Clint Claessen

🔊 🔊 🚨 🚨

Last Friday (Sept. 19), @andduer.bsky.social, @mcortinaescudero.bsky.social, @yavuzmehmet.bsky.social and I had the great pleasure of hosting an excellent workshop at @unisalzburg.bsky.social on how to use Large Language Models #LLMs to boost International Relations research. A short 🧵1/4

Our new study in
@jwpp.bsky.social
shows that online abuse on social media discourages young women from pursuing political offices—before their actual political careers have even begun.

with @dahoehmann.bsky.social and Tomoko Latteier

www.tandfonline.com/doi/epdf/10....

We present our new preprint titled "Large Language Model Hacking: Quantifying the Hidden Risks of Using LLMs for Text Annotation". We quantify LLM hacking risk through systematic replication of 37 diverse computational social science annotation tasks. For these tasks, we use a combined set of 2,361 realistic hypotheses that researchers might test using these annotations. Then, we collect 13 million LLM annotations across plausible LLM configurations. These annotations feed into 1.4 million regressions testing the hypotheses. For a hypothesis with no true effect (ground truth $p > 0.05$), different LLM configurations yield conflicting conclusions. Checkmarks indicate correct statistical conclusions matching ground truth; crosses indicate LLM hacking -- incorrect conclusions due to annotation errors. Across all experiments, LLM hacking occurs in 31-50\% of cases even with highly capable models. Since minor configuration changes can flip scientific conclusions, from correct to incorrect, LLM hacking can be exploited to present anything as statistically significant.

🚨 New paper alert 🚨 Using LLMs as data annotators, you can produce any scientific result you want. We call this **LLM Hacking**.

Paper: arxiv.org/pdf/2509.08825

Mooi dat de SRV podcast terug is! @hakhverdian.bsky.social stukroodvlees.nl/aflevering-1...

Excited to announce the publication of my article in Comparative Political Studies on how crises shape the ideological legitimation strategies of authoritarian regimes. journals.sagepub.com/doi/full/10....

Peilingen zijn nuttige instrumenten voor kiezers, politici, en media. Ze kunnen wezenlijke effecten hebben, vooral via de wijze waarop media er - juist én onjuist - over rapporteren.
Ook in de campagneperiode hoop ik dat journalisten en duiders de peilingen gebruiken waarvoor ze geschikt zijn:

Plaatjes van de electoraatjes 2025: NSC - StukRoodVlees Op 29 oktober zijn er verkiezingen voor de Tweede Kamer. In deze serie gaan wij aan de hand van een aantal grafieken in op de vraag wie de kiezers van de vijftien grootste partijen volgens de peilinge...

Veel zijn het er niet meer. Maar wat kenmerkt de kiezer van NSC? @mrooduijn.bsky.social start onze serie de plaatjes van de electoraatjes 2025 stukroodvlees.nl/plaatjes-van...

Vacancy — PhD Position on Improving Social Media Using Large Language Models The Institute for Logic, Language and Computation (ILLC) at the University of Amsterdam is inviting applications for a fully funded PhD position in the NWO VIDI project "Improving Social Media Using L...

🚨 PhD Position at the University of Amsterdam 🚨

Join my team as a computer scientist / computational social scientist working on LLMs, social media, and politics.

We offer freedom, impact, and an inspiring environment at one of Europe's leading universities.

🔗 werkenbij.uva.nl/en/vacancies...

BJPolS abstract discussing the dynamics of mainstream and radical political rhetoric regarding anti-immigrant policies, highlighting differences in societal responses and enforcement norms based on the source of the rhetoric.

NEW -

How Mainstream Politicians Erode Norms - cup.org/4lfeHvD

"we find that statements by mainstream politicians lead to more norm erosion than similar statements by radical-right politicians"

- @valentimvicente.bsky.social, Elias Dinas & @dziblatt.bsky.social

#OpenAccess

A New Season for Respect the Marble Introducing "Etched in Marble: Conversations with Writers" & "Sunday Writer's Digest"

My Substack on writing, Respect the Marble is back! 🎉

I’ll be launching several new things this autumn, outlined here: catherineeunicedevries.substack.com/p/a-new-seas...

A quick preview 👇

🧵

In poli sci, "the top 20% of departments produced 75% of all faculty and the bottom 50% accounted for less than 5% of all tenure track faculty members at a research university. 49 programs did not have a single graduate placed in a TT position at a PhD-granting department in the past 10 years."

🚨Check out our new Preprint & give Sean a follow! :)👇

Opinie: Een direct gekozen minister-president blaast het parlementaire stelsel op Nederland mist een manier om kiezers invloed te geven op het belangrijkste politieke ambt. Maar een direct gekozen premier is niet de oplossing, betoogt politicoloog Tom van der Meer.

Ook het voorstel van Jort Kelder voor een direct gekozen premier is helemaal niet verstandig. Het zou ons parlementaire stelsel opblazen.
www.volkskrant.nl/columns-opin...
Wie kiezers echt een stem op de macht wil geven, kiest liever voor effectievere en constructievere middelen.

