The Non-Expert Problem and Climate Change Science

[Jack Frost]

Interesting read from the creator of Dilbert.

http://blog.dilbert.com/post/154082416051/the-non-expert-problem-and-climate-change-science

I am a non-discriminatory poster that cites PhDs in Atmospheric Science and cartoonists.

Candidly, the cartoonist is not as intellectually engaging but quite entertaining.

 

[Jack Frost]

 

By the way, this has a little something for everybody.

[WidreMann]

He's really attacking a strawman. I mean, I guess it's true that they are people who just say "98% of scientists agree, so there" and I guess there are some skeptics/deniers who actually post something of substance in response. But a lot of them time, the deniers are just making claims about Al Gore or how it snowed last winter and the non-deniers actually point out some very real evidence about change in temperatures or sea ice, etc. The non-experts, of course, can only go so far.

What's more disturbing about his post, howerver, is the idea that expert opinions are bunk because it's really just groupthink and experts are wrong all the time anyway. Of course, he's never taken the time to consider, from his list of examples, exactly who was promoting these ideas and who the actual experts were. A lot of times, there's "science" journalism that completely misrepresents reality and instead of blaming them for misreporting, people blame the experts who were misreported. The problem isn't the experts, it's the non-experts -- the exact opposite of what he's claiming. And while it's true that groupthink is a real thing, it can only go so far in the face of real evidence. The problem again with his argument is that there really is a ton of good evidence that the climate is changing (even skeptics have come to this conclusion, against their better wishes), and that there's also some good evidence that humans are involved. Details vary, of course, and I'm open for different POVs on exactly what will happen and how and on what time frame. Yet another problem with his argument is that he does not say anything about the very real groupthink and identity-based POV behind the denialist movement. What is their motivation? Why do they continue to hold out against mounting evidence and so strongly? Why is nothing convincing, even though they rarely have anything that supports their POV except some sort of idea that maybe the data is wrong or maybe the models are wrong? He says *nothing* about that pernicious problem. Therein lies the real issue: the denialist movement is not a movement of skeptics, or people who have an alternative, but valid view. It's a movement of reactionary nay-sayers. He conveniently ignores this.

Also, the pollsters actually did quite well with the election. They got that Clinton won the popular vote. They were close on most states, and the states that gave Trump the election either weren't being polled at all, or were polling very close (PA). The mistake people made was that they assumed that the country would be as stupid and ass-backwards as to elect someone like Trump, with no relevant skills, no concrete reasonable proposals, built on reactionaryism and nativism. The mainstream thought, apparently wrongly, that even though Hillary had skeletons and wasn't the most likeable person, that people would, at the end of the day, choose someone with experience and well-thought out plans over someone who could even run successful companies and main claim to political fame was leading a movement that denied Obama's right to be president (and was wrong). This is not revelatory of anything except the baseness of humanity. But of course Scott Adams enjoys it because it's always fun sticking it to the system, regardless of the consequences. Same with trying to take down the climate science "establishment" -- at humanity's peril.

[gtg947h]

On 4/22/2017 at 0:32 PM, WidreMann said:

He's really attacking a strawman. I mean, I guess it's true that they are people who just say "98% of scientists agree, so there"

The actual statistic, from the paper (Cook et al, 2013) that apparently started it, is more like "97% of abstracts of climate change studies where a specific conclusion is stated".  Which is about 32% of the total abstracts in the study.   Another 66% of the abstracts do not explicitly take a position; polling of the authors of those papers shows 54% "support the consensus" but doesn't necessarily imply the "primary cause".

Somehow that got turned into "97% of scientists agree" with the clear implication that "humans are the primary/majority cause" was what they're agreeing to.  There's a subtle but important distinction being ignored, and though it doesn't affect the actual science it does leave a bad taste in the mouth. 

And yes, there are people who "just say" that or some equivalent--quite a number of them, really.  The average Joe, no matter their stance on the matter, lacks the scientific (and mathematical/statistical) background to really understand the research or the technical feasibility of proposed solutions--they just believe because they're told to or it fits the beliefs of their "in group".

