One weakness Bluesky has compared to Threads is that we simply don’t have as many massive idiots making stupid posts generating screenshot-worthy replies
07.03.2026 04:26
👍 64
🔁 13
💬 3
📌 0
One weakness Bluesky has compared to Threads is that we simply don’t have as many massive idiots making stupid posts generating screenshot-worthy replies
Awesome YouTube tutorial by Johnny van Doorn, "Theory and Practice of Bayesian Inference Using JASP". Link to the video: www.youtube.com/watch?v=VhIm...
Blogpost:
www.bayesianspectacles.org/youtube-lect...
Interested in finding out more about @jaspstats.bsky.social? The Psych Methods Group at the University of Amsterdam is running four hands-on workshops (in person or online) this summer (@ejwagenmakers.bsky.social, @fbartos.bsky.social). More information at jasp-stats.org/workshops/ but to sum up:
26 participants, 17 new modules...the JASP hackathon was a success and we're looking into making it an annual event!
jasp-stats.org/2026/02/24/r...
On 16 and 17 February, the #JASP team organised a #hackathon where 27 participants from 6 countries learned how to contribute to their own modules to the JASP ecosystem!
Special thanks to @ejwagenmakers.bsky.social, @pabrod.bsky.social, Stefan Verhoeven & Zowi Mens for making the event a success!
vazul: An R Package for Analysis Blinding: https://osf.io/mp54s
New: "Effect of Artificial Intelligence on Learning: A Meta-Meta-Analysis" by Wagenmakers and colleagues revealing evidence for "severe publication bias and extreme between-study heterogeneity" in existing meta-analyses of the effects of AI on learning: osf.io/preprints/ps...
Shameful action on behalf of the editors. Or should we blame Springer? @springernature.com what is your position?
I hate this: "The Editors have retracted this article... the Mini-Mental State Examination (MMSE) instrument was used without proper permissions" link.springer.com/article/10.1... The MMSE is a series of 11 simple questions. A perfectly good paper is retracted because you can copyright 11 questions
"Discovering Statistics Using JASP", the first of four summer workshops, is now open for registration. Join Johnny van Doorn and myself in sunny (well, we can hope) Amsterdam or online!
jasp-stats.org/2026/02/03/h...
Bayesian Thinking for Toddlers -- now as a cartoon, courtesy of @viktorbeekman.bsky.social
You can read the cartoon at www.bayesianspectacles.org/bayesian-thi...
"So why do journals even bother with mandatory Data Availability Statements, if they do not seem to care what is in them?"
Indeed...difficult to get used to these practices.
I wrote a blog for the Meta-Research Center expressing my infinite frustration about not getting data. What else is new, you might think? Well, I added an extra layer of annoyance directed at the journals who do NOTHING to enforce promised data sharing.
metaresearch.nl/blog/2026/2/...
New preprint by @semihaktepe.bsky.social 🎉
We compare ANOVA/SDT/GLMM for binary judgments in 20 datasets of the truth effect. #lme4
Main conclusion:
"GLMMs are a theoretically sound and practically robust method and thus superior for analyzing binary judgments in social and cognitive psychology.”
#ICYMI: The seats for the Build Your Own #JASP Module #hackathon are quickly running out. It is not only free, we are even offering travel grants!
Want to join us in Amsterdam on 16 and 17 February?
Register here ➡️ jasp-stats.org/2026/01/27/a...
Does it mean that AI/LLMs do not help at education? I personally don't think so. I'm using the AI every day and find it incredibly useful. It would be odd if they didn't help at learning at all. However, the current empirical base does not substantiate strong claims.
We just posted a preprint with a comprehensive meta-meta-analysis of the effects of AI/LLMs on learning.
TLDR:
- 1,840 effect sizes
- extreme between-study heterogeneity
- extreme publication bias
- small average effects (three times lower than usually reported)
(osf.io/preprints/ps...)
A British Museum photo showing the reverse side of a silver stater coin from the ancient city of Eretria on the island of Euboea, Central Greece. The coin is roundish in shape, and in the centre there is a relief image of an octopus within a recessed square. The octopus has a bulbous head with a small round eye on either side. Beneath the head, eight arms curve gracefully, symmetrically arranged with four arms on each side, curling into a tight spiral at the tip. Diameter: 26 millimetres. Weight: 8.670 grammes. Dated circa 500-480 BC.
Ancient Greek silver stater (coin) with an octopus on the reverse side. 🐙
Minted in Eretria, Euboea, Greece, 500-480 BC.
Photo: British Museum www.britishmuseum.org/collection/o...
#Archaeology
#ReliefWednesday
Ten new JASP modules -- or perhaps more?
The JASP Hackathon will take place Feb 16-17 at the eScience center in Amsterdam.
This is our last call for applications. Teams are welcome; support to offset travel costs is available.
jasp-stats.org/2026/01/27/a...
For JRR it would have to be a reanalysis of a specific study (and there would need to at least one other team to do a different reanalysis)
Are you interested in adding Bayesian analyses to your next project so you can finally start interpreting your null results? Then join us tomorrow where Dr. Johnny van Doorn from @jaspstats.bsky.social will give a workshop how easy Bayesian stats can be using JASP! Sign up below!
US court allows ICE to arrest and pepper-spray peaceful protestors in Minnesota
Appeals Court Stays Restrictions on Federal Tactics in Minnesota The Eighth Circuit granted the Trump administration’s request to block, at least for now, a lower court’s injunction limiting how federal agents interact with protesters in the state.
One court decision, two different headlines.
Guardian on the left, NY Times on the right. Which one more directly conveys information and the stakes to the reader?
I love my profession deeply, but some of them are not covering themselves in glory these days. If a mob boss draws a sharp blade languidly across your palm and says it would be a real shame for you to lose your pinky, your headline isn't "Godfather rules out hacking off finger to settle debt."
That is exactly what you can do with the BF
If you want to speak of evidence, it is strictly required to report a BF ;-)
"Interest" in SESOI relates to practical use (the P in ROPE). The reason why SESOI for shock therapy is different than for a sugar pill is utility.
I do understand that some people misuse SESOIs as a stand-in for a semi-Bayesian prior, encoding it to reflect their design-dependent expectations. But if you want a Bayesian omelette...
And you would not change SESOI based on your design, because "interest" is defined in real-world utilities: e.g., how much improvement is worthwhile; this should not depend on the design of the clinical trial.