EPSA have announced that they will hold a conference in July 2026.

😵‍💫 We understand that there might be some confusion about EPSS and EPSA.

👉🏽 So we thought we would clarify some things.

A short 🧵

2024 was a turbulent year in German politics: four elections, a surging far right, and the coalition's collapse. Our new Political Data Yearbook piece (with @lucykinski.bsky.social) traces the events leading to the 2025 snap elections.

@hhu.de @sceus.bsky.social

Wow. A group of top scholars at Harvard just sent a letter to its president, Alan Garber, warning against surrendering to Trump.

Signatories include Steven Levitsky, Dani Rodrik, Ryan Enos, Theda Skocpol, and Steven Walt.

Someone forwarded it to me. Read it here:

The European Political Science Society is now accepting paper & panel proposals for its annual conference!

📢 Call for Papers: EPSS 2026 – Belfast

🗓️ June 18–20, 2026

📍 ICC Belfast

📬 Deadline: Nov 7, 2025

🧵

#MHAWS: Mirya Holman's Aggressive Winning Scholars Newsletter | Substack Surviving and thriving in academia. Click to read #MHAWS: Mirya Holman's Aggressive Winning Scholars Newsletter, a Substack publication with thousands of subscribers.

Just discovered this gem on writing retreats by the brilliant @mirya.bsky.social

One question she addresses is: how I can make my normal life like a writing retreat?

More of this please!

If you're into smart reflections on writing, give this a read. 👇🏽

miryaholman.substack.com?utm_source=d...

Fake headline in NYT font: Full Transcript: Mamdani Told Waitress 'You Too' After She Said 'Enjoy Your Meal'

This site is gold:

mamdanitimes.com

Until another party do us part? Party members’ electoral disloyalty in Portugal This article investigates whether and why party members sometimes vote for other parties. Although often taken as parties’ most loyal voters, recent evidence suggests that members’ vote is not unco...

Our paper has now been assigned a volume and issue number at JEPOP. Marco Lisi and I examine how spatial proximity, intra-party democracy, and party activism explain why some party members—often seen as their party’s most loyal voters—defect at the ballot box.

www.tandfonline.com/eprint/CMGCM...

A graphic featuring a red text "POLITICAL ANALYSIS" on a yellow background with the hashtag "OpenAccess" at the bottom.

#OpenAccess from the latest issue of @polanalysis.bsky.social -

Positioning Political Texts with Large Language Models by Asking and Averaging - cup.org/4leSKhe

- @glemens.bsky.social & @ainagallego.bsky.social

#LLM

How do populists define “the people”? @tiondelisa.bsky.social examines how right-wing populists construct ingroup identity through opposition. Read OPEN ACCESS: buff.ly/GEvXWhB

@polstudiesassoc.bsky.social @uoypolitics.bsky.social @sagepub.com #polsky #polsci #ECRs

Selecting and ranking female candidates under PR: Evidence from a two‐stage conjoint experiment with party elites Does gender affect candidate selection and list placement under proportional representation (PR)? Existing research argues that PR systems have a positive effect on women's representation due to a mo...

Now out with volume number!
“Selecting and Ranking Female Candidates under PR” (with Michael Jankowski @cerpintaxt.bsky.social) in European Journal of Political Research (doi.org/10.1111/1475...)

Intra-party Competition, Geographic Responsiveness and Incumbent Deselection in Closed-List PR | The Journal of Politics: Vol 0, No ja

Another piece on candidate selection of mine is forthcoming in @thejop.bsky.social : Intra-party Competition, Geographic Responsiveness and Incumbent Deselection in Closed-List PR” (doi.org/10.1086/734244)

Reminder: Great job opportunities at Aarhus University!

Please help spread the word before everyone disappears for summer break! 🏖️

Book cover of the second edition of Foundations of European Politics A Comparative Approach

📢 🇪🇺📘 Exciting news! The second edition of our textbook "Foundations of European Politics: A Comparative Approach" - with @catherinedevries.bsky.social @sarahobolt.bsky.social and @jbslapin.bsky.social - is out at Oxford UP! 🧵👇 1/5

Yet another great post I can only recommend reading! This is a perfect application of growth mindset. It’s very hard to not regard yourself as a ‘static’ binary, but instead see your skill power as something dynamic and optimize what Catherine calls the ‘stretch’. Reminded me of this picture

🚨 Big News for European Political Science 🚨

We’re thrilled to announce the launch of the European Political Science Society (EPSS): a new, member-led, not-for-profit association built to support our scholarly community.

🔗 epssnet.org

Here’s a thread with everything you need to know.

🧵

(2/3)On Friday 27 June (13:10 - 14:50, Room 1A.11) I present joint work with @ta-huikuri.bsky.social on Women in International negotiations as part of the panel 'Gender Representation: rhetoric and parliamentary speeches' chaired by @clint0475.bsky.social