Revealing public opinion towards COVID-19 vaccines with Twitter Data in the United States: a spatiotemporal perspective

Background The COVID-19 pandemic has imposed a large, initially uncontrollable, public health crisis both in the US and across the world, with experts looking to vaccines as the ultimate mechanism of defense. The development and deployment of COVID-19 vaccines have been rapidly advancing via global efforts. Hence, it is crucial for governments, public health officials, and policy makers to understand public attitudes and opinions towards vaccines, such that effective interventions and educational campaigns can be designed to promote vaccine acceptance.

Objective The aim of this study is to investigate public opinion and perception on COVID-19 vaccines by investigating the spatiotemporal trends of their sentiment and emotion towards vaccines, as well as how such trends relate to popular topics on Twitter in the US.

Methods We collected over 300,000 geotagged tweets in the US from March 1, 2020 to February 28, 2021. We examined the spatiotemporal patterns of public sentiment and emotion over time at both national and state scales and identified three phases along the pandemic timeline with the significant changes of public sentiment and emotion, further linking to eleven key events and major topics as the potential drivers to induce such changes via cloud mapping of keywords and topic modelling.

Results An increasing trend of positive sentiment in parallel with the decrease of negative sentiment are generally observed in most states, reflecting the rising confidence and anticipation of the public towards vaccines. The overall tendency of the eight types of emotion implies the trustiness and anticipation of the public to vaccination, accompanied by the mixture of fear, sadness and anger. Critical social/international events and/or the announcements of political leaders and authorities may have potential impacts on the public opinion on vaccines. These factors, along with important topics and manual reading of popular posts on eleven key events, help identify underlying themes and validate insights from the analysis.

Conclusions The analyses of near real-time social media big data benefit public health authorities by enabling them to monitor public attitudes and opinions towards vaccine-related information in a geo-aware manner, address the concerns of vaccine skeptics and promote the confidence of individuals within a certain region or community, towards vaccines.