Faces of Biased Selectivity: A Latent Profile Analysis to Classify News Audiences and Their Selection Biases in the U.S. and UK

Abstract

The overload of news in today’s digital information environment can lead to biased media exposure on the individual level—for example, based on the confirmation of preexisting attitudes, attractiveness of negative news, and familiarity with sources. To better understand such news patterns and to whom these selection biases apply, this study identifies different classes of (biased) news audiences and explores several antecedents. A survey in the U.S. and UK presented respondents with multiple vignettes in the form of political news headlines that were altered to reflect (1) confirmation bias, (2) negativity bias, and (3) source bias. Respondents’ likelihood of selecting these biased news items was used to classify individuals into audience profiles. The results of a latent profile analysis provide three distinct and theoretically meaningful classes of news audiences that vary in terms of selection biases: avoiders, confirmers, and informers. As an important contribution, we show how these profiles are driven by the political attitudes and news preferences of audiences.