Journal of Communication

Arrington, Michael, Irvin

doi: 10.1111/j.1460-2466.2011.01591.xpmid: N/A

Originating at least as far back as the ancient Greeks, sizable bodies of research in communication studies, philosophy, and psychology have investigated processes of deception in human interaction. More recently, scholars have also inquired into the process of self-deception. However, extant research implies that the two phenomena are mutually exclusive, as evidenced by the lack of overlap between the two. In an effort to counter that assumption, Clancy Martin, a leading scholar on Nietzsche's theory of deception, has assembled a roster of 14 leading deception and ethics scholars in The Philosophy of Deception. Some of the contributors examine one type of deception, while others consider the two together. All of the studies point to the central argument of the book—that regarding deception of the self and deception of others together will allow researchers to understand each type of deception in greater detail. The first half of the book investigates what Martin calls the “hows” of deception and self-deception. In one of the book's strongest chapters, Paul Ekman, arguably the world's leading expert on lies and lie detection, investigates six possible explanations why most people are unable to detect attempts to deceive. From there, he shifts his discussion to the more specific area of lie detection via facial microexpressions. Harry Frankfurt, author of the popular books On Bullshit and On Truth, adds insight from the field of classical philosophy. Frankfurt contests the Kantian argument that lies undermine human society, arguing that lies have just as much potential to keep a society together. However, perhaps the most notable scholarly contribution in this section comes from philosopher Mark Wrathall, whose chapter offers a phenomenology of deception, insightfully complementing other approaches to this study of deception. Building on the prior work of Merleau-Ponty, Heidegger, and Nietzsche, Wrathall investigates the concept of perceptual deception and questions the responsibility of deceived individuals in their perception of deceptive messages. The second half of the book moves from the practice of deception to theory and ethics. A highlight of this section emerges from two chapters that offer conflicting views. In one essay, James Edwin Mahon considers Immanuel Kant's assertion that lies are inherently immoral acts, working from a narrow definition of lying. In the next chapter, Daniel Sussman critiques Kant's assumption of truthfulness as an unconditional, universal duty; he then offers an equally compelling argument for lying as an act of self-defense based on a broader definition of the term. The book's latter half also includes other notable essays. Amelie Rorty offers the book's strongest argument for viewing deception and self-deception together, describing the self by employing the metaphor of a medieval city with independently functioning neighborhoods. By combining the research of philosophers James Friedrich, Yaacov Trope, and Akiva Liberman (“FTL”), Alfred Mele provides the framework for a new “FTL” theory of self-deception. His theory attempts not only to define self-deception but also (by integrating self-deception into a theory of lay hypothesis testing) to provide insight into how it occurs. Communication researchers who are fortunate enough to find The Philosophy of Deception will be pleased to see a collection that informs both the interpersonal communication process of deception and the intrapersonal communication process of self-deception. However, Oxford University Press might have unwittingly overlooked the book's utility for communication scholars by marketing it primarily—nearly exclusively, in fact—as a philosophy of mind text. The publisher's website features the book in its philosophy section, under the “human nature” subheading, but omits it from the site's communication listings. It would be a shame if these essays did not contribute to contemporary communication scholarship merely because of poor marketing. Taken together, the essays in The Philosophy of Deception move beyond simplistic views of deception and consider the importance of deception and self-deception to communication. Martin's compilation suggests that these vilified phenomena are, at worst, necessary evils that are “crucial to our social interaction” (p. 5). In offering this addition to ongoing scholarly conversations about deception, Martin and his colleagues encourage readers to consider the nature and value, not only of truthfulness but also of truth itself. © 2011 International Communication Association

Willis-Chun,, Cynthia

doi: 10.1111/j.1460-2466.2011.01590.xpmid: N/A

In Places of Public Memory, editors Greg Dickinson, Carole Blair, and Brian Ott approach this study of public memory through the related topoi of rhetoric, memory, and place, taking seriously the task of reading places of memory through these three lenses simultaneously. The authors seek to answer several related questions: What function does (or could) rhetoric serve in regard to memory studies? What would it mean to take seriously the notion of “place” as it relates to memory? And, most importantly, how might an intersectional approach combining rhetoric, memory, and place serve to challenge and enlighten scholars working in these fields? In their introduction, the editors begin to tease out the tensions and possibilities created when memory and rhetorical studies meet. Throughout the book, rhetoric is positioned as a corrective for existing studies of public memory, and it begins with a set of fundamental assumptions, which are that “memory is rhetorical and that memory places are especially powerful rhetorically” (p. 2). This claim makes clear that rhetoric and memory are mutually constitutive, as rhetorical concerns are central to struggles over public memory, while at the same time places of public memory can be used as rhetorical resources. They further argue that place is essential to public memory because the situatedness and materiality of places of public memory contribute to their symbolic value as well as the possibilities for their signification. Blair, Dickinson, and Ott use the introduction to challenge many of “public memory's consensual premises” which they suggest lead to fundamental disconnects when rhetoricians accede to them too easily (p. 12). The authors explain how rhetoric, memory, and place function both separately and in combination, offering a cogent explanation of rhetoric's role in public memory places and their analysis. Beyond the book's contribution in carefully situating the rhetorical in terms of memory studies, Place of Public Memory is a valuable read because of the uniqueness of the places examined in later chapters. Dickinson, Blair, and Ott arranged the book into three sections—rhetoric, memory, and place—based on which of these topos is most prominent in each of the chapters. Section 1, which focuses on rhetoric, includes three essays. The first is by Bryan C. Taylor, who examines nuclear museums' post-Cold War navigation of nuclear technology, weaponry, and policy in the face of “nuclear ambivalence” on the part of the public (p. 67). He argues that “American nuclear museum rhetoric has promoted the interests of particular groups (such as national security officials and corporate contractors) over their competitors” (p. 67). Taylor analyzes multiple museum sites, taking into consideration the ways that local experiences of nuclear development influence exhibits—in short, he reflects on the ways in which the places of these museums impact the narratives the museums tell. Moving from the museal to the monumental, Victoria J. Gallagher and Margaret R. LaWare analyze The Fist (more formally, the Monument to Joe Louis), an enormous sculpture of a muscular forearm hung in downtown Detroit. The authors meld text and context admirably, studying The Fist as an artifact of visual and material culture even as they consider how its creation and presence “in the heart of a city radically reconfigured by the effects of racial strife” reveals the tensions of the period in which it was conceived and that persist today (p. 87). In a shift from the material to the ephemeral, Gregory Clark analyzes the National Jazz Museum in Harlem (NJMIH), an institution that as of yet has no actual museum building: “the museum's exhibits are, for now, Harlem itself—or, rather, people of the Harlem community who embody the musical heritage as well as the contemporary jazz culture of this place” (p. 121). Although Clark devotes much of his essay to an examination of the NJMIH's programming and plans for a more traditional museum space, he makes clear that “its primary goal is to describe the rhetorical power inherent in cultural experiences that are essentially aesthetic in their form” (p. 114). Clark argues that it is important for memory scholars to consider not just the institutions and exhibits that constitute public memory places, but the meaning that visitors derive from encounters with them and the attitudes that are created (and coached) thereby. Section 2 of the book focuses on memory and includes chapters from John Bodnar on memorialization of the Bataan death marches and Cynthia Duquette Smith and Teresa Bergman on tourist practices at Alcatraz Island. Bodnar contends that while national memorials to World War II (WWII) such as the Marine Corps (Iwo Jima) Memorial and the National World War II Memorial in Washington, DC are fervently heroic celebrations of patriotism, the same is not true of more local memorials. He notes that “the more memorials were tied to local experiences the more they seemed reluctant to accept the idea that the tragic aspects of the war should be completely forgotten” (p. 145). He concentrates his essay on memorials to the American defeat at Bataan, paying particular attention to the influence of local histories as they played out in these places of public memory. As much as Bodnar analyzes the memorials themselves, however, he commits himself to consideration of later WWII memorialization in these localities as it relates to the existing Bataan commemoration. For example, in 1999 the Santa Fe city council sought to mark the site of a nearby internment camp, a move that drew protest as a supposed slight to Bataan veterans. This outrage was not shared by all individuals, however, “Some vets … were more forgiving, and one indicated that when he had returned from the war he had noticed the guard towers at the local internment camp because they reminded him of his own imprisonment” (p. 150). Bodnar's essay makes clear that understandings of WWII memorialization as monolithic are misguided, and suggests that greater attention should be paid to local memorials' contributions to public memory. Whereas Bodnar compares the local and the national, Smith and Bergman examine Alcatraz Island as a local place of public memory nationally known for two major reasons but which is cast in terms of only one through its tourist experience. Smith and Bergman note that although some indications of the Native American Occupation of Alcatraz that occurred in 1969 are evident in the tour, the bulk of the tourist experience is dedicated to the island's history as a federal prison that housed infamous criminals. They contend that although the Alcatraz Island tour contributes to the public memory of this place, “[e]qually important here is what visitors do not get to experience” (p. 179). Although traces of the occupation are present on the island, visitors are not given a meaningful framework to interpret them. In particular, Smith and Bergman claim that because memories are most persistent when made material through sight, sound, touch, and interactivity, visitors to the island “are better able to relate to Al Capone and other notables whose cells they can actually peer into or enter” and whose presence on the island is explained in detail through tour audio (p. 179). The final section of Places of Public Memory focuses on memory, and features chapters by Michael S. Bowman on Jedburgh, Scotland's Mary Queen of Scots House and Visitor Centre (MQSH), Bernard J. Armada on the National Civil Rights Museum (NCRM) and a Memphis woman's countermemorial, and Eric Aoki, Greg Dickinson, and Brian L. Ott on the Draper Museum of Natural History. Bowman investigates the MQSH as a place where public memory has come to reside, regardless of its authenticity. He notes that although there is historical evidence placing Mary Queen of Scots in Jedburgh, there is no certainty that she stayed in the MQSH itself. In fact, the MQSH plays upon this tension in its exhibits, as it simultaneously offers evidence to authenticate materials and uses “narrative [that] leads one to draw the inference that the item must be genuine while cagily avoiding many any such claim, save for the identifying caption” (p. 197). Bowman interviewed several visitors during his time at the MQSH and found that they were savvier than might have been expected. He explains, “As visitors tour the building, they tack back and forth between a state of flow and reflexivity—that double consciousness inherent to performance—where one part of them is walking in Mary's footsteps, following her trail, while another is wholly aware that Mary may never have walked here at all” (p. 211). In this way, visitors seem to take pleasure in the very uncertainty that surrounds the MQSH, appreciating both its potential authenticity as a place of public memory and its refusal to position itself as such. In contrast to the MQSH, Armada's chapter begins from authenticity—the Lorraine Motel, site of Martin Luther King Jr's assassination and the NCRM—but quickly shifts to consider not the NCRM itself, but a countermemorial staged by Jacqueline Smith, an African American woman who “has lived on the street in protest since she was evicted from the Lorraine Motel in 1988” (p. 217). Armada considers the NCRM in light of its expansion (completed in 2002) as it impacts “the placement of bodies—those of visitors' [sic], that of Smith—that influences how memory is perceived in the museum” (p. 217). He notes that the NCRM's expansion has allowed the museum to remove Smith from her formerly prominent site of protest across the street from the balcony where King was murdered to a location that renders her nearly invisible to museum visitors. He argues, “In effect, then, pushing Smith out of sight also pushes her out of mind and discourages visitors from questioning the museum's hegemonic practices” (p. 227). Armada suggests that the museum's repeated displacement of Smith allows it to maintain the message of post-Civil Rights harmony, while at the same time furthering efforts at gentrification. He contends that sidelining Smith's protest denies visitors the opportunity to make sense of competing memories on their own, a process that he argues is essential in memorialization. Aoki, Dickinson, and Ott note a similar tension at the Draper, where both narratives and the physical layout of the museum conspire to situate visitors as “master naturalists” that bestow meaning upon nature through their exploration and protection of it. The authors note that the Draper uses directed movement to guide visitors through the museum in a manner that creates a sense of traversing various habitats and that limits visitors' ability to experience the museum in their own ways. Furthermore, the use of displays that simulate nature is sanitized, such that “[a]lthough the sensory content of the Draper echoes ‘real nature,’ the visitor is never fully vulnerable to the powerful vicissitudes of nature itself” (p. 253). Visitors are continually positioned as masters of the natural domain, overlooking sprawling vistas contained by helpful maps, meeting unthreatening stuffed beasts, and navigating the Greater Yellowstone region having gained the memory of engaging with nature without facing the perils (large and small) that actually doing so would entail. As any worthwhile book does, Places of Public Memory opens up as many avenues of inquiry as it pursues. Throughout the book, a few persistent themes arise that beg further consideration. For example, multiple chapters point to the question of why some memories “stick,” whereas others seem to fade so readily. Both Armada's piece on the NCRM and Smith and Bergman's essay on Alcatraz Island begin a preliminary investigation of how these sites have managed to privilege particular memories by all but obliterating others that are equally tied to these places. This leads to a second theme worthy of further thought, which is to deliberate more fully on the “public” of public memory. In their introduction, Blair, Dickinson, and Ott recognize that this as “a quite elastic concept,” and mention that rhetoricians have long been preoccupied with the notion of publics and publicity (p. 5). To this end, it seems useful to amend the question of “what sticks” to “what sticks and for whom?” Just as the uninitiated have no framework to understand the graffiti remnants of the Native American occupation, it seems equally likely that tourists lacking context for Smith's countermemorial at the NCRM would have little ability to understand or identify with her protest. Approaching the study of public memory places with a concomitant appreciation that the publicity of these places and the different publics engaging them have a significant impact on the way they are viewed seems a clear next step. The role of these places of public memory as (potential) tourist destinations is another avenue that warrants further exploration. Although some of the sites explored in the book seem to have (d)evolved into the mundanity of the everyday (such as the Monument to Joe Louis, which Gallagher and LaWare suggest has become a flash in the rearview mirror for Detroit commuters rather than a destination in its own right), others exist only by virtue of their ability to attract visitors (e.g., Alcatraz Island, nuclear museums, the Draper) and are therefore constructed toward this end. Although public memory scholars of all stripes have become adept at recognizing the political entanglements that accompany memory sites, as a whole they seem less attuned to the equally important political economy of such sites. Taking seriously the commercial side of public memory will allow scholars to better understand these places as situated within a framework that requires them to simultaneously remain attractive to tourists and somehow exist above the touristic fray. This kind of investigation would also open the door for scholars to consider how places of public memory might function differently for locals who see and may engage with them regularly versus visitors whose treks to the sites function as pilgrimages to experience history. Finally, the essays in Places of Public Memory seem to point to the concepts of materiality, physicality, and presence. Although Clark's essay on the NJMIH clearly shows that museums need not follow the brick-and-mortar tradition to succeed in capturing the imaginations of their publics, the very fact that the NJMIH is working on plans to create such a building suggests that there is a cachet associated with the material. Furthermore, the NJMIH's plans to create “Jazz Way,” an immersive experience wherein visitors could learn about the history of jazz while “local musicians working as docents would both converse and perform” for these visitors suggests that the museum hopes to capture the ephemeral (music, memory) through recourse to the immediate and material (pp. 124–125). Likewise, both Alcatraz Island and the Draper attempt to capture at least part of an authentic experience—in first case, by asking visitors to literally take the places of absent prisoners, if only for a moment, and in the latter, by creating displays that draw nature into the constructed space of the museum. Despite its tenuous relationship to authenticity MQSH functions similarly, allowing visitors to walk where Mary walked, or at least in a reasonable facsimile of where she might have trod. These places of public memory thus trade on the idea of presence by creating exhibits and tours that give visitors the feeling of “being there,” of taking part in the memory itself. Studies on this element of public memory could build upon existing research on memory, history, and authenticity, making performative aspects of memory places a concern for public memory scholars. I have no doubt that Places of Public Memory will inspire investigation in these and other areas, and look forward to seeing how it influences the future of rhetorical studies of public memory. © 2011 International Communication Association

