Media Bias and Political Polarization

The project tries to understand how different reporting strategies by newspapers affect news demand, belief updating, and political polarization. The project encompasses four key subprojects, each addressing distinct aspects of the topic, such as exploring the impact of slanted language on political polarization and estimating the relative weight of accuracy concerns versus belief confirmation. The research employs cutting-edge methods like large-scale information provision experiments, browsing data on news consumption linked to survey responses, and the use of recent developments in artificial intelligence. By unraveling the "black box" of news demand, this project seeks to offer profound insights into media consumption's role in shaping political views and polarization, potentially guiding the development of policies for a more informed and less polarized society.

The project, called POLARIZE, aims to investigate how media bias and different reporting styles affect belief formation and political polarization. POLARIZE will employ state-of-the-art methods, including large-scale information provision experiments, individually-linked browser data on actual news consumption, and modern artificial intelligence techniques to generate clean experimental variation in reporting styles. The project will include four interrelated subprojects to examine the causal effect of slanted language on political polarization, the impact of different reporting styles on memory and belief updating, the effects of opinion news versus straight news on political polarization, and how different forms of media bias affect the demand for new. By opening the "black box" of news demand, POLARIZE will improve our understanding of how to optimally regulate media markets and design evidence-based policies to promote a more informed and less polarized society.