Filter Bubbles, Echo Chambers, and Epistemic Bubbles in English Young People

Particularly since the shock popularity and victories of Brexit and Donald Trump, there has been increased concern that citizens exist in democratically dysfunctional ideological bubbles, where they only hear likeminded perspectives. Researchers examined the extent of these bubbles - particularly through analysis of digital platforms such as search engines and social media. Our understanding of this ‘embubblement’ and its implications for democracy is hindered by the relative lack of qualitative research on filter bubbles and young people, and the overly simple way media exposure is often measured, which fails to consider the context (and assumes, for example, that all cross-cutting exposure is ‘good’). The study explores embubblement in this marginal, perhaps high-risk group, who get more news online and are considered more impressionable. This mixed-methods digital ethnography contains a 10-wave cohort study, diary study hybrid. One day a month for 10 months, English participants aged 16-18 (n=20) captured any political communication they encountered across all mediums - online and in-person. Descriptive statistics using regression analysis suggest strength of partisanship positively correlates with embubblement, though no participants were strongly embubbled (even strong partisans). No statistically significant correlations emerged between embubblement and increased embubblement over time or political polarisation. However, embubblement positively correlated with degree of political engagement. Ethnography explored what causes embubblement. Embubblement occurred rarely, influenced by structural factors: ‘socialising agents’, including family, peers, education, media, and events. This thesis makes a new contribution, a typology of factors shaping embubblement, incorporating an agent-centred approach. Main factors were agreeable news sites, apps and hyperpartisan social media communities. The research addresses questions of agency – for example, a user making a new TikTok account, to reset personalisation algorithms after realising the existing ones were radicalising her. Implications for schools and policy-makers are addressed through recommendations on how to encourage political engagement without embubbling citizens.