Love methods papers where the authors cover an estimator’s properties in painstaking detail, then add a footnote:
“We assume X, Y, Z, which are unrealistic, but relaxing them is too complex”
Thanks. I suppose it’s up to us applied researchers to deal with those gaps then?
Opened Bluesky for some USA vs Canada hockey banter.
Everyone’s discussing R pipes. 😑
Novel, non-robust methods instead of something tried and tested? Sign me up! Sounds like a perfect fit for research
Not even daemonless or rootless containers? I thought they were natively supported by slurm atleast
Ironically, I was gonna demonstrate that a project that I archived using docker 7+ years ago still runs without issues. But I cannot remember where I stored the archive 🤷♂️
I guess I should've put it on a satellite.
No, everyone should not learn how to run their own Kubernetes cluster on arch. This should be handled centrally by the university when projects are archived.
Yes, archive OCI (Open Container Initiative) images for each project. Then we can simply docker / podman pull & run, without installing and fixing broken system or R package deps. This is much more future-proof and convenient than any R-only solution
Fun fact about LMMs: when you have missing data, power also depends on the correlation between intercepts and slopes. In the clip, I’m varying the amount of random slope variance.
Just added to PowerLMM.js v0.3: Interactive power contour plots!
Visualize how statistical power changes across parameter combinations.
powerlmmjs.rpsychologist.com?view=contour
A
Try it out and let me know what you think, it is still a bit experimental
Just added to PowerLMM.js v0.3: Interactive power contour plots!
Visualize how statistical power changes across parameter combinations.
powerlmmjs.rpsychologist.com?view=contour
Wasn't that happy with how performance crashed as time points grew, so I dusted off some linear algebra to optimize the calculations. Benchmark showed a ~1Mx speedup using new implementation (with 100 time points) 😅
Screenshot of https://powerlmmjs.rpsychologist.com/
🎉 @rpsychologist.com 's PowerLMM.js is the online statistics application of the year 2025 🎉
powerlmmjs.rpsychologist.com
- Calculate power (etc) for multilevel models
- Examine effects of dropout and other important parameters
- Fast! (Instant results)
Dammit 😅
Let's make it happen. I'm ok with doing whatever I want for 1 week
Who’s gonna pay @rpsychologist.com to make a mega tool converting all g*power analysis types to an interpretable web tool like this?
Sign me up 🫡
v3 will recruit the participants for you
Thanks 🙏
GitHub version is still functional (dev branch)
No immediate plans to re-submit to CRAN, but open to collaborations/co‑maintainers.
Plus:
- New welcome screen
- Guided help tour
- View correlation matrix as a heat map
- and other UI improvements
☕
Calculations have been validated against R.
Validation notes: github.com/rpsychologis...
Share your analysis as a URL, or store it as a JSON file.
For example:
powerlmmjs.rpsychologist.com?n1=13&time_e...
Power analysis summary report, that can be printed or saved as a PDF
Hypothesis region visualization. Interactive display of the chosen null and alternative hypothesis.
New release of PowerLMM.js! Browser-based power analysis for longitudinal models with dropout.
Now includes:
- Power analysis summary report
- Reproducible & shareable configs (URL/JSON)
- Calculations validated against R
- Hypothesis region visualization
powerlmmjs.rpsychologist.com
