SKM 2023 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 18: Social Systems, Opinion and Group Dynamics I
SOE 18.4: Talk
Thursday, March 30, 2023, 16:15–16:30, ZEU 260
Analysing the structure of opinion spaces — •Eckehard Olbrich1 and Sven Banisch2 — 1Max Planck Institute for Mathematics in the Sciences — 2Karlsruhe Institute for Technology
Data for understanding opinion dynamics arise in a variety of contexts: from voting patterns and multi-item surveys to hashtags use and users participation in different online groups. Despite their different origins they have a similar mathematical structure: a matrix with rows representing members of a population and columns representing e.g. items of a survey or political issues. Recently various novel methods have been proposed to make sense of correlational patterns in such data in order to identify them with cultural schemata [3] or dimensions of an underlying political space [1]. In this contribution we will compare different such methods including (1) issue bundles [1], (2) latent space models [2] and (3) correlational class analysis [3]. We will discuss their advantages and disadvantages using data sets from a survey on attitudes towards Corona measures and from Swiss public votes.
[1] E. Olbrich, and S. Banisch, The rise of populism and the reconfiguration of the German political space, Frontiers in Big Data 4, 731349 (2021).
[2] F. Gaisbauer, A. Pournaki, S. Banisch, and E. Olbrich, Grounding force-directed network layouts with latent space models. arXiv:2110.11772 (2021).
[3] A. Boutyline, Improving the measurement of shared cultural schemas with correlational class analysis: Theory and method. Sociological Science 4.15 (2017): 353-393.