Are you curious about open peer review, but don't know where to start? ⏰️ This is your last chance to register for tomorrow's Community Call, where our speakers will introduce this concept, walk you through its ins and outs, and answer your questions during the Q&A. Join us! buff.ly/7MpqKYB

I'm looking for a cite on the potential of hidden moderators as responsible for failed replications. This was a regular dialog in peak replication crisis, but can't find/recall a specific cite arguing or laying out this possibility. Maybe a perspectives piece?

Now out in the American Sociological Review We present the first large-scale assessment of the structure and evolution of temporalities expressed in U.S. climate change news coverage (2000 to 2021). For this, we analyzed more than 23,000 statements about climate change effects and actions. 🧵 1/

My dotfiles now manage my AI coding agents 🤖 Same instructions for Claude Code and OpenCode, symlinked from one repo. Here's how I set it up: https://drmo.site/CCrf37

Your model is only as good as its assumptions. 📊 But what happens when your data breaks the rules? Let’s dive into how to check your model assumptions—and exactly how to fix those pesky violations: 🧵👇 easystats.github.io/performance/... #rstats #easystats #performance

in my decision making class I run a version of The Endowment Effect that goes like this: When students come in, sealed envelopes are waiting on their chair-arms. The main slide that greets them says 'NO TOUCHING'

The critique of unmeasured confounding is often levied in a lazy/broad way. It is trivially true in any observational study. But if the critic can't think of a plausible such confounder and posit a reasonable direction/magnitude of its bias then they're not doing productive science.

All your p-values are wrong Nice blog by Guillaume Rousselet garstats.wordpress.com/2026/02/27/p...

I've been testing Claude to see how well it can "vibe out" a stat. power app that I've already coded completely myself - so I know what I want. It mostly gets things right with animations (those are easily verifiable) but looking into the backend stats code is nightmare inducing (see pic).

Mein Buch #Einsamkeit erscheint am 25. März! Die ersten Termine für Lesungen stehen bereits fest. (Details im Thread) 📅 17.03.2026 📍München 📅 18.03.2026 📍Berlin 📅 19.03.2026 📍Leipzig 📅 20.03.2026 📍Leipzig 📅 26.03.2026 📍Hannover www.fischerverlage.de/buch/prof-dr... #Lesereise #Sachbuch

Great new paper led by @bringmannlaura.bsky.social, highlight the need to collect qualitative data in ESM / EMA research. #PsychSciSky 🧪 www.nature.com/articles/s44...

I wrote a short blog post that describes selection on significance in plain language and then proposes and criticizes two alternatives (selection on precision and registered reports). This is me working through my thoughts online, so feedback is very welcome. ryancbriggs.net/blog/the-pro...

New Paul Meehl Graduate School course: Advanced Basic Statistics Workshop: Deepening Your Understanding of Standard Statistical Analyses March 27, 2026 Sign up: paulmeehlschool.github.io/2026-02-23-a...

Aside from the bonkers horse analogy, this argument belies an unfortunately common perspective that the reason why teachers and professors assign homework is because it somehow benefits us for that work to get done.

Am I understanding this right? The standard grading scale in Denmark goes –3, 00, 02, 4, 7, 10, 12 ! This is amazing. I kinda love it. I mean I hate it. But I love it.

Lots to say on this, but one thing recently came to my mind is the isomorphism between “wide format” & “long format” for panel datasets. The latter treats time as a variable, which confuses the issue, at least conceptually. DiD with wide format is clear and here: journals.sagepub.com/doi/10.1177/...

The northeast blizzard ain't stopping us! You won't want to miss the Moral Psychology Pre-conference at @spspnews.bsky.social in Chicago this week! It should be an incredible day of sharing research and networking. Hope to see you there.

I'm hiring a postdoc at @cmu.edu (w/ far.ai & @dgrand.bsky.social + @gordpennycook.bsky.social)! How do LLMs shape human beliefs — and what do we do about it? AI safety meets behavioral science. Open to technical and social science backgrounds.

There's promise in using LLMs for code review, but it's tricky things to make sure it's not overwhelming. I was looking at this new experimental package by Simon Couch and I really love how it allows you to review code iteratively. #rstats #ai #llms github.com/simonpcouch/...

Interesting start to the week: Paper rejected with the justification that the journal does not accept pre-registrations or manuscripts that are based on pre-registrations 🤔🤪 #Registration #PreRegistration #RegisteredReport

The lack of uptake of sequential analysis shows how irrational scientists are, and how their methods are driven by norms. Sequential analyses give you more flexibility and are more efficient than a single hypothesis test, and yet, they are still very rare. lakens.github.io/statistical_...

Does your R-script work on your computer? Want to make it fully reproducible on ANY system, now and in the future? In this video, I demonstrate how to turn R scripts into reproducible, version-controlled projects with worcs, renv, targets, and testthat to verify reproducibility. youtu.be/3mgRFMr5APU

spending my sunday evening once again attempting to draw a DAG for diff-in-diff

Remember that you compute power for the smallest effect size of interest. Not the effect you hope for. The smallest effect that you do not want to miss. This is in line with the basic idea of conditional error control: *If* there is an effect large enough to matter, we are likely to detect it.

Wonderful to see this replication effort in the physical sciences using the models of many labs, preregistration, and transparency that have benefitted other fields. And, an investment of $9.5 million to do it! www.nature.com/articles/d41...

