Richard McElreath | Lightnews

Richard McElreath 🐈‍⬛

@rmcelreath.bsky.social

17K followers 1.3K following 1.1K posts

Anthropologist - Bayesian modeling - science reform - cat and cooking content too - Director @ MPI for evolutionary anthropology https://www.eva.mpg.de/ecology/staff/richard-mcelreath/

Videos

Richard McElreath is an American professor of anthropology and a director of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. He is an author of the Statistical Rethinking applied Bayesian statistics textbook, among the first to largely rely on the Stan statistical environment, and the accompanying rethinking R language package. .. more

Psychology 22%

Sociology 21%

Posts Replies Media Videos

Pinned

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · Sep 19

If you hate statistics like I do, then you'll love my free lectures. Putting science before statistics, 20 lectures from basics of inference & causal modeling to multilevel models & dynamic state space models. It's all free, made with love and sympathy. 🧪 #stats www.youtube.com/playlist?lis...

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 34m

My department is holding its annual xmas movie night and there is only one valid choice, the most xmas movie ever made

online form show six traditional Christmas movies to vote for, but only "Die Hard" is chosen

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 42m

I'd still like to make these lectures into a small book, but there are already two book projects between me and doing that. One day maybe.

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 44m

Teaching some actual anthropology this week and next

Title slides for 9 lectures on human ecology and culture
1. Human adaptation
2. Human life history
3. Human families
4. Food and people
5. Cooperation and magic
6. Technology
7. Politics and power
8. Cultural evolution
9. Inheritance and inequality

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 1h

False. There is no such thing as an introductory statistics textbook

Andrew Gelman et al. @statmodeling.bsky.social · 1d

What intro stats textbook to use?
statmodeling.stat.columbia.edu/2025/11/09/w...

What intro stats textbook to use? | Statistical Modeling, Causal Inference, and Social Science

statmodeling.stat.columbia.edu

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 20h

Link to reanalysis for those who haven't heard this story before (PDF): marketing.wharton.upenn.edu/wp-content/u...

marketing.wharton.upenn.edu

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 21h

If I had more time today, I would make a thread of published nonsense ad hoc pipelines. So just one: the 1985 hot hand fallacy paper by Gilovich et al justified its bogus estimator with nothing but intuition. It was 30 years before someone bothered to check it with synthetic data/analysis.

Reposted by Karl Andraczek

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 21h

So many nonsense ad hoc pipelines could be prevented by requiring that they work on synthetic data.

I tend to think of experiments as special cases of inference, since most of the problems I work on cannot be studied in experiments. But I get that many researchers see experiments as base analogy.

Adam Kucharski @adamjkucharski.bsky.social · 1d

"Validate With Simulated Truth: A first habit is to test whether an analytical pipeline can recover known conditions."

Very good advice below. So much COVID nonsense (e.g. 'immunological dark matter') basically came down to a non-identifiable model that hadn't been properly tested.

Modelling Like an Experimentalist

Dahlin et al. (2024) apply experimental thinking to a model of mosquito-borne disease transmissions.

onlinelibrary.wiley.com

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 1d

Fax is also acceptable

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 1d

The relevant bit of grump from the quoted post (left). And age distribution of Max Planck directors (right)

The biggest problem is the leadership of the institutes, the directors, people like me. Most of us got appointed by being successful in a system that lacks the reforms that are needed. And the job encourages a kind of competitive narcissism that views complaints about the system as weakness. Why would you trust us to implement reform?

The good news is that a lot of directors are going to retire in the next 10 years. More than 100 of them. If hiring is done right, a lot of change can be accomplished. But we’ll have to be less focused on hiring people based upon how many Nature papers they’ve published. Instead we’ll need to hear that they have a persuasive vision for how to improve the conduct of research in their field. No more Brotgelehrte with high h-indices, please.

age/gender distribution of Max Planck directors. Left is men. Right is women. Vertical axis is age. Horizontal is count.

