The inherent originality and vast temporal and spatial reach in the expressions on social media data have made it a potentially important tool for researchers to examine consumer behaviors, attitudes, and perception in real-time. Leveraging data from one social media platform, Twitter, we build a model for identifying people with language markers of depression, anxiety and psychological stress to explore the relationship between student loan debt, gender, and mental health among individuals vulnerable to mental illness. The objective of this research is to demonstrate the efficacy of Twitter posts as a source of data for studying mental health issues. We do this by comparing the student loan-related sentiments and emotions of Twitter users whose posts indicate markers of depression, anxiety, and/or psychological stress with those who express no markers. Consistent with a growing number of studies on mental health consequences of student loan debts (Deckard et al., 2021; Jessop et al., 2020; Rodney & Mincey, 2020), findings from our study indicated disproportionate volumes of negative sentiments and emotions about student loans among Twitter users with markers of depression, anxiety, and psychological stress across genders. 

Accepted Oral Presentation