[Air-L] Online Seminar Series of the Data Methods Initiative

[Marcel Garz]

The Data Methods Initiative (https://datamethodsinitiative.org/) is a collaborative academic group committed to promoting the use of computational methods (e.g., deep learning, natural language processing, computer vision) in political science, communications, internet research, and other disciplines.
Next talk: "Linking individual, meso, and macro data: A relational comparative approach to survey-based research" by Qinfeng Zhu (University of Groningen) on November 5, 1:15 - 2:45 pm CET, on Zoom: https://ju-se.zoom.us/j/67448948642
In this research seminar, Qinfeng will present a relational comparative framework that infers structural patterns inductively by analyzing individual-level survey data, which can be linked to meso-level data (e.g., media outlets) and macro-level data (e.g., country characteristics). The talk will showcase how social network analysis is used to examine audience overlap between alternative and mainstream media outlets, by linking survey data on media consumption with content analysis of the outlets. Additionally, Qinfeng will demonstrate how non-metric multidimensional scaling can be applied to participants' thermometer ratings of political party supporters, to identify different dimensions of affective polarization and explore how these are influenced by media and political systems. The seminar will highlight how these computational methods can be applied to traditional survey data, offering valuable insights into the relationship between individual behavior, media systems, and political contexts.