29

The Rise and Fall of the Limited Effects Model

Peter Simonson

Something named the “limited effects model” of mass communication did not emerge before the 1970s, at the very moment it was being called into question. Indeed, “limited effects” seems to have been a locution favored, if not invented, by its critics, who sometimes mischaracterized it and exaggerated its dominance as a paradigm governing media research in the previous decades. The story of its rise, fall, and continued life extends from the 1940s to the present and provides a window into broader stories about the intellectual and institutional development of media studies in the second half of the twentieth century. It is a story about the perceived mainstream of social scientific media studies in the postwar era, which intersects with a number of the other foundations, moments, and approaches discussed in this volume. It grew up in an assemblage centered upon Columbia University, whose sociology program in the 1940s would come to rival – and eventually surpass – the University of Chicago's. The assemblage centered around Paul Lazarsfeld (1900–1976) and the Bureau of Applied Social Research (known as the Office of Radio Research before 1945), heirs like the University of Chicago to Enlightenment projects of scientifically oriented knowledge production and ideologically committed to pluralistic liberalism and social democracy. They would develop their so-called limited effects model through empirical studies of decision-making behavior related to the media of the early 1940s – radio, newspapers, magazines, and motion pictures. Emerging from studies funded by commercial and government agencies, the model articulated with a distinct political economy of research and thinking. Built upon the intellectual labor of women and ethnic outsiders, the model trails a distinctive and mostly unacknowledged social history. Named and taken to task in the 1970s by Marxian critical and cultural scholars (as well as critics from within the ranks of objectivist social science), the limited effects model reveals some of the main fault lines and boundary disputes of media studies since then. Revisiting it allows us to check and correct our collective memories of the field and also to reflect on elements of classic Columbia research that have been revived and extended in our own time.

First a word about definitions. The notion of “limited” media effects indexes several overlapping but analytically distinct ideas, frameworks, and propositions which are not always distinguished in discussions of the model. On the one hand, “limited” can mean minimal, as in the idea that “the media are not very important in the formation of public opinion,” as Todd Gitlin (1978, p. 207) put it in his carefully documented and very influential critique. “Limited” in this sense could be contrasted with “powerful” effects, a meaning Elisabeth Noelle-Neumann (1973) operated with implicitly when she introduced her alternative paradigm for understanding media, soon to be called the “spiral of silence” theory. Alternatively, “limited” can refer to an absence of direct effects, meaning that media influence on audiences is generally limited by social and psychological filtering mechanisms – “[s]elective attention, selective perception, selective retention, selective recall,” for instance, in addition to membership in groups whose beliefs and cultural stereotypes condition response to mass media (Bauer, 1973, pp. 143, 148). These mediating devices, often captured under the headings of selectivity and interpersonal relations, are said to limit the kind and amount of influence that media might exercise on individuals. More precisely, selectivity and interpersonal relations are said to limit the ability of the media to change people's attitudes, opinions, and behaviors. In both inflections, “limited effects” has typically been linked to the two-step flow model of communication – the idea that media campaigns and information flow to local opinion leaders who in turn pass them on to others via face-to-face talk. In its classic formulations in the 1940s and 1950s, the two-step flow model was linked with research data that indicated that interpersonal influence exercised a stronger force than mass media on the opinion formation and decision-making of individuals. In comparison with people around us, then, media were said to have relatively limited effects. Critics would jump on that conclusion, the research methods that fed it, and the broader view of media, power, and society that underwrote it.

In this chapter, I will sketch the rise, fall, and recent revivals of components of the limited effects model. I'll launch the tale in the 1970s, when Gitlin and others named and historically amplified its significance as they advanced their own alternative understandings and theoretical approaches. From there, I will turn back to the emergence, consolidation, and diffusion of the two-step flow and limited effects ideas at Columbia and elsewhere from the 1940s through the 1960s. Among other points, I will show how it was built on an empirical base of women's talk, took initial shape as one model among several at Columbia, and rose to a position of rhetorical power – though not, I argue, clear or overwhelming dominance. I'll then go on to trace developments and criticisms of the semi-independent two-step flow and limited effects ideas in the 1960s and early 1970s, before Gitlin and other paradigm-shifting critics weighed in. From there, I consider ways that the “limited effects model” came into symbolic focus and did historical and rhetorical work in the 1980s and after, in addition to reflecting on ways that some of its core insights have reappeared in ways we have come to think about audiences, communicative flows, and networks today.

“Limited Effects”: Birth in Death

Before 1975, the term “limited effects” was rarely, if ever, used to name a model or paradigm for understanding media. It entered academic parlance over the next several years, and within a decade it had become a commonplace in media studies. It gained that status through work done by critics, some of whom marshaled the label explicitly, others who energetically questioned the findings, focus, and overarching framework it pointed to. Among the earliest adopters of the term was Steven Chaffee, the Wilbur Schramm-trained student of mass communication and politics whom one observer called “the outstanding communication research scholar of his generation” (Alexander, 2001).¹ Operating from the horizons of positivist, objectivist social science, Chaffee took a shot at the model in the ground-clearing introduction to his edited 1975 volume, Political Communication. “This book has been produced by social scientists who are willing to assume, in some degree or another, that the study of political communication needs to be approached from fresh intellectual perspectives, and with new tools,” he announced (p. 15), before moving on to specify his intellectual target:

Chaffee had been investigating the media's role in political socialization of the young for several years at that point, and had come to recognize the part it played in increasing political knowledge for those who paid attention to it (e.g. Chaffee, Ward, Scott, & Tipton, 1970). Now he was naming his foe, not yet sure whether to hyphenate the phrase “limited effects,” but about to push forward with a series of criticisms of the model's deficiencies in understanding the fuller range of relations among media and politics, culminating in a thoroughgoing critique-cum-historical reconstruction of the model under the rubric of “the beginnings of political communication research in the United States” (Chaffee, 1977, 1978; Chaffee & Choe, 1980; Chaffee & Hochheimer, 1985; Chaffee & Wilson, 1977).

While Chaffee criticized “limited effects” as a model for political communication, the young critical media sociologist Todd Gitlin would cast it in broader relief – as the dominant model for media research writ large – in his thoroughgoing and influential critique, “Media Sociology: The Dominant Paradigm” (1978). Gitlin didn't use the new “limited effects” label, but his eloquent opening salvo made clear what his target would be: “Since the Second World War, as mass media in the United States have become more concentrated in ownership, more centralized in operations, more national in reach, more pervasive in presence, sociological study of the media has been dominated by the theme of the relative powerlessness of the broadcasters” (p. 205). Shortly thereafter, Gitlin made a powerful identifying assertion: “The dominant paradigm in the field since World War II has been, clearly, the cluster of ideas, methods, and findings associated with Paul F. Lazarsfeld and his school: the search for specific, measurable, short-term, individual, attitudinal and behavioral ‘effects’ of media content, and the conclusion that media are not very important in the formation of public opinion. Within the whole configuration,” he went on, “the most influential single theory has been, most likely, ‘the two- step flow of communications’” (p. 207).

Gitlin proceeded to unleash a devastating critique of Lazarsfeld, the dominant paradigm, and the field that was shaped by them, which he traced from The People's Choice (Lazarsfeld, Berelson, & Gaudet, 1944), the classic Columbia study of the 1940 presidential election campaign, through Klapper's “definitive compilation of the field's early stages” in The Effects of Mass Communication (1960). He devoted most of his attention however to Elihu Katz and Lazarsfeld's Personal Influence (2005), based upon a 1945 study in Decatur, Illinois. Gitlin (1978) called it “the founding document of an entire field of inquiry” (p. 208), codifying the two-step flow model, its attention to short-run changes in individual attitudes and behavior, and its conclusion that media power was limited in comparison to interpersonal influence. Gitlin traced the model back to its origins in marketing research and its self-proclaimed advance over what he called an “earlier ‘hypodermic’ theory” in which society is conceived as a “mass society, and mass communications ‘inject’ ideas, attitudes, and dispositions toward behavior into passive, atomized, extremely vulnerable individuals” (p. 210). He critiqued its “administrative” mentality attuned to the needs of marketers and political campaigners and directing its narrow, positivist attention to influences on individual decisions to buy soap or vote for a candidate, and he argued that its conclusions about limited media power were unsupported by the facts marshaled for it. More broadly, he claimed that the view had the effect of deflecting attention from media as they operated at the macro level – e.g., in defining shared reality, establishing and undercutting the legitimacy of institutions and movements, and shaping political agendas and public images. The paradigm reflected an “abstracted empiricism [. . .] concretely founded on the prevailing political and commercial culture” (p. 240), and indirectly doing “its share to consolidate and legitimize the cornucopian regime of mid-century capitalism” (p. 245).

