Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Crémieux Profile picture
@cremieuxrecueil
Sep 5, 2024 • 13 tweets • 5 min read • Read on X
Scrolly
How bad are Richard Lynn's 2002 national IQ estimates?

They correlate at r = 0.93 with our current best estimates.

It turns out that they're really not bad, and they don't provide evidence of systematic bias on his part🧵 Image
In this data, Lynn overestimated national IQs relative to the current best estimates by an average of 0.97 points.

The biggest overestimation took place in Latin America, where IQs were overestimated by an average of 4.2 points. Sub-Saharan Africa was underestimated by 1.89 pts. Image
Bias?

If you look at the plot again, you'll see that I used Lynn's infamously geographically imputed estimates.

That's true! I wanted completeness. What do the non-imputed estimates look like? Similar, but Africa does worse. Lynn's imputation helped Sub-Saharan Africa! Image
If Lynn was biased, then his bias had minimal effect, and his much-disdained imputation resulted in underperforming Sub-Saharan Africa doing a bit better. Asia also got a boost from imputation.

The evidence that Lynn was systematically biased in favor of Europeans? Not here.
Fast forward to 2012 and Lynn had new estimates that are vastly more consistent with modern ones. In fact, they correlate at 0.96 with 2024's best estimates. Image
With geographic imputation, the 2012 data minimally underestimates Sub-Saharan Africa and once again, whatever bias there is, is larger with respect to Latin America, overestimating it.

But across all regions, there's just very little average misestimation. Image
Undo the imputation and, once again... we see that Lynn's preferred methods improved the standing of Sub-Saharan Africans.

There's really just nothing here. Aggregately, Lynn overestimated national IQs by 0.41 points without imputation and 0.51 with. Not much to worry about. Image
The plain fact is that whatever bias Lynn might have had didn't impact his results much. Rank orders and exact estimates are highly stable across sources and time.

Want to learn more? See:
See this too, and note that it depicts rank correlations:
See this as well, on how very diverse data produces the same results:
And finally, see for the source of our current best estimates.sebjenseb.net/p/most-accurat…
It also might need to be noted: these numbers can theoretically change over time, even if they don't tend to, so this potential evidence for meager bias on Lynn's part in sample selection and against in methods might be due to changes over time in population IQs or data quality.

It might be worth looking into that more, but the possibility of bias is incredibly meager and limited either way, so putting in that effort couldn't reveal much of anything regardless of the direction of any possible revealed bias in the estimates (not to imply bias in estimates means personal biases were responsible, to be clear).
Some people messaged me to say they had issues with interpreting the charts because of problems distinguishing shaded-over colors.

If that sounds like you, don't worry, because here are versions with different layering:
Image
Image

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Crémieux

Crémieux Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @cremieuxrecueil

Crémieux Profile picture
Sep 1
One of my favorite studies on the validity of psychological measures was a survey that included 15 commonly-used measures.

Virtually all of them were found to be invalid for making comparisons between groups. Image
The only scale passing muster was the Need for Cognition scale.

Measurement invariance was assessed for age and sex and the degree of measurement invariance violations was not computed. That degree could be problematic or fine, but we don't know.

Either way, lots of invalidity.
And commonly-used measures are likely going to be the ones that are better-vetted.

My experience with measures that people make up for their own studies is that they're usually much worse than measures that have at least had some level of validation.
Read 4 tweets
Crémieux Profile picture
Sep 1
After I posted this thread, I was given all the raw data.

So, here are the zero-sum moral circles, where the categories are explicitly non-overlapping and giving moral units to a higher category does not include a lower-level category.

First, conservatives: very family-centric! Image
In case you're unfamiliar, these are the levels.

Participants were instructed that the moral units to allocate were like currency they can spend on others and allocate to different moral circles, and that a higher-level circle does **not** mean allocating to a lower one. Image
If you pick "16" in this exercise, then you take a moral unit away from levels 1-15.

Here's how this worked out for moderates. Curious result: they're a little more family-focused than conservatives! Image
Read 15 tweets
Crémieux Profile picture
Sep 1
Here's a fun alternative to the moral heatmap way of getting at in-groups.

Just ask people what communities they identify with most.

Liberals identify with the globe first, the nation second, and their local community last. Conservatives go nation, local, then global. Image
This study also gave participants the option to either pocket or donate another $5 for completing the survey.

They were given an international, national, or local charity option.

70% of liberals and 56% of conservatives donated something. Both liked local charity the most. Image
This result is curious because, in the real world, conservatives tend to donate more. No more often, but in greater amounts, so they overall give more, even controlling for income.

The reason has to do with religion. That explains the entire small conservative charity bump. Image
Read 5 tweets
Crémieux Profile picture
Sep 1
The moral circles study gave two versions of its circle task:

- One with a limited number of moral units, and thus moral allocation was zero-sum

- One where participants had unlimited moral units to distribute however they liked

Replot thread🧵

Everyone, zero-sum first. Image
The authors provided values on concern for humans versus nonhumans, and the results had to be scaled to be proportional.

