Utilising interaction features on Twitter/now X to understand the 'messaging' of migration amongst non-elite users

Abstract

Different aspects of interactions on social media – communication and action – imply distinctive ways of knowing the social world. I present a new methodological approach that utilizes ‘big’ social media data to understand politically salient issues such as the ‘messaging’ of migration on Twitter/now X? An iterative abductive interpretivist analytical strategy drawing on computational and qualitative social science techniques was applied to a corpus of 47,978 tweets created over five months around the time of lifting of temporary controls on free movement from Romania and Bulgaria to the UK in January 2014. Initial computational network analysis on the retweet action feature revealed a small number of highly influential users and a large proportion of isolated users (non-elites) who were never retweeted. Given paucity of understanding of how elite narratives on migration are absorbed, accepted or contested by non-elites, the next stage involved qualitative thematic analysis of a sub-sample of actual tweets (communication) from non- elites to understand meaning-making in views expressed. Qualitative analysis confirmed presence of highly polarised immigration attitudes amongst non-elites but also revealed their values and beliefs about national belonging. These findings prompted questions about what or who influences these values amongst non-elites and whether there are any structural differences in information flows amongst anti- and pro-immigration users. In the third stage, computational surface thematic mapping of different aspects of communication and action in the whole corpus revealed importance of the entire media environment but also differences in the presence or lack of echo-chambers amongst those expressing anti or pro-immigrant sentiments. This article demonstrates the potential of cross-disciplinary analytical strategies when investigating politically salient issues on social media.

More information

Identifiers

ORCID for Bindi Shah: orcid.org/0000-0002-5571-9755

Catalogue record

Export record

Altmetrics