new

The One-Step Flow of Communication

Author(s): W. Lance Bennett and Jarol B. Manheim

Source: The Annals of the American Academy of Political and Social Science , Nov., 2006, Vol. 608, Politics, Social Networks, and the History of Mass Communications Research: Rereading Personal Influence (Nov., 2006), pp. 213-232

Published by: Sage Publications, Inc. in association with the American Academy of Political and Social Science

Stable URL: https://www.jstor.org/stable/25097863

REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/25097863?seq=1&cid=pdf- reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

and Sage Publications, Inc. are collaborating with JSTOR to digitize, preserve and extend access to The Annals of the American Academy of Political and Social Science

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

The One-Step Flow of

Communication

By W. LANCE BENNETT

and JAROL B. MANHEIM

This analysis explores the transformation of public communication in the United States from a two-step flow of messages passing from mass media through a social mediation process, to a one-step flow involving the refined targeting of messages directly to individu als. This one-step flow reflects both a transformation in communication technologies and fundamental changes in the relations between individuals and society.

Opinion leaders who played a pivotal role in the two step paradigm are increasingly less likely to "lead" because they are more likely to reinforce latent opin ions than to reframe them. And because the mass

media in the one-step flow are increasingly fragmented and differentiated, they contribute to the individualiz ing process through shrinking audiences, demographi cally driven programming, and transmitting targeted political advertising and news spin.

Keywords: one-step flow; two-step flow; strategic communication; campaign; social networks; targeting; narrowcasting; data mining; persuasion; public information

Fifty years ago, the Free Press published Elihu Katz and Paul F. Lazarsfeld's (1955) soon-to-be-classic study, Personal Influence: The Part Played by People in the Flow of Mass Communications. Though the concept had ear lier roots (Berelson, Lazarsfeld, and McPhee 1954, especially chap. 6; Lazarsfeld, Berelson, and Gaudet 1944), it is to this volume that con temporary scholars turn for the classic state

ment of the "two-step flow of communication" hypothesis. In the first step, messages are issued by the mass media to what is, to all out

ward appearances, a more or less homogeneous mass audience. In the second, innumerable small group interactions powered by horizontal opinion leaders interpret and contextualize these mediated messages for their participants,

who then internalize the resulting content. The end result is a more or less differentiated understanding of the message across various social boundaries. This was a nontrivial finding.

For society at large, one implication was that there were limits to the potential influence of

DOI: 10.1177/0002716206292266

ANNALS, AAPSS, 608, November 2006 213

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

214 THE ANNALS OF THE AMERICAN ACADEMY

mediated communication?limits set by each individual's knowledge networks and social interactions, some of which might be systematic in character, others products of chance. Yet at the same time, mediated communication was poten tially so integral to interpersonal communication?providing it with both stimuli and information?that the two might no longer be distinguishable. This meant that mass society could never be entirely homogeneous but also that individual social locations could never be entirely isolated from mass influence. Even in his recent work, Katz (Kim, Wyatt, and Katz 1999) has continued to argue the cen tral importance of personal conversation?even the most casual of talk?in the democratic process. Indeed, he and his coauthors viewed such discourse as the very definition of democracy (Kim, Wyatt, and Katz 1999, 361).

For political communication professionals?a class of operatives that was only beginning to develop at the time, but was destined to grow into a determinative force in American politics?the implications were similarly consequential. For they suggested that effective political communication?by which we mean influence? required not merely the management and control of mediated messages but also understanding and controlling how these messages would be processed through social interaction before their effects would become manifest. In other words, it became apparent that the so-called "water cooler effect"?by which mass mediated messages reach audience members who were not directly exposed to them through secondary interactions with friends and colleagues?was not merely a means of expanding the audience for a given message but was a potentially success-critical mechanism for assigning it meaning. Perhaps more than any other single research finding, this realization led communicators to incorporate the study of opinion dynamics with that of message effects in developing their art. The heavy reliance we see today on integrated programs of focus group research and public opinion polling in any effort at mass political persuasion is a direct outgrowth of that realization.

But that was then; this is now. Our thesis is that society, communication tech nologies, and individual communication habits have changed fundamentally in ways that affect how individuals receive and process information. These social and technological changes directly challenge the underlying assumptions of the two step flow hypothesis because they have isolated increasing numbers (though surely not all) of today's citizens from the very groups that traditionally provided vital cues

W. Lance Bennett is a professor of political science and Ruddick C. Lawrence Professor of Communication at the University of Washington, where he also directs the Center for Communication and Civic Engagement (www.engagedcitizen.org). The general focus ofhis work is with how communication processes affect citizen engagement with politics. Publications include Mediated Politics: Communication in the Future of Democracy (Cambridge) and News: The Politics of Illusion (Longman). He has received the Ithiel de Sola Pool distinguished career award and lectureship and the Murray Edelman Career Achievement Award in Political Communication, both from the American Political Science Association.

Jarol B. Manheim (PhD, Northwestern, 1971) is a professor of media and public affairs, and of political science, at George Washington University, where he was the founding director of the School of Media and Public Affairs. His research interests are in the area of strategic political communication. His work has appeared in the leading journals of political science, journalism, and mass communication.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 215

for interpreting information. The combination of social isolation, communication channel fragmentation, and message targeting technologies have produced a very different information recipient than the audience members ofthe Eisenhower era.

Even as individuals have become less likely to participate in groups, they have gained greater command of their own information environments, often partici pating in multiple, fluid social networks oriented to self-expression, generally organized around lifestyles (Giddens 1991; Bennett 1998). These networked individuals are hard to reach in large numbers with effective messages, as evi denced by the spiraling costs of marketing campaigns in both politics and consumer arenas. At the same time, these networks are often worth the expense of targeting as evidenced by the efforts to reach so-called "soccer moms" in the 2000 elections and "Nascar dads" in 2004. Given the decline of group identifica tions and loyalties, individuals, even when reached, are hard to hold, whether in terms of attention or adoption of message content. Witness the rise of so-called permanent campaigns aimed at branding individuals to products, ideas, candi dates, leaders, and policies.

