What Predicts Selective Exposure Online: Testing Political Attitudes, Credibility, and Social Identity

Abstract

The online environment offers nearly unlimited sources and information, giving people unprecedented agency over selection. This article offers a test of three factors predicting selective exposure to opinionated news online—prior political attitudes, source credibility, and individual social identity based on race (Study 1) and gender (Study 2)—testing their independent and interactive effects. Two original selection studies on samples of adult Americans offered articles that were (1) pro- or counter-attitudinal with regard to individual attitudes on two political issues, (2) from high- versus low-credibility sources conceptualized in two different ways, and (3) featuring participants’ social in-group or out-group. Unobtrusively logged behavioral selection data suggest that prior political attitudes, and their strength and importance in Study 1, more strongly predict pro-attitudinal exposure than both source credibility and the demographically based social identity. Both studies additionally reveal nuanced interactions between the tested factors.