Events

Abstract: I will present a book project that aims to provide a comprehensive overview of geographic voting patterns in Germany, based on neighbourhood-level data. Such fine-grained data capture people’s everyday experiences and conversation networks much better than, for example, data at the regional- or county-level.

As guiding heuristic to understanding variation in local voting patterns, I present a cluster and typology approach. In a multiparty system like Germany, local voting patterns are complex, and single-variable information, such as which party wins the most votes, gives only limited insight into people’s voting behaviour. I use voting data from the German federal elections up to 2021 at the level of the ca. 100,000 on-site and postal ballot districts (‘neighbourhoods’) and use Latent Class Analysis to identify four types of neighbourhoods based on each party’s local vote share.

I will then analyse the types of neighbourhoods along three dimensions: (1) politically left vs. right, (2) segregated vs. integrated/representative, and (3) homogeneous vs. diverse. The typology aligns remarkably well with geography, in particular the urban-rural continuum, East-West differences, and centre-periphery patterns within major cities. Overall, the analyses aim to provide a thorough and accessible examination of Germany's political geography and social structure through the lens of voting behaviour.

The Sociology Seminar Series for Trinity Term is convened by Juliana de Castro Galvao, Pablo Geraldo and David Kretschmer.  For more information about this or any of the seminars in the series, please contact sociology.secretary@nuffield.ox.ac.uk.