54 Broadcasting versus Narrowcasting: Do Mass Media Exist in the Twenty-First Century?

Wither mass communication? In a 2001 article titled “The End of Mass Communication?” Chaffee and Metzger observed that a profound change was occurring such that the defining features of mass communication that had been in place since the early twentieth century were being undermined by significant changes in the communication technologies used to produce, disseminate, and exhibit media content in the twenty-first century. Specifically, whereas “mass communication” of the past could be characterized in terms of the production of broad-appeal content by a handful of large and powerful media institutions, and by little content choice at the audience level due to a limited number of available channels and formats, contemporary media present a vastly different environment to media consumers. The “new” media environment not only offers a massive amount of information from a multitude of sources that can transmit their content over many channels but also places more control over both content creation and selection in the hands of audience members themselves. These changes precipitated a shift from mostly broad-appeal programming to a sizable portion of content produced to appeal to more narrow audiences. As a result, Chaffee and Metzger (2001) argued that contemporary media are “demassifying” mass communication.

Demassification is particularly evident in recent media production and consumption patterns. The one-to-many model of media content production that has traditionally defined “mass” communication and set it apart from other forms of communication now finds itself having to accommodate newer models of many-to-many communication. Opportunities afforded to users of digital media for producing and distributing their own media content to large audiences (i.e., “user-generated content”) undermine the long-held assumption that large organizations are required to produce and distribute media content at a mass scale. In other words, whereas mass media producers of the past were “big and few,” today they are also “small and many” (Chaffee and Metzger, 2001).

Moreover, the enormous popularity of Web-based platforms such as YouTube, blogs, and other forms of digital and social media may divert audience attention from media content delivered via traditional media channels. Indeed, there is evidence to show that this is happening in the realm of news and political information. Data from a recent State of the News Media annual report, for example, show that while digital platforms for news are gaining in audience share, all other mass media sectors—including cable news for the first time—are experiencing declines in their audience numbers (Rosensteil and Mitchell, 2011). And, as the plethora of choices expands, audiences will continue to be stretched thinner across all the options of media sources and channels. In sum, then, the two key defining characteristics of mass communication historically, namely the size of mass media producers and the size of the audience for any particular media content, appear to be “demassifying” in large part as a result of changes in the communication technologies used to produce and deliver media content.

This chapter examines these changes in the media landscape to address the question of the continuing relevance of mass media at the dawn of the new millennium. It begins by tracing the history of media content production, distribution, and consumption first from broadcasting to narrowcasting, and then to more recent trends toward “hyperpersonalization” in the digital media environment. The chapter then focuses on what these changes mean for politics and for political communication theory. It concludes by attempting to answer the question “Do mass media exist in the twenty-first century?” and by posing some questions about the future of mass media that are intended to serve as a call for research into the changing nature, circumstances, and effects of mass communication in the contemporary media environment.

The original “broadcast model” of media content production to mass audiences was a rational response to early mass media systems’ economic and physical resource scarcity. Under such constraints the best way to attract the largest portion of the audience share was to produce the programming with the broadest appeal. But a significant effect of the explosion of channels made available by digital networked communication technologies in the waning decades of the twentieth century has been to diffuse the audience for any particular media product, as discussed earlier. The result, which again was a rational response to changing market conditions, has been toward increasingly niche-oriented media content delivered over increasingly specialized channels (Owen and Wildman, 1992). This trend away from “broadcasting” and toward “narrowcasting” (see Massey, 2004) and its sociopolitical impacts has been discussed by several scholars (e.g., Bennett and Iyengar, 2008; Mendelsohn and Nadeau, 1996; Prior, 2007; Ranney, 1990; Sunstein, 2001, 2007, 2009; Turow, 1997).

Although the shift from broadcasting to narrowcasting is evident across several genres of media content, news provides a particularly interesting case study. In the heyday of traditional mass communication in the twentieth century, major media outlets, including the leading urban newspapers and television network news broadcasts, dominated the news audience share in most US media markets. For example, the audience for the three major network newscasts (ABC, CBS, and NBC) in the mid-twentieth century typically reached over seventy million Americans (Prior, 2007). That has changed dramatically, with only about a fifth of Americans now tuning in on an average weeknight (Rosensteil and Mitchell, 2011). Although data show steady declines in audiences for network news over the last thirty years, during which time the networks lost 55.5 percent of the share of viewership they once had, this does not mean that most people have abandoned news altogether. Instead, they appear to be gravitating to other news platforms (Rosensteil and Mitchell, 2011).

