iiinova — where your digital self keeps evolving

Reposted by Paul Meiners

Stefanie Walter @stefwalter.bsky.social · 02/10/2026

Have you ever wondered about what we know about the broad patterns, the causes, and the consequences of contestation of IOs? We – @jonastallberg.bsky.social, Michael Zürn and I – did and therefore set out to take stock of the state of the art. Our review is out now in @iojournal.bsky.social.

cambridge.org

Contestation of International Organizations: Patterns, Drivers, and Responses | International Organization | Cambridge Core

Contestation of International Organizations: Patterns, Drivers, and Responses

12412

Reposted by Paul Meiners

Sabrina J. Mayer @sabrinajmayer.bsky.social · 29/09/2026

Es gibt soviel Forschung und dabei noch mehr Halbwissen -- ein studienbasierter Überblick mit möglichen Stellschrauben gegen die Erfolge radikal-rechter Parteien.

0154

Paul Meiners @pmeiners.bsky.social · 07/09/2026

Yeah, I feel like participation was a much more central topic in the 2000-2010s.

011

Paul Meiners @pmeiners.bsky.social · 02/09/2026

absolutely not being serious here

020

Paul Meiners @pmeiners.bsky.social · 02/09/2026

Just ask an LLM, ezpz problem solved

210

Reposted by Paul Meiners

D. Hicks @danhicks.bsky.social · 23/08/2026

This is an incomplete understanding of academic freedom, and partly confuses academic freedom with freedom of speech. (🧵 adapting some things from this paper, esp. §5: philsci-archive.pitt.edu/28499/. Highly recommend the papers by Dea and Kronfeldner that I cite there!)

philsci-archive.pitt.edu

412141

Reposted by Paul Meiners

vincentholst.bsky.social @vincentholst.bsky.social · 13/08/2026

In 2023, @nature.com published 'Papers and patents are becoming less disruptive over time', receiving world-wide media attention. Our Matters Arising, published after a 32 month delay (more on that soon), shows that the reported decline can largely be attributed to dataset artefacts. 🧵

The average CD_5 index per year for Web of Science. The original Park et al. decline (top curve) becomes essentially flat (bottom curve) when removing papers with CD_5=1. Those papers largely correspond to dataset artefacts.

9352173

Reposted by Paul Meiners

Danbischof @danbischof.bsky.social · 05/08/2026

📄 New working paper out: osf.io/preprints/so... We (@thiloalbers.bsky.social @felixkersting.bsky.social and Fabian Kosse) try to understand why some citizens who are objectively well off still feel left behind, and why this matters for populist attitudes.

24818

Reposted by Paul Meiners

jon ben-menachem @jbenmenachem.com · 05/08/2026

Frankly, I don’t think frontier models should be used for research at all. Must be an “archivable” model (open weight or open source) that will be available indefinitely for reanalysis.

3172

Reposted by Paul Meiners

Tim Bale @timbale.bsky.social · 03/08/2026

This by @jamesrdennison.bsky.social and @andrewpgeddes.bsky.social looks brilliant - and very timely! global.oup.com/academic/pro...

New book - cover - 'What Europeans Think About Immigration and Why It Matters'

13110

Reposted by Paul Meiners

Dan Silverman @dmsilverman.bsky.social · 28/07/2026

Surveys like this always alarm people Fortunately, recent research from cognitive psychology shows that the reason people are so wildly wrong when estimating the size of minority groups is largely NOT because of prejudice — but rather how humans think about unknown proportions of any kind

111543

Reposted by Paul Meiners

Adam Berinsky @adamberinsky.bsky.social · 29/06/2026

Boy this was a long road, but glad it's finally out! @dgrand.bsky.social @robbwiller.bsky.social @arechar.bsky.social (along with Michael N. Stagnaro, James N. Druckman, not tagable)

05524

Reposted by Paul Meiners

Andrea Vik @andreavik.bsky.social · 11/04/2026

New publication! Unelected representation may influence how satisfied people are with democracy, opening up important discussions on the role activists play in shaping the legitimacy of liberal representative democracies. You can find out more below 👇

052

Reposted by Paul Meiners

Claes de Vreese @claesdevreese.bsky.social · 11/04/2026

🛑 Chatbots are not fit for giving voting advice 🗳️ Latest study from @liberties.eu on the Hungarian election shows lack of transparency and reliability in the advice. Our research shows that 10-15% by now use chatbots for advice, for young voters up to one quarter. www.liberties.eu/en/stories/h...

liberties.eu

Should AI Tell You Who to Vote For?

