Damien Bol
banner
damienbol.bsky.social
Damien Bol
@damienbol.bsky.social

Prof in Political Behavior in Sciences Po Paris. Ex boom-bap rapper, I wear white kicks ALL THE TIME

www.damienbol.eu

Political science 60%
Economics 14%
Pinned
🚨 New paper🚨 on how AI can help us doing research (or not) with my office mate @phbono.bsky.social

We test whether LLMs (ChatGPT, MistralAI) can be used to ideologically position parties in real time.

journals.sagepub.com/doi/10.1177/...

1/3
Can ChatGPT accurately identify the position of parties? A validation study with an expert survey in France - Damien Bol, Pierre-Henri Bono, 2025
We evaluate whether ChatGPT can be used to estimate the ideological positions of parties in real time. Compared to other methods, ChatGPT can be used at all tim...
journals.sagepub.com

Take-aways: (i) support for violence against politicians is low (phew), (ii) largely unrelated to affective polarization, and (iii) we can reduce it with simple and scalable interventions.

A paper that gives us some hope

End/

We've also a vignette experiment where we test whether a simple intervention can decrease support for violence: encouraging people to consider politicians' perspective and reminding them about the importance of pluralism. It does for the most pervasive form of violence

5/

Conjoint shows that support is lower when the politician is a woman (fits stereotypical views), but partisanship don't. Negative correlation disappears when controlling for individual variables. The profile of violent citizens are low-educated young men who dislike all parties

4/

Similar to other studies using these types of survey questions, we find that people don't support the most extreme forms of violence (physical assault), but they do support "milder" forms like online insult. Though, overall support is rather low, no more than 10% support.

3/

We rely on latest recommendations (@seanjwestwood.bsky.social and colleagues) and survey citizen support for concrete violence forms against (hypothetical) politicians using a conjoint in France (very high polarization, think about US) and french-speaking Belgium (much much lower).

2/
New WP on political violence in democracies with the fantastic @dianebolet.bsky.social and @bjarneck.bsky.social. Sadly very topical, but with some positive results

osf.io/preprints/so...

1/
The Political Behaviour Stream of the MSc in Political Science @lsegovernment.bsky.social is welcoming applications for entry in 2026/27: www.lse.ac.uk/study-at-lse... Benefit from teaching and supervision from world-leading scholars, weekly research seminars & an active alumni network:
MSc Political Science (Political Behaviour)
Study political opinions and behaviours with the MSc Political Science (Political Behaviour), offering rigorous training and diverse career paths.
www.lse.ac.uk
🚨 New LSE job in political science 🚨

We're looking for a *Full or Associate Professor of Political Science and Public Policy* to join the LSE School of Public Policy

Please share!

jobs.lse.ac.uk/Vacancies/W/...
Full or Associate Professor of Political Science and Public Policy
Full or Associate Professor of Political Science and Public Policy, , <p style="text-align: center;"><em><span>LSE is committed to building a diverse, equitable and truly inclusive university</span></...
jobs.lse.ac.uk
First-time authors: Get the reviewer on your side in the first 3 pages. From there on, they either look for areas to improve or reasons to reject.

Reposted by Damien Bol

Le CEVIPOF lance une collection de #notes dédiées à l'enquête #FracturesFrançaises, pour comprendre les transformations profondes du paysage politique et social FR.

NOTE2 : Anne Muxel : La #génération est-elle encore un marqueur des fractures politiques françaises?
www.sciencespo.fr/cevipof/site...
🚨 New paper in @thejop.bsky.social

Why do politicians often misperceive what citizens' policy positions are?

@simonotjes.bsky.social and I study ~10,000 estimates of public opinion by politicians in Denmark & the Netherlands to uncover the sources of these (mis)perceptions

Thread 🧵1/10
🚨 new paper out w/ @simonchauchard.bsky.social in BJPS @bjpols.bsky.social on the link between misinformation and religiosity- does religious belief cause people to endorse more misinformation? And if so, what can we do about it?

thread below ⬇️
New WP thread! 🤓 Vietnam's solar panel exports boomed between 2018 and 2022. With my PhD student Meng Yu Ngov, Trang Thu Tran, and Gaurav Nayyar we look at how this industry success can be attributed to attracting FDI and importing parts and components, as well to Chinese subsidy spillovers.👇

