Thursday assorted links


7 How can you read that dreck? I'm sure there might be something of interest there, after wading through all the emotional feeling garbage.

Am I supposed to feel sorry for someone whose paper couldn't be replicated? She did aTED talk for God's sake!

Unlike postmodernists, at least social psychologists take evidence, statistics, findings and arguments — and ultimately, refutation — seriously. You know, generally speaking.

Social psychologists deal mostly with narratives while pretending to be hard scientists. They aren't.

Read 7b, it's excellent.

"Dominus also writes, “Gelman considers himself someone who is doing others the favor of pointing out their errors, a service for which he would be grateful, he says.” This too is accurate, and let me also emphasize that this is a service for which I not only would be grateful. I actually am grateful when people point out my errors."

I think Gelman is an honest broker, and his combination of statistical prowess and understanding of how science works (quote above) make him incredibly valuable in the current environment.

That was a good read. Gellman was right on in a thoughtful manner.

7. Yep, I doubt Tyler read the whole thing. Here's an excerpt:

"That morning of the troubling text, Cuddy logged onto her computer and discovered that Carney had posted on her website a document (then quickly published on New York magazine’s site) that seemed intended to distance its author forever, in every way, from power posing. “I do not believe that ‘power pose’ effects are real,” she said. Not only had she stopped studying power poses, “I discourage others from studying power poses.”

She listed a number of methodological concerns she had, in retrospect, about the 2010 paper, most of which, Cuddy says, Carney had never raised with her. In an email a few months earlier, Carney had clearly told Cuddy that she thought the study’s data was flimsy, the sample was tiny, the effects were barely there. But Cuddy said she had never received notice that this kind of renunciation was coming."

And that's why we have a replication crisis!

".... wading through all the emotional feeling garbage."

yeah, lately the New York Times articles have markedly strayed into the verbose literary format rather than the direct journalistic mode --- fake news flows easier as creative literature

As someone to whom social psychology is foreign, I did appreciate the offhand definition of "demand effect," which is generally the first thing that pops into my head when I read about these absurd studies - but I hadn't known the term for it.

But of course, it's not a problem!

Gotta disagree with Derek. I'm in the field (personality and social psychology) and very familiar with the train wreck (the non-replicability crisis) as it was developing. I think the article was accurate. Tough break for Amy, but she gambled on very iffy data, tried to milk it for more than it was worth (which wasn't much) and got called on it. Look of the bright side, lots of people manage to live happy lives without tenure at Harvard.

The NYT piece goes thru contortions to avoid pointing out explicitly that Cuddy did and continues to milk her bad science. In a post-truth world, that's not so unusual - as long as you don't pretend to be a scientist. I personally think "social science" (as it currently is practiced) is NOT science. Cuddy is a great example; she failed to take responsibility (until recently) for the paper she authored. That's really bad science. She also seems to have been (if I read between the lines) quite resistant to listening to and learning from her paper's critics. Also not good science (but not unusual, either). I suspect her claim that she wasn't able to run a replication study is an outright lie. The sad thing here is that it took so long to kick her butt out of academia and move her into her true calling: motivational speaking. I am highly dubious of those "scientists" who claim they didn't know any better. Cherry picking isn't some new concept invented in 2010. The "social sciences" have aided and abetted that rot for decades; the schools, the journals, and the professional organizations. It's with no little schadenfreude that I see them reaping a little of what the profession so richly (imho) deserves. It's with some amazement that the repercussions have been so long in coming and so mild. Drain the swamp, comes to mind. Easier said than done. Gelman's response is also disingenuous. He goes out of his way - repeatedly - to say that doing bad science is not necessarily because one is a "bad person". It's a strawman argument. Doing bad science IS a clear indication that one is a bad scientist. Bad science isn't based on what the person knows or has been taught, it is based on what the person SHOULD HAVE known and SHOULD HAVE been taught. Which raises an interesting point: I wonder how much thesis work is replicable? Wouldn't that be an interesting measure of the quality of the field? (As compared to the fraction of published papers that are replicable.) I wonder if it's ever been done. I doubt it. I also doubt it ever will. Science is not a profession that is capable of rapid change. To paraphrase: bad science is most often buried with its adherents.
There is one thing even more vital to science than intelligent methods;
and that is, the sincere desire to find out the truth, whatever it may be.
- Charles Pierce
Science is not belief, but the will to find out. - author unknown
Most people say that is it is the intellect which makes a great scientist.
They are wrong: it is character. - Albert Einstein
based on the above, how should Cuddy be judged?
A scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die and a new generation grows up that is familiar with it. - Max Planck
Science is as human an activity as politics or law making. Just as law making has been likened to sausage making, so is science - full of well-meaning fools, blundering savants, careful idiots, and careless genius. Enthusiasm and loyalty, faith and distrust, and, especially, inflated egos are part and parcel. NYT suggest that scientific discourse is polite and respectful. OMG. That's like saying sex is neat and tidy.