Gitlin's critique flew at the dawning of a new moment in media studies, and what he called “the beginning of the decomposition of the going paradigm itself” (p. 206). As evidence, he cited Chaffee's Political Communication (1975), Maxwell McCombs and Donald Shaw's “Agenda-Setting Function of the Mass Media” (1972), and “in England, the alternative approach of cultural studies,” mentioning Stuart Hall's work among others (Gitlin, 1978, p. 246, n.2; compare Gouldner, 1976, pp. 149–150). Hall would weigh in with his own influential critique of Lazarsfeld a few years later, pivoting from it to announcing “the rediscovery of the ideological dimension” by “the critical paradigm in media studies” then being revived by cultural studies (Hall, 1982, p. 65). Gitlin might have drawn evidence for the decomposition of the paradigm from other sources as well, including the right-leaning German public opinion researcher and theorist Elisabeth Noelle-Neumann (1973), who had advocated a “return to the concept of powerful mass media.” Drawing on openings provided by the British sociologists Jeremy Tunstall (1970) and Halloran, Elliot, and Murdoch (1970), Noelle-Neumann asked if researchers were asking the right questions about media, and sought (like Gitlin and others would) to broaden the frame to consider macro-level issues. Observing the news media's ability to create a climate of public opinion and macro-level cumulative effects, she called into question the idea of audience selectivity, and argued that the media had the power to initiate what she would soon call a “spiral of silence” whereby those who perceived themselves to hold minority positions not favored in journalistic coverage would remain quiet about them, and the supposedly dominant views would grow in support and relative standing. Her critique meant that the limited effects model was attacked from left, right, and center; and from both within and beyond the borders of objectivist social science.

As it emerged from critiques and characterizations in the 1970s and early 1980s, the “limited effects” model was given a history. It was said to have arisen in the 1940s, marking the end of a more speculative period of propaganda research that tended to inflate media power (sometimes captured by the “hypodermic” image), and the beginning of a new era of methodologically sophisticated empirical social scientific study. It was associated with Paul Lazarsfeld (nearly always the central character in the story), Columbia University, and the Bureau for Applied Social Research. It came into focus across three books – The People's Choice (Lazarsfeld et al., 1944), Personal Influence (Katz & Lazarsfeld, 2005), and The Effects of Mass Communication (Klapper, 1960), the first introducing the two-step flow model, and the second two consolidating it into the limited effects paradigm. It was said to have dominated the field – either of media studies writ large, or political communication more specifically. And it was widely considered to be inadequate as a theory – insufficiently supported by its own evidence, narrowly focused on short-term individual-level effects brought about by purposive media campaigns, and blind to broader collective and institutional dimensions of media in modern societies. While there was truth in the history and characterizations of the model, there were overstated, misleading, and strategically selective elements as well. To get a clearer sense of the origins and development of the model, I'd like to turn back to the 1940s and 1950s and sketch pieces of the social, intellectual, and rhetorical history of its central texts and key components.

Birthing the “Two-Step Flow”: The People's Choice and Its Columbian Rivals

In the summer of 1940, a dozen or so women living in Erie County, Ohio laid the groundwork for what others would call the “dominant paradigm.” They were working with Elmo Wilson and Hazel Gaudet, who had taken up residence in Sandusky and were overseeing a major study conducted by Lazarsfeld's Office of Radio Research (ORR), which had just moved from Newark to Columbia. Also collaborating was Elmo Roper, the pioneering marketing and public opinion survey researcher who at the time headed the Fortune Survey. Funded by the Rockefeller Foundation and money Lazarsfeld drew from marketing and consulting work done at the ORR, the Erie County study was an ambitious effort at tracking the interplay of political propaganda, public opinion, and voting behavior over the course of the 1940 presidential election campaign.

Between May and November, the specially trained local women knocked on the door of every fourth house in Sandusky and the rural towns of the county, recruited 3,000 local citizens, and interviewed each one multiple times during the unfolding presidential campaign. They gathered a wealth of information about each respondent, recorded their political preferences and communicative involvement in the campaign, and followed up with open-ended questions when someone reported changing their vote intention in any way. War raged in Europe, France fell to the Nazis, and President Roosevelt campaigned for an unprecedented third term in office against the Midwestern Republican businessman Wendell Wilkie. The Erie County women made use of a still-new method of public opinion research recently codified by Lazarsfeld and Marjorie Fiske (1938), the “panel technique,” which utilized multiple interviews of the same respondents over time. In contrast with polling data, panels had the advantage of being able to track changes in opinion as they unfolded during a political campaign. The research team hoped to correlate changes with “exposure to campaign propaganda in all the media of communication – the press, radio, personal contacts, and others” (Lazarsfeld, Berelson, & Gaudet, 1944, p. 5).

Published four years later, The People's Choice offered a snapshot of how campaign propaganda and informal face-to-face communication intersected with opinion formation and voting in a real-world setting. Scattering quotes from respondents with statistical charts of survey data (designed by the advertising agency McCann Erickson) and larger-bore narratives about the geographical setting and historical context for the campaign, Lazarsfeld, Berelson, and Gaudet (1944) charted among other things the ideological differences (their term) between the parties, the appeal of class-based argument, and the gendered ecology of attention and social influence during the campaign season. The book became a classic work in voting studies and public opinion research.

While change in voter intentions particularly interested them, they identified and discussed at least three major categories of effects brought about by what they called “the speeches, the events, the writings, the discussions, the total propaganda output” (p. 101). “The campaign activated the indifferent, reinforced the partisan, and converted the doubtful” (p. 101), in addition to creating a “bandwagon effect” (which sounds something like Noelle-Neumanns spiral of silence) (pp. 107–109). Reinforcement was the main effect (occurring in 53% of the panel members), but more than 30% of the effects fell into categories that don't fall comfortably within the limited effects model – conversion and activation (p. 103).

Beyond a study of voting behavior and opinion formation, The People's Choice is a book about “propaganda,” as it reminds us from start to finish. The book's two lasting ideas – opinion leaders and the two-step flow – emerge out of that problematic. In that regard, the book indexes a transition moment between two paradigms and overarching signs – propaganda and communication (on which, see Sproule, 1997, pp. 217–223; Collins, Chapter 27, this volume). That transition would continue through most of the 1940s, with Columbia's work straddling both worlds – the civicly engaged and publicly critical model of propaganda research, and the more removed or disengaged model of objectivist social science (compare Sproule, who puts Columbia firmly on the latter side). The book revolves around propaganda, as propagated through both mass media and face-to-face life. Through the 1940s, social democratic intellectual engagement vied with professionalized liberal objectivism as orienting ideologies governing communications and media research at Columbia (Simonson, 2010, pp. 141–161; Simonson & Weimann, 2003).

“Opinion leaders” straddled both worlds – more exposed to campaign propaganda in the media, and more engaged in talking about it with others. The concept was applied to anyone who answered yes to one of two questions raised by the interviewers: “Have you tried to convince anyone of your political ideas recently?” or “Has anyone asked your advice on a political question recently?” (Lazarsfeld, Berelson, & Gaudet, 1944, p. 50). (Twenty-one percent of their sample were designated “opinion leaders.”) They played a key role in “the two-step flow of communications” (plural in this first iteration), which named the way that “ideas often flow from radio and print to the opinion leaders and from them to the less active sections of the population” (p. 151). When local citizens reported that they changed their minds about something, they often made reference to interpersonal conversation, the authors found. The two-step flow helped explain that specific finding: “The person-to-person influence reaches the ones who are more susceptible to change, and serves as a bridge over which formal media of communications extend their influence” (p. 152).

