According to a new study published in the International Journal of Press/Politics, exposure to partisan and centrist news websites does not enhance polarization, finding that politics and partisan news account for a miniscule proportion of overall information consumption.
Over the past several decades, America has become increasingly politically polarized, with some blaming this trend on the emergence of partisan media. Some studies have shown that exposure to partisan news outlets can reinforce prior attitudes and out-party hostility; however, these results have largely stemmed from surveys and experiments.
Study authors Magdalena Wojcieszak and colleagues argue that this provides limited insight into the polarizing effects of partisan media in the real world, writing “it is crucial to offer over-time evidence on these effects in naturalistic settings, where people can select content from an unrestricted set of available alternatives.”
The researchers relied on two studies that combined two-wave panel surveys, each comprised of distinct samples of American adults. Study 1 participants included 303 American Facebook users, while Study 2 included a quota sample of 904 U.S. adults. Prior to completing the study survey, participants submitted their internet browsing data at two-time points, for a total of twelve months’ worth of browsing data; this totaled over thirty-seven million visits to various websites.
Collecting participants’ browsing data enabled researchers to map changes in respondents’ actual exposure to centrist and partisan news, and track their effects on polarization, both attitudinally and affectively.
Attitude polarization was measured by gauging participants’ agreement with liberal or conservative solutions to various issues or policies that were salient at the time the studies were conducted (April 2018 for Study 1 and April 2019 for Study 2). Study 1 explored topics such as immigration, gun control, sexual assault, and Islam, while Study 2 asked about the economy, environment, as well as immigration and gun control.
Affective polarization was measured by prompting participants to provide ratings on a “feeling thermometer” toward out-partisans on a scale ranging from “cold and unfavorable” to “warm and favorable”.
Study 1 also measured negative trait ratings and social distance, while Study 2 measured the latter and included a novel measure gauging participants’ understanding of out-partisans.
Wojcieszak and colleagues report “robust null findings”, such that, real-world online exposure to liberal and conservative news websites, regardless of whether they were congenial or dissimilar, did not enhance participants’ attitude nor affective polarization. In other words, participants’ policy attitudes did not become more extreme, nor did they become more hostile toward out-party supporters. Partisan news exposure had no polarizing effects among Democrats, Republicans, or strong partisans, with null effects emerging to political articles within partisan and centrist sites.
The researchers argue that their null findings challenge “the popular narrative that partisan news is to blame for the ills of contemporary US politics,” suggesting their results provide a more accurate representation of the reality of how limited the effects of partisan news are in the real-world. Politics and partisan news accounted for a very small fraction of respondents’ online activities, comprising less than 2% of the trace data – which are “nearly unnoticeable in the overall information and communication ecology of most individuals.” Thus, their polarizing effects are also limited.
The most prominent limitation the authors note is that this study did not account for the overall communication and information ecology of participants – such as news exposure on multiple computers, mobile devices, and offline use (e.g., listening to the radio in the car, watching TV at home). They suggest the lack of mobile data is particularly important given a growing number of Americans consume news via mobile devices (as opposed to using a desktop or laptop computer).
As well, the researchers could not capture the partisan pages that respondents follow on social media, or content shared by the individuals they follow. Thus, their research likely underestimates participants’ mere exposure to partisan news; however, it captures behavior that is arguably more meaningful – actual engagement with news via clicking a URL.
The study, “No Polarization From Partisan News: Over-Time Evidence From Trace Data”, was authored by Magdalena Wojcieszak, Sjifra de Leeuw, Ericka Menchen-Trevino, Seungsu Lee, Ke M. Huang-Isherwood, and Brian Weeks.