If you know even a little of the history of science, you'll know that there have been many scientists who have made genuine contributions while also having horrible characters, pathological narcissism, patterns of trying to destroy the careers and lives of competing scientists, etc. Some have even committed the grave sin of profiting from their science or touting their own results. Meanwhile, there have been ineffective scientists who were modest, monkish, and self effacing. And if you've done science in an academic setting, you'll know that there are written and unwritten rules about how the game is played. The replication crisis has had the useful effect of pointing out how the 'science game' itself has led to serious corruption of the process, often by well meaning people. This is partly why I left science for medicine years ago.

It appears to me that Cuddy's great sin in all this is not that she published, and defended, some mediocre science. Were that it, most university science and economics departments would need to be gutted, post haste, and with prejudice. It's that she became popular, and made money. The jumped the academic status queue. That is apparently unforgivable to a lot of people.

"Some have even committed the grave sin of profiting from their science or touting their own results."

Michael Mann comes to mind on this point. But he managed to profit without making genuine contributions and managed to profit even though he was exposed as being both ineffective and a fraud.

For an advocate of rigorous statistical practice, your sampling is a little suspect. You might think twice before using one case in one literature to generalize across all of social science -- an immense category which includes multiple fields and methodologies and research traditions.

Not to mention that the medicine game would not, if we are generalizing here, come out smelling like roses in comparison to the science game. Far from it.

Amy should have picked a filed where being obviously wrong does not matter and is rewarded. Climate 'science' comes to mind. You get to cherry pick your data and can even 'adjust' it if it is convenient for your hypothesis. Not only do you get tenure but the government pays you millions for your efforts. You might even get to hang around with Leo and Al and talk about how you care about the children.

# 2: First association for me was "The Lady Vanishes" rather than "Close Encounters".

For me it was Berlioz's Symphonie Fantastique...


I've been to restaurants that provide discounts for doing this. Of course they mark up before that so everything comes out even. Not sure if it's a gimmick or if they're trying to encourage more/less time spent in the restaurant. Sitting at a table on your phone doesn't generate much new revenue and in fact prevents new revenue from walking in the door, whereas people having conversations tend to order more to eat/drink? I don't know.

That reminds me of a story I saw here maybe a year ago about a chef whose restaurant was getting worse and worse reviews. People complaining about slow service and cold food. He had been monitoring those numbers, and time to be seated was the same it always had been, and time from order to delivery was the same. But, so many people were sitting on their phones instead of reading the menu that time from seating to ordering had increased, purely because people were sending the waiter away for more time. Also, the time they spent taking pictures of their food and Instagramming it was making a significant impact in time between delivery and actually eating.

The joys of smartphone culture.

The moving fridge was done years ago.

#5: I blame Global Warming.

#7: American journalism is wordy. Are they paid by the word?

Still, it's always amusing to see a journalist pointing out that someone else's work is bogus.