The appearance of cable and satellite technologies in the 1980s and 1990s offered the first serious alternatives to broadcast network news, and both these and more recent digital networked communication technologies accelerated the decline in both television news viewership and print newspaper readership. Perhaps more importantly, however, the fracturing of audiences for traditional news outlets prompted product differentiation and experimentation with new forms of news,¹ including more partisan news outlets (Bae, 1999, 2000; Iyengar and Hahn, 2009). Fox News is the most obvious example, and Iyengar and Hahn (2009) argue that Fox News emerged in part by audience fragmentation resulting from channel proliferation in the 1990s. Mullainathan and Shleifer (2005) further show that there is economic incentive for ideological specialization in news products, as audiences tend to prefer information that confirms their beliefs, and media organizations that respond to audience demand stand to reap greater profits. In other words, niche news is an effective competitive strategy in an increasingly crowded news marketplace.

The rise of media channel proliferation and narrowcasting increased audience choice in media content and, consequently, audience members’ opportunities for selective exposure (Chaffee and Metzger, 2001; Iyengar and Hahn, 2009). Although niche media (e.g., radio, magazines) predate cable, satellite, and Internet media technologies, and while audience selectivity has been observed with traditional mass media content as well (see Sears and Freedman, 1967 for a review), the appearance of partisan news affords greater opportunity for a particular type of audience selectivity—selective exposure to ideologically congenial news and public affairs information.

Several scholars argue that this type of selectivity may significantly reduce citizens’ exposure to political difference (see, for example, Bennett and Iyengar, 2008; Sunstein, 2001). Also, while narrowcasting and increased consumer control over media exposure reduced the size of the available audience share for all types of programming, demand for mainstream news information, which used to provide an “information commons” or shared context for receiving public affairs information, has shrunk over the years, particularly among younger audiences (Bennett and Iyengar, 2008).² Even among politically interested citizens, niche media provide attractive alternatives to traditional news outlets, while facilitating politically uninterested audiences to tune out completely. Bennett and Iyengar (2008, 707) consequently question whether the concept of mass media has been “made obsolete by audience fragmentation and isolation from the public sphere.” For many of the same reasons, Chaffee and Metzger (2001) similarly wondered whether history will show that “mass media” was a purely a twentieth-century phenomenon.

The latest development in the move from broadcasting to narrowcasting, coupled with niche news and the expansion of opportunities for selective exposure afforded by modern media technology, is what Bennett and Iyengar (2008, 723) call “the personally mediated society.” Similar to Negroponte’s (1995) concept of the “Daily Me,” which refers to news content that is customized according to individuals’ personal tastes, Bennett and Iyengar argue that the diversified information environment allows individual news consumers to limit their news information to, for example, include only those sources and perspectives that share their ideological views. Indeed, Turow’s historical analysis of media marketing strategies shows a progression toward what he calls the “hypersegmentation” of audiences by media firms and advertisers in order to take better advantage of the narrowcasting opportunities made possible by channel proliferation in the contemporary media environment (Turow, 1997). Thus, whether it is by their own hand via selective exposure, or by the hand of media firms and their advertisers via target marketing, media audiences are increasingly likely to receive information that is customized to their personal tastes, interests, and political viewpoints to the possible exclusion of other information.

The personally mediated society may be exacerbated by Web 2.0 applications, such as social media including social networking platforms (e.g., Facebook), social bookmarking sites (e.g., Digg or Reddit), and political blogs. These platforms and applications allow for a new level of customization that is based on personal interests that we both consciously and inadvertently express via our activity on the Web, including the sites we visit, the search terms we use, the interests and actions we disclose in online social networks, and even those of our peers with whom we are connected via our various online communities. Writing about this development, Sunstein (2007, 4) comments that:

Negroponte’s prophecy was not nearly ambitious enough. As it turns out, you don’t need to create a Daily Me. Others can create it for you. If people know a little bit about you, they can discover, and tell you, what “people like you” tend to like—and they can create a Daily Me, just for you, in a matter of seconds.

