follow CCP

Recent blog entries
popular papers

Science Curiosity and Political Information Processing

What Is the "Science of Science Communication"?

Climate-Science Communication and the Measurement Problem

Ideology, Motivated Cognition, and Cognitive Reflection: An Experimental Study

'Ideology' or 'Situation Sense'? An Experimental Investigation of Motivated Reasoning and Professional Judgment

A Risky Science Communication Environment for Vaccines

Motivated Numeracy and Enlightened Self-Government

Making Climate Science Communication Evidence-based—All the Way Down 

Neutral Principles, Motivated Cognition, and Some Problems for Constitutional Law 

Cultural Cognition of Scientific Consensus
 

The Tragedy of the Risk-Perception Commons: Science Literacy and Climate Change

"They Saw a Protest": Cognitive Illiberalism and the Speech-Conduct Distinction 

Geoengineering and the Science Communication Environment: a Cross-Cultural Experiment

Fixing the Communications Failure

Why We Are Poles Apart on Climate Change

The Cognitively Illiberal State 

Who Fears the HPV Vaccine, Who Doesn't, and Why? An Experimental Study

Cultural Cognition of the Risks and Benefits of Nanotechnology

Whose Eyes Are You Going to Believe? An Empirical Examination of Scott v. Harris

Cultural Cognition and Public Policy

Culture, Cognition, and Consent: Who Perceives What, and Why, in "Acquaintance Rape" Cases

Culture and Identity-Protective Cognition: Explaining the White Male Effect

Fear of Democracy: A Cultural Evaluation of Sunstein on Risk

Cultural Cognition as a Conception of the Cultural Theory of Risk

Main | TEDx in Vienna »
Tuesday
Dec042018

More *doing* science communication on science of science communication

 

Read the essay here.


PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (70)

Dan,

What intellectual capacities—or if one prefers, cognitive virtues—should the citizens of a modern democratic society possess?

Probably OK intro sentence for demographics of SciAm readers, but this does begin to sound a bit close to a plea for voter literacy testing. Because "should...possess" could either be stand-alone normative or indicating improvement in function. Maybe I'm just sensitive since still recovering from the recent election's abundance of fraud claims.

What an ordinary member of the public thinks about climate change, for example, has no impact on the climate. Nor does anything that she does as a consumer or a voter; her individual impact is too small to make a difference. Accordingly, when she is acting in one of these capacities, any mistake she makes about the best available scientific evidence will have zero impact on her or anyone she cares about.

Problem with this is that politicians are even more polarized than their constituents on these topics, and they do have power.

One doesn’t have to be a Nobel prizewinner to figure out which position one’s tribe espouses.

True, but open to criticism that many in US don't accurately figure that out.


Missing was any indication of how uncommon science curiosity is.

December 4, 2018 | Unregistered CommenterJonathan

new SEP page on reproducibility:

https://plato.stanford.edu/entries/scientific-reproducibility/

December 4, 2018 | Unregistered CommenterJonathan

Obviously just wasting my breath at this point, but....

Simply put, as ordinary members of the public acquire more scientific knowledge and become more adept at scientific reasoning, they don’t converge on the best evidence relating to controversial policy-relevant facts. Instead they become even more culturally polarized.

Given that, as far as I have been able to tell, you don't have longitudinal data to fit the pattern you explicitly describe, I still fail to understand why you choose such phrasing.

But given that you are consistent in following this practice, I have to assume there is something going in here that I can't understand.

December 4, 2018 | Unregistered CommenterJoshua

If we had a less gerrymandered, more perfect democracy, on key issues of controversy, this would not be true:

"What an ordinary member of the public thinks about climate change, for example, has no impact on the climate. Nor does anything that she does as a consumer or a voter; her individual impact is too small to make a difference. "

The point, at any rate, in participating in a democracy is that ones vote, when combined with the votes of like minded individuals, can make a difference. And that sometimes, one individual making a well supported principled stand, can make all the difference.

This past election had a number of very close votes. A near local to me example: https://www.coloradoindependent.com/2018/11/08/first-transgender-colorado-arvada/

This is reflective of the fact that we are in the midst of cultural transitions, away from a reliance on fossil fuel energy and old line industrial processes and towards new structures, not as of yet highly defined. And that our society is generally more willing to acknowledge and empower it's diversity,. But also, substantial portions of our citizens favor retention of the status quo.

