I've been working on developing materials for my 'research group'. Mostly it is items that I would have found helpful as a student and general recommendations (including on LLMs) I made it all publicly available in the hopes it can be useful to others. I also welcome any feedback or suggestions

GitHub - pzivich/LabGroupMaterials: Collection of my lab group materials for students and trainees

Collection of my lab group materials for students and trainees - pzivich/LabGroupMaterials

github.com

Back on my blogpost shenanigans. Today, I decided to sketch out some thoughts I've been having about navigating the uncertainties inherent to interpersonal communication. www.i4replication.org/blog/the-bou...

My mental model for what a valid DOI can look like got a significant update today doi.org/10.1002/(SIC...

Two‐sided confidence intervals for the single proportion: comparison of seven methods

Simple interval estimate methods for proportions exhibit poor coverage and can produce evidently inappropriate intervals. Criteria appropriate to the evaluation of various proposed methods include: c...

doi.org

Mark has kindly invited me to do this interview. How I got into metascience, how my path diverged from the mainstream and more if you're interested. I enjoyed thinking about his questions. Hope you enjoy reading!

Mark Rubin@markrubin.bsky.social · 2w ago

Q&A with @devezer.bsky.social about metascience, open science, and her work in this area.

Psychological Science is running a special issue on "Advancing Psychological Science using Large Language Models", co-edited by me and @ruben.the100.ci. Check out the calls for submissions below! www.psychologicalscience.org/publications...

Trying to visualize a bunch of simulations and got annoyed that R was taking 2 minutes to render each plot, so I decided to do it in Julia and now it's been 20 minutes of waiting for packages to precompile 😮‍💨

I have found a great new #stats teaching hack to find new open datasets for teaching. I've been going to the Collabra: Psychology journal website, and typing in an analysis type (e.g., logistic regression) into the search bar. Both quick and high-quality data! online.ucpress.edu/collabra

English Version of "How to read Mathematics?" on the ArXiv now: arxiv.org/abs/2607.24136 German Version here: web.mathi.uni-heidelberg.de/ggt/schwer #math #teaching #highereducation #mathematics #mathsky

How to read Mathematics? A study guide

Reading a math textbook isn't like reading a novel. You've probably already figured that out on your own. But how *do* you read mathematical texts? And how can you actually learn from a book or lectur...

arxiv.org

Petra Schwer@petraschwer.bsky.social · 3w ago

You aim to understand a mathematical text all by yourself? I finally made an English version of my users guide on "How to read math?" You might want to download it here: web.mathi.uni-heidelberg.de/media/How_To... #math #readingmath #mathematics #matheducation #science #mathsky

The underlying study has no control group and only enrolls people with poorly controlled asthma. It's just pre/post comparison of asthma symptoms. Results are entirely consistent with reversion to the mean. Or seasonal change, for that matter. Come on people, we can do better!

Sammy Roth@sammyroth.bsky.social · 3w ago

Insane new research. Scientists gave asthma patients free electric stoves to replace their gas stoves, then checked to see if symptoms improved. Not only did symptoms improve, ditching gas was *better than medicine.* Details in today's newsletter: www.climatecoloredgoggles.com/p/electric-s...

This is an interesting and good take on multiverse analyses…

ArXiv Paperboy (Stat.ME+Econ.EM)@paperposterbot.bsky.social · last mo.

arXiv📈🤖 Multiverse analysis, abdication of responsibility and manufacturing of doubt By Modr\'ak

I get absurdly giddy when I see someone taking the time to try and put a precise meaning on terms that everyone just seems to throw around like it's all very obvious. Definitions! Such a wonderful concept!

Mark Rubin@markrubin.bsky.social · last mo.

Useful table of theory-related terms by @philippmusfeld.bsky.social et al. www.researchgate.net/publication/...

Join our Many-Analysts Simulation Study—yes, you read that correctly—accepted as Stage 1 Registered Report in (AMPPS) @psychscience.bsky.social. We're trying to find out how researchers' simulation design choices shape methodological conclusions. Further info: osf.io/8vcxh/files/...

Huh, learned a couple new things today! (that people think a fan can be harmful when it's too hot, and that this is indeed the case at certain threshold... which I guess makes perfect sense if one considers how convection ovens work)

Mridul K. Thomas@mridulkthomas.bsky.social · last mo.

Since we're in a heat wave: Despite apparently popular belief, running an electric fan helps even when it is hotter than 35C / 95 F. Humidity matters but 35C is a bad rule of thumb. Take it from someone who has spent a LOT of time above 35C.

