Computational Social Science Lab at the University of Pennsylvania
Philadelphia, PA
- Tweets 347
- Following 61
- Followers 612
- Likes 236
👩💻🧑💻👨💻 Want to apply LLMs to text analysis in your research?
Join Timothy Dörr (@Timothy95Doerr) and Baird Howland tomorrow for a hands-on workshop! 💻
They’ll share their work on how news media describe violence in international conflict, then guide participants through using LLMs for content analysis and text extraction.
No prior computational experience needed. Food will be served!
📅 Sept. 9, 2–5 PM
📍 ASC 300 @AnnenbergPenn
💻 Bring a laptop and Google account! No installation needed!
2
5
263
What happens when scientific claims go beyond what a study can establish?
In new research published in Nature Human Behaviour, Calvin Isch, Timothy Dörr (@Timothy95Doerr), Neil Fasching, Grace Jennings, and Duncan Watts (@duncanjwatts) find causal claims in 46% of 194,631 cross-sectional studies and show that causal wording makes readers more likely to infer causation.
They also find that asking LLMs to simplify research or discuss its practical implications can increase causal overclaiming 🧵⬇️
nature.com/articles/s41562-0…
1
9
17
2,126
The broader lesson is that improving science is not only about collecting better data. It is also about making sure the claims built on that data accurately reflect the evidence underneath them, and that those limits are preserved when research is communicated and summarized
1
31
📖 Read the open-access paper to explore the full findings nature.com/articles/s41562-0…
29
Congratulations, Dr. Samar Haider! 🎓🖥️
@samarhdr
Samar successfully defended his dissertation, "A Computational Approach to News Production", studying how computational methods can help us systematically understand what news organizations choose to cover and how they cover it.
1
2
17
9,592
At the CSS Lab, Samar has also been a leading member of our Media Bias Detector mediabiasdetector.seas.upenn…
1
4
323
Measures like "stops🛑" are building blocks for studying human mobility. But how reliably can we identify them in sparse GPS data?🛰️📍
At the CSS Showcase, Paco Barreras, Andrés Mondragón, and Caroline Chen shared work on the robustness of stop detection in sparse location data
1
3
5
514
The results show that different approaches can achieve similar accuracy when tuned to individual trajectories, but vary in robustness across a dataset. Smaller spatial thresholds can split or miss stops, while larger ones can merge neighboring stops
1
41
This work is part of the broader NOMAD effort to expand access to human mobility data and develop methods, tools, and guidance for analyzing it reliably!!
🛰️Learn more about NOMAD nomad.seas.upenn.edu/
9
