My take on multilevel models based on the literature: YOU HAVE TO USE MLMS! Except they have assumptions that are never met. & when those assumptions aren't met they are biased. & the 'right' models never converge. & other approaches are wrong bc they aren't MLMs. But you have to use them. #stats

I am happy to announce the publication of our new work on the impact of arbitrary analytical choices on type I and type II error rates. We simulated reaction time data in a conflict task and analyzed the notable CSE effect in a multiverse manner. Worrying results: link.springer.com/article/10.3...

Academia.edu is changing its Terms and Conditions, and they're so outrageous (AI training; signing over ownership to them etc) I closed my account today. A thread with some practical tips if you want to delete your account (5 steps):

Proposal for how to fix family wise error rates. For every uncorrected p value you must add an extra letter to the claim. “Eating chocolate maaaaaaaaay be associated with lower rates of stroke”

Preschool-aged kids whose parents are working class are less likely to be called on when they raise their hand compared to kids with middle/upper class occupations. Early socialization indeed. | From @lewisdoyle.bsky.social & @andreicimpian.bsky.social www.pnas.org/doi/pdf/10.1... #psych #phdsky

Say it with me: Demand effects do not exist in online studies. Participants in online studies don't care about our research. They aren't trying to be good subjects. They want their $. #psych #meta

Researchers: Stop Thinking You Know How Things Have Changed, Go Measure It! Your intuitions are biased. The past isn't how you think it is. Measurement Invariance is a mindset & a must. Our new paper on intuitive vs real change in society/people. osf.io/4v6mf_v1/dow... #psych #phdsky #sociology

RIP Firefly

PsyArXiv's amazing team of 100+ moderators has now approved all preprints that meet the requirements outlined in the updated PsyArXiv policies (is.gd/paxpolicy). Thank you to everyone who volunteered, this was a true community effort! #PsychSciSky

There is a growing inequality in reading for pleasure. Those who read are reading for longer now than 20 years ago. BUT when asking people what they did in the past 24 hours fewer said they read (28% to 16%). From Jessica Bone www.cell.com/iscience/pdf... #psych #phdsky

How do we assess content learning w/out accidentally tapping a general cognitive ability for memory/knowledge? Seems Θ in IRT knowledge scale will covary w/a general memory/knowledge ability but when teaching that's not what we want. We want to know our scale is assessing learned knowledge. #stats

Algorithms are not necessary for the creation of echo chambers. All that is required is people muting or leaving when they have a bad interaction. Then an echo chamber arises with minimal preferences. From @pettertornberg.com #psych arxiv.org/pdf/2508.10466

Biggest concern in mediation is mediator & outcome confounding, even when X is random. Comparing methods: how the indirect effect acts when errors of M & Y are allowed to correlate is a great option. From Alvarez-Bartolo & MacKinnon journals.sagepub.com/doi/pdf/10.1... #stats #psych #phdsky

If the administration releases the data for the causes-of-autism study that is supposed to be reported this month, then it could be useful to have a crowdsourced many analyst/multiverse investigation ready to go that leans-in on pre-commitment to analysis plans. Interested? Alternatives?

Here is a how-to guide for writing the next social change or technology panic article. Included is a fillable template, a how to guide to running a correlational study nearly guaranteed to work, suggested language (e.g. save the children), & more. Free! #phdsky osf.io/cvnhu_v1/dow...

Are you doing research on impression formation, face perception, personality judgment, or related topics? Then you might be interested in joining our collaborative study! Follow the link for more information: tilburgss.co1.qualtrics.com/jfe/form/SV_... #socialpsyc #PsychSciSky

Final call for global collaborators on this replication collaboration on @bioxphi studies — moral psychologists and x-phi researchers wanted! especially based in SOUTHERN HEMISPHERE! global-bioXphi | moral science lab www.mscilab.com/globalbioxphi/

Other nautical terms we still use: ➡ Flagship:...The ship with the flag; I should have known this one ➡ Taken aback: When a ship's sails are caught aback, pushing the ship backwards ➡ Above board: On or above the open deck, in plain view ➡ Loose cannon: Literally a loose cannon on the ship

As a totally separate point: the effect of choice didn't replicate in humans in a large-scale preregistered replication! And the Cohen's ds in the original paper range from 1.3 to 4.17. journals.sagepub.com/doi/full/10....

It's that time of year again where the Yahoo Sports app asks for permission to send me notifications and then in 24-hours acts like Charles Boyle in a groupchat and blows up my phone so bad I have to remove notification access again until next year when I forget this lesson again.

New preprint! A Crowdsourced Megastudy of 12 Digital Single-Session Interventions for Depression in American Adults What's the best way to help someone overcome depression with just 10 minutes of their time online? We ran a trial (n=7,505) testing 12 diverse interventions. osf.io/preprints/ps...

