Thursday, April 04, 2013

Blog Posts Are Not Science

A commenter named Frank popped in here to say:
"David: You correctly complain that Steve McIntyre's blog articles are not peer-reviewed science. They are, however, "peer reviewed" by literally hundreds of climate scientists and highly educated readers who would love to catch Steve McIntyre making a serious mistake."
McIntyre's readership is a self-selected group. Such groups are known to exhibit significant biases, which is why polls taken from them, such as the Internet polls some newspapers now run, are meaningless.

If that population was the group participating in a clinical trial, its results would be laughed out of any journal in existence. For good reason -- it is not representative, and McIntyre's readership certainly does not constitute anything like anonymous peer review from experts.

Blog posts are not science. In certain hands, though, they are harassement, which is mostly what their blogs are about.

23 comments:

Anonymous said...

McIntyre did make some serious mistakes, and it was pointed out many times on numerous blogs. Obviously those who say he hasn't do not read science-based blogs.
-Dan j Andrews

David Appell said...

That's exactly why his results are not up to the standards of peer reviewed science.

He has a blog, nothing more.

Dano said...

I'm amused by what they have to resort to these days. If that is their top thought process now, we should all rejoice.

Best,

D

gallopingcamel said...

McIntyre says that Mike Mann used inverted Tiljander proxies in Mann et al., 2008.
http://climateaudit.org/2009/09/03/kaufmann-and-upside-down-mann/

Who is right? Mann or McIntyre?

Answer the question honestly and you may realize that peer review of "Climate Science" has been corrupted.

Unknown said...

McIntrye says .... and then we are invited to go to McIntyre's site for an "independent" assessment of the evidence?

I think Dave can rest his case.

Frank said...

David: Please don't create straw men. I never claimed that McIntyre's readers were a representative group of climate scientists or interested amateurs. I simply said that AMONG his readers are many active scientists and competent amateurs with a strong incentive to catch him making a serious mistake. Although disagreements abound, the relative absence of unambiguous mistakes and falsehoods noted in comments at CA or posts at other sites like RC (I do read both sides) suggests to me that McIntyre's posts have survived a scientific review. (The bulk of people who comment at CA tend to agree with Steve in general, but often seriously debate and attempt to reproduce controversial aspects of his work.)

On the other hand, the professional peer review given to a journal article is often done by only two anonymous experts in the field of the paper, usually chosen by the editor from a list of qualified reviewers provided by the author of the paper. No reviewer ever attempts to reproduce any of the complicated analysis discussed in the paper. A scientist submitting a paper to PNAS (ie a member of the NAS) chooses his own reviewers, revises his paper if necessary, and submits the paper (along with the reviews) to the editor.

There is plenty of room for disagreement about the relative effectiveness of both types of review. What seems clear to me, however, is that a number of papers McIntyre and company have examined should have received better peer review before they were published.

Furthermore, climate scientists in general are a self-selected group. The current generation were students when concern about AGW developed and many entered the field hoping their work would contribute to "saving the planet". Those involved in actually drafting the IPCC's Summaries for Policymakers and similar documents (rumored to be about 15 from each Working Group) are a more highly select group with the authors of the last report picking the authors of the next report from candidates proposed by governments. The scientist's draft is modified and approved by national representatives, but those representative don't have the knowledge to challenge the scientific expertise of the authors.

David Appell said...

I don't buy it, Frank. You're trying to have it both ways, claiming climate scientists are biased, except for those who read McIntyre's site.

Clearly his readership is self-selected. And frankly most scientists don't put much stock in blog posts; the published literature is their bread and butter.

By the way, AGW is an established fact, so it's natural people interested in understanding it better would go into that field. It's like complaining that people who believe in quarks go into physics.

Blog posts aren't science.

Anonymous said...

If Frank was right, there would be countless examples of peer-reviewed papers being withdrawn or easily rebutted by follow-up papers; and there would be countless examples of blog science being used to produce peer-reviewed papers or their self-evident truth leading to the creation of such papers.

