COMMUNICATION AND PUBLIC OPINION PLUS C x A CHANGE ?

Three central themes that have persisted throughout the history of research on communication and public opinion are examined in light of past, present, and future research. These themes include (1) ongoing concerns surrounding the political diversity of the communication environment; (2) selective exposure to political communication; and (3) the interrelationship between mass and interpersonal political communication. We explore the importance of these themes with an emphasis on how technological changes have made them, if anything, more relevant today than they were when first identified as central concerns of the discipline. To say that communication technology has changed dramatically since the last Public Opinion Quarterly anniversary issue in 1987 is to state the obvious. But technological change alone does not necessarily dictate changes in the locus of scholarly concern. In this article, we suggest that many of the same themes that Elihu Katz identified on the 50th anniversary ofPublic OpinionQuarterly— which were themselves drawn from the earliest studies of communication and public opinion in the 1940s and 1950s—are as relevant today as ever. In this article, we review three major themes that connect studies of communication and public opinion from the past to those in the present and likely future. First, scholarly activity reflects an ongoing interest in the political diversity of the communication environment. To what extent does the information environment approximate the democratic ideal of a marketplace of ideas? In this case, the research question is largely a descriptive one, asking to what extent the state of the communication environment is as one would like it to be. Beyond diversity, two additional themes follow directly from the earliest empirical DIANA C. MUTZ is the Samuel A. Stouffer Professor of Political Science and Communication at the University of Pennsylvania, Philadelphia, PA, USA. LORI YOUNG is a Ph.D. student at the Annenberg School for Communication at the University of Pennsylvania, Philadelphia, PA, USA. *Address correspondence to Diana C. Mutz, University of Pennsylvania, 208 S. 37th Street, Room 217, Philadelphia, PA 19104, USA; e-mail: [email protected]. doi: 10.1093/poq/nfr052 The Author 2011. Published by Oxford University Press on behalf of the American Association for Public Opinion Research. All rights reserved. For permissions, please e-mail: [email protected] at U niersity of Penylvania L irary on N ovem er 9, 2012 http://poqrdjournals.org/ D ow nladed from studies of media and public opinion at Columbia University in the 1940s and 1950s. An emphasis on selective exposure was proposed by the Columbia researchers as an explanation for the lack of persuasive media effects found in their studies. Now the same theme is the focus of intensified research interest because of the proliferation of television networks and Internet news. A third theme emanating from the Columbia studies was interpersonal communication. Over the past 25 years, studies of interpersonal communication have experienced a massive renaissance in the form of studies of deliberation and social networks. Moreover, one of the very earliest theories about interactions between mass and interpersonal communication—the two-step flow—is now more relevant than ever before. Although the conclusions that scholars draw may not be the same as 25 or 50 years ago, in the process of reviewing current work, we have found that what is important to scholars of the past and present is uncannily familiar. Our three themes reflect scholarly interests during the past and present eras. Interestingly, throughout the period of Public Opinion Quarterly s existence, the dominant public understanding of the importance of communication to public opinion has remained relatively constant and straightforward. In the United States and Europe, the public generally believes that media of all kinds consistently have large persuasive effects on political opinion (e.g., Parisot 1990; Schudson 1996). This fact is fascinating, if only because this perception has remained constant despite the waxing and waning of academic assessments over the years. As Warren Weaver of the New York Times put it in the foreword to Patterson and McClure s (1976) seminal book, The Unseeing Eye: The Myth of Television Power in National Politics, ‘‘If this pioneering book does not send shock waves through the broadcasting industry, the major political strategists of both parties, and the political science community, nothing ever will’’ (p. 18). It thus appears likely that nothing will ever change this perception in the eyes of political strategists and the mass public. Perceptions of the media s political importance have not waned among the public, and the amount spent on communicating in order to influence public opinion has only increased (Bennett and Iyengar 2008), evidence notwithstanding. Based on this assessment, it seems fair to say that academic research remains largely irrelevant to what transpires in the real world (see Green and Smith 2003). Political consultants may shift spending from one form of communication to another, or from one strategy to another, but they remain true believers in the persuasive powers of the media (e.g., Vavreck 2009). Paradoxically, direct persuasive effects are the one area of research on communication and political opinion where academic interest has not flourished (see Neuman and Guggenheim 2011). Communication and Public Opinion 1019 at U niersity of Penylvania L irary on N ovem er 9, 2012 http://poqrdjournals.org/ D ow nladed from For a combination of reasons, these types of effects remain elusive. However, our ultimate concern is not with whether direct persuasive effects or effects of any other kind are minimal or massive, as these terms have limited meaning or utility. Instead, we are interested in how the themes we have identified have evolved during the years since the last Public Opinion Quarterly anniversary. These