Interaction effects are very common in psychological research, but they are typically hard to replicate. A suspected reason for this is measurement error. Because of the poor replicability, a common recommendation is to not model interactions. 🧵

Did a 90 minute webinar workshop for APA on depression measurement. They were kind enough to allow me to share this on youtube. youtu.be/5OwYjWao1hs

What Scale is Best? Understanding Sources of Validity Evidence for Depression Measurement

YouTube video by Eiko Fried

youtu.be

Important work. “The emerging literature blurs the line between integrity filtering, based on necessary but insufficient signals of validity like reproducibility of stated results or lack of fake citations, and epistemic filtering, which uses machine-detectable signals to judge scientific quality.”

Jessica Hullman@jessicahullman.bsky.social · 2mo ago

Using AI to support peer review seems unavoidable, but what quality checks should AI implement? We can take some lessons from metascience on the hard reward design problem that is AI review. I wrote a paper synthesizing a few points the emerging lit seems at risk of confusing. 1/

How do you make LLMs actually good at explaining new math? It's like reading a badly written reference for people who already know the subject. When I ask a question, it never matches my level. If I try to rephrase to test my understanding, it just sycophantically agrees.

Issue 50 of #rdmweekly is out, also marking the 1 year anniversary of the newsletter! 🎉 Thank you to everyone who has followed this newsletter this past year, and here's to another year of learning together! rdmweekly.substack.com/p/rdm-weekly...

RDM Weekly - Issue 050

A weekly roundup of Research Data Management resources.

rdmweekly.substack.com

Posting this for myself, mostly.

Ryan Moulton@moultano.bsky.social · 2mo ago

I am so sick of punchy final sentences now. Let it end with a whimper god damnit. If what you told me is good I won't care. Not everything needs a bon mot attempt.

It was my turn on the I4R blog this week. It mostly consists of links to things other people have said and it's not very long. That was a deliberate and surprisingly excruciating choice to commit to. www.i4replication.org/blog/the-con...

You thought correcting for heterogeneity is enough? What if the heterogeneity is heterogenous?? Hm? Did you think about that?!

Wolfgang Viechtbauer@wviechtb.bsky.social · 2mo ago

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

The math part went completely over my head, but I'm *very* glad I stuck through. I think the advice Prof. Litt gives at 18:44 is widely applicable to any context in which one wishes to use LLMs in a helpful way: "If it's at all hard to check correctness [of the output], just throw it away"

Sam Power@spmontecarlo.bsky.social · 2mo ago

Interesting viewing: youtu.be/7WGtiwvxQRs?...

Building the smoothbp R package, I had an intuition about what causes funnel pathology in hierarchical models. I was wrong. But chasing that intuition led somewhere more fundamental to the centring vs non-centring decision than I had hoped. New paper: arxiv.org/abs/2606.19655

A Flat Connection: The Pooling Factor and the Geometry of Centring in Hierarchical MCMC

Standard MCMC diagnostics ($\hat{R}$, effective sample size, divergence counts) detect whether a chain has mixed, but not why it has not. We ask whether the centring/non-centring obstruction in hierar...

arxiv.org

💯 (See also: www.the100.ci/2024/12/01/w...)

Demetri@phdemetri.bsky.social · 2mo ago

Don't fall in love with your paper drafts

I was wondering where the term 'nuisance parameter' was first introduced and came across this very nice website mathshistory.st-andrews.ac.uk/Miller/mathw...

Took some time to slow down and write some good documentation and tests for a little side project, and as usual this process highlighted several areas for improvement. I can't think of a time this *hasn't* happened.

new diamond open-access Journal of Statistical Ecology launched 🎉🎉🎉 jse.centre-mersenne.org @fredbarraquand.bsky.social @oaggimenez.bsky.social @marieaugermethe.bsky.social @vianeylb.bsky.social (apologies to any Bsky editors I missed)

I was browsing a psychology journal when I noticed this strange header. There's no further explanation I could find and clicking it doesn't lead anywhere, it's just... ominously hovering up there.

Wonderful news! Journal of Research on Research is launched. "if you are a researcher who researches any aspect of research, then this is the journal and community for you." Worth reading the whole editorial for a transparent and thoughtful account of this amazing community-driven effort.

Research on Research Association@roration.bsky.social · 2mo ago

We're excited to launch J·ROR, the Journal of Research on Research. A new open-access home for research on how research is funded, organised, conducted, communicated, and evaluated. Our first editorial: www.tandfonline.com/doi/full/10.... #ResearchOnResearch #MetaScience #OpenScience #STS

#statstab #543 How hazard ratios can mislead and why it matters in practice Thoughts: Another *insert effect size measure* has issues paper. If you use HRs you better study up. #observational #hazardratio #effectsize #misconceptions #causalinference link.springer.com/article/10.1...

How hazard ratios can mislead and why it matters in practice - European Journal of Epidemiology

Hazard ratios are routinely reported as effect measures in clinical trials and observational studies. However, many methodological works have raised concerns about the interpretation of hazard ratios ...

link.springer.com