The opinion leader and two-step flow ideas took shape within a book that functioned in the propaganda debate as a kind of democratic apologetic. Opinion leaders, the text announced (and Lazarsfeld consequently amplified), were “found in all occupational groups” (p. 50). Never mind that the chart they produced showed enormous differences among occupations in the prevalence of designated “opinion leaders” (44% of salesmen as opposed to 13% of housewives and 15% of the unemployed, for instance), the official rhetoric of opinion leadership suggested a democratic otherwise. The two-step flow played a similar function, suggesting that propaganda from the mass media might be refracted through the good sense of locals who pass it on to their friends. The moral of the story was clear in the book's upbeat, democratic close: “In the last analysis, more than anything else people can move other people. From an ethical point of view this is a hopeful aspect in the serious social problem of propaganda. The side which has the more enthusiastic supporters and which can mobilize grass-root support in an expert way has a great chance of success” (p. 158). Limited media effects, in other words, were coupled with a vision of the power of grassroots organizing, and the hope that democratically supported propagandas might win the day.

When it came out in 1944, The People's Choice was far from the most notable study of media produced by Columbia. That title belonged to Radio Research 1942–1943 (Lazarsfeld & Stanton, 1944), which gave a varied and compelling picture of a field the editors saw on its way to merging “with the study of magazines, newspapers, films, and television into one broader discipline of communications research” (p. vii). Articles by Herta Herzog, Frankfurt School émigré-in-residence Leo Lowenthal, and others addressed everything from the characters and “listening gratifications” of soap operas to the popular music industry and the historical and cultural meaning of biographies in popular magazines, in addition to a series of articles about methodological issues and applications of the so-called Lazarsfeld–Stanton Program Analyzer, a kind of primitive polygraph machine that allowed researchers to determine when members of small test audiences liked or disliked parts of a radio broadcast or film.

With its range of media, topics, methodologies, and conceptual orientations, Radio Research 1942–1943 ushered in modern media studies. Meanwhile, when Radio Research emerged from the presses, Lazarsfeld's colleague and longtime collaborator Robert K. Merton was in the midst of a study of a series of all-day radio marathons conducted by the patriotic icon and celebrity Kate Smith, which would yoke in-depth qualitative interviews with middle-range and larger-bore sociological theory and symbolic and ideological analysis (on middle-range theory, see Boudon, 1991). The result, Mass Persuasion (Merton, Fiske, & Curtis, 1946), signaled the birth of a cultural sociology of media, manifest with restrained critical civic impulses that channeled both Marx and Dewey. Neither Radio Research nor Mass Persuasion worked with the two-step flow/opinion leader ideas, nor did either feed straightforwardly into the limited effects paradigm.

Through the 1940s, media study at Columbia was a rich and internally diverse project – committed not just to “effects” studies, but also to probing qualitative inquiry into audiences' social and psychological “gratifications” as well as the “functions and dysfunctions” of mass communication. “Effects” was Lazarsfeld's preferred idiom, reflecting long-standing interest in determinants of individual action, his statistical acumen in investigating them, and a theoretical sensibility that blended European positivism and US pragmatism and rallied around the ideal of objectivist science as a potentially meliorative social force. In focusing on effects, he aligned himself with both the University of Chicago's Douglas Waples and the polymath political scientist Harold Lasswell (1948), who generated the architectonic scheme for communication as a study of “who, says what, in which channel, to whom, with what effect?” While campaign research like The People's Choice tended to focus on short-term media effects, Lazarsfeld actually aspired to address long-term and institutional effects as well (Katz, 2001; Lazarsfeld & Merton, 1948). Merton was less apt to write about “effects,” as was Herta Herzog (who left the ORR in 1943 to head the research department at McCann Erickson, where she disseminated media research techniques into advertising and marketing [see Gladwell, 1999]). Besides the Kate Smith study, Merton's functionalist analysis is best exemplified in his superb 1948 essay with Lazarsfeld, which posits three middle-range concepts to describe ongoing media processes – enforcing social norms, conferring public status, and narcotizing citizens (Lazarsfeld & Merton, 1948; see Simonson & Weimann, 2003). Herta Herzog (1940, 1941, 1944), meanwhile, pioneered the study of audience gratifications, which she worked out through a deep understanding of psychology as understood by Karl and Charlotte Buhler, Hadley Cantril, and Freud. Bernard Berelson (1949) would move the gratification paradigm in more social dimensions in his classic study of a 1945 newspaper strike. Through the 1940s, then, the limited effects paradigm wasn't even dominant at Columbia, much less the rest of the overlapping fields of propaganda, public opinion, and mass communications research.

There was diversity at Columbia, but also a family resemblance across research projects which distinguished them from other approaches represented in this volume and from earlier versions of social research at Columbia. Lazarsfeld and Merton set the tone for an objectivist, scientifically guided style that blended sophisticated methodology, carefully articulated middle-range concepts, and a blend of qualitative and quantitative empirical research guided by them. It was both more professionalized and more advanced methodologically and theoretically than the publicly oriented work of the man who helped hire them, Robert S. Lynd, co-author with his wife Helen of the landmark Middletown studies of the 1920s and 1930s that presented US citizens with a social portrait of themselves. Lazarsfeld and Merton operated in a different intellectual style, also well distinct from that practiced by the Frankfurt School's Max Horkheimer and Theodor Adorno, who took up residence at Columbia for a time in the early 1940s in one of those meetings-of-the-paradigms that has become canonical in historical tales of media research (e.g., Morrison, 1978; see also Wheatland, 2009). Lazarsfeld (1941) himself tried to mediate the differences between what he termed “critical” and “administrative” media research, and both Merton and the Frankfurt School's Leo Lowenthal made efforts to do some of the same (on which, see Merton, 1968, pp. 493–509). In contrast with the Frankfurt group, the Columbians were less historical in their writings on media (though Merton's work typically cast mass communication in some historical perspective), less apt to situate media within a larger social order, more empirically positivist, and more attuned to audiences than industries (for a fascinating exception from Columbia, see Smith, 1945). Marxian and Freudian insights are scattered across classic Columbia writings from the 1940s, but as trace elements among an eclectic array of theoretical influences that also include Durkheim, Gabriel Tarde, Adolphe Quetelet, Talcott Parsons, and the pragmatists. It is worth remembering, too, that whereas Horkheimer's Institute for Social Research had a rich endowment that supported it, Lazarsfeld's Bureau needed to support itself in an era when foundations and funding agencies were just beginning to provide grants to the social sciences. This need would affect the kind of research the group did. The People's Choice was funded by the Rockefeller Foundation, which was broadly concerned with US culture and civic life; by Life magazine, which was interested in the influence of its publications; and by income generated by commercial marketing research done by the Bureau. A large subsequent study in Decatur, Illinois – about which more in a moment – was in turn funded largely by the magazine publisher Macfadden Publications, financially driven by the marketability of the two-step flow model to commercial funders (Morrison, 2006).

Beyond its differences from the Frankfurt group, Columbia differed from Chicago in several ways, too. For one thing, Columbia conducted far more funded research than its Midwestern rival, organizing larger-scale research-team projects of the sort represented by their community based voting and social influence studies. (The most famous team-based media project at Chicago in the postwar era – Kurt and Gladys Lang's [1968] classic study of the live-crowd vs. televisual experience of the MacArthur Day parade – was conducted by graduate student volunteers who set themselves up around the parade route.) Besides their differing political economies, Chicago-style research tended to work in a broadly symbolic interactionist tradition, attending to cultural meanings, dramatic enactments, and community formations related to media and communication. Merton's Kate Smith study and Herzog's early work on soap opera listeners certainly attended to meaning-making and related cultural issues, but the mainstream of empirical research at Columbia attended more to flows of influence and behavioral effects brought on through mass media. Beyond the specific details and fault lines at Columbia that I have been charting, it is important to keep an eye on the bigger picture as well.

