Profile avatar
pgmj.bsky.social
PhD & lic. psychologist. Research and applied work at RISE Research Institutes of Sweden & Karolinska Institutet. R package for Rasch psychometrics: pgmj.github.io/easyRasch/ #openscience, #prevention, #psychometrics, #rstats, #photo
212 posts 2,224 followers 1,122 following
Regular Contributor
Active Commenter
Posts 24 Comments 26

#statstab #350 Communicating causal effect heterogeneity By @matti.vuorre.com Thoughts: Cool guide on properly communicating uncertainty in effects. #bayesian #uncertainty #ggplot #r #brms #tidybayes #heterogeneity vuorre.com/heterogeneit...

submitted 4 days ago • 1 comment

Sometimes when I talk to intervention scientists about #measurement #psychometrics I get the argument “even if this questionnaire has bad psychometrics, everyone else in my field uses it and I want to be able to compare my results with theirs”. Let’s talk about how that is a misunderstanding 🧵 1/6

submitted 5 days ago • 2 comments

🚨 New study reveals that when used to summarize scientific research, generative AI is nearly five times less accurate than humans. Many haven't realized, but Gen AI's accuracy problem is worse than initially thought.

submitted 8 days ago • 10 comments

New preprint: ”Uncovering Measurement Issues in Collective Efficacy: A Rasch-Based Evaluation”: osf.io/preprints/os... We find inadequate measurement properties for both subscales in this widely used survey questionnaire on collective efficacy. #psychometrics #rasch

submitted 8 days ago • 0 comments

🚨New R package! {easymediation}🚨 The *Simplest* and *Most Correct* Way to Do Causal Mediation Analysis Are you tired of explaining mediation analysis to your colleagues? Just send them this package. github.com/rpsychologis...

submitted 22 days ago • 6 comments

Undrar om han kan komma och tala lite förstånd med Uffe o co.

submitted 22 days ago • 0 comments

I thought exactly the same many, many times. Reviewers entirely ignore the methodological part of the paper and take its results at face value, but they will debate the "framing" of it in every little detail.

submitted 25 days ago • 8 comments

Took the opportunity to flex my #marginaleffects muscle and write a little guide for computing ranks and rank-based contrasts from ordinal regression models. Check it out! #rstats

submitted 26 days ago • 3 comments

Our new paper is out in Statistics in Medicine! We developed a joint model for multiple (un)bounded longitudinal markers, recurrent events, and competing risks, applied to US CF patient registry data. Joint work with @drizopoulos.bsky.social and @erandrinopoulou.bsky.social. doi.org/10.1002/sim....

submitted 30 days ago • 0 comments

Vignettes of Ipswich GFX50R vs Rolleicord (I'm only kidding! The Widelux version I haven't shown you yet wins hands down... 😜) #StaringAtWalls #filmphotography

submitted 26 days ago • 0 comments

Really looking forward to this. It's been six years since his last book.

submitted 28 days ago • 0 comments

once again expressing my gratitude for the amazing R packages we have these days, my life would be so much harder without glmmTMB and marginaleffects

submitted 29 days ago • 2 comments

Yet another paper showing that CBT is ineffective as universal prevention. Importantly, this is not an argument against universal prevention in general. To me, it seems that early interventions targeting self-regulation and cooperation/social skills are more promising and should be studied further.

submitted 32 days ago • 1 comment

priorsense 1.1.1, for prior and likelihood sensitivity analysis, is now in CRAN (n-kall.github.io/priorsense/) There is now an easier way to select which priors are power-scaled using the new prior tags feature in brms (2.22.11+), which allows focusing the analysis on specific priors. 1/3

submitted 33 days ago • 1 comment

We are live! Introducing the "Journal of Robustness Reports" – a Diamond Open-Access journal dedicated to publishing short reanalyses of empirical findings. Check out our website and blog post about the journal: 🌐 scipost.org/JRobustRep 📄 www.bayesianspectacles.org/introducing-...

submitted 40 days ago • 2 comments

1. LLM-generated code tries to run code from online software packages. Which is normal but 2. The packages don’t exist. Which would normally cause an error but 3. Nefarious people have made malware under the package names that LLMs make up most often. So 4. Now the LLM code points to malware.

submitted 45 days ago • 116 comments

I had reason to revisit this thread, and it contains a nice discussion with some great papers and resources. Friends don't let friends fit random-intercept-only models, when there's also room for random slopes. #rstats

submitted 47 days ago • 4 comments

📚😅🎉 Yay!! I just submitted the complete manuscript of my upcoming book to the publisher! Learn to easily and clearly interpret (almost) any stats model w/ R or Python. Simple ideas, consistent workflow, powerful tools, detailed case studies. Read it for free @ marginaleffects.com #RStats #PyData

submitted 47 days ago • 18 comments

For 23&Me users: This article has directions for deleting your data, and it applies to everyone. You don’t have to live in California to access California privacy. Gift link

submitted 64 days ago • 2 comments

PsyArXiv can deliver preprint posting and reading services to its users at no charge thanks to the generous support of hero institutions! Let's give them a round of applause for their support and encourage others to follow suit!👏👏👏 blog.psyarxiv.com/2025/01/03/m...

submitted 69 days ago • 1 comment

New blog post! In which I explain the issue with mediation analysis and sketch out one way to deal with the underlying causal inference problem -- in just a bit over 1,000 words! If you have never found the time to read up on this, now is your chance. www.the100.ci/2025/03/20/r...

submitted 68 days ago • 27 comments

Läser ett utdrag ur Johannes Klenells ”Bananrepubliken Sverige - Hur politiker, välfärdskapitalister och PR-byråer utarmar demokratin” i DN. En bok som borde stämma S-kongressen i Göteborg i maj till självkritik och en tydlig reglering av relationen till lobbyindustrin. www.dn.se/kultur/johan...

submitted 68 days ago • 1 comment

Don't trust "AI" and also realize that the push to have "AI" in everything happened because some tech people wanted to get rich and other tech people didn't want others to get rich without them, and now they've all spent too much money on a bad product and are trying to scramble out of the hole.

submitted 70 days ago • 253 comments

📊 #psy6135 #rstats Today's topic: Visualizing Uncertainty Topics: Visualizing distributions, Error bars, Bayesian stuff, fitted curves, ... Thx to @elibryan.bsky.social, @mjskay.com, @clauswilke.com, @cedricscherer.com and other folks I stole from. 💻 Slides-- friendly.github.io/6135/lecture...

submitted 74 days ago • 2 comments