iiinova — where your digital self keeps evolving

Wolfgang Viechtbauer @wviechtb.bsky.social · 11/10/2026

Wow, strongly reminds me of the cathedral in Aachen, Germany.

Photo of the octagon of the cathedral in Aachen, Germany.

Wolfgang Viechtbauer @wviechtb.bsky.social · 06/10/2026

Just participated in a fantastic workshop by Kenneth Koslowski and Fabian Münch on the mlts package (cran.r-project.org/package=mlts) for multilevel latent time series models. This opens up the world of DSEM to R users! Check out the package and corresponding paper: doi.org/10.1037/met0... #RStats

cran.r-project.org

mlts: Multilevel Latent Time Series Models with 'R' and 'Stan'

Fit multilevel manifest or latent time-series models, including popular Dynamic Structural Equation Models (DSEM). The models can be set up and modified with user-friendly functions and are fit to the...

090

Reposted by Wolfgang Viechtbauer

code spaghettifier @singlewordname.bsky.social · 19/09/2026

A computer science prof actually tested this digitally against a bunch of completely buckwild chess algorithms. Random movement comes out to about 478 ELO, though that's buoyed by a couple truly deranged opponents youtu.be/DpXy041BIlA

youtu.be

30 Weird Chess Algorithms: Elo World

YouTube video by suckerpinch

2395

Reposted by Wolfgang Viechtbauer

Kelly Bodwin @kellybodwin.com · 16/09/2026

👏👏👏 to @liomys.mx for calling out the issue of academic works citing the IDE instead of the language/packages!

A slide containing many snippets of text from academic papers saying "analysis was performed with RStudio"

4358

Reposted by Wolfgang Viechtbauer

Transactions on Machine Learning Research @tmlrorg.bsky.social · 16/09/2026

TMLR has faced a deluge of submissions, necessitating stricter desk rejection policies due to limited reviewer capacity Co-EiC Nihar Shah reached out to authors of 10 papers slated for desk reject. Could they answer questions about their *own* submission? medium.com/@TmlrOrg/ask...

medium.com

Asking Authors About Their Own Papers

By Nihar B. Shah

116965

Wolfgang Viechtbauer @wviechtb.bsky.social · 11/09/2026

So TWO things that are slow - the uptake of novel (or really not so novel) methods and the start up time ...

010

Wolfgang Viechtbauer @wviechtb.bsky.social · 11/09/2026

Fair point. I just think we should point out the benefits of open alternatives instead of bashing SPSS users. And one benefit is not having to wait a minute for things to start up 😉

001

Wolfgang Viechtbauer @wviechtb.bsky.social · 09/09/2026

It's got a certain retro charm to it ...

000

Wolfgang Viechtbauer @wviechtb.bsky.social · 09/09/2026

As an R user, I know it's tempting to dunk on SPSS, but I try to avoid such software snobbery. Occasionally, I even start up SPSS to make software comparisons. But that it takes almost 50 seconds to start up SPSS (on a freshly booted Windows machine) is just bonkers. #RStats #SPSS

4100

Wolfgang Viechtbauer @wviechtb.bsky.social · 26/08/2026

The High Fens before (2026-08-14) and after (2026-08-24) the wildfire. So much beautiful nature destroyed. #wildfire #highfens #hohesvenn

Satellite image of the High Fens on 2026-08-14.

Satellite image of the High Fens on 2026-08-24 showing the damage caused by the wildfire.

020

Wolfgang Viechtbauer @wviechtb.bsky.social · 24/08/2026

Standard issue cat recharging via solar power. #cat

A gray tabby cat sitting on a table in the sun.

2210

Reposted by Wolfgang Viechtbauer

The R Foundation @r-foundation.bsky.social · 19/08/2026

We invite #RStats users to contribute to a fund dedicated to helping the family of Tomáš Kalibera following his untimely death in April. Find out more and donate here: www.r-project.org/foundation/k...

r-project.org

R: Tomáš Kalibera Family Fund

04434

Wolfgang Viechtbauer @wviechtb.bsky.social · 17/08/2026

Image from the Sentinel-2 satellite from yesterday, showing the burning wildfire nearby in the High Fens (about 6km from where I live). #wildfire #highfens #hohesvenn

Image from the Sentinel-2 satellite from yesterday, showing the burning wildfire in the High Fens.

