Despite reports of the ‘near death of the (Christian) religion’ following the 2021 Census for England and Wales, almost two thirds of British adults still identify with a religious community, participate in religious activity, and/or hold religious beliefs, and an emergent literature documents its continuing influence on political attitudes and voting behaviour in Britain (Kolpinskaya forthcoming; Kolpinskaya and Fox 2019; 2021; Clements 2015; Tilley 2015; McAndrew 2020). Data limitations nonetheless continue to hinder this field, with data beyond religious identification not usually collected in surveys and no data available to study the effects of contextual religion (i.e., the religious features of one’s community) on social behaviour.

This project will combine individual level data from Understanding Society, the British Election Study and the ESRC/JSPS Covid19 Study with aggregate and spatial data from the 2021 Census, places of worship directories and school censuses to produce the first multi-level dataset of religion in Britain. While we will use this new data to study the relationship between religion and voting behaviour in the 2024 general election, we expect the dataset to be a valuable resource for future social research. This presentation will outline our strategy for combining the individual and community level data, the detail of validation checks and analytical models, and present illustrative results, for feedback to improve the production of the new dataset and our analysis.

Co-authors

Dr Ekaterina Kolspinakay; Professor Daniel Stevens, University of Exeter;

Dr Steven Pickering, University of Amsterdam, NL;