Badaracco,, Claire

doi: 10.1111/j.1460-2466.2011.01589.xpmid: N/A

Joel Best is a master of writing pithy, concise books about important megatrends in contemporary mass culture. Among his earlier titles, readers would have enjoyed Flavor of the Month: Why Smart People Fall for Fads, More Damned Lies and Statistics, and Stat-Spotting: Field Guide to Identifying Dubious Data (all published by University of California Press). This volume, about being a winner in the era of what he calls “Status Abundance,” and the construction of social class in the Facebook era, is another good read with far-reaching implications for the field of the sociology of communication, mass culture, and education. Whether it is parents boasting on a bumper sticker about their child's excellence, or faculty sitting through long honors ceremonies where nearly everyone gets a prize, Best's hypothesis that we live in an age of Status Abundance is plainly visible in mass culture. His assertion that status and class are conflated forms the rationale for this mediation and explication of how we live surrounded by the need for and demands of self-congratulation. Best dissects the symbolic from the paradigmatic, and implies throughout the six-chapter book that our culture is shaped by its powerful beliefs in its own positivism, whether constructed, manufactured, or mediated. He cites Michael Barone's trenchant observation about the demographics of self-congratulation, that there are two periods of life: Soft America through the school years including higher education and Hard America from ages 18 to 30, when people are subjected to competition and accountability (p. 3). Best provides a vernacular with which to analyze the categories of status and its operation in social groups, and concludes his first chapter with the assertion that “Inflation” exists. Prizes are abundant, from Nobel to Academy Awards, the Marines to the Iowa State Fair: Best's second chapter describes the justifications and the skyrocketing number of prizes in all categories awarded in the past 60 years. The net effect is to treat esteem held by others and self-esteem as commodities, part of the audience to whom applause is critical. The most curious examples are from higher education, where the trend is a deflation of the Phi Beta Kappa honor society, long the highest congratulatory honor for college grads. Because of prize proliferation, Best argues, it becomes difficult to keep the prizes straight, and among the Facebook generation, some resistance has emerged, with students turning down invitations to join. The popularity of other honor societies, particularly among first-generation college students, has led to the deflation of the original most highly esteemed prize. Best argues that critics who argue that prize proliferation leads to deflation of social esteem are “easy to ignore” (p. 57), precisely because of the massification of congratulatory culture and because the audience demographics for prizes is changing. Best does not extend his argument into the question of how information is revalued and reclassified by the Googlization of data, but this is the elephant in the reading room, and the implications of his argument are clear, leaving the reader to deduce the rest. In Chapter 3, the longest chapter and the center argument in the book, Best explores the impact of this mass culture on the business of being a student. In this arena, two schools of thought prevail, he argues. The first is that people think about education as mastery of skills, of content, and the second school regards education as providing opportunity for social mobility. Opportunity advocates, for example, worry about racial inequalities in education and is suspicious of the mastery argument. Both ways of thinking about education result in differences in viewing educational public policy, and have far-reaching implications, as Best suggests. The conflation of status and social class, he mentions in the first chapter, is derived from this division in the prevailing thought about why people need education, and about how they ought to obtain it. More profoundly, he raises the consequence for our public culture if people are unable to attain basic mastery of important life skills. Again, while Best does not ride the theme too hard, there is a sleeper issue about technology as a distraction from learning: like the prize industry, when information is abundant, its real value is deflated. The fourth chapter is devoted to explicating our concept of heroism, as it shapes our perception of popular shows designed to raise the ratings of broadcast network news shows such as CNN's Heroes broadcast, where ordinary people were singled out as news stories with extraordinary impact. Typical of this broadcast was the interactive voting mechanism arranged for audience involvement, an extension of backpack journalism, where the audience for whom the show is intended is, in fact, shaping the outcomes of the show. This example leads into deeper water, into our public consciousness about 9/11, about how to honor military heroes fallen in war, about the notion of victims and survivors in natural disasters. Best cites the instance of Pat Tillman, certainly a parable for this age. First, a NFL football hero, then a Marine killed by friendly fire. Best does not go into the larger problem, where his death was valorized as part of the U.S. war effort in Iraq, and the documentary produced by his immediate family which exposed this valorization by the Bush Administration. The challenge to heroism permeates our age, then, for a number of reasons, which Best explores. Best extends this analysis to our politicians and elected officials who overpromise and overstate, only to be compromised later as the truth emerges that they fudged their resume or claims to war valor, or in other ways fall from grace. As a result, the mass audience that follows celebrities and invests belief in constructed reputations that become unmanageable, where governors, or corporations, all are part of the social processes in the congratulatory culture. The fifth chapter goes over rather familiar ground, about the ranking and ratings such as the U.S. News and World Report and its consequence for the reputation management of higher educational institutions, but raises the specter at the end of the chapter that there is a rebellion brewing against these rankings constructed on the basis of faculty–student ratio, grades, endowments, and the number of grants and awards accrued by faculty. Best argues that when people pay attention to ratings, the consequences are profound, because their choices pull them into a competitive arena driven by ratings. It is not the status and rankings he is objecting to, then, but the process of mediation which brings it about, the advertising and marketing communications arm of the news industry which has little knowledge about higher education itself that is determining consumers' choices. As those choices can make or break small schools, they become profound shapers of the educational infrastructure of the country. Finally, Best concludes the book with a chapter reflecting on the significance of this argument. Status exists because we construct it, believe in it, and treat it as a real social denominator, he argues. Meaning is not absolute, he argues, but is manufactured. This is a fine small volume, interesting and provocative in its implications, and would be an excellent addition to a graduate class and to a personal library. © 2011 International Communication Association

Balaji,, Murali

doi: 10.1111/j.1460-2466.2011.01592.xpmid: N/A

In the decade since Naomi Klein's groundbreaking No Logo, media scholars have tried to add to a continuum of scholarship on branding. However, notions of branding had, for the most part, focused on how corporations used branding as a means of cultivating consumption and building identities in global consumption markets. Branding, at least in media scholarship, has been inextricably tied to the increased studies of globalization and its impact. Much of today's scholarship builds off the works of Klein and Andrew Wernick's Promotional Culture, which 20 years after its publication continues to heavily influence critical scholarship of media industries. But in Blowing Up The Brand, a collection of essays largely from a NYU symposium of the same name, editors Melissa Aronczyk and Devon Powers expand the discourse of branding and what it means for both producers and consumers of culture. No longer confined to cultural and corporate industries to be consumed by masses, branding takes place in almost every nook of global society, as these essays attempt to articulate. One of the most compelling notions explored in this book is actually a challenge to Klein's thesis on anticorporate activism and its role as an outside force in disrupting multinational conglomerates. Aronczyk and Powers and Graham Knight argue that activists have indeed co-opted branding as a way of creating recognition for their causes, and as a result, social and political movements must rely on the same brand valuations that No Logo's primary villains—Nike, Gap, Reebok, and Shell—used to build their corporate identities in the 1980s and 1990s. As Knight warns in his piece, “if branding offers activist organizations benefits, these benefits come at a potential price internally with respect to relationships with constituents and supporters” (p. 191). As other essays in the volume seek to show, branding is rooted not only in production but also in performance and discursive constructions as well. The rhetorical possibilities of branding, as explained John Corner, as well as its political/spatial uses, as shown in essays by Gabrielle Consentino, Waddick Doyle, and Miriam Greenberg, remind us that branding is no longer just a tool of the multinationals, but an integral part of both public and private culture. The anthology's strengths are not just in how branding has permeated into all aspects of global institutions, but how individuals engage—consciously or otherwise—in the same branding strategies as multinationals. One of the emerging areas of scholarship explored in essays by Jefferson Pooley and Jonathan Gray, for example, are branding in the digital age. Although this book was published before Twitter's rise, there is enough source material for scholars of digital branding to use to expand this nascent area of research. Although it might be one of the most comprehensive works on branding and promotional culture, Blowing Up The Brand has its shortcomings. For starters, there continues to be a paucity of critical scholarship on how branding impacts audiences, both on a micro- and macro-level. With global brands such as Apple, Google, and even Facebook instantly recognizable in both households and public streets, it would have been worth adding some mention of consumer responses to the shifts in how branding has been deployed. Another area that the volume touches upon, but does not fully articulate, is the idea of celebrity branding. As recent works by Turner (2010) and Marshall (2010) show, celebrity branding is both inextricably tied to established systems of cultural production and distribution, as well as to McRobbie's (2002) notion that the boundaries between private and public have become indistinguishable. Although the volume seeks to show some of this dynamic in the introduction and in Consentino and Doyle's essay on Silvio Berlusconi, an additional essay or two on how public figures and celebrities use branding would have been useful. Another area of concern is the fact that neither the editors nor the contributors mention how branding is deployed upon or used by marginalized populations. Still, this volume provides for a great start to the next decade's worth of research on branding and promotional culture, as Wernick and Klein's pieces did in the 1990s and 2000s, respectively. By publishing this collection of essays, Aronczyk and Powers have given media scholars a strong push to examine branding in ways that both nuance and challenge the cultural production/consumption model. References Marshall , P. D . ( 2010 ). The promotion and presentation of the self: Celebrity as marker of presentational media . Celebrity Studies , 1 , 35 – 48 . Google Scholar Crossref Search ADS WorldCat McRobbie , A . ( 2002 ). From clubs to companies: Notes on the decline of political culture in the speeded up creative worlds . Cultural Studies , 16 , 516 – 531 . Google Scholar Crossref Search ADS WorldCat Turner , G . ( 2010 ). Approaching celebrity studies . Celebrity Studies , 1 , 11 – 20 . Google Scholar Crossref Search ADS WorldCat © 2011 International Communication Association

Vraga, Emily, K.;Edgerly,, Stephanie;Wang, Bryan, M.;Shah, Dhavan, V.

