Saturday, September 29, 2012

Strained Silver and strange predictions on climate

It's not the usual lopsided intellectual battle we discuss here at Eli's.  This is Michael Mann criticizing Nate Silver's somewhat skeptical take of climate prediction capability in Silver's new book:

It's not that Nate revealed himself to be a climate change denier; he accepts that human-caused climate change is real, and that it represents a challenge and potential threat. But he falls victim to a fallacy that has become all too common among those who view the issue through the prism of economics rather than science. Nate conflates problems of prediction in the realm of human behavior -- where there are no fundamental governing 'laws' and any "predictions" are potentially laden with subjective and untestable assumptions -- with problems such as climate change, which are governed by laws of physics, like the greenhouse effect, that are true whether or not you choose to believe them.
As usual, I'll leave the heavy lifting to someone else, Mann in this case.  Also as usual, I haven't read Silver's book, so maybe there's more to it.  What I can add, however, is that it's helpful to look at climate predictions by Silver himself and by a denialist he credulously supports.

Three years ago, Nate offered to bet climate denialists on a monthly basis over whether the temperature in their hometown was one degree above or below the historical average.  As I said at the link, this was a somewhat aggressive bet offer that could've been vulnerable to letting his opponents rely on a short-term seasonal prediction of colder temps to game the system against him.  It would be interesting to see if he discusses his past bet offer and why he's critical of predictions that are much less affected by random noise.

Second is Silver's enthusiasm for the discredited Scott Armstrong, a crackpot climate denier.  In that case, there was a prediction and a betting market created by Armstrong's fans at InTrade, a skewed and unfair prediction that they still managed to lose spectacularly (link goes to a series of posts on Armstrong and the bet).

Like Mann, I'm a Fan of Nate, but he whiffed on this one.

One more thing:  Nate apparently wrote something about Gavin Schmidt (no relation) and his unwillingness to get involved in betting over climate models.  As someone who is willing to bet over climate, here's my response about climate denialists who won't bet over their predictions:
Of course any particular skeptic might honestly not be interested in betting, but the widespread lack of interest tells you something.
There's a difference between an individual's disinterest in betting versus the widespread disinterest among denialists as a community (with honorable skeptic exceptions) in putting their money where their mouths are.


susan said...

I used to be a Nate Silver fan, but Mike Mann was way too kind on this one. Silver, along with the Obamas and Revkins of this world, to be paying too much attention to the corporate PR spread by the shills and dupes of their world.

David B. Benson said...

Tarnished silver.

Lars Karlsson said...

If you want to see just how crackpotty Armstrong is, look here.

Brian said...

Lars - wow, just wow.

david lewis said...

Dan Kahan has a letter in the latest Nature Climate Change that throws some light on the phenomenon mentioned ("not the usual lopsided intellectual battle") in the discussion here.

Kahan says what he was trying to study was whether "public apathy over climate change" was due to a "deficit in comprehension". He says his study "found no support" for that thought.

One thing he discovered in the process was that "members of the public with the highest degrees of science literacy and technical reasoning capacity were not the most concerned about climate change. Rather, they were the ones among whom cultural polarization was greatest". He concluded: "public divisions over climate change stem not from the public's incomprehension of science but from a distinctive conflict of interest between personal interest individuals have in forming beliefs in line with those held by others with whom they share close ties and the collective one they all share in making use of the best available science to promote common welfare".

Chris Mooney has interviewed Kahan twice for Point of Inquiry.

Naomi Oreskes points out that in earlier days debates with deniers weren't preposterous "lopsided intellectual battles" with types such as Watts, Monckton, or Bastardi. Frederick Seitz was a former President of the NAS. William Nierenberg was a former director of the Scripps Institute of Oceanography. Robert Jastrow was once the director of Goddard Institute for Space Studies. Yet their role in denying the obvious was central.

Lotharsson said...

Kahan et al was proferred by a denialist on the threads at Shaping Tomorrow's World about Lewandowsky et al's paper arguing that conspiracist ideation predicts climate science rejection. (IIRC he ultimately chose to ... depart the discussion rather than abide by the comments policy and his comments were deleted by the moderators.) He cited the paper as support for his claim that the more scientifically literate one was, the less alarmed one was about climate science findings (and his linked claim that non-alarm was the correct evidence-based position).

If my reading of Kahan et al is correct, then I'd suggest one interpret "highest degrees of scientific literacy" very cautiously. The questions measuring scientific literacy peaked at roughly mid-high school level, and the numeracy questions at about high school. Both sets of questions were rolled into a single "scientific literacy" metric, and it's difficult to see how this could reliably distinguish the real-world highest degrees of scientific literacy and numeracy from moderate (i.e. high-school) levels. (The proportion of the sample scoring very highly on these measures looks pretty small too, but the lack of discrimination between genuinely high levels and what others might call moderate seems like a far more serious caveat.)