[Communicators . . . will. . . have substituted their own audience selection and

targeting skills for the role formerly assigned to peer group interaction. This is the

one-step flow of communication

In these and other respects, today, the communication process is aimed at the individual or at the direct messaging of assembled networks of like demograph ics. The mass context is still important, but more as an affirming echo chamber than as a social cueing system. The technologies of focus grouping and polling are still relevant, but they are increasingly augmented or supplanted by massive data bases aimed at identifying and characterizing individual members of the mass audience and at delivering messages directly to these individuals through the most efficient and narrowest possible channel. Whether that channel is direct mail, targeted telemarketing, receiver-sensitive Web sites, specialized e-mail lists, or some other means, the objective is the same?to fit the boundaries and fram ing of the message to the needs, wants, expectations, beliefs, preferences, and interests of the audience member. In the aggregate, such messages may form a global statement of position or action, but for the individual audience members only the most relevant and potentially persuasive portions of that global entity are visible. To the extent that communicators are effective in achieving this objective,

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

216 THE ANNALS OF THE AMERICAN ACADEMY

they will, for all practical purposes, have substituted their own audience selection and targeting skills for the role formerly assigned to peer group interaction. The availability and content of each message will have been shaped upon transmission to anticipate and replace the social interaction component of the two-step flow. This is the one-step flow of communication.

The key phrase here, of course, is "to the extent that communicators are effec tive." Clearly, the effectiveness of even the most sophisticated one-step communi cation efforts is limited, and there are, to be sure, residual pockets of two-step communication scattered throughout contemporary society. Some of these pockets are sustained by highly localized social networks and by the identification of indi viduals with them. As Huckfeldt and his colleagues (1995) suggested, some of these "social cells" remain closed to outside influence, while others are better integrated with external information networks. Even relatively coherent local "social cells" are under pressure from the continued development and growing efficiencies of targeted information and communication technologies that deliver strategically personalized communication (Turow 1997). Indeed, self-initiated technological applications such as viral marketing may be redefining personal networks them selves in ways that open them to increased outside penetration (see, for example, Subramani and Rajagopalan 2003). Looking at society more broadly, it is also clear that older demographics are more likely than younger ones to belong to member ship organizations that serve as transmission belts in the two-step flow (Putnam 2000). Our overall argument is about trends that point to the decline of a social membership society and the rise of a lifestyle network society in which information is less likely to be cued by proximate authoritative opinion leaders.

The achievement of a one-step flow among large-scale publics is typically nei ther simple nor inexpensive. It requires, at the very least, a marriage of basic communication skills with the availability of vast and highly differentiated data on members of the prospective audience, the ability to manipulate and organize the data in ways that permit the linkage of specific cases with specific attributes, the capability to identify the physical or electronic location of the specific individuals (or resident locations of like demographics) in the database, and access to some suitable form of narrowcasting technology. That list of prerequisites may appear to be insurmountable, and in the era ofthe two-step flow, it was. But today, half a century later, each of these elements is readily available to those able to afford them. Both the hardware and software now exist to reach large numbers of indi viduals (generally targeted as demographic clusters) directly with messages that appeal to their personal emotions and their very identity (Turow 1997). Their use produces an interaction, not among members of peer groups, but between the technology and the individual audience member.

This means that the Republican voter opening her door or agreeing to talk on the phone with a partisan canvasser may receive a different political message than her similarly Republican neighbor, based on their known differences in issue interests, political donation histories, religious or sexual preferences, or con sumer purchasing patterns. Moreover, her responses or donation levels may be entered in a wireless data transmission device after the contact has ended and

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 217

uploaded into the party's district or national database. As the database is fueled by other interactions, it is programmed to change incrementally in real time, adjusting the configurations of target populations, message impact evaluations, and next approach strategies. At least that is the scenario for canvassing of the near future being offered by both parties. Another slice of that scenario has each party claiming a database of upwards of 150 million individuals, with anywhere from several to several hundred pieces of information on each person. These information bits may include education; employment; sexual preference; dona tion histories; organization memberships; purchasing patterns; and other private traces left behind through credit purchases, marketing survey responses, or Internet surfing behavior, to name but three potential sources. These data are mined by companies that resell them to communication clients such as governments, interest organizations, political consultants, parties, and corporations. Howard (2006) suggested that a less transparent, less public political culture may result from these developments.

We believe that the underlying relations between individuals, society, sources, and messages are, indeed, changing in ways that suggest the need to formulate a new communication paradigm. But we also wish to hold open the question of whether the communication technologies that serve this paradigm must neces sarily be developed and deployed in ways that isolate individuals and fragment society. Our concerns in this article are twofold: how better to understand the two-step paradigm in relationship to its changing social and historical contexts and whether the emerging one-step process can be theorized and researched in ways that offer insights about more and less positive developments in social and democratic communication processes.

Social Reality and Communication Paradigms

Communication models are rooted in assumptions about the social structures inhabited by the individuals and groups being studied. Researchers may fail to for

mally theorize these deeper social structures, but they rely on assumptions about them in designing research and interpreting results. For example, as elaborated below, the two-step flow rests on assumptions about a pluralist social order that both invite intuitive acceptance of the model and guide its empirical applications. Those assumptions may have been reasonable in 1950, but they seem less reason able today. We suggest that the model seems most appropriate to understanding communication flows in American society from roughly the early decades of the twentieth century until the 1980s, a period that various scholars have associated

with the rise and fall of a civil society defined by federations of social groups from national service organizations to bowling leagues (Putnam 2000).

We see the upward arc ofthe two-step flow (long before it had been identified formally) in the dawn of modern public relations, when Edward L. Bernays and others on the Committee on Public Information during World War I realized that the mass propaganda messages aimed at mobilizing support for the war were best

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

218 THE ANNALS OF THE AMERICAN ACADEMY

delivered by a legion of "four minute men," those respected members of local communities who stood up in theaters and other town meeting places to deliver scripts from on high as their own (Ewen 1996).

