Lecture B09 of Stat Rethinking 2026: Generalized Linear Madness. Two examples of analyses that build up from scientific models first. #1 allometry predicts weight from height with ZERO parameters. #2 Vibrant ODEs for time series instead of cold dead lag models. Fully coded and animated with love.

Like many methodologists, I often find myself arguing with someone that their method/epistemology makes no sense. That someone thinks I am arguing against their *conclusions*. But I am arguing they have no justification for their conclusions. Is there a philosophical term for this problem/confusion?

"Evolution Evolving is a landmark book written by stellar scientists who have brought together numerous interesting and relevant examples that show why development matters for evolution." Thankyou @kampourakisk.bsky.social for your wonderful review in American Biology Teacher. See shorturl.at/2Gso8

Blog post on a new paper (w/@dingdingpeng.the100.ci & @statmodeling.bsky.social) reflecting on the value of multiverse analysis 10 years in. Like many attempts at science reform, it provides clear conceptual value but is tricky to theorize & can even be counterproductive if taken too seriously

Max entropy is the statistical version of "Measure twice, cut once." Multiverse analysis in practice is "Cut a bunch of times, and pick the ones you like the best."

Every time I am forced to use Matlab I can only conclude it is a language designed by and for sociopaths Shout out to this @olivia.science blog post that I sent out on a neuro listserv back in grad school, that a prof there is apparently still mad about neuroplausible.com/matlab

“la statistique bayésienne” — everything sounds classier en français

A biologically parameterized model of growth and honest measurement error in the mix. This paper is a quantum leap in handling diverse and messy cross-cultural and archaeological data on human growth, where papers usually fit arbitrary polynomials and ignore measurement error. Link in quote

Has anyone tried teaching *only* Bayesian inferential #stats for undergrad psych (or other UG sosci programs)? Like completely foregoing NHST, confidence whatevers, p values, CLT... I'd be very interested in seeing a syllabus of such a course/ program!

So I gave this workshop today and I think it went pretty well. The best comment afterwards was "thanks for presenting statistics as someone who is not dead inside". I will aim to write this up as a blog post or preprint when I get some time (after teaching finishes later this month).

*drums in the deep* work flow . work flow . work flow *growing closer* Work Flow . Work Flow . Work Flow *menacing now* WORK FLOW WORK FLOW WORK FLOW

my course notes on a bayesian workflow for (single agent) cognitive modeling are now fully revised and online: fusaroli.github.io/AdvancedCogn... Predictive checks, updating checks, sensitivity analyses and simulation based calibration in @mc-stan.org Feedback is very welcome!

*deep inside the pentagon* Analyst: Sir, all is go but we still need to disable the statisticians. They might be trouble in some tail scenarios. Chief: Agreed. What do you suggest? Analyst: I know just the thing... *goes online and starts posting about interpretations of confidence intervals*

New postdoc opportunity in our lab!

A Rock Pigeon with a threatening aura

I know armageddon is starting but I am giving a workshop tomorrow

Lecture A09 - I get mildly ranty about the low quality of papers on discrimination and somehow also introduce generalized linear models for events and illustrate post-stratification. The theme continues next week with modeling sensitivity to unmeasured confounding. I will try to be less ranty.

Yes please let's normalize errors, their discovery, their correction, and their integration into the literature. "Corrections Not Found: Post-Publication Integrity in Usable Security and Privacy Research" (osf.io/preprints/ps...)

🎉 It's livestream day 🎉 Join me at 1700 (CET; 1600 UCT) today for two hours of GAM goodness 🤤 📽️ Youtube: youtube.com/live/A9U8e1K... Hit the Notify me 🔔 to get a reminder when I go live #RStats #mgcv #statistics #GAMs #DataScience 🧪

good morning everyone project your personal imposter syndrome onto this gif ur welcome

Red Oval by Vasily Kandinsky, 1920 https://botfrens.com/collections/212/contents/137250

What a pre-conference workshop on Bayesian statistical methods—a powerful toolkit for our community! 📊 Many thanks to Daniel Redhead and Ramona Roller for guiding us through STRAND and to Michael Chimento for the introduction to STbayes. And thank you to all participants for their energy! 🙌

I told several non-German colleagues today that the German word for placenta is "mother cake" (Mutterkucken) and they all hate me now

Sometimes I assign applied statistics homework problems that are pretty difficult. Students spend hours over days on them. But you know what is much much harder? Research

A neat blog post walking through implementing Bayesian Decision Theory in @pymc.io with code! #statssky 📈📉 daniel-saunders-phil.github.io/imagination_...