 Pariser (2011) similarly argues that as algorithmic gatekeepers and recommendation systems replace human ones, each communication we receive will be chosen in advance based on algorithmic filters, which themselves are based on the traces that we leave of ourselves via our online actions. This creates a form of “hyperpersonalization” of news and other media content, as well as a new form of preselected selective exposure that is outside the direct control of audience members. A case in point is that the exact same Google search performed by two people now returns quite different results. Moreover, Facebook envisions a future in which all of the information we receive online is filtered through our personally tailored social networks such that, rather than querying a search engine like Google as we seek advice on what camera to buy or dentist to see, we will instead query our “social graph,” which consists of our network of friends, colleagues, peers, and family with whom we are connected online (Vogelstein, 2009). Given that humans tend to associate with others who share similar attributes and attitudes (e.g., Byrne, 1971), this may further limit the diversity of information that people are exposed to and create an echo chamber effect as they engage with information and others online.

An important question stemming from these trends asks what these shifts in the media landscape mean for politics and for political communication theory? As it turns out, there is considerable research to help answer this question, but it is not without controversy. Two major areas of scholarship pertaining to the first part of this question are whether narrowcasting leads to greater audience selective exposure, and whether selective exposure will foster political polarization that serves to undermine democracy. Another area of scholarship addresses the second part of the question by examining whether new forms of media content (e.g., niche news) and patterns of their consumption necessitate entirely new theorizing and expectations about media effects on political outcomes. These issues are discussed next.

Several scholars have written about the impact of selective exposure due to the rise of narrowcasting and hyperpersonalization of media content on democracy, arguing that exclusive attention to like-minded others exacerbates rifts in society between ideologically divergent groups, and thus impedes the cooperation that is needed to address the complex problems facing modern society (e.g., Mendelsohn and Nadeau, 1996; Stroud, 2010; Sunstein, 2001, 2007, 2009). Stroud (2010), for instance, looked at news consumption behavior over time, and found strong evidence that repeated selective exposure to attitude-consistent information resulted in increased political polarization. Sunstein (2007) argues that selective exposure to ideologically congenial information undermines the preconditions for a well-functioning democracy, which include exposure to political difference and a shared information commons. Without these, he says, common understandings between groups will be more difficult to achieve and society will have a much harder time in addressing its social problems. Others have similarly discussed the implications of channel proliferation on creating political attitude extremity and partisan echo chambers (e.g., Jamieson and Cappella, 2009).

Of course, whether the most drastic political implications are ever realized is contingent on the extent to which narrowcasting encourages audience members’ selective exposure to exclusively attitudinally congruent news information in the first place. This topic has received a good deal of attention over the years. While early research on selective exposure in general found weak or mixed evidence for the phenomenon (see, for example, Kinder, 2003; Sears and Freedman, 1967), more recent empirical work provides strong evidence for its prevalence today (Iyengar and Hahn, 2009; Stroud, 2007, 2008, 2010, 2011; Tewksbury, 2005). Although this research has largely settled the debate over people’s tendency to selectively attend to attitude-consistent political information, disagreement concerning whether they actively avoid attitude-discrepant information—which is an important component of normative views of the effects of selective exposure—continues. Research by Stroud (2010) and by Iyengar and Hahn (2009) showing evidence of selective avoidance is countered by studies showing no evidence for the phenomenon (e.g., Garrett, 2009a, 2009b; Kobayashi and Ikeda, 2009; Webster, 2007). Holbert, Garrett, and Gleason’s (2010, 22) overview of this literature concludes that despite having a preference for congenial information, people do engage with discrepant viewpoints under certain circumstances. Thus, these scholars conclude that “selective exposure and encounters with attitude-discrepant information can co-exist.”

There are also questions about the prevalence of selective exposure and avoidance across the political spectrum. One of the circumstances under which selective exposure appears to occur is with news consumers who possess strong political attitudes (Brannon, Tagler, and Eagly, 2007). Indeed, Knobloch-Westerwick and Meng (2009) found that partisanship interacts with selectivity. In their study, although people chose attitude-consistent sources of news information, and spent more time reading news from these sources than news from counter-attitudinal sources, this effect was more pronounced for people who felt strongly about their beliefs. Evidence for selective avoidance by stronger partisans is also seen in the blogosphere, where studies have found a tendency for blog readers to avoid those that challenge their ideological views (Johnson, Bichard, and Zhang, 2009), and for bloggers to avoid linking to political videos that challenge their ideological positions (Wallsten, 2011). The question of whether selective avoidance operates only or primarily among strong partisans, while the majority of citizens receive a fairly balanced diet of attitudinally consistent and inconsistent information, is important because this would weaken claims that narrowcasting will necessarily have drastic negative repercussions for democratic society.