The downward arc of the paradigm coincides with the rising tide of studies indicating that the social cues people needed to interpret messages from mass

media sources were often embedded in the media context itself. Indeed, the later decades ofthe two-step flow paradigm (roughly from the 1980s through the mid 1990s) saw a notable decline in research findings that supported earlier wisdom about the minimal effects of mass communication (Klapper 1960), and a rise in findings that mass media content produced substantial direct effects without notable secondary social processing (Iyengar and Kinder 1987; Iyengar 1991; Mutz 1998). In this twilight period ofthe two-step flow, the group-based society declined, yet the mass media continued to reach large audiences through relatively few channels. Under these circumstances, journalists and other content develop ers such as entertainment producers became practicing sociometricians, writing social and political cues into the demographically targeted narratives we call news and entertainment. In the public information sphere, social cues were coded in terms of recognized news sources (organization leaders, public officials) or ideo logical labels (Zaller 1992).

Both media and society have continued to change since the wave of studies showing direct mass media effects such as agenda setting, framing, and priming, among others. One change is that conventional mass media reach smaller audi ences, while niche media attract increasing numbers, making it harder to send effective generalized messages but easier to target specialized appeals. Another change is that individuals in fragmented late modern societies have assumed more responsibility for managing their own emotional and cognitive realities, often apart from group influence processes (Giddens 1991; Bennett 1998; Mutz 1998). This implies less reliance on interpersonal influence particularly in the area of politics and public affairs, a factor that may be associated with both the declining importance of politics for many people, and the tendency to avoid polit ical talk in many social relationships (Mutz 1998; Eliasoph 1998).

All of this suggests that we are at an important moment in communications history for thinking carefully about (1) how we define the models that may become the next paradigm and (2) what we would like those knowledge struc tures to do. Accumulated experience with the old paradigm offers some guid ance. While we believe that there were, indeed, some broad correspondences between the two-step flow and the observable social structure of the United States, we also think that there are some important cautionary tales to be told.

One ofthe leading critics ofthe two-step flow paradigm, Todd Gitlin (1978), argued that the two-step model was more the product of broad conceptions of social pluralism shared by social scientists in the early 1950s than it was ever empirically supported by strong or consistent findings. Gitlin argued that plural ism and related behavioral research orientations were congenial to funding agencies and communications industry support for the early Lazarsfeld studies. In Gitlin's critique, the two-step flow was the perfect antidote for unsettling

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 219

critical-theory-based notions of a mass society in which media were strong, intrusive forces, akin to hypodermic injections.

Gitlin's (1978) reading ofthe Katz and Lazarsfeld (1955) studies suggests sev eral anomalies that might have challenged their conclusions about the two-step process. For one thing, the interpersonal influence findings were based on data gathered largely before the rise of broadly available television programming, which has become the sine qua non of mass media influence. Perhaps more curi ous is the fact that Katz and Lazarsfeld brushed aside their own acknowledged finding that in the area of public affairs, fully 58 percent of opinion changes were made without remembered personal contact, meaning that the main effect was presumably from the mass media. Consistent with this anomalous finding, a bare 38 percent of reputed influentials and influencees in the area of public affairs confirmed that they had in fact made an attempt to influence someone or been influenced (Gitlin 1978, 219).

Since four areas of information flow were studied by Katz and Lazarsfeld (1955) (fashion, marketing, movies, and public affairs), the major discrepancy in the area of public affairs was apparently easier to ignore in declaring general empirical support for the model. Yet subsequent decades of research by other scholars addressing a variety of information flows, in an array of settings, and published in leading journals, also failed to produce consistent support for the model, resulting in any number of proposed amendments and revisions (examples include Burt 1999; Harik 1971; Weiman 1982). The most careful studies tended to discover multiple flow patterns, from the preferred horizontal flow of the two-step model, to, in some cases, a half dozen different flow direc tions, including direct media effects (Robinson 1976). Yet none of these pertur bations in the paradigm shook the standing of the simple and compelling two-step flow idea. They were brushed aside in a manner typical of the normal science process operating in paradigms as described by Kuhn (1962). Gitlin (1978, 205) argued that it was hard to challenge this "dominant paradigm" of

media sociology when the likes of Daniel Bell talked about the "received knowl edge" of "personal influence" and pronounced Personal Influence "the standard work" (see Bell 1975, 218).

In the ensuing decades after the publication of Personal Influence, research into political sociology, social psychology, and communication accelerated, pro ducing a rich trove of knowledge about human behavior in general and human behavior in politics in particular. We?the social scientific community?knew much less in 1955 than we know today, and we were far less sophisticated in our thinking. And that, in turn, makes the lasting influence of Personal Influence all the more remarkable. Consider, for example, the study of attitudes and attitude change. It was after the two-step flow was promulgated that social psychologists developed such theories as Festinger's (1957) cognitive dissonance, Katz's (1960) functional analysis, belief-based models of attitude change such as Fishbein's (1967a, 1967b) summative model, Sherif and Sherif's (1967) social-judgment involvement approach, or Petty and Cacioppo's (1986) elaboration likelihood model, to name but a few of the more prominent subsequent conceptual

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

220 THE ANNALS OF THE AMERICAN ACADEMY

advances. Similarly, the study of public opinion benefited only later from Kelman's (1961) work on compliance, identification, and internalization of meaning; Converse's (1964) analysis of belief systems; McCombs and Shaw's (1972) breakthrough study of agenda setting; or even such essential building block notions as indexing (for example, Tannenbaum 1955) or framing (devel oped earlier but perhaps best characterized in Entman [1993]). In political sociology, Robert Lane had yet to write Political Life (1959); and C. Wright Mills,

who played an early role in Katz and Lazarsfeld's project and actually served as a field-worker on the Decatur study, was only then at work on his best-known book, The Power Elite (1956). Berger and Luckman's (1966) reality was still a social construct away. Even though some of this work, including some ofthe early work, directly challenged the dominant paradigm, most of it continued to motor along side for some time, causing little disturbance.

The two-step paradigm was further supported by the failure of early election studies to find substantial campaign media effects, and so, political science came under the influence of the communication variable-averse Survey Research Center at the University of Michigan, which we know today as the Inter University Consortium for Political and Social Research (ICPSR), the early work of which was typified by Campbell et al. (1960). To this day, political science has been slow to develop a robust agenda of work on media and politics at the core of the field. Meanwhile, mass communication research would soon reinforce the dominant paradigm under Joseph Klapper's statement of the minimal effects hypothesis in his The Effects of Mass Communication (1960), a view that was still prevalent twenty years later, prompting James Lemert to pose the rhetorical question in his own book title, Does Mass Communication Change Public Opinion After All? (1981).