Open Ai's ChatGPT and Google's Gemini do not provide reliable political guidance on the 2026 Hungarian elections, our latest research finds.

32212

Paul Meiners @pmeiners.bsky.social · 07/04/2026

Maybe he’s only talking about post-stratification?

010

Reposted by Paul Meiners

Charles Logan @charleswlogan.bsky.social · 01/04/2026

My university: "Copilot is Microsoft's AI-powered productivity service that uses large language models (LLMs) to help you create content, analyze information, summarize documents, and complete tasks more efficiently." Microsoft: LOL

231489570

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(9/9) Although many Europeans want direct elections of the European Commission, the lead candidate process is unlikely to improve legitimacy perceptions. Without attachment to and engagement with procedural reforms, they might fall flat.

000

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(8/9) Again, we find no evidence for strong backlash against overturning the lead candidate system. Even when fully informed and after adding another procedural reform, respondents do not react strongly.

Treatment effects for both main outcomes. No large differences can be found

100

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(7/9) As another procedural reform, we add a hypothetical primary in which respondents could choose their own candidate. Then, we varied whether member state governments would refuse to accept the candidate chosen by the newly elected European Parliament, similar to the events in 2019.

100

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(6/9) In order to differentiate between these explanations, we ran a survey experiment right before the European elections in 2024. We showed respondents different scenarios how their vote might matter.

100

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(5/9) Looking at survey data collected right before and right after, we find no evidence of a strong backlash. There are three possible explanations: 1. Europeans were not sufficiently informed. 2. The process is not important to Europeans. 3. Europeans were never attached to the candidates.

A: Marginal treatment effects on satisfaction with the process of the EC president selection, 95% confidence intervals. B: Marginal treatment effects on satisfaction with the way democracy works in the EU, 95% confidence intervals. We find no meaningful effects.

110

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(4/9) We argue that this came as a surprise to European voters. The sudden nomination of von der Leyen is our "unexpected event" that allows us to test how attached citizens are to the process. If citizens cared a lot, we would expect to see a negative reaction.

We examined Google search interest and Wikipedia page views for von der Leyen around the time of the 2019 European elections. There was a clear spike in attention when she was nominated.

100

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(3/9) We first take a look at the failure of the process after the European elections in 2019. Back then, von der Leyen was chosen by the EU member states to be their candidate for the head of the commission, even though she was never in the running during the electoral campaigns.

100

Paul Meiners @pmeiners.bsky.social · 30/03/2026

(2/9) The key idea here is that a lead candidate system increases the alignment between citizens' democratic expectations and institutional reality. However, if these procedures are not followed through, they could have a negative effect on legitimacy perceptions.

100

Paul Meiners @pmeiners.bsky.social · 30/03/2026

Judging from what I've read, this is already the new norm.

000

Paul Meiners @pmeiners.bsky.social · 29/03/2026

(4/4) There isn't much data on attitudes toward aggressive or retributive economic measures (especially in Europe). It would be interesting to see how the situation develops as China uses its economic weight more and more.

100

Paul Meiners @pmeiners.bsky.social · 29/03/2026

(3/4) In a survey experiment, we tried a few framing alternatives, but they did not significantly affect the overall results.

Average treatment effects and 95% confidence intervals from a 2x2 factorial experiment. Only the high cost treatment has a (small) effect on support for trade sanctions against China.

110

Paul Meiners @pmeiners.bsky.social · 29/03/2026

(2/4) Despite ongoing geopolitical tensions between the US and China, Europeans may not perceive themselves as being on one "side" or the other. Even those who want to increase trade with the US do not necessarily support restrictions on trade with China.

Bar chart showing support for trade restrictions against China against support for trade with China and support for trade with the US.

100

Reposted by Paul Meiners

Matt Blackwell @mattblackwell.bsky.social · 26/03/2026

mariozechner.at/posts/2026-0...

Text from linked article “There's a much more important difference between clanker and hu-man. A human is a bottleneck. A human cannot shit out 20,000 lines of code in a few hours. Even if the human creates such booboos at high frequency, there's only so many booboos the human can introduce in a codebase per day. The booboos will compound at a very slow rate.
Usually, if the booboo pain gets too big, the human, who hates pain, will spend some time fixing up the booboos. Or the human gets fired and someone else fixes up the booboos. So the pain goes away.
With an orchestrated army of agents, there is no bottleneck, no human pain. These tiny little harmless booboos suddenly compound at a rate that's unsustainable. You have removed yourself from the loop, so you don't even know that all the innocent booboos have formed a monster of a codebase. You only feel the pain when it's too late.

