average of all recorded situations at the same time in the past under the same conditions.

    Choosing whether a climate scientist chosen at random will support or reject the consensus on climate change is also predictable. The results from the Consensus Project are fairly indicative. Such predictions are based on the probabilities that are involved in making the prediction or determining the outcome.

    What deniers/obstructionists like TWFYSYWD seem to want is 100% certainty when in science there is none. That is why confidence intervals are used in making scientific assessments. There might be chaos in the phenomonen but that does not necessarily mean that it is entirely random or unpredictable. In a coin toss, the outcomes are entirely random and somewhat chaotic with H and T occuring around 50% of the time for each as more and more trials are conducted. However, that does not mean we will know exactly how many heads will come up, tails will come up, or whether the coin will come up on it's side, which is, although very remote, not entirely impossible. Coin on its side just doesn't occur with the same equally likely manner that the H/T outcomes do. There is no theoretical 100% certainty even with a coin toss.

    Outcomes related to climate change, while perhaps being based on a somewhat chaotic phenomonen,  will always be based on the tendencies determined by the underlying scientific principles of the phenomonen. So if you heat the planet by increasing the level of greenhouse gases, you will get more chaotic turbulence filtering through the weather systems which will ultimately impact on the climate. The idea that climate scientists can't precisely predict what is going to happen from their climate models is not a reason to negate the fundamentals of the argument. A statistical aggregation with related probabilities from all climate models is more likely to yield an accurate prediction and the IPCC does a faily good job of aggregating the research.

    Personally, I prefer the paleoclimate data to indicate what is likely to happen. From what I understand, there is only a 4 degree or so global temperature difference between an ice age with all those huge continental wide ice sheets and an inter-glacial, with the polar regions we see today. I don't think returning the planet to the CO2 levels of the past when there no icecaps and it was up to 4-6 degrees warmer, by burning all known recoverable reserves of fossil fuels, is such a bright idea. As for those who think that climate change is all natural and it just magically changes, well I would suggest that you don't have any real scientific understanding of what "natural" actually means. Climate scientists, while not necessaly having a complete understanding, do have a pretty good understanding of what is happening and why. Based on the probabilites, I think the Earth has a problem that needs to be addressed. 

    0 0
  16. PhilippeChantreau - "Do we really have to bother with this nonsense?"

    While the Nth repetition of a denialist meme is rather tiresome to those who have been participating in forums such as SkS, I suspect that the occasional bit of nonsense is very instructive for undecided readers. Because they so clearly demonstrate the paucity of pseudo-skeptic positions.

    1 0
  17. mancan @18, what deniers seem to what is:

    1)  A clear acknowledgement that 'science is a process, not a body of knowledge' so that the consensus of scientists is irrelevant to science and should be ignored; and

    2) A clear acknowledgement that only that which is 100% certain in science is an acceptable basis for policy.

    Leaving aside that science is both a process and a body of knowledge, the second requirement above presuposes that science is a complete body of knowledge, ie, one that cannot require further expansion or correction.  Their two desiderata, therefore, rely on contradictory and inaccurate conceptions of science.

    1 0
  18. 1) A clear acknowledgement that 'science is a process, not a body of knowledge' so that the consensus of scientists is irrelevant to science and should be ignored; and


    Interesting. I just had this exact argument with Ms Nova. She refused to accept that Science was a noun as well as a verb (The noun was the original definition, even).

    0 0
  19. I endorse AGW, but I am confused about the calculation of 97%.  The set of e-mail responses includes "no position on AGW (35.5%)".  As I see it, there are two possible viewpoints among the 35.5% having "no position":

    "irrelevant" — The cause of global warming is not part of the discussion. 

    "undecided" — The cause of global warming is considered, but the evidence does not support a yes or no vote for AGW.

    As I see it, the "undecided" papers should be counted as opinions, but "irrelevant" papers should be omitted.  If so, then there should be three fractions: yes, undecided, and no, which sum to 1.  Then the yes votes would be less than 97%.

    Kindly explain the logic and tell me where I am going wrong.

    0 0
  20. If you change 'irrelevant' to 'unstated', that's exactly what the authors did.

    There was an 'uncertain' pot as well as a 'no position' pot.
     Very few papers were 'uncertain'

    0 0
  21. Tristan @20.

    The size of the two different 'pots' are relevant but they do apply to different data.

    Concerned @19.

    The self-assessment asked authors to rate their entire paper rather than just the abstract so it was different things being assessed as well as the assessors being different. The self-assessment survey asked for ranged from 1 = Explicit endorsement with quantification through 4 = No Expressed Position to 7 = Explicit rejection with quantification. There was thus no means of differentiating between 'irrelevant' and 'undecided'.