Broad scholarly belief in a pluralist American social structure also surely helped general acceptance of the model and discouraged its modification or ongoing critical specification. We are not suggesting here that the group dynam ics underlying pluralistic theories were unimportant in the mass communication process of earlier years. Indeed, they almost certainly continue to operate even today for those who hold strong social memberships. Our point is that the strength of the underlying social science consensus surrounding pluralism and behavioral research may have led to an overly simplified notion of mass commu nication based on discounting the many anomalous findings that challenged the paradigm. Those anomalies, if examined more carefully, might have challenged communication theory to accommodate other social processes and more direct media effects that may have existed all along. Indeed, there may have been more continuous evolution of theory from a two-step dominant model to a one-step flow as both society and media continued to change. Herein lies a cautionary lesson for thinking about the next paradigm: if we are at a point where both social structure and media technologies are changing rapidly once again, we will be well served by more careful specifications of the underlying social and technological factors involved.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 221

The One-Step Flow: Society, Technology, and the Audience

Observers such as Joseph Turow (1997) have credited the rise of refined, highly personalized marketing technologies themselves with the breakup of soci ety. While Turow has drawn useful attention to the transformation of various modes of communication into marketing exercises, it seems to us unlikely that communication, alone, can account for changes as fundamental as the decline of social group membership or the isolation of individuals from each other. The individuation of social experience and the decline of broad social memberships, even at the all-embracing levels of class and church that we see in European societies, are changes that are sweeping most postindustrial democracies (Inglehart 1997; Bennett 2005). Indeed, social theorists such as Giddens (1991; see also Bennett 1998) have linked the impact of global economic change to social structure, resulting in the increasing isolation of individuals and the rise of new emotional and identity processes around self-expression and lifestyle. It cannot be surprising that, as these lifestyles become central, the media habits that are at the core of so much of them also begin to proliferate and fragment around personal realities.

[T]he emergence of new technologies and targeting models for reaching people seems as much an effect of a fragmenting

social order as a cause.

Given these developments, the emergence of new technologies and targeting models for reaching people seems as much an effect of a fragmenting social order as a cause. The ongoing interaction between technology and social receptivity is notable in the development of media and audience relationships that point to increasing individuation in the delivery and reception of information. Silicon-based computing, digitalization, miniaturization?these and other core technological developments have changed not only the nature and use of our machines but the relationship between machines and human society. That is as true of politics and democratic practice as it is of any other human endeavor. Indeed, the changes have been so extensive and of such a type that they must, necessarily, call into question even so venerable a theory as the two-step flow. Consider the following:

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

222 THE ANNALS OF THE AMERICAN ACADEMY

J. Evolving media formats. In 1955, television was primarily a black-and-white version of radio, with visual renditions of such radio dramas as the Lone Ranger, and with former newspaper reporters reading (literally) the events of the day to a camera. Telephones were hardwired, and long-distance communication often required the intervention of an operator. Cell towers were elements of prison architecture, computers?the few that existed?filled entire floors of buildings, and the World Wide Web was not even a glimmer in the eyes of scientists working for the Advanced Research Projects Agency (ARPA), the Defense Department agency that spurred development ofthe Internet, which itself was not established until 1958 as part of the reaction to the Soviet Union's launch of Sputnik. This matters because different communications media have different attributes, dif ferent patterns of human interaction, and different implications for political life. Compare the appointment-based society that gathered around network broadcast news with the emergence ofthe podcast society increasingly driven by personalized, on-demand news aggregators.

2. Evolving individual media use habits. Herbert Krugman (1965, 1966, 1971; Krugman and Hartley 1970), in a series of experimental studies in the 1960s and 1970s, demonstrated that people processed information from television differ ently from the way they consumed from print media. Differences were evident in such biometrics as galvanic skin response and brain wave patterns, which con sistently showed that television viewing was a relatively passive activity, less psy chologically involving than reading. The prevailing form and effect of human interaction with the Internet and related new technologies is still to be deter

mined, though there is early evidence that Internet use enhances participation, information seeking, and social interaction rather selectively (Shah, Kwak, and Holbert 2001; Tewksbury 2003; Hardy and Scheufele 2005). The mobility of the cellular culture, which has seen actual declines of 3 to 4 percent per year in the number of residential telephone subscribers in recent years, has rendered the direct-dial area code, a twentieth-century technological wonder in its own right, a pseudo-geographic anachronism in the twenty-first. More to the point, cell phones have rendered individuals reachable virtually anywhere, anytime, and not just with personal calls but with content ranging from music to political informa tion. Such technologies have enabled citizens to send and receive information instantly in organizing demonstrations and protests, enabling them to make news and integrate information and action in real time. As the movement expression coined by Indymedia goes in the one-step era, people can "be the media" (Rheingold 2002). And though the Web had existed only since 1995, and then in a nascent form, by 2000 more than half of all Americans under the age of fifty five were entering the online world at least once a week (U.S. Census Bureau 2000). Today, individuals not only are receivers of information, but also produc ers and distributors it in shifting and often faceless knowledge networks such as blogs, lists, and action alerts forged through social technology rather than personal influence.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 223

3. Evolving social distribution of media. Just as there have been micro-level differences in the ways people interact with and process information from vari ous media, there are macro-level differences in the roles that such media play in society. As new media become available, the potential develops for substantial changes in the relative influence of more traditional forms. We saw this occur in the 1960s, when television emerged as the primary source of news about politics. In 1952, for example, 32 percent of Americans reported that they relied on tele vision as their primary source of campaign news while 28 percent relied on radio. Just a dozen years later, in 1964, television was the medium of choice for 58 percent ofthe public, and radio for less than 4 percent. Newspapers remained in the low to-mid 20s throughout the period. (Nimmo 1970, 116). By the 1990s, Americans

were spending far more time interacting with their television sets than with news papers?roughly seven times as much annually (U.S. Census Bureau 1999). Now, it appears that dramatic shifts away from broadcast television, along with other

media formerly known as "mass," are under way. Between 1993 and 2004, for example, regular consumption of newspapers reported in national surveys dropped from 58 to 42 percent, radio from 47 to 40 percent, local television news from 77 to 59 percent, and television network newscasts from 60 to 34 percent. In contrast, between 1994 and 2004, regular consumption of news from the Internet rose from 0 (1994 was the last pre-Web year) to 29 percent (Pew Research Center for the People and the Press 2004). Not only are the media for information consumption becoming more personalized, but the information for mats are also shifting. These shifts away from mass media experiences may account for the broad questioning of the objectivity of the press and the filtering of facts according to personal values and incipient political interests.

