Comparative Approaches to Mis/Disinformation| Fighting Zika With Honey: An Analysis of YouTube’s Video Recommendations on Brazilian YouTube

Abstract

This study focuses on misinformation about the Zika virus on YouTube in Brazil. Whereas most studies in health communication have so far conducted manual content analysis to understand the scope of health-related misinformation, we use a computational approach to understand the formation of information clusters via YouTube’s video recommendation algorithm on a platform level. Through network analysis and topic modeling of 20,499 YouTube videos, we are able to show a clear distinction between Portuguese-language and English-language YouTube recommendations in regard to this topic. The most prominent videos of both communities feature mostly accurate and correct information about Zika, but both communities are prone to faulty and potentially dangerous misinformation. We show that this harmful misinformation is not separate from the communities that feature correct information. In addition, we suggest that health misinformation surrounding both Zika and vaccinations seems to be more prevalent in YouTube’s video recommendation long tail.