Neither would seem to be the case, although there are some examples of peer-reviewed papers being withdrawn or easily rebutted (all of them, as far as I am aware, being produced by so-called sceptical scientists).
Apart from WUWT Watts having used his website to ultimately produce a peer-reviewed paper (whose result was the opposite of his previous constant claims), i can't think of any other blog science having being worthy of peer-reviewed production. Why not? Not conspiracy, I hope...

Frank said...

David: Of course, everyone who reads McIntyre's site is self-selected. Most of McIntyre's come because they agree with him. However, some of his readers chose to come to try and catch him making a serious scientific mistake or deliberately lying. You know because the go back to their own blogs and write posts trying to prove that McIntyre is wrong. A search for "McIntyre" on RealClimate alone produces 183 results and 10 screens of linked articles! A few scientists even comment at ClimateAudit, while other email information that he posts. I'm asserting that McIntyre's opponents read his blog and provide a form of scientific peer review, especially for people like me who read blogs from both points of view.

If you need further evidence that climate scientists read ClimateAudit and sometimes appear to even "borrow without attribution" from what they read, try these posts. They contain scientist's emails (obtained legally through FOI) discussing what they have read at ClimateAudit and what they will do about it. The paper under discussion was removed from the journal's website (it hadn't yet appeared in print) after discussion of the paper at ClimateAudit.
http://climateaudit.org/2012/10/28/gergis-et-al-correspondence/
http://climateaudit.org/2012/10/30/karoly-and-gergis-vs-journal-of-climate/

Now, let's stop pretending that climate scientists read ClimateAudit and pay attention to its scientific content. Climate scientists may reject most of what they read at ClimateAudit, but they do read and respond. Of course, the published literature remains is scientist's bread-and-butter. If it were freely available, I'd certainly spend more time reading the published literature (research paid for with my tax dollars) rather than blogs.

It turns out that McIntyre is also peer-reviewed by his supporters. As I write, a dozen or more readers have downloaded the data archive from Tingley's latest hockey stick paper and are writing and sharing code to understand precisely how the data was analyzed to produce the key graphs in that paper. It's a group project, a sophisticated version of a journal club in grad school trying to digest the conclusions and limitations of a new paper. Last night, speculation was that thermometers are one of the "proxies" used to "reconstruct" temperature and the blade of the hockey stick comes from the thermometer data. You really should stop by and watch the process. Some people there actually want to know what conclusions should be drawn from the data in Tingley's paper and how Tingley reached his conclusions. Since they have taken apart a dozen or more reconstructions based on many of the same proxies, they sound over-confident about what they will find.

David Appell said...

Frank: McIntyre's incessant need to insinuate less than pure motives to the scientists whose work he writes about, and his endless needling and mocking (as in the title of his most recent post: "More from the Junior Birdmen") isn't worthy of proper scientific scholarship. Many scientists and others simply do not want to get involved with a person who stoops to that kind of discourse -- why would they? Through such tactics McIntyre has chosen to keep himself out of the real game and mostly irrelevant to science. It might get him blog traffic, but it does not get respect for his work, which in any case is often of a very uneven quality (which, if a real scientist did it, would be career threatening; but bloggers pay little price for it precisely because their readership is self-selected and, therefore, highly biased).

gallopingcamel said...
This comment has been removed by a blog administrator.
David Appell said...

Peter Marcombe: Due your blatant lying, your comments are not permitted here.

Frank said...

David: You've changed the nature of your objections to McIntyre's blog. Are you thereby admitting I was correct in asserting that the science in McIntyre's blog is read by many professional scientists and knowledgeable amateurs, who provide a form of peer-review for those who look at a variety of blogs? Or are you simply trying to change the subject?

