iiinova — where your digital self keeps evolving

Daniel Lakens @lakens.bsky.social · 06/10/2026

😂 It is really pretty smooth nowadays, but we appreciate the trauma you are willing to suffer 🤣

Dermot Lynott @dermotlynott.bsky.social · 05/10/2026

Reposted by Daniel Lakens

Delighted to announce the Irish Reproducibility Network's latest online event . Prof. Lai Ma (UCD) will talk about "JD Bernal's Preprint Proposal at the 1948 Royal Society Scientific Information Conference" Oct 20th, 1200. Registration is free, via EventBrite www.eventbrite.ie/e/irish-repr...

eventbrite.ie

Irish Reproducibility Network - Talk Series

Join Prof. Lai Ma (University College Dublin) for Talk 3 of the IRN Talk Series

11010

Daniel Lakens @lakens.bsky.social · 05/10/2026

(Sharing this because some people on this platform are saying such dumb things about this replication study, I would strongly recommend people to listen to more informed viewpoints).

Peer Community In @peercommunityin.bsky.social · 05/10/2026

Reposted by Daniel Lakens

The latest PCI Webinar is now freely available online! Many thanks to Daniël Lakens for an insightful webinar entitled “Essential aspects of well-designed studies”. For those who couldn’t join us live, the recording is now available: peercommunityin.org/pci-webinar-...

peercommunityin.org

PCI Webinar Series - Peer Community In

The PCI webinar series is a series of seminars on research practices, publication practices, evaluation, scientific integrity, meta-research, organised by Peer Community In

041

Daniel Lakens @lakens.bsky.social · 05/10/2026

If you want to listen to a podcast about a team taking a solid and thoughtful approach to replicating a classic study, thinking carefully about translation, replicating multiple versions of the same hypothesis, etc, listen to Rolf Zwaan's "Research as it Happens" shows.acast.com/research-as-...

shows.acast.com

Research as it Happens

Listen to episodes and learn more about Research as it Happens. What does science look like while it’s still unfolding?In Research as it Happens, we go inside a large-scale psychology project, where…

1146

Reposted by Daniel Lakens

Sajedeh Rasti @sajedehra.bsky.social · 05/10/2026

We are excited to announce that we are going to broadcast the second day of Meta-Research Symposium 2026 (this Friday on October 9th) for those who were interested but unable to join in person. You can join the meeting using this link: teams.microsoft.com/l/meetup-joi...

teams.microsoft.com

Join conversation

1159

Reposted by Daniel Lakens

Judith ter Schure @judithterschure.bsky.social · 05/10/2026

Meer onderzoeksjournalisten in het wetenschappelijk debat! Wetenschappers zouden moeten opletten hoe onderzoeksjournalisten te werk gaan. Fouten meteen corrigeren. Bepaalde uitgevers de rug toe keren. Beroepseer hooghouden met focus op kritiek geven en ontvangen. neerlandistiek.nl/2026/10/geen...

neerlandistiek.nl

Geen onbevoegden in het wetenschappelijk debat

Wetenschappers vertellen elkaar graag gruwelverhalen over alle onwetenden die zich op internet zomaar met hun vak bemoeien, en die zo de wetenschap ondergraven.

121

Daniel Lakens @lakens.bsky.social · 05/10/2026

We are excited to announce that we are going to broadcast the second day of Meta-Research Symposium 2026 (this Friday on October 9th) for those who were interested but unable to join in person. JOIN: teams.microsoft.com/l/meetup-joi... PROGRAM: docs.google.com/document/u/0...

docs.google.com

Meta Research Symposium 2026

Meta Research Symposium 2026 8-9 October, 2026, TU/e Eindhoven Conference website: https://paulmeehlschool.github.io/workshops/symposium2026/ Times are in central European time. To check the time…

01710

Reposted by Daniel Lakens

Henrik Singmann @singmann.bsky.social · 02/10/2026

New preprint: When should you use linear mixed models (LMMs), and when is repeated-measures (RM) ANOVA enough? LMMs offer greater flexibility. But does that make them a better default? My paper argues for statistical parsimony: use RM-ANOVA when both methods are appropriate. osf.io/preprints/ps...

osf.io

When to Use Mixed Models and When RM-ANOVA - An Argument for Statistical Parsimony

When should researchers use linear mixed-effects models (LMMs) versus repeated-measures analysis of variance (RM-ANOVA) for analysing repeated-measures data? I first outline the conceptual and statist...

17317

Reposted by Daniel Lakens

Peer Community In @peercommunityin.bsky.social · 01/10/2026

It’s today! Join us today for a webinar with Daniël Lakens on: “Essential aspects of well-designed studies” 🕐 4 PM (GMT+1) 👉 Register now and find out more: univ-cotedazur.zoom.us/meeting/regi... More information on PCI Webinar series: peercommunityin.org/pci-webinar-...

univ-cotedazur.zoom.us

https://univ-cotedazur.zoom.us/meeting/register/IMJIOCUDSzajhO6qQFXYqQ

063

Reposted by Daniel Lakens

Matthew Russell @matteomics.bsky.social · 28/09/2026

Sometime the real potential benefit of "reproducibility" is really auditability. The possibility of laying out the data and the steps for critical appraisal. The reproducibility makes that audit possible, it is not a magic test on its own.

