Political Communication and Public Opinion

It is a remarkable time to be a scholar of public opinion and political communication. Not only has the political media environment been fundamentally transformed in the past 20 years, but the rate of ongoing transformation shows no sign of slowing. Digital platforms like Facebook (and its subsidiaries such as Instagram and WhatsApp) are increasingly powerful curators of news and political content. Algorithms designed by platform companies enable high levels of personalization and fragmentation of media audiences; platform-initiated changes to algorithms and newsfeed features can open, or close, the floodgates of visitors to news media websites; the social actions of digital media users are increasingly tracked and turned into data for use in shaping future content visibility—and to classify users for sale to political advertisers. Growth in public concern about information disorder has placed a spotlight on questions that have long been at the heart of public opinion and political communication research: Who sees news and political content? Which intermediaries in the political sphere have the power to shape content exposure and public opinion? What are the impacts of the structure of our information environments on democratically desirable—and undesirable—outcomes? Who are the haves and have-nots in political engagement facilitated by digital platforms? Political communication researchers have responded to the challenges posed by our chaotic media environment with an explosion of methodological and theoretical creativity. No longer can we rely on influence to move from elite content producers to the citizenry in simple “two-step” flows—if we ever could. Ours is a time of a multiplicity of content flows, curated on social media by many actors: politicians and news organizations, but also individuals themselves through networks of sharing, alongside processes of algorithmic curation undertaken by platforms themselves (Chadwick 2013; Thorson and Wells 2016). Establishing who sees what political content, through what pathway, to what effect is hard to untangle under conditions of highly personalized, algorithmically curated newsfeeds, and when patterns of news use are increasingly fragmented (Edgerly 2015). While it has never been an easy task to measure media exposure and effects on public opinion and engagement, transformations in the media environment have certainly increased the level of difficulty, even while offering the tantalizing possibility of using digital trace data for measurement of key concepts. Our hope for this special issue is to highlight some of the most exciting new work in public opinion and political communication that grapples head-on with changes in the media environment. Below, we highlight three themes that emerged across the papers included in this issue: methodological innovation, changing patterns of content production and circulation, and an emphasis on explicating inequalities in political information exposure and engagement. The rise of digital media has been accompanied by an increase in datafication, the process by which our social actions are translated into data for use in algorithmic prediction and behavioral tracking—with perhaps dangerous implications for a broad array of democratic outcomes, including news exposure and production (Webster 2011; van Dijck 2014; van Dijck, Poell, and De Waal 2018). At the same time, these digital traces of user behavior provide researchers with some of the best sources of data to test the relationship between digital media use and effect. Scholars have begun to use digital trace data to measure classic public opinion concepts such as exposure, information seeking, affect, issue opinions, and political interest. Many of these studies pair digital trace data with user surveys to explain not only what digital media users are seeing, but why (e.g., Stier et al. 2019). Such studies have helped push back empirically against early assumptions of digital politics research, such as the filter bubble hypothesis that social media will lead to high levels of partisan media siloes (Moeller and Helberger 2018; Bruns 2019). Kraft, Krupnikov, Milita, Ryan, and Soroka leverage a large body of social media data to examine patterns of news sharing across platforms. Their focus is on news sentiment—the negativity or positivity of the story. They find that the news stories people choose to read on the web are more negative than the stories that are shared via social media. Their results suggest that sharing patterns are related to self-presentation concerns, such that news curators are more likely to highlight the positive when they share news with others. Bryanov, Watson, Pingree, and Santia focus their methodological innovation toward creation of complex experimental stimuli. The researchers built a platform that mimics the experience of consuming news via a digitally curated news aggregator. This approach allows the authors to test hypotheses about the impact of being embedded in more and less partisan news streams on satisfaction with the platform as well as exposure to mainstream (versus partisan) content. Their findings do not suggest strong effects of partisan selective exposure, but instead reveal more subtle challenges related to increased dosage of exposure to partisan content: Those whose newsfeeds contained more partisan content matching their preferences were more likely to be satisfied with the platform itself and were less likely to engage with mainstream news content. Finally, no single research method holds a monopoly on understanding the new, complex information environment and its democratic implications. Not all methodological innovation is computational in nature. As the studies in this special issue exemplify, we gain insight through diverse and creative approaches to studying the information environment. Shannon McGregor draws on interviews with political campaign professions to build theory about the use of social media to construct representations of public opinion. Her rich, context-aware study pushes us to rethink our understanding of what can be known about the public from social media, and the ways in which portrayals of public opinion are contingent on complex interactions between elites and everyday citizens on platforms like Twitter. In doing so, McGregor reminds us of the need for qualitative data