There is reason to believe that these and similar trends matter. For example, we know from the ipost-Personal Influence literature on attitude change and per suasion that certain kinds of learning are facilitated by low levels of psychological involvement. If, as Krugman (1965, 1966, 1971) suggested, television viewing is a more passive experience than reading, it follows that the shift to television as the primary source of political information would have facilitated certain kinds of persuasive efforts and impeded others. In an era dominated by print, for example, media consumption required and rewarded attentive participation in the communication process. In an era dominated by television, inattentive par ticipation was presumed and was quite possibly satisfying in and of itself. In the first instance, the political persuader employing the dominant medium would employ cues designed to stimulate understanding in social context, and perhaps even self-actuated information seeking, elements of the two-step world. But in the second, she would employ cues designed to bypass the critical faculties of the audience member and to preselect and link the message with some desired set of perceptions or values that had been targeted in advance and that not only required no social interpretation but effectively discouraged seeking one. That was the first step into a one-step world in which, in a sense, there is no sociology.

Differences in the means of exchange between audience members and their media aggregate into systemic differences in the means of exchange between the

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

224 THE ANNALS OF THE AMERICAN ACADEMY

FIGURE 1 HOMOGENIZATION OF THE FOX NEWS AUDIENCE

40

35 ^^? 30 ^^^*^^

25 ^

20 ^^r^^^^ " 15

10

5

0 I-.-1 2000 2002 2004

-Republicans - - Democrats

Source: Based on Pew Research Center for the People and the Press (2004).

polity and its citizens. Manheim (1976), Kubey and Csikszentmihalyi (1990), Putnam (2000), Bennett and Manheim (2001), and others have offered a variety of arguments over the implications that such ongoing reconfiguration may have for the nature of power, the conduct of politics, and the character of daily civic life. Similarly, though we have less evidence in hand as to the psychological dynamics of virtual interaction with the news, such scholars as Davis (1999), Norris (2001), and

Bimber (2003) have suggested that the most recent shift in media preferences, incorporating the Internet and related technologies, may have equally significant implications for civil society and for the place of individual citizens in it.

Under this scenario of changing social relationships and media patterns, even "old media" provide new opportunities for effective one-step communication. Consider, for example, the homogenization ofthe audience for Fox News, widely regarded as appealing primarily to conservative viewers. As illustrated in Figure 1, as recently as 2000 18 percent of both Republicans and Democrats, respectively, reported that they watched the cable network regularly. By 2004, however, a clear and expanding gap had emerged, with Republicans doubling their own reported viewership and nearly doubling that of their Democratic counterparts. The result is an increasingly homogeneous audience and one that offers persuaders a venue for narrowcasting their messages?distributing them to large numbers of like-minded recipients. In the world of professional persuaders, messages that are distributed through such specialized channels can be nuanced in ways designed to (de)mobilize

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 225

one audience segment?in this case Republicans?while minimizing the risk of such collateral effects as (de)mobilizing their opponents?here, Democrats?or having inconsistent messages reach diverse audience segments. From the per spective of the audience member, this narrowcasting reduces the diversity of information that is on offer and does so in ways that align directly with his or her presumed needs, wants, and expectations. This mapping of content into mind-set looks much less like a Texas two-step than it does dancing in front of the mirror.

In the case of new media, the opportunities for personalizing messages are immensely greater, and their exploitation can be immensely more sophisticated. Consider, for example, the marketing practices of online retailer Amazon.com, as summarized in part by Waller (2003). When a customer selects a product for view ing, a Web page appears with information about that product. That information includes, among other elements, a list of other products purchased by other cus tomers who also purchased the item in question. We can think of these as "affin ity" purchases. This information is derived from the merchant's database of sales information and can be generated for any book or other product that Amazon sells. Underlying the selection is a potentially vast pool of data that capture clus ters or networks of interrelated products. The process is called data mining, and it can be used in a variety of ways. The same data, for example, can support con struction of a list of all purchasers of any given book or any combination of affin ity products. These lists can be drawn on the basis of demographic, economic, psychographic, sociographic, or other information that is included in the database.

[Tfhere is considerable potential for. . . direct personal messaging to be rendered coherent and connected through technologies . . . that enable people to explore and evaluate linked topics and meet others "face to face" to form

online and offline communities.

Extending this example one more step will make the larger point. Regular cus tomers of Amazon.com also know that whenever they log on to their accounts, they are greeted by suggested purchases, the selection of which is based on their own recent purchasing or browsing behavior. More than that, customers have the option of receiving regular e-mails from the company advising them of new prod ucts that are judged likely to appeal to them. And they also have the option of vol untarily adding to the information the company stores about them by responding

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

226 THE ANNALS OF THE AMERICAN ACADEMY

to marketing surveys, thereby enriching Amazons data mining capabilities. This combination of data mining, audience differentiation, and communication out reach creates a unique, specialized, affinity-based, but highly individualized and personalized shopping experience for each Amazon customer. Add the ease with which the databases of one organization can be exchanged with those held by others, and the potential to exploit personalized information for purposes of per suasion is magnified many times (Howard 2006).