    The self-asessment results did a reasonable job of supporting the detailed findings of the Abstract assessment process, if anything strengthening the consensus finding. With the results so similar, it is likely relevant that the Abstract rating process only found a tiny proportion of 'undecided' Abstract, Tristran's "very few papers" (from Table 3 of Cook et al (2013) 0.3% of all abstracts were rated thus, ie "Uncertain on AGW"). It is thus likely that this tiny proportion would also have been found had the self-assessment allowed for such a catagorisation.

    However an author approached over this self-assessment process may well consider that if their paper fell into this 'undecided' category and they themselves were not convinced by the evidence of AGW that it should be rated more as a mild rejection of AGW rather than entirely neutral. And likely visa versa. Were there such an effect, again it does not support the idea of a large volume of 'undecided' literature being missed by the survey.

    0 0
  22. Concerned, suppose you have a class of 20 students; 10 boys and 10 girls... and you want to calculate the percentage of the boys who are white. The Cook paper methodology would have you count the white boys (say there are 7) and divide by the number of boys total... to get 70%. Your proposed methodology would instead have us find that only 35% (7/20) of the boys are white, 15% of the boys are non-white, and 50% of the boys are not boys.

    In short, you are asking why values not relevant to the calculation aren't included. That's a disturbing level of logical dissonance.

    Note that by such 'logic' the consensus level is essentially reduced to a factor of the size of the pool of papers reviewed. To take the extreme; the percentage of all papers on every subject imaginable which state a position on any given single issue is going to be near 0%. So your methodology of 'include non-relevant data' would indicate that there is approximately 0% consensus on everything. For example, approximately 0% of all papers ever written clearly state that gravity exists. Ergo there is 0% consensus that gravity exists.

    See 'where you went wrong'?

    0 0
  23. CBDunkerson @22,

    I don't believe your analogy is at all fair to Concerned's question. I have agreed with you when you have made similar statements in the past, but those statements were made in response to a different question.

    Putting aside the details of the 97% consensus paper (I am not sufficently familiar with the details), it is true that the relevant denominator to consider, when calculating a consensus, is the number of papers considering a position, including those that eventually adopt a conclusion of "undecided" after such consideration - not the number of papers adopting a position. Papers that do not consider the question at all are not at issue here, which is why your gravity analogy is unfair. If there were a large number of undecided papers (I believe there were not), that would indeed affect the validity of the consensus.

    If 1000 authors explicitly wrote about the possibility of an El Nino in 2015 and 97 of them said one was on the way, 3 said we were definitely not having an El Nino in 2015, and 900 said it was too early to call it, there would not be a 97% consensus we were having an El Nino in 2015. The fact that another million papers were on other topics would be irrelevant.

    0 0
  24. Leto, your statement that "Papers that do not consider the question at all are not at issue here" is simply incorrect. That is precisely the issue.

    The 35.5% figure Concerned cited actually came from the self-rating portion of the study. In that phase, abstracts for a total of 2142 papers were rated by their authors. Of those, 1342 (62.7%) endorsed the consensus, 39 (1.8%) rejected the consensus, and 761 (35.5%) did not address the issue.

    1342 / (1342 + 39) = 97% consensus

    Why not factor in the 761 abstracts, or some unspecified fraction of them, as 'undecideds'? Because this wasn't looking at opinions, but rather content... what the text of the abstract actually said. Either that text addressed the issue, or it did not... and for 35.5% of the papers their authors said that the issue wasn't covered.

    In short, 'the fact that another 761 papers were on other topics is irrelevant'.

    0 0
  25. CBDunkerson @24,

    Your first analogy about boys is not correct, as Leto said. Of course boys would be the denominator, as that is how the question was posed.

    What I was driving at is that the data should be presented as a histogram of those who have considered the question and have an opinion on the cause of global warming.  If some value the X out of the 35.5 % who adddressed and considered the issue were undecided on the cause, those X people have an opinion on the cause.  Then the histogram would have three values, yes AGW = 1342/(1342+39+X), undecided = X/(1342+39+X), and no AGW = 39/(1342+39+X).

    My basic question is, what is the value of X?  You are assuming that all 35.5% are irrelevant, so X is 0.  The words used to describe the 35.5% were "no position".  That is not the same as "irrelevant".   Some of those "no position" people could have considered the cause and be undecided.

    The problem here is the use of language.  "no position" is ambiguous.  I am asking how many of those "no position" people, X,  have considered the question, perhaps very carefully, and are undecided.  That value X affects the histogram.

    If there were 7 categories of opinion, then the results should be shown as a 7-column histogram, not yes or no.

    0 0
  26. Concerned @25.

    You are incorrect to suggest that the self-assessment is assessing the opinion of the authors. The survey is asking for a self-assessment of their work, not the authors personal opinion of AGW.