I agree with your distaste for titles like "More from the Junior Birdmen", as well as the needling and mocking tone. It certainly has hurt McIntyre's credibility many academic climate scientists and made it difficult to get his work published. In one post he wrote about being invited to speak at a dendroclimatology meeting by a session organizer who valued his work, but who was forced to rescind the invitation when no one else was willing to appear at the same session. Does the unwillingness of the paleoclimatology community listen and respond to McIntyre's criticisms say more about McIntyre or about the scientists who are advising our policymakers?

David Appell said...

Frank: No, I still don't think the self-selected readership of McIntyre's blog constitutes anything like peer review.

And I certainly understand why no scientist would want to appear at a session with him.

Frank said...

David: You say that the review McIntyre receives does not "constitute anything like peer review". However, the first sentence in my comment - the comment that prompted this post criticizing me - said: "You correctly complain that Steve McIntyre's blog articles are not peer-reviewed science." However, I have provided evidence proving that climate scientists and knowledgeable amateurs who disagree with McIntyre read his blog and respond in various forums. This provides technically competent readers who read a variety of sources with evidence suitable for forming their own judgments. Although it takes more work, this provides readers with a far better understanding of the issues than journalist like you provide through a magazine like Scientific America. All of the articles on climate change and other environmental issues now seem to come not from scientists, but from journalists linked to or possibly paid by organizations like Climate Wire and Propublica.

The problem with climate science is that policy advocates - who are often also climate scientists - are eager to use the latest scientific new to push for legislation and regulation. As a Steven Schneider, one of the founders of climate science, wrote: "... we are not just scientists but human beings as well. And like most people we'd like to see the world a better place, which in this context translates into our working to reduce the risk of potentially disastrous climatic change. To do that we need to get some broadbased support, to capture the public's imagination. That, of course, entails getting loads of media coverage. So we have to offer up scary scenarios, make simplified, dramatic statements, and make little mention of any doubts we might have." This is politics, not science. In the same passage, Schneider explained how scientists should behave (at least when talking to each other): "as scientists we are ethically bound to the scientific method, in effect promising to tell the truth, the whole truth, and nothing but — which means that we must include all the doubts, the caveats, the ifs, ands, and buts." When journalists like you are dealing with climate scientists who have become policy advocates, you clearly can't expected them to "tell the truth, the whole truth..." as you might expect a physicist or biologist to do. Scientists who are policy advocates need to be treated by journalists like you in the same manner as politicians and lawyers (whose behavior they emulate*); you interview advocates from both sides of an issue and try to present the public a sensible composite of the agreed-upon facts and areas of debate.

*The AGU has now even assembled a rapid-response team to put the alarmist spin any controversy that arises during the 24-hour news cycle.

Frank said...

David: I left a long reply that disappeared for some reason. The first sentence of my comment - the one that prompted this post - explicitly said that ClimateAudit was not peer reviewed science. My point, which you don't address, was that some readers can make reasonable decisions about the reliability of something important they see at ClimateAudit by reading comments there and posts at forums like RealClimate. Both sites often provide links to primary sources.

When I group refuses to listen to a critic, that can say something about the critic and the group. Activist climate scientists want to be the prosecutors of anthropogenic GHGs, the jury, and prevent witnesses from testifying about problems with the evidence.

David Appell said...

Frank: Your comment was only held for moderation, only because I do for comments on all blog posts older than 10 days to reduce comment spam. It's now published.

David Appell said...

Frank wrote:
Activist climate scientists want to be the prosecutors of anthropogenic GHGs, the jury, and prevent witnesses from testifying about problems with the evidence.

This statement is an exact demonstration of the kind of bias that brings you (and others) to ClimateAudit in the first place. It's exactly why the readership there is self-selected, biased, and why comments and discussions there cannot be trusted as anywhere near objective.

David Appell said...

Frank: Your quotation of Stephen Schneider is out of context, and it's WELL KNOWN to be out of context. Shame on you.

