Protzko

Sometimes, philosophers of science write more about the practices scientists actually do (descriptive); sometimes they write more about what scientists should do (normative). Too often, scientists reading philosophy of science confuse what philosophers says we do with what we should do. Don't.

PSA for the American (North, Middle, and South, I presume) collaborators: Europe just did the whole clock changing thing. Prepare for some meeting time confusion if you meet at fixed times

Your brain and mind are also not a computer program, a set of relay switches, a series of hydraulic pipes, a muscle, a blank slate, a charioteer with two horses, an ideal Bayesian updater, a...

Seed hacking is a thing.

Our scale search machine got a major update! Now with 31% more scales, 21% more items, 58% more coverage of measures by use, and 100% more goth!

Applications are open for the Besample Dissertation Grant 2026–2027! 🌍 🎓 Besample will again fund 15 best multi-country exploratory studies (N=500), then 3 main awards (N=3,000). Open to dissertation-level Ph.D. students in social and behavioral sciences: grants.besample.app?utm_source=f...

Thinking LLMs can replace RAs, PhD students, & post Docs shows you think their job is to do your grunt work & learn through osmosis. Student/early career researchers should be trained, not used a means to enhance your career. If you think you can replace your trainees you aren't training them.

Great advice here on retaining copyrights and sharing materials when publishing!

any statistician born after 1983 can't hand-calculate sums of squares... all they know is be Bayesian, eat hot chip & calculate marginal effects

this whole "professors indoctrinate our youth" nonsense needs to stop. That line of argument infantilizes grown adults. The reality is that if we faculty could successfully indoctrinate students, they'd all know the syllabus for each course chapter and verse. Good luck with that, I always say.

Where does everyone store their Open Educational Materials when OSF storage is no longer available? #EduSky #OpenScience crowd - any pointers?

Day 2 of #BTSCON2026 is underway! It’s never to late to join your big team science friends in a zoom room near you 🥰 bigteamscienceconference.github.io

#BTSCON2026 starts soon! We can’t wait to celebrate another year of big team science with all of you! 🥰 @psysciacc.bsky.social @manybabies.org @manyprimates.bsky.social @manydogsproject.bsky.social @manylanguagesc.bsky.social @cos.io @forrt.bsky.social @abrir.bsky.social @reproducibilitea.org

Hi Accelerators, The October 2026 newsletter is now live! Read about PSA highlights, opportunities for our members, and project updates: psysciacc.org/news/2026-10... - Sara Eftekhari, PSA Newsletter Coordinator

Why social sciences are hard(er): 1) We can’t break down problems into small experiments 2) Many ?'s about humans already solved, so we're left with the harder ones I can think of others reasons but an interesting take! From Mohseni, Herrmann, Orona www.sciencedirect.com/science/arti... #phdsky

Agreed. Science needs an International Memorandum of Understanding to protect access and ensure collaboration. Researchers are not their governments. osf.io/preprints/ps...

I dearly hope "shaping the program" doesn't mean: "assigning speakers & not accepting proposals for talks". Metascience can be such a great conference, I wouldn't want to see it turn into a conference where only the friends of the organizers & local hosts are really the ones who get speaking slots

I’ll be presenting next week @btscon.bsky.social on integrating undergraduate students into big team science projects in an advanced course. Come check out my presentation on Tuesday 10/6 at 2pm Eastern!

I have this memory that either JASP or JAMOVI or something like that found the computation of (if I remember) Bayesian RMANOVA or similar were incorrect. Meaning, if you ran those analysis in that program, it was wrong. Anyone else remember this? I can't find evidence this happened. #stats #meta

So many conferences for international societies just so happen to be in Canada this summer. I wonder why that could be. Maybe a different country is acting foolish regarding people from other countries?

An agent completes a survey, this is not ground breaking news, yet its led to another set of 'online testing is dead' posts. Remember capability ≠ deployment. Just because it can be done doesn't mean it is being done Also, mouse tracking catches it easily. Design studies with this in mind.

Come work on a PhD in Metascience with me! There’s just a little over one week left to apply!

Someone sent me this meme about peer reviewer comments and it is 👌🏻 #PhDSky #AcademicSky

Let's say that pDoom is actually high. Seems that that means we should be trying to flood the internet with BS and AI slop at an even more rapid rate, so that model collapse might occur prior to model world takeover.

