Lightnews — Scholar-powered news

Richard McElreath 🐈‍⬛

@rmcelreath.bsky.social

17K followers 1.3K following 990 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 Media Videos Starter Packs

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...

28 220 820

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 10h

Most of us get 80% of the value from the first 20% of the material, so having that 20% in one place is a big contribution

1 3

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 12h

Looks like a good guide - the general data cleaning part is a lean intro to some very common issues in all sorts of data. Would be great if every phd who touches raw data was offered a short course in these basics (in R or Python or whatever HipsterScript) cleaning-data-r.ala.org.au/2_general-cl...

2 8 32

Reposted by Richard McElreath

Max Planck Institute for Demographic Research (MPIDR) @mpidr.bsky.social · 13h

💻Join online! 🎧
Süßmilch Lecture at the Max Planck Institute for Demographic Research (MPIDR)
Putting Science Before #Statistics
🗣️Richard McElreath 🐈‍⬛ MPI-EVA Leipzig
📅Oct 21, 2025
🕒3 pm CEST
✏️Sign up now:
https://ow.ly/uxfc50Xb5Yx

6 10

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 14h

For the right niche audience, you know who you are (www.neue-bachgesellschaft.de/mosaik-sonde...)

1 10

Reposted by Olivier Morin

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

"Dunbar's Number" is a zombie that lives forever in the science press it seems. Estimates of Dunbar's Number with 95% intervals, for a range of model specifications (from doi.org/10.1098/rsbl... ):

5 23 62

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

Well, I could go full Bayes and maybe reject the SEM as well. Presumably the proxies come from a distribution, and we could use partial pooling to do better in realistic scenarios with some more scientific information. But these are hard problems always in finite samples, whatever estimator.

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

I guess the conclusion you reach is the same as what I'd tell an editor when reviewing: The only admissible strategy is the SEM. The bias in the other models is potentially huge.

For extra fun, we could make it nonlinear and try to find an identification strategy using a DAG!

1 6

Reposted by Richard McElreath

Grant Sanderson @3blue1brown.com · 3d

Ever since I made a video about Fourier Transforms, one of the most requested topics on the channel has been its close cousin, the Laplace Transform.

I've been having a lot of fun animating a mini-series about this topic, and the main part is now out.

youtu.be/j0wJBEZdwLs

But what is a Laplace Transform?

YouTube video by 3Blue1Brown

youtu.be

9 67 410

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

The real Butlerian Jihad is not againt AGI, but against AI slop? Happy weekend to all www.youtube.com/watch?v=_zfN...

AI Slop Is Destroying The Internet

YouTube video by Kurzgesagt – In a Nutshell

www.youtube.com

4 5 32

Reposted by Richard McElreath

Ursula K. Le Guin Bot @leguinbot.bsky.social · 6d

Infinite are the arguments of mages.

3 24

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 7d

I guess because software allows it, ppl keep trying to estimate diversification rates on phylogenies. This is not in principle possible, because infinite combinations of diversification and extinction rates can explain almost any tree. Short, clear recent-ish paper: www.nature.com/articles/s41...

2 10 38

Reposted by Richard McElreath, Rebecca Sear, Corey Sparks

Antonino Polizzi @polizzan.bsky.social · 8d

✍️ Just out: Annette Baudisch & I delve into the formal demography of fertility, birth, & reproduction timing. 👶 We use methods from mortality research to summarize when in the parental life course children are born. 📆 1/n
doi.org/10.1080/0032...
@sdu.dk @oxforddemsci.bsky.social @mpidr.bsky.social

Fertility, birth, reproduction: Connecting formal demographic frameworks

The conventional framework of fertility research conceptualizes childbirth from the mother’s perspective. From her perspective, birth is an uncertain and potentially recurring event. In contrast, t...

doi.org

1 12 44

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 8d

This is a nice paper. It faces down a common problem: in order to explain to why a common approach cannot work even in theory, we first need to teach a framework in which regression is not magic that tells us which variables on right cause the variable on left. It's exhausting.

Anyway great paper!

Peter Tennant @pwgtennant.bsky.social · 9d

academic.oup.com/ije/article/...

Analyses of ‘change scores’ do not estimate causal effects in observational data

AbstractBackground. In longitudinal data, it is common to create ‘change scores’ by subtracting measurements taken at baseline from those taken at follow-u

academic.oup.com

3 16 78

Reposted by Richard McElreath, Jill A. Jacobson

Peter Tennant @pwgtennant.bsky.social · 9d

academic.oup.com/ije/article/...

Analyses of ‘change scores’ do not estimate causal effects in observational data

AbstractBackground. In longitudinal data, it is common to create ‘change scores’ by subtracting measurements taken at baseline from those taken at follow-u

academic.oup.com

1 3 26

Reposted by Richard McElreath, Gordon Hodson, Martin Tomko , and 2 more Richard McElreath, Gordon Hodson, Martin Tomko, Barrett, Patrick Präg

Julia M. Rohrer @dingdingpeng.the100.ci · 9d

Happy to announce that I'll give a talk on how we can make rigorous causal inference more mainstream 📈

You can sign up for the Zoom link here: tinyurl.com/CIIG-JuliaRo...

3 56 140

Reposted by Richard McElreath, Paul Goldsmith-Pinkham, Barrett

Aki Vehtari @avehtari.bsky.social · 9d

I'm looking for a doctoral student with Bayesian background to work on Bayesian workflow and cross-validation (see my publication list users.aalto.fi/~ave/publica... for my recent work) at Aalto University.

Apply through the ELLIS PhD program (dl October 31) ellis.eu/news/ellis-p...

ELLIS PhD Program: Call for Applications 2025

The ELLIS mission is to create a diverse European network that promotes research excellence and advances breakthroughs in AI, as well as a pan-European PhD program to educate the next generation of AI...

ellis.eu

34 45

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

Someone sent this to me and I want to hug, um, share it with all of you (src @theunderfold.bsky.social )

1 9 110

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 12d

Thinking about and discussing this more with colleagues, I'd really like a continuous time solution, like the Gillespie algorithm but for perfect conterfactuals as defined in thread below. I can't find that this has been done, but somehow believe some rogue chemist has already worked it out

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 13d

Are we doing simulations wrong? This paper convinced me we are. doi.org/10.1098/rstb... Usually we run 2 sets of "worlds" w and w-out intervention. Gives large uncertainties that include negative (harm) effects of interventions that are actually always positive (beneficial)!

Figure 5. Time series showing cumulative number of cases averted at each time caused by the intervention calculated using our method (single-world) and a
standard method. Shaded regions denote 90% confidence intervals. Note that there is more variation in the middle of the epidemic, so it may seem as though the
number of cases averted is large during those times. (Online version in colour.)