Public Perceptions of COVID-19 Vaccines: Policy Implications from US Spatiotemporal Sentiment Analytics

G G Md Nawaz Ali; Md Mokhlesur Rahman; Md Amjad Hossain; Md Shahinoor Rahman; Kamal Chandra Paul; Jean-Claude Thill; Jim Samuel

doi:10.3390/healthcare9091110

Public Perceptions of COVID-19 Vaccines: Policy Implications from US Spatiotemporal Sentiment Analytics

Healthcare (Basel). 2021 Aug 27;9(9):1110. doi: 10.3390/healthcare9091110.

Authors

G G Md Nawaz Ali¹, Md Mokhlesur Rahman^{2

3}, Md Amjad Hossain⁴, Md Shahinoor Rahman⁵, Kamal Chandra Paul⁶, Jean-Claude Thill⁷, Jim Samuel^{8

9}

Affiliations

¹ Department of Computer Science and Information Systems, Bradley University, Peoria, IL 61625, USA.
² The William States Lee College of Engineering, University of North Carolina at Charlotte, Charlotte, NC 28223, USA.
³ Department of Urban and Regional Planning (URP), Khulna University of Engineering & Technology (KUET), Khulna 9203, Bangladesh.
⁴ Department of Accounting, Information Systems, and Finance, Emporia State University, Emporia, KS 66801, USA.
⁵ Department of Earth and Environmental Sciences, New Jersey City University, Jersey City, NJ 07305, USA.
⁶ Department of Electrical and Computer Engineering, University of North Carolina at Charlotte, Charlotte, NC 28223, USA.
⁷ Department of Geography and Earth Sciences and School of Data Science, University of North Carolina at Charlotte, Charlotte, NC 28223, USA.
⁸ Department of Business Analytics, University of Charleston, Charleston, WV 25304, USA.
⁹ E.J. Bloustein School of Planning & Public Policy, Rutgers University, New Brunswick, NJ 08901, USA.

Abstract

There is a compelling and pressing need to better understand the temporal dynamics of public sentiment towards COVID-19 vaccines in the US on a national and state-wise level for facilitating appropriate public policy applications. Our analysis of social media data from early February and late March 2021 shows that, despite the overall strength of positive sentiment and despite the increasing numbers of Americans being fully vaccinated, negative sentiment towards COVID-19 vaccines still persists among segments of people who are hesitant towards the vaccine. In this study, we perform sentiment analytics on vaccine tweets, monitor changes in public sentiment over time, contrast vaccination sentiment scores with actual vaccination data from the US CDC and the Household Pulse Survey (HPS), explore the influence of maturity of Twitter user-accounts and generate geographic mapping of tweet sentiments. We observe that fear sentiment remained unchanged in populous states, whereas trust sentiment declined slightly in these same states. Changes in sentiments were more notable among less populous states in the central sections of the US. Furthermore, we leverage the emotion polarity based Public Sentiment Scenarios (PSS) framework, which was developed for COVID-19 sentiment analytics, to systematically posit implications for public policy processes with the aim of improving the positioning, messaging, and administration of vaccines. These insights are expected to contribute to policies that can expedite the vaccination program and move the nation closer to the cherished herd immunity goal.

Keywords: COVID-19; Public Sentiment Scenarios framework; Twitter; coronavirus; public policy; sentiment analysis; spatiotemporal analysis; textual analytics; vaccine.