The very same technologies, when applied in a political rather than a com mercial context, and by electoral, programmatic, ideological, or other kinds of inherently political actors rather than by merchants, have the potential to deliver information to individuals across a society in such a way that each person receives it through channels and embedded in cues that make clear its relevance for her or him individually. Where implemented, this process can be designed to replace in whole or in part the sociological dynamic of the two-step flow. And, precisely because it is based on interest, preference, and need data provided by the tar geted individual himself or herself, it may also be more effective than peer group exchanges in anticipating and responding to the personalized concerns of each audience member. Though definitive data are not yet available on this last point, Zarsky (2006) has made much the same argument with respect to the advertising in particular. As he put it,

For a successful. . . campaign that will have an optimal effect on a consumers percep tion, attention and comprehension, advertisers require knowledge about their target consumers, and the ability to target them directly. In the offline world . . . these attempts to collect and tailor information are inferior to the online dynamic. . . . Since the dynamics of online advertising offers advantages both in collection and in tailoring, it will prove more effective in meeting advertisers' objectives?achieving influence by crossing the crucial barriers of perception, attention and comprehension (every one of which requires a specific dataset that the online environment makes available). Meeting these objectives will also be made possible by applying . . . data mining technology

which will assist content providers in identifying the preferences, traits and other personal attributes of every user. (p. 218)

Many fear that these communication modes, from automated recommenda tion systems to the burgeoning blogosphere, mainly produce self-centered churn

with little coherent social impact. Yet there is considerable potential for such direct personal messaging to be rendered coherent and connected through tech nologies such as Really Simple Syndication (RSS) feeds, action lists, and meet-up sites that enable people to explore and evaluate linked topics and meet others "face to face" to form online and offline communities.

The push technologies pioneered by online retailers and adapted by commer cial and political marketers can be either stealthy or transparent, depending on the privacy standards, norms, and agreements prevailing between technology developers and their audiences. Thus, individuals may be given the choice to opt in or out of databases, and they may be given choices about which of their response behaviors in credit purchases or in online environments are shared with data miners and their known or unknown clients. Or they may not.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 227

Communication Technologies and Political Choice

The evolving communication technologies and applications that support the one-step flow are not entirely deterministic in terms of whether they must isolate individuals or short-circuit potentially useful social interaction such as delibera tion and critical exchanges of ideas by the public. There are choices being made daily about how the one-step process becomes routinized. We would do well to recognize and assess those choices.

The factor that most favors the predominance of isolating and divisive commu nication technologies in the one-step era is the rise of professional communication strategists advising parties, candidates, leaders, interest organizations, and their issue campaigns (Blunder and Kavanaugh 1999). The resulting culture of commu nication favors a top-down, war room, strategic control relationship between client and target audience (Howard 2006). Insofar as shaping audience responses by pushing emotional buttons is thought to favor client success more than, say, initiat ing open dialogues between candidates and voters, client interests will trump democratic values. As one ofthe founders of modern consulting, Joseph Napolitan, remarked when one of the authors challenged an audience of political consultants to consider the democratic consequences of some of their practices, Why should consultants be obligated to worry about democracy when their primary obligation is to advance the immediate goals of their clients? (remarks made at the annual

meeting ofthe International Association of Political Consultants, Malta, November 9-12, 2001). A notable result of this mentality is the rise of what Bennett (1989) has called a "managerial democracy" in which citizens are symbolically moved in and out ofthe political picture as strategic elements of communication campaigns.

[Tfhe most important point about information transmission in the one-step era is not that

campaigns are permanent or ubiquitous. The real point is that campaigns are definitive.

The managerial polity and its communication regime have developed not of necessity, but through the gradual accretion of choices born of a culture of opin ion management. The result leaves politicians hard-pressed to ask citizens for sacrifice or commitment because such things are not born of communication

manipulations alone. Consider in this light Patterson's (2002) observations about the political landscape facing the citizen in post-9/11 America:

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

228 THE ANNALS OF THE AMERICAN ACADEMY

The organizations they might have joined fifty years ago are now run by professional managers, who solicit dues but do not invite active participants. The war in Afghanistan was fought by professional soldiers, the bioterrorism threat was relegated to professional scientists, homeland security was entrusted to professional law enforcement and intelli gence officers. Even the appeal to sacrifice was missing. Americans were urged to keep spending, lest the economy sink. . . . The modern campaign, too, is a thoroughly pro fessional operation. Voters have no expectation that they will be asked to do anything more than send a check and vote on election day. (p. 186)

Another consequence of the chosen forms of one-step communication flow is what many scholars and pundits have noted as the unboundedness of campaign related communication. The so-called "permanent campaign" is a characteriza tion that is most commonly, but not exclusively, applied to electoral campaigns. Even issue advocates and activists become caught up in permanent campaigns, largely because the entry of numerous groups and individuals in many issue net

works decentralizes the control of campaigns and makes them harder to turn on and off (Bennett 2003).

But the most important point about information transmission in the one-step era is not that campaigns are permanent or ubiquitous. The real point is that cam paigns are definitive. The campaign style of communication?its intellectual bases, assumptions, practices, effects, and consequences?literally defines politi cal communication (and, in important ways, political organization) in the current era. And this is true in venues far removed from contests for public office. Bimber (2003, 104-7) referred to the current communication and information age as "post bureaucratic," indicating that social networks and the information that flows over them are increasingly hard to distinguish in organizational terms, particularly in contrast to an earlier era in which information was largely man aged by bureaucratic organizations that channeled its social flow.

Conclusion: Normative Implications of a New Paradigm

The evolution of an ethos of managed or engineered consent that connects the two-step and one-step eras helps highlight an important distinction between the empirical and normative aspects of communication paradigms. Critiques such as Gitlin's (1978) argue that many ofthe choices about specifying the two-step flow model were not just weakly supported empirically, but they were strongly nor mative, meaning they conveyed a sense of inevitable objective reality when other possibly valid models of communication in society were available. There is a sim ilar danger in the way we think about the one-step flow. Many practitioners today argue that there are few alternatives to the increasingly costly and privacy-invading methods used to reach individuals in fragmenting societies. The stark alternative is seen as chaos and lack of cohesiveness in the important process of engineering consent.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 229

But there are, in fact, fairly clear alternative ways in which we could think about how the one-step flow could operate in democracies. There are practical choices to be made about technologies that render messages more or less engaging, transpar ent, public, exclusive, divisive, or cohesive. These choices are not just technologi cal; they are also political. They affect whether citizens become further isolated through communication processes that emphasize unidirectional and highly manipulative information flows or whether citizens share known social networks through the transparent interactive capabilities that are available through the same technologies. Just as there is little reason to ascribe patterns of social fragmentation to communication processes alone, there is equally little reason to think that fragmented societies require communication processes that further compound the fragmentation. For example, it is just as possible to employ deliberative models of polling (see Fishkin 1991) that bring individuals together to explore ideas in public as it is to implement stealthy communication practices that feed messages to indi viduals based on known emotional preferences aimed at short-circuiting such delib eration. Similarly, it is just as possible to write software that enables recipients of messages to share their reactions and meet online or offline to discuss them, as it is to send individual messages soliciting or triggering individual responses.