Anonymous said...

Learning is a slow process if you have to keep re-discovering the same things over and over. Let's recall Robert Park's 2000 book Voodoo Science on the subject of outrageous "scientific" claims and the people who make them and the people who believe them. In Chapter 6 (Perpetuum Mobile) Park takes several pages on one free energy scheme because he went to a promotional event, and thanks to long hot delays spent several hours with the believers.

"It was classic flimflam." (p 129) But finally on page 132:

"It is easy to dismiss the people who packed that stuffy makeshift auditorium in Hackensack for almost five hours as foolish, and even to feel that they deserve to be fleeced. But I came away with the impression that these people were somewhat more knowledgeable about technology than the average citizen, and mistrust of authority is not at all unreasonable; all sorts of outrageous claims are made in the name of science. Extending mistrust of scientific claims to include mistrust of the underlying laws of physics, however, is a reckless gamble. And yet, as we will see in the next section, people who have technical backgrounds and hold highly responsible positions fall into the same trap."

I think the lesson is simply that it is easier for those with a little more knowledge than average to convince themselves that they know more than the real experts. Thus they may be slightly more inclined to believe flimflam and to reject either the basic laws of thermodynamics or their extended application on a planetary scale.

Anonymous said...

The unsigned comment just above is mine -
Pete Dunkelberg

Jeffrey Davis said...

IOW, climate denial coincides with the rise in the disbelief in common welfare (aka civilization). Which coincides with the phenomenon of all these politicians and plutocrats who obviously believe that it's better to reign in Hell than to serve in Heaven.

Who wouldn't rather live in Copenhagen than in Somalia? Well, it turns out that isn't a rhetorical question.

a_ray_in_dilbert_space said...

David Lewis,
I'm afraid I have to agree with Lotharsson and Pete. Kahan's definition of "scientifically literate" is laughable. A C student in high school would be a genius according to him.

I think one of the most important things you can learn in a class is how little you know. Our educational system utterly fails to teach this--in part because the teachers themselves know too little to realize how little they know.

Anonymous said...

Hey a-ray, I'm glad you agree with the gist of my comment, but please don't join the War on Teachers (which is a large part of the War on Education). It's just too easy to roll off a slur like that without having any idea of the larger anti-social forces you're joining.

The very shortest reply to what you said is that there is a wide gulf between what teachers would like to teach and what circumstances require or even allow them to teach. They are required by law to teach their state standards. That alone would keep them busy. Then they have to teach standardized tests in self defense. These are two different things. They also have to devote much time to simply giving tests instead of teaching anything.

Then comes what teachers want to teach: students. Of course teachers want all their students, and perhaps especially those who are struggling, to develop their growing minds in the best way.

random notes


david lewis said...

Rather than get details about Kahan's work from a commenter at SKS try Kahan's Supplementary Material which is freely available from Nature Climate Change. The first interview Mooney did with Kahan for Point of Inquiry is here.

Mooney called Kahan's theory "cultural cognition" and asked him to explain it:

Kahan: "[ I study ] how do we know how the world works. How do we figure out whom to trust to tell us what to believe on all kinds of matters that obviously have an impact on our lives that we're not in a position to figure out for ourselves."

"We have to have some way basically to get the message. Scientists are collecting lots of data. How are we supposed to know what it is that they've found out and whether we can trust it? Cultural Cognition is a theory that connects that process to certain kinds of group commitments and values that people have that basically create networks of confidence and trust. So people 'get the message' from people who are part of this group. And usually the groups - there are different groups and we have different measures for them - usually they converge, i.e. they come to the same understanding about what scientists are saying, and they're right."

Kahan illustrates this point later saying look at the response to H1N1, there isn't some great fault line between left and right over whether people should get vaccinated, there's a stampede to get it, and he claims "most" issues where the public needs to agree on what most of the scientists who are actually in the position to know believe about a certain issue are like this.

"But some of the most interesting conflicts we have are when the groups people are relying on to help them 'get the message' about what science can tell them get their wires crossed".

In the Point of Inquiry interviews Mooney is obviously primarily concerned about if Kahan's work can be useful to those trying to convince the civilization to take climate science seriously. He constantly criticizes Kahan to his face saying there's something wrong with his theory if it doesn't find something uniquely and totally peculiar about the "right" wing of the political spectrum compared to the "left" in his studies, because of the widespread rejection of climate science by supporters of right wing politics in the US which has led to the unified public rejection of climate science by elected Republicans which looks so preposterous to outside observers, i.e. bunnies interested that the planet continue to be a place suitable for carrot growing.

