K isn't real! What are we doing here? Why is a textbook wasting their time with these racist freaks and their nonsense

Men witness false bear; bear false witness

Yeah, some of the potential applications of LLMs really do seem incredible (“help you remember many of the beautiful moments in your life that you’ve forgotten”) but are totally untenable in the current environment of functionally zero regulation of data sharing (to say nothing of gov surveillance)

My lab (u.osu.edu/spal/) and Ken Fujita’s lab (u.osu.edu/fujita/) at OSU are hiring a joint lab manager (a full-time postgraduate researcher) to start this summer!

Come work in two incredibly fun labs in a thriving social psychology department.

Application and more information soon to come!

here we go, my friends. no homework episode 2. please let everyone know we're having fun.

1/
"Silicon samples" are becoming more and more common in research and polling.

One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want.

The updated version of this preprint is now online!

THREAD🧵

arxiv.org/abs/2509.13397

ggsql: A grammar of graphics for SQL Introducing ggsql, a grammar of graphics for SQL that lets you describe visualizations directly inside SQL queries.

I am excited beyond description to lift the veil on what we have been working on in 2026:

Please meet ggsql! A new extension of the SQL language for creating visualisations using the grammar of graphics. Read all about it in the blog post or visit the website at ggsql.org

Why was I invited to Beast Studios? YouTube video by Folding Ideas

Why was I invited to Beast Studios?
youtu.be/0dwagg5wYY4

This is a really good column and really good advice for people entering the job market

Lawcop needs to come out of retirement for one last job

I forget who it was, but there was a academic on Twitter who had a fun recurring bit “did Stapel get more citations than me this year?” And the answer was always yes.

Each year, people speak 338 less words per day (on average). These effects accumulate. In 2019, people were speaking 28% less words each day than in 2005 (!).

journals.sagepub.com/doi/full/10....

🚨New preprint and our results are rather concerning..

We find the "boiling frog" equivalent of AI use. Using large-scale RCTs, we provide *casual* evidence that AI assistance reduces persistence and hurts independent performance.

And these effects emerge after just 10–15 minutes of AI use!

1/

This paper claims that watching sports causes zero-sum thinking, with experimental and longitudinal evidence
psycnet.apa.org/fulltext/202...

Yeah totally! I can def see people interpreting all of these carefully distinct questions as variants of 'how happy are you?' (or, alternatively, as opportunities to voice that they're happy).

bsky.app/profile/theo...

Agreed about the cool work, and I had a similar question: is it that in our theories these distinctions are meaningful but our measurements of them get overwhelmed by positivity, or is it that the theoretical distinction isn’t meaningful?

Damn you’re really well sourced

I’ve never seen an LLM do cocaine before

Pic showing the top users on bluesky by engagement, with Mrs. Betty Bowers in the #3 spot

Number go up

Oh nice, thanks for this! Yeah Joe’s journey here was very similar to mine

(If your model performs worse than just predicting the mean value for every data point, then your R-squared will, apparently, be negative)

A few months ago I ran a model that had a negative R-squared. I didn’t even realize that was possible.

Come share your research at NYU!

The NYU Social Psychology Program is inviting speakers for 2026-2027! If you’re passing through NYC next year and are interested in giving a research talk nominate yourself using this short form by MAY 15th: docs.google.com/forms/d/1yTI...

I should probably note that this was in the early-mid 2000s and hopefully things have improved (but the ed department was very mad at our department for a while, from what I hear)

Nature meta-research project puts claims in social-science paper to the test. Refs in last post

I'm interested in Econ and Psych so I focused on that:

Econ had about the same rate of "not reproducible" analyses as Psych and a worse rate then Political Science.

Yeah I think that nails it. It’s also a space with many very-well-intentioned and motivated-to-help people coupled with (imo as an outsider) slightly weaker norms of data stewardship, which can lead to blinders + replicability issues

At Ohio State they had a policy where one dissertation member was from an external dept (not chosen by the student) and had veto power. One prof had to exercise that power because an ed student basically said “well, the p-value is above .95, so there’s only a 5% likelihood this is due to chance”

The psychology research on learning and memory is also pretty good, but most of those are lab studies rather than classroom interventions to boost performance (which are both harder to collect the data and also academic performance heavily multiply determined).

SCORE | Center for Open Science SCORE shows that there is no shortcut to producing credible research findings, and there is no single indicator of trustworthiness. Research progress depends on transparency, rigor, and establishing r...

SCORE, a collaboration of 865 researchers, is now released as three papers in Nature, six preprints, and a lot of data (cos.io/score/). SCORE examined repeatability of findings from the social-behavioral sciences and tested whether human and automated methods could predict replicability.