doi: 10.1111/j.1460-2466.2011.01581.xpmid: N/A

Abstract Changes in the information society, especially the rise of blogs, have refocused attention on questions of media modality, source identification, and motivation in online environments. We manipulate the structure of a blogger's critique on a news story (global vs. interspersed) and the partisan target of the blogger (Democrats vs. Republicans) in an experiment embedded in an online survey. Our results support our expectations: The more difficult story format decreases the ability of less motivated readers to correctly identify the source of their information, without affecting the motivated. These effects of structure on source identification are democratically consequential when people rely on blogs for facts about public affairs without the proper cautionary caveats regarding the credibility of the source. Political blogs play an increasingly large role in political communication, both in terms of the aims and topics addressed by bloggers—acting as a check to traditional news media and politicians—and in the growing number of blog readers (Farrell & Drezner, 2008). One of the distinguishing characteristics of blogs is the ability to incorporate traditional media stories into blog commentary (McKenna & Pole, 2008; Woodley, 2008). The political blog Daily Kos, for example, regularly integrates large portions of current news stories into their postings. The repurposing of mainstream news content exemplifies the more complex environment blogs present, where readers manage multiple sources of information within a single blog posting. These changes in the information society, especially the rise of blogs, have refocused attention on questions of media modality and source identification in online environments. The organization of information and commentary on political blogs, especially those focusing on critiques of mainstream news content, highlights novel issues of news information processing in mediated communication contexts, yet recalls earlier work on information structure and memory (Cohen, 1957; Eveland, Cortese, Park, & Dunwoody, 2004a; Haugtvedt & Wegener, 1994). And while issues of structure, processing, and motivation have been explored within controlled laboratory settings, few studies to date have examined the relationship among these factors in an online environment outside the lab. The studies that have focused on this topic have investigated structural differences within the same source—that is, an offline versus online copy of a newspaper (Tewksbury & Althaus, 2000)—or structural differences attributed to the website design—that is, the use of hyperlinks (Eveland et al., 2004a). This study explores a new area of online structure effects—the processing of information when a blogger repurposes a mainstream media article into their commentary in a highly juxtaposed and interruptive manner. With the heightened intertextuality of the contexts through which we receive news, the question of how information structure and individual motivation intersect to influence message processing becomes increasingly important. We suspect that structural differences in how news is repurposed will have the greatest influence among those with a lower motivation to process. To examine these issues, we test how the argument structure of blog commentary on press accounts affects readers' ability to correctly identify the source of information, extending research into source identification by linking it to structural and motivational differences. We expect that changes in structure will have differential influence depending on whether people are motivated to process the information carefully. Structural changes should be particularly consequential among those who dedicate low levels of effort to message processing, for whom breaks in the flow of the news account create barriers during encoding. These effects of structure on source identification are democratically consequential when people rely on blogs for facts about public affairs without the proper cautionary caveats regarding the credibility of the content. Literature review In a digital environment, where news can be repurposed and repackaged by bloggers, source identification becomes paramount. Most consequentially, journalists and bloggers abide by different standards of fact checking and norms of objectivity (McKenna & Pole, 2008; Perlmutter, 2008). Individuals who encounter news information through political blogs, especially in the form of author commentaries on referenced news texts, have to navigate a more complex environment. Being able to correctly identify the source of incoming information is vital to processing the news responsibly and maintaining skepticism about unverified information. Source confusion between news and political ads has been documented, especially when the persuasive message is designed to resemble the news (Yegiyan & Grabe, 2007). Furthermore, over time, a less credible source can sway opinions as much as a credible source (Hovland & Weiss, 1951–1952), suggesting the importance of studying audience information processing in these emerging mixed media environments. But these changes in how information is structured and presented in online settings are not likely to affect everyone equally, as people bring with them a variety of motivations and skills when processing information. People can differ in their inherent enjoyment of and willingness to engage in navigating complex information environments (Cacioppo & Petty, 1982; Haugtvedt, Petty, & Cacioppo, 1992), which may stimulate some, rather than others, to correctly identify the sources of new information despite the structure in which it is found. Furthermore, individuals' motivation to process information in these mixed media environments may be partly determined by the interplay between personal and contextual factors. For example, individuals tend to be motivated to protect previous beliefs and opinions in the face of contradictory arguments, which often leads to more thorough processing (Ditto & Lopez, 1992; Kunda, 1987, 1990). Regardless of the reason, motivation to process should prove influential in determining the effects of structure on different processing goals. These issues are highlighted in the referential nature of the increasingly large body of blog commentary. A prevalent style adopted by political bloggers is to cite, quote, or hyperlink to a news story and then offer their own perspective (Farrell & Drezner, 2008; McKenna & Pole, 2008). Two contrasting modes exist for the presentation of these intersecting texts: (a) global presentation of the news story followed by comments or (b) interspersed, interruptive juxtaposition of news and commentary. While these two modes are by no means the only forms of presentation available online, they represent two extremes in presentation style when bloggers critique a particular news story. Furthermore, these competing structural formats were selected for their real-world validity, as these formats can be seen on popular political blogs such as The Daily Kos and Little Green Footballs. In this study, we focus on this structural difference to examine how the manner in which information is organized and the motivations that readers bring to such communication contexts affect the subsequent encoding process. Source identification An important precursor to perceptions of information credibility is the ability to correctly identify its source. The general term source can refer to “a variety of characteristics that, collectively, specify the conditions under which a memory is acquired” (Johnson, Hashtroudi, & Lindsay, 1993, p. 3). In a news story, for example, the journalist may cite facts and opinions from a variety of sources he or she interviews. Or in a blog entry, the poster may cite and comment on a news article, thereby offering two general sources of information to the audience: blog and news. In this study, we are specifically concerned with the latter case (i.e., the media channels through which the audience acquires information) because this juxtaposition is especially pertinent within the blogosphere, where channels of varying levels of credibility coexist on the same page. Therefore, source identification is defined here as the ability of readers to accurately identify the original media channel in which specific information appears. Given the importance of correct source identification to decision making, the possibility of source confusion has concerned scholars for decades. For example, one study suggests that while people appear to differentiate between high and low credibility sources in the immediate aftermath of exposure, over time, this difference between the persuasive impact of the sources disappears (Hovland & Weiss, 1951–1952). Beyond persuasive impact, Yegiyan and Grabe (2007) found that although no immediate difference in source identification emerged between three types of television messages—conventional political ads, news-like political ads, and news stories—a week later, subjects were more likely to attribute information from news-like ads to news. But time may not be a necessary prerequisite for source confusion: Mares (1996) manipulated the visual similarity of news and fictional content and found evidence for both fiction-to-news and news-to-fiction source confusion. These studies demonstrate that message structure can impact source identification and persuasion, which has important implications for the study of communication effects in a digital media environment. To understand structural effects, it is necessary to examine how source identification is stored in memory. Source identification is especially taxing on cognitive resources—for example, it requires higher levels of attention than fact recall (Johnson, Kounios, & Reeder, 1992)—because it demands not only storage of the fact, but also maintaining a link between that fact and its source (Johnson et al., 1993). Studies on source identification draw heavily upon the source monitoring framework (Johnson, 1997), which is defined as “the set of processes involved in making attributions about the origins of memories, knowledge, and beliefs” (Johnson et al., 1993, p. 3). This framework highlights that many important memory characteristics are established at the time of memory formation, such that the structure in which individuals first encounter information should be particularly consequential. The ways political bloggers refer back to news articles, either through global presentation or through an interspersed format, appear to differ along these dimensions, making it especially relevant to consider how changes in structure influence individuals' ability to correctly identify the source of information. Structure The effects of the structure of information have long been studied “off line.” For example, the widely studied topics of primacy and recency effects demonstrate the importance of order in conveying information (Asch, 1946; Kassin, Reddy, & Tulloch, 1990). A more nuanced look at this phenomenon incorporates the “chunking” of information, or whether or not information clearly blocked or grouped into segments corresponding to valence (Petty, Tormala, Hawkins, & Wegener, 2001). The chunking of information has important effects on persuasive outcomes, though these effects are dependent on individuals' motivation to process (Petty et al., 2001). The concept of “chunking” is useful in demonstrating that the grouping of certain ideas can impact how participants make sense of the information they encounter. Moving beyond these traditional “offline” studies of information structure, more recent studies have taken advantage of the unique environment provided online by investigating structural differences on the Internet. For example, Tewksbury and Althaus (2000) found the differing structures of The New York Times and The Times on the Web impacted how much information participants could recall, with individuals reading the print version demonstrating higher recall. With regards to the structure of the Internet, Eveland and Dunwoody (2001) developed the theory of structural isomorphism, which asserts that the structure of the Internet has the ability to mimic the associative nature of human memory. Hyperlinks, for example, prompt readers to ask “how are these two things related?” thus enacting the node-link structure of human memory (Eveland & Dunwoody, 2002). In their examination of website structure, Eveland et al. (2004a) found a linear, print-like design produced a higher level of factual knowledge, while participants in the more user-dependent hypertext structure were able to convey a “denser” level of understanding by making connections between multiple issues. Intriguingly, the condition which required more work from participants, the user-dependent design, did not produce a higher level of knowledge, but did create a deeper level of understanding, suggesting that motivation and ability may play a role in the processing of nonlinear information structures. Despite studies looking at the structural effects of online news sites (Barnhurst, 2002; Eveland et al., 2004a; Eveland, Marton, & Seo, 2004b; Tewksbury & Althaus, 2000), more research needs to focus on the impact of structure in the blogosphere, especially as it becomes a more prominent source of political news (Pew, 2008; Rainie, Cornfield, & Horrigan, 2005). A blogger can potentially create associations between pieces of information depending on the structure of their critique. For example, in a structure where the news story and blog commentary are intertwined, the blogger can link a specific aspect of the news story to a specific commentary, thus facilitating the natural associations which structural isomorphism supports (see Eveland & Dunwoody, 2001). However, in more global structures, distilling specific points of criticism may be more difficult, as the news story information and the response commentary are distinguished into separated forms. Moreover, these differences in format should not affect everyone equally. Need for cognition When looking at the effects of changes in structure, research has consistently demonstrated people differ in their ability and motivation to process information (Chaiken, 1980; Petty & Cacioppo, 1979, 1986). One important individual difference that has been shown to influence reasoning and processing is need for cognition, or the “differences among individuals in their tendency to engage in and enjoy thinking” (Cacioppo & Petty, 1982, p. 116). People who are higher in need for cognition demonstrate a willingness to engage in effortful processing of material, a greater enjoyment of navigating complex material, and more knowledge of relevant ideas and arguments (Cacioppo & Petty, 1982; Cacioppo, Petty, Feinstein, & Jarvis, 1996). Need for cognition appears to be a stable individual difference in people's motivation to process information (Cacioppo et al., 1996; Cacioppo, Petty, & Morris, 1983). Need for cognition and its effects may be linked to the processing strategies used in evaluating and interpreting new information. For example, researchers suggest that individuals high in need for cognition are habitually more likely to use an effortful central route of processing, while those with a lower need for cognition are more likely to use peripheral cues (Axsom, Yates, & Chaiken, 1987; Haugtvedt et al., 1992; Priester & Petty, 1995). These different modes of processing suggest that structural changes can affect people with high versus low need for cognition differently. Generally, people with a high need for cognition demonstrate more effortful processing despite differences in cues that can discourage people with a low need for cognition, while individuals low in need for cognition can become motivated to use a central processing route by increasing the relevance of the story or their skepticism of peripheral cues (Axsom et al., 1987; Priester & Petty, 1995). Beyond altering story cues, changes in message structure also impact these groups differently, leaving people with a higher need for cognition less influenced by these variations. In one example, the order of ideas in a presentation only affected the opinions of people with a lower need for cognition, while those with a higher need for cognition remained immune (Cohen, 1957). Further, an experiment on “chunking,” or breaking up information to allow time for consideration, produced different attitudinal effects depending on need for cognition, with people higher in need for cognition demonstrating more effortful processing, especially under the “chunked” condition (Petty et al., 2001). Partisan incongruence Another individual difference that has proven consequential when evaluating news coverage is partisan identification, or more precisely partisan congruence or incongruence with the encountered content. Partisan identity, itself, plays an important role in determining individuals' perceptions of the news as a whole, as well as influencing their sources of news (Dalton, Beck, & Huckfeldt, 1998; Eveland & Shah, 2003; Pew, 2007). This identity is likely to become more salient when the group or its positions are attacked, such as in a political blog. Previous research has suggested that incongruent messages, or messages that disagree with a previously held stance or opinion, prompt greater attention, as people attempt to argue against incongruent information and selectively seek out congruent information (Ditto & Lopez, 1992; Edwards & Smith, 1996; Kunda, 1987, 1990; Lord, Ross, & Lepper, 1979; Taber & Lodge, 2006). This process is characterized by not only more thorough and effortful attention to the message itself, but also a more careful search through memory for ideas that disconfirm the incongruent information (Edwards & Smith, 1996; Kunda, 1990). While need for cognition speaks to an enduring predisposition to process, partisan identity is more situational, a function of whether individuals encounter information that disagrees with their political positions and prompts defensive processing. Nonetheless, both should heighten the motivation to process, making these individuals better equipped to handle complex structural changes. Hypotheses and research questions Although the Internet may alter traditional format and structure with the juxtaposition of different sources of media content, we contend that these changes will not affect everyone equally. Differences in how blogs structure repurposed news content should influence whether people accurately recall the source of new information, but these differences may be greater for some than for others. Differences in people's motivation to process information should greatly affect whether they are willing to put forth the mental effort to understand more challenging, interruptive, and interspersed formats (Chaiken, 1980; Petty & Cacioppo, 1986). When two authors, the news story and the blogger, are intertwined, we should see that motivation becomes especially important for recalling the source of new information. People with a higher need for cognition, given their enjoyment of complex tasks and their higher level of engagement with the material (Cacioppo & Petty, 1982; Cacioppo et al., 1996; Haugtvedt et al., 1992) should be unaffected by these changes in structure, as they thoroughly navigate the message. Conversely, those with a lower need for cognition will expend less mental effort and thus should find the interspersed commentary more confusing. Therefore, we hypothesize: H1a: Among individuals who are lower in need