0257

Paul Meiners @pmeiners.bsky.social · 16/03/2026

Great overview. I assume exchangeability is important here, right (wasn't mentioned explicitly in the post)?

110

Reposted by Paul Meiners

Nathaniel Haines @natehaines.bsky.social · 15/03/2026

New blog just dropped! This one is all about estimators—we cover James-Stein, classical test theory, empirical Bayes, penalized regression, and hierarchical models, showing how they all can be used to do a better job than sample stats alone 🤓 haines-lab.com/post/how-to-...

haines-lab.com

How to Estimate a Mean, and What It Means for Science | Computational Psychology

Introduction If I asked you to estimate 30 means, you would probably compute 30 sample means. And you would be provably wrong (well.. maybe not wrong, but at least provably inefficient).

56524

Paul Meiners @pmeiners.bsky.social · 14/03/2026

I mean, if you take away the CIs, it’s just low-tech EDA. Loop over all bivariate regressions, find the biggest R2 difference between linear and squared model.

010

Reposted by Paul Meiners

Jason Koebler @jasonkoebler.bsky.social · 12/03/2026

I met with AI data labelers in Kenya who are organizing their colleagues to fight the brutal working conditions and horrible pay given to the workers at the "bottom of the AI supply chain." They believe the NDAs they've signed are unenforceable so are speaking out: www.404media.co/ai-is-africa...

404media.co

'AI Is African Intelligence': The Workers Who Train AI Are Fighting Back

Kenyan workers are still the underpaid labor behind AI training, moderation, and sex chatbots. The Data Labelers Association is fighting back.

3563193016

Reposted by Paul Meiners

Per Engzell @pengzell.bsky.social · 10/03/2026

Strictly speaking, a postdoc is just a doc who posts

67212

Reposted by Paul Meiners

Olli Bäckström @ollibackstrom.bsky.social · 02/03/2026

I'm not gonna lie: Macron and his generals singing the Marseillaise to a nuclear weapon is a bit chilling.

1022943

Paul Meiners @pmeiners.bsky.social · 23/02/2026

This could be an inspiration: arxiv.org/abs/2505.035...

Partially directed SWIG assumed to generate data for a cannonical 2x2 DID.

011

Reposted by Paul Meiners

Grant McDermott @gmcd.bsky.social · 20/02/2026

I think the main issue is that many people, quite reasonably tbf, don't like the default base data.frame print method... But this is easy to override! gist.github.com/grantmcdermo...

A nicer print.data.frame method showing column types, as well as a subset of rows. Inspired by data.table's print method.

0195

Reposted by Paul Meiners

Kevin Elliott @kjephd.bsky.social · 17/02/2026

Arendt suggests Eichmann is a clown *rather than* a monster. But why not both—a clown AND a monster? It seems to me a key part of the horror of fascism is precisely its pervasive clownishness. There's a mind-rending indignity in having to take seriously rulers who are fundamentally unserious.

813829

Reposted by Paul Meiners

Dr Stylite @columnist.bsky.social · 12/02/2026

The art of politics is not to do what people think. If it was, there literally wouldn’t be any politics. The art of politics is to convert public opinion in the direction of your policies. Not so subordinate your policies to opinion.

22788138

Reposted by Paul Meiners

Cyrus Samii @cdsamii.bsky.social · 03/12/2025

A blog post giving a more thorough take on survey experiments and the credibility revolution: cyrussamii.com?p=4168

cyrussamii.com

Survey experiments and the credibility revolution – Cyrus Samii

67737

Reposted by Paul Meiners

Alexander Wuttke @kunkakom.bsky.social · 14/11/2025

The billions of research funds channeled into the pockets of Elsevier and other commercial publisher in return for very little actual value is one of academia's big inefficiencies that will eventually be replaced by more attractive Community-run alternatives

05119

Reposted by Paul Meiners

Python Software Foundation @python.org · 27/10/2025

TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-... 🧵

python.org

The official home of the Python Programming Language

12363772727

Reposted by Paul Meiners

Matt Graham @mattgraham.bsky.social · 27/10/2025

After a huge post-election flip in economic perceptions, I thought Democrats and Republicans might be lying to pollsters to send a partisan message — but I was wrong! New in the Journal of Experimental Political Science (open access): doi.org/10.1017/XPS....

39241

Paul Meiners @pmeiners.bsky.social · 21/10/2025

Don't worry, surely some statistical prediction machine will easily replace them. You just need an intern to speak english to a computer, right?

010

Reposted by Paul Meiners

Sven Hegewald @svenhegewald.bsky.social · 21/10/2025

Super happy to see this out in @jeppjournal.bsky.social 🎉🎉

1144