Just typed "shouldntve" to a fried and realized non-Americans eyes would likely start twitching

Did you register to help moderate PsyArXiv preprints but haven’t heard from us yet? Check your spam folder and add [email protected] to your contacts! We are onboarding moderators in batches to avoid overloading the system, but rest assured we will get to everyone. #PsychSciSky

Help us show there is support for offering registered reports at one of Psychology's leading method's journal! chng.it/TwwnVBScVb

Want to help moderate preprints on PsyArXiv? We're hosting a training tomorrow morning. Register here: us06web.zoom.us/j/8114764450... More trainings are coming as well! Already registered using the other form? Check your spam folder -- some of our email invitations are going to spam. #PsychSciSky

Generalizability issue: we use participants in our studies & the type of people who are participants are a different type of person than non-participants. ⬆️age, intelligence & education, conventional ⬇️sociable, arousal-seeking, interested in religion, self-disclosing osf.io/gh2ck_v2/dow... #phdsky

The ManyTones project is looking for several lead positions. See poster for details. 📢📣✨ Get involved! 🎉🤝🌟

Still can't upload new versions of preprints to psyarxiv. Site just keeps buffering with a charming blue dot. ...am I alone in this?

Looking for an example from #physics #chemstry #ecology #medicine or any non-social science of changing causal strength. Where at t1 causal effect is X and at t2 a causal effect is different. That doesn't come from modeling (e.g. Network, SEM...) but in the same system causal effects change.

All of Statistics is secret linear algebra. We often don't do the linear algebra or don't know we are doing it because of assumptions. Here is a more accessible resource for what you're really doing when you analyze human subjects data.

PsyArXiv is seeking new moderators to help combat an increase in AI submissions! If you've ever posted a preprint to PsyArXiv, please consider joining. Minimum commitment 1h/month, there's a training session this Monday @ 1pm ET. More info here: forms.gle/9LB1rEtxHAeZ... #PsychSciSky

"If your books were uploaded to LibGen/fed into the LLMs, add your name to the potential class action lawsuit by tomorrow." Details: LibDem database: www.theatlantic.com/technology/a... Attorney form: www.lieffcabraser.com/anthropic-au... (This information comes via a reliable Discord source.)

Kids These Days...behavior problems aren't changing. In over 418,000 children from nationally-representative samples, child behavior problems are pretty similar as in the 1980s, with most changes being improvements, not declines. Our work led by Zsofia Takacs #psych #phdsky osf.io/63egm_v1/dow...

Oups. 😄 It seems my close colleague forgot to cut my joke at the end of the sentence... And then not the editor, nor the people in charge of editing considered necessary to change it before publication! #SciencePublication #FunButSad #SadButTrue #FrenchAcademic

Why is this a problem, and can't we just ignore this low quality stuff? 1. It potentially pollutes research reviews. Plus other AIs then gobble up the garbage. 2. It wastes resources. Each paper has a cost, even if that cost is not borne by authors or readers.

The difficulty with writing methods articles aimed at non methods savvy researchers is they are reviewed by methods savvy researchers who insist everybody knows this & nobody does the thing you are trying to get researchers to stop doing. So you have to keep showing that people are doing the thing.

Have you ever seen this call for SIPS EC nominees and thought "it's probably not for me"? Don't be so sure! If you think you might be interested but have questions, feel free to DM or email me, I'll do my best to answer them :) #PsychSciSky

Fadeout is everywhere. In 11 studies that randomly assigned weight-loss drugs, there was significant weight regain after discontinuing. There was, however, some persistent weight loss 1 year after stopping. From Han Wu & Linong Ji link.springer.com/content/pdf/... #phdsky

🚨 Deadline Extended! 🚨 We’ve extended the #BigTeamScience conference submission deadline to August 8! 🧠🌎 Don’t miss your chance to be part of this global, interdisciplinary event! docs.google.com/forms/d/e/1F...

The abstract submission deadline for the pre-Psychonomics symposium on individual differences in cognition (SIDIC), to be held on Nov. 20, is due on Friday, August 1, TWO DAYS FROM TODAY! Here's the direct link to the abstract submission information: caliberlab.wixsite.com/sidic2025/su...

Hey folks: A little while aback, @jschleiderphd.bsky.social and I created a repository of SOPs to help folks prepare for grad applications. If you'd like to add your SOP, feel free to use this form: docs.google.com/forms/d/e/1F... Of note, mask as you desire as I will be posting them as is.

For those interested in the one replicable Construal Level Theory study: People think it is more useful to think about WHY they would perform a high-level task (e.g. delay discounting) and more useful to think HOW they would perform a low-level task (e.g. Proofreading).

Since the 1990s, rates of anorexia nervosa have been falling. Yet people and practitioners way overestimate not only the rate, but also this change. From my student Erin Sullivan osf.io/gjncz_v1/dow... #psych

PSA: Birth rates are not fertility rates. Birth rates include the entire population as the denominator, and fall with population aging, even if young adults are having the same number of kids as before.

If you just got an email from APA encouraging you to come to this session, YOU SHOULD! Because the "Open Science and Methodology panel member" is ME! Come say hi! Learn about registered reports, and other awesome open science things at APA!

Just got 3 error messages that osf is having problems registering my study. Sure glad they got rid of the files window in order to bring us this new function...

Yoooo! Another Dutch people doing unethical things in train stations study just dropped! Very retro! But seriously, Dutch people let white people skip paying 67% of the time but only 49% of the time if the skipper was middle eastern. From Kasper Otten journals.sagepub.com/doi/pdf/10.1... #psych

What is SES? We experimentally test the big 3 components in lay & expert participants. Once you dissociate $ from job from education, it's mostly just money (but not exclusively!). Experts also overvalue education. This was one of my favorite projects! osf.io/76w9m_v1/dow... #psych #econ

Incredibly important results from @markfabian.bsky.social and coauthors on cognitive interviewing of life satisfaction ratings. Does not exactly inspire confidence in the scales like Cantril Ladder