Nicole Schwitter

A banner advert for the University of Warwick, which says "Beyond ignores curiousity". This is a spelling error.

The University of Warwick has paid a lot of money, I'm guessing, for their internet banner ads on the Guardian and elsewhere.

But they can't spell "curiosity". This isn't a US-UK variant. They just, genuinely, have a massive spelling error in their banner ads.

🚨 @haukelicht.bsky.social and I are hiring! Join the GAPED project! 👇
Postdoc (Computational Text Analysis)
🔗 www.uibk.ac.at/de/politikwi...
Predoc (Survey Experiments)
🔗 www.mzes.uni-mannheim.de/en/news/deta...
🗓 Deadline: Aug 22, 2025
Please share widely! #PoliSky

🚀 New postdoc position for Platform Data & Computational Social Science
Join us to drive @gesis.org' efforts around the Digital Services Act (DSA) and conduct research with online platform data
💼 TV-L EG 14 | Location: Cologne
📌 Apply now: www.gesis.org/en/institute...
#DSA #CSS #DataScience

🎥 All keynote talks from IC²S² ‘25 are now available on our YouTube channel!
#ic2s2 #css

This will get taken up as a problem with the model (it gave a bad answer), when it really should be taken up as a problem with the use (people shouldn't turn to LLMs for this kind of question, though I'm sure some do).

Sometimes, a descriptive figure is worth more than a thousand model coefficients: the importance of data description in social research Many social research questions aim at understanding relationships between different phenomena, and increasingly complex multivariate statistical models are often employed to address these questions...

Finally got this banger of a title out: 𝑆𝑜𝑚𝑒𝑡𝑖𝑚𝑒𝑠, 𝑎 𝑑𝑒𝑠𝑐𝑟𝑖𝑝𝑡𝑖𝑣𝑒 𝑓𝑖𝑔𝑢𝑟𝑒 𝑖𝑠 𝑤𝑜𝑟𝑡ℎ 𝑚𝑜𝑟𝑒 𝑡ℎ𝑎𝑛 𝑎 𝑡ℎ𝑜𝑢𝑠𝑎𝑛𝑑 𝑚𝑜𝑑𝑒𝑙 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡𝑠.
If your descriptive figures and your model tell different stories, trust the figure. (And maybe rethink the model.)
👉 📄 www.tandfonline.com/doi/full/10....

Thinking about using #CSS methods to study #racism, #stereotypes or #hate speech in text? 📐

👉 Check out my first dissertation paper co-authored by @fabiennelind.bsky.social and @hajoboo.bsky.social just published in Annals of the ICA! @icahdq.bsky.social 🥳

🔗 doi.org/10.1093/annc...

Quite a strong final statement: "Descriptive research is important and it is a pity that the general obsession with causal estimates disincentivizes researchers from attempting to publish careful and detailed description."

Hendrik Erz PhD Student at the IAS, Linköping University (Sweden)

+++ ANNOUNCEMENT +++

I'm currently on the job market, looking for Postdoc opportunities (end of 2025/early 2026).

If you are — or know someone who is — in need of a political sociologist/computational social scientist, please get in touch or share my CV!

More info: www.hendrik-erz.de

„KI verändert schon jetzt, wie wir denken"

Ein Experiment mit jungen Forschern an der Universität Mannheim zeigt, wie tiefgreifend ChatGPT & Co das wissenschaftliche Schreiben verändern – und dass das noch längst nicht alles ist.

Im Wiarda-Blog: www.jmwiarda.de/blog/2025/06...

🧠🚀 What happens when early-career social scientists use ChatGPT for academic writing? Clearer and more coherent text - but no boost to depth, originality, rigour, or literature use, according to our field experiment.
📄👇

Call for papers: Generative AI and Social Science Research

journals.sagepub.com/topic/collec...