There is a lot more at stake here for individuals than "identity". Our socio-economic structures are being redefined. I don't think that we are going to make progress unless we are willing to acknowledge that some individuals have a lot to lose in this transition, and make accommodations accordingly. For example: https://www.theguardian.com/environment/2018/oct/26/spain-to-close-most-coal-mines-after-striking-250m-deal

Cultures offer an established version of history and an expected vision of the future. In the US, because we are boxed in to a choice between "D" and "R", we tend to think of that as offering choices between sides on major issues. In a time of cultural change, and in one in which the reins of power are largely held by a wealthy and powerful few, this is less true. In the past election, key groups of swing voters were those who did not vote in the past election, or had voted for Obama and then Trump. A full third of the voters here in Colorado are registered as Independents. this rejection of the established parties does not put them half way between D and R. If we are searching for explanations as to how people navigate contentious issues, we need to use a political spectrum that is more in tune with where people actually are.

December 4, 2018 | Unregistered CommenterGaythia Weis

they don’t converge on the best evidence relating to controversial policy-relevant facts.

Is that uniformly true? For example, how about HIV causes AIDS, or species evolve without divine intervention, or women who are raped aren't likely to get pregnant, or abortion puts women at higher risk for mental health illnesses, or vaccines cause autism, or homosexuality is genetically determined, etc.?

December 4, 2018 | Unregistered CommenterJoshua

What a coincidence that this drops on my radar the same day as Dan's SciAm article!:

Rethinking the link between cognitive sophistication and identity-protective bias in political belief formation - From Dan's frenemies Gord Pennycook and David Rand:

https://psyarxiv.com/yuzfj/

In this paper, we investigated the hypothesis that cognitive sophistication facilitates identity- protective processing in political belief formation. Our findings suggested that cognitively sophisticated individuals deferred more to their prior beliefs—rather than to their political identities per se—when reasoning about information in the political domain. Furthermore, benchmarked against a Bayesian agent, we found evidence that these individuals were overall less—not more—biased in their belief updating after receipt of such information. These results highlight a somewhat more optimistic perspective on the role of cognitive sophistication in political belief formation than prior work: That cognitive sophistication may be deployed to assess and integrate new evidence in light of what the person currently believes to be true, rather than to disregard and resist identity-threatening evidence per se. From a practical perspective, however, deference to prior beliefs may be similarly problematic for the prospect of achieving convergence on true beliefs in politics. One factor determining this assessment is the quality of the information environment (e.g., on social media, claims made by politicians). Where prior beliefs about political issues are constructed on the basis of misinformation and bad evidence, deference to prior beliefs will cement false beliefs. The upshot highlights the paramount importance of safeguarding the integrity of the information environment to which people are exposed over the long-term.

vl;ossf (very long, only skimmed so far) - But, MTurk trigger warning applies to some of the studies. However, note that one of the findings of Many Labs 2 was that good replicable results shouldn't fear the MTurk (or worry that much about WEIRD).

December 4, 2018 | Unregistered CommenterJonathan

Jonathan -

Thought you might like this.

https://www.journals.uchicago.edu/doi/10.1086/697253

December 4, 2018 | Unregistered CommenterJoshua

Jonathan -

Speaking of which....

Was listening to latest Klein podcast with a con about immigration..

They mentioned briefly that the trends of more divided party-based partisanship likely leads to a mis-perception of the extent to which people are partisan (i.e., people are more than the sum of the parts of their party affiliations). Maybe there is a limit to the value of information about how demz and conz are more likely to think that their counterparts are devils incarnate.

I want to think about that some more. I think there's something to it.

December 4, 2018 | Unregistered CommenterJoshua

Joshua,

Yes - will read that - thanks. Have read other things like it - and you probably guessed was partially behind my point above that Americans don't really understand the positions/compositions of their respective tribes (and more so about outgroup tribes - especially considering that many stances are reactionary). Would be nice to know if higher cognitive abilities enhance or reduce these position/composition mistakes. Probably reduce?

December 4, 2018 | Unregistered CommenterJonathan

Jonathan: 'But, MTurk trigger warning applies to some of the studies.'

