Magnus Johansson

Interpreting Confidence Intervals Interactive visualization of confidence intervals

#stats Here's a nifty interactive demo of confidence intervals by
Kristoffer Magnusson

rpsychologist.com/d3/ci/

GitHub - pgmj/RstudioQuartoIntro: A brief intro course on reproducible data analysis with Quarto and R A brief intro course on reproducible data analysis with Quarto and R - pgmj/RstudioQuartoIntro

I did a similar thing, but more generally as an intro to R (for psych researchers), where they download qmd-files that also contain some exercises that are based on modifying existing code: github.com/pgmj/Rstudio...

Looks like the Voigtländer Bessa L, which I think Cosina built.

Paper on statistical power necessary for interaction effects
doi.org/10.1177/2515...

Thanks, Wendy!

It's true! #openaccess doesn't have to mean that the author pays fees. We index journals that charge APCS (article processing charges) and/or any other fees. The 14107 journals we index don't charge any fees at all! So no hidden charges!

Take a look: doaj.org/u/idD0kN

Nice to see dynamic cutoffs used to evaluate model fit metrics!

The stupidity, the criminal vandalism, the wanton destruction of information involved in dichotomisation
statmodeling.stat.columbia.edu/2026/02/14/t...

A great read, including damning examples of how @thelancet.com refuses to clean up the scientific literature when they publish papers that are easily shown to be invalid.

This chapter on multicollinearity is a zinger, especially the central point in Section 23.4. If your focus is on model *predictions*, multicollinearity isn't necessarily a problem, and it might even be good. Well posted, @chrisadamsecon.bsky.social!

Yeah the 40 could be my one and only lens. I used to have the 20, 40 and 90, a super compact and performing trio.

Those Voigtländer SL lenses for Nikon are all really nice.

Is this PF as described within ACT?

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/...

I switched from Spotify to Qobuz some time ago, mostly for ethical reasons (pays artists better, etc). But the improvement in sound quality is really obvious when using a decent listening rig. And the curated playlists and magazine articles are really nice, especially on an iPad or similar.

That‘s the spirit.

Unexpected to see the Cochrane Risk of Bias tool v2 (2019) more or less asking for tests of baseline differences in RCTs when this is generally considered bad practice, for instance by the CONSORT guidelines.

Ja bra. Men blev det 18 av 81? www.mitti.se/nyheter/natl...

The file drawer could be vast.

Yes, and other aspects. Generally wary of research on your own product. Independent research replicating findings would be crucial.

I see many takes on the new Anthropic LLM-assisted programming study but no discussion about potential bias due to their financial interests in their own product?

I have made an R package for test equating using the Leunbach method. Thanks to the gracious sharing of DIGRAM source code by Svend Kreiner and some assistance of Claude Opus 4.5, the package `leunbachR` is now available on GitHub: pgmj.github.io/leunbachR/in...
#rstats #psychometrics

All my #rstats ebooks are now in the quarto format, and hosted on Quarto Pub. Their old bookdown links will all be dead by the weekend.

1/3

I have a strong belief that scientific instruments like questionnaires should not be have the same copyright protection as other text. Yes you can assert authorship, no you can't stop people using it.

A black and white image of a large oak tree trunk with 'Wake Up' graffitied on it. In the background there are more oak trees receding into the distance.

My favourite from 2025 using @ilfordphoto.com film. This (I think) is going to be the cover of my next zine...

Ilford Ortho+ 80 developed with Ilfotec HC 1+31. Taken with a Bronica SQAi & 80mm.

#fridayfavourites #2025fave #filmphotography

📢 The most recent volume of the Annual Review of Psychology is now online! The most read article so far is "How Do Psychologists Determine Whether a Measurement Scale Is Good? A Quarter-Century of Scale Validation with Hu & Bentler (1999)" by @dmcneish.bsky.social.
arevie.ws/4jRYY6B

RegCheck RegCheck is an AI tool to compare preregistrations with papers instantly.

Comparing registrations to published papers is essential to research integrity - and almost no one does it routinely because it's slow, messy, and time-demanding.

RegCheck was built to help make this process easier.

Today, we launch RegCheck V2.

🧵

regcheck.app

I've never experienced anything like this in my life, where people keep telling me that "we have to use AI or else we'll be left behind" but they don't actually tell me exactly what problem they are hoping to solve with the "AI" or how we would know it was useful. But we must use it! Apparently.

Frodo: I wish so much unbelievably stupid shit had not happened in my time Gandalf: So do all who live to see such times. This is all exceptionally stupid, though

Here it is! #rstats

This is quite a post

www.planetearthandbeyond.co/p/reality-is...