Personal Influence and Two-Step Flows in the 1950s

In the summer of 1945, women in Decatur, Illinois performed the communicative labor that led a decade later to the publication of Katz and Lazarsfeld's Personal Influence. which more than any other text narrowed the conceptual landscape of Columbia media research and established the limited effects model. The study was funded by Macfadden Publications, publisher of, among other periodicals, the popular lowbrow women's magazine True Story and the general interest Liberty magazine, on whose editorial pages Bernarr Macfadden had once boasted that he had “more influence politically on the masses than any one individual in the United States” quoted in Simonson, 2006, p. 12). True Story had commercial interest in the two-step flow idea, along with the finding that opinion leaders operated in all social strata – both of which served as fodder to persuade advertisers that there were valuable multiplier effects to running ads with them. Time magazine had funded a similar study a year earlier, searching out what Merton would re-name “influentials” in his administrative and scholarly write-ups of his field study in Dover, New Jersey (Merton, 1943, 1949).

In January 1945, Lazarsfeld and Merton hired a young C. Wright Mills to oversee field research in Decatur (Summers, 2006. pp. 28–29). He would work with Helen Dinerman and Thelma Ehrlich, two researchers at the newly renamed Bureau of Applied Social Research, trained in the panel technique by Jeannette Green. They would recruit some 15 local women to conduct repeated interviews with 800 Decatur women to determine how they made up their minds about marketing and political decisions (Douglas, 2006). It was the first time Mills had overseen a large empirical study, and it would be the wedge that drove him and Lazarsfeld apart, setting in motion one of the defining intellectual breaks in postwar sociology (Summers, 2006). Among their differences, Mills found the structure of opinion leadership and interpersonal influence to be less democratic than Lazarsfeld had been trumpeting since The People's Choice.

After more than a decade of travails following the field study in Decatur, Personal Influence was published. Through rhetorically compelling narratives that served to magnify its own significance, Personal Influence invented one of the most persistent – and inaccurate – accounts of the history of media research, and helped convince a number of readers that mass media were not in themselves important social forces. Though it nowhere uses the term “limited effects,” Personal Influence laid out a problematic and tropes that – amplified by Katz, Klapper, and other former Columbia graduate students – settled into the limited effects model. It was really two books in one. Part I, “The Part Played by People: A New Focus for the Study of Mass Media Effects,” was a revision of Katz's dissertation, which was in turn an extension of a report he wrote for a major but unpublished study of US television funded (and later killed) by the Ford Foundation (Katz, 1953; Morrison, 2000; see also Buxton, 2009). It presented a tale of discovering “intervening variables” in effects of mass communication, including the variable of interpersonal relations and opinion leaders, and consolidated Columbia research on the topic, including their second community based voting study (Berelson, Lazarsfeld, & McPhee, 1954). It also included a highly influential history of the field, which purportedly grew from an era when the audience was conceived as “an atomistic mass of millions of readers, listeners, and movie goers prepared to receive the Message,” and “every Message [was seen] as a direct and powerful stimulus to action which would illicit immediate response” (Katz & Lazarsfeld, 2005, p. 16). This image would later be associated with the “hypodermic model” Gitlin referenced, though it was essentially a straw man that referenced no actual earlier work. As Jefferson Pooley (2006) and Debora Lubken (2008) have convincingly shown, Personal Influence invented a flawed “other” to which it claimed superiority. Something analogous would happen in the 1970s when the “limited effects model” was rhetorically mobilized by its critics.

Meanwhile, the second part of Personal Influence, “The Flow of Everyday Influence in a Midwestern Community,” presented and interpreted the decade-old data from Decatur in a manner that has been roundly criticized by social scientists and critics since. Relying on self-reports of the Decatur women, it concluded that interpersonal influence was more important than mass media in decisions made about shopping, movie-going, fashion, and public affairs. It analyzed opinion leadership in those four realms, and concluded with consideration of the two-step flow and interpersonal influence. Like The People's Choice, Personal Influence fits nicely into a liberal pluralist worldview and presented a reassuring picture to mid-century US citizens – opinion leaders came from all social strata and through face-to-face conversation helped limit the force of mass media; the book functioned as a democratic apologetic in the face of fears of a mass society (McCormack, 2006; Scannell, 2006). Some reviewers criticized the evidence advanced as support for the two-step flow theory, but most accepted the book's claim that interpersonal outweighed media influence in individual decision-making. Kurt and Gladys Lang (2006) have argued that the book “heralded a clear shift in the rhetoric” of sociologists and political scientists surrounding the importance of mass communication and toward a “‘minimalist’ view of media effects” that pushed out alternative conceptions (pp. 165, 167). These included the broadly symbolic interactionist perspectives developed by the Langs and others at the University of Chicago, which mapped alternative paradigms of media sociology (Lang & Lang, 1961, 1968).

Though the minimalist view of media power shifted the academic rhetoric, the model actually contained two overlapping ideas, one of which could also support theories of more robust media power. On the one hand, the claim that selectivity and group norms limited the direct influence of mass media fed a conclusion supported across Columbia research, captured in a statement from a summary article of 1957: “Mass communication does not serve as a necessary and sufficient cause of audience effects, but rather functions among and through a nexus of mediating factors and influences [. . . which] typically render mass communication a contributory agent, but not the sole cause, in a process of reinforcing of existing conditions” (Klapper, 1957, p. 458). “Reinforcing existing conditions” could be taken as the start of a theory of media and ideological hegemony, a direction Lazarsfeld and Merton (1948) feinted toward a decade earlier (Simonson & Weimann, 2003). But Personal Influence dialed the resolution of the viewing frame down to the individual, shortterm, and decision-making level, occluding the bigger picture, and downplaying the relative significance of mass media as a social force.

The two-step flow idea could be put to other uses, however. Beyond suggesting the interpersonal mediation and therefore limits of media power, it also pointed toward diffusion, larger-scale communicative flows, and social mechanisms of enhancing media effectivity. In this iteration of the two-step flow, media campaigns combined with word-of-mouth diffusion to cascade through social networks. Running alongside rhetoric of comparatively limited media influence, then, Columbia researchers also lay down a vision of communication in a networked society. In this regard, the Columbia paradigm merged with diffusion studies conducted by rural sociologists, with work in international development, and with psychological warfare efforts coordinated by the US government in collaboration with communication researchers (Simpson, 1994), all of whom looked at conversations as potential multipliers of media influence.

The opinion leader, interpersonal influence, and flow-of-communication ideas had entered the state-sponsored Cold War arsenal of psychological warfare by the late 1940s. Frankfurt School émigré and former Bureau affiliate Leo Lowenthal had become research director of the Voice of America, which funded opinion leadership and audience studies in Norway, Sweden, and other European and Latin American countries in 1948–1949, explicitly understood as a contribution to “psychological warfare” (Klapper & Lowenthal, 1951). Two years later, the Voice of America funded similar survey work in the strategic Near East countries of Turkey, Lebanon, Egypt, Syria, Jordan, and Iran, collecting information about media behavior and seeking out opinion leaders in towns and cities. At the same time, the Bureau prepared a survey research manual, Are We Hitting Our Target?, for an immediate predecessor of the US Information Agency, whose mission was to influence foreign publics in the US national interest; it guided opinion leader studies in the Philippines and elsewhere (Simpson, 1994, p. 73). The Near East study would be published as Daniel Lerner's (1958) Passing of Traditional Society, an infamous articulation of modernization theory as well as a fascinating historical portrait of communication, politics, and society in those six still-important countries. Snapshots of opinion leadership are captured throughout (e.g., pp. 26–28, 185–196, 245–246, 333–338).

In the meantime, University of Washington sociologist Stuart Dodd and graduate students like Melvin DeFleur advanced the mass communication and interpersonal flow model through studies of airplane leaflet drops for the US Air Force and covertly funded by the CIA. Their Project Revere study (1951–1953) was a rather remarkable series of field studies of mass leaflet drops in the Pacific Northwest, which built off and considerably extended the general idea of communicative flows laid out in The People's Choice. Dodd and his students meticulously charted geographical diffusion pathways, as well as maximally efficient ratios of leaflets dropped to reception of the message (DeFleur & Larsen, 1987, pp. vii–xxxiv; Dodd, 1952; Lowery & DeFleur, 1995, pp. 213–237). Their work gave rise to a series of additional studies and publications that advanced the mediated - and - interpersonal diffusion model of mass communication, with specific attention to news stories (e.g., Danielson, 1956; Deutschmann & Danielson, 1956; Larsen & Hill, 1954;) and rumor (e.g., DeFleur, 1962). Meanwhile, rural sociologists at Iowa State University studying the adoption of modern farming practices developed their own pictures of the roles played by mass media and face-to-face contacts in the diffusion of information (Beal & Bohlen, 1957; Rogers, 1962; Rogers & Beal, 1958; see also Valente & Rogers, 1995). Through the 1960s, the model of flows contributed to what Everett Rogers (1978) would call the dominant paradigm of development research, partly organized around the idea that “mass communication was [. . .] a very powerful and direct force for development.”

