Marek Vranka

Understanding Maximum Likelihood Estimation, now in 3D!: rpsychologist.com/likelihood #dataviz #stats #rstats

ok I guess?

CrossRef, a major gateway from journal articles to scholarly database, implemented CRediT back in July, but what we hadn't noticed was that the blogpost that introduces it mentions that CRediT has been "made available in multiple languages", which I'm pretty sure is a reference to our work!

Glad to see our article in latest @ejisbisa.bsky.social issue. @smetanamichal.bsky.social, @laurensukin.bsky.social, @mvranka.bsky.social & I use survey experiments to show how public opinion can sometimes influence US & UK elite views on nuclear weapon use. #OpenAccess: doi.org/10.1017/eis....

Why is Psych Science doing a bunch of special issues? We want to get more submissions on these topics! Authors may not submit papers on these topics because they don’t see many examples in our journal. Hopefully special issues can help! www.psychologicalscience.org/publications...

Two more retractions and a discussion of what we learned. 🧵 #CRAVEFiles

so, I've always suspected my wife has a lot of tabs open and was actually a bit disappointed that when I installed an extension that counts them on her firefox, it said she >only< had like 393... turns out, the counter was not able to display more than three digits and she actually has 3930+🥲

This document is going to be an assigned reading in college classes that cover this moment in time, there's a lot happening in a few paragraphs... www.ahmath.org/statements

as one of those interested but unable to join, I am very excited (and happy!) about this announcement! (so much I am even willing to undertake the traumatizing experience of opening ms teams)

I would still prefer OSF or a relevant clinical trial registry for study registration. Study plans or protocols shared in other repositories are nice, but they are not "study registrations." Registrations need to be in a formal registry, following Box 2 here: osf.io/preprints/me...

"They 'trust me.' Dumb fucks." - same energy...

no need for AI to be better at peer review than humans, it could complement human reviewers especially in parts humans suck at / are too lazy to do thouroughly - like checking reporting, stats, preregistration deviations etc. let's leave the "not original enough" judgment for human Rev 2...

New post and game by me! Some people claim that AI will cure all diseases in 10 years. Others are more skeptical, arguing that experimentation and clinical trials are the bottleneck. Who's right? I designed a game to find out.

With the recent mainstreaming of AI concerns, I'm very disappointed that we haven't heard more about supporting proof of personhood tools and techniques, which some have worked on for more than a decade. news.mit.edu/2024/3-quest...

I usually don't love the central role Harvard has to play in all our higher ed convos, but I think Anastasia Berg is right to point out that this is a grotesque abdication of the responsibility they so proudly accept as a representative of the liberal arts. www.chronicle.com/article/harv...

I used to love this time of year, hearing from students applying to work in my lab. No more. Every day a fresh batch of nearly identical emails with nonsensical remixes of the text on my lab website. It makes me so sad.

Machine learning and generative LLMs are subject to many of the same drawbacks and issues as general empirical research in the typical publishing system. In this piece now published at NHB we argue that, like for empirical research, ML and LLMs can benefit from the use of Registered Reports.

It is an error. We asked them to correct it. I hope they do soon!

Trump’s Golden Dome scam is all gold and no dome. The program gave contractors $3.2 billion to develop space interceptors that we already know can’t possibly work. There are several excellent studies out there showing why. Here’s a new one I strongly recommend. thebulletin.org/2026/09/defe...

this amazing #metasci event (again!) somehow flew under my radar - paulmeehlschool.github.io/workshops/sy... - although submissions are closed, registration is still open, so check it out! (I will squeez my very limited budget and try joining as well!) @youngmetascience.bsky.social

Modern academic structures are constraining scientific creativity A new paper says research is structured around short funding cycles, productivity-based evaluation criteria, and risk-averse frameworks that favor predictable and non-transformative research outputs. www.nature.com/articles/s41...

this is really cool replication idea!

Great news! Some of us have been waiting a long time for major funders to not just shovel $ into the maws of the big publishers, and instead support the long tail of scholar-controlled journals that are run on a shoestring

🚨 New preprint: In this paper we make the case for measuring social media use at the level of naturally occurring sessions (i.e., beginning when the app/site is opened and ending when it is closed). Across 7 days and 161 participants this gave us nearly 4,000 paired sessions of use 🧵

I am afraid this is the reason why fraud is actually much more widespread - lots of cheats simply fly under the radar, avoiding scrutiny completely...