That above all we must always strive for the utmost in empathy—rather than “mere” decency—is always the subtextual tone in NYT pieces like that one. It’s like there’s an emotional-ideological template.

Mostly whne "sisters" write about eachother.

I blame Henrietta Lacks. The lady book clubs flocked to that one like white on rice.

Sorry, but science is about finding out the truth, not empathy. People who publish results that rely on dodgy data and use inadequate methods cannot pretend to be real scientists. The social sciences fail miserably because the methodology is badly flawed and the hurdles are far too low for most 'empirical' work to have much meaning. The idea that we can look at one variable while ignoring the variability of thousands of other variables that have an impact while trying to draw conclusions about complex systems is about as fraudulent as we can get. Most academics ignore the methodological problems because their fields have exactly the same issues.

#1 Conversation? Or perhaps just some peace and quiet from all the noise of people yammering away on their phones seemingly everywhere all the time. Sounds like a nudge to get going toward a social norm of less noise-making. Or perhaps a filter driving away the yammering minority and so creating a quiet space for the untapped market of hordes wanting calmer digestion. A 'Get Smart' cone of silence would work even better (if there were such a thing.)

3. One of my favorite facts about pop music:
In 1967, the year of Sgt. Pepper's, Disraeli Gears, John Wesley Harding, etc., plus debuts from The Doors, Janis Joplin and Aretha Franklin, the Billboard album charts were dominated by...The Monkees. The Prefab Four had the No. 1 album 29 out of 52 weeks. They were No. 1 with three different albums for the first six months of the year, except for one week when Herb Alpert was on top.

The Monkees' songs were better.

Neil Diamond wrote the #1 Monkee's hit (I'm a Believer)

Interesting their other biggest hit (Last Train to Clarksville) was a disguised anti-VietnamWar song... about a guy getting drafted and leaving home. (end lyric states, "I don't know if I'm ever coming home")

Undeniably good pop songs.

#3 - I like Billy Joel ("6. Billy Joel – 82.5 million units"). Not necessarily because of his music, though his pop hits were not bad, but because he just had another kid, at age 67 or so. G.O.A.T.!

# 1 - Under the PAP totalitarian system, Singaporeans exchanged their souls for a mess of pottage. The PAP has drowned the most heavenly ecstasies of religious fervour, of chivalrous enthusiasm, of philistine sentimentalism, in the icy water of egotistical calculation. It has resolved personal worth into exchange value, and in place of the numberless indefeasible chartered freedoms, has set up that single, unconscionable freedom — Free Trade. In one word, for exploitation, veiled by religious and political illusions, it has substituted naked, shameless, direct, brutal exploitation. The PAP has torn away from the family its sentimental veil, and has reduced the family relation to a mere money relation.

I disagree. The family unit in Singapore is doing much better than in most other places. Look at the US and Europe and you see that the illegitimacy rate in Singapore is much lower. People are educated and work hard. While that presents a problem because women do not marry down, that has nothing to do with losing their souls or not having fun.

3. Quite a few surprises (for exclusion as well as inclusion). For inclusion, Garth Brooks at number 2: "I've got friends in low places Where the whiskey drowns and the beer chases My blues away And I'll be okay I'm not big on social graces Think I'll slip on down to the Oasis Oh, I've got friends In low places". That was a happier time, back when Garth Brooks' fans spent their leisure time in the Oasis Bar rather than at Trump political rallies and white nationalist marches.

Asshole much?

I'm sure in his own mind, it's everyone else that's rude and illogical.

It was happier for you. That is all that matters.

I drink my gin and beer at home. Big glass, pour in jigger of gin, pour in bottle of Yuengling. It works for me. The money saved is well-spent on ammunition, guns, and hunting gear. I got a "Trump2020" sticker on the ass-end of my pick-up truck, and gun racks in the cab. .

Still deplorable after all these years.

I feel your pain, man. It's going to have a rough seven-and-a-half years for hate-filled, liberals.