Columbia sociologists rarely cited the Washington group's work, however, nor much from Chicago, choosing instead to solidify their own lineage as it passed from Lazarsfeld and Merton to their favorably placed and influential network of former students through what Terry Nichols Clark (1998) has called the “Columbia sociology machine.” Even then, however, Columbia media studies varied, with some more favorably placed and better amplified by the machine than others. A number of excellent studies lay clearly outside the realm of Lazarsfeldian effects or two-step flow studies – including Leila Sussmann's (1956) study of mail to the president, Warren Breed's (1955, 1958) critical functionalist accounts of news, and Charles Wright's (1959) synthetic introduction to a functionalist sociology of mass communication. Katz and others consolidated the two-step flow and diffusion ideas, however, and underscored the argument from Personal Influence that media power to bring about change was relatively limited (Coleman, Katz, & Menzel, 1957' Katz, 1957, 1960). This line of second-generation Columbia thinking would eventually rise to symbolic dominance.

Consolidation and Challenge in the 1960s

Joseph Klapper's Effects of Mass Communication came out in 1960, consolidating the conceptual vocabulary of “effects” by surveying a wide but also limited body of social scientific research on mass media. A broader and more subtle book than typically portrayed in the collective memory of media studies, Klapper's volume charted a range of effects of persuasive communication, from reinforcement (most prevalent) to the creation of opinion on new issues as well as conversion, along with specific consideration of portrayals of crime and violence, “escapist media,” child audiences, and media-induced passivity. Klapper, who had taken his doctorate at Columbia and then gone to work in the research departments of CBS and General Electric, wrote as a skeptical social scientist surveying certifiable knowledge on the topic. His easy-to-read introduction summarized findings that provided fodder for a limited-effects understanding, including the much-repeated line, “Mass communication ordinarily does not serve as a necessary and sufficient cause of audience effects, but rather functions among and through a nexus of mediating factors and influences” (Klapper, 1960, p. 8). The book's conclusion would warn against “the danger [. . .] to go overboard in blindly minimizing the effects and potentialities of mass communications” (p. 252), identify ways that media can exercise “extensive social effects” by operating through elites (p. 253), and call for more attention to “the probable but unmapped interplay between the mass media and cultural values” (p. 255). As Roger Brown (1970) observed, the book's claims about limited media power received much more attention than its feints toward more powerful effects, an interpretation and uptake which in turn served the interests of CBS and other broadcasters faced with criticism about their offerings and accusations of their deleterious social consequences. In this respect, the paradigm did indeed deflect criticism from powerful media industries, as Gitlin and other critics would argue.

Klapper's book was widely cited, though its real influence is harder to discern. The Columbia two-step flow and effects models were both extended and criticized in the 1960s. Raymond Bauer (1964) condensed the limited effects hypothesis into the image of the “obstinate audience,” which he flatly declared “a defense [. . .] of advertising and mass media” (p. 327), admitting a charge that critical researchers would later make of the paradigm. Lazarsfeld and Menzel (1963) posited a “multi-tep flow of communications,” while others distinguished relay and reinforcement functions of opinion leaders (Deutschmann & Danielson, 1960), drew attention to differences between diffusing information and persuading when it came to two-step flows (Troldahl, 1963), or refined the theory by adding categories like “opinion avoiders” and “opinion sharing” (Troldahl & Van Dam, 1965; Wright & Cantor, 1967). Other research called the two-step flow model into question, though, including studies of news diffusion for major events like the Kennedy assassination (Greenberg, 1964; Mendelsohn, 1964; Spitzer, 1964–1965). Agricultural diffusion researchers found the two-step flow model to be overly simplified and misleading, and found that both opinion leaders and followers were directly influenced by media sources, though at different stages of the adoption process (van den Ban, 1964). Others argued that the two-step flow model had never been adequately confirmed empirically (Arndt, 1968). From the perspective of Chicago sociology, Kurt and Gladys Lang (1968) offered a more culturalist alternative, building on a decade and a half of pioneering work definitively outside the Columbia orbit. Interestingly, in the definitive compilation of media effects research of the era, the two-step flow model received very little attention (Schramm & Roberts, 1971).

By the early 1970s, it was clear to empirical researchers that “the two-step hypothesis appears to explain very few communication situations and is likely too simplified a concept for great utility in explaining the process of communication” (Bostian, 1973). Katz himself had turned away from short-term effects and diffusion to revive Herzog's gratifications paradigm (Blumler & Katz, 1974; see also Katz, 1968), and Noelle-Neumann (1973) had called for a return to the concept of powerful mass media. From England, Jeremy Tunstall's (1970) Media Sociology collection pivoted away from the US effect and flow models (p. 5) and toward two dozen essays that cast media in broad social and institutional terms. From the international New Left, Richard Hoggart and Stuart Hall's Centre for Contemporary Cultural Studies was about to enter its most creative and influential phase in Birmingham, while Latin Americans were working out their own Marxian-inspired theories of dependency, domination, and resistance. In the United States, as left-leaning US sociologists invented and attacked “mainstream sociology” (Calhoun & VanAntwerpen, 2007), a new generation of media sociologists led by Gaye Tuchman (1978) cast broad attention to the institutional power, “symbolic annihilations,” and ideological dimensions of the media, at the same time that James W. Carey (1975) was laying the groundwork for US-style cultural studies. By the time Gitlin (1978) weighed in with his influential critique, the limited effects model was clearly on its down slope, with critics both within and outside objectivist social science.

Limited Effects Since the 1980s

By the mid-1980s, the “limited effects model” had become a rhetorical commonplace in mass communication and media studies, used to talk about, separate, normatively rank, and (de)legitimate competing theories and orientations. The “limited effects” locution was given a phantom history, with references to Klapper and other sources where it didn' occur (e.g., Moschis & Moore, 1982). It would take its place in a mythic historical narrative of the field anchored on the far end by the “hypodermic” model of direct media injections into the masses, to which “l imited effects” was a supposed response (Lubken, 2008). As these misrememberings were occurring, Elihu Katz (1980, 1987) substantively responded to and dialectically incorporated Columbia's critics into the effects tradition, extending the idea from Personal Influence that all studies of media in some sense concern themselves with effects. Working with an expansive theoretical tent, he distinguished theories of limited and powerful effects, charted a history of their development, and defended Lazarsfeldian-style communications research as it had developed after the classic work of the 1940s and 1950s. Adding further counterweight, former Columbia Yale social psychologist and persuasion researcher William McGuire (1986) savaged the claims and empirical evidence in dismantling what he called the “myth of massive media impact” and defending the idea of limited effects (cf. Zaller, 1996).

Katz and McGuire were responding to the fact that the limited effects model had come to serve as a discredited alternative against which some alternate approach was framed as superior – be it cultural studies (Curran, Gurevitch, & Woollacott, 1982; Hall, 1982), political economy (Schiller, 1989), humanistic rhetorical studies (Gronbeck, 1984), or competing social scientific theories like agenda setting (McCombs, 1981), political socialization (Chaffee & Hochheimer, 1985; Moschis & Moore, 1982), or information processing (Entman, 1989). In short, in the 1980s, “limited effects” came to serve as a symbolic boundary marker with considerable rhetorical power in teaching, textbooks, and research about media.

