SODAS Data Discussion 1 (Spring 2026)
Copenhagen Center for Social Data Science (SODAS) aspirers to be a resource for all students and researchers at the Faculty of Social Sciences. We therefore invite researchers across the faculty to present ongoing research projects, project applications or just a loose idea that relates to the subject of social data science.
Two researchers will present their work. The rules are simple: Short research presentations of ten minutes are followed by twenty minutes of debate. No papers will be circulated beforehand, and the presentations cannot be longer than five slides.
Discussion 1
Presenter: Jason W. Burton
Title: Stated Versus Revealed Preferences in Online Content Consumption
Abstract:
People’s exposure to content online is often shaped by engagement-optimizing recommender systems that purport to show users the content they want and value. On social media platforms, for example, recommender systems typically predict which posts a user is likely to engage with (e.g., by “liking” or dwelling on a post) and then promote those posts in the user’s news feed. Central to this engagement optimization is the assumption that users’ actions reveal the kind of content they want. However, revealed preferences, inferred from users’ engagement behavior, often diverge from stated preferences users explicitly report when asked. Critics argue that by privileging revealed preference, recommender systems cater to myopic impulses rather than deliberative, long-term aspirations and wellbeing. Yet, relatively little is known about when discrepancies between stated and revealed preferences might arise in online environments and which psychological constructs, traits, and mechanisms might explain such discrepancies. In this talk, I present an experimental design and analysis plan to test whether and to what extent different psychological factors associate with discrepancies between stated and revealed preferences in content consumption within a simulated social media environment.
Discussion 2
Presenter: Moritz Johanning
Title: Under the Neighborhood: Why Your Next-Door Neighbor Matters Most
Abstract:
The neighborhood is a fundamental concept in social and geospatial data analysis. However, empirical studies diverge on what constitutes a neighborhood. In this talk, I argue for modeling neighborhoods as networks of individuals rather than bounded spatial units. Combining this definition with fine-grained Danish registry data, I present descriptive evidence that similarity between neighbors decays sharply with proximity. Similarity is highest among immediate next-door neighbors and declines sharply with neighbor rank. This micro-local clustering is not detectable in standard aggregate measures of neighborhoods. The second part of the presentation addresses the causal challenge of disentangling this micro-level sorting into peer influence and residential sorting. I outline a research design to identify these dynamics and present initial findings as well as planned next steps in this project.