Cooking with Mischka: Pizza with bolognese and olives. Warm here today so the yeast were partying. Made a big batch of bolognese in morning, so used thin spread as sauce. Mischka received turkey again

No matter the size or number, cats are always cats

Odd that the English name for this piece is "Birches" when the original title is Shawm (or Schalmei), the instrument that the figure on the left is player. (See www.guggenheim.org/artwork/1836)

Whenever UK politics floods my feed, it reads to me like: "Angry Fay Defeat Poshies and Gray Cloaks in Smashing Lower Stabbington Faff-Election" I vastly prefer this to USA news naturally

Bayesians assemble

I see blue horse, I repost

Figure 1 in this paper is both informative and hillarious. The collider on self-perceived causal inference skill is too real

"the paper’s main outcome variable cannot possibly be constructed as described in the paper." Retraction of 2023 paper that did not use the reported variables. The replication report is astonishing www.nature.com/articles/s41... Pre-publication peer review remains undefeated in laundering bullshit

Cooking with Mischka: Chocolate chip cookies. sugar, brown sugar, butter, egg, salt, baking soda, vanilla, chocolate. Mischka waited patiently and received some turkey

Per protocol analysis strikes again! Folks, if you randomize but then don‘t analyze some of the people who got randomized (maybe because they didn’t adhere to instructions, maybe because they dropped out), randomization will no longer do all the heavy causal inference lifting.

This recent RCT of an "AI stethoscope" claims the technology "shows promise" for diagnosing cardiovascular conditions. It does not. It is a textbook example of the risks of conducting unprincipled 'per protocol analyses'. Once again, peer review at a major medical journal has failed. 🧵 1/

Since I was astonished to discover the Danish grading scale yesterday, today I remembered they say 90 as "half fives". Wait that sounds insane, I must be misremembering this

Couldn't agree more - "We need more case studies for ecologists to emulate and for norms to shift so "regression and storytelling" becomes unpublishable." I have a running (yet incomplete) list of applications on my course website to help enable this: github.com/LauraDee/Eco... 1/2

If we are making lists of non-ignorable "complexity" in biology, how about the interaction of selection and development? Many biologists find it boring (unnecessary complexity) or threatening (old naive adaptationism called into question). @mauriciogforero.bsky.social does good work in this area!

Interested in using my textbook, Modeling Social Behavior, but wishing code was in a more general programming language than NetLogo? The incomparable Ben Fried has translated all modeling code into JULIA, utilizing the excellent Agents.jl package. github.com/BenjaFried/m...

I saw this when it was a preprint? They have a unique version of the estimand-estimator-estimates workflow, will be useful for may biologists, not just ecologists. We need more case studies for ecologists to emulate and for norms to shift so "regression and storytelling" becomes unpublishable.

This seems worthwhile. I'd join but I'm sitting here in central EU fighting the Wissenschaftszeitvertragsgesetz

Am I understanding this right? The standard grading scale in Denmark goes –3, 00, 02, 4, 7, 10, 12 ! This is amazing. I kinda love it. I mean I hate it. But I love it.

Is this a Lisp joke? Don't see those often!

Sausage tossing, Tony Hawk, and wine tasting - Lecture A08 of Statistical Rethinking 2026. Introduction to MCMC with lots of animations, then MCMC diagnostics in example of item response (IRT) model of wine judges. www.youtube.com/watch?v=N_LR...

Great to see Germany's largest research society support research assessment reform: less garbage bibliometrics like impact factor and h-index, less emphasis on over-production of papers no one wants to read, more expert assessment of quality and rewards for teamwork.

I could plagiarize this paper today and nobody would be able to tell the content is from 1970 (open pdf). academic.oup.com/jrsssa/artic...

Wenn Wissenschaft unsere Demokratie verteidigen will, braucht’s vor allem: Solidarität. Die wird aber durch Fehlanreize im Wissenschaftssystem strukturell verhindert, allen voran durch Wettbewerbisierung unter Institutionen & Individuen, die „alle gegen alle“ bedeutet. Was lässt daran ändern?

This squid is me.

My post-doc Charlotte King will be starting her own Minerva research group in September and is looking for PhD students. Click here if you are into experimental biomechanics, apes and hominin evolution. www.eva.mpg.de/career/posit...