This suggests another implication of how we think about the paradigm shift in the message-audience-response relationships of the contemporary era. The ways in which communicators choose among the technological alternatives for reach ing fragmenting audiences may, in turn, have profound interactive effects on those audiences and their subsequent social and political relationships. It is important to emphasize here that technologies exist that have the potential to create mutual understanding and peer-to-peer communication across various issue and demographic divides (Price and Capella 2002; Iyengar, Luskin, and Fishkin 2004). Indeed, many social-movement networks have employed trans parent social networking technologies with the aim of creating sustainable social bonds through the communication process (Bennett 2003).

In the current transition between two historic public communication eras, it appears that the chosen emphasis is more toward the stealthy technologies that isolate individuals than toward transparent networking technologies that may unite citizens in common cause (Bennett 2004). The likely explanation for this, as noted earlier, is that many of the communication consultants who steer much of the democratic process today do not work for social or democratic values but for the political success of partisan clients (Blunder and Kavanaugh 1999). But let us recall that the professionalization of politics is itself a product of particular per spectives and knowledge developed within the two-step paradigm. Recognizing the transition of that paradigm into a fundamentally different one-step flow of communication will lead scholars and practitioners to generate alternative per spectives and knowledge that will similarly shape the future. The choices of pro fessionals, if left to their own devices and driven by their current motives, will shape the polity in less salutary ways for the next half century. Our choices as scholars should guide them.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

230 THE ANNALS OF THE AMERICAN ACADEMY

References Bell, Daniel. 1975. The end of American exceptionalism. The Public Interest 41:193-224. Bennett, W. Lance. 1989. Marginalizing the majority: Conditioning public opinion to accept managerial

democracy. In Manipulating public opinion, ed. Michael Margolis and Gary Mauser, 321-61. Pacific Grove, CA: Brooks/Cole.

-. 1998. The uncivic culture: Communication, identity, and the rise of lifestyle politics. PS: Political Science & Politics 31:741-61.

-. 2003. Communicating global activism: Strengths and vulnerabilities of networked politics. Information, Communication &* Society 6 (2): 143-68.

-. 2004. Political communication and democratic governance: From mass society to personal infor mation networks. Paper presented at the conference on Democracy in the Twenty-First century: Prospects and Problems, University of Illinois, Urbana, October 24-26.

-. 2005. Civic learning in changing democracies: Challenges for citizenship and civic education. Paper presented at the annual meeting ofthe American Political Science Association, Washington, DC, September 1-4.

Bennett, W. Lance, and Jarol B. Manheim. 2001. The big spin: Strategic communication and the transfor mation of pluralist democracy. In Mediated politics: Communication in the future of democracy, ed. W. Lance Bennett and Robert M. Entman, 279-98. New York: Cambridge University Press.

Berelson, Bernard R., Paul F. Lazarsfeld, and William N. McPhee. 1954. Voting: A study of opinion for mation in a presidential campaign. Chicago: University of Chicago Press.

Berger, Peter L., and Thomas Luckman. 1966. The social construction of reality: A treatise in the sociol ogy of knowledge. Garden City, NY: Doubleday.

Bimber, Bruce. 2003. Information and American democracy: Technology in the evolution of political power. New York: Cambridge University Press.

Blumler, Jay G., and Dennis Kavanaugh. 1999. The third age of political communication: Influences and features. Political Communication 16 (3): 209-31.

Burt, Ronald S. 1999. The social capital of opinion leaders. Annals ofthe American Academy of Political and Social Science 566:37-54.

Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald E. Stokes. 1960. The American voter. New York: John Wiley

Converse, Philip E. 1964. The nature of belief systems in mass publics. In Ideology and discontent, ed. David Apter. New York: Free Press.

Davis, Richard. 1999. The web of politics: The Internet's impact on the American political system. New York: Oxford University Press.

Eliasoph, Nina. 1998. Avoiding politics: How Americans produce apathy in everyday life. New York: Cambridge University Press.

Entman, Robert M. 1993. Framing: Toward clarification of a fractured paradigm. Journal of Communication 43:51-58.

Ewen, Stuart. 1996. PR! A social history of spin. New York: Basic Books. Festinger, Leon. 1957. A theory of cognitive dissonance. Stanford, CA: Stanford University Press. Fishbein, Martin. 1967a. A behavior theory approach to the relations between beliefs about an object and

the attitude toward the object. In Readings in attitude theory and measurement, ed. Martin Fishbein, 389-400. New York: John Wiley.

-. 1967b. A consideration of beliefs and their role in attitude measurement. In Readings in attitude theory and measurement, ed. Martin Fishbein, 257-66. New York: John Wiley.

Fishkin, James. 1991. Democracy and deliberation: New directions for democratic reform. New Haven, CT: Yale University Press.

Giddens, Anthony. 1991. Modernity and self-identity: Self and society in late modern society. Stanford, CA: Stanford University Press.

Gitlin, Todd. 1978. Media sociology: The dominant paradigm. Theory and Society 6:205-53. Hardy, Bruce W., and Dietram A. Scheufele. 2005. Examining differential gains from Internet use:

Comparing the moderating role of talk and online interactions. Journal of Communication 55:71-84.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

THE ONE-STEP FLOW OF COMMUNICATION 231

Harik, Ilia F. 1971. Opinion leaders in the mass media in rural Egypt: A reconsideration ofthe two-step flow of communication hypothesis. American Political Science Review 65:731-40.

Howard, Philip. 2006. New media campaigns and the managed citizen. New York: Cambridge University Press. Huckfeldt, Robert, Paul Allen Beck, Russell J. Dalton, and Jeffrey Levine. 1995. Political environments,

cohesive social groups, and the communication of public opinion. American Journal of Political Science 39:1025-54.

