“Dark under the light”? The differences in the Overlooked emotions of insiders versus the intensified emotions of outsiders on Weibo and Twitter during the COVID-19 pandemic

This study analyzes social media data related to COVID-19 to identify geographic and temporal differences in emotional expressions. Utilizing the Linguistic Inquiry and Word Count tool, we examined emotional expressions of anxiety and anger among insiders near the epicenter and outsiders during three phases of the pandemic. Our dataset was collected from Weibo (482, 428 posts from Jan 1st, 2020, to April 7th, 2020) and Twitter (48,759 posts from Jan 19th, 2020, to May 5th, 2020). Insiders from China near the epicenter predominantly exhibit anxiety, peaking in the early phase, while insiders from the United States near the epicenter predominantly exhibit anger, peaking in the early phase as well. Outsiders from China express more anger, with a nadir during lockdown, while outsiders from the U.S. express more anxiety, especially during lockdown. Despite similar emotional intensity, insiders’ anger-related posts from Weibo and Twitter suggest less virality than anger-related posts from outsiders. However, insiders’ posts with anxiety words are more likely to go viral on Twitter but less likely to go viral on Weibo compared with outsiders’ posts with anxiety words. These findings enhance our understanding of risk perception and highlight the necessity for emotional regulation strategies. These insights could help develop targeted non-pharmacological interventions to reduce anxiety among insiders and anger among outsiders during public health crises.