Each allocation proportion is treated as a radial coordinate. To give it spread, we assign an angular coordinate θ = π/4 to tilt it like the original.

Cons: Image
To get some space on this, we add 10 degrees of Gaussian spread to give the cloud thickness.

Thus, each participant becomes a point (x = radial coordinate, r * cosθ, y = r * sinθ).

Did you notice conservatives being very human-focused? Here are moderates: Image
Read 17 tweets
Crémieux Profile picture
Aug 31
I'm delighted by how upset this benign observation made some people, because the same thing happened with the survivorship airplane meme.

If you're unfamiliar, it's this:

The supposed origin of the image is Abraham Wald's observation that the British Royal Air Force (RAF) was reinforcing the wrong parts of planes that returned from raids on the Germans. The military was noticing where the bullet holes were in planes that returned. The fact that those planes made it back suggested that those areas of the plane were the sturdiest, and reinforcement should instead be done on the areas without bullet holes.

This is a wonderful way to illustrate the concept of survivorship bias. It's so useful that it's come to be the canonical example in many classrooms, and the image has been seen by billions. The image went viral online as a way of illustrating survivorship bias. For example, you'll regularly see the image posted in response to someone making a mistake that's due to a failure to understand survivorship bias.

When this image first started going viral, one of the common responses to it was to state that the image was not, in fact, one ever seen by Wald or used by the RAF, and that it was actually just an illustrative mockup based on another mockup by @cameronmoll from 2005. The issue with that statement is that, after a short while of being viral, almost no one knew the origin of the image, and almost no one claimed it was actually an image used by Wald of anyone in the RAF, so it doesn't matter. The image is still an excellent way to understand survivorship bias.

A good question then, is why anyone would care that this clearly illustrative image wasn't actually used by Wald or the RAF. I'm going to wager that, for most people who made that argument, they were just missing the point. But, for some, they might just be regurgitating what they saw other people saying in response to people who wrongly claimed that it was a diagram used by the RAF. People like to do that—they like to repeat what they believe to be smart arguments, even when the context makes their point irrelevant.

The moral heatmap is in this stage of mimesis, where there's still a large mass of holdouts who haven't accepted that the meme just is the meme regardless of the study the diagram comes from. You see these holdouts everywhere, but as memes spread, they become less common. They exist for scientific papers and even for basic words. Some examples follow:

"Alpha" and "Beta": Supposedly pieces of wolf status hierarchies, these words now just mean you're a "Chad" or a "Virgin", a winner or a loser. The research on wolves didn't work out and the concepts don't hold up there, but it doesn't matter one bit, because these words now have a meaning separate from their misconceived origin. If someone says 'X is alpha!' or 'Y is a beta!' you don't win the argument by saying 'Actually, those parts of wolf status hierarchies don't exist in the real world' you just look retarded, because the words now describe something real: losers and winners! (With some added nuance that comes from sentiment attached to alpha/beta.)

The Dunning-Kruger Effect: This is supposedly the psychological bias where people with low knowledge/ability/awareness/etc. are overconfident. In some permutations, of the phenomenon, experts are underconfident, but it's irrelevant. What people understand the phenomenon to be isn't real: it's a statistical illusion resulting from binning a continuous variable with a raised intercept and an imperfect correlation between confidence and knowledge.

But, if you bring this up to debunk someone saying "Dunning-Kruger" to suggest someone is an overconfident buffoon, you just look retarded, because the words now describe something real.

The Banality of Evil: This is the idea that anyone can be made to do great evil, particularly through the influence of just following directions from authority figures. This was supposed to explain the Holocaust. Banality was supposedly confirmed in a series of experiments that took place at Harvard in the 1960s. In the Milgram experiments, students were told to shock someone they couldn't see, even as the shock intensity kept escalating and the person behind the wall screamed out louder and louder. They supposedly took part in this because the test administrator—an authority figure—was urging them along.

But in reality, the experiments were misdescribed and participants resisted more than Milgram said. Subjects also didn't go along with the experiments nearly as often if they believed them to be real. They also just didn't comply, applying weaker shocks when the experimenter was urging stronger ones.

The Banality of Evil is not real, but it doesn't matter. If you say we know it's not real, you are being retarded, because the concept still has utility in the expanded set of cases it's applied to these days, and in being used as a touchpoint to explain 'going along with orders.'

"Left-Brain/Right-Brain": This is the idea that the left and right hemispheres of the brain divide logical thinking from creative thinking, and that certain personalities have a given dominant hemisphere. The idea is untrue, but if you call this out when someone says something like 'That's very left-brained of you!', then you are being retarded, because left/right brain has entered the popular lexicon and it now refers to personality regardless of if its origins describe some real neural locallization.

"Reptile Brain": Carl Sagan popularized this one. This one comes from a now-discredited model of the evolution of the human brain, from the brain reptiles have—basal ganglia—to one other mammals have, allowing emotion—with the addition of the limbic system—to the one we humans have, allowing higher thought—with the addition of the neocortex.

