COMPutational Analysis of Social System Lab

This study compared Twitter social-network characteristics of HBCU students before and during the COVID-19 pandemic, focusing on those who posted or reshared depression-related content versus those who did not. By quantifying structural properties—such as average friend count, reciprocity, assortativity, modularity, and community entropy—the authors found that, during the pandemic, students tweeting about depression increased their average number of friends (from 1,194 to 1,371), strengthened mutual connections (reciprocity 0.78 → 0.80), and showed higher homophily with similarly affected peers (assortativity 0.60 → 0.70). Affiliation communities became more aligned with real-world groups (entropy 1.0 → 0.5) among the depression group, while non-depression users maintained stable community structure (modularity 0.75 → 0.76). These patterns highlight opportunities for targeted social-media interventions to support Black college students’ mental health.