Finally, there is debate about the effects of selective exposure on political participation. Stroud (2011) concludes from her research that selective exposure influences how average people engage with politics. She argues that on the one hand, citizens may become increasingly polarized as a result of using media that coheres with their political beliefs, causing them to experience political frustration and apathy, and perhaps culminating in decreased voter turnout. On the other hand, however, she also finds that partisan selective exposure may encourage participation at the individual level by encouraging political engagement. Sunstein (2009) similarly argues that political extremism can be good in that it can bring like-minded individuals together, spurring heightened political involvement and collective action. As the title of his book suggests, like minds can both unite and divide.³

In addition to debates on the extent to which selective exposure exists and leads to political polarization, there is contention over whether the shifting foundations of modern mass communication media are creating fissures in our central and long-held theories of political communication effects (Bennett and Iyengar, 2008; Chaffee and Metzger, 2001; Holbert et al., 2010; Metzger 2009; Neuman and Guggenheim, 2011). This first took the form of Chaffee and Metzger’s (2001) proposition that several of our core media effects theories are challenged by demassification brought on by digital networked media. Metzger (2009) further suggested media effects theories that assume mass exposure to relatively uniform content (e.g., exposure to a limited set of news information and perspectives emanating from three dominant broadcast networks), such as agenda setting, cultivation theory, and the spiral of silence, are either untenable in the contemporary media environment or that their basic premises should be seriously reexamined.

The debate over media effects in the new media landscape has centered more recently on whether the effects of political communications in this environment are likely to be minimal or substantial. Bennett and Iyengar (2008) argue that the trend toward narrowcasting is likely to produce only minimal effects of political communication on audiences, at least in the realm of persuasive media effects. This is true for several reasons, they say. First, channel proliferation encourages selective exposure to attitudinally congruent media content (and selective avoidance of dissonant information), which serves to reinforce rather than change news consumers’ preexisting attitudes. In addition, given most people’s preference for entertainment media content, all but the most politically interested citizens will be exposed to less news information overall and so collectively, the electorate will know less about and thus participate less in politics. Third, it will be difficult for survey researchers to prove that audiences have been persuaded by a media message when those audiences self-select messages that confirm their views, and media audiences will be more likely to resist messages that challenge their preexisting attitudes. This means that reinforcement effects will be found even among those who are exposed to counter-attitudinal information. As an example of the last point, they discuss survey data showing that many Republicans continued to believe that Iraqi President Saddam Hussein possessed weapons of mass destruction even after receiving overwhelming information to the contrary.

While Holbert et al. (2010) agree with much of Bennett and Iyengar’s analysis, they disagree with their conclusion. They instead argue that recent changes in the media environment facilitate “pull” rather than “push” media, which means that those who do choose to expose themselves to political information in high-choice media environments may be more motivated to process that information, and thus will be affected by it. They rely on theories of persuasion such as the Elaboration Likelihood Model, which show that, under conditions of high motivation to process information (i.e., central route processing), people are more receptive to influence, and that central route processing produces attitude change that is more permanent and resistant to counter-persuasion attempts. Thus, they argue that the effects of political communication in the new media environment may not be so minimal after all.

Perhaps most important, these discussions have called for new theorizing, as well as for refinement of existing theory, to accommodate serious changes in the contemporary media landscape. This stands as the field’s greatest challenge at the moment. The way forward will include devising new methods and measures of media exposure, as well as considering carefully what has and has not changed about both media content and audience experience when theorizing about effects in the altered media landscape. While some effects may be diminished by the expansion of media choice, others will be enhanced, and most will likely change in interesting ways. For example, although mainstream media’s ability to set the public agenda may wane as people turn to individually customized news diets, the power of the public to set the media and policy agendas via blogs and user-generated news may increase. In addition, peer-to-peer agenda setting via online social networking and other forms of social media becomes an interesting and important area to study.

Similarly, as technology increasingly enables media audiences to connect with those who share similar views, mechanisms driving the spiral of silence may disappear. At the same time, however, “reinforcement spirals” may take their place and result in stronger in-group allegiances and out-group suspicion (Slater, 2007). Alternatively, new venues for spirals of silence to emerge may appear, particularly within social media environments such as Facebook, whose usage is often motivated by a desire for popularity and where pressure to conform to peer opinion may be quite powerful (see also Metzger, 2009; Weimann, Weiss-Blatt, Mengistu, Tregerman, and Oren, 2014; Perloff, 2015 for fuller discussions of these and other theoretical extensions to traditional media effects theories due to technological changes).

