Browse By Day
Browse By Time
Browse By Person
Browse By Mini-Conference
Browse By Division
Browse By Session or Event Type
Fifty years ago, there was a captive audience for news; the three major network newscasts reached the vast majority of the adult audience. Today, the media market has changed dramatically, affording consumers considerable control over their exposure to public affairs information. The proliferation of news outlets, many offering content with explicit partisan slant, has revived the concept of selective exposure and the corollary expectation that partisans will gravitate to news providers aligned with their preferred party. Indeed, media scholars (e.g. Sunstein, 2017) typically cite the availability of “biased” news as an explanation for the significant intensification of party polarization in the post-Internet era.
We advance the literature on selective exposure by documenting individuals’ media diet and then how that diet changes in the immediate aftermath of events that threaten the standing of partisans’ favored party. In response, partisans engage in a form of defensive selectivity by which they become especially reliant on in-party sources. This pattern confirms classic psychological theories of cognitive consistency and dissonance avoidance.
We use a novel dataset that tracks individuals’ daily web browsing behavior between August and December 2016. We first show the extent to which an individual’s news consumption relies upon a single source. We then characterize how this media consumption shifts in response to major campaign events. To examine this shift, we identify major exogenous shocks to the web browsing time series – unexpected events that attracted widespread media attention such as the release of the tape containing Donald Trump’s comments about his willingness to grope women, James Comey’s letter reopening the investigation into Hillary Clinton’s emails, and the results of the 2016 election.
Our results provide new insights into how selective exposure affects individuals’ information seeking behavior. We demonstrate that individuals avoid information when it is particularly negative for their preferred party. Building on this analysis, we use a large scale text analysis to identify the content that individuals are consuming and how that consumption shifts after the onset of negative news. Our results lead to more nuanced and dynamic insights into how individuals process the news.