Global Psychological Consequences of COVID-19: A Longitudinal Sentiment Analysis of Social Media Data

Global events such as the COVID-19 pandemic have profoundly affected physical, psychological, economic, and social domains. While physical influences can be estimated directly, psychological consequences remain difficult to measure because constructs such as emotion and well-being are unobservable. This project takes advantage of the longitudinal social media data to investigate global emotional changes during the pandemic. Specifically, we (1) examine how potential factors influence emotional trajectories across countries, (2) apply lexicon-based and large language model (LLM)-based sentiment analysis to analyze the data, and (3) develop an R software package to advance longitudinal text analysis in psychology and public health research.

Image