Crémieux Profile picture
Sep 5, 2024 13 tweets 5 min read Read on X
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

Jul 26
I don't think people realize just how wacky things have gotten.

First: The White TFR is ~1.55 and the Black TFR is ~1.53.

Second: Even the stereotypically extreme Hispanic TFR is now below-replacement, at ~1.98. Image
It's hard to overstate just how much things do not stay the same.

It's not even just in the U.S. where fertility rates are now shocking.

This graph shows Saudi Arabia. They're Muslims, so they must be having kids, right? No, they're barely above-replacement.Image
Even the Mormons aren't having kids!

In fact, we're approaching a decade of Utahan fertility being below-replacement! Image
Read 4 tweets
Jul 24
The White House just released a really good executive order on cleaning up America's streets, re-institutionalizing insane people, and ending open air drug abuse and the problems it creates.

Here's a quick overview🧵 Image
The first section is the one I'm most excited for. An alternative name for it could be "Bring Back The Asylums"

It instructs the administration to make it possible to involuntarily commit crazy people again

That crazy hobo pushing a cart full of urine bottles? He's going away! Image
The next section is one that you'll need to familiarize yourself with if you're interested in 'what happens next'.

This was a never achieved goal in Trump-I.

The idea is to compel cities to do what you want by withholding, barring, and giving discretionary funds for compliance. Image
Read 15 tweets
Jul 24
What comes after GLP-1RAs make everyone skinny?

What comes after myostatin inhibitors make everyone buff?

One new candidate is:

Safe, cheap, and easily-administered injections that locally remove fat. A new drug that just passed through phase 2 seems to do just that🧵 Image
The new drug is called CBL-514.

It has a counterpart on the market in the form of deoxycholic acid injections—brand name Kybella.

Kybella is FDA-approved, and it works: it helps people to get rid of their double chins. But there's a catch. Image
Kybella, unfortunately, is not all that safe, and though many patients swear by it, there are notable side effects.

This is predictable, since the way Kybella works is through cytolysis: causing cells to die by rupturing them, releasing their contents, causing inflammation. Image
Read 17 tweets
Jul 23
Pancreatitis is a commonly mention potential side effect of GLP-1RAs, but the evidence for it is poor.

In trials, it doesn't tend to show up more often than in placebo groups, and in high-quality comparisons, risk typically isn't notably elevated.

Risk might even be reduced! Image
There's generally bupkes for pancreatic cancer risk and the evidence for increased pancreatitis risk is similarly poor.

Here's a summary of evidence from before the study whose result I plotted up there: Image
Remember this trial?

It used very high doses of semaglutide and still failed to find any elevation of pancreatitis risk:
Read 8 tweets
Jul 22
Many Founding Fathers wrote under pseudonyms

Pseudonyms afforded the protection needed to write things that were controversial, to engender debate over things they didn't themselves believe in, and to encourage focus on ideas over reputations

Thread of their known pseudonyms🧵 Image
Alexander Hamilton, James Madison, and John Jay all wrote under the name Publius, after the Roman consul Publius Valerius Poplicola.

This shared authorship became known after Hamilton died, but the individual authors of the Federalist Papers Publius entries remain debated. Image
Image
Image
Image
John Jay and John Stevens, Jr. shared the Americanus pseudonym when writing various Federalist essays. Image
Image
Read 20 tweets
Jul 22
One of my favorite papers in recent years included this diagram.

It shows the impact of controlling for three different types of variables: confounders, colliders, and mediators.

With confounders, control is good. With the others, you ruin your result by controlling. Image
If you have variables with measurement error, you can run into another problematic variable: the proxy.

Proxy variables can make all of these distortions much worse and much more difficult to deal with. Image
The paper makes this simple observation: statistical control requires causal justification. That's actually the title.

They gave several DAG-based examples. Consider this one: is edutainment a confounder or a mediator? Should you control for it, or would that bias your estimate? Image
Read 9 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!

:(