I am not anti LLM across the board. But if your use case starts with the necessary assumption that the model is a vessel of all human knowledge, then you are consulting an oracle not doing scientific research.

Regular reminder to use R Projects and the `here` package. Future you will be happy tidyverse.org/blog/2017/12...

1. Transparency is necessary for credibility 2. Transparency is hard to change 3. Require transparency* 4. Transparency is not magic 5. Journals are part of problem 6. Expect more from journals 7. Peer review is not magic 8. A crisis can look a lot like „normal“ science 9. Meta-analysis is not magic

The most precious commodity you have is your attention. You don’t have to waste it on poor-faith debates or arguments with strangers if you don’t think they’ll be productive. You can prioritize the things that matter to you and make your life richer.

Are you using mediation analysis? Take a look at this cool paper!

🚀 1 week to go! Dive into the world of computational social science at ODISSEI Conference 2025 – workshops, demos & research projects that spark innovation. ⚡🧠 📍 Utrecht, 4 Nov → odissei-data.nl/event/odisse...

Science!

Open Science Workshop - detecting errors and misconduct in science Dates: 20-22 October Location: Radboud University, Nijmegen, The Netherlands Registration link: tinyurl.com/3xkd8twf ONLY PHYSICAL ATTENDENCE POSSIBLE! NO ONLINE POSSIBILITIES.

Really interesting online event on Aug 27, 2025 08:00AM (Amsterdam time).

Teaching is done, course evaluation survey responses are in. "Clear communication, great content. I acquired not only academic knowledge but also 𝘂𝘀𝗲𝗳𝘂𝗹 𝘀𝗸𝗶𝗹𝗹𝘀 𝗶𝗻𝗰𝗹𝘂𝗱𝗶𝗻𝗴 𝗴𝗶𝘁𝗵𝘂𝗯."

Do you want to go to a metascience conference where AI will not solve all problems, you can ask critical questions, and El$evier is not invited? Sign up for the mailing list for the 8th Perspectives on Scientific Error conference in 2026 in Leiden! perspectivesonscientificerror2026.wordpress.com

I finally got to adding the slides from the SIPS workshop to the OSF project. The GIF below would be an accurate depiction of me in the past days, but only if money were replaced by emails and to-do lists.

Source: Figure 1 Rohrer 2024 @dingdingpeng.the100.ci (compass.onlinelibrary.wiley.com/doi/10.1111/...) The blue line of wishful thinking is my own edit.

I don't care about claims for the link between causal inference knowledge and academic career success, but I sure hope that the arrow from lack of single-minded focus is as causal as it gets. It has to be, right? It's just a tiny arrow, there's nothing in between.

when trump and co say they're seizing control of research because of the reproducibility crisis, it's pretty similar to when they say they're enacting trans bathroom bans and sports bans because of violence and discrimination against women.

I think in part the higher education funding crisis and the LLM-teaching crisis actually strike at the same problem, which is that education doesn't work without a broad, society-wide sense that the *actual content* of education - not the credential - is valuable.

Data Colada has submitted a motion for sanction against Francesca Gino and her former lawyer team. I hope the court rules in their favor. It will be a powerful message that sleuths can expose fake research without fearing legal consequences. storage.courtlistener.com/recap/gov.us...

I’ve seen situations in which people try to correct the scientific record and then critics act like that was superfluous or just not the right way to use one’s energy because “science is self-correcting.” This just came to my mind

Do future-you a favor and start using version control now!

Papers published in Psych Science following January 1st, 2024 are candidates for I4R to computationally reproduce, stress-test and replicate. This is an official collaboration with @psychscience.bsky.social. journals.sagepub.com/doi/10.1177/... We are looking for reproducers and replicators!

Stop the budget cuts in education in the Netherlands! #rotterdam

#FAIRS who else is attending Fostering Accountability for the Integrity of Research Studies (FAIRS)?

