Go as we are: the neighbourhood and its political effects

Thesis

We might think our local identity is secondary to our other geographic identities, like our national identity. Yet, this is not the case. Many voters consider it as important, or more so, than these higher-order identities. This thesis argues that the `neighbourhood' is one of our foremost geographic identities, that it is a social identity, and that its effects are more widespread than previously understood. Many studies of place-based identity ignore lower-order identities. Those which do not are often not generalisable because they use limited samples or study only one type of identity, such as rurality. Others measure identity using non-survey-based measures, and political science generally focuses only on radical right voting. Over seven chapters, this thesis models neighbourhood identity and its effects on three key areas of political behaviour: participation, affective polarisation, and candidate choice. Firstly, it shows that those factors which influence it are those which most affect how we form social relationships on the neighbourhood level: social homogeneity, physical distance and time. It finds that it predicts some forms of participation, though mostly low-cost behaviours, such as voting. It shows that, while evidence for the effect of geographic sorting on polarisation is mixed, this is because it is, in part, jointly moderated by voters' place-based identity and contact with their in-group. Finally, while many studies have suggested that voters choose candidates who display descriptive localism because of their own place-based identity, this work, for the first time, demonstrates causally that they do. These effects are generally modest, however. To do this, it innovatively primes place-based identity and combines this with a forced-choice conjoint. This thesis brings together diverse data, including novel surveys, panel studies and micro-level voting figures across Britain, Europe and the US, modelling on small geographic units, to provide more credible causal evidence. It raises important questions about how we understand place-based identity, the units and the data we use, and the political effects we study.

Authors: Ward, AJ

• DOI: 10.5287/ora-zvpzgj85n
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: place-based identity; public opinion; elections
• Subjects: Political sociology; Political psychology