Inglehart, Ronald. 1997. Modernization and postmodernization: Cultural, economic, and political change in 43 societies. Princeton, NJ: Princeton University Press.

Iyengar, Shanto. 1991. Is anyone responsible? Chicago: University of Chicago Press. Iyengar, Shanto, and Donald Kinder. 1987. News that matters: Television and American opinion. Chicago:

University of Chicago Press. Iyengar, Shanto, Robert Luskin, and James Fishkin. 2004. Deliberative public opinion in presidential

primaries: Evidence from an online deliberative poll. Paper presented at the conference Voice and Citizenship, University of Washington, Seattle, April 23-24.

Katz, Daniel. 1960. The functional approach to the study of attitudes. Public Opinion Quarterly 24:163-204. Katz, Elihu, and Paul F. Lazarsfeld. 1955. Personal influence: The part played by people in the flow of mass

communications. New York: Free Press.

Kelman, Herbert C. 1961. Processes of opinion change. Public Opinion Quarterly 25:57-78. Kim, Joohan, Robert O. Wyatt, and Elihu Katz. 1999. News, talk, opinion, participation: The part played

by conversation, in deliberative democracy. Political Communication 16:361-86. Klapper, Joseph T. 1960. The effects of mass communication. New York: Free Press. Krugman, Herbert E. 1965. The impact of television advertising: Learning without involvement. Public

Opinion Quarterly 29:349-56. -. 1966. The measurement of advertising involvement. Public Opinion Quarterly 30:583-96. -. 1971. Brain wave measures of media involvement. Journal of Advertising Research 11:3-10.

Krugman, Herbert E., and Eugene L. Hartley. 1970. Passive learning from television. Public Opinion Quarterly 34:184-90.

Kubey, Robert, and Mihaly Csikszentmihalyi. 1990. Television and the quality of life: How viewing shapes everyday experience. Hillsdale, NJ: Lawrence Erlbaum.

Kufm, Thomas. 1962. The structure of scientific revolutions. Chicago: University of Chicago Press. Lane, Robert E. 1959. Political life. New York: Free Press. Lazarsfeld, Paul F, Bernard Berelson, and Hazel Gaudet. 1944. The people's choice. New York: Columbia

University Press. Lemert, James B. 1981. Does mass communication change public opinion after all? A new approach to

effects analysis. Chicago: Nelson-Hall. Manheim, Jarol B. 1976. Can democracy survive television? Journal of Communication 26:84-90. McCombs, Maxwell D., and Donald E. Shaw. 1972. The agenda-setting function of mass media. Public

Opinion Quarterly 36:176-87. Mills, C. Wright. 1956. The power elite. New York: Oxford University Press. Mutz, Diana. 1998. Impersonal influence: How perceptions of mass collectives affect political attitudes.

New York: Cambridge University Press. Nimmo, Dan. 1970. The political persuaders: The techniques of modern election campaigns. Englewood

Cliffs, NJ: Prentice Hall. Norris, Pippa. 2001. Digital divide: Civic engagement, information poverty, and the Internet worldwide.

New York: Cambridge University Press. Patterson, Thomas. 2002. The vanishing voter: Public involvement in an age of uncertainty. New York:

Random House.

Petty, Richard E., and John T Cacioppo. 1986. The elaboration likelihood model of persuasion. In Advances in experimental social psychology, vol. 19, ed. Leonard Berkowitz, 123-205. New York: Academic Press.

Pew Research Center for the People and the Press. 2004. News audiences increasingly politicized. http://people-press.org/reports/display.php3?ReportID=215.

Price, Vincent, and Joseph Capella. 2002. Online deliberation and its influence: The Electronic Dialogue Project in Election 2000. IT 6- Society 1:303-29.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms

232 THE ANNALS OF THE AMERICAN ACADEMY

Putnam, Robert D. 2000. Bowling alone: The collapse and revival of American community. New York: Simon & Schuster.

Rheingold, Howard. 2002. Smart mobs: The next social revolution. Cambridge, MA: Perseus. Robinson, John P. 1976. Interpersonal influence in election campaigns: Two-step flow hypothesis. Public

Opinion Quarterly 40:304-19. Shah, Dhavan V, Nojin Kwak, and R. Lance Holbert. 2001. "Connecting" and "disconnecting" with civic

life: Patterns of Internet use and the production of social capital. Political Communication 18:141-62. Sherif, Carolyn W, and Muzafer Sherif 1967. The social judgment-involvement approach to attitude and

attitude change. In Social interaction, ed. Muzafer Sherif, 342-52. Chicago: Aldine. Subramani, Mani R., and Balaji Rajagopalan. 2003. Knowledge-sharing and influence in online social net

works via viral marketing. Communications ofthe Association for Computing Machinery 46:300-307. Tannenbaum, Percy H. 1955. The indexing process in communication. Public Opinion Quarterly 19:292-302. Tewksbury, David. 2003. What do Americans really want to know? Tracking the behavior of news readers

on the Internet. Journal of Communication 53:694-710. Turow, Joseph. 1997. Breaking up America: Advertisers and the new media world. Chicago: University of

Chicago Press. U.S. Census Bureau. 1999. Media usage by consumers, 1992-1999. Statistical abstract of the United States.

Figure 18.1. Washington, DC: U.S. Census Bureau. -. 2000. Multi-media audiences?Summary 2000. Statistical abstract ofthe United States. Table 911.

Washington, DC: U.S. Census Bureau. Waller, Angie. 2003. Data mining the Amazon. N.P. Weiman, Gabriel. 1982. On the importance of marginality: One more step into the two-step flow of com

munication. American Sociological Review 47:764-73. Zaller, John R. 1992. Nature and origins of mass opinion. New York: Cambridge University Press. Zarsky, Tal Z. 2006. Online privacy, tailoring, and persuasion. In Privacy and technologies of identity?A

cross-disciplinary conversation, ed. Katherine Strandburg and Daniela Stan Raicu, 209-24. Berlin: Springer.

This content downloaded from �������������128.220.8.15 on Sat, 20 Aug 2022 02:58:53 UTC��������������

All use subject to https://about.jstor.org/terms