Abstract
Since the 2016 US presidential election, the production, distribution, consumption, and influence of false and otherwise misleading news has emerged as a topic of urgent interest among researchers, journalists, and policymakers. Underlying this widespread concern are two assumptions: first, that news-related content broadly construed comprises a significant fraction of the total information diet of ordinary Americans; and second, that the prevalence of fake news is comparable to, or even greater than, that of mainstream news (Silverman 2016; Vosoughi, Roy, and Aral 2018). In this project we propose to examine both these assumptions by quantifying the prevalence of different categories of content—non-news, news in general, and fake news—on Facebook, Twitter, and in a representative panel of web users. In addition, we will examine how these fractions vary by region, age, gender, and ideological leaning. Our objective in answering these questions is to place the ongoing debate around the impact of fake news and social media platforms on democratic processes in the broader context of the total information diet of Americans, as well as to suggest further directions for media research in general.
Principal Investigator
Duncan Watts
Principal Researcher and Partner, University of Pennsylvania
Participants
Hunt Allcott
Associate Professor of Economics, New York University
David Rothschild
Economist, Microsoft Research