A Social Network Analysis of Pro-maskers and Anti-maskers on Twitter during the COVID-19 Pandemic (Preprint)

BACKGROUND Online debates surrounding face masks as a precautionary measure against COVID-19 pandemic have been raging on social media platforms like Twitter. Users against or in favor of masks have be vocal in sharing their opinions about the masks wearing, practices, and consequences. Through a social network analysis (SNA), we unearth the important contributors to the debate on Twitter surrounding pro and anti-masks. OBJECTIVE The aim of this study is to develop an understanding of the content and influencers on Twitter related to anti-mask and pro-mask debates. METHODS In this study 18,000 tweets related to keywords that were either pro mask and anti-mask from 12th December 2020, to 18th December 2020, and 18,000 tweets from 9th of December to the 18th of December 2020. The two datasets were analysed using social network analysis in Gephi, meanwhile NodeXL was used to produce network metrics and identify the key users, websites, and content within the data. RESULTS Discussions for both the cases revolved around prominent political figures and their supporters. The anti-mask discussions contained frequent mentions to Donald Trump’s Twitter account which made the account the most influential within the anti-mask network. On the contrary, for tweets that were pro-mask the most influential user was Joe Biden. Our network visualization further identified several clusters within the network and found that influential accounts appeared with sizeable clusters. Keywords used by Twitter users for pro-masking revolved around words that were in support of masking. Whereas tweets related to anti-masks appeared to be broader and included words related to news stories of popular figures not wearing face masks. For pro-mask discussions, popular websites also appeared to support the idea of wearing masks whereas the anti-mask data contained references to news stories of popular figures not wearing masks and of news stories related to a study which noted not wearing a mask was safer than wearing a used mask. CONCLUSIONS The current study highlights how discussions around pro-mask and anti-mask revolved around the key political figures. Our results are likely to be of interest to public health authorities involved in encouraging the public to wear masks. The study highlighted key users that could be targeted for information dissemination about face masks. The study further provided insight into narratives and online resources shared through pro-masking and anti-masking accounts.