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 1d

I'm participating in an "Open Science Exploratory Roundtable" in the Max Planck Society later this week, so bumping my grumpy thoughts about open science again. I'm as curious as everyone else to see what a club of competitive narcissists who succeeded under the status quo (the MPG) can manage!

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · Jul 9

How can we reform science? I have some ideas. But I am not sure you’ll like them, because they don’t promise much. elevanth.org/blog/2025/07...

Which Kind of Science Reform

What hope is there for science reform, if we can't agree on what to reform? Right now, principles are more important than practices.

elevanth.org

Reposted by Richard S.J. Tol, Richard McElreath, Wim Thiery , and 7 more Richard S.J. Tol, Richard McElreath, Wim Thiery, Francesco Saraceno, Will Lowe, C. Kirabo Jackson, Jonathan Birch, David Bartram, Ann Bartow, Olúfẹ́mi Táíwò

Jan Rosenow @janrosenow.bsky.social · 1d

Solar’s price drop is astonishing: panels are now 98% cheaper than when I first analyzed them in 2004.

Today, building a fence with solar can be cheaper than using wood.

Reposted by Richard McElreath

Adam Kucharski @adamjkucharski.bsky.social · 1d

Modelling Like an Experimentalist

Dahlin et al. (2024) apply experimental thinking to a model of mosquito-borne disease transmissions.

onlinelibrary.wiley.com

Reposted by Richard McElreath

Andrew Gelman et al. @statmodeling.bsky.social · 1d

What intro stats textbook to use?
statmodeling.stat.columbia.edu/2025/11/09/w...

What intro stats textbook to use? | Statistical Modeling, Causal Inference, and Social Science

statmodeling.stat.columbia.edu

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 2d

If I understand right, that sounds reasonable. Treatment assignment depends upon complexity, and complexity influences time to surgery conditional on assignment. An interesting example for modeling prob of treatment assignment!

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 3d

In big impact, low citation corner: Tu Youyou won the Nobel Prize in 2015 for work on anti-malaria compounds that have gone on to save millions of lives, but her 1999 paper on the topic still doesn't even have 100 citations.
en.wikipedia.org/wiki/Tu_Youyou
Paper: pubmed.ncbi.nlm.nih.gov/11721477/

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 3d

I try to practice forgiveness, so I will meditate on that. But as a general meta-science note, in one case the evidence against Pruitt was that he left in a spreadsheet a formula that copied data from one treatment and added a constant to it in another treatment. It's like:

still from The Wire in which Stringer Bell scolds someone for "taking notes on a criminal fucking conspiracy"

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 4d

This is great. I am not a fan of the oscillator example - makes it seem like a physics niche thing. But so useful for us population-thinkers. I might write up a post using a population dynamics differential equation example. If I can finish cloning myself. Yeah. www.youtube.com/watch?v=FE-h...

Why Laplace transforms are so useful

YouTube video by 3Blue1Brown

www.youtube.com

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

Link to report: doi.org/10.1098/rsos...

Reposted by Ingo Rohlfing

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

A spicier opinion is that academia taking back publishing is not necessarily a path to innovation and efficiency. Do you associate universities with efficiency? The problem with for-profit publishing is not the profit, it's the oligopoly power of major publishers. Anti-trust in our lifetime?

Reposted by Juan Rocha

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

I am slow to react to this recent Stockholm Declaration on scientific publishing. A lot of it sounds good, but I don't see how we get from here to there. I worry nothing substantial will happen until the cost disease kills the host.

3. Incentive systems to merit quality, not quantity

Reward scientific achievement based primarily on quality—not quantity measures of today’s ‘reputation economy’ metric system [8].

Discourage ‘salami-slicing’ of results to obtain ‘least publishable units’; reduce the abundance of review articles (meta-analyses) and commentaries that over-burden reviewers and can too easily be generated using AI.

Rankings of academic achievement with metrics such as citation counts, ‘journal impact factor’ (JIF) and ‘publication number’ must be used responsibly because they often do not correlate with research quality [9] (see also CoARA—Reforming Research Assessment). 7

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

Lots more Celtic words in old French I've been told, and of course very very many place names in France are Celtic. Even "Paris"!