 

On 4/22/2017 at 0:32 PM, WidreMann said:

What's more disturbing about his post, howerver, is the idea that expert opinions are bunk because it's really just groupthink and experts are wrong all the time anyway. Of course, he's never taken the time to consider, from his list of examples, exactly who was promoting these ideas and who the actual experts were. A lot of times, there's "science" journalism that completely misrepresents reality and instead of blaming them for misreporting, people blame the experts who were misreported. The problem isn't the experts, it's the non-experts -- the exact opposite of what he's claiming. And while it's true that groupthink is a real thing, it can only go so far in the face of real evidence. The problem again with his argument is that there really is a ton of good evidence that the climate is changing (even skeptics have come to this conclusion, against their better wishes), and that there's also some good evidence that humans are involved. Details vary, of course, and I'm open for different POVs on exactly what will happen and how and on what time frame.

Yes, there's mass-market "science" journalism that gets things wrong, or (worse?) dumbs things down to the point that it insults the intelligence of a scientifically-literate reader.  Because let's face it--most of what's presented to the general public is a neat, tidy package.  "This will happen in X years".  It's too clean and convenient.  Minimal or no discussion of uncertanties, potential confounding factors, assumptions, etc. 

But with the bolded part, that's not really discussed.  The overwhelming popular picture is that there's one firm determined scenario, one answer explicitly defined.  We don't get ensemble plots, confidence values, etc. in general media, and that causes problems.

I say this because I know quite a number of engineers--people with a good solid scientific background, and education/training in modeling, data analysis, and simulation--that are actual skeptics, not "deniers"; people saying "okay, that's plausible, but what about...?"  followed by questions on methodology, data collection and reduction, model sensitivity, extrapolation of chaotic systems, etc.  Some have asked "how can someone like that not believe this?" and I think I know why. 

These people aren't climate or weather experts, but they do know enough basic physics and have enough experience creating and running models and analyzing experimental data that the first thing they're going to do when presented with a result being hyped like this is try to poke holes in it.  That's what engineers do; it's how we reconcile ourselves with observed data.  We try everything we can think of to disprove what's there, and if it stands up to that, we move forward. 

Unfortunately, start asking questions like that, and you generally don't get answers.  Rather, you're a "denier".  You're told to shut up, get in line, and believe it because "97% of scientists agree".  This doesn't sit well with engineers, who are all too familiar with things like management/marketing pressure to provide neat, tidy data that looks good and meets the deadline.  Most any engineer will qualify statements with caveats, conditions, and limitations, and the more they're pressed for absolutes, the more qualifications you get.  When we see statements without those qualifications and conditions, our first assumption is that something's being left out, or someone's lying. Or both.  I understand that there's not a lot you can to to answer dumb questions from people with no clue, but I think the valid questions are getting ignored in the noise. 

Same thing goes for a lot of the proposed "green" solutions, actually.  Too many proposals seem to gain traction more because they generate warm fuzzy green feelings than out of technical and economic feasibility; others get tossed out due to irrational public fears.  It frustrates us and leaves us wondering what the actual objective is.

I think it would do a lot of good to address these things and get people like engineers on board.  They'll do a lot better to sell your concept and solutions than research scientists working in a lab or liberal arts majors.

On 4/22/2017 at 0:32 PM, WidreMann said:

Yet another problem with his argument is that he does not say anything about the very real groupthink and identity-based POV behind the denialist movement. What is their motivation? Why do they continue to hold out against mounting evidence and so strongly? Why is nothing convincing, even though they rarely have anything that supports their POV except some sort of idea that maybe the data is wrong or maybe the models are wrong? He says *nothing* about that pernicious problem. Therein lies the real issue: the denialist movement is not a movement of skeptics, or people who have an alternative, but valid view. It's a movement of reactionary nay-sayers. He conveniently ignores this.

It's a salesmanship problem. 

It's because most of the exchange in the public sphere has less in common with scientific discussion and more in common with religious argument, at this point.  If you're in favor, you believe it because it's SCIENCE!!--but you must believe all of it in full, without question, because you're told to, or you're a heretic denier!  And the prophecies predictions of follow-on effects--that is, selling through fear--sound an awful lot like a religious fundamentalist ranting about the Apocalypse.  Either way, a bunch of people are gonna die!  And like a religious push, this approach puts wind in the sails of its base supporters, but is completely counterproductive when it comes to trying to convince those who aren't already on board.

