The Covid-19 Infodemic and the Efficacy of Corrections

Abstract

Social media platforms have taken unprecedented steps to combat misinformation about Covid-19. However, critics question whether the most common strategy – labeling and alerting readers to misinformation – successfully counters misinformation or paradoxically reinforces inaccurate beliefs. Using a pair of survey experiments, we examine the efficacy of two different corrections in reducing accuracy misperceptions and social media sharing that spreads false claims. Simply flagging fake headlines had little effect on subjects’ accuracy assessments and social media responses. Corrections explicitly countering false claims with factual information were more effective, though many respondents reported that they believed and would share the false information. Despite the increasing politicization of America’s pandemic response and polarization more generally, corrections to false claims with and without partisan valence were equally effective. We found no evidence of partisan backfire effects. The efficacy of corrections did vary with social media usage and age, a proxy for digital literacy. However, these moderating effects were inconsistent and often conflicting. Our results suggest that more aggressive corrections are essential to combating the Covid-19 infodemic; backlash risks are minimal and milder corrections ineffective.