The number of people doing science opportunistically is great. In 2010, these people p-hacked. In 2026, they do crap preregistrations. Just like we figured out ways to prevent p-hacking, we are finding ways to prevent opportunistic preregistration. The solution is always the same: transparancy.

Slightly-updated verson of tenzing.club now deployed, let us know of any issues. Thanks to floze-the-genius for chipping in and adding the enhancement! github.com/marton-balaz...

I got my hands on a new #socPsy textbook by Hayes & Bradshaw, and of course first thing I checked was how they handle QRPs and replication crisis. and boy, was I disappointed - lots of hand-waiving, downplaying and even a completely made-up term "PHarking" (likely a mash-up of p-hacking and HARKing)

two years ago, Ariely was invited to an event in Czech Republic where he promoted his new book about misinformation, lies, and trust in science... absolutely unbelievable!

apropo, has anyone checked data for Ariely's masturbation paper or that one with large stakes in India? there is no way in hell those are not made up as well... maybe @steamtraen.eu knows?

you can't put into crisis a field that is still in the original crisis 😘

An influential paper by Ariely & Wertenbroch (2002) reported two studies with tampered data. Post 1 of 2. datacolada.org/138

What are some examples of *non-fraudulent* major errors in scientific papers that had major consequences, perhaps after taking a long time to be corrected?

this is an important clarification for everyone who was worried about how changes at OSF would affect uploading files during preregistration:

The Center for Open Science recently announced its decision to phase out OSF Projects. We want to share how those changes will affect PsyArXiv, a psychology preprint server run and maintained by SIPS volunteers but hosted on the OSF platform. improvingpsych.org/2026/08/11/w...

this is Arina Bones levels of prescience! maybe we all really knew the future all along!? ... but kudos for clearly stating the limitations - it seems that predicting actual behavior in the field is much more tricky!

we started to work on this eons ago (like 2023?) and a lots happened in the meantime, but I believe the main principles we identified are still relevant even today...

🔔🔔 Russian polls have shown ~75% support for the war in Ukraine since February 2022, essentially flat. New working paper with Kirill Chmel and @nikitasavin.bsky.social shows that number is stable partly because it stopped measuring what people actually think. 🧵 ssrn.com/abstract=711...

The need for this also a major finding of our qualitative work on reproducibility journals.plos.org/plosone/arti...

sounds like a dream job!

our new survey exp methods preprint comparing video/text+images/long text/short text vignettes on emotion elicitation, policy support, manipulation check success and survey enjoyment brings some clear answers (ppl like videos and short text elicits weaker emotions), as well as some new puzzles!

Nice replication with Norwegian twin data of @pamherd.bsky.social et al.'s polygenic-score-based result that heritability of education increased for women over time with increasing gender equality. www.nature.com/articles/s41... by @olerogeberg.bsky.social @kph3k.bsky.social @torkildl.bsky.social

WE ARE HIRING! Come work with us (me + @malte.the100.ci) on developing and validating RegCheck. PhD or postdoc applications welcome. Folk with backgrounds in meta-science, psychology, (pre)clinical trials, and/or economics are especially invited to apply. jobs.unibe.ch/job-vacancie...

Looking for a #SAS user who is interested in computational reproducibility to join a many-analysts-style paper. Run some #stats in SAS, get authorship, contribute to the world of reproducibility. If this is you or you know someone DM or email me (protzko@gmail) #phdsky

almost silly how after all those years of talking about DPTs, we only now finally test for mechanisms of how S2 actually spots & corrects errors of S1... before that, it was simply:

📢NEW RESOURCE: A new web resource and accompanying report to explore the vulnerabilities to political interference of 33 key UK scientific public bodies & 11 scientific advisory committees! unchecked.uk/public_body_... 1/7

⚡ Metaventory: building better science ⚡ Had many discussions about the need for mapping out all that's meta-science: people, tools, events, jobs, simulations, etc. in one place. Metaventory aims to help connect those interested in meta-science & improving science. metaventory.vercel.app

Integrative experiments - where effects are tested across auxiliary assumptions - are a more useful tool than multiverse analysis and many analyst studies. If you like to ponder the philosophical basis of statistical procedures, you might enjoy this paper! osf.io/preprints/ps...