An actor-based approach to understanding radical right viral tweets in the UK

ABSTRACT

Radical right actors routinely use social media to spread highly divisive, disruptive, and anti-democratic messages. Assessing and countering such content is crucial for ensuring that online spaces can be open, accessible, and constructive. However, previous work has paid little attention to understanding factors associated with radical right content that goes viral. We investigate this issue with a new dataset (the ‘ROT' dataset) which provides insight into the content, engagement, and followership of a set of 35 radical right actors who are active in the UK. ROT contains over 50,000 original entries and over 40 million retweets, quotes, replies and mentions, as well as detailed information about followership. We use a multilevel model to assess engagement with tweets and show the importance of both actor- and content-level factors, including the number of followers each actor has, the toxicity of their content, the presence of media and explicit requests for retweets. We argue that it is crucial to account for role of actors in radical right viral tweets, and therefore, moderation efforts should be taken not only on a post-to-post level but also on an account level.

KEYWORDS:

• Radical right
• social media
• polarisation
• Twitter
• far right

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Dataset available at https://zenodo.org/record/6393085.

2 Quotes are retweets where the retweeter also makes a comment.

3 Due to account suspensions, 17,697 original entries were no longer available at the end of the period. Otherwise, only 2.6% of original entries had been deleted.

4 ROT contains mentions of the actors which are not in direct response to an actors’ content (e.g., a Twitter user writing a new tweet in which they @ mention one of the actors). We do not use them for our analyses as these mentions are not associated with original entries produced by the actors.

5 Note: we do not only collect content produced by Twitter accounts set to ‘private’.

6 https://www.perspectiveapi.com/.

7 The automatic detection of abusive or toxic content poses several challenges. We comment on many of these in the discussion section.

8 These correspond to the percentage of actors’ followers who were suspended or deleted three months after data collection.

9 The ROT dataset will be released shortly by the authors.

10 July 10, 12, 13, 15, 17, 18, 20, 23, 25, 27, 29, 31 and August 5, 6.

11 blaiklockBP, PrisonPlanet, DVATW, GerardBattenUK, LeaveEUOfficial, Michael Heaver, Henrik Palmgren, LanaLokteff.

Additional information

Funding

This study was supported by Wave 1 of The UKRI Strategic Priorities Fund under the EPSRC [grant number EP/T001569/1].