for cognition, those exposed to an interspersed commentary will have significantly lower total source identification compared to those exposed to a global commentary, but this difference will not be evident among people with higher need for cognition. Individual differences can also intersect with situational factors to alter people's motivation to process—for example, whether or not a message is congruent with partisan identity. Research suggests that people engage in more effortful processing when faced with incongruent or disconfirmatory information, as they seek to bring the new ideas in line with their preconceived beliefs and attitudes (Edwards & Smith, 1996; Kunda, 1987, 1990; Taber & Lodge, 2006). This motivated reasoning (Kunda, 1990), also referred to as defensive processing, is found when individuals are exposed to incongruent information. Such processing should limit the effects of changes in structure, while this same motivation to carefully process will not be apparent with exposure to congruent information. Therefore, we propose: H1b: Among individuals who are exposed to blog commentary congruent with their partisan identity, those exposed to an interspersed commentary will have significantly lower total source identification compared to those exposed to a global commentary, but this difference will not be evident among people exposed to incongruent blog commentary. Of course, given that the message juxtaposes two separate sources—a news story and a blogger's commentary—it may be that source identification mistakes are centered among one of these two sources, rather than the other. Differences in credibility between the sources may play a role in this variability: Research suggests that opinionated language is perceived as less credible (Hamilton & Hunter, 1998). While we expect motivation to play the same moderating role in navigating structural content, we examine whether this process is identical for news and blogger source identification. Therefore, we ask: RQ1a: Will the structure of the commentary have differential effects on news and blogger source identification among those lower in need for cognition? RQ1b: Will the structure of the commentary have differential effects on news and blogger source identification among those exposed to a congruent partisan message? Methods Our hypotheses were tested in an experiment embedded in a web survey in which participants viewed a fictitious news story about global climate change policy accompanied by commentary from a political blogger. Participants, who received extra credit, were undergraduate students at a large university in the Midwestern United States. The study (N = 877) was fielded in the spring of 2007. Respondents reported a liberal ideology for both social and economic issues and almost two-thirds of the sample identified themselves as Democrats (including those leaning Democrat). The students were predominantly female (about two-thirds), and averaged just over 20 years old. Study design Respondents read a news story about global climate change policy written to emulate journalistic practice by providing a balanced summary of two positions on climate change policy. The article described both a liberal policy position for addressing climate change, endorsing mandatory caps on emissions, as well as a contrasting conservative position, supporting voluntary emissions reductions and technological innovation by industry. The news story was attributed to the Associated Press and the content remained consistent across all experimental conditions. We chose to attribute our story to the AP, rather than a specific news source, to limit the influence a specific newspaper's label, and related perception of credibility, may have on processing. The news story was embedded in commentary from a fabricated political blogger named “Curt” within a structure that emulated the basic elements of a conventional blog post. The study used a 2 (structure of presenting information) × 2 (target of the blogger's attack) between-subjects design.1 The structure manipulation altered the format in which the news story was conjoined with the blogger's commentary, using either a “global” or “interspersed” format. In the global condition, participants viewed the entire news story, followed by the blogger's critique of that story. The interspersed condition did not change the content of the blogger's critique or the news story but the two were intermingled, with the blogger's critique interrupting the news story and commenting on specific chunks of the narrative (Appendix A). The blogger's commentary was also manipulated to critique either the Republican or Democratic policy position presented in the balanced news story. The partisan incongruence manipulation was created by comparing participants' reported political party identifications with the manipulated political target of the blogger's critique. Participants who reported Democrat affiliation and viewed blogger commentary critiquing the Republican position on global climate change policy were in the partisan congruence condition, as were Republican participants who read blog critiques of Democratic policy. Republicans who read critiques of the Republican position and Democrats who viewed critiques of the Democratic position were considered to be in the partisan incongruence conditions. Participants who reported either “Independent” or third-party affiliation were excluded from the analysis, leaving a total N of 773. Measures Source identification To measure source identification, respondents were asked to identify the source of a series of facts and claims as originating from either the news story or the blogger's comments (see Appendix B for question wording). The correctly identified items were summed to create a total source identification score (M = 4.38, Max = 8.00, SD = 1.88). The correct identification of items from the news story (M = 2.40, Max = 4.00, SD = 1.09) and blogger's commentary (M = 1.98, Max = 4.00, SD = 1.16) were separated for follow-up analysis. Need for cognition This construct was measured by two items on an 11-point scale that asked respondents about their enjoyment of problem-solving and their preference for complex problems (see Cacioppo & Petty, 1982). These items were averaged to create an index (M = 6.10, SD = 1.88, r =.50, p <.001). This index was split at the median to compare those with lower and higher need for cognition. The median split occurred at 6.0 on an 11-point scale, classifying 51.7% of respondents as lower in need for cognition and 48.2% as higher in need for cognition. General media trust To focus on the relationship between structure, motivation, and source identification, we controlled for general media trust, which is linked to news exposure and need for cognition (Tsfati & Cappella, 2005). The questions specifically gauged feelings toward mainstream media, as previous research has shown a negative relationship between mainstream media trust and preference for alternative media sources such as blogs (Johnson & Kaye, 2004). This variable was measured using two items on an 11-point scale asking about mainstream media trust and balance, which were averaged to create an index (M = 5.10, SD = 1.82, r =.67, p <.001). Results To test our hypotheses and research questions, we used a series of two-way ANCOVAs. In each of these tests, both the civility of the blogger's commentary and respondent's level of general media trust are controlled. To ensure that the effects of source identification are occurring separate from differences in learning from the manipulations, we first test their effects on fact recall—or accurate recall of information in the news story and blogger's commentary. We expect that the interaction between structure and motivation will not affect learning. The data show that this is the case: There is no main effect of structure nor is there an interaction with either need for cognition or partisan incongruence on fact recall. In all conditions, people demonstrate the same levels of learning, ensuring that any effects of the manipulations on source identification occur independently of the learning of new material in different conditions. Next, we tested our hypotheses about the effects of the intersection of need for cognition and partisan incongruence with structure on total source identification. H1a predicted that exposure to the interspersed structure would hinder source identification for people with lower need for cognition compared to the global condition, but would not affect people with higher need for cognition. The data support this hypothesis (see Table 1 for significance levels of the omnibus tests of the interactions). Planned comparisons demonstrate that among people with a lower need for cognition, those who viewed the interspersed blogger's commentary were significantly less likely (p =.007) to correctly identify the source of the information (M = 4.07, n = 204) compared to those who saw the global blogger's commentary (M = 4.56, n = 197), while no significant difference was apparent between global (M = 4.43, n = 188) and interspersed (M = 4.54, n = 184) commentary among those with a higher need for cognition (Figure 1). Table 1 ANCOVAs for Source Identification . Total Source Identification . News Source Identification . Blogger Source Identification . Civility 0.283 0.434 0.058 Media trust 1.903 1.008 1.637 Structure 2.059 3.588† 0.292 Need for cognition 1.629 0.159 2.809† Partisan incongruence 5.193* 0.224 10.338** Structure × Need for cognition 6.619* 3.452† 5.756* Structure × partisan incongruence 5.333* 2.735† 4.693* . Total Source Identification . News Source Identification . Blogger Source Identification . Civility 0.283 0.434 0.058 Media trust 1.903 1.008 1.637 Structure 2.059 3.588† 0.292 Need for cognition 1.629 0.159 2.809† Partisan incongruence 5.193* 0.224 10.338** Structure × Need for cognition 6.619* 3.452† 5.756* Structure × partisan incongruence 5.333* 2.735† 4.693* † p <.10. * p <.05. ** p <.01. Open in new tab Table 1 ANCOVAs for Source Identification . Total Source Identification . News Source Identification . Blogger Source Identification . Civility 0.283 0.434 0.058 Media trust 1.903 1.008 1.637 Structure 2.059 3.588† 0.292 Need for cognition 1.629 0.159 2.809† Partisan incongruence 5.193* 0.224 10.338** Structure × Need for cognition 6.619* 3.452† 5.756* Structure × partisan incongruence 5.333* 2.735† 4.693* . Total Source Identification . News Source Identification . Blogger Source Identification . Civility 0.283 0.434 0.058 Media trust 1.903 1.008 1.637 Structure 2.059 3.588† 0.292 Need for cognition 1.629 0.159 2.809† Partisan incongruence 5.193* 0.224 10.338** Structure × Need for cognition 6.619* 3.452† 5.756* Structure × partisan incongruence 5.333* 2.735† 4.693* † p <.10. * p <.05. ** p <.01. Open in new tab Figure 1 Open in new tabDownload slide Effects of need for cognition and structure on total source identification. Figure 1 Open in new tabDownload slide Effects of need for cognition and structure on total source identification. Further, H1b predicted the same interaction would occur between partisan incongruence and structure. We again found a significant interaction and performed a pairwise comparison to test our hypothesis that differences in source identification would emerge only among those exposed to a congruent partisan message. The results support this hypothesis: For people exposed to the congruent partisan message, an interspersed structure significantly (p =.005) decreased source identification (M = 3.99, n = 188) compared to a global structure (M = 4.51, n = 194), while no significant differences emerged among those exposed to incongruent commentary between the interspersed (M = 4.63, n = 200) and global (M = 4.48, n = 191) structures (Figure 2). Figure 2 Open in new tabDownload slide Effects of partisan incongruence and structure on total source identification. Figure 2 Open in new tabDownload slide Effects of partisan incongruence and structure on total source identification. We next test our research questions, which asked whether the interaction between motivation and structure functions the same for news and blogger source identification. For news source identification, we looked at the pairwise comparisons to see if the effects on source identification are again centered among those with a lower need for cognition. We see a similar pattern: Among those with a lower need for cognition, an interspersed structure significantly decreases (p =.011) news source identification (M = 2.26) compared to a global structure (M = 2.53), while the difference among those with a higher need for cognition is not statistically significant for the interspersed (M = 2.42) versus global (M = 2.44) structures (Figure 3). Figure 3 Open in new tabDownload slide Effects of need for cognition and structure on news source identification. Figure 3 Open in new tabDownload slide Effects of need for cognition and structure on news source identification. We also look at the pairwise comparisons for blogger's source identification to investigate this interaction between need for cognition and structure. In this case, the pairwise comparisons suggest among those with a lower need for cognition, the interspersed commentary marginally (p =.051) decreases blogger source identification (M = 1.81) compared to the global commentary (M = 2.03), but we see no difference among those with a higher need for cognition between the interspersed (M = 2.12) and global (M = 1.99) structures (Figure 4). Figure 4 Open in new tabDownload slide Effects of need for cognition and structure on blogger source identification. Figure 4 Open in new tabDownload slide Effects of need for cognition and structure on blogger source identification. Finally, we test the interaction between partisan incongruence and structure against both news and blogger source identification. For news source identification, we see a similar pattern in the pairwise comparisons. Again, a significant difference emerges in the partisan congruence condition (p =.008) between the interspersed (M = 2.25) and global (M = 2.54) structures, while this difference is not significant for the individuals exposed to incongruent commentary between the interspersed (M = 2.43) and global (M = 2.43) structures (Figure 5). Figure 5 Open in new tabDownload slide Effects of partisan incongruence and structure on news source identification. Figure 5 Open in new tabDownload slide Effects of partisan incongruence and structure on news source identification. Looking at blogger source identification, the pairwise comparison suggests that for those exposed to a politically congruent message, the interspersed structure (M = 1.73) significantly decreased blogger source identification (p =.039) compared to the global structure (M = 1.98), while among those exposed to an incongruent message, the difference between the interspersed (M = 2.20) and global (M = 2.05) structures is not significant, fitting the overall pattern observed (Figure 6). Figure 6 Open in new tabDownload slide Effects of partisan incongruence and structure on blogger source identification. Figure 6 Open in new tabDownload slide Effects of partisan incongruence and structure on blogger source identification. Discussion This article provides a preliminary exploration of how structural differences online, specifically the more intermingled structure of blogs, can affect people's processing of new information. It also extends previous research into the impact of motivations on processing into a new context: the close juxtaposition of news and commentary apparent on political blogs like The DailyKos and Little Green Footballs. This study suggests that while changes in format do not significantly affect people's fact recall, it can affect their ability to correctly identify the source of their learning. Among people with a lower motivation to process, specifically those with a lower need for cognition or those confronted with congruent partisan messages, the change in structure, from a global—that is, a clear distinction between a news story and blogger's commentary—to an interspersed structure—that is, a more interruptive format in which the article and commentary intersect—affects their ability to correctly identify information, leading to misidentification of whether “facts” emanated from a news report or from blog commentary. However, those motivated to process the information thoroughly, either by individual differences or contextual circumstances, are able to navigate the more complex intermingled structure without confusion. This pattern holds for individuals' ability to correctly identify the information coming from the news story and from the blogger—in other words, those with a higher motivation to process are not more likely to misidentify information from either source. On the other hand, for those not motivated to process as thoroughly and carefully, an interspersed structure made it more difficult to identify the source of ideas regardless of whether they came from the repurposed news report or the blogger's commentary. Although the notion that those less motivated to process will have greater difficulty in accurately recalling information is hardly novel, our article demonstrates two possible techniques by which this can occur: an individual difference in need for cognition and a contextual difference in information congruence. And while these individual differences may be relatively static, the shift in motivation at the contextual level suggests greater attention must be paid to the environment in which information is received. These motivational shifts that occur as a result of contextual cues may be especially consequential in the new media environment, particularly when combined with a more difficult intermingled structure. Furthermore, the resulting effects on accurate information recall may have important democratic implications as people transfer to more news consumption in online contexts. Therefore, this study extends previous research to suggest that the new information environment online, in which objective news content is often juxtaposed with opinionated commentary, can have consequences for people's information processing strategies. Although the tendency of individuals with stronger motivations to use a more careful form of processing is well documented (Chaiken, 1980; Petty & Cacioppo, 1979, 1986), it has yet to be studied in the increasingly used medium of political blogs (Pew, 2008). Further, research into the differences in structure between online and offline content has examined levels of factual knowledge and information recall (Eveland & Dunwoody, 2001, 2002), but this study examines source identification in a context in which assertion and fact are more closely mingled. As political blogs often use traditional media sources as a platforms upon which to critique and comment (McKenna & Pole, 2008), it is democratically necessary to consider whether people can distinguish between these different sources in forming their opinion. If people are confused about the source of their information, they may be unable to accurately judge its credibility or appropriately integrate this new information into their existing knowledge structures. Furthermore, the difficulties in interpreting and identifying online information do not affect everyone equally, raising concerns about learning from cross-cutting media environments especially among those who are not highly motivated to process news. The results suggest that an interspersed structure may