Interested in how we can tackle the scourge of research fraud? And meeting people in the delightful city of Oxford? Join us at the FAIRS meeting in April - in-person and online options. More details and booking form: www.sjcfairsmeeting.com (turn off VPN if it doesn't load)

Our latest paper on #Neuroforecasting is out at @pnasnexus.org (w/ Brian Knutson and Lester Tong) A short 🧵 tinyurl.com/e4279exp

I'll be at Social and Behavioural Data Science Center at the University of Amsterdam for a couple of weeks. I'm giving a lecture on Monday Feb 24th that anyone is invited to and hope to meet up with some people while I'm in town to talk methods and measurement. sites.google.com/view/mellenb...

Aaaaaaaa!

Modern-Day Oracles or Bullshit Machines? Jevin West (@jevinwest.bsky.social) and I have spent the last eight months developing the course on large language models (LLMs) that we think every college freshman needs to take. thebullshitmachines.com

On my way! Really looking forward to the next few days Check this out: errorsin.science Double check your code! Make your future self proud 💪

Asking PIs about their stats is like

One of the weirder scholarly practices regarding generative AI that seems to have been normalized is citing chatbots. I say normalized because many univs & scholarly associations recommend it as an element of proper scholarship. But it doesn't make sense when you consider what a citation means. 1/

Hi, can you help me? I want to develop a model that makes risk predictions. Use logistic regression. Can I use some more modern techniques, like AI? Use a neural network with single non-hidden feed forward layer that outputs to a single dimension using a sigmoid activation function.

🚨 Calling everyone in or near the Netherlands who cares about the future of higher education 🚨 📣 Protest 📅 Thursday 🕐 13:00 📍 Utrecht 🟥 linktr.ee/woinactie

Larry is here!

Just because your code executes without error doesn't mean there's no mistake in it. Do yourself a favor and write code thinking that another human will have to read it and understand it. "another human" includes you in X months from now xkcd.com/1695/

Don't make a habit of protesting, but this protest in #Utrecht seems worth it... English: www.tilburguniversity.edu/current/budg... Nederlands: www.tilburguniversity.edu/nl/actueel/b...

Don't give your DNA to any company. The company can disappear, your data can be sold to anyone to do anything, and it's not only your data, it's also the data of your family members. #Privacy. Thanks for the quote, @zsk.bsky.social www.bbc.co.uk/news/article...

Join the protest in Utrecht, nov 14, 13.00-15.00. www.aob.nl/actueel/camp...

Plan and document now! Future-you will be grateful, or at least less annoyed compared to if present-you thinks "oh, this is obvious. I will remember all these details, no need to write anything down."

Really excited for this conference

Classic past-you vs present-you clash

Want to do a PhD on the neuroscience of (consumer) decision making? Come join our team! www.rsm.nl/about-rsm/wo... #neuroscience #vacancy #fmri #eeg #neuroeconomics

Interested in doing a marketing PhD? Come work with us! The marketing group at Rotterdam School of Management, Erasmus University seeks highly motivated PhD students looking to study topics in marketing (quant or CB). More info: www.rsm.nl/about-rsm/wo... #marketingsky #econsky #marketing

Find out more in the PhD Virtual Open House Day Opt1: Nov 28, 4pm-5.30pm CET OR Opt2: Nov 29, 10am-11.30 CET You will meet some of our faculty and students, hear them talk about their research, and have the chance to ask questions. Sign up here: erasmusuniversity.eu.qualtrics.com/jfe/form/SV_...

Interested in science? Do you want to do research on decision making? Check the open PhD positions in the marketing group at Rotterdam School of Management, Erasmus University - quant/CB www.rsm.nl/about-rsm/wo... - decision neuroscience www.rsm.nl/about-rsm/wo...

I strongly recommend thinking before collecting data. "More data" is not guaranteed to be useful, and it is not at worst useless. More data can lead to problems for future-you, so collect more only if you have a good reason to do so. Just because you can, doesn't mean you should get more data!