Charting Westminster’s bubble: an ideological map of Britain’s digital elite

Abstract

The social networking site, Twitter (now rebranded as X), is a microblogging platform which has garnered a reputation for being a platform for the elites. Not only are its users disproportionately members of socially privileged groups – younger, wealthier, and more highly educated – relative to other social networking sites, it has become the favourited social media platform for many high profile media and political elites. This is no less the case than in the United Kingdom (UK) where many of the country’s politicians, journalists, media outlets, and political commentators have taken to the site en masse over the last decade and a half. The recent digitisation of Britain’s commentariat has presented an opportunity to trace the networks of political and media elites in the UK and exploit them to gather a better understanding of important offline phenomena. Using advanced quantitative and computational techniques, this three-paper thesis leverages original large-scale digital data from the Twitter networks
of UK political and media elites to address three key concepts: (1) intra-party
competition; (2) media representation; and (3) dyadic representation.

Generating a formally validated set of left/right ideological estimates of UK Members of Parliament (MP) and a wider set of elite accounts that follow them, this thesis strives to build an ideological map of the UK’s digital elite. Paper 1 of this thesis uses these left/right estimates to model candidate endorsement in the September 2022 Conservative Party leadership contest, confirming that Liz Truss drew support from the further right of the party. Paper 2 assesses ideological representation in the guest selection of seven flagship political programmes on the UK’s six major T.V broadcasters between 2022 and 2024, finding that each of the seven shows selected from the right of the average elite Twitter user. Paper 3 makes use of contemporary developments in small area estimation in the form of multilevel regression with poststratification to assess the dyadic relationship between an MP’s Twitter profile and their respective constituencies, establishing a within-party responsiveness to the left/right position of their constituents along a social dimension.

More information

Identifiers

ORCID for Conor Gaughan: orcid.org/0000-0002-0774-0111

ORCID for Will Jennings: orcid.org/0000-0001-9007-8896

ORCID for Rebecca Hoyle: orcid.org/0000-0002-1645-1071

Catalogue record

Export record