Retweets as Trust: The Sharing of Pandemic-Related Information from U.S. Government Accounts
The rapid onset of the COVID-19 pandemic created a desire for information about the virus and government-directed response efforts. Twitter is a popular mechanism for disseminating crisis information, but questions exist about what government sources citizens trust enough to share their posts (“retweet”) as well as the types of language which encourage sharing. This study examines trust in U.S. government officials and organizations during the onset of the COVID-19 pandemic. The sharing of U.S.- and COVID-specific tweets from government accounts acts as a proxy for trust, representing implicit confidence in these units to provide credible information. This study investigated whether significant differences in the retweet frequency of government sources existed based on the source (public officials, public agencies, federal-level, or state-level, Republican-led states, or Democrat-led states) and differentiated by time period (pre- or post-declaration). Retweets were also examined to determine whether linguistic factors predicted the retweet frequency. The post-declaration period was found to have significant impact, with more active federal and state accounts. Significant group differences suggest greater trust in state-level sources versus federal sources (especially Democrat-led states), as well as public officials versus public agencies. Emotion-related language was also shown to predict the frequency of retweeting, though impacts differed based on the source.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of conflicts of interestNone
Corresponding author email@example.com
Lead author country
- United States
Lead author job role
- Higher Education Faculty 4-yr College
Lead author institutionPenn State University
Ethics statementThis study was not considered human subjects research according to Penn State Office of Research Protection policies.
- Yes, I agree to Advance terms