Here is the first sentence of that quote that you choose to leave out:

"On the one hand, as scientists we are ethically bound to the scientific method, in effect promising to tell the truth, the whole truth, and nothing but — which means that we must include all the doubts, the caveats, the ifs, ands, and buts."

http://en.wikipedia.org/wiki/Stephen_Schneider#Media_contributions

David Appell said...

The Schneider quote did not come "in the same passage," it came *before* the passage you quoted.

Your quote is completely out of context.

David Appell said...

Frank wrote:
The AGU has now even assembled a rapid-response team to put the alarmist spin any controversy that arises during the 24-hour news cycle.

This also reveals your bias. Apparently you think it's OK for certain people to correct what people are saying about the science, but not for others to do so.

You've turned out to be a great example of the bias I mentioned in the first place. Thanks.

Frank said...

David: If you actually read what I wrote, you would find that I quoted later in my discussion exactly the same passage you claim I omitted. “Shame on you” for falsely accusing me of quoting material out of context.

Our society attempts to arrive at "justice" in law and "best policy" in politics by an adversarial process. No one expects politicians or attorneys to tell the truth, the whole truth and nothing but the truth, etc., as Schneider tells us ethical scientists should. We expect politicians and lawyers to focus on the strong points of their position and do everything possible to hide weaknesses. Society has evolved mechanisms that protect itself from this duplicity: 1) By iron-clad tradition, both sides are always given equal opportunity to present their position or case and respond to opponents. 2) Ethical journalists are expected ask probing questions, to accurately report the arguments made by both sides, to assess the reliability of both, and accept nothing at face value. These protections are essential to the proper functioning of our society.

Scientific “truth” traditionally hasn't been pursued by the adversarial system. Scientists don’t want to waste their time carefully auditing each other's work (the way McIntyre does). To make progress under these conditions, scientists must, as Schneider says, “tell the truth .... including all of the caveats, ifs ands and buts”. Or as Feynman described in Cargo Cult Science, ethical scientists must disclose all of the weaknesses in their own work, because they are the ones who understand those weaknesses best.

Scientists also traditionally convey a shortened version of the "whole truth" to the public. The job of explaining science to the policymakers and journalists is difficult enough without the deliberate obfuscation that is the bread-and-butter of politicians and lawyers. These behavior differences between scientists and politicians have contributed to science’s reputation as one of the most honest and ethical professions. (www.gallup.com/poll/1654/honesty-ethics-professions.aspx)

This is why I accuse activist climate scientists of wanting to be both the prosecutors of anthropogenic GHGs as well as the jury. Since scientists come from a profession trusted to tell the unbiased truth, the protections normally available in adversarial systems have been abandoned, especially by journalists like you. Scientists who disagree with the consensus are denied a forum, because they might interfere with “making the world a better place”. The “caveats, if, and, and buts” normally present in scientific discourse are omitted for the same reason.

Are concerned scientists seeking to make the world a better place supposed to handicap themselves by “telling the whole truth”? Of course not. No citizens in this country have more right to express their opinions to policymakers. Like other policy advocates, they should be free to adopt whatever tactics they think will be most effective. But first, they owe their audience and their fellow scientists an honest statement about whether they are speaking as scientists - whose ethical obligations Schneider has clearly defined - or speaking as policy advocates - who by tradition are free to tell scary stories, make simplified dramatic statements, and hide doubts. There is no ethical double bind: Speak as a scientist or explain why you must speak as a knowledgeable policy advocate: The caveats may be too complicated. The danger from the worst-case scenario or most likely scenario may be too great. But whatever an activist scientist does, they shouldn’t don the mantle of impartial science and drag MY profession into the ethical mud of politics and the law! When activists feel obligated to adopt the tactics of lawyers and politicians, it's time to stop pretending that what comes out of their mouths is science.

David Appell said...

Frank: You divided up Schneider's comment and put his first part last, as if it were an afterthought. That's misleading and of context.

Nothing you've written convinces me my thoughts on CA are wrong -- in fact, you've reinforced them.

Good luck.