Spatial Distribution Of Twitter Discussion Topics Regarding Covid-19 And Related Public Health Policies

ASU Author/Contributor (non-ASU co-authors, if there are any, appear on document)
Harrison Brown (Creator)
Institution
Appalachian State University (ASU )
Web Site: https://library.appstate.edu/
Advisor
Kara Dempsey

Abstract: Since the onset of the COVID-19 pandemic, general public opinion in the United States has shown a decrease in trust towards public health organizations and public health campaigns. Using social media as a lens for public opinion, this paper highlights the spatial distribution of the most widespread discussion topics among Twitter users regarding COVID-19, vaccinations, mask mandates, social distancing, quarantines, and shelter-in-place procedures. Using Topic Modeling, this study examined more than 5 million Tweets from January 1, 2020, to January 1, 2022.This study performs Topic Modeling, using Latent Dirichlet Allocation (LDA) as an intermediate step in analyzing the spatial distribution of topics across geographic scales. Analyzing the Twitter data provided 8 latent topics, from COVID-19 misinformation to vaccine requirements at school. Through Twitter geolocation information, each topic shows a unique spatial distribution across the contiguous United States. Analyzing and interpreting the prevalence of these topics allows policymakers to better understand sentiments regarding COVID-19 and related public health campaigns at the community and state level.

Additional Information

Publication
Thesis
Brown, H. (2022). Spatial Distribution Of Twitter Discussion Topics Regarding Covid-19 And Related Public Health Policies. Unpublished Master’s Thesis. Appalachian State University, Boone, NC.
Language: English
Date: 2022
Keywords
Twitter, COVID-19, Topic Modeling, GIS, Social Media

Email this document to