Aside from this, the example questions of study 1, the open mindedness test, the election results test, do not seem to be probing direct views of strongly culturally conflicted science issues that, in theory at least, have an evidence based position which IPC skews for each side. For such highly polarised / contested science issues in the US, likely 'prior belief' about the issue will be much more aligned to political side anyhow. Are they even measuring the same thing, or issues that are more tractable to cognition due to their indirect / lighter weight nature?

December 4, 2018 | Unregistered CommenterAndy West

Andy,

Yeah - noticed that too. The questions are historical sociopolitical, not scientific. I wonder if they did this because it is easier to get a larger supply of these. Haven't read far enough to see how they determine prior belief and such - am under assumption that they are able to separate prior belief from identity cognition somehow...

My initial guess is that historical sociopolitical questions are less likely to show cognition effects. Having high CRT doesn't imply to me that one knows any more about how health care premiums behaved under the ACA.

December 4, 2018 | Unregistered CommenterJonathan

Jonathan, thanks, my own reading quick / shallow so far, but at least I know it's not just me thinking along those lines 0:

December 4, 2018 | Unregistered CommenterAndy West

Joshua,

https://www.psypost.org/2018/12/study-finds-large-shift-towards-acceptance-of-evolution-among-mormon-undergraduates-52694

(contains longitudinal goodness)

December 4, 2018 | Unregistered CommenterJonathan

The "science curiosity" model's explanatory power completely fails in the recent events in France. Demonstrators include both right and left (from Rassemblement National to Anarchists) but notably no POCs, who wisely stay away, fearing the all-white crowd will unite against them if they show up.

The "climate-protecting" fuel tax increase was suspended, but I predict the uprising will continue and spread.

December 5, 2018 | Unregistered CommenterEcoute Sauvage

NB POC=person of color

December 5, 2018 | Unregistered CommenterEcoute Sauvage

One problem I see with studies 1 and 2 in the Tappin et al. paper is that, in their normative Bayesian model, the researchers do not consider the (non)independence of evidence. They assume that every subject will (or should) perceive the signal as being completely independent of all evidence the subject will have used to establish their prior. They also provide no means for subjects to determine the independence of the signal vs. their prior evidence (such deprivation is typical in real world cases as well).

This is where the nature of the questions as real/fake news items instead of science might really make a difference. The nature of news items is such that in most cases if one is familiar with the item, a signal that agrees with one's prior will often not be independent of that familiar prior evidence. Furthermore, if one is both familiar with the item and with debate about its truth, then both types of signals (agreeing/disagreeing) will often not be independent of prior evidence. Subjects in these cases - in order to be optimal Bayesians - have to consider these possibilities somehow. This should cause overweighting of priors.

So, what does this mean vs. their findings? Tappin et al. find that higher CRT folks are more likely to be Bayesians that apparently don't do much of the above signal filtering. Is it that the news items used all sound novel to these subjects? Or, are they aware (higher CRT -> more aware) that they can't fulfill the task in an ideal manner due to this issue and instead merely go through the motions (higher CRT -> more able to perform such motions to researchers satisfaction)? Or, are lower CRT folks trying harder to do the signal filtering?


Another problem with those first 2 studies is that the authors don't consider the formation of a strong novel prior during the test for subjects that have never encountered the news item before to be indicative of partisanship instead of proper Bayesianism. They assess proper Bayesianism using only the difference between the prior and the posterior. But, especially for subjects new to the news item that demonstrate high priors, the bias is likely baked into the prior. If, upon hearing for the first time that Trump declined a handshake from a disabled kid at an anti-ACA rally, I immediately produce a strong prior, then any small but proper Bayesian updating I do that follows isn't saying much about the control my partisan identity has over my belief formation.

December 5, 2018 | Unregistered CommenterJonathan

Matthew Motta et al. on flu vaccine myth debunking:

https://theconversation.com/countering-misinformation-about-flu-vaccine-is-harder-than-it-seems-106479

December 6, 2018 | Unregistered CommenterJonathan

Dan's SciAm article relies on arguments which appear obvious at first sight, but fail to account for game theory. This becomes clearer by an example from the barrage of exhortations on climate change - which after half a century has signally failed to show any practical success. We read:

https://www.washingtonpost.com/national/health-science/a-kind-of-dark-realism-why-the-climate-change-problem-is-starting-to-look-too-big-to-solve/2018/12/03/378e49e4-e75d-11e8-a939-9469f1166f9d_story.html

"...“If you’re driving on a highway and the car in front of you stops short, and you slam on brakes and realize that you’re going to hit the guy no matter what, that’s not the time to take your foot off the brake,” said John Sterman, a professor of management at the Massachusetts Institute of Technology’s business school. “And you certainly don’t step on the gas.”"