hinder the process of correctly identifying the source of information among those less engaged in processing—those lower in need for cognition or in a politically congruent message environment. The fact that this occurs despite equal learning from both types of structure suggests that those low in motivation to process may “learn” from bloggers' repurposing of news content but may also mislearn information if they are unable to distinguish which ideas have been vetted through an editorial process and which ones are the unfiltered views of a partisan blogger. If an interspersed structure suppresses source identification among low information processors, they may be accepting information as factual without carefully considering its merits. This may have important implications for democratic functioning and decision-making given the changing nature of online news environments. These findings become especially troubling if people are using the Internet to seek out more confirmatory sources of information. The Internet in particular may offer individuals the opportunity to selectively attend to information that is congruent with their previous opinion, especially politically (Adamic & Glance, 2005; Adams, 1961; Garrett, 2009). If, as this study suggests, people exposed to politically congruent information are not motivated to process information carefully, they could be especially prone to mingling objective and opinionated commentary in their search for facts. Therefore, it becomes vital to more carefully study source confusion online, as the Internet and political blogs may be not only intermingling content but also allowing people to selectively expose themselves to agreeable information, promoting further confusion in information sources. Conversely, it is possible that blog readers are likely to be individuals higher in need for cognition. If only motivated readers were attending to political blogs, this would lessen the impact of our study. However, our study suggests the relationship between need for cognition and blog readership is modest (r =.102, p <.01). Our data also suggest that individuals with both higher and lower need for cognition are engaging in blog readership online—in fact, nearly 39% of those with lower need for cognition in our study report reading political blogs, while roughly 59% of young voters in 2008 turned to political blogs for information (Pew, 2008). In conjunction with confirmatory information seeking online, many individuals are susceptible to the source confusion noted in this study when exposed to intermingled content. Of course, there are several limitations to consider involving the study manipulations and sample that future research should address. While our study produced a clear pattern of results, the effect sizes are relatively small. However, this study examined exposure to a single political blog entry at a single point in time. If people consume blogs as part of their standard diet, the cumulative effects of misremembering facts are likely to be substantial (Kull, Ramsey, & Lewis, 2003–2004). Also, while many of the facts were specific to the fabricated policy, individuals could have known some of the others (such as those dealing with the Kyoto Accords) without reading the story. Although these differences in pre-existing knowledge would be randomized across conditions and would not affect our pattern of findings, they may create a floor effect for misidentification, limiting our ability to recognize the source confusion individuals experienced. Two more general limitations concern the use of a college sample and the issue of global climate change. First, the college sample limits our ability to generalize to the population. However, in studying blog effects it is important to note that several studies have found that blog readers tend to be college educated (Kaye, 2005; Rainie, 2005). Furthermore, recent research suggests there is mixed empirical evidence regarding the age, gender composition, and media consumption patterns of blog readers, although the effects of education are relatively consistent (Eveland & Dylko, 2007). Therefore, the use of a college sample may more closely approximate the population most likely to consume news through political blogs. Second, the issue of global climate change could be of greater importance to a college sample and, consequently, our results may be indicative of the specific issue rather than the manipulations. However, global climate change is also an issue does not split cleanly along party lines, especially among college students. Measures of support for government action against global warming were moderately correlated with party identification (r =.35, p <.001), which suggests that some individuals did not favor the partisan position being advocated. Despite this, the salience of the political attack on the party should have created partisan incongruence, even for those who did not agree with the issue position. Future studies should test multiple issues, especially those with a more divisive impact, when exploring the effects of blog structure to ensure our results are not unique to this issue. We expect the effects of partisan incongruence will be heightened with an issue that more clearly divides partisans, as it should multiply the sources of incongruence. Ultimately, this study points to an important new direction for media studies. Although the study of blogs has become more prevalent in mass communication theory (Farrell & Drezner, 2008; Kaye, 2005), it is important to consider how these increasingly important forums repurpose news content and how the structure of such repurposing impacts information processing. Blogs operate in tandem with traditional news media, often borrowing, reworking, and responding to their content (McKenna & Pole, 2008). Future research should advance this study's findings by looking more carefully at a variety of structures available online and different motivations that could influence processing. If blogs are encouraging people to accept information without carefully considering its source or its veracity, this could have important implications in a democracy that depends on a rational and thoughtful public. 2 Notes 1 " This 2 × 2 design was extracted from a larger experimental design. The full design included a third manipulation related to the civility of the blog post. This variable did not interact with the structural or congruence manipulation and is included in the analysis only as a statistical control. 2 " Brackets indicate slight changes in question wording to match the slightly altered stimulus for the commentary attacking either the Democratic and Republican position. The correct answer is indicated in bold after the question. Appendices Appendix A: Screen shots of structure manipulations Appendix B: Question wording Political identification Which of the following best describes your party affiliation? Strong Democrat, Democrat, Independent-leaning Democrat, Independent, Independent-leaning Republican, Republican, Strong Republican Need for cognition Here are some of the statements that people find useful in describing themselves. For each of them, please indicate how accurately it describes you. I really enjoy a task that involves coming up with new solutions to problems. I prefer complex problems to simple ones. General media trust Please indicate your level of agreement with the following statements: I trust the information I find in the mainstream news media. Most mainstream news media present information in a balanced way. Source identification According to what you just read, attribute each of the following statements to the appropriate source (news article or blogger's commentary):2 Rep. [Gill's/Hunt's] proposal would require greenhouse-gas emissions to return to 2004 levels by 2012. News article Rep. [Gill's/Hunt's] proposal would result in a 1° decrease in surface temperature by 2020. Blogger's commentary The [Democratic proposal/Republican] is estimated to cost $2,600 per household per year. Blogger's commentary U.S. emissions of carbon dioxide have increased 1% per year since 1990. News article It is [uncertain/certain] whether global temperature increases are due to human activity. Blogger's commentary Technological innovations and voluntary reductions have already begun slowing the growth of greenhouse gases. News article [Most of the nations that signed the Kyoto accord have failed to meet its requirements/A majority of Kyoto signatories have made progress toward reducing greenhouse gas emissions]. Blogger's commentary President Bush renounced the Kyoto accord in 2001. News article References Adamic , L. A. , & Glance , N. ( 2005 ). The political blogosphere and the 2004 election: Divided they blog . Communications of the ACM , 36 – 43 . OpenURL Placeholder Text WorldCat Adams , J. S . ( 1961 ). Reduction of cognitive dissonance by seeking consonant information . Journal of Abnormal and Social Psychology , 62 , 74 – 78 . Google Scholar Crossref Search ADS PubMed WorldCat Asch , S . ( 1946 ). Forming impressions of personality . Journal of Abnormal and Social Psychology , 410 , 258 – 290 . Google Scholar Crossref Search ADS WorldCat Axsom , D. , Yates , S., & Chaiken , S. ( 1987 ). Audience response as a heuristic cue in persuasion . Journal of Personality and Social Psychology , 53 , 30 – 40 . Google Scholar Crossref Search ADS PubMed WorldCat Barnhurst , K. G . ( 2002 ). News geography and monopoly: The form of reports on U.S. newspaper Internet sites . Journalism Studies , 3 , 477 – 489 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , & Petty , R. E. ( 1982 ). The need for cognition . Journal of Personality and Social Psychology , 42 , 116 – 131 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , Petty , R. E., Feinstein , J. A., & Jarvis , W. B. G. ( 1996 ). Dispositional differences in cognitive motivation: The life and times of individuals varying in need for cognition . Psychological Bulletin , 119 , 197 – 253 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , Petty , R. E., & Morris , K. J. ( 1983 ). Journal of Personality and Social Psychology , 45 , 805 – 818 . Crossref Search ADS Chaiken , S . ( 1980 ). Heuristic versus systematic information processing and the use of source versus message cues in persuasion . Journal of Personality and Social Psychology , 39 , 752 – 766 . Google Scholar Crossref Search ADS WorldCat Cohen , A. R . ( 1957 ). Need for cognition and order of communication as determinants of opinion change. In C. I. Hovland (Ed.), The order of presentation in persuasion (pp. 79 – 97 ). New Haven, CT : Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dalton , R. J. , Beck , P. A., & Huckfeldt , R. ( 1998 ). Partisan cues and the media: Information flows in the 1992 presidential election . American Political Science Review , 92 , 111 – 126 . Google Scholar Crossref Search ADS WorldCat Ditto , P. H. , & Lopez , D. F. ( 1992 ). Motivated skepticism: Use of differential decision criteria for preferred and nonpreferred conclusions . Journal of Personality and Social Psychology , 63 , 568 – 584 . Google Scholar Crossref Search ADS WorldCat Edwards , K. , & Smith , E. E. ( 1996 ). A disconfirmation bias in the evaluation of arguments . Journal of Personality and Social Psychology , 7 , 5 – 24 . Google Scholar Crossref Search ADS WorldCat Eveland , W. P. , Jr., Cortese , J., Park , H., & Dunwoody , S. ( 2004a ). How Web site organization influences free recall, factual knowledge, and knowledge structure density . Human Communication Research , 30 , 208 – 233 . Google Scholar Crossref Search ADS WorldCat Eveland , W. P. , Jr., & Dunwoody , S. ( 2001 ). User control and structural isomorphism or disorientation and cognitive load? Learning from the Web versus print . Communication Research , 28 , 48 – 78 . Google Scholar Crossref Search ADS WorldCat Eveland , W. P. , Jr., & Dunwoody , S. ( 2002 ). An investigation of elaboration and elective scanning as mediators of learning from the Web versus print . Journal of Broadcasting & Electronic Media , 46 , 34 – 53 . Google Scholar Crossref Search ADS WorldCat Eveland . W. P. , Jr, & Dylko , I. ( 2007 ). Reading political blogs during the 2004 election campaign: Correlates and political consequences. In M. Tremayne (Ed.), Blogging, citizenship, and the future of media . New York : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Eveland , W. P. , Jr., Marton , K., & Seo , M. ( 2004b ). Moving beyond “just the facts”: The influence of online news on the content and structure of public affairs knowledge . Communication Research , 31 , 82 – 108 . Google Scholar Crossref Search ADS WorldCat Eveland , W. P. , & Shah , D. V. ( 2003 ). The impact of individual and interpersonal factors on perceived news media bias . Political Psychology , 24 , 101 – 117 . Google Scholar Crossref Search ADS WorldCat Farrell , H. , & Drezner , D. W. ( 2008 ). The power and politics of blogs. Political Choice , 134 , 15 – 30 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Garrett , R. K . ( 2009 ). Echo chambers online? Politically motivated selective exposure among Internet news users . Journal of Computer-Mediated Communication , 14 , 265 – 285 . Google Scholar Crossref Search ADS WorldCat Hamilton , M. A. , & Hunter , J. E. ( 1998 ). The effect of language intensity on receiver evaluations of message, source, and topic. In M. Allen & R. W. Preiss (Eds.), Persuasion: Advances through meta-analysis (pp. 99 – 138 ). Cresskill, NJ : Hampton Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Haugtvedt , C. P. , Petty , R. E., & Cacioppo , J. T. ( 1992 ). Need for cognition and advertising: Understanding the role of personality variables in consumer behavior . Journal of Consumer Psychology , 1 , 239 – 260 . Google Scholar Crossref Search ADS WorldCat Haugtvedt , C. P. , & Wegener , D. T. ( 1994 ). Message order effects in persuasion: An attitude strength perspective . Journal of Consumer Research , 21 , 205 – 218 . Google Scholar Crossref Search ADS WorldCat Hovland , C. I. , & Weiss , W. ( 19511952 ). The influence of source credibility on communication effectivenss . The Public Opinion Quarterly , 15 , 635 – 650 . Google Scholar Crossref Search ADS WorldCat Johnson , M. K . ( 1997 ). Identifying the origin of mental experience. In M. S. Myslobodsky (Ed.), The mythomanias: The nature of deception and self-deception (pp. 133 – 180 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Johnson , M. K. , Hashtroudi , S., & Lindsay , D. S. ( 1993 ). Source monitoring . Psychological Bulletin , 114 , 3 – 28 . Google Scholar Crossref Search ADS PubMed WorldCat Johnson , T. J. , & Kaye , B. K. ( 2004 ). Wag the blog: How reliance on traditional media and the Internet influence credibility perceptions of weblogs among web users . Journalism and Mass Communication Quarterly , 81 , 622 – 642 . Google Scholar Crossref Search ADS WorldCat Johnson , M. K. , Kounios , J., Reeder , J. A. ( 1992 , July). Time course studies of reality monitoring and recognition. Paper presented at the 33rd annual meeting of the Psychometric Society, St. Louis, MO. Kassin , S. M. , Reddy , M. E., & Tulloch , W. F. ( 1990 ). Juror interpretations of ambiguous evidence: The need for cognition, presentation order, and persuasion . Law and Human Behavior , 14 , 43 – 55 . Google Scholar Crossref Search ADS WorldCat Kaye , B. K . ( 2005 ). It's a blog, blog, blog, blog world . Atlantic Journal of Communication , 13 , 73 – 95 . Google Scholar Crossref Search ADS WorldCat Kull , S. , Ramsey , C., & Lewis , E. ( 20032004 ). Misperceptions, the media, and the Iraq war . Political Science Quarterly , 118 , 569 – 598 . Google Scholar Crossref Search ADS WorldCat Kunda , Z . ( 1987 ). Motivated inference: Self-serving generation and evaluation of causal theories . Journal of Personality and Social Psychology , 53 , 646 – 647 . Google Scholar Crossref Search ADS WorldCat Kunda , Z . ( 1990 ). The case for motivated reasoning . Psychological Bulletin , 108 , 480 – 498 . Google Scholar Crossref Search ADS PubMed WorldCat Lord , C. G. , Ross , L., & Lepper , M. R. ( 1979 ). Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence . Journal of Personality and Social Psychology , 37 , 2098 – 2109 . Google Scholar Crossref Search ADS WorldCat Mares , M. L . ( 1996 ). The role of source confusions in television's cultivation of social reality judgments . Human Communication Research , 23 , 278 – 298 . Google Scholar Crossref Search ADS WorldCat McKenna , L. , & Pole , A. ( 2008 ). What do bloggers do: An average day on an average political blog . Public Choice , 134 , 97 – 108 . Google Scholar Crossref Search ADS WorldCat Perlmutter , D. D . ( 2008 ). Blogwars . Oxford, England : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Cacioppo , J. T. ( 1979 ). Issue involvement can increase or decrease persuasion by enhancing message-relevant cognitive responses . Journal of Personality and Social Psychology , 37 , 1915 – 1926 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , & Cacioppo , J. T. ( 1986 ). The elaboration likelihood model of persuasion . Advances in Experimental Social Psychology , 19 , 123 – 205 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Tormala , Z. L., Hawkins , C., & Wegener , D. T. ( 2001 ). Motivation to think and order effects in persuasion: The moderating role of chunking . Personality and Social Psychology Bulletin , 27 , 332 – 344 . Google Scholar Crossref Search ADS WorldCat Pew Research Center for the People & the Press. ( 2007 ). Views of press values and performance. Report from the Pew Research Center for the People & the Press. Pew Research Center For the People & the Press. ( 2008 ). More than a quarter of voters read political blogs. Report from the Pew Research Center for the People & the Press. Priester , J. R. , & Petty , R. E. ( 1995 ). Source attributions and persuasion: Perceived honesty as a determinant of message scrutiny . Personality and Social Psychology Bulletin , 21 , 637 – 654 . Google Scholar Crossref Search ADS WorldCat Rainie , L . ( 2005 ). Pew Internet and American life project. Retrieved from http://www.pewinternet.org/pdfs/PIP_blogging_data.pdf. Rainie , L. , Cornfield , M., & Horrigan , J. ( 2005 ). The Internet and campaign 2004 . Washington, DC : Pew Internet & American Life Project . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Taber , C. S. , & Lodge , M. ( 2006 ). Motivated skepticism in the evaluation of political beliefs . American Journal of Political Science , 50 , 755 – 769 . Google Scholar Crossref Search ADS WorldCat Tewksbury , D. , & Althaus , S. ( 2000 ). Differences in knowledge acquisition among readers of the paper and online versions of a national newspaper . Journalism & Mass Communication Quarterly , 77 , 457 – 479 . Google Scholar Crossref Search ADS WorldCat Tsfati , Y. , & Cappella , J. N. ( 2005 ). Why do people watch news they do not trust? The need for cognition as a moderator in the association between news media skepticism and exposure . Media Psychology , 7 , 251 – 271 . Google Scholar Crossref Search ADS WorldCat Woodley , D . ( 2008 ). New competencies in democratic communication? Blogs, agenda setting and political participation . Public Choice , 134 , 109 – 123 . Google Scholar Crossref Search ADS WorldCat Yegiyan , N. S. , & Grabe , M. E. ( 2007 ). An experimental investigation of source confusion in televised messages: News versus entertainment . Human Communication Research , 33 , 379 – 395 . Google Scholar Crossref Search ADS WorldCat © 2011 International Communication Association