to help us make sense of public opinion in the digital age. Another innovative line of research highlighted in this special issue explores the ways that the content creation and circulation practices of elite and non-elite actors alike are changing alongside pressures to become visible across digital platforms (Chadwick 2013; Jungherr, Posegga, and An 2019). Here, we see rich empirical work investigating motivations for creating and sharing political content on social platforms, as well as studies of the shaping role of network logics on the behavior of politicians and news organizations (Kreiss and McGregor 2018; Nielsen and Ganter 2018). The digital media environment is creating new opportunities for elite influence, but also for motivated individuals to engage in information seeking, content creation, and other forms of self-expression. This line of scholarship reminds us of the importance of the impossibility of separating studies of content production and circulation from analyses of media effects on public opinion—and that expression effects may be as important as consumption effects in the digital era (Pingree 2007). An excellent complement to Kraft et al.’s work on affect and news sharing is Daniel Lane’s investigation of expressive citizenship norms as motivators of online political talk. Based on a survey of young adults in the United States, Lane’s findings push back against the dominant image of “expression-focused citizenship” on social media. His results show that expression norms are among the least important norms and are only modestly associated with political expression on social media. Lane calls for an expanded research agenda on political expression focused on the shaping role of political identities and socialization, particularly among marginalized youth, in the development of expressive civic norms. Kosmidis and Theocharis examine the connections between exposure to incivility on social media and emotional responses. They find that exposure to uncivil interactions on Twitter can spark enthusiasm, especially among partisans who share the views of the Twitter commentary. Their studies remind us that there are good reasons for elites to intentionally provoke uncivil exchanges on platforms like Twitter—such exchanges are both “harmful and profitable” because they can lead to emotional responses that prepare audiences for mobilization. Hmielowski and colleagues explore the dynamics of offline and online communication behaviors around elections. They combine a panel survey with Google Trends data (a useful methodological innovation) to show that political information seeking, media use, and discussion increase over the course of election cycles. Further, they point out this progress is not linear—campaign events such as debates and conventions drive sharper increases in political communication behavior. One implication of these findings, as they note, is to add to the call for political communication scholars to think outside of linear models of political behavior and effects. A final theme in recent public opinion and political communication scholarship that is brought forward in the special issue is a renewed attention to inequalities in news exposure and political engagement via digital platforms. The hope of early scholarship on digital media and public opinion was that the internet would ease inequalities between political information haves and have-nots, and mobilize new citizens into political action. These hopes have largely not been borne out. As predicted by the more pessimistic early observers of digital politics, digital divides in political uses of new media remain, especially between those high and low in political interest (Bimber 2003; Prior 2007). It now seems clear that the new media environment provides new spaces to engage with news and politics, but that these opportunities are not equally distributed. As studies in this special issue demonstrate, attention to inequalities in information exposure and engagement has remained a focal area for political communication—and likely will for some time. A critical space for examining inequalities is in the rate of news exposure among different groups. Sabine Geers explores the possibility of digital divides in news consumption across the population of Dutch adolescents. Adopting a news repertoire approach, she uses survey data from Dutch youth (ages 16 to 21) to identify four types of news consumers: minimalists, omnivores, traditionalists, and online news users. Her findings confirm digital divides in news usage based on interest, but contradict prior research findings that education levels are important differentiators of news use patterns. Jennifer Oser and Shelley Boulianne conduct a meta-analysis of repeated-wave panel data studies that test the relationship between digital media use and political participation. They find strong evidence that digital media use is more likely to reinforce the political engagement of the already engaged, rather than mobilizing new participants into action. They tie their findings to a call for more research on the long-term impact of digital media use for participation inequalities. Finally, studies in this special issue remind us that research on the impact of digital media on society has to date been overwhelmingly focused on the United States and Western democracies. This focus has led to a failure to recognize that the relationships between digital media use and political outcomes are highly contingent on national contexts and variations in media systems. This perspective is exemplified in the article from Toff and Kalogeropoulos, who examine patterns of news avoidance across 35 countries. Their findings point to the complex interplay between individual-level explanations for avoiding news and contextual factors, shaping what the authors call “cultures of news consumption.” These results invite future research to consider how news avoidance can hold different meanings and democratic implications across different cultures of news consumption. Altogether, the papers in this special issue highlight the new opportunities, and accompanying inequalities, for exposure to news and political information in today’s media environment. The papers reflect a diverse set of approaches and foci for studying the complexities of information sharing, political influence, and media choice that shape democratic outcomes.