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

Yes a name I think popularized in England by the Normans, who spoke French, but were actually Danish. Richard I the Lionheart was I guess an exemplar.

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

To update my list of Celtic words used in English, I am being told that "iron" is an early Celtic borrowing into early Germanic, as Germanic ppl (South Scandinavians at the time) moved into central Europe. So like 1000-500 BCE.

Another borrowing from same time is rich/Reich!

iron(n.)

Middle English iron, iren, yron, from Old English iren, variant (with rhotacism of -s-) of isen, later form of isern, isærn "the metal iron; an iron weapon or instrument," from Proto-Germanic *isarn (source also of Old Saxon isarn, Old Frisian isern, Old Norse isarn, Middle Dutch iser, Old High German isarn, German Eisen).

This perhaps is an early borrowing of Celtic *isarnon (compare Old Irish iarn, Welsh haiarn), which Watkins suggests is from PIE *is-(e)ro- "powerful, holy," from PIE *eis "strong" (source also of Sanskrit isirah "vigorous, strong," Greek ieros "strong"), on the notion of "holy metal" or "strong metal" (in contrast to softer bronze).

rich(adj.)

Old English rice "strong, powerful; great, mighty; of high rank" (senses now obsolete), in later Old English "wealthy;" from Proto-Germanic *rikijaz (source also of Old Norse rikr, Swedish rik, Danish rig, Old Frisian rike "wealthy, mighty," Dutch rijk, Old High German rihhi "ruler, powerful, rich," German reich "rich," Gothic reiks "ruler, powerful, rich"), borrowed from a Celtic source akin to Gaulish *rix, Old Irish ri (genitive rig) "king," from Proto-Celtic *rix, from PIE root *reg- "move in a straight line," with derivatives meaning "to direct in a straight line," thus "to lead, rule" (compare rex).

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

Philosophers of biology will continue to have fun charting and explaining what evolutionary biologists are up to with all this "organismal design" stuff! There are no signs yet we are approaching a stable consensus, to my mind.

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

The concept of "fitness" is central to evolutionary biology but it's not entirely worked out. There are multiple definitions, doubts about predictive power, problems with internal consistency. Here's a paper from last year attempting to solve some of these problems. doi.org/10.1093/evol...

Abstract

A long-standing problem in evolutionary theory is to clarify in what sense (if any) natural selection cumulatively improves the design of organisms. Various concepts, such as fitness and inclusive fitness, have been proposed to resolve this problem. In addition, there have been attempts to replace the original problem with more tractable questions, such as whether a given gene or trait is favored by selection. Here, we ask what theoretical properties the concept fitness should possess to encapsulate the improvement criterion required to talk meaningfully about adaptive evolution. We argue that natural selection tends to shape phenotypes based on the causal properties of individuals and that this tendency is, therefore, best captured by a fitness concept that focuses on these properties. We highlight a fitness concept that meets this role under broad conditions but requires adjustments in our conceptual understanding of adaptive evolution. These adjustments combine elements of Dawkinsian gene selectionism and Egbert Leigh’s “parliament of genes.”

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 5d

It's complicated. The "whole organism" perspective has some things going for it I think. Cool paper from 2019 on the topic doi.org/10.1098/rspb...

The strategic reference gene: an organismal theory of inclusive fitness | Proceedings of the Royal Society B: Biological Sciences

How to define and use the concept of inclusive fitness is a contentious topic in evolutionary theory. Inclusive fitness can be used to calculate selection on a focal gene, but it is also applied to wh...

doi.org

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 6d

exactly, I don't think there is an accessible explanation that is also right. similar to trying to explain a p-value without mathematical expressions. it's not surprising people end up believing false things afterwards.

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 6d

Could say the same about "sexual selection" I guess. The mathematical models don't much resemble the ev-psych narratives most of us tell in intro courses.

Light upyour news

Sign in to Lightnews

Sign up to start reading

Connect Bluesky

Connect with Bluesky

Light up
your news