After Sagan's popularization, people started to use being reptile brained as an insult. You can allege someone's actions are due to their reptile brain, making them a primitive. Though this concept and a lot of its support is now discredited, it led to good theorizing and discoveries, and if you respond to someone saying you're reptile-brained for being dumb, then you are being retarded, because the term now has a separate meaning from the theory it originated from.

"Marshmallow experiments": This refers to a very influential experiment where kids were told they could have a marshmallow now or have two if they waited. The kids who waited were supposedly vastly more successful in life. This is an interesting way of conveying that people who exercise more self-control are likely to be more successful later on. The experiment didn't itself hold up, but people now use the term "Marshmallow experiment" to refer to things where having self-control matters. For example, 'life is a series of marshmallow experiments'. Replying to this by saying that the experiment didn't hold up is retarded, because it's now a shorthand for delayed gratification.

"Lemmings": These cut little animals supposedly jump off of a cliff and kill themselves. But in real life, they don't do that. That was just a myth made by Disney. Nowadays, the name of the animal is often used to refer to people engaging in self-injurious or suicidal behaviors. You can point out that lemmings don't actually kill themselves, but you'll just look retarded, because a 'lemming' now refers to something besides the animal.

"A Frog Doesn't Notice It's Being Boiled": Kind of says it all. They do notice, but it doesn't matter, because when people use this phrase, they're almost never referring to actual frogs being boiled, they're referring to situations where people are haplessly unaware of dangerous changes around them. If you correct people by saying that frogs do notice being boiled, then you looked retarded, because again, that is not what people are really referring to, it is a turn of phrase.

"Eskimo Have 100 Words for Snow": This funny phrase was meant to humorously illustrate cultural relativism, but people started taking it literally. Now it's mostly not used as a fact about Eskimo culture, but as a stand-in for 'cultures vary' and sometimes 'people think too much about what they're overexposed to', and if you point out that the Eskimo don't have all those distinct words for snow when someone uses it like that, you're being retarded.

"You have the memory of a goldfish!": People believe goldfish have short, three-second memories. This isn't true, but it's entered the popular lexicon. If you say someone has the memory of a goldfish, you don't look smart by replying that 'actually, goldfish remember many things in the long term', you just look retarded, because people generally are not referring to actual goldfish memory span, they're saying you have a short memory.

"She's a Type-A personality": Some people have claimed that there are two main personality types: A, and B. Type A personalities are ambitious, competitive, and thrive under pressure, while Type B personalities are relaxed, patient, and adaptable. These don't really exist, but if you correct someone saying that a given person is "Type A" to refer to their ambitious personality, then you are being retarded, because their statement isn't based off of the theory, it's a broad description of a person's perceived personality as being a certain way.

Tons of these concepts have proliferated and entered the public consciousness. New ones enter it all the time, and I think we should generally welcome them if the concept has a real referent worth being able to talk about more clearly, which is what the concepts provide us with. The person who just can't accept this, who has to point out that these things aren't real, just doesn't get that memes evolve. They're the same sort of person who also points out things like:

- People misuse the word "literally"
- People misuse the word "ironic"
- People misuse the word "decimate"
- People misuse the word "peruse"
- People misuse the word "spazz"
- People misuse the word "approximate"
- People misuse the word "nauseous"
- People misuse the word "factoid
- People misuse the word "bigot"
- People misuse the word "nonplussed"
- People misuse the term "begs the question"
- Economists are misusing the word "identification"
- People misuse the term "enormity"
- People REALLY misuse the term "moot point"
- Etc.

But language evolves, and the misuses of literally decimating spazzes help us to understand one another better. They can also help people to signal affiliations, make their use in jokes, etc.

I propose that the people who feel compelled for whatever reason to object to memes and words that've evolved beyond the use they're trying to bring them back to are suffering from dysmimesis. Mimesis refers to the representation of the real world in art and literature—what the moral heatmap now does—and dysmimesis refers to the act of objecting to mimetic drift, or the dispositional urge to protest or 'correct' the evolved use of a meme, word, symbol, or practice as it spreads—i.e., resistance to mimetic/semantic drift and a wish to restore an earlier, 'proper' form.

P.S. I've used the word "retarded" a lot throughout this. It formally refers to people having adaptive behavior deficits, but almost everyone just uses it as a synonym for stupid. If you don't get that, then, well, you're a dysmimetic retard.

More reading on Banality, Marshmallow Experiments: cremieux.xyz/p/the-vast-emp…Image
Image
Image
Image
Oh, and, yes, I am aware that the rationalist community is a frequent origin for terms like "marshmallow experiments" as they're now used. That's one of the things I like about rats!
'Do you feel smart when you object to the evolution of language and art?' Image
Image
Image
Image
Read 6 tweets
Crémieux Profile picture
Aug 30
There are massive intelligence differences across populations. Image
Also, it only takes twice as long for a variant at a constant selection pressure to reach fixation in a population of 10,000 as in a population of 100.

Where is he getting the idea that 5,000 years is short? With rising populations, that can easily mean accelerated evolution.Image
If you want to learn more about existing differences (and they are real!), see: cremieux.xyz/p/national-iqs…Image
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(