It's certainly not helped by the political associations the issue has taken on; it's well-established at this point that peoples' political views will usually override their normal thought process on a particular issue when it's brought into a political context.  I don't know how it became linked to political parties, but it did.  And once it did, once it became linked to one general political "pole", substantial opposition was all but inevitable.  It's simple politics.

If we want to break the stalemate, it has to become non-political and non-religious.  Unrelated political objectives can no longer be justified under the "climate change!" banner.  Proposed solutions should be driven by economic and technical feasibility, not desired socioeconomic outcomes or punitive measures for politically-unfavored industries.  Any tax or fee schedule (e.g. carbon tax) imposed should come with inviolable requirements that all funds be spent on solutions, not dumped into the general coffers.  Ditch the guilt and fear approach; it doesn't work.  Whenever possible, use solutions and ideas that provide other benefits (economic, physical, etc.)--I'm reminded of an article I read a while back, where the author was confused why the people he was reporting on supported "green" measures but were "deniers".  He correctly guessed the answer--because something tangible was in it for them--but completely missed the boat when he said "they're doing it, but still, we have to make them believe!" Hell, if you have to, sell your plan to dramatically reduce hydrocarbon use as a plan to turn the screws on the Middle East and make the country independent.  It's a twofer.

Unfortunately, it seems that this is a non-starter.  It's apparently as important that everyone be made to "believe" as it is to fix the problem.  And the harder it's pushed like that, the more resistance there will be.

[ORH_wxman]

Much of the climate debate in the media and on the blogs/social media are not what actually is being debated in the scientific journals....and that is unfortunate. Because there's some legit debate in the literature...sensitivity, attribution, longer term extrapolation just to name a few...but generally the problems arise when you get a legit study that finds some range of possibilities....and the media/blogs headline the most extreme scenario for clicks/viewers whether that extreme scenario is very alarmist or very benign depending on the media's demographic audience.

What;s often left out is that the scenario being pimped is on the far end of the distribution curve of potential outcomes. We're then left with strawmen arguments from groups of people who have very little actual knowledge of the scientific literature.

[gtg947h]

13 hours ago, ORH_wxman said:

Much of the climate debate in the media and on the blogs/social media are not what actually is being debated in the scientific journals....and that is unfortunate. Because there's some legit debate in the literature...sensitivity, attribution, longer term extrapolation just to name a few...but generally the problems arise when you get a legit study that finds some range of possibilities....and the media/blogs headline the most extreme scenario for clicks/viewers whether that extreme scenario is very alarmist or very benign depending on the media's demographic audience.

What;s often left out is that the scenario being pimped is on the far end of the distribution curve of potential outcomes. We're then left with strawmen arguments from groups of people who have very little actual knowledge of the scientific literature.

Yep.  And it's all written at a 5th-8th grade level, which is insulting to a lot of people.  There's nothing really in between--either you get the journal article that's almost incomprehensible even to someone with a STEM background who's not a climatologist or meteorologist, or you get general media editorial-as-news.  I think they've gone so far trying to convert the average Joe that they've lost a lot of people who'd otherwise be on their side, and I think even a lot of people without STEM backgrounds can tell they're being talked down to. 

It's almost like science has become the new religion to a large part of the public--it's cited blindly ("because Science says so", "scientists agree that...", "this is Science so it's true" sounds an awful lot like "because God says so", "the Word of God says...", "it's true because the Bible says so") and you're supposed to believe it at face value.  It gets simplified, sloganized, and televangelized.  And like religion, it's astounding how many of the most strident followers/believers/adherents really don't understand it beyond an elementary school level.  We want people to decide and act based on scientific principles and hard data, but then we push it like an act of faith.  "Just believe it because it's true; Science says so".