    Also the suggestion that a histogram would be a more appropriate presentation of the results - I find this unconvincing. The reason a paper is rated, say,1 - Explicit endorsement with quantification rather than 2 - Explicit endorsement without quantification is more to do with the type of analysis being employed rather than the finding of that analysis. There is in all this the truth that there are very few pieces of research 1991-2011 attempting to prove what all climatologists know to be true. A tiny handful of researchers continue the struggle to undo this truth but without result. These denialist climatologists may say this is 'without result so far' or 'without an accepted result so far' but I have not seen a thing in a decade to suggest the truth is in any way in doubt.

    Your histogram can be gleened from the uploaded self-assessment data I linked to @21 (although the total is six short):-

    1 - 242
    2 - 557
    3 - 539
    4 - 759
    5 -   25
    6 -    5
    7 -    9

    0 0
  27. Concerned @25

    (Correction:  I have just noted that I used the numbers of respondents rather than the number of papers in calculating the values below.  I don't have time now, so will recalculate later if you have not already done so before hand.)

    A subsidiary survey found that among abstracts rated as neutral on AGW (category 4), only 0.5% expressed the position "... that human's role on recent global warming is uncertain/undefined" (category 4b).  On the simplest assumption, therefore, only 0.5% of the 759 category 4 papers from the author self ratings would similarly fall into an "uncertain category" had one existed, so X would be 2, yielding a 96.26% endorsement rate rahter than the 96.38% actually obtaind for the paper.

    Arguably that is too simplistic a way to project the results of the subsidiary survey onto the author ratings.  Presumably the reduced percentage of category 4 ratings in the author ratings would have come from truly neutral (category 4a) papers rather than from those with a discernable opinion in the abstract (category 4b).  That being the case, we should project the percentage of category 4b papers as a percentage of all papers.  Doing so lifts the value for X to 3.99.  

    Arguably the category 4b papers would also have expanded in relative number in the same way as the affirming and rejecting papers did.  On that basis, we should multiply our projected X by the ratio of papers taking a positon in the author rating divided by the ratio of papers taking a position in the abstract ratings, ie, by 1.96 yielding a value for X of 7.8 which we can round to 8.

    Using this value for X, the endorsement percentage for author ratings drops to 95.4% with an projected uncertain (category 4b) rate of 1.02%.

    So, even with this most generous estimate of X from the only available data, the endorsement rate scarcely falls.  Specifically, it does not drop outside the uncertainty interval of the original estimate.  And it does not drop to values which would support any contesting of the idea that there is a scientific consensus on AGW.

    0 0
  28. CBDunkerson,

    My ealier post was about the fairness of your analogy, and explicitly did not go into the percentages of the consensus paper, which I will also not touch on this occasion.

    Concerned raises two issues:

    1) Relative to a debated scientific point, papers could potentially be divided into four categories: yes, no, undecided and irrelevant (Y, N, U, I). Concerned suggests that the headline percentage of consensus is best captured by Y/(Y+N+U), where U refers to papers that have actually attempted to consider the question, not merely mentioned a search term. This appears reasonable. He explicitly rejects I from the equation.

    2) He wonders if this affects the cited 97% consensus.


    My comments relate to the first issue. While it may be that Concerned has made mistakes in attributing papers to the categories Y, N, U and I, and that his percentages include papers from the I category, it is not fair to accuse him of misunderstanding the issue so badly that he needs to be told that the consensus ratio should not include I in the denominator. Your analogies suggested he was unaware of this, and in that sense were unfair and distracting.

    0 0
  29. Concerned @25, I have now recalculated the values discussed in my preceding post using the papers rather than author values.  The results are:

    X= Multiplier Value End% Unc%
    Neutral 0.50% 3.82 96.90% 0.29%
    All 0.33% 7.17 96.62% 0.58%
    Adj 193.77% 13.90 96.20% 1.00%

    In each case, in calcuting the endorsement percentages, I rounded the value of X up.  As you can see, even the most generous projection still makes little difference, only reducing the endorsement percentage by 1%.  So, while Cook et al (2013) would have been improved had authors been given the option of an "uncertain" rating, it would have made no difference to the headline results.

    0 0
  30. Thank you all for your comments. I understand now that the number of undecided (or X) opinions is small, so X has minimal impact on the 97% conclusion. Thanks for the calculations, Tom Curtis.

    I read the paper more carefully while composing a response to an AGW denier troll, whose post was deleted (thanks, Moderator) while I was writing. Now I understand where the numbers come from and how the opinion categories were defined. I also note some confusion in the presentation, as I'll describe below.