I think it's interesting that Garth Brooks was originally pilloried as a clean pop star making country music indistinguishable from mainstream pop. By the standards of modern country music, however, Garth Brooks' earlier work is gritty and difficult. He rendered himself obsolete.

White nationalist marches? You can do better.

6 is funny because different people report the same numbers as good or bad news. In this case, the marketplace is more difficult for consumers, the government spends more, yay? I guess if you assume players can navigate the Adventure game to appropriate Bronze plans.

North. North. East. Drop sword. Buy insurance.

Open window. West. West. Open trophy case.

Consumers have a hard time because there is no free market in insurance and in health care. Access was easier before government got into the healthcare business. Trump has not gone far enough. He needs to get rid of all of the regulations and let consumers find the best care that they can afford.

#7, what a confused article -- human interest, personalities, women's issues, science, statistics all jumbled together in an article that just won't quit. Don't think you really need an advanced degree in statistics to recognize that a lot or perhaps most social science research is nonsense.

#3. That's a great picture of Robert Plant flashing an impish grin at Obama.

"5. “Today, humanity lives in a relatively quiet volcanic period.” (NYT)"

Thanks Obama!

A good read on the replication crisis:

After suffering a major brain injury, Amy Cuddy danced professionally, graduated college, studied at Princeton, became a professor at Harvard, and published a bestselling book. No wonder she is a sought after speaker, and no wonder she come across much better than her critics.

If one wants to crusade against poor analytical methods, it is wise to double check relevant emails before putting up a highly critical blog post. As for Gelman, I’ve read a number of his posts with interest, but the way he talks about Cuddy’s brain injury is enormously off-putting.

I have nothing intelligent, or even stupid to say about social psychology’s replication crisis, but if this article provides a modestly accurate depiction of many of the people who work in the field, then I think the discipline needs to pay attention to rudimentary human virtues, before concerning itself with inadequate quantitative methods.

Overall, being someone who enjoys thinking, and sometimes blogging as a contrarian; and who relishes logical dissections of established arguments, I viewed the article as a call for rigorous self-assessment.

Sorry, my comment immediately below was supposed to be a reply to this one.

Having read a lot by Andrew Gelman, I have never seen anything even remotely insensitive or cruel about Cuddy's brain injury, or for that matter anything about her non-science-related activities. Moreover, from all of his writings he seems genuinely (and tirelessly) helpful and plain-spoken. What are you referring to? Could you please cite some post of his?

I was referring to his discussion of the brain injury that is described in the NYT piece.

The NYT quote is first from a commenter, not Gelman: "“I’ve wondered whether some of Amy Cuddy’s mistakes are due to the fact that she suffered severe head trauma as the result of a car accident some years ago.” Gelman replied, “A head injury hardly seems necessary to explain these mistakes,” pointing out that her adviser, Fiske, whom he has also criticized, had no such injury but made similar errors." I've spent the last several minutes trying to find this exchange on Gelman's blog, using both Google and the blog search bar, and have not succeeded. Anyway, assuming it's an accurate quote I think your reading of it is uncharitable -- in response to a commenter rather snarkily pointing at brain injury, Gelman draws attention, as he consistently does, to the issue of methodological flaws *not* at all being unique to Cuddy, but rather commonly found in the work of many "successful" researchers. I would recommend reading Gelman's blog -- it's far better than you might assess from few-sentence NYT sound bites. It's true that he obsesses a bit over people like Fiske and Brian Wasink, but overall it's really a treasure of insight.

#7 The whole point of replication is that it takes the politics and personalities and religious beliefs out of science. The article raises interesting issues but is a good example of why replication is necessary since it attempts to muddle all of those together in painting a sympathetic portrait of a scientist who hasn't really fully come to grips with these issues. There is also a weird subtext that runs through feminist thinking that academic postions are prizes that aren't awarded fairly to women by men, not meritocratic positions.

Comments for this post are closed