I'd like to relate a story from the past couple of days as an example.  My wife has never remotely been anywhere on the "denier" side, by anyone's measure, and as part of her degree she took a lot of first- and second-level science classes at Georgia Tech.  I was discussing some of the points mentioned in my previous post--that intelligent people with reasonable questions (remember, asking intelligent questions doesn't mean "you're full of crap", it means "I'm doing my due diligence before I buy into your idea") were being shut down, ridiculed, and lumped in with those who aren't reasonable--and she got to wondering... "where did that '97% of scientists' figure actually come from, anyway?  What scientists, what were they asked, and when was it done?  I know how you can use and abuse statistics, so I want to see what this figure really was."

So, she found the paper (Cook et al, 2013), see my rough summary of it above. And her reaction surprised even me.

"So I looked this up, and I started reading this... this isn't even remotely '97% of scientists agreeing' on anything!  I can't believe how badly this has been misrepresented, by everyone!  I felt hot and my face started burning as I was reading it, because that's not what this says at all and I had bought into it!  That 97% is a subset of journal articles, many of which are written by the same authors!"

"And why is it that every time I read an article on [climate change], I feel like I'm being talked down to like a 5th grader?  I'm not stupid.  I have a good grasp of science, and a really good handle on statistics and how to misuse them, but all I get is the Disneyland version?  I'm on their side, FFS, so why do I feel insulted?"

 

[drstuess]

I think it is the general polarization of most issues which is at fault in sapping the intelligence out of the discussion, and it is a feedback loop, where one side (either denial or alarmist) reports something extreme either for clicks(sensationalization) or dumbing it down. This causes people on the other side to radicalized or shut off rational discussion, which repeats and furthers the cycle.

I think there is also the element that certain people will never listen to science if it conflicts with their convenience, no matter the soundness and neutrality. This creates the opportunity for people in the media to search for and report the answers they and their audience want to hear.

Overall, this creates an environment where people are more likely to sensationalize because they believe if they don't, the other side still will and the consensus will shift away from them.

Sent from my SM-G930V using Tapatalk

[skierinvermont]

His examples of experts being wrong are almost all wrong.

His statement of facts regarding climate change are also wrong.

Not surprising, given he is a non-expert.

[skierinvermont]

On 5/4/2017 at 4:35 AM, gtg947h said:

So, she found the paper (Cook et al, 2013), see my rough summary of it above. And her reaction surprised even me.

"So I looked this up, and I started reading this... this isn't even remotely '97% of scientists agreeing' on anything!  I can't believe how badly this has been misrepresented, by everyone!  I felt hot and my face started burning as I was reading it, because that's not what this says at all and I had bought into it!  That 97% is a subset of journal articles, many of which are written by the same authors!"

"And why is it that every time I read an article on [climate change], I feel like I'm being talked down to like a 5th grader?  I'm not stupid.  I have a good grasp of science, and a really good handle on statistics and how to misuse them, but all I get is the Disneyland version?  I'm on their side, FFS, so why do I feel insulted?"

 

Yes it's a subset - it's a subset where authors actually have to respond and correct in response to peer-review. And yes, you do see a lot of the same authors. But there are probably 1000s of authors total publishing articles consistent with AGW. Some are more prolific. When you become more familiar with the issue you become more comfortable with this fact because you see these authors correcting the mistakes of some fringe authors that publish in fringe journals without real peer-review and the errors by the fringe authors never get corrected.

I've read lots of peer-review... I've never found anything in a journal with a proper peer-review process that is skeptical of AGW. So at least from my personal experience I have no problem accepting that 97% statistic. In my experience, it is more like 100%. 

Authors like Spencer and Christy published nonsense in "journals" and blogs all the time but then when they actually go through peer-review process it gets cleaned up and the conclusion changes because they are forced to think critically about their own work.

[gtg947h]

On 5/8/2017 at 6:36 PM, skierinvermont said:

Yes it's a subset - it's a subset where authors actually have to respond and correct in response to peer-review. And yes, you do see a lot of the same authors. But there are probably 1000s of authors total publishing articles consistent with AGW. Some are more prolific. When you become more familiar with the issue you become more comfortable with this fact because you see these authors correcting the mistakes of some fringe authors that publish in fringe journals without real peer-review and the errors by the fringe authors never get corrected.

I've read lots of peer-review... I've never found anything in a journal with a proper peer-review process that is skeptical of AGW. So at least from my personal experience I have no problem accepting that 97% statistic. In my experience, it is more like 100%. 