The opinion leader and two-step flow components of the model have in the meantime taken on new life in network theory and studies of Internet-related diffusion. Heinz Eulau (1980) and Katz's student Gabriel Weimann (1982, 1994) joined Columbia ideas to a new generation of network analysis techniques. Others incorporated part of the classic model into increasingly sophisticated network models that tracked diffusion (Valente, 1996), social capital (Burt, 1999), and marketing and social epidemics (Watts, 2007; Watts & Dorrs, 2007). Featured as the Number One “Breakthrough Idea” of 2007 in the Harvard Business Review, Duncan Watts's notion of the “accidental influentials” shifted attention from influential individuals to network structures that allow influence to cascade widely. Campaign researchers continue to find variations on the two-step flow idea useful (e.g., Hornik, 2006; Southwell & Yzer, 2007). It has provided a tool for mapping Internet-related diffusion and influence patterns related to medicine, politics, and climate change campaigns, among other topics (Case, Johnson, Andrews, Allard, & Kelly, 2009; Nisbet & Kotcher, 2009; Norris & Curtice, 2008). The Columbia flow paradigm lives on, and helps to explain one family of mediated social processes. More recently, it has been joined by a revival of the idea of limited effects within political communication and persuasion research; reflecting on “the continued detachment of individuals from group-based society, and the increasing capacity of consumers to choose from a multitude of media channels,” Lance Bennett and Shanto Iyengar (2008), for instance, have argued that we are entering “a new era of minimal consequences” (pp. 708, 725; see also Bennett & Iyengar, 2010; Holbert, Garrette, & Gleason, 2010).

Conclusion

The rise and fall of the limited effects model turns out to be a more complex and interesting story than might be suggested by its place in the collective imagination of contemporary media studies. Let me end by summarizing and reflecting on what I take to be some of the lessons of this historical episode. I have traced some of the intellectual, institutional, and rhetorical origins and fault lines of the model and shown how these developed over time. It grew out of Columbia's Office of Radio Research in the early 1940s, mediated through women interviewers in three small provincial cities, and taking shape through the objectivist language and middle-range theory of Columbia sociology. The “two-step flow of communications” was one of those middle-range theoretical concepts, linked in its development to the conceptual problematics of propaganda and public opinion, the new research method of the panel interview, and funded investigations that served the purposes of media organizations and commercial marketing as well as progressive tolerance campaigns and the social scientific pursuit of new knowledge. The two-step flow came to be linked to a view of the mass communication process that posited a number of intervening psychological and social mechanisms limiting the ability of purposive media campaigns to change people's attitudes and behaviors. This finding led to the conclusion that most of the time media play a reinforcement role. Together, these semi-independent elements came together into what its critics named the limited effects model.

Since the 1950s, the flow, limiting mechanisms, and reinforcement ideas have all found places in a range of research and thinking. This traverses work done both within the Columbia lineage and by critics of it. The “discovery” of people as intermediaries and disseminators fed into both research and practice in diffusion, development, marketing, and psychological warfare. It is one of the birth points for understanding communication networks and flows. The idea that audiences can be “obstinate,” resisting the efforts and messages that come to them through mass media, has in turn fed images of the active audience in uses and gratifications research. Since the 1970s, active audiences have also been a staple in cultural studies, though rarely with positive acknowledgment of the Columbia tradition that preceded it. Audiences who resist or put media to active work in their social environments have been heroes across wide swaths of media studies, which from a grander historical vantage point complexifies and disperses core insights of 1940s Columbia work. Reinforcement, meanwhile, is itself a core concept in Marxian and other critical media studies, worked out through Althusserian, Gramscian, Foucauldian, and other theoretical idioms. The Columbians came to their conclusion from very different political and intellectual horizons, but their baseline conclusions about media's power to reinforce dominant attitudes, public opinions, and behaviors clearly pointed in complementary directions (Katz, 1980, p. 134). That they rarely drew that conclusion – Lazarsfeld and Merton (1948) being an exception – is of course significant, and supports their critics' charge that their middle-range theory was sociologically myopic and obfuscating. Still, as Katz (1987, 2005) has skillfully shown, a good number of contemporary research programs can be cast as extensions of early Columbia insights. Though the limited effects model was itself rather limited, it still flows through our histories and contemporary work.

NOTE

1 Rothschild (1978) cites a 1973 Association for Education in Journalism (AEJ) conference paper by Jay Blumler and J. M. McLeon, “Communication and Voter Turnout in Britain,” that uses the term “limited effects.”

REFERENCES

Alexander, M. (2001). Steven Chaffee, Professor Emeritus of Communication, dead at 65. Stanford Report (18 May). Retrieved April 17, 2010, from http://news.stanford.edu/news/2001/may23/chaffee-523.html

Arndt, J. (1968). A test of the two-step flow in diffusion of a new product. Journalism Quarterly, 45, 457–465.

Ban, A. W. van den. (1964). A revision of the two-step flow of communications hypothesis. Gazette, 10, 237–249.

Bauer, R. (1964). The obstinate audience. American Psychologist, 19, 319–328.

Bauer, R. (1973). The audience. In I. de Sola Pool, W. Schramm, F. Frey, N. Maccoby, & E. B. Parker (Eds.), Handbook of communication (pp. 141–152). Chicago, IL: Rand McNally

Beal, G. M., & Bohlen, J. M. (1957). The diffusion process. Special Report No. 18 (Agriculture Extension Service, Iowa State College), 1, 56–77.

Bennett, W. L., & Iyengar, S. (2008). A new era of minimal effects? The changing foundations of political communication. Journal of Communication, 58, 707–731.

Bennett, W. L., & Iyengar, S. (2010). The shifting foundations of political communication: Responding to a defense of the media effects paradigm. Journal of Communication, 60, 35–39.

Berelson, B. (1949). What “missing the newspaper” means. In P. F. Lazarsfeld & F. M. Stanton (Eds.), Communication research 1948–9 (pp. 111–129). New York, NY: Duell, Sloan & Pearce. Reprinted in J. D. Peters & P. Simonson (Eds.). (2004). Mass communication and American social thought: Key Texts, 1919–1968 (pp. 254–262). Lanham, MD: Rowman & Littlefield.

Berelson, B., Lazarsfeld, P. F., & McPhee, W. (1954). Voting: A study of opinion formation in a presidential campaign. Chicago, IL: University of Chicago Press.

Blumer, J., & Katz, E. (Eds.). (1974). The uses of mass communications: Current perspectives on gratifications research. Beverly Hills, CA: Sage.

Bostian, L. R. (1973). The two-step flow theory: Cross-cultural implications. In M. Prosser (Ed.), Intercommunication among nations and peoples (pp. 122–132). New York, NY: Harper & Row.

Boudon, R. (1991). What middle-range theories are. Contemporary Sociology, 20(4), 519–522.

Breed, W. (1955). Social control in the news room: A functional analysis. Social Forces, 33, 326–335.

Breed, W. (1958). Mass communication and socio-cultural integration. Social Forces, 37, 109–116. Reprinted in J. D. Peters & P. Simonson (Eds.). (2004). Mass communication and American social thought: Key Texts, 1919–1968 (pp. 417–425). Lanham, MD: Rowman & Littlefield.

Brown, R. L. (1970). Approaches to the historical development of mass media studies. In J. Tunstall (Ed.), Media sociology: A reader (pp. 41–57). London, UK: Constable.

Burt, R. (1999). The social capital of opinion leaders. Annals of the American Academy of Political and Social Science, 566, 37–54.

Buxton, W. J. (Ed.). (2009). Patronizing the public: American philanthropy's transformation of culture, communication, and the humanities. Lanham, MD: Lexington Books.

Calhoun, C., & VanAntwerpen, J. (2007). Orthodoxy, heterodoxy, and hierarchy: “Mainstream” sociology and its challengers. In C. Calhoun (Ed.), Sociology in America: A history (pp. 367–410). Chicago, IL: University of Chicago Press.

Carey, J. W. (1975). A cultural approach to communication. Communication, 2, 1–22.

Case, D. O., Johnson, J. D., Andrews, J. E., Allard, S. C., & Kelly, K. M. (2009). From two-step flow to the Internet: The changing array of sources for genetic information seeking. Journal of the American Society for Information Science and Technology, 55(8), 660–669.

Chaffee, S. H. (Ed.). (1975). Political communication: Issues and strategies for research. Beverly Hills, CA: Sage.

Chaffee, S. H. (1977). Mass media effects: New research perspectives. In D. Lerner & L. M. Nelson (Eds.), Communication research: A half century appraisal (pp. 210–241). Honolulu, HI: University Press of Hawaii.

