2020 US Presidential Election and COVID-19 Sentiment Correlation​

Tristram Dacayan

Co-Presenters: Maliha Haider

College: The Dorothy and George Hennings College of Science, Mathematics and Technology

Major: Computer Information Systems (M.S.)

Faculty Research Mentor: Ching-yu Huang

Abstract:

The COVID-19 pandemic has affected both the physical and emotional well-being of the public, sparking intense debates on political topics, particularly on social media platforms like Twitter. This study explores the relationship between political affiliation and sentiment toward COVID-19 policies in social media during the 2020 U.S. presidential election. Using a dataset of 1.727 million election-related tweets from Kaggle and 364,254 COVID-19-related tweets from IEEE Dataport, we analyze sentiment trends and their correlation with public support for presidential candidates. Our methodology involves an Extract, Transform, Load (ETL) process to clean and preprocess the datasets, followed by sentiment analysis, time series analysis, regression modeling, and Pearson correlation testing.We hypothesize that higher negative sentiment in COVID-19 discussions correlates with increased negativity in election-related discourse. Additionally, we expect Democratic supporters to express more positive sentiment toward COVID-19 policies compared to Republican supporters. Our goals are to understand how political beliefs shaped opinions on COVID-19 policies and whether pandemic-related sentiment influenced election discussions. Using sentiment analysis and correlation techniques, we aim to reveal trends in public opinion and how major events impact political discourse. Our findings will help explain the connection between health crises and political sentiment during elections.