Edy, Jill, A.;Snidow, Shawn, M.

doi: 10.1111/j.1460-2466.2011.01584.xpmid: N/A

Abstract News coverage following Al Gore and George W. Bush's appearances on the Oprah Winfrey Show during the 2000 presidential campaign shows journalism asserting its authority to manage political discourse despite competition from alternative media. Analysis using Foucault's concepts of knowledge, power, and discipline reveals journalism affirming its continuing relevance and integrates insights from the framing and paradigm repair traditions. Journalists rejected Oprah as political discourse but reframed its elements to meet news criteria established by institutional journalism. Using negative stereotypes of women as political actors, journalists also disciplined Winfrey's “mostly female” audience for failing to adequately enact citizenship. Journalists thus both reasserted authority to manage political discourse and set standards for citizenship that positioned journalism as necessary to democracy. A growing body of literature addresses the social consequences of what might be called infotainment politics. Research has documented voters' use of nontraditional sources of political information (Pew Center for People and the Press, 2000) and the different styles of information presentation that distinguish alternative and traditional media (e.g., Baym, 2005; Jones, 2001) as well as debating the effects of soft news and infotainment on users (e.g., Baum, 2003; Brewer & Cao, 2006; Prior, 2003). This study, in contrast, considers how journalism asserts its social relevance and responds to entertainment media's threat to its institutional authority as the legitimate arbiter of political discourse. To do so, it examines a relatively early challenge from the current “new media” era: two key candidate talk show appearances during the 2000 campaign. Shortly after their respective party conventions, both George W. Bush and Al Gore made hour-long appearances on the Oprah Winfrey Show where they were interviewed by Winfrey in front of a live studio audience. Political figures have appeared on entertainment television programs since the 1960s, but these high-profile appearances represented a potent challenge to journalism's authority to moderate candidates' appearances before the public. By 2000, there was a growing trend of candidates “doing” talk shows, which undermined journalists' power to manage political discourse. Moreover, the length of the interviews, the ability of voters to contrast the candidates' appearances on the show, the size of Winfrey's audience, and Winfrey's reputation for socially engaged “talk” made the candidate interviews a near-miss to journalism (especially Sunday morning talk). Thus, this case is prominent enough to generate adequate data for analysis while at the same time offering opportunities for comparison with later joint or contrasting appearances in nonnews forums, such as Pastor Rick Warren's candidate symposium in 2008. This analysis of news media reaction applies Michel Foucault's (1980) social theory in the context of social institutions. Foucault equates the production of knowledge with the exercise of power and assumes an inevitable relationship between power and resistance. While some of the discursive practices analyzed could be described as examples of media framing, much framing research is notoriously descriptive of frames rather than analytical of the social processes that generate them. Studies that do consider institutional struggles over frames typically paint the news media as villains (e.g., Patterson, 1993) or victims (e.g., Bennett, Lawrence, & Livingston, 2007) in political processes. Foucault's theory, in contrast, emphasizes the continuous nature of social power and thus encourages examining how institutional power (including the news media's) is maintained over time. Similarly, while some of the news media's reactions to political leaders' appearances in popular media fit existing descriptions of paradigm repair (e.g., Bennett, Gressett, & Haltom, 1985; Berkowitz, 2000; Hindman, 2005; McCoy, 2001), Foucault's theory generates novel insights. It suggests where challenges to journalistic authority come from and enables scholars to move beyond considering how journalism revalidates its practices to explore how journalism disciplines audiences to reaffirm the necessity of journalistic institutions. In Discipline and Punish (1977), Foucault argues that power is more than corporal punishment, for knowledge of oneself as an object of economic production and as a subject in relation to other social institutions is also a source of social control. Thus, although journalism does not have direct, physical control of its audience, it can apply disciplinary mechanisms, setting the parameters of normal and deviant behavior, which serve to reinforce the institution as a source of social authority (Foucault, 1980; Peer, 1992). News coverage of Bush's and Gore's appearances on the Oprah Winfrey Show argued that journalism was uniquely qualified to inform the public about politics, and any political discussion that did not include it was fatally flawed. In Foucauldian terms, the news media asserted their authority to define real political discourse as that produced by journalism and distinguish it from lesser forms of candidate self-presentation. But the news did more than defend its own practices and denigrate those of the entertainment media. It also disciplined the candidates and, more surprisingly, Oprah's audience, suggesting that those who used nonnews forums for political information were not “real” citizens. That is, in justifying its continuing relevance, journalism not only refigured political discourse into its own terms but also defined an audience who needed that discourse. Journalism's disciplinary resources Political communication scholarship has identified journalism as a social institution (Cook, 1998; Patterson, 1993; Sparrow, 1999) playing a vital role in the political process. Its legitimate status is denoted among both scholars and professionals by the fact that journalistic political discourse is the standard against which “alternative” forms of political discourse are compared. This is why although the Oprah Winfrey Show is a mainstream television program with an audience about the size of that for a network nightly news program, in this context, it can be considered alternative. From a Foucauldian perspective, the institution of journalism is a set of social practices that produce and define knowledge and that legitimate ways of knowing about political and public life. It exercises disciplinary power in two ways. By defining “normal” political discourse, it reinforces its own authority to define and categorize “normal” and to discipline behavior that does not fit. Thus, in exercising its disciplinary power, journalism generates both dominant and subjugated knowledges. These subjugated knowledges are not institutionally maintained but emerge at the point where the dominant knowledge is resisted (Foucault, 1980). Second, journalism has some of the characteristics of Foucault's panopticon, in which those in authority remain invisible while those subjected to authority experience “a state of conscious and permanent visibility that assures the automatic functioning of power.” The observed are “the object of information, never a subject in communication” (1977, 201, 200). Journalists' professional norms require that they observe the social world without appearing in their own stories. In responding to the challenge represented by the Oprah Winfrey Show, they not only observed and publicized the behavior of candidates; they observed and publicized the behavior of citizens, Winfrey's audience. Foucault argues that the panopticon is socialization mechanism that impels its subjects to internalize expected behavior. Disciplining the candidates While paradigm repair approaches typically see challenges to journalistic authority as emerging from accidents or errors (e.g., Bennett et al., 1985; Berkowitz, 2000; Hindman, 2005), Foucault (1980) assumes that resistance is a natural by-product of the exercise of power. A social system's dominant knowledges are always shadowed by what he calls “subjugated knowledges,” ways of knowing that are not affirmed by existing social institutions or norms. Thus, rather than treating challenges to news media authority as transient or accidental moments of paradigm repair, Foucault's perspective suggests that such challenges can be seen as purposeful rejections of journalism's way of knowing. Journalistic ways of knowing are similar to scientific ways of knowing (Carey, 1987). Reporting embraces scientific norms such as objectivity, empirical observation (see Zelizer, 1992), and valuing quantification as a means of knowing (see Herbst, 1993). Knowledge generation paradigms such as horse-race coverage and game framing of politics (Patterson, 1993) adopt not only the social scientific practice of polling but the “objectified” language of science, a language of “is” (empirical observation) and “will” (hypothetical prediction). The objective and rational knowledges produced by journalism are a poor fit for moral and persuasive discourses of “should” and “ought” (Glasser & Ettema, 1989; Jones, 2001) and for the production of image. As campaigns are chiefly engaged in persuasive discourse and image building, campaigning generates a form of subjugated knowledge that challenges journalism's way of knowing. Transfigured through the news, the language of campaigns is rendered scientific by considering how the candidates' discourse serves their larger campaign strategy and evaluating its effectiveness. This disciplinary power produces journalistic forms of political knowledge, transforming campaign discourse into news discourse, and rendering the campaign's way of knowing deviant. Such discipline has taken on a near-ritual form that reaffirms journalism as an authoritative source of political discourse (Tuchman, 1972). As journalism distorts efforts at creating the type of knowledge campaigns generate, candidates may resist journalism by seeking alternative forms of media better able to produce subjugated knowledge. Candidates' increasing tendencies to make appearances in nonnews media forums in 2000 and beyond can usefully be conceptualized as a form of resistance to the political knowledge produced by journalism. Here again, although the Oprah Winfrey Show is part of mainstream popular culture, the political knowledge it generates can be thought of as subjugated because it does not adhere to existing standards of political discourse. Foucault encourages us to think of this struggle over framing not in terms of electoral outcomes or democratic processes but rather in terms of the struggle for institutional legitimacy and power. Journalism asserts institutional authority to both define and generate political knowledge and in doing so disciplines candidates. Disciplining the audience Work in paradigm repair has shown how journalists discipline their own, by firing a reporter (Hindman, 2005) or forcing an editor to apologize for his work (McCoy, 2001), for example, as part of the process of validating existing news norms. However, conceptualizing moments of crisis in journalistic practice as acts of resistance to journalistic ways of knowing suggests that the institution must do more than validate its own practices in order to maintain its social authority. The continuing relevance of journalism as a social institution relies on its authority to generate political knowledge. This requires not just the validation of norms within the profession but the discipline of political actors who resist those norms, including the public. As Carey (1987) observes, without a public, there is no need for journalism. A public is not merely an audience; it is an audience that “requires” the knowledge journalism produces. To generate one, journalism defines good (or normal) citizenship as consumption of journalistic output and disciplines those citizens who turn to alternative venues for information by defining them as deviant.1 Thus, journalism disciplines not only candidates but audiences who use subjugated knowledges rather than the legitimated political knowledge news produces. To define citizenship, journalism draws not only on its own institutional resources but on the disciplinary power of a hegemonic status quo. The culturally powerful Progressive era model of citizenship, which requires high levels of information and rational decision-making on the part of voters, is one definition of a “good citizen” (Schudson, 1998) and a key disciplinary resource. However, other disciplinary knowledge exists in the form of negative stereotypes that have been used to deny political equality to various demographic groups. In the case of the Oprah Winfrey Show, the full extent of the news media's discipline of the audience cannot be appreciated without considering how they represented women's citizenship.2 Studies have examined the news media's representation of women as candidates and leaders (e.g., Devitt, 2002; Heldman, Carroll, & Olson 2005; Kahn & Goldenberg, 1991), but virtually no research has considered how contemporary news discourse constructs women's role and performance as citizens. Historically, however, women have been defined as unworthy of political citizenship. Kent (2004) summarizes early thought on the political role of women in liberal democracies: “Women inhabited a separate, private, domestic sphere, one suitable for the so-called inherent qualities of femininity: emotion, passivity, submission, dependence, and selflessness” (p. 102). Living in the private sphere, women were primarily responsible for the education and moral development of their children and were considered ill-suited for public action. Bernard (1979) offers the obverse side of this claim: “Participation in the polity would coarsen and degrade women and unfit them for the gentler roles of wife and mother” (p. 280). This argument that women were the guardians of moral values and that their chief occupation was raising children would ultimately be used by women agitating for the right to vote. They argued that their presence would help to “clean up” politics and public life (Bernard) and that if they were capable of raising good citizens, they were surely capable of being good citizens (Ramsey, 2000). However, echoes of this way of thinking reverberate to this day in what are defined as “women's issues.” The even older prejudice that women are emotional and incapable of the rational thought required to make political decisions dates back to Aristotle's writings on politics. He believed that women were enslaved by their bodies, making them similar to animals and therefore the opposite of men, who, as heads of their households, were capable of putting aside personal concerns in order to achieve public goods (Kent, 2004). At least since the French revolution, women's chastity was chief among the moral virtues that fitted them for life in the private sphere, and their sexual behavior—unlike that of their husbands—subject to state control (Kent). Faced with political discourse that had drifted into a nonnews forum, the news media drew on these disciplinary resources. Disciplining the candidates for producing subjugated knowledge, they rejected conversations with Oprah Winfrey as not authentically political, reasserting their authority to define news while at the same time recasting the interviews through news perspectives in order to render them political. Disciplining the audience that used an alternative source of political knowledge, they reinforced a rational, scientific model of citizenship and denigrated those who consumed the knowledge generated by the Oprah Winfrey Show as irrational, emotional, and sexually motivated. Methods: Identifying news coverage of the Oprah Winfrey Show appearances To capture the way the candidates' appearances on the Oprah Winfrey Show were covered by the traditional news media, the “All News” portion of the Lexis-Nexis database for the first 3 days following each candidate's appearance was searched. For Gore's appearance, search terms were “Al Gore” and “Oprah Winfrey” appearing in the same article or transcript between September 12 and 14, 2000. Search terms for news coverage of Bush's appearance were “George Bush” and “Oprah Winfrey” appearing in the same article between September 20 and 22, 2000. Most of the coverage of their appearances appears in these three day windows, and expanding the time frame did not add substantively to the data set. Non-American newspapers, Associated Press services other than their main wire, and specialty publications such as business publications or those devoted to a single profession were excluded from the search results in order to focus on how the news media presented the talk show appearances to a general, mainstream, American audience. Duplicate articles (as when more than one newspaper carried the Associated Press wire copy, or when the same article was included in several editions of the same paper) were also excluded. Articles that mentioned the candidates' names and Winfrey's but did not mention their appearance on her show were excluded, as were articles devoted to the weekly television ratings and those that mentioned the Oprah Winfrey Show appearances as part of a calendar of candidate events. Reduced to unique, substantive news articles and transcripts about the candidates' appearances on the program, the data set included 53 articles and transcripts about Gore's appearance and 106 articles and transcripts about Bush's. Although there are significantly more articles about Bush's appearance (possibly because by the time he appeared, there seemed to be a trend toward candidates “doing” talk shows), analysis shows that the coverage of each candidate's appearance was remarkably similar. Rather than describing the coverage, this analysis remains theoretically focused on assertions of journalistic authority and discipline. In capturing the institutional response of journalism, no distinction is made between news outlets or individual journalists, nor is distinction made between “reporters” and “commentators.” Instead, commonalities across the profession are highlighted. Discipline of the candidates was associated with journalists either asserting the right to define news and, by extension, political information, or using material from the Winfrey interviews to generate news discourse. Discipline of the audience was associated with journalists contrasting audience behavior to definitions of good citizenship or recalling stereotypes that have been used to marginalize or exclude women from political participation. Findings: Forms of discipline The analysis that follows considers how journalism affirmed its social relevance and institutional authority as the sole, legitimate source of political information in the face of an evident challenge from popular culture. Defining political discourse Journalists' most consistent criticism of the candidates' Oprah Winfrey Show appearances was also the broadest in terms of disciplining the participants. They argued that the programs lacked substance. Under this criticism, the candidates were disciplined for circumventing journalists' tough questions, Winfrey was disciplined for her less-than-rigorous political questioning, and the audience was disciplined for demanding this sort of content. This form of discipline establishes politics as the rightful domain of journalism by contrasting the self-perceived hard-hitting questioning of news reporters to the soft questions of Winfrey. Media critic and reporter Howard Kurtz accused the candidates of avoiding reporters and the substantive questions they would ask: “Candidates like the talk show circuit for a reason. The questioning, shall we say, is not what you would get with Tim Russert or Ted Koppel” (Inside Politics, 2000). He continued: Chatting with the likes of Oprah, of course, is no substitute for answering serious press questions on the road to the White House….The voters may be entertained by hearing the candidates talk about their favorite cereal, but it's not exactly a high-fiber diet. (“Inside Politics”) New York Times reporter Alison Mitchell (2000) made the point more subtly with the observation that the Bush campaign had sought a way to “speak to voters for an hour ‘unfiltered’ by the news media.” Her use of quotation marks defends her profession more than it preserves the language of Bush's campaign staff and suggests that Winfrey, unlike a “real” reporter, would be unable to hold her guest politically accountable or prevent him from manipulating her audience. A New York