032

Reposted by Wolfgang Viechtbauer

Johan Larsson @jolars.co · 15/08/2026

I just released version 0.19 of Arity: a formatter, linter, and language server for R! The headline of this version is that Arity now formats and lints your DESCRIPTION. The release also includes improvements to roxygen2 formatting as well as many bug fixes. #rstats github.com/jolars/arity

github.com

GitHub - jolars/arity: Language server, formatter, and linter for R

Language server, formatter, and linter for R. Contribute to jolars/arity development by creating an account on GitHub.

1102

Wolfgang Viechtbauer @wviechtb.bsky.social · 14/08/2026

Wildfire burning nearby in the High Fens. Another wildfire is burning nearby in Hürtgenwald. #wildfire #highfens #hurtgenwald

Foto at night over a forest towards a wildfire in the distance illuminating the sky in red.

000

Wolfgang Viechtbauer @wviechtb.bsky.social · 12/08/2026

Built a pinhole projector with the kids to get ready for the #solareclipse tonight. We will only get about 89% occlusion, but should still be fun to see.

Photo from inside the pinhole projector showing the dot of the sun.

060

Reposted by Wolfgang Viechtbauer

František Bartoš @fbartos.bsky.social · 05/08/2026

For the revision for our "Meta-Analysis with JASP, Part I: Classical Approaches" tutorial, I recorder a series of YouTube videos walking through the examples in JASP www.youtube.com/watch?v=mbou... First time doing this. Do you find walkthroughs like this useful? Let me know any suggestions!

youtube.com

Meta-Analysis with JASP, Part I: Classical Approaches (Example 1)

YouTube video by JASP Statistics

095

Reposted by Wolfgang Viechtbauer

Alfredo Sánchez-Tójar @asanchez-tojar.bsky.social · 29/07/2026

🐦 Thrilled to share our latest research, a milestone: the first first-author paper from @shreyadimri.bsky.social, published in @royalsocietypublishing.org Why do birds add fresh green nest material to their nests? 🧵 🔗 doi.org/10.1098/rspb...

12911

Wolfgang Viechtbauer @wviechtb.bsky.social · 24/07/2026

That would be my guess. Most of the followers are male ...

010

Wolfgang Viechtbauer @wviechtb.bsky.social · 24/07/2026

What's up with all these apparently fake data science accounts? bsky.app/profile/ds-w... bsky.app/profile/ds-f... bsky.app/profile/ds-j... bsky.app/profile/ds-b... bsky.app/profile/ds-n... ... What's the endgame here?

190

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/07/2026

I don't know of any paper that states this, but given that the BLUPs are unbiased and the group-level means vary randomly around mu, it follows that the mean of the BLUPs are unbiased for mu.

030

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/07/2026

Assuming that the data generating mechanism is mui ~ N(mu, tau) and yij ~ N(mui, sigma), then I think it must be the case that the mean of the BLUPs is unbiased for mu (since the BLUPs are unbiased for mui). This will be the case whether groups are balanced or not.

120

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/07/2026

But that won't be the case when the data are unbalanced.

120

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/07/2026

With overall population mean, do you mean the intercept?

100

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/07/2026

But coming back to y ~ 1 + (1 + group). The mean of y within groups is unbiased for the group-level means and so are the BLUPs (intercept + u-hat_i), but the BLUPs have lower variance.

100

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/07/2026

Hint: What does the U in BLUP stand for?

100

Wolfgang Viechtbauer @wviechtb.bsky.social · 12/07/2026

Oh no, you broke my workflow when I need to generate ALL integers quickly!1! (xkcd.com/1172/)

xkcd.com

Workflow

140

Reposted by Wolfgang Viechtbauer

MikeKSmith @mikeksmith.bsky.social · 07/07/2026

Inspiring talk from Dariia Mykhailyshyna about teaching and surveying the R community in Ukraine while in the midst of a war. #rstats #useR2026

1409

Wolfgang Viechtbauer @wviechtb.bsky.social · 30/06/2026

I only see "Video not found". Might just be me, but would love to see what you have done!

110

Wolfgang Viechtbauer @wviechtb.bsky.social · 23/06/2026

Got it - thanks to all who replied!