Thanks for the shoutout!
Our review of 69 experimental studies on disability discrimination in hiring shows:
✳️ Big gaps in callback rates
✳️ Some conditions get all the attention
✳️ We need more theory + intersectionality

Congratulations again 🥳

Norrköping campus at sunset

We are hiring postdocs in Computational Social Science
📍SweCSS, Norrköping, Sweden
⏰Deadline June 3
🔗https://liu.se/en/work-at-liu/vacancies/26854
Please apply // help us spread the word

International Roundtable on Computational Social Science: Using AI-Generated Images in Vignette Experiments Välkommen till International Roundtable on Computational Social Science med Nicole Schwitter från University of Mannheim. Seminarium är öppet för allmänheten och hålls på engelska. Kontakta madelene.t...

Join us on Thursday, 15 May, at 14.30 CET for the International Roundtable on Computational Social Science with @nschwitter.bsky.social 🔹Using AI-Generated Images in Vignette Experiments 🔹More info: liu.se/en/event/202...

Anomalies & data problems are found in the Hungarian EU SILC income data from around year 2017. It is very likely that the data was manipulated to decrease the poverty rate.

Annamaria Tatrai and Andras Gabos conducted a very thorough investigation of the issue: www.valaszonline.hu/2025/04/01/e...

Thrilled our global data ecosystem audit was accepted to #ICLR2025!

Empirically, it shows:

1️⃣ Soaring synthetic text data: ~10M tokens (pre-2018) to 100B+ (2024).

2️⃣ YouTube is now 70%+ of speech/video data but could block third-party collection.

3️⃣ <0.2% of data from Africa/South America.

1/

text table

Here's an update on sociology (and some demography) journal editorial polices with regard to sharing preprints before submission and after acceptance. A big improvement from the last edition of this table. With one exception.

Details of the publication with abstract which reads: Honor killings, which occur when women are perceived to have broken purity norms and bring “dishonor” to their family, pose profound moral and societal problems and underrecognized sociological puzzles. Given the immense cost, why do families murder their own daughter, niece, or cousin? Conversely, given the tragic consequences, why are norms broken in the first place? Drawing on accounts of honor killings, we characterize the key actors, actions, and incentives, and develop two interlinked theoretical models, one on norm-enforcement and another on norm-breaking. The former specifies the conditions under which honor norms should hold, the latter, counterintuitively, predicts that honor killings occur most frequently when honor norms are contested; not when they are strictest. Analyzing data from 24 countries and ~26,000 individuals and building a unique dataset of honor killings from Turkey, we find support for the hypotheses. Honor norms are stronger when laws offer leniency for honor killings, families’ loss of reputation is more consequential, and community cohesion is higher. Actual killings have an inverse-U-shaped link with the prevalence of honor norms. Our work advances the theoretical understanding of honor norms and killings and offers one of the most comprehensive empirical analyses of the factors influencing honor killings.

Figure 1 from the paper. Within-Family Interaction and Probabilities of Interaction Outcomes Note: Panel A shows an interaction within a family. Panel B shows the probability of breaking a norm (p2) in the game tree in panel A; the probability of the existence of an honor norm in the repeated norm-enforcement game (p1, i.e., conditions given in Hypotheses 1, 2, and 3 obtain); and the probability of honor killing (p3), which is the product of the former two probabilities, as a function of repeated norm-enforcement game parameters.

Figure 3 from the paper. Support for Honor Killings; Coefficients from Multilevel Regressions with Random Intercepts for Countries Data source: PEW (2013) World Muslims Survey. Note: N (response) = 26,458. N (country) = 24. Countries that adopted laws that allow leniency for honor-based violence: Algeria, Egypt, Iraq, Jordan, Lebanon, Morocco, Niger, Palestinian Territories. Countries that have adultery laws: Afghanistan, Pakistan.

Figure 5 from the paper. Results of Poisson Models Predicting the Monthly Number of Honor Killings Note: Predictive margins for the association between support for honor norms and honor killings (top panel) and the average marginal effect of passing law 6284 according to support for honor norms (bottom panel).

📢 My paper w Aron Szekely on honour-based violence is published online (Open Access): journals.sagepub.com/doi/epub/10....