The discussions of the future media landscape and its effects on audiences imply that a dichotomy exists between broadcast and narrowcast content by suggesting that narrowcasting is replacing general-appeal content. Some even imply that narrowcasting and hyperpersonalization threaten the very survival of the “mass” media. These concerns are overblown. The reality is that the contemporary media environment can and does include both specialized and mass appeal content. Also, as mentioned earlier, while partisans may gravitate more readily toward one-sided news, the majority of the audience so far does not avail itself of niche news, and typically prefers more “objective” journalism found in traditional and online versions of mainstream newspapers and nonpartisan cable or television news stations (Takeshita, 2006). Also, only about 9 percent of Americans obtain news via blogs regularly, and they tend to supplement these sources with other news sources as well (Pew Research Center, 2010). Indeed, many people consider blog credibility to be suspect. Consequently, demand still exists for mass-appeal news and political information, and economic incentives for media firms to produce broad-appeal programming are still in place, even in multiple channel markets (Waterman, 1992).

In fact, setting newspapers aside, most mass media organizations (e.g., traditional broadcast and cable television firms such as CBS or CNN) are in many ways bigger and more powerful than ever before, and they now venture far outside the national borders that once contained them. They have also managed to escape the delivery channels that used to bound and distinguish them, now often offering the same content delivered over a variety of print, broadcast, cable, and digital channels. When readership of their online versions is factored in, the major newspapers, television, and cable news outlets still have a stranglehold over content production and command a large percentage of the audience share for news (Pew Research Center, 2010). At the same time, mainstream media are incorporating narrowcasting into their products in several ways, including developing their own blogs, affording opportunities for user commentary and user-generated news on their websites (e.g., CNN’s iReport), and maintaining a presence in social networking sites. In other words, broadcast and narrowcast content can and do coexist. An interesting question is whether this new hybrid form of news offers the best scenario for democracy by offering greater and more enticing chances for political engagement while providing enough of an information commons to avoid polarization?

So, one thing that is clear is that mass media do still exist in the twenty-first century and are likely to survive for quite some time. They are, however, changing as a response to emerging information technologies, and mass media news firms of the future may ultimately bear little resemblance to what they looked like at the beginning of the twenty-first century. For example, as news consumers have become agnostic about the channel through which they receive news information, consuming broadcast, cable, or print news content all via the same digital medium, news organizations have been forced to abandon their channel-specific formats by offering a mix of text (print) and video (broadcast, cable) journalism, as well as interactive formats of news information and commentary through their websites. This has created considerable duplication in the industry that now threatens the survival of even the leading news brands. To achieve efficiencies, one possibility is that search engines such as Google, Microsoft’s Bing, and Yahoo! could replace the former channel-specific news monoliths such as CBS News, CNN, and the New York Times to become the mainstream news brands of the future.

Another thing that is clear is that communication scholars must change their definitions of mass communication to better describe what is happening in today’s media landscape. As Napoli (2008, 2) contends, mass communication still has relevance today only if it is redefined with a new interpretive approach that is not exclusive to institutional communicators and “that allows the term ‘mass’ to extend to both the senders and receivers of messages.” In other words, whereas mass communication was traditionally defined by the number receivers of a message, digital forms of distribution allow for masses to communicate with masses. So, while Napoli’s suggestion retains the one-to-many aspect of mass communication at its core, it enables the term “mass” to describe both the number of receivers and the number of sources in the definition of mass communication. Digital networked media necessitate this more expansive definition of mass communication.

As the previous sections demonstrate, the rapid and dramatic shifts in the media environment that have taken place since the 1990s beckon reconceptualizations of mass communication and mass media effects. Many challenges loom for political communication scholars seeking to understand the scope, nature, and impacts of the new forms of mass communication and media content available to audiences. The following questions represent some starting points in this line of inquiry:

This chapter posed the question of whether mass media will continue to endure in an environment of vast choice and increased audience control over both media content production and its reception. Implications of the shift from broadcasting to narrowcasting, and to hyperpersonalization of content, invite political communication scholars to reexamine long-held theoretical assumptions about whether, where, and how media effects are likely to occur and to devise new methods to study them. While the chapter made clear that mass media will likely endure, it also raised new questions that will have to be addressed as scholars strive to understand the nature and effects of mass communication in the future.