Social Data Mining in the Post-pandemic Era (SocDM 2022​)

The COVID-19 pandemic has brought vaccination from the purview of parents and the health-compromised to everyone in the public. However, vaccination hesitancy continues to limit the impact of this highly effective intervention. Anti-vaccination views pervade online social media, fueling distrust in scientific expertise and increasing vaccine-hesitant attitudes. The pandemic brought the vaccination discourse onto the world stage, necessitating effective countermeasures to tackle the spread of potentially harmful misinformation on a global scale. Unfortunately, social media discourse has been shown to form “echo chambers” wherein individuals receive information they are likely to agree with, and surround themselves with likeminded others. In this talk, I will present the analysis of large vaccine-related Twitter messages in 18 languages from 2019 to 2021. First, I will focus on Italy, one of the first and most affected European nations by the virus. In particular, we use network analysis to show the clusters of like-minded users, with those sharing anti-vaccination content communicating rarely with those promoting the vaccines. We show that this separation persists throughout the pandemic. However, using content analysis we find that over time the two groups speak on increasingly similar topics. Then, I will turn to 28 countries, and show that many display such polarization in their vaccine debates. Moreover, we show that the anti-vaccination clusters tend to be connected across countries, making a kind of international Twitter no-vax echo chamber. US users are central in this network, while Russian users also become net exporters of misinformation during vaccination roll-out. Interestingly, we find that Twitter’s content moderation efforts, and in particular the suspension of users following the January 6th U.S. Capitol attack had a worldwide impact in reducing misinformation spread about vaccines. I will finish with a discussion of the limitations of this methodology, privacy concerns around this sensitive topic, and thoughts for future research in the area.