Do small effects really accumulate to make big effects? Not really. And certainly not when there are other causes on the thing you are looking at. Which, you know, is always. From Berggren & @jkflake.bsky.social journals.sagepub.com/doi/pdf/10.1... #psych

Now in press! Kids These Days!...aren't really changing that much; at least, not on child behavior problems since 1980s. Some improvements, some decrements, but In > 418,000 kids from nationally-representative samples, no real change. from Zsofia Takacs & Boglarka Vekety doi.org/10.1007/s105...

As you get more experienced in publishing research on all sides it's amusing how much easier it is to spot sections or even just sentences in a paper you can tell were put there because of a reviewer.

Congratulations to the project “The Advantage of Big Team Science: Lessons Learned From Cognitive Science,” which received🏆2026 SIPS commendation! It offers a synthesis of strategies for successfully navigating the BTS landscape, including logistical, ethical, and epistemological challenges.

Rakow’s new paper found that “around 90% of articles that discuss replication and over 80% of articles with author-identified replications are missed by searching databases for replicat* because databases hold few details per article”. 8/

Maybe about academia, maybe really about the decline of Talenti

Reading a study right now that's talking about dual processing and includes an experimental treatment*demographic moderation and all I can think is "oh, that's so 2010" . Sure enough, paper was published in 2010.

Just a reminder that posting code/materials on your personal website is not a long term solution and not compliant with most sharing standards. I've run into two papers today from the same lab that uses this practice, all the links are broken. Will wait for replies to emails I guess....

We’re hiring! Wisdom & Culture Lab @uwaterloo.ca seeks a Research Project Manager for a 12+ country project on cultural narratives, economic games & decision-making, via Wise Judgment Consortium. Please apply/share! 👇 uwaterloo.ca/wisdom-and-c... #AcademicJobs #ResearchJobs #CulturalPsychology

Far too many interventions never find out how long the effects last. In this new pre-print we explain you should always try & find out how long intervention effects last. Complete with practical how to advice & warnings of unique difficulties. In short, always follow-up. doi.org/10.31234/osf...

Skills are replaced between generations, not "lost" scienceblog.com/g-basic-life...

We use correlation (r) a lot. We use it for test-retest reliability Prediction/memory accuracy Trait stability, But we should not. In a new simulation study, we show r is upwardly biased & doesn't tell you what you want. We show you what does (Table 2). osf.io/preprints/ps... #psych #stats

The bottom line here is that Prolific hosts very different kind of studies, but I worry that they are all being binned together in a way that is misleading. As a researcher, though, it feels important to me to be aware of what I am actually competing against on the platform.

There has been a lot of helpful resources on new places to share data and code with the closing of osf. But what about preregistering? Since you can't upload files (like code) to osf post 2/27 it seems we need a new prereg option. What are people planning on moving to?

Should we still use the term "preregistration"? Our take here: osf.io/preprints/ps... tl;dr: we think yes! With @lspitzer.bsky.social, @lukaswallrich.bsky.social, and @josefinaw.bsky.social

Gotta love when Zotero sees a paper with dozens of authors and thinks: "well, this is a nice web page with no authors and is definitely not a journal article" so you have to input the dozens of authors by hand...

I think the OSF project infrastructure contributed more than anything else to increasing transparency in psychology research and we'll have to fight to keep up the same standards!

The following R script downloads your public projects from the OSF. Set the variable "me" to your OSF user ID, which you can find at osf.io/profile after logging in.

What do researchers believe the replication rate of their field is? What do they think is an acceptable %? Lots of debate but not a lot of data; so we asked! 1,826 researchers in 11 fields said they thought ≈56% of studies in their field are replicable, <≈ 60% is a problem. osf.io/preprints/ps...

Not-a-spoiler-alert: you never see demand effects in online general population studies

Meta Research Symposium 2026, 8-9 October 2026, TU/e Eindhoven paulmeehlschool.github.io/workshops/sy... Seems like a wonderful event for ECRs doing meta-science. I'll give a workshop on "How to use AI tools, agents, LLMs, and automation to improve open science practices" Hope to see you there.

Hard to beat list right here

Figma doesn’t mean anything but Sigma is annoying right now and I went to college during the Ligma era.