Kahan holds his ground, saying from his point of view, notwithstanding how vital or urgent the issue is, climate is just another issue. For instance, he says, how can the National Academy of Sciences have concluded decades ago as they did that deep geologic disposal of high level nuclear waste is an acceptable method of disposal yet so many in the "left" continue to believe that the scientific consensus on this issue is not what the NAS says it is?

What Kahan is studying is what is similar about the "left" and "right" and other groupings of individuals in how they get it so wrong on what scientists believe about particular issues. He is interested in what can be done about it.

He presents his insight on the climate issue as food for thought.

a_ray_in_dilbert_space said...

Oh, no, I am far from anti-teacher. My complaint has more to do that teachers are often sent in unprepared.

I used to train science teachers in Africa and then in Appalachia. My experience has been that many teachers are scared to death of science, and most of their training does not prepare them to teach science.

Lotharsson said...

"Rather than get details about Kahan's work from a commenter at SKS try Kahan's Supplementary Material which is freely available from Nature Climate Change."

Yes, I had already read it. Similarly I also read the paper itself rather than any interviews with Kahan.

The Supplementary Material did not change my conclusions about lack of ability to discriminate (actual) "high" scientific literacy because the SkS comment accurately reported the questions.

david lewis said...


I can see why you criticize Kahan given his statement in his published letter: "members of the public with the highest degrees of science literacy and technical reasoning capacity were not the most concerned about climate change".

I read that sentence completely differently. I assumed Kahan wasn't looking to separate out persons of genuinely advanced scientific understanding. My reading of him prior to when I read this published letter had convinced me he wouldn't be interested in a group like that.

He's interested in the general public and why they believe what they believe.

His classification schemes are all fairly crude. He chops up the public into groups and tests theories he comes up with.

I thought he was referring to a classification system for scientific literacy he found useful that he could use to rank a random group of the general public into subgroups such as "don't have a clue", "know something about science", and "know quite a bit", or something like that.

I read his sentence in his published letter as a guy saying he's found that the more scientifically literate a subgroup gets the more polarized on climate it is.

I felt Kahan would have a good defense for why he used the questions he did. I became curious to find out more about it after reading your comments.

Kahan cites the National Science Foundation as the source for his set of questions. The questions are taken from Science and Engineering Indicators 2010 Chapter 7, in a subsection entitled "Public Knowledge about Science and Technology". The Kahan questions are a subset of a set of questions that has been used since 1979 by the "GSS" which in the acronyms section is listed as "General Social Survey Survey of Graduate Students and Postdoctorates in Science and Engineering" (Survey Survey is the way it reads). The NSF says something about the GSS sticking with this standard question set so there will be data that is "fully comparable" over time, but they also present other data in the section gathered from answers to other questions that reflect that science evolves over time. Kahan says this 1979 set of questions has been used in many countries.

The NSF report cited states: "in light of using a small number of questions... generalizations about American's knowledge of science should be made cautiously".

But Kahan isn't making generalizations about American's knowledge of science.

He's looking for an explanation for why the American general public is so polarized about specific scientific issues.

I only mentioned him as I've found, after initially being very skeptical of him, that what he's got to say is worth thinking about.

a_ray_in_dilbert_space said...

David Lewis,
I refer to the continuing NSF study as the "Prove Americans are Idiots" study.

One of the most important things you learn--or at least should learn--upon becoming an expert in a field is that expertise matters. A truckdriver has no business pontificating on Electroweak unification even if he can quote Ohm's Law. And what is more, being able to quote Ohm's law gives him no more understanding of the subject than Kim Kardashian.

I would contend that Kahan has merely rediscovered that a little (very little) knowledge can be a dangerous thing.

Lotharsson said...

"I read his sentence in his published letter as a guy saying he's found that the more scientifically literate a subgroup gets the more polarized on climate it is."

I don't, unless that sentence, particularly the concept of "more scientifically literate", is significantly qualified. It's only "more" up to a point, and that point isn't very far along the spectrum from the "completely illiterate" end. Once that qualification is made (and it wouldn't likely be assumed by a random reader who came across his unqualified statement), then I have no problem with it.

(And I note that the denialist citing Kahan was trying to argue that Kahan was evidence that the more scientifically literate one was, the more likely one was to be "not alarmed" about climate science findings - which is waaaaaaaaay off base).

"But Kahan isn't making generalizations about American's knowledge of science."

Well, he's certainly making something that looks very much like a generalisation about the relationship between American's knowledge of science, and something else. In that case the same caution (or qualification) seems to be warranted.