After watching Richard McElreath's lecture on measurment models, I got comparatively more excited about my working paper with Yaroslav on accounting for non-classical measurement error in belief-updating experiments and my foray into measurement and time-series econometrics. A thread. #EconSky

📣 Join @katiecorker.bsky.social (ASAPbio), @thewilkybarkid.bsky.social, and @neurosarda.bsky.social (@prereview.bsky.social) on March 5th for the Community Call! Speakers will discuss the ins and outs of doing open peer review as part of a team🤝 Register 👇️ buff.ly/5I0wC4l

I've been interested when we were going to reach the capability threshold for this. Once the tires have been kicked, all social science journals should adopt computational reproducibility on submission. Huge reduction in reviewer burden by eliminating a whole class of errors up front.

How to draw propensity scores (PS) in DAGs? Some (me also) claim it is like "treatment -> PS <- covariates", since in order to compute PS we need both treatment and covariates. This view has confused me for so long, and now I think I was wrong. My letter here: track.smtpsendmail.com/9032119/c?p=...

The Big 5 test is about twice as accurate as the Meyers-Briggs Type Indicator for predicting life outcomes, placing the usefulness of the MBTI test halfway between science and astrology. Any psychologist will tell you, the Meyers-Briggs is mostly bullshit. powerofusnewsletter.com/p/why-are-we...

Bots have made their way to Prolific experiments. Our lab has stopped online testing of adults entirely now for this reason - we want to know if what we study is real. Probably data collected 2-3 years ago are ok, but moving forward we just can't know. www.pnas.org/doi/10.1073/...

Great blog by @deevybee.bsky.social on how often PubPeer entries are preceding highly cited retracted articles. Fully agree with Prof. Bishop's description of PubPeer. I am glad that this platform exists because it is a non-annoying way to point out errors/problems.

Recent work has shown how vulnerable online survey research is to LLMs. Motivated by this, we examined our online Posner cueing data from Prolific. It's concerning. We now must carefully consider when (or whether?) online behavioral data can be trusted. see our comment: www.pnas.org/doi/10.1073/...

It's funny, frustrating, and satisfying in equal measure to spend all day on a DAG, fitting propensity weights, checking covariate balance, and finally fitting a weighted model, then getting almost the same exact estimand from a linear model you fit in <20 seconds 😎🔫 #statsky #rstats #episky

Some more details on the omnibus test for whether the ADRF is flat, as I mention in this test. We have an estimate for the ADRF at each treatment value `a`, and jointly they have a multivariate normal distribution. This makes the ADRF estimate a Gaussian Process. 1/8

Are you a junior faculty member interested in spending 2-4 weeks at Princeton Psych? Consider applying for our Microsabbatical program! It’s a fully funded visit for professional development and creating long-term collaborations. psych.princeton.edu/diversity/mi...

I'm so excited to announce the first release of my newest #Rstats package, {adrftools}! This package facilitates estimation, visualization, and testing for the causal effect of a continuous (i.e., non-discrete) treatment. 🧵 1/10 #statssky #episky #causalinference

📣Join the SIPS Awards Committee!📣 The Committee is responsible for: 💠Soliciting awards nominations, 💠Selecting winners, 💠Proposing new award types. All in collaboration with the Executive Committee. Learn more 👉

== Developing tools for understanding and improving science using AI (vibe) coding == A talk I gave at Weizmann Institute of Science about AI (vibe) coding solutions for understanding and improving science is now online: www.youtube.com/watch?v=1IBm... 🧵👇

Have you ever wondered what you would find if you could keep your eyes on a bee for more than a few meters? Us, too! preprint (with videos!) + thread 🧵 Precise, individualized foraging flights in honey #bees 🐝 revealed by multicopter drone-based tracking www.biorxiv.org/content/10.6... 1/9

poor bee GOLD_07, whose embarrassing mistake has been immortalised in a way beyond his comprehension

We ♥ Free Software! On Love Free Software Day, huge thanks to the maintainers of tools we rely on and teach about daily: OpenAlex, R, Quarto, Git, renv, Zotero, RDMO, formR.. Free software is collective infrastructure for open, reproducible research. Thank you! #OpenScience #FOSS #Reproducibility

If you set out to test a hypothesis, you should preregister it. If you deviate from a preregistration, report a table with all deviations, and evaluate the consequences for the validity and severity of the test. As a reviewer, ask for such a table! online.ucpress.edu/collabra/art...

Currently a lot of attention is going to get people to preregister, to check preregistrations, and noting deviations. But the most important step will be evaluating the impact of the deviations! There is almost no work on this. Good topic for a young metascientist!

We're pleased to have @betanalpha.bsky.social join us in Princeton tomorrow (Feb 17th) to talk about generative modeling! Talk: 11am Workshop (with Stan): 12:15pm The zoom is open to public (see flyer). @princeton.edu

CAUSAL GRAPHS ➡️ HAPPINESS. 😍 Agree? & Work in/around Berlin? 🐻 Sign up for the 2026 Applied Causal Graphs Workshop! 🌟 Deadline Feb 28th ⬇️ applied-causal-graphs.de This 3rd edition is hosted by the great folks at Uni Potsdam 💡

Great piece from @dsquareddigest.bsky.social, which touches on one of the big theories for why English-speaking countries do especially badly at housebuilding: Adversarial and litigious common law systems (Anglo) vs judge-led civil law systems elsewhere. samf.substack.com/p/build-the-...

Today as always, @citizencohn.bsky.social is a must read. Here he is on the massive danger to public health of the latest anti-vax actions by Trump’s and RFK’s FDA. open.substack.com/pub/thebulwa...

This is the grammar of dispersion measures in #tidyplots 🤩 You can mix and match as you like. #rstats #dataviz #phd

In standard scientific reports in psych 96% of first mentioned main hypotheses are supported. In Registered Reports, this is 46%. The 96% is clearly biased (given true H1 rate and power). Lack of transparency means we do not know the true baserate. journals.sagepub.com/doi/10.1177/...