We explain this puzzling phenomenon applying game-theory and testing it using large-scale data on norms and actual femicides.

@sriucl.bsky.social @uclsociology.bsky.social

The Wikimedia Foundation, which owns Wikipedia, says its bandwidth costs have gone up 50% since Jan 2024 — a rise they attribute to AI crawlers.

AI companies are killing the open web by stealing visitors from the sources of information and making them pay for the privilege

1/ Excited to share our new preprint “A Collective Field Experiment Disentangling Participation in Online Political Discussions” – osf.io/preprints/so...

At this point, it is obvious that it is not safe for scholars to travel the US. There are no easy solutions for conferences scheduled in the US this year but it should be cleary to everyone involved that they cannot just go ahead as planned.

The flyer shows the overview of the programme for the Text-as-Data Spring Term Speaker Series 2025, which usually takes place on Wednesdays at 5pm Berlin time.

The #TaDa Speaker Series is back for the spring 🎉 We're looking forward to an exciting line-up of talks by @prashantgarg.bsky.social, @miriamschirmer.bsky.social, @chdausgaard.bsky.social, @taniseceron.bsky.social, @lukashetzer.bsky.social, and Catarina Pereira! More infos at tada.cool & on Slack ⬇️

2025-03-05 | Input talk | Nicole Schwitter (University of Mannheim) Digital Data, Contextual Complexities, and Lessons for Social Scientists more Hybrid event [A5, 6, Room A231 + Zoom] March 05, 2025, 13:45-15:15 Abstract With increasing digitalisation, data is everywhere - (seemingly) waiting to be harvested, by social scientists, from historic archives and search trends to Facebook friendships and open-source collaborations. Yet beneath this public surface lies the usual complexity of data generation, raising important questions about whether the observable actions represent meaningful concepts. This talk will serve as a cautionary reminder for social scientist to scrutinise their data, investigate and describe the data-generating context and community, and to (also) apply formal theory when measuring constructs. I will be drawing on several well-published examples and highlight their shortcomings to underscore these points and to suggest how we can – and need to – do better. Presenter(s) Nicole Schwitter is a postdoctoral researcher at the MZES and holds a PhD in Sociology from the University of Warwick, where she studied offline interactions in online communities using Wikipedia as a case study. Her research explores (inter)group relations and social dynamics, with a strong interest in applying novel computational methods to social science questions.

🚨 Upcoming: "Digital Data, Contextual Complexities, and Lessons for Social Scientists"

👤 @nschwitter.bsky.social (University of Mannheim)

🗓️ Wed, March 5, 13:45-15:15 CET

📺 Register for the live stream: us02web.zoom.us/meeting/regi...

🔗 socialsciencedatalab.mzes.uni-mannheim.de/page/events/

Registration for our conference "Computational Social Science:
AI and Society – Exploring Inequality in the Digital Age" now open!

🗓 May 15-16th, 2025 (University of Mannheim/MZES/MCDS)
👉 Esp. for young scholars interested in CSS and/or inequality
🔍 computational-social-science.org/workshops/20...

Want to learn about computational social science *for free* and launch interdisciplinary research projects? We are so excited to announce there will be *26* Summer Institutes in Computational Social Science this year! Apply to one of them here: sicss.io/locations

Logo for Wiki Workshop 2025 mentioning it will take place virtually on May 21 and 22, 2025.

📣 The Call for Papers for the 12th edition of Wiki Workshop is out.

Submit your 2-page extended abstract of your research about Wikimedia projects by March 9! All submissions are non-archival (ongoing, completed, already published works are welcome).

More info: meta.wikimedia.org/wiki/Wiki_Wo...

Can men & women be friends?

A phenomenal paper analyses 1.3 trillion Facebook connections!!!

open.substack.com/pub/dralicee...

At #DataFest Germany, bachelor’s and master’s students form teams to analyze an extensive dataset and compete for exciting prizes and recognition in four categories:

🏆Best Insights
🏆Best Visualization
🏆Best Use of Outside Data
🏆Best Data Quality Insights

Stay tuned for this year’s winners!