Authors like Spencer and Christy published nonsense in "journals" and blogs all the time but then when they actually go through peer-review process it gets cleaned up and the conclusion changes because they are forced to think critically about their own work.

But the problem is that the 97% statistic is wrong.  What is being quoted as a fact ("97% of scientists") actually is not.  This is a problem because the entire point is to show "we have good hard data that is unimpeachable"; but if we then you go and oversimplify your data like this and try to make it sound like something it isn't... what does it say about everything else?  "Oh yeah, well, technically this isn't really the right number, but it doesn't really matter, our other numbers are good, we promise!" 

I quote from the abstract of the paper:

Quote

We analyze the evolution of the scientific consensus on anthropogenic global warming (AGW) in the peer-reviewed scientific literature, examining 11,944 climate abstracts from 1991-2011 matching the topics "global climate change" or "global warming".  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.  In a second phase of this study, we invited authors to rate their own papers.  Compared to abstract ratings, a smaller percentage of self-rated papers expressed no position on AGW (35.5%).  Among self-rated papers expressing a position on AGW, 97.2% endorsed the consensus.  For both abstract ratings and authors' self-ratings the percentage of endorsements among papers expressing a position on AGW marginally increased over time. Our analysis indicates that the number of papers rejecting the consensus on AGW is a vanishingly small proportion of the published research.

Of note is that the "no position" category includes things like "while the extend of human-induced global warming is uncertain...". 

Give someone this paper, including the caveats and the methodology, and it's a lot more convincing than a dumbed-down "97% of scientists agree".  And that's my point--the whole thing has gotten so dumbed-down that it is turning off people who know about data collection, modeling, lying with statistics, and so on. 

[skierinvermont]

On 5/12/2017 at 9:04 AM, gtg947h said:

But the problem is that the 97% statistic is wrong.  What is being quoted as a fact ("97% of scientists") actually is not.  This is a problem because the entire point is to show "we have good hard data that is unimpeachable"; but if we then you go and oversimplify your data like this and try to make it sound like something it isn't... what does it say about everything else?  "Oh yeah, well, technically this isn't really the right number, but it doesn't really matter, our other numbers are good, we promise!" 

I quote from the abstract of the paper:

Of note is that the "no position" category includes things like "while the extend of human-induced global warming is uncertain...". 

Give someone this paper, including the caveats and the methodology, and it's a lot more convincing than a dumbed-down "97% of scientists agree".  And that's my point--the whole thing has gotten so dumbed-down that it is turning off people who know about data collection, modeling, lying with statistics, and so on. 

 

So the one critique you make of the paper actually is quite good until you read the paper really closely. If the "no position" category includes a lot of papers that say things like "while the extent of human-induced global warming is inconclusive..." that would be a major problem with the paper. Because that's not really "no position." Saying there's that much uncertainty is an implicit rejection of the scientific consensus which says there's not that much uncertainty. When the authors aggregated positions 1-7 into 3 categories position 4b should have had its own "inconclusive" category or been included in the "Reject" category. I'm not sure why they separated position 4b from position 5.

 And after my first reading of the paper I agreed with you. It's entirely possible that a big chunk of the 66% "no position" is actually papers that implicitly disagree with the consensus. 

However, after reading carefully, you'll see that the caption for Figure 1 states that only .5% of the 66% no position is the "uncertain" 4b position. .5% of 66% is .3%. If we lump the .3% that are uncertain or inconclusive about AGW in with the .7% that reject it we get 1% of papers that reject the consensus. This also increases the % of papers reject to 2.9% from 1.9% out of papers that take a position. 



So, after reading carefully, I find the 97.1% of papers simplified statistic to be a good representation of the findings of the paper. It's also consistent with the "self ratings" of authors and with the findings of other studies.

 

I would generally agree the science reporting is often overslimplified, exaggerated, and/or misrepresented. But this doesn't appear to be a good example of that phenomenon. I find the 97.1% statistic to be a good representation of the paper and of the truth. From personal experience I find the % to be closer to 100%.

[bobbutts]

I don't need to read any more of his drivel after reading several during election season.  Scott Adams does a decent job selling alt-right bs, that's about it.