That's not at all obvious. If the car stopping suddenly ahead is a PC-type Prius or other small car, I'm in a big Mercedes (steel box passenger compartment, safety belt, airbag that works) I will initially and instinctively press on the brake, true. But in the next millisecond, figuring I'll be responsible for the crash no matter what - and suspecting all drivers of PC cars of being scammers to begin with, probably out to collect from my insurance the replacement price of their vehicles - I will take my foot of the brake and total the car ahead. I'm certainly not going to risk a skid, especially on a wet road, in order to minimize damage to the other car and driver.

What appears to be obvious rarely is. And the singular ineffectiveness of the 97% consensus message is due to the fact green technologies are scams.

December 6, 2018 | Unregistered CommenterEcoute Sauvage

In Tappin et al., does figure 6B center (pg 43) look strange to anyone else? Where did those climate-skeptics-more-open-minded prior folks with high CRT come from such that they don't influence 6A significantly? Are they a very small group compared to the rest? Maybe will have to dig into supplemental materials for answers...

December 6, 2018 | Unregistered CommenterJonathan

Maybe will have to dig into supplemental materials for answers...

Of the 1201 subjects, 208 are climate-skeptics-more-open-minded prior folks and 79 answered more than half of the CRT questions correctly. This goes down to just 7 that answered all CRT questions correctly - 3 in one treatment, 4 in the other. So, maybe 6B center looks different because underpowered? For comparison with 6B left, there are 680 believers-more-open-minded prior folks and 354 answered more than half of the CRT questions correctly, and this goes down to 45 that answered all correctly.

December 7, 2018 | Unregistered CommenterJonathan

"So, maybe 6B center looks different because underpowered?"

Presumably that's why the error bars are wider on the middle graph in 6B?

December 7, 2018 | Unregistered CommenterNiV

One reason why, but not a full explanation - note the 95% CI regions in 6A middle are just as wide. Also, the regions in 6B middle don't explain the skew.

December 7, 2018 | Unregistered CommenterJonathan

"Another problem with those first 2 studies is that the authors don't consider the formation of a strong novel prior during the test for subjects that have never encountered the news item before to be indicative of partisanship instead of proper Bayesianism. They assess proper Bayesianism using only the difference between the prior and the posterior."

There's a lot of more complex hypotheses they don't consider. Problem is, the Bayesian analysis starts getting messy the more possibilities you incorporate.

Let's take a still very simple example of a test between two hypotheses about the real world (H1 and H2) where the observational sources can be reliable or unreliable (Hr and Hu). In the simple Bayesian model, the reliability of the source appears in the likelihood calculation, telling you what P(Obs|H) is. Biased and unbiased sources have different probabilities of reporting each outcome assuming each hypothesis. But the usual Bayesian framework has no way of dealing with uncertainty about the statistical model, let alone updating it.

However, we can incorporate the uncertainty about the observer model by incorporating it into the hypothesis. For example, suppose we take four hypotheses Hr1 = "source is reliable and real world is in state 1", Hr2 = "source is reliable and real world is in state 2", Hu1 = "source is unreliable and real world is in state 1", and Hu2 = "source is unreliable and real world is in state 2".

We can do Bayesian updating using this model with the following scheme:

P(Hr1|E) = [P(Hr1)/P(E)] P(E|Hr1)
P(Hr2|E) = [P(Hr2)/P(E)] P(E|Hr2)
P(Hu1|E) = [P(Hu1)/P(E)] P(E|Hu1)
P(Hu2|E) = [P(Hu2)/P(E)] P(E|Hu2)

Assuming the hypotheses are exhaustive and mutually exclusive, we can just multiply our vector of priors by the vector of likelihoods, and normalise so the sum of probabilities is 1.

If the source is reliable then we get a good measurement - the meter reads 'a' if H1 is true, and 'b' if H2 is true, with 90% reliability.