    In the analysis of abstracts, by "(4a) No position" the authors meant what I was calling "irrelevant" and by "(4b) Uncertain" the authors meant what I was calling "considered but undecided", or X. I think the abstract part of the presentation would have been cleaner if category 4 had consisted of only 4b, so every one of the seven categories was an opinion of sorts on AGW. I think category 4a should have been renamed and set aside. Separation of 4 into 4a and 4b reflects the process used for a second phase refinement of the abstract analysis, but it makes the presentation harder to follow.

    In the case of the self assessment emails, the situation is further clouded by defining category 4 as: "Neutral: paper doesn't address or mention issue of what's causing global warming." This is evidently a rewording of category 4a, but the word "neutral," which could express an opinion, and the phrase "doesn't address or mention" are not the same thing. Thus, in the self-assessments, there is no explicit mention of category 4b, uncertain / considered but undecided / X, but I think there should have been.

    Anyway, all is good with me now. I agree with the 97% for the abstracts, possibly corrected to 96% for for the self assessments.

    0 0
  31. As we see through this thread, it's commonly claimed that 97% of climate scientists accept human caused climate change.
    Climate change deniers and various conservatives have vehemently denied this and it turns out they are right. The 97% number is not correct.

    Excerpt from the most recent issue of Skeptical Inquirer.

    ---
    The Consensus on Anthropogenic Global Warming.
    "In 2013-2014, only four of 69,604 publishing climate scientists rejected anthropogenic global warming. The consensus
    on anthropogenic global warming is not 97 percent, as is widely claimed; it is above 99.9 percent"
    -
    "I used the Web of Science to review the titles and abstracts of peer-reviewed articles from 2013 and 2014, adding the search topic "climate change" to "global climate change" and "global warming."
    Of 24,210 abstracts, only five--one in 4,842 or 0.021 percent--in my judgement explicitly rejected AGW. Two of the articles had the same author, so four authors of 69,406 AGW. That is one in 17,352, or 0.0058 percent.
    This result would allow the claim that 99.99 percent of scientists publishing today accept AGW. To be conservative, I prefer to say above 99.90 percent.
    Excluding self-citations, only one of the five rejecting articles has been cited and that article only once.
    Remember that the 99.9 percent figure does not represent what we usually mean by consensus: agreement of opinion. Rather it is derived from the peer-reviewed literature and thus reflects the evidence therein. It tells us that there is virtually no publishable evidence against AGW. That is why scientists accept the theory.
    The consensus on AGW is not 97%. Instead, publishing scientists are close to unanimous that "global warming is real, man-made, and dangerous" as President Obama put it."
    In another article this author notes: "Anthropogenic global warming is as much the ruling paradigm of climate science as plate tectonics is of geology and evolution is of biology."
    ---
    James Lawrence Powell is executive director of the National Physical Science Consortium. He has been president of three colleges and of the Franklin Institute and the Lose Angeles County Museum of Natural History. He is also a former member of the National Science Board.
    Excerpt from article: "The Consensus on Anthropogenic Global Warming." Published in Skeptical Inquirer, Nov/Dec 2015, pg 42.

    Excellent article and magazine. Check it out.

    0 0
  32. Dar Dedar @31, I do not have access to the Skeptical Enquirer in print so I am going to ask for several points of clarrification, but before that two points of criticism:

    1)  First, the time between obtaining results of a study and seeing it published are in the 1 to 2 year range.  It may take substantially longer to undertake the research leading to the publication.  Therefore a two year publication window may be distorted for this sort of survey simply because of the timing of results.  Therefore I would not accept a firm figure for the consensus for any literature survey with so small a time frame.

    2)  Far more concerning is the evident reasoning.  Specifically, he finds just 4 authors rejecting AGW and concludes that there is 99.9% consensus in favour of AGW.  The thereby assumes that nobody is on the line, ie, undecided on the issue.  Given that Cook (2013) found that 1% of those abstracts indicating a position on AGW were uncertain, that is an unwarranted assumption.

    Worse, JL Powell only considers explicit rejections of AGW.  He does not consider the percentage expressing no opinion which should be excluded from calculalation of the headline result.  Nor does he consider implicit rejections (0.45% of all abstracts; 1.35% of abstracts expressing an opinion).  He has in fact adopted the same flawed strategy of the denier critics of the Cook et al paper who claim the real endorsement level is only 0.54% (the percentage of explicit, numerically quantified endorsements from all abstracts).  The reasoning is no more valid in service of a good cause than in service of a bad one.

    Moving on to the clarrifications, JL Powell excludes duplicate 'skeptical' authors.  Did he also exclude all duplicate authors from the other papers?  Did Powell classify all the abstracts by himself, and if so did he explicitly read all abstracts or classify by word search?

    0 0

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.



The Consensus Project Website

THE ESCALATOR

(free to republish)

Smartphone Apps

iPhone
Android
Nokia

© Copyright 2018 John Cook
Home | Links | Translations | About Us | Contact Us