091

Reposted by Daniel Lakens

ingmarvisser.bsky.social @ingmarvisser.bsky.social · 28/09/2026

New paper 🍼📄 ManyBabies 3 is out (open access, registered report in Developmental Science). 30 labs, 33 languages, 839 infants set out to replicate a textbook finding: 7-month-olds learn abstract rules from speech (@garymarcus.bsky.social et al., 1999). We didn’t find it. 🧵

37847

Reposted by Daniel Lakens

Marek Vranka @mvranka.bsky.social · 26/09/2026

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...

162

Daniel Lakens @lakens.bsky.social · 27/09/2026

You should probably also send your students to my course then :)

000

Daniel Lakens @lakens.bsky.social · 27/09/2026

You can come to my master course then. It will start again next year september.

100

Daniel Lakens @lakens.bsky.social · 25/09/2026

They are indeed not science.

Daniel Lakens @lakens.bsky.social · 24/09/2026

I work in the social sciences an use the HD method. The developments I referred to are Mayo's severe testing.

101

Daniel Lakens @lakens.bsky.social · 21/09/2026

For 1 and 3, it is useful to keep in mind most scientists are incompetent, so of course. 2 has been done for at least 30 years (I would say sufficiently). 4 is probably true but those fields as a consequence has serious problems.

000

Daniel Lakens @lakens.bsky.social · 21/09/2026

What do you think about 'the scientific method' as 'procedures that will tell you that you are wrong when you are wrong, and right when you are right'? Basically, severe testing as Mayo discusses it. This is not to say that is the only procedure we use, nor that we apply it perfectly.

320

Reposted by Daniel Lakens

Peer Community In @peercommunityin.ecoevo.social.ap.brid.gy · 21/09/2026

Don’t miss our next PCI Webinar! Join us for a webinar with Daniël Lakens on: “Essential aspects of well-designed studies” 📅 October 1st | 🕐 4 PM (GMT+1) Want to learn more about the quality of science, sample size justifications and best pratices in science? Join the discussion and discover […]

ecoevo.social

Original post on ecoevo.social

012

Reposted by Daniel Lakens

Constantin Späth @cspaeth.bsky.social · 19/09/2026

Great thoughts on replications! @uyguntunc.bsky.social I’d vote for an additional educational version that is, as far as possible, jargon-free and suitable for undergraduate students. :) bsky.app/profile/phil...

152

Daniel Lakens @lakens.bsky.social · 20/09/2026

It is not a you problem but a field problem. We should not work in small groups. We should have organized ourselves into much larger collaborations. In these collaborations there are always tasks to do for undergraduates interested in doing research.

Daniel Lakens @lakens.bsky.social · 19/09/2026

I read this once because of the title. The contents explain why it has been cited so rarely.

Dan Quintana @dsquintana.bsky.social · 19/09/2026

Reposted by Daniel Lakens

New preprint! 🎉 I analysed 1660 papers from 4 psychology journals and found materials sharing went from 9% of papers in 2015 to 82% in 2025, and these materials *do* get downloaded — a median of 135 times each. BUT shared code is often hard to run. doi.org/10.31234/osf... Let's walk through it 🧵

Two-panel figure. Panel a is a flow diagram tracking 1,611 empirical psychology articles from publication year (545 in 2015, 552 in 2020, 514 in 2025) to repository-link type: 785 link an OSF project, 85 link another platform, and 741 link no repository. Of those with an OSF link, download counts were retrieved for 670 and not retrieved for 115. Panel b is a line chart of the share of empirical papers linking OSF across 2015, 2020 and 2025. The overall rate, shown as a dashed black line, rises from 9% to 57% to 82%. All four journals rise steeply and end close together: Psychological Science 92%, JESP 89%, JML 83%, Cognition 76%, with Psychological Science highest throughout.

Three-panel figure. Panel a: ridgeline plot of downloads per file by material type on a log scale, with the percentage never downloaded labelled for each — archive 16% of 545 files, documents 20% of 2,970, code 12% of 3,471, other 17% of 1,646, data 21% of 10,051, media 35% of 2,120, images 35% of 4,102. Most files cluster between 1 and 10 downloads, with long right tails past 100. Panel b: ridgeline plot of downloads per paper by journal, log scale, with dashed median lines; Psychological Science is highest, then JESP, JML and Cognition. Panel c: stacked bars showing, for documents, data and code separately, the share of papers by download band (0, 1–10, 11–100, more than 100) in 2015, 2020 and 2025. The share exceeding 100 downloads falls sharply over time in all three types, from roughly two-thirds in 2015 to a quarter or less in 2025, as the 1–10 band grows.