If the source is unreliable, (a partisan for H2,) then they're going to fake the data to look as if H2 was true whatever the real state of the world.

This gives us the following likelihoods.

P(E=a|Hr1) = 0.9
P(E=a|Hr2) = 0.1
P(E=a|Hu1) = 0.1
P(E=a|Hu2) = 0.1

P(E=b|Hr1) = 0.1
P(E=b|Hr2) = 0.9
P(E=b|Hu1) = 0.9
P(E=b|Hu2) = 0.9

So if we get lots of 'a's in our sequence of observations, the vector of beliefs converges on (1,0,0,0). We conclude that the source is reliable, and H1 is true.

If we get a lot of 'b's in our sequence of observations, the vector of beliefs converges on (0,1/3,1/3,1/3), we think H2 is probably true, but we are in severe doubt over the reliability of the source. Our prior beliefs are weak.

Now we come across a brand new source. We know nothing about them, whether they're reliable or unreliable. We split our four hypotheses into eight, and expand our table of likelihoods as follows:

P(E=a|Hrr1) = 0.9
P(E=a|Hrr2) = 0.1
P(E=a|Hur1) = 0.9
P(E=a|Hur2) = 0.1

P(E=a|Hru1) = 0.1
P(E=a|Hru2) = 0.1
P(E=a|Huu1) = 0.1
P(E=a|Huu2) = 0.1

P(E=b|Hrr1) = 0.1
P(E=b|Hrr2) = 0.9
P(E=b|Hur1) = 0.1
P(E=b|Hur2) = 0.9

P(E=b|Hru1) = 0.9
P(E=b|Hru2) = 0.9
P(E=b|Huu1) = 0.9
P(E=b|Huu2) = 0.9

So in the case where we got lots of 'a's before and converged close to (1,0,0,0), and the new source reports 'a' too, our vector of prior beliefs is (1/2,0,0,0,1/2,0,0,0) which we multiply by the top half of the table, and upweight the first and downweight the fifth. We're going to converge on (1,0,0,0,0,0,0,0). Both sources are reliable and H1 is true.

However, if the new source reports 'b', then we multiply by the bottom half of the table, and downweight the first hypothesis while upweighting the fifth. The first source is reliable, the second is not, and H1 is still true! We have discounted the new evidence based on our prior beliefs!

Now consider what happens if the original source reported a long sequence of 'b's. Our prior belief is that H2 is probably true, but we're a lot more open-minded. Our vector of prior beliefs is (0,1/6,1/6,16,0,1/6,1/6,1/6). The new source reporting 'a' multiplies by the top half of the table, and we see only Hur1 is still upweighted. We conclude the first source was unreliable, the second reliable, and H1 is true. We have allowed the new evidence to change our mind.

However, if the new source also reports 'b', we multiply by the bottom half of the table, the only option newly downweighted is Hur1. The first source has gained credibility because of the confirmation by a second source.

Of course, this scheme is still far too simple to represent real human thought processes. Unreliable sources are all assumed partisans for one side rather than having partisans on both sides, and there should be a continuous spectrum of reliability. Partisans may be reliable on some questions and not on others, depending on other recognisable features of the question. Hypotheses may be extremely complex, and a long way from the simple H1/H2 choice at the core of the decision.

The Tappin paper is very interesting, in that it is moving in the direction of constructing a more complex and realistic model of Bayesian reasoning to compare with, but it's still a long way short of considering all the possibilities. I'd also be a lot less happy about the differences between figures 6 and 7 than they are. Yes, both separately fit their theory, but it's strong evidence that the populations being sampled are not independent of the variables under study, and that puts everything else into question. Of course, we can find plenty of possible explanations that don't threaten their theory, but the question needs to be answered, not hand-waved away. Apart from that, though, (and bearing in mind I've only skimmed it too,) very good!