Reposted by Damien Bol

Thanks for the SO. We tried estimate the effect of having a 4th candidate (as a separate RDD), but there were too few treated districts. That said, as France is becoming increasingly fragmented, there will be more and more data points
A recent paper by @damienbol.bsky.social and @riaivandic.bsky.social provides some inspiration: in France's two-round system, the top two parties go through to the second round, but so do any others that pass a given threshold (currently receiving votes from 12.5% of registered voters)
Ria Ivandic (@riaivandic.bsky.social)
AP @ University of Zagreb. Associate at @cep-lse.bsky.social. Working on research in domestic abuse, gender economics and political economy. She/her. https://sites.google.com/site/riaivandic/
riaivandic.bsky.social

Reposted by Damien Bol

A recent paper by @damienbol.bsky.social and @riaivandic.bsky.social provides some inspiration: in France's two-round system, the top two parties go through to the second round, but so do any others that pass a given threshold (currently receiving votes from 12.5% of registered voters)
Ria Ivandic (@riaivandic.bsky.social)
AP @ University of Zagreb. Associate at @cep-lse.bsky.social. Working on research in domestic abuse, gender economics and political economy. She/her. https://sites.google.com/site/riaivandic/
riaivandic.bsky.social

Et ça s’insère dans le débat général sur : est-ce que chaque recherche (entendu chaque papier) doit proposer une nouvelle théorie et la tester? C’est intenable compte tenu des attentes en termes de publications et pas souhaitable car les theories doivent être testées et restées pour être validées
📣 We are hiring! 👇

This PhD position is part of a research project that examines how growing income inequality can reshape the way democracy functions by influencing both political discourse and public opinion.

More info here:
www.academictransfer.com/en/jobs/3542...

Il y aurait probablement moins de vote utile avec le jugement majoritaire ou le vote par approbation. Mais il faudrait les voir à l’œuvre dans le temps.

Dans les pays utilisant du ranked-choice par exemple, les partis finissent souvent par donner des consignes de vote stratégique à leur supporters.

A noter qu'aucun mode de scrutin n'est immunisé face au vote stratégique, même la proportionnelle. Voir par exemple ici : journals.sagepub.com/doi/10.1177/...

ou là : journals.sagepub.com/doi/10.1177/...
Sage Journals: Discover world-class research
Subscription and open access journals from Sage, the world's leading independent academic publisher.
journals.sagepub.com

Le vote stratégique existe donc bel et bien en 🇫🇷, même au 1er tour. Cela soulève la question démocratique des inégalités de représentation.

A lire ici : shs.cairn.info/revue-revue-...

@cevipof.bsky.social @edr-sciencespo.bsky.social

Fin/
Le vote stratégique dans une France multipolarisée
Le cas des élections législatives de 2024
shs.cairn.info

Sans surprise, les électeurs stratégiques restent proche de leur parti préféré (exemple : Reconquête ➡️ RN)

Fait marquant : au 1er tour, le vote stratégique est surtout pratiqué par les électeurs plus âgés et politisés. Cela contribue donc à un déficit de représentation pour les autres.

4/
Avec les enquêtes CEVIPOF, nous quantifions le vote stratégique aux législatives 2024.

👉 Il y en a plus au 2nd tour, quand le choix est restreint.
👉 Mais même au 1er tour, 7-8% ont voté utile plutôt que pour leur parti préféré.

C’est comparable aux scrutins à un tour britanniques ou canadiens.

3/

On entend souvent que le scrutin à deux tours rend le vote stratégique anodin en 🇫🇷 : on vote avec son cœur au 1er tour, avec sa tête au 2nd. Mais cela suppose seulement 2 grands blocs. Ce n’est plus le cas. Résultat : même au 1er tour, voter utile est crucial (cf. élimination de Jospin en 2002).

2/
Très heureux de voir notre article publié dans la Revue Française de Science Politique, en collaboration avec @jfdaoust.bsky.social, toujours aussi brillant ✨

Après avoir beaucoup travaillé sur le vote utile, aussi appelé vote stratégique, nous testons nos méthodes sur le cas français.

1/
#polisky #psjminfo We're hiring for three faculty positions in the Department of Political Science at the University of South Carolina. All 3 positions are part of an internal departmental cluster hire in the thematic area of Problems of Democracy. A 🧵on the positions. 1/
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

🧵
Join our team! 🚨

We are hiring 2 postdocs for the new Advertising Democracy project at Aarhus University.

We are looking for researchers who are passionate about advertisements & citizens’ democratic values in Western democracies.

Feel free to share!👇🏾

bss.au.dk/en/about-aar...
Postdoctoral positions in Political Science at Aarhus University – Advertising Democracy - Vacancy at Aarhus University
Vacancy at Department of Political Science, Aarhus University
bss.au.dk

ejpr.onlinelibrary.wiley.com

Kicking off the @sciencespo.bsky.social @ox.ac.uk @iast.fr workshop in political economy. This year at beautiful Nuffield College in Oxford