Pandemic Personas: Analyzing Identity Signals in COVID-19 Discourse on Twitter

Scott Leo Renshaw, Samantha C. Phillips, Michael Miller Yoder, Kathleen M. Carley

Abstract: The first year of the COVID-19 pandemic coincided with significant social and political changes. This article presents an exploratory analysis of Twitter users’ self-representations in the context of COVID-19. While some identities remained stable throughout the year, others appear to have been influenced by external events such as the Black Lives Matter protests and the U.S. presidential election. We also examine how users’ political identities are expressed in tweets, finding that right-leaning users are more likely to mention left-leaning identities, although both conservative and liberal-leaning users showed increased out-group focus in the period leading up to and after the election. Our study provides a comprehensive overview of user personas in a critical discourse topic, shedding light on identity signaling on social media. We end by identifying several clear opportunities for future research on social identity self-presentation on social media.