December 7, 2018 | Unregistered CommenterNiV

Amusing paper on the frequentist vs. Bayesian (and others) argument:

https://psyarxiv.com/ue5wb/

December 7, 2018 | Unregistered CommenterJonathan

https://fivethirtyeight.com/features/psychologys-replication-crisis-has-made-the-field-better/

December 7, 2018 | Unregistered CommenterJonathan

prediction markets and peer belief survey results for Many Labs 2: scientists are pretty good chicken sexers (though probably not as good as Dan thinks lawyers are):

https://psyarxiv.com/sgjaz/

December 7, 2018 | Unregistered CommenterJonathan

Real-time testing of public opinion on carbon taxes now underway in Paris - courtesy of Russian TV
https://www.youtube.com/watch?v=kMN2TkDlhmM

December 8, 2018 | Unregistered CommenterEcoute Sauvage

Another live feed, also RT but via German transmission - the French government is interfering with the local stream, but the Germans aren't - so far. Fascinating.
https://www.youtube.com/watch?v=wzdxEM8Gh2s

December 8, 2018 | Unregistered CommenterEcoute Sauvage

....Live feed continuing uninterrupted on German RT link - posters on chat stream, blocked by the French, have now joined the Germans. Posts in Italian, Dutch, Spanish, English are also appearing intermittently - all in support of the gilets jaunes in Paris.

December 8, 2018 | Unregistered CommenterEcoute Sauvage

Ecoute -

Condolences. Such an injustice.

https://www.washingtonpost.com/local/public-safety/jury-set-to-begin-deliberations-in-james-a-fields-jr-car-ramming-trial/2018/12/06/65d38748-f9b3-11e8-8c9a-860ce2a8148f_story.html?utm_term=.ef2243c76b66

December 8, 2018 | Unregistered CommenterJoshua

Podcast discussion of the human "justification machine" - much of which is closely related to motivated reasoning:


https://art19.com/shows/the-ezra-klein-show/episodes/5bce189f-80ba-4893-bb53-7b82c90cdaed

December 8, 2018 | Unregistered CommenterJoshua

Another all-white crowd gathered today, this time in London. Protesting migrant invasion of Europe and government failure to deliver on Brexit vote (of course caused by aforementioned migrant invasion). Police anticipates possible violence by counter=demonstrators, few of whom are visible so far unless those waving foreign flags count.
https://www.youtube.com/watch?v=Nm1Szb0gTgk

December 9, 2018 | Unregistered CommenterEcoute Sauvage

Ecoute -

Do you align yourself with Tommy Robinson, and the EDL?

If so, that would indeed help explain how you would argue that antisemitism had nothing to do with the man who went into a synogogue and mass murdered Jews.

December 9, 2018 | Unregistered CommenterJoshua

Will the reading-comprehension-challenged here kindly note there's no pro-Robinson demonstration, there's a pro-Brexit demonstration. There's also a counter-demonstration, backed by "multiculturals" and groups with names including "anarchist feminists". Unlike the pro-Brexit crowd they include assorted Antifa with faces covered.

Real-time updates for the Antifa march on The Guardian by reporter Haroon Siddique.

December 9, 2018 | Unregistered CommenterEcoute Sauvage

Perhaps clicking on links requires instructions? Robinson is speaking now; his main supporters are front and center waving large Israeli flags for reasons unclear - it's a pro-Brexit rally. But here is the previous link again, in case:
https://www.youtube.com/watch?v=Nm1Szb0gTgk

December 9, 2018 | Unregistered CommenterEcoute Sauvage

Ecoute -

Do you align yourself with Tommy and the EDL?

December 9, 2018 | Unregistered CommenterJoshua

Ecoute -

I'm still trying to figure out how you can believe that antisemiticism had nothing to do with Bowers' mass murder of Jews in a synogogue.

If you do align with the EDL and Tommy, that might help explain your reasoning.

December 9, 2018 | Unregistered CommenterJoshua

Robinson's financial support almost exclusively originates with neocons. He's a convicted fraudster barely tolerated on the Right.
https://www.theguardian.com/uk-news/2018/dec/07/tommy-robinson-global-support-brexit-march

December 9, 2018 | Unregistered CommenterEcoute Sauvage

Thx. How about the EDL?

December 9, 2018 | Unregistered CommenterJoshua

Also, I was wondering what criteria you might use to distinguish "barely tolerated on the right" from "supported by the right."

December 9, 2018 | Unregistered CommenterJoshua

On "science of science communication", humorous updates from the ongoing UN global warming gathering in Poland - held at the heart of the Silesian coal basin, in a convention hall tastefully decorated by mounds of the local product:
https://www.nbcnews.com/news/world/protesters-demand-tougher-action-global-warming-during-u-n-climate-n945691

"A group wearing polar bear costumes was expelled from the march after suggesting that fossil fuels should be replaced by nuclear power, a technology that many environmentalists object to."