Four-panel figure. Panel a: statistical languages detected among 333 papers with retrievable code — R 88%, SPSS 12%, Stata 4%, SAS 1%. Panel b: code red flags among those 333 papers — 34% hard-code an absolute path, 40% reference a missing file — above documentation among 672 OSF-linked papers — 21% have a README, 30% are documented by README, description or wiki. Panel c: among 562 Elsevier papers with no repository link, 44% (245) host at least one journal supplementary file but only 14% (77) host data, code or an archive. Panel d: composition of those 448 hosted files — documents 52%, data 21%, media 9%, other 7%, archive 6%, code 3%, images 1%. The code panels are green, the journal-supplement panels blue.

Coefficient plot (download predictors)

Dot-and-whisker plot of three standardised predictors of OSF download volume, each with a 95% confidence interval. Repository size (number of files) has the largest effect at about 0.76, citations about 0.38, and altmetric attention about 0.12. All three intervals sit entirely above zero, so each predicts more downloads, with repository size roughly twice the effect of citations and six times that of attention. X-axis: standardised effect, −0.2 to 1.0.

4265113

Daniel Lakens @lakens.bsky.social · 18/09/2026

Lovely day at the start-up meeting for OpenScienceNL funded projects at Leiden. Great day for networking, and being inspired by what Dutch (meta)scientists are up to.

040

Reposted by Daniel Lakens

Jan van Gemert @jvgemert.bsky.social · 14/09/2026

Our fifth workshop on Meta-science for Machine Learning is on Sept 22. metascienceforml.github.io Registration info in the link.

metascienceforml.github.io

Metascience for Machine Learning

A repository for collecting information about Metascience for machine learning.

044

Reposted by Daniel Lakens

European Journal of Political Research @ejprjournal.bsky.social · 16/09/2026

Research is strongest 💪 when others can understand, evaluate, and build on it 🏗️ That's why our new Reproducibility Guidelines provide clear guidance on transparency, data, code, and research material requirements 🤩 Read the policy here: buff.ly/SdtAPcB

cambridge.org

Research transparency

Welcome to Cambridge Core

033

Reposted by Daniel Lakens

Johannes Breuer @johannesbreuer.com · 16/09/2026

Keynote by @lakens.bsky.social + roundtable w/ Hella von Unger & Christian Merz. 📍 @cais-research.bsky.social (Bochum, Germany) & online 🔗 Further info & registration: eveeno.com/320501318

eveeno.com

Symposium: More than compliance checks and harm prevention?

033

Exactly - metacheck will check dozens of possible mistakes (in filenames, values in datasets, column labels) and I do not think I could check these manually in medium to large datasets even if I wanted, and I would never remember to check them all)

000

But I am curious to hear what you think would solve the problem you encountered? A better design for the coding form? I don't think the explanation would be as I am dedicating months of my life to solve this exact problem.

000

that's ok. We both try our best to improve science, and we can see what worked in 2 decades. I am all in on automatic checks. I see no other solution.

220

Horizons Magazine @horizons-en.bsky.social · 16/09/2026

As human factors happens to be my expertise, I think there is no distinction between cultural and procedural. We have a system where the people who create the data you will analyze will never - never ever - have the time to learn the skill to not do this wrong. Automation is the only solution.

120

Reposted by Daniel Lakens

Both basic and applied research suffer from overpromising rhetoric, says integrity researcher and psychiatrist Joeri Tijdink. He calls for more honest communication in which #reliability, #rigour and #reproducibility are the main product of science. 👇 www.horizons-mag.ch/2026/09/03/a...

horizons-mag.ch

Admitting uncertainty is the real breakthrough

Horizons (EN) - Back to basic research

0115

The Metacheck module 'data_check' will check for this (capitals or not within the same column) and warn users to fix it. Maybe telling the people who made this to run metacheck first would prevent you a lot of time!

Oh! I did not notice (because I often use this journal as an example of a highly cited journal). That is just a glitch in the system, that is not accurate.

I have seen some of her talks, but nothing I considered worthwile. But then I have a very high bar. She is not a flat earther, but if you follow the decoding the gurus guys on X, you will see what I mean.

She is not a metascientist, but someone who makes her money from the number of people who watch her videos. I have also seen men on YouTube saying things about a flat earth. I am sure you do not identify with those people either.

It is an absolute shame (and disaster) what is happening to US science. I can only hope your next election starts the road to a better government! (my wife can vote in the US elections and will do her share)

This blog is ok, but still largely missing the point. Yes, weak theory and testing are risk factors. But the important risk factors are publication bias, low power and p-hacking. Which people widely admit they engaged in. Actually, clinical trials (testing, weak theory) have no crisis.

100

Can I ask where you are getting these ideas from? What are you reading? I am so surprised to hear - even assuming you are an American - that people think these things! If you know metascientists who say CERN is in crisis, let me know. That sounds ridiculous.

220

I have certainly not heard anyone criticize particle physics. The other way around. The Psychological Science Accelerator named itself after CERN because it is an example to emulate. Just like we learn from medicine, as it is far ahead of the social sciences.

Sorry, but who is doing this? I have never heard any metascientist argue for system level reform? Who is making up that we do this? Whoever it is, they are making things up. Very curious to see the quotes where people say this, especially in their full context.

200