Chaffee, S. H. (1978). Presidential debates – are they helpful to voters? Communication Monographs, 45, 330–346.

Chaffee, S. H., & Choe, S. Y. (1980). Time of decision and media use during the Ford–Carter campaign. Public Opinion Quarterly, 44, 53–69.

Chaffee, S. H., & Hochheimer, J. L. (1985). The beginnings of political communication research in the United States: Origins of the “limited effects” model. In E. M. Rogers & F. Balle (Eds.), The media revolution in America and in Western Europe (pp. 267–296). Norwood, NJ: Ablex.

Chaffee, S. H., Ward, L. S., & Tipton, L. P. (1970). Mass communication and political socialization. Journalism Quarterly, 47, 647–659.

Chaffee, S. H., & Wilson, D. G. (1977). Media rich, media poor: Two studies of diversity in agenda-holding. Journalism Quarterly, 54, 466–476.

Clark, T. N. (1998). Paul Lazarsfeld and the Columbia sociology machine. In J. Lautman & B.-P Lecuyer (Eds.), Paul Lazarsfeld (1901–1976): la sociologie de Vienne à New York (pp. 289–360). Paris, France: L'Harmattan.

Coleman, J. S., Katz, E., & Menzel, H. (1957). The diffusion of an innovation among physicians. Sociometry, 20, 253–270.

Curran, J., Gurevitch, M., & Woollacott, J. (1982). The study of the media: Theoretical approaches. In M. Gurevitch, T. Bennett, J. Curran, & J. Woollacott (Eds.), Culture, society and the media (pp. 6–25). London, UK: Methuen.

Danielson, W. A. (1956). Eisenhower's February decision: A study of news impact. Journalism Quarterly, 33, 433–441.

DeFleur, M. L. (1962). Mass communication and the study of rumor. Sociological Inquiry, 32, 51–70.

DeFleur, M. L., & Larsen, O. N. (1987). The flow of information: An experiment in mass communication. New Brunswick, NJ: Transaction. (Original work published 1958)

Deutschmann, P. J., & Danielson, W. A. (1960). Diffusion of knowledge of the major news story. Journalism Quarterly, 37, 345–355.

Dodd, S. (1952). Testing message diffusion from person to person. Public Opinion Quarterly, 16, 247–262.

Douglas, S. J. (2006). Personal Influence and the bracketing of women's history. Annals of the American Academy of Political and Social Science, 608, 41–50.

Entman, R. M. (1989). How the media affect what people think: An information processing model. Journal of Politics, 51(2), 347–370.

Eulau, H. (1980). The Columbia studies of personal influence: Socialnetwork analysis. Social Science History, 4, 207–228.

Gitlin, T. (1978). Media sociology: The dominant paradigm. Theory and Society, 6(2), 205–253.

Gladwell, M. (1999, March 22). True colors: Hair dye and the hidden history of postwar America. The New Yorker, 70–81.

Gouldner, A. W. (1976). The dialectic of ideology and technology. New York, NY: Continuum.

Greenberg, B. S. (1964). Person-to-person communication in the diffusion of news events. Journalism Quarterly, 41, 489–494.

Gronbeck, B. (1984). Functional and dramaturgical theories of presidential campaigning. Presidential Studies Quarterly, 14(4), 486–499.

Hall, S. (1982). The rediscovery of “ideology”: Return of the repressed in media studies. In M. Gurevitch, T. Bennett, J. Curran, & J. Woollacott (Eds.), Culture, society and the media (pp. 56–89). London, UK: Methuen.

Halloran, J. D., Elliot, P., & Murdoch, G. (1970). Demonstrations and communications: A case study. London, UK: Penguin.

Herzog, H. (1940). Professor quiz: A gratification study. In P. F. Lazarsfeld (Ed.), Radio and the printed page (pp. 64–93). New York, NY: Duell, Sloan & Pearce.

Herzog, H. (1941). On borrowed experience: An analysis of listening to daytime sketches. Reprinted in J. D. Peters & P. Simonson (Eds.). (2004). Mass communication and American social thought: Key Texts, 1919–1968 (pp. 139–156). Lanham, MD: Rowman & Littlefield.

Herzog, H. (1944). What do we really know about daytime serial listeners? In P. F. Lazarsfeld & F. N. Stanton (Eds.), Radio research 1942–1943. New York, NY: Duell, Sloan & Pearce.

Holbert, R. L., Garrette, R. K., & Gleason, L. S. (2010). A new era of minimal effects? A response to Bennett and Iyengar. Journal of Communication, 60, 15–34.

Hornik, R. (2006). Personal Influence and the effects of the national youth anti-drug media campaign. Annals of the American Academy of Political and Social Science, 608, 282–300.

Katz, E. (1953). The part played by people: A new focus for the study of mass media effects. New York, NY: Bureau of Applied Social Research.

Katz, E. (1957). The two-step flow of communication: An up-to-date report on an hypothesis. Public Opinion Quarterly, 21, 61–68.

Katz, E. (1960). Communication research and the image of society: On the convergence of two traditions. American Journal of Sociology, 65, 435–440.

Katz, E. (1968). On reopening the question of selectivity in exposure to mass communication. In R. P. Abelson (Ed.), Theories of cognitive consistency: A sourcebook (pp. 788–796). Chicago, IL: Rand McNally.

Katz, E. (1980). On conceptualizing media effects. In T. McCormack (Ed.), Studies in communication (vol. 1, pp. 119–141). Greenwich, CT: JAI Press.

Katz, E. (1987). Communications research since Lazarsfeld. Public Opinion Quarterly, 51 (supplement), S25–S43.

Katz, E. (2001). Lazarsfeld's map of media effects. International Journal of Public Opinion Research, 13(3), 270–279.

Katz, E. (2005). Introduction: Lazarsfeld's legacy: The power of limited effects. In E. Katz & P. F. Lazarsfeld, Personal influence (pp. xv–xxvii). New Brunswick, NJ: Transaction.

Katz, E., & Lazarsfeld, P. F. (2005). Personal influence. New Brunswick, NJ: Transaction. (Original work published 1955)

Klapper, J. (1957). What we know about the effects of mass communication: The brink of hope. Public Opinion Quarterly, 21(4), 453–474.

Klapper, J. (1960). The effects of mass communication. Glencoe, IL: Free Press.

Klapper, J., & Lowenthal, L. (1951). The contributions of opinion research to the evaluation of psychological warfare. Public Opinion Quarterly, 15, 651–662.

Lang, K., & Lang, G. E. (1961). Collective dynamics. New York, NY: Thomas Y. Crowell.

Lang, K., & Lang, G. E. (1968). Politics and television. Chicago, IL: Quadrangle.

Lang, K., & Lang, G. E. (2006). Personal Influence and the new paradigm: Some inadvertent consequences. Annals of the American Academy of Political and Social Science, 608, 157–178.

Larsen, O. N., & Hill, R. J. (1954). Mass media and interpersonal communication in the diffusion of a news event. American Sociological Review, 19, 426–433.

Lasswell, H. (1948). The structure and function of communication in society. In L. Bryson (Ed.), The communication of ideas (pp. 37–52). New York, NY: Harper & Brothers

Lazarsfeld, P. F. (1941). Administrative and critical communications research. Reprinted in J. D. Peters & P. Simonson (Eds.). (2004). Mass communication and American social thought: Key Texts, 1919–1968 (pp. 166–173). Lanham, MD: Rowman & Littlefield.

Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1944). The people's choice. New York, NY: Duell, Sloan & Pearce.

Lazarsfeld, P. F., & Fiske, M. (1938). The “panel” as a new tool for measuring opinion. Public Opinion Quarterly, 2(4), 596–612.

Lazarsfeld, P. F., & Menzel, H. (1963). Mass media and personal influence. In W. Schramm (Ed.), The science of human communication: New directions and findings in communication research (pp. 94–115). New York, NY: Basic Books.

Lazarsfeld, P. F., & Merton, R. K. (1948). Mass communication, popular taste, and organized social action. In L. Bryson (Ed.), The communication of ideas. Reprinted in J. D. Peters & P. Simonson (Eds.). (2004). Mass communication and American social thought: Key Texts, 1919–1968 (pp. 230–241). Lanham, MD: Rowman & Littlefield.