000

Wolfgang Viechtbauer @wviechtb.bsky.social · 22/06/2026

The standard random-effects model in meta-analysis assumes that the amount of heterogeneity is homoscedastic. This assumption may not be true. I derived and examined various tests for heteroscedastic heterogeneity. See osf.io/cavhy for a preprint describing these methods. #MetaAnalysis #RStats

osf.io

OSF

13112

Wolfgang Viechtbauer @wviechtb.bsky.social · 22/06/2026

As another alternative to the standard forest plot for visualizing many effect size estimates in a meta-analysis, one can use 'orchard plots' (Nakagawa et al., 2021). Example code to create such figures can be found here: www.metafor-project.org/doku.php/plo... #MetaAnalysis #RStats

An orchard plot showing many effect size estimates clustered by a categorical moderator variable, with pooled estimates (with 95% CIs/PIs) within subgroups superimposed.

2287

Wolfgang Viechtbauer @wviechtb.bsky.social · 18/06/2026

In large meta-analyses, standard forest plot can become excessively large. One solution is to group the estimates together within studies. I just added an example to the metafor website illustrating this possibility: www.metafor-project.org/doku.php/plo... #MetaAnalysis #RStats

A forest plot where the effect size estimates within studies are grouped together.

0347

Wolfgang Viechtbauer @wviechtb.bsky.social · 15/06/2026

The updated data and code for my chapters can be found here: www.wvbauer.com/doku.php/pub... (they will also be updated on the book website soon).

020

Wolfgang Viechtbauer @wviechtb.bsky.social · 15/06/2026

The third edition of the Open Handbook of Experience Sampling Methodology is out! The book includes 14 chapters that cover all aspects of designing, running, and analyzing the data from an ESM/EMA study. The book is freely available here: www.kuleuven.be/samenwerking...

Cover of the Open Handbook of Experience Sampling Methodology.

13112

Reposted by Wolfgang Viechtbauer

Alfredo Sánchez-Tójar @asanchez-tojar.bsky.social · 07/06/2026

Had a fantastic time last week in Trondheim at the @eshackathon.bsky.social 🇳🇴 A full week discussing evidence synthesis & how to improve it with a brilliant and knowledgeable interdisciplinary team. Highly energising! 🙌 Thank you to @drmattg.bsky.social for hosting us & taking such good care of us

072

Wolfgang Viechtbauer @wviechtb.bsky.social · 02/06/2026

Indeed. This is why I think the consequences for such behavior should be very severe.

000

Wolfgang Viechtbauer @wviechtb.bsky.social · 02/06/2026

Isn't it nice when actual humans make mistakes for a change? 😉

010

Wolfgang Viechtbauer @wviechtb.bsky.social · 02/06/2026

This should even apply to all journals from the same publisher. The publisher should also approach the university of the authors and demand disciplinary consequences. If these are not made, then the entire university should be banned.

020

Wolfgang Viechtbauer @wviechtb.bsky.social · 02/06/2026

It wastes the time of editors and reviewers, who are already under severe time pressure. Basically, this breaks the entire system. I do not know who the authors are. However, the journal does. IMHO, such behavior should lead to a lifetime ban from ever submitting an article to the journal again.

131

Wolfgang Viechtbauer @wviechtb.bsky.social · 02/06/2026

So the authors were either too lazy or incapable of reading the generated text to detect this inconsistency. When I realized what was going on, I was livid. I certainly see the usefulness of LLMs for certain tasks, but this is abhorrent.

140

Wolfgang Viechtbauer @wviechtb.bsky.social · 02/06/2026

I recently reviewed a paper for a top journals in statistics that was written front to back with a LLM. To make things worse, the conclusions drawn from the simulation study did not match what the results actually showed and instead just conformed to what the authors intended to show.

1114

Wolfgang Viechtbauer @wviechtb.bsky.social · 14/05/2026

Yeah, looking forward to finally having the time to learn R. Unless I go blind with that monitor placement ...

030

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/05/2026

I prompted it to imagine what my ideal retirement will look like. Not sure who is writing in my notebook - definitely not me, because my handwriting is terrible 🫣

000

Wolfgang Viechtbauer @wviechtb.bsky.social · 13/05/2026

I asked ChatGPT to imagine what my ideal retirement will look like. This is what it came up with.

A relaxed retired academic sits in a cozy lakeside study surrounded by books, chess pieces, Linux laptops, and R code, enjoying coffee while overlooking mountains and a peaceful European village.

380

Reposted by Wolfgang Viechtbauer

Mattias Villani @matvil.bsky.social · 12/05/2026

R-folks, is the RTMB package the go-to autodiff package in R now? Limitations? Fast? Alternatives? #RStats #r4ds

012