Travis A. Kelly

Little research has currently been done focusing on how the COVID-19 pandemic has affected the mental and emotional processes of people in the United States. The few studies that have been published so far have focused on self-report measures to analyze the impact. While these publications are a good first step, more research needs to be done using methods controlling for the bias often found in self-report data. As very little research has been used with naturally occurring data, the present study focuses on this approach. The study uses code in R to pull content from Twitter to analyze certain emotions based on the linguistic content. Results from this study showed that there is a significant change in depression, suicidal ideation, emotional distress, and/or uncertainty in different areas. This includes changes in these categories from one year to the next, changes in the way people talk about COVID-19, and changes in the way people talk on Twitter during COVID-19 but before vaccine distribution and after vaccine distribution began in the United States. Implications of this study include the need for increased awareness and further research on how COVID-19 has impacted mental health.
Keywords: Big data, depression, emotional distress, Twitter, COVID-19