themes continue to reflect a concern for societal harm, but not as a result of the kind of persuasion that preoccupies the public. In the 50th-anniversary issue, Jim Beniger (1987) observed that the term ‘‘mass society’’ had lost its currency, along with enthusiasm for studying direct persuasive effects (see also Neuman 1991). But, as we document here, the mass society theme of societal impoverishment due to changing patterns of communication has nonetheless retained its power and significance in scholarly writing. In this current era, however, the kind of effect on public opinion that is of greatest concern to scholars is no longer changing people s opinions so much as reinforcing them and making them more extreme. Reinforcement was prominent in the Columbia studies as well, but these scholars were heavily criticized for considering it a ‘‘non-effect’’ (e.g., Gitlin 1978). As with the telephone, films, and television, inflated hopes about the effects of each new communication technology have been followed soon thereafter by a fear of imminent cultural decline along the lines suggested by mass society theory (see Mutz 1998). The most recent ‘‘new media’’ have been no exception in this regard. The most obvious technological change in communication since the 50th-anniversary issue has been the rise of the Internet. And, once again, scholars have exuded both high hopes and grave predictions. On the positive side are those who predict the rise of a stronger participatory culture as a result of the Internet (e.g., Jenkins 2006), an ‘‘empowering’’ of citizens in contrast to passive television viewing (e.g., Lawson-Borders 2006). But, just as Putnam (1995) and others blamed television for robbing people of their social networks, the Internet is our new whipping boy for fears about the loss of social relationships: We re moving from a world in which you know all your neighbors, see all your friends, interact with lots of different people every day, to a functional world, where interaction takes place at a distance . . . the more hours people use the Internet, the less time they spend with real human beings. (Nie and Erbring 2000, p. 1) 1. The lack of evidence of strong effects has been blamed on measurement issues, the inadequacy of research models, and limited sample size (Zaller 1996, 2002). However, the jury is out on the extent to which these factors uniquely limit evidence of persuasive effects. For example, many measures of media exposure are no worse than other commonly used survey items in other research areas (see Dilliplane, Goldman, and Mutz 2010; Bartels 1993). To the extent that they are, there is no reason why media effects should not be held to the same standards of empirical evidence as other theories. 1020 Mutz and Young at U niersity of Penylvania L irary on N ovem er 9, 2012 http://poqrdjournals.org/ D ow nladed from One cannot mistake the strong echo of mass society themes in this recent literature. The feared negative effects follow familiar themes involving the production of a socially isolated and atomized population, and the decline of a quality public sphere. As evidenced by books with titles such as The Future of the Internet and How to Stop It and The Filter Bubble: What the Internet Is Hiding from You, our current digital era appears to be no different in this regard (see Zittrain 2008 and Pariser 2011, respectively). Likewise, some scholars have characterized the current era as a ‘‘neo-dark age’’ fostering social fragmentation (through niche programming); realignment to narrow, parochial interests; lack of social cohesion; cultural stasis; and a decline in civic engagement (e.g., Kornegay 2009). Few have predicted a far more modest impact (but, for an important exception, see Neumann 1991). Perhaps these fears will prove to be similar to initial fears of other new media— that is, somewhat overblown—perhaps not. Regardless, now as in the past, a great deal of the panic emanating from ‘‘new media’’ critics is borne of fear rather than evidence. In many ways, the role of research on communication and public opinion has been largely one of complicating simplistic initial assumptions. Complex findings do not cater to punchy headlines and thus seldom receive the same level of attention as apocalyptic warnings. Nonetheless, researchers trudge onward, adding to our general understanding of the role of communication in the formation and change of public opinion, even when their more modest conclusions do not generate headlines. In this essay, we produce a similarly mixed picture of the combined impact of these technological changes—including both the Internet and cable television—on public opinion. Diversity: Be Careful What You Wish For Twenty-five years ago, a dominant theme in communication researchers critique of the news environment of that era was its political blandness. With television carrying the mantle of ‘‘most important medium’’ for political communication, the focus was on evening news broadcasts and the limited range of political perspectives they provided. Indeed, the nightly news programs of ABC, CBS, and NBC were widely criticized for being largely the same broadcasts covering the same narrow range of issues from the very same angles. Interestingly, professional journalistic norms promoting objectivity were perceived to be at the root of this problem. Although the motive behind these practices was acknowledged to be avoidance of bias and adherence to ‘‘just the facts,’’ by avoiding partisanship, journalists were accused of skirting political argument altogether. Many scholars of that era were extremely critical of the lack of opinion incorporated in the news. As Bennett (1996) argued, ‘‘If anything, the press is guilty of taking its claims of objectivity or fairness too seriously and, therefore, failing to recognize that these professional standards boil down to an extremely passive stance toward news’’ (p. 163). Communication and Public Opinion 1021 at U niersity of Penylvania L irary on N ovem er 9, 2012 http://poqrdjournals.org/ D ow nladed from Likewise, many political scientists of this era pushed for greater partisanship and political argument. The American Political Science Association Committee on Political Parties (1950) advocated a ‘‘responsible two-party system,’’ that is, ‘‘political parties which provide the electorate with a proper range of choice between alternatives of action’’ (emphasis in original, p. 15). At the time, political parties were heavily criticized for offering insufficiently distinct political alternatives. The major parties were deemed Tweedledee and Tweedledum, leading to a situation in which ‘‘alternatives between the parties are defined so badly that it is often difficult to determine what the election has decided even in its broadest terms’’ (pp. 3–4). Coalitions that cut across party lines were to be avoided because they would ‘‘deprive the public of a meaningful alternative’’ (p. 19), and scholars regularly eschewed the lack of party loyalty among politicians. When it came to the press, communication scholars of the previous era often espoused a romantic longing for the good old days, that is, the political press prior to the 20th century, when most newspapers were actively associated with and advocated on behalf of political parties. Instead, corporate profits now trumped a desire for political expression. As Bagdikian (1983) lamented in his classic critique of corporate ownership, nowadays media ‘‘aren t started with the desire of someone to express what he believes; [instead], they become bland to avoid controversy’’ (pp. 112, 85). Likewise, editors were chastised for being ‘‘too quiet, too bland, . . . too safe’’ (Diamond 1993, p. 393). Rather than serving the interests of fairness and avoidance of bias, objectivity itself was often seen as an impossible ‘‘double standard,’’ a ‘‘chimera’’ that simply reinforced the status quo. Professional journalists were dubbed ‘‘the best and the blandest’’ (Diamond 1986, p. 18), and a press with known bias or partisanship was seen as infinitely preferable to one with undefined, surreptitious politics. Anything was surely better than apolitical pablum: In the early days of the American republic, the news was anything but objective. Most newspapers were either funded by, or otherwise sympathetic to, particular political parties, interests, or ideologies. Reporting involved the political interpretation of events. People bought a newspaper knowing what its political perspective was and knowing that political events would be filtered through that perspective. In many respects, this is a sensible way to approach the news. If one knows the biases of a reporter, it is possible to control for them in interpreting the account of events. Moreover, if reporting is explicitly politically oriented, different reporters can look at the same event from different points of view. Finally, since political events generally convey political messages, an overtly political reporting style is more likely to draw these messages out than to let them slide by unnoticed (with the risk that they might pass for broad, non-partisan perspectives). (Bennett 1996, p. 147) 1022 Mutz and Young at U niersity of Penylvania L irary on N ovem er 9, 2012 http://poqrdjournals.org/ D ow nladed from From this discussion of the past, it should be clear that while the general diversity of political voices has always been valued by students of political communication, there has been little agreement on what kind of press best serves that end. The kind of diversity that critical scholars had in mind 25 years ago was one incorporating multiple advocacies rather than professional neutrality. Instead of an emphasis on facts and objectivity, the diversity they wanted was one of antagonistic vested interests, one in which reporters were advocates for their positions without any pretense of neutrality. This view of diversity is probably closest to the one John Stuart Mill had in mind when advocating a marketplace of ideas: . . . so long as popular truth is one-sided, it is more desirable than otherwise that unpopular truth should have one-sided assertors too; such being usually the most energetic, and the most likely to compel reluctant attention to the fragment of wisdom which they proclaim as if it were the whole. (Mill 1859, p. 67) Whether Mill s conception of the marketplace works best remains to be seen; some formal models suggest that the public as an aggregate will be less informed by this system than by a monopoly with politically moderate reporting (e.g., Stone 2011). Regardless, the recent proliferation of channels on cable television seemed to promise precisely the possibility of many opinionated voices. And the Internet offered a megaphone (or at least a microphone) to even more people to espouse political arguments. Given the suspicions with which scholars regarded the professionalized ‘‘objective press’’ of that era, one might assume that by incorporating more partisan voices, contemporary political television would be viewed as a genuine step in the right direction. After all, today no one complains that Fox News, MSNBC, and the various networks all look alike in their coverage, or that people with radical viewpoints lack ways of communicating with their compatriots. Pleas for more politicized news content are rare indeed. Nonetheless, the current media environment has not been heralded as an unabashed improvement either. Instead, the rise of partisan media outlets and proliferation of channels of communication has given rise to an entirely new set of concerns. The political media currently on offer to citizens include a panoply of voices, both partisan and otherwise. What is perhaps most impressive is the sheer amount of information and opinion one can access if one is so inclined. The old-style network news programs remain, but have been joined by myriad styles of political and public affairs news, including roundtable discussions, talking (and screaming) heads, interview programs, and comedy news, to name just a few. Indeed, asking the average citizen whether he or she watches, reads, or listens to ‘‘news’’ these days is the classic example of a bad survey question because the very definition of what constitutes ‘‘news’’ is in flux. Because scholars have yet to come to grips with all of Communication and Public Opinion 1023 at U niersity of Penylvania L irary on N ovem er 9, 2012 http://poqrdjournals.org/ D ow nladed from these recent changes, we know little about where people are getting their exposure to political information and argument, and whether the source makes any difference. Two concerns have received the bulk of attention with respect to the huge increase in media choice afforded to U.S. citizens. The first is that the choice between political and nonpolitical media will lead to an increasing gap between the political ‘‘haves’’ and ‘‘have-nots.’’ In this case, the choice that is relevant is between news and public affairs content as opposed to purely entertainment programs. A second form of choice that raises eyebrows for entirely different reasons is the choice among different sources of political and public affairs news. CHOOSING AMONG GENRES For many Americans, the central choice is whether to pay attention to political media at all. As Prior (2007) points out in Post-Broadcast Democracy, politics is not inherently interesting for the bulk of the American public. When a representative sample of respondents was randomly assigned to choose among either the traditional early evening broadcast options or not watching at all, around 80 percent chose one of the broadcast news options, and 20 percent chose not to watch anything. But when an equivalent sample was given those same options along with the broader selections available to cable subscribers (e.g., a sitcom, drama, science fiction, reality show, or sports), only 35 percent chose a news program, and 9 percent chose not to watch anything. Not surprisingly, the majority chose non-news programming. The implication, according to Prior (2003), is a smaller inadvertent audience for news; given the choice, citizens who are not particularly interested in politics will opt out and watch sitcoms or dramas instead, decreasing their exposure to political information. Less incidental exposure lowers political knowledge and participation among those only marginally interested. As such, the diversity of available genres does not appear to serve democratic interests. On the other hand, the proliferation of programs within the soft-news, ‘‘infotainment’’ genre has meant that more people are exposed to political content through programming outside traditional news (Baum 2002, 2003a, 2003b). The net impact of these two forces is unclear. But, to be fair, this is not the kind of diversity of choice that most political theorists have in mind when they highlight the benefits of a marketplace of political ideas. It is the diversity of political ideas, rather than genres, that is of concern. CHOOSING AMONG POLITICAL NEWS SOURCES Choice has increased not only among political and nonpolitical media, but also within the realm of political news sources. The diversity of perspectives offered is now broader than in the broadcast era. Most observers and scholars concur 1024 Mutz and Young at U niersity of Penylvania L irary on N ovem er 9, 2012 http://poqrdjournals.org/ D ow nladed from that the partisan diversity of political information sources has also increased relative to the broadcast era. In this case, however, the main scholarly concern is that choices will be based on pre-existing prejudices and partisanship, thus leading to news exposure that strictly reinforces pre-existing political views. Reinforcement was viewed by Lazarsfeld et al. as ‘‘a lower-order affair compared to persuasion or mobilization’’ (see Gitlin 1978, p. 216); it was akin to no effect at all. Nowadays, however, reinforcement is at the heart of scholarly concern about the impact of new media. Two possible consequences are at the root of most hand-wringing over the fragmentation of news exposure along partisan lines. The first concern, voiced most strongly by Sunstein (2007), is that the public will lack common ground to use as a basis for discussion and debate. According to this line of thought, the increased choices offered by Internet news and cable television have combined to deprive the American public of shared experiences, the kind previously provided by general-interest media sources without strong partisan identities. If audiences are receiving entirely different versions of the news of the day, they may have difficulty discussing issues of mutual concern. A second concern stemming from partisan fragmentation of news audiences is the fear that modern communication technologies will fuel extremists fires. There are more diverse political voices in the marketplace, to be sure, but to the extent that people expose themselves exclusively to one set of those voices— those with whom they already agree—then political views can be expected to polarize, and thus governing becomes more difficult, and extremism more likely. By exposing themselves strictly to echoes of their own voices, partisans may become less willing to compromise, and less likely to consent to being governed by another faction. Importantly, based on the scenarios described above, the perils of a politically diverse communication environment are assumed to be driven not by the kind of political media on offer (the concern of the previous generation), but by the actions of individual citizens whose choices among plentiful sources paradoxically lead them to less, rather than greater, political diversity of exposure. What is less clear from evidence accumulated to date is (1) whether those politically interested enough to be motivated to choose likeminded political content are exposed exclusively to likeminded content; and (2) how much of this selection process is driven by ongoing, active choice on the part of the individual. We turn next to address these issues in the context of the long history of research on selective exposure.