From same link, this other gem:
"The Climate Action Network, an umbrella group for environmental organizations, on Saturday gave its Fossil of the Day award to the United States after Washington's diplomats objected to linking human rights to climate change."
---------------------------------------

Linking human rights to migration to climate change?! A guaranteed winner - hope it's adopted by the Left.

December 9, 2018 | Unregistered CommenterEcoute Sauvage

The underlying message is an inclusive and uplifting one. In an era of increasing polarization and identity politics, our findings may positively contribute to the societal conversation and reinforce the conviction that good things happen when people of different backgrounds, cultures, and ethnicities come together to work towards shared goals and the common good.

https://www.nature.com/articles/s41467-018-07634-8

December 9, 2018 | Unregistered CommenterJonathan

Jonathan -

Would that provide support, also, for an increase in viewpoint diversity?

December 9, 2018 | Unregistered CommenterJoshua

Joshua,

They didn't study viewpoint diversity. Not surprising, considering how they mined their (massive) data - would have been very hard without auxiliary survey data to detect viewpoints.

December 9, 2018 | Unregistered CommenterJonathan

I saw that. Was wondering if you think we could extrapolate?

December 9, 2018 | Unregistered CommenterJoshua

Joshua,

Wouldn't think so, because coauthors prob admired each other's previous work. Rivals probably stay that way.

December 9, 2018 | Unregistered CommenterJonathan

"our findings may positively contribute to the societal conversation and reinforce the conviction that good things happen when people of different backgrounds, cultures, and ethnicities come together to work towards shared goals and the common good."

The 'reinforced conviction' being a political one, yes?

If they'd had some viewpoint diversity, someone might have pointed out the obvious alternative interpretation of their result. Researchers have to attract collaborators based on the quality of their work, and there is some "assortative matching" going on where the poorest can only attract collaborators locally, while the high-flyers attract attention and partners from a wider area. Top researchers in a field seek out international collaborations with the top researchers from other countries. Finding partners and setting up collaborations across borders is more difficult and expensive, so people only do it when it's really worth it. You can fund a visiting professor from abroad if they're globally famous as a brilliant researcher. It's not so easy to do so for a nonentity. I'd have thought that would be obvious to anyone who had spent time in a university research department, but I don't know. Maybe that's just my political/intellectual perspective?

I find it ironic (and a bit sad) that academics are still doing research looking for correlations between intelligence and race.

"Not surprising, considering how they mined their (massive) data - would have been very hard without auxiliary survey data to detect viewpoints."

People have certainly done research on the subject of political diversity in academia. You might not get the sample size with other sources (although I'd be suspicious of results that need massive samples to be detected), but I think it's a more interesting and under-researched question. Do you get better or worse results from liberal-conservative collaborations than you do from pure liberal-liberal or conservative-conservative? Would you expect anyone to have looked?

December 10, 2018 | Unregistered CommenterNiV

Why would liberal--conservative be a particularly interesting or informative brand of viewpoint diversity? I can think of many interesting brands, particularly with respect to investigating specific research questions.

December 10, 2018 | Unregistered CommenterJoshua

NiV
"..I find it ironic (and a bit sad) that academics are still doing research looking for correlations between intelligence and race..."

Where are those academics located? Mere mention of the topic is enough to unleash the "blank slate" equalitarian mob. Recent example:
https://quillette.com/2018/12/07/academics-mobbing-of-a-young-scholar-must-be-denounced/

These two academics tried to set up a conference on a "downstream" topic, and faced violent opposition:
https://quillette.com/2018/12/08/what-happened-when-we-tried-to-debate-immigration/

These academic fights leave all combatants weakened, to the benefit of the (excluded) Right, sole remaining supporter of free speech.

December 10, 2018 | Unregistered CommenterEcoute Sauvage

Am just guessing, but I think AlShebli et al. would have included political diversity and viewpoint diversity of other kinds if they were able to mine it from their dataset as easily. And if not them, then Heterodox Academy probably has this on their to-do list.

December 10, 2018 | Unregistered CommenterJonathan

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>