Dataset for "Battle for Britain: Analyzing Events as Drivers of Political Tribalism in Twitter Discussions of Brexit”

In this study, we investigate how Brexit tribalism has unfolded over time on Twitter. The dataset contains a corpus of tweets posted to Twitter during a period of 32 months following the 2016 UK European Union membership referendum. The tweets were selected as a result of searching for keywords: firstly for "Brexiteer" and "Remainer" and secondly for "Brextremist" and "Remoaner". The CSV file in this dataset contains both sets of results. There are two columns in the file: timestamp and tweet text, which will be sufficient to replicate our process. Tweet IDs were removed to preserve user anonymity. First, we characterize the nature of the discussion by comparing language use patterns between tweets containing Brexiteer/Remainer and Brextremist/Remoaner keywords. We find that Brextremist/Remoaner are more commonly used in a derogatory way. We also find that all four group identity keywords are used more frequently over time, suggesting an increase in tribal interactions. Finally, we find evidence of a relationship between real‐life Brexit events and spikes in tribal responses online.”