Lazarsfeld, P. F., & Stanton, F. N. (Eds.). (1944). Radio research 1942–1943. New York, NY: Duell.

Lerner, D. (1958). The passing of traditional society: Modernizing the Middle East. Glencoe, NY: Free Press.

Lowery, S. A., & DeFleur, M. L. (1995). Milestones in mass communication research: Media effects (3rd ed.). White Plains, NY: Longmans.

Lubken, D. (2008). Remembering the straw man: The travels and adventures of Hypodermic. In D. W. Park & J. Pooley (Eds.), The history of media and communication research: Contested memories (pp. 19–42). New York, NY: Peter Lang.

McCombs, M., & Shaw, D. (1972). The agenda-setting function of the mass media. Public Opinion Quarterly, 36, 176–187.

McCombs, M. E. (1981). The agenda-setting approach. In D. D. Nimmo & K. R. Sanders (Eds.), Handbook of political communication (pp. 121–140). Beverly Hills, CA: Sage.

McCormack, T. (2006). As time goes by. Annals of the American Academy of Political and Social Science, 608, 179–192.

McGuire, W. (1986). The myth of massive media power: Savagings and salvagings. In G. Comstock (Ed.), Public communication and behavior (vol. 1, pp. 173–257). Orlando, FL: Academic Press.

Mendelsohn, H. (1964). Broadcast vs. personal sources of information in emergent public crises: The presidential assassination. Journal of Broadcasting, 8, 147–156.

Merton, R. K. (1943). Time readership and the influence structure of Dover, NJ (an exploratory study). Office of Radio Research, Columbia University. RKM, personal files.

Merton, R. K. (1949). Patterns of influence: A study of interpersonal influence and of communications behavior in a local community. In P. F. Lazarsfeld & F. M. Stanton (Eds.), Communication research 1948–9 (pp. 180–222). New York, NY: Duell, Sloan & Pearce.

Merton, R. K. (1968). Social structure and social theory. New York, NY: Free Press. (Original work published 1949)

Merton, R. K., Fiske, M., & Curtis, A. (1946). Mass persuasion: The social psychology of a war bond drive. New York, NY: Harper & Brothers

Morrison, D. E. (1978). Kultur and culture; The case of Theodor W. Adorno and Paul F. Lazarsfeld. Social Research, 44, 331–355.

Morrison, D. E. (2000). The late arrival of television research: A case study in the production of knowledge. In H. Tumber (Ed.), Media power, professionals, and policies (pp. 127–148). London, UK: Routledge.

Morrison, D. E. (2006). The influences influencing Personal Influence: Scholarship and entrepreneurship. Annals of the American Academy of Political and Social Science, 608, 51–75.

Moschis, G. P., & Moore, R. L. (1982). A longitudinal study of television advertising effects. Journal of Consumer Research, 9(3), 279–286.

Nisbet, M. C., & Kotcher, J. E. (2009). A two-step flow of influence? Opinion-leader campaigns on climate change. Science Communication, 30(3), 328–354.

Noelle-Neumann, E. (1973). Return to the concept of powerful mass media. Studies of Broadcasting, 9, 67–112.

Norris, P., & Curtice, J. (2008). Getting the message out: A two-step model of the role of the Internet in campaign communication flows during the 2005 British general election. Journal of Information Technology and Politics, 4(4), 3–13.

Pooley, J. (2006). Fifteen pages that shook the field: Personal Influence, Edward Shils, and the remembered history of mass communication research. Annals of the American Academy of Political and Social Science, 608, 130–156.

Rogers, E. M. (1962). The diffusion of innovations. Glencoe, IL: Free Press.

Rogers, E. M. (1978). The rise and fall of the dominant paradigm. Journal of Communication, 28, 64–69.

Rogers, E. M., & Beal, G. M. (1958). The importance of personal influence in the adoption of technological changes. Social Forces, 36, 329–335.

Rothschild, M. L. (1978). Political advertising: A neglected policy issue in marketing. Journal of Marketing Research, 15, 58–71.

Scannell, P. (2006). Personal Influence and the end of the masses. Annals of the American Academy of Political and Social Science, 608, 115–129.

Schiller, H. I. (1989). Culture, Inc. New York, NY: Oxford University Press.

Schramm, W., & Roberts, D. F. (1971). The process and effects of mass communication (rev. ed.) Urbana, IL: University of Illinois Press.

Simonson, P. (Ed.). (2006). Politics, social networks, and the history of communications research: Rereading Personal Influence. Annals of the American Academy of Political and Social Science, 608.

Simonson, P. (2010). Refiguring mass communication: A history. Urbana, IL: University of Illinois Press.

Simonson, P., & Weimann, G. (2003). Critical research at Columbia. In E. Katz, J. D. Peters, T. Liebes, & A. Orloff (Eds.), Canonic texts in media research: Are there any? Should there be? How about these? (pp. 12–38). Cambridge, UK: Polity.

Simpson, C. (1994). Science of coercion: Communication research and psychological warfare 1945–1960. New York, NY: Oxford University Press.

Smith, J. S. (1945). How to study control of the media of communications. Bureau of Applied Research Report B-0248, Columbia University, New York, NY.

Southwell, B. G., & Yzer, M. C. (2007). The role of interpersonal communication in mass media campaigns. In C. S. Beck (Ed.), Communication yearbook 31 (pp. 420–463). New York, NY: Taylor & Francis.

Spitzer, S. P. (1964–1965). Mass media vs. personal sources of information about the presidential assassination: A comparison of six investigations. Journal of Broadcasting, 9, 45–50.

Sproule, J. M. (1997). Propaganda and democracy: The American experience of media and mass persuasion. Cambridge, UK: Cambridge University Press.

Summers, J. H. (2006). Perpetual revelations: C. Wright Mills and Paul Lazarsfeld. Annals of the American Academy of Political and Social Science, 608, 25–40.

Sussmann, L. (1956). FDR and the White House mail. Public Opinion Quarterly, 20(1), pp. 5–16. Reprinted in J. D. Peters & P. Simonson (Eds.). (2004). Mass communication and American social thought: Key Texts, 1919–1968 (pp. 401–408). Lanham, MD: Rowman & Littlefield.

Troldahl, V. (1963). Mediated communication and personal influence: A field experiment. (Unpublished PhD dissertation.) Minneapolis, MN, University of Minnesota.

Troldahl, V., & Van Dam, R. (1965). Face-to-face communication about major topics in the news. Public Opinion Quarterly, 29, 626–634.

Tuchman, G. (1978). Making news: A study in the construction of reality: New York, NY Free Press.

Tunstall, J. (Ed.). (1970). Media sociology: A reader. London, UK: Constable.

Valente, T. W. (1996). Social network thresholds in the diffusion of innovations. Social Networks, 18(1), 69–89.

Valente, T. W., & Rogers, E. (1995). The origins and development of the diffusion of innovations paradigm as an example of scientific growth. Science Communication, 16(3), 242–273.

Watts, D. J. (2007). The accidental influentials. Harvard Business Review, 85(2), 22–23

Watts, D. J., & Dorrs, P. S. (2007). Influentials, networks, and public opinion formation. Journal of Consumer Research, 34(4), 441–458.

Weimann, G. (1982). On the importance of marginality: One more step into the two-step flow of communication. American Sociological Review, 47, 764–773.

Weimann, G. (1994). The influentials: People who influence people. Albany: State University of New York Press.

Wheatland, T. (2009). The Frankfurt School in exile. Minneapolis, MN: University of Minnesota Press.

Wright, C. R. (1959). Mass communication: A sociological perspective. New York, NY: Random House.

Wright, C. R., & Cantor, M. (1967). The opinion seeker and avoider: Steps beyond the opinion leader concept. Pacific Sociological Review, 10, 33–43.

Zaller, J. (1996). The myth of massive media impact revived: New support for a discredited idea. In D. C. Mutz, P. M. Sniderman, & R. A. Brody (Eds.), Political persuasion and attitude change (pp. 17–78). Ann Arbor, MI: University of Michigan Press.