Times critic negatively contrasted the talk show appearances with other, more traditional political forums like debates (James, 2000). CNN's Tucker Carlson equated Winfrey's show with Jerry Springer's talk show and World Wrestling Federation programs as targeted publicity outlets for candidates (Inside Politics, 2000). The interviews were said to mark candidate efforts at image building rather than at addressing the issues: “Such alternate television appearances are meant to reach voters who are not glued to CNN, but they reach them with a hollow message: vote for me, I'm a good guy” (James). Evidence of this lack of substance was provided in a number of stories about the talk show appearances, where discussion of the candidates' favorite breakfast cereals, sandwiches, books, dreams, and sleeping attire was commonly reported. Washington Times writer Jennifer Harper (2000) equated such image building with emotional (as opposed to rational) appeals: “‘Feels’ is the operative word here. The talk show landscape is, after all, far from the madding polls, the horse race, and journalists who ask policy questions.” Here, Harper associates polls and horse race coverage, with their sheen of “science” (see Ginsberg, 1986; Herbst, 1993; Peer, 1992) with journalism and opposes them to the emotional forum of entertainment television, a point to which we shall return. She goes on to cast such appearances as a threat to the entire political system: “Indeed, there is a risk that talk show chatter can disintegrate into prepackaged tripe, void of substance and gravitas. Some think that talk show appearances trivialize not only the campaign process but the very office of president itself” (Harper). Some journalists blamed candidates for this lack of substance (James, 2000), but others accused Winfrey of asking “softball” (Cain, 2000) questions to keep her audience entertained: “The talk show diva … frequently interrupted when the vice president lapsed into his stump speech, steering him back to more viewer-friendly matters.” (Slater, 2000) “Ms Winfrey, who could be as tough as any political reporter if she wanted to, knows her audience too well to expect it to sit still for a heavy policy discussion.” (James, 2000). “Gore often tried to steer the conversation into policy areas. But just as often, Winfrey steered the talk into more entertaining areas…” (Shepard, 2000). Transforming the appearances into news As reporters dismissed the candidates' Oprah Winfrey Show appearances as lacking substance and failing to provide real political discourse, they nevertheless subjected the candidates' performances to the criteria they used to judge performances in forums they did define as political, such as convention addresses or debate performances. Once refracted through the news media, candidates' public appearances, regardless of how nontraditional a forum they use, look basically the same. They are depicted offering calculated performances designed to win over voters, and the quality of their performances can be judged by how effective they are at attracting support. In Foucauldian terms, news media produced journalistic knowledge and repressed alternative ways of knowing. Two transformations of the appearances into news were made: one passive and the other an active reconstruction of the subjugated knowledges into the dominant paradigm of horse-race coverage and game framing. The first transformation of the talk show appearances highlights politically substantive comments made by the candidates on the talk show. Co-optation of newsworthy information from the talk show appearances allows journalists to reclaim their position as the purveyors of political information, as the substantive information only becomes political once it makes it into the news. Even as their colleagues claimed the talk show appearances lacked substantive content, some reporters used statements Gore made during his Oprah Winfrey appearance in news coverage of a policy issue. Gore had appeared with Winfrey shortly after the release of a Federal Trade Commission (FTC) report that accused the entertainment industry of marketing products such as films and videogames to underage users. A front page article in the Los Angeles Times treated Gore's statements as it might any stump speech: “Gore blasted Hollywood during an appearance on ‘The Oprah Winfrey Show,’ later pledging at a campaign stop to give the industry six months to shape up before he pursues legislation to punish deceptive marketing” (Shiver & Fiore, 2000; see also Eller & Bate, 2000). Gore's comments sounded even more substantive in a New York Daily News article: “Appearing on ‘The Oprah Winfrey Show,’ Gore called for ‘an immediate ceasefire’ by the industry—or, he said, it will face accusations of false advertising” (Bazinet, 2000). In the context of commentaries on Winfrey's interviews, Gore's statements lacked substance, but when they reappeared in traditional journalistic forums, they became news. While treating talk show moments as news only when they are presented through the news might be termed a passive form of Foucauldian discipline, the second transformation of the talk show appearances to news is much more assertive. Many stories about the candidates' appearances with Winfrey evaluate their performances in the context of campaign strategy; they are game framed (Patterson, 1993). Two dimensions of game framing are especially evident in the reporting: The Oprah appearances are situated in the greater context of campaign strategy, and the success of the strategies is evaluated. Ironically, these frames ignored the substantive content of the interviews as they transformed the (subjugated) knowledge of the Oprah Winfrey Show into news. First, journalists discussed how the talk show appearances fit into the candidates' campaign strategy. Not only did reporters accuse the candidates of seeking to avoid tough questions; they also described the candidates' efforts to “court” the “women's vote” with their appearances. An MSNBC journalist observed, “Make no mistake, Bush's appearance on ‘Oprah’ is part of a calculated strategy to reach out to women voters, now the crucial swing voters in this election” (The News, 2000). Other journalists noted that the Oprah appearances gave the candidates a chance to reach voters “they can't normally get to” (News Recap, Today Show, 2000). Typical of game-framed coverage, these kinds of stories not only discussed campaign strategy but described who was ahead and who behind in public opinion polls of demographic groups likely to affect the outcome of the election. Campaign strategy was also revealed in stories that described the candidates' appearances on the Oprah Winfrey Show as strategic performances: “Well behind Al Gore with women voters, George Bush took his turn today… with a sly imitation of his opponent's famous convention hug and kiss. This is a blatant pitch for Oprah's millions of female viewers” (Mitchell Report, 2000; see also La Ganga, 2000). Both Bush and Gore were accused of avoiding Winfrey's questions for strategic reasons. Gore, asked by Winfrey if he had ever stolen anything, said he did not think so. Journalists interpreted the hedge as an indication that he probably had but would not admit it. Bush was asked if there had ever been a time when he needed forgiveness. He declined to provide specifics, explaining that he was running for president. Both these events appear several times in the coverage. In general, journalists interpreted the construction of positive images by the candidates as a ploy: “Behind their smiles, the candidates are fiercely playing the politics of personality” (James, 2000). Second, journalists evaluated candidate performances on the Oprah Winfrey Show in terms of winning and losing. In Bush's case, the origin of this story may have been his campaign staff; communication director Karen Hughes was reported describing Bush's appearance as a “home run” (Balz, 2000). A Fox News correspondent added, “The kiss helps. In fact, this whole show helps” (Special Report, 2000). A Dallas Morning News entertainment writer was more circumspect, but still thinking along the same lines: “[Bush] may have succeeded in making a favorable impression that money can't buy” (Bark, 2000). A New York Post story tried to put things in perspective by offering a “tale of the tape” match-up of the candidates' appearances, comparing them on dimensions such as “best one-liner,”“biggest pander,” and “best moment” (Orin, 2000). Of course, the sports metaphors are also a common element of game framing. Disciplining the audience Journalists did not simply discipline the candidates by rejecting the Oprah Winfrey Show as “real” political discourse. They disciplined the audience as well by claiming that, in turning to nonnews forums for political information, audience members were not fulfilling their obligations as citizens. Schudson (1998) traces these reporters' underlying expectations that good citizens are highly informed and keep abreast of public affairs to a Progressive era model of citizenship that, he reminds us, is neither natural nor necessary to democratic politics. However, in defining “real” citizens as highly informed and engaged in public affairs, the news media discipline audiences by setting standards difficult to meet without attending to the news, reaffirming journalism's social relevance while undermining the legitimacy of alternative sources of political information. Some elements of this disciplinary rhetoric have already been presented: the journalists' accusations that Winfrey steered the conversation away from “real” political talk for the sake of her audience, which demanded to be entertained. Other journalists also blamed the audience for the campaigns' shift into popular culture forums. Kurtz suggested that the politically disinterested had lowered the level of political discourse: “It's a way of reaching folks who aren't political junkies and particularly women. It provides pretty pictures for the networks to replay” (Inside Politics, 2000). Disciplining women In addition to the relatively generic accusation that voters do not pay enough attention to the news, journalists aimed barbs at Winfrey's audience and the women voters that audience was said to personify. News stories commonly forged a strong link between the Oprah Winfrey Show and the women's vote. They typically estimated Winfrey's audience at between 7 and 22 million, “mostly women.”USA Today said, “Al Gore, seeking to solidify his gains with female voters, picked a perfect venue Monday with a one-hour appearance on The Oprah Winfrey Show” (Bennedetto, 2000). CBS's Bill Plante claimed, “George W. Bush has been ambushed at the gender gap. That is why he was on daytime TV with Oprah Winfrey, hoping to win over some of the more than 20 million women in her audience” (Early Show, 2000). Reporters blamed women's lack of political engagement for the actions of the candidates: If women voters won't come to the candidate, then the candidate must go to them. So George W. Bush went to the queen of daytime TV, Oprah Winfrey, to court her huge audience—22 million, mostly women—submitting to the kind of personal confessional he usually shuns. (Evening News, 2000) In suggesting that women are politically inattentive, journalists are disciplining not Oprah viewers but women as a social group. Their discipline of women for the transgression of using nontraditional information sources drew upon historic arguments opposing women's participation in public life, defining “women's issues” as those that emerge from the private sphere and suggesting that women made political decisions based upon subjugated knowledges that were irrational and deviant. A private sphere of women's issues In constructing Winfrey's show as the lynchpin to a strategy for winning the “women's vote,” the journalists contributed to a very traditional image of women. Winfrey appears on daytime television, so, video recorders notwithstanding, one would expect her audience to lean toward stay-at-home mothers and women who work part-time as opposed to women working full-time or in the professions. In other words, reporters equate women voters with those women who enact relatively traditional roles, suggesting that real women inhabit the private sphere of home and family, whereas real citizens are attuned to news and public affairs. Journalists also defined “women's issues” in ways consistent with the traditional belief that the private sphere is their legitimate domain. After reporting Bush's claim that he expected to do well with women voters because “I know what I'm talking about when it comes to educating children,” the Associated Press put the quote in context by saying, “Bush has spent much of the past week trying to stem Gore's gains among women voters, stressing cradle-to-grave domestic issues” (Raum, 2000). The Atlanta Journal–Constitution led its story on Gore's Oprah appearance, “Al Gore used an appearance on Oprah Winfrey's television show Monday to promote ‘family-friendly’ politics and try to sustain his increased popularity among female voters” (Shepard, 2000). The Dallas Morning News described the Gore campaign's strategy as seeking to “widen their advantage among women voters by focusing on education and ensuring that there is a government role in curbing the marketing of violent entertainment to children” (Bark, 2000). Health care and Social Security also made the list of “women's issues.” Although public opinion polls do show that women are interested in these issues, two additional points are worth noting. First, the issues typically identified with women in the news media's coverage of the candidates' appearances with Winfrey are issues in which women themselves are invisible. That is, they invoke women's roles as nurturers of (and advocates for) others, such as children and aging parents. Mention of issues on what might be called a “women's agenda” that did not involve their roles as family caretakers, such as equal pay and women's health issues, was rare. The second point to be made about the identification of “women's issues” is that such a definition constructs women as a special interest group, distinct from both men and other demographic groups who “own” other issues. Constructing women as a special interest relies not on their minority status, for more women than men vote, but rather on the assumption that men's political attitudes are normal, while women's are distinguished to the extent that they are different from men's. Defining women as a special interest links them to a particularly disliked aspect of American politics. Moreover, in asserting the ability to define women's issues, journalists assert the ability to define citizenship and create an audience for the knowledge they produce. Women as emotional While a limited and traditional range of women's issues did figure into the news media's coverage of the Oprah Winfrey Show appearances, some journalists tied the lack of substance they saw in the candidates' performances to women's emotional nature. The lead of a Dallas Morning News story said: “Surrounded by Oprah's audience, famously wonkish Al Gore was all feelings Monday: family and parenting and his ‘partnership’ with Tipper” (Slater, 2000). Many journalists seized upon Bush's tearful moment when describing his wife's pregnancy and the birth of his daughters. Fox News' Morton Kondracke observed, “[T]he moment he started talking about his daughters—that was a wonderful ‘Oprah’ moment, and it was entirely genuine…. To the extent that I understand what appeals to women voters, it seemed to me to be a homerun” (Special Report, 2000). A Salon.com commentator argued that Bush's display of emotion made his Oprah appearance more effective than Gore's: “Bush cried on the Oprah Winfrey Show …. And so, if Al Gore went on Oprah last week to prove he was a real person, George W. Bush proved he was a real Oprah guest, which is even better” (Skinner, 2000). It is useful to consider these constructions of women's political thinking in light of the contrast developed by Washington Times reporter Harper (2000), linking journalism to rational thinking and science while connecting talk shows with emotion. In these formulations, “real” political discourse appears in rational, scientific forums traditionally dominated by men, while forums designed to appeal to women are emotional and not authentically political. Thus, the Oprah Winfrey Show was depicted as an extension of the private sphere of women and not part of the public realm of politics. The women who attended to it were not genuinely political because they relied on emotion rather than reason to make decisions; therefore, they were not real citizens. Real citizens watched news. Women as sexual Journalists covering the candidates' forays into the private realm of women's talk used the language of sexual attraction, which takes the discipline of women as emotional and irrational to a further extreme. Not only are women too emotional to participate in politics; they are concerned with sex and sensuality, which have no place in political discourse. Bush and Gore were said to be “wooing,”“courting,” and “charming” women voters with their Oprah appearances. Larry King was especially clear about the double entendre, opening his show with: “Tonight, Al Gore and George W. Bush are happily married men, but they’re courting other women on the campaign trail” (Larry King Live, 2000). Another common double entendre was the use of the word “pander”—a word associated with the commercial sex trade—to describe the candidates' attempts to win women's votes. Of course, the candidates were also said to be “courting” and “pandering” to other demographic groups, such as families and seniors, but only with women were the metaphorical and literal meanings of this language so closely bound. Journalists used other phrases to suggest that the campaign had moved not just into the private sphere but into the bedroom, singling out for attention moments from the interviews that could be constructed as titillating. The Boston Globe said Gore was allowed to “purr and flirt” with Winfrey's audience (Nyhan, 2000), and the Atlanta Journal–Constitution said “provocative questions… led to titters from the mostly female studio audience” (Shepard, 2000). Kondracke described Winfrey's “audience lovingly looking at Al Gore” (Special Report, 2000). Some news outlets attached sexual innuendo to Gore complimenting Winfrey's shoes, as in this sensual passage from the Dallas Morning News: “I don't have a magazine or a publishing house,” he said, alluding to her more expansive business endeavors. Then pointing to her red stilettos he said: “And I don't have red boots.” Winfrey laughed, leaned back in her luxurious yellow leather chair and lifted one leg to show off her ruby red footwear. (Slater, 2000) The 2000 election was called “The Year of the Kiss” owing to Gore's apparently impassioned greeting of his wife at the Democratic National Convention and Bush's more modest greeting of Winfrey. Winfrey had teased Gore about not kissing her, so when Bush kissed her cheek, it became one of the signature moments of the interview. Margaret Carlson of Time magazine likened the image of Bush kissing Winfrey in the next day's papers to “the cover of a bodice-ripping romance novel” (Inside Politics, 2000). A Washington Post staff writer also embellished the story a bit: “He began the show with a kiss, a serious plant on the cheek that brought an ‘Oh, yes!’ from the seemingly surprised Winfrey. ‘My pleasure,’ Bush said with a smile as he settled into the easy chair beside her” (Balz, 2000). Both Brit Hume of Fox News and CNN's Carlson described the kiss as a “public display of affection,” or PDA, a euphemism typically applied to teenagers and thus one that implies immaturity in both the women Winfrey represents and the candidates. Journalists also sexualized other elements of the candidates' interaction with the host. Gore was said to have “played footsie” with Winfrey (Brown, 2000) while Bush sat “knee-to-knee” with her (McFeatters, 2000). Winfrey's “provocative” questions and the responses they drew were also a popular topic in the news. Several reports described Gore's near-admission that he slept in the nude in response to Winfrey's question about “his favorite thing to sleep in.” News outlets also reported that in response to Winfrey's question about his favorite dream, Bush said nothing but blushed and raised his hand. The Bush campaign staff claimed that it was a silent reference to taking the oath of office as president, but news media reports suggested that what he had in fact been thinking about was too naughty to mention: “Communications director Karen Hughes was asked after the show whether the blushing Bush was thinking about, oh, something personal, maybe racy even” (LaGanga, 2000). Sexualizing the candidates' appearances on the Oprah Winfrey Show gave journalists the best of all worlds: they titillated their own audiences and ridiculed, reproached, and disciplined the candidates and Winfrey's audience, equating the political “transgressions” with sexual transgression. Women who made political choices based on sexual attraction were not good citizens. Conclusions The 2000 presidential campaign was a moment in American political history when the candidates challenged the norms of journalism, sitting for an hour with Oprah Winfrey and engaging in alternative forms of political discourse. Faced with political journalism's aura of scientism and rationality, the candidates used alternative venues and topics by which to connect with citizens. Journalism, in turn, reasserted its social relevance and institutional authority as the legitimate producer of political knowledge and disciplined those who digressed from the defined norm. First, journalists used their institutional (disciplinary) power to define knowledge, dismissing the Oprah Winfrey Show as inauthentic political discourse while at the same time reframing elements of that discourse as “real news.” By claiming that the hour-long discussions lacked reason and contained emotional and sexual appeals, the news media asserted their authority as the legitimate and sole owners of political discourse and disciplined the candidates for generating subjugated knowledge. For journalists, the Oprah Winfrey Show appearances were evidence that other (entertainment) venues were not appropriate places to discuss politics because they failed to create rational discourse (Harper, 2000; James, 2000). Similarly, they were critical of the woman who was once a journalist herself for failing to ask questions that promoted rational discourse. It was her fault that the public lost the chance to learn about policy, from which they would have later been able to make an informed decision (James, 2000; Shepard, 2000; Slater, 2000). Second, journalists disciplined the show's audience and the social group that audience was said to personify for failing to act as good citizens. Journalists disciplined Winfrey's audience and women in general for failing to live up to standards of citizenship that require political information produced by the news media (Evening News, 2000; Inside Politics, 2000), offering support for Carey's (1987) observation that journalism needs a public. They blamed women voters for the candidates' digression into a nonstandard forum, drawing upon both the Progressive-era model of informed citizens (Schudson, 1998) and even older models in which femininity rendered one unworthy of enfranchisement or life in the public sphere. As part of this discipline, journalists perpetuated traditional stereotypes that define women as guardians of the private sphere, having no identity outside their roles as surrogates. In an even more extreme form of discipline, some suggested that women's use of nontraditional media disqualified them as actors in the public sphere (Balz, 2000; Brown, 2000; Inside Politics, 2000; Larry King Live, 2000; McFeatters, 2000; Slater, 2000). Women were portrayed as sexually motivated, incapable of acting as rational citizens, and easy prey for exploitative pandering by candidates. While these results do not contradict findings previously generated in the paradigm repair and framing traditions, they do offer unique insights unavailable to either perspective. Thinking of journalism as an institution and using Foucault's conceptualization of the production of social knowledge illustrates how power is created through discourse and offers a richer picture of the social environment in which journalism constructs and defends its authority and relevance. Beyond simply revalidating its practices in the wake of accidents and errors, journalism must continuously reassert its social authority both to define authentic political discourse and to discipline social actors who challenge journalistic ways of knowing to maintain its institutional power. In doing so, the profession both creates knowledge and defines an audience that requires that knowledge. Perhaps most importantly, Foucault's concept of discipline brings the previously invisible public back into the picture and gives life to Carey's (1987) key insight that journalism, and for that matter democracy, require a public. This analysis offers some insight into how journalism creates one. There is a long cultural and political history that grants journalists special status when it comes to acquiring and disseminating information about public affairs. Every day, reporters claim the ability to identify news, and it is not a large step from there to claim they can identify authentic political discourse. However, this analysis demonstrates that journalism also asserts the authority to define citizenship in ways that make necessary, even ideal, the discourse it creates. If institutions maintain power by reaffirming the ability to identify deviance, it makes sense that journalism engaged in this disciplinary action not only for the noble purpose of creating better-informed citizens, but also to help preserve its position as sole owner of political discourse. Nevertheless, there is also good reason to imagine that these journalistic attempts at discipline are ineffective and may even backfire. By couching the talk show appearances in their traditional style and manner, in which they analyzed each candidate's strategy and motives for appearing on the show and used game frames to guess which hour was most successful, the journalists again engaged in the very behavior that encourages candidates to avoid news venues. Nonnews forums may be popular among candidates, not just to reach voters but to build a discourse better suited to their goals. Disciplining voters for failing to act like good citizens could also have negative repercussions. Maintaining that only the news media are capable of protecting citizens from candidates' manipulations suggests that citizens are incapable of making informed political decisions on their own. Such a claim may further alienate audiences from a news media that insults their capacity to perform basic democratic tasks. Likewise, disciplining women by marginalizing their political agenda and rejecting ways of knowing associated with femininity may drive them toward venues that embrace their perspectives on politics and public life rather than demanding rational, scientific citizenship. Not only can such attempts to discipline the public sphere backfire for journalism itself; they may have broader negative social consequences. Claims that journalism is necessary to protect citizens from politicians assume the duplicity of candidates (for a discussion of the negative impacts of this, see Gitlin, 1996; Patterson, 1993). Upholding the Progressive-era model of highly informed citizenship, with news as the key information provider, may be a drag on political participation (Schudson, 1998). In disciplining those who turned to alternative sources for political information, journalists suggest to news consumers that their fellow citizens cannot be trusted to make good decisions, which could have negative impacts on social capital (Putnam, 2000) as well as political participation (Downs, 1957). This analysis suggests several fruitful avenues for further research. First, it raises questions about how traditional news media are laying claim to a niche in a new information environment in which they do not have a meaningful monopoly on political information. The nature and success of such claims are likely to help define the nature of journalism and impact the economic viability of news organizations. Second, it raises questions about how the news media depict citizenship, both generally and in the context of specific social groups. Schudson (1998) has documented the evolution of citizenship models over time, and other scholars have debated the nature of citizenship in a globalized, capitalistic world, but it is the media, and particularly the news media, that present images of citizenship to the public on a daily basis. What do those images include? How might they affect the practice of citizenship? How might they affect democratic functioning? Acknowledgment The authors thank Michael X. Delli Carpini and the anonymous reviewers for their thoughtful criticisms of the manuscript. Notes 1 " Both Berkowitz (2000) and McCoy (2001) briefly reference public behavior in their descriptions of paradigm repair, but both center their analysis on how members of the journalistic profession were disciplined and punished. 2 " Until the mid-20th century, legal precedent and cultural practice disconnected voting rights from citizenship rights, but this essay embraces the modern perspective that equates the two. References Balz , D . ( 2000 , September 20). Dreaming of resurgence; On ‘Oprah’ relaxed Bush displays retooled approach . The Washington Post. Retrieved from Lexis-Nexis database. Bark , E . ( 2000 , September 20). Bush plays it safe; Oprah also strong in nonpartisan effort . The Dallas Morning News. Retrieved from Lexis-Nexis database. Baum , M. A . ( 2003 ). Soft news goes to war: Public opinion and American foreign policy . Princeton, NJ : Princeton University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Baym , G . ( 2005 ). “The Daily Show:” Discursive integration and the reinvention of political journalism . Political Communication , 22 , 259 – 276 . Google Scholar Crossref Search ADS WorldCat Bazinet , K. R . ( 2000 , September 12). Showbiz warned Prez says police marking to kids . Daily News (New York). Retrieved from Lexis-Nexis database. Benedetto , R . ( 2000 , September 12). Reaching out to female voters Vice president focuses on family in ‘Oprah’ chat . USA Today. Retrieved from Lexis-Nexis database. Bennett , W. L. , Gressett , L. A., & Haltom , W. ( 1985 ). Repairing the news: A case study of the news paradigm . Journal of Communication , 35 ( 1 ), 50 – 68 . OpenURL Placeholder Text WorldCat Bennett , W. L. , Lawrence , R. G., & Livingston , S. ( 2007 ). When the press fails: Political power and the media from Iraq to katrina . Chicago, IL : University of Chicago Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Berkowitz , D . ( 2001 ). Doing double duty: Paradigm repair and the Princess Diana what-a-story . Journalism , 1 , 125 – 143 . Google Scholar Crossref Search ADS WorldCat Bernard , J . ( 1979 ). Women as voters: From redemptive to futurist role. In J. Lipman-Blumen & J. Bernard (Eds.), Sex roles and social policy . ( 279 – 286 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Brewer , P. R. & Cao , X. ( 2006 ). Candidate appearances on soft news shows and public knowledge about primary campaigns . Journal of Broadcasting and Electronic Media , 50 , 18 – 31 . Google Scholar Crossref Search ADS WorldCat Brown , M . ( 2000 , September 20). Oprah puts Bush's brains to the test . Chicago Sun Times. Retrieved from Lexis-Nexis database. Cain , A . ( 2000 , September 12). Oprah's softballs allow Gore to score . The Washington Times. Retrieved from Lexis-Nexis database. Carey , J. W . ( 1987 ). The press and public discourse . The Center Magazine (March/April), 4–15. Cook , T. E . ( 1998 ). Governing with the news: The news media as a social institution. Chicago, IL : University of Chicago Press . Devitt , J . ( 2002 ). Framing gender on the campaign trail: Female gubernatorial candidates and the press . Journalism and Mass Communication Quarterly , 79 , 445 – 463 . Google Scholar Crossref Search ADS WorldCat Downs , A . ( 1957 ). An economic theory of democracy . New York, NY : Harper . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Early Show. ( 2000 , September 20). CBS [Television broadcast]. Transcript retrieved from Lexis-Nexis database. Eller , C. & Bate , J. ( 2000 , September 12). Company Town; The biz; Hollywood licking its wounds after mauling in FTC report . Los Angeles Times. Retrieved from Lexis-Nexis database. Evening News. ( 2000 , September 19). CBS [Television broadcast]. Transcript retrieved from Lexis-Nexis database. Foucault , M . ( 1977 ). Discipline and punish: The birth of the prison . New York, NY : Vintage Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Foucault , M . ( 1980 ). In C. Gordon (Ed.). Power/knowledge: Selected interviews and other writings 1972–1977 . New York : Pantheon . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ginsberg , B . ( 1986 ). The captive public: How mass opinion promotes state power . New York, NY : Basic Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gitlin , T . ( 1996 ). Television's anti-politics: Surveying the wasteland . Dissent , 43 , 76 – 85 . OpenURL Placeholder Text WorldCat Glasser , T. L. & Ettema , J. S. ( 1989 ). Investigative journalism and the moral order . Critical Studies in Media Communication , 6 , 1 – 20 . OpenURL Placeholder Text WorldCat Harper , J . ( 2000 , September 20). Campaigns all ‘talk’ with Arsenio, Oprah: Candidates feed off tele-minded voters . The Washington Times. Retrieved from Lexis-Nexis database. Heldman , C. , Carroll , S. J., & Olson , S. ( 2005 ). “She brought only a skirt”: Print media coverage of Elizabeth Dole's bid for the Republican presidential nomination . Political Communication , 22 , 315 – 335 . Google Scholar Crossref Search ADS WorldCat Herbst , S . ( 1993 ). Numbered voices: How opinion polling has shaped American politics . Chicago, IL : University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hindman , E. B . ( 2005 ). Jayson Blair, the New York Times and paradigm repair . Journal of Communication , 55 , 225 – 241 . OpenURL Placeholder Text WorldCat Inside Politics. ( 2000 , September 20). Cable News Network . [Television broadcast]. Transcript retrieved from Lexis-Nexis database. James , C . ( 2000 , September 22). The 2000 campaign: The talk show front—Critic's Notebook; Blurring distinctions while chasing laughs . The New York Times. Retrieved from Lexis-Nexis database. Jones , J. P . ( 2001 ). Forums for citizenship in popular culture. In Hart , R. P. & Sparrow , B. H. Politics, discourse, and american society: New agendas . ( 193 – 210 ). Lanham, MD : Rowman & Littlefield . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kahn , K. F. & Goldenberg , E. N. ( 1991 ). Women candidates in the news: An examination of gender differences in U.S. Senate campaign coverage . Public Opinion Quarterly , 55 , 180 – 199 . Google Scholar Crossref Search ADS WorldCat Kent , S. K . ( 2004 ). Gender history: Law and politics. In T. A. Meade & M. Wiesner-Hanks (Eds), A companion to gender history . (pp. 87 – 108 ). Malden, MA : Blackwell Publishing . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC LaGanga , M. L . ( 2000 , September 20). Campaign 2000; The softer side of Bush: George W. visits Oprah . Los Angeles Times. Retrieved from Lexis-Nexis database. Larry King Live. ( 2000 , September 21). Cable News Network . [Television broadcast]. Transcript retrieved from Lexis-Nexis database. McCoy , M. E . ( 2001 ). Dark alliance: News repair and institutional authority in the age of the Internet . Journal of Communication , 51 , 164 – 193 . Google Scholar Crossref Search ADS WorldCat McFeatters , A . ( 2000 , September 20). Bush and Oprah go knee-to-knee . Pittsburgh Post Gazette. Retrieved from Lexis-Nexis database. Mitchell , A . ( 2000 , September 21). Campaign 2000: The Governer's wife; A political wall flower has a full dance card . The New York Times. Retrieved from Lexis-Nexis database. Mitchell Report, MSNBC ( 2000 , September 20). Bush gets a ‘kiss’ of his very own. Hotline . Retrieved from Lexis-Nexis database. The News, MSNBC ( 2000 , September 20). Bush gets a ‘kiss’ of his very own. Hotline . Retrieved from Lexis-Nexis database. News Recap. ( 2000 , September 22). The Today Show [Televised broadcast]. Transcript retrieved from Lexis-Nexis database. Nyhan , D . ( 2000 , September 13). Gender gap is zapping Bush . The Boston Globe. Retrieved from Lexis-Nexis database. Orin , D . ( 2000 , September 20). Bush's smoochy coup: Greets Oprah with big kiss . The New York Post. Retrieved from Lexis-Nexis database. Patterson , T. E . ( 1993 ). Out of order . New York, NY : A. Knopf . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Peer , L . ( 1992 ). The practice of public opinion polling as a disciplinary mechanism: A Foucauldian perspective . International Journal of Public Opinion Research , 4 , 230 – 242 . Google Scholar Crossref Search ADS WorldCat Pew Research Center for the People and the Press. ( 2000 , February). The tough job of communicating with voters . Retrieved from http://people-press.org/reports/display.php3?ReportID=46. Prior , M . ( 2003 ). Any good news in soft news? The impact of soft news preference on political knowledge . Political Communication , 20 , 149 – 172 . Google Scholar Crossref Search ADS WorldCat Putnam , R. D . ( 2000 ). Bowling alone: The collapse and revival of American community . New York, NY : Simon and Schuster . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ramsey , E. M . ( 2000 ). Inventing citizens during World War I: Suffrage cartoons in The Woman Citizen . Western Journal of Communication , 64 ( 2 ), 113 – 148 . Google Scholar Crossref Search ADS WorldCat Raum , T . ( 2000 , September 22). Bush proposes ‘medical moonshot' to find cures . The Associated Press. Retrieved from Lexis-Nexis database. Schudson , M . ( 1998 ) The good citizen: A history of American civic life . Cambridge, MA : Harvard University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Shepard , S . ( 2000 , September 12). Oprah playfully spars with ‘fun guy’ Gore . The Atlanta Journal and Constitution. Retrieved from Lexis-Nexis database. Shiver , J. , Jr. & Fiore , F. ( 2000 , September 12). Heat on Hollywood as FCC ponders curbs on violence . Los Angeles Times. Retrieved from Lexis-Nexis database. Skinner , D . ( 2000 , September 20 ). Bush gets a ‘kiss’ of his very own. Hotline . Retrieved from Lexis-Nexis database. Slater , W . ( 2000 , September 12). Gore charms Oprah, but leaves her disappointed (no kiss) . The Dallas Morning News. Retrieved from Lexis-Nexis database. Sparrow , B. H . ( 1999 ). Uncertain guardians: The news media as a political institution . Baltimore, MD : Johns Hopkins University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Special Report Roundtable. ( 2000 , September 20). Fox Special Report with Bret Hume [Televised broadcast]. Transcript retrieved from Lexis-Nexis database. Tuchman , G . ( 1972 ). Objectivity as strategic ritual: An examination of newsmen's notion of objectivity . American Journal of Sociology , 77 , 660 – 679 . Google Scholar Crossref Search ADS WorldCat Zelizer , B . ( 1992 ). Covering the Body: The Kennedy assassination, the media, and the shaping of collective memory . Chicago, IL : University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Author notes * " An earlier version of this article was presented at the annual meeting of the American Political Science Association, September 2–4, 2004, Chicago, IL, USA. © 2011 International Communication Association

Bennett, W., Lance;Wells,, Chris;Freelon,, Deen

doi: 10.1111/j.1460-2466.2011.01588.xpmid: N/A

Abstract Communication aimed at promoting civic engagement may become problematic when citizen roles undergo historic changes. In the current era, younger generations are embracing more expressive styles of actualizing citizenship defined around peer content sharing and social media, in contrast to earlier models of dutiful citizenship based on one-way communication managed by authorities. An analysis of 90 youth Web sites operated by diverse civic and political organizations in the United States reveals uneven conceptions of citizenship and related civic skills, suggesting that many established organization are out of step with changing civic styles. Two narratives have dominated scholarly debate about the role of communication in youth political participation. One story flows from Putnam's (2000) argument that the rise of a passive television culture and declining group memberships has created a “generational displacement” from politics and public life that is unlikely to be ameliorated by new forms of online civic action. Another narrative depicts young people as “digital natives” at the forefront of participatory media that may promote new forms of engagement in public life (e.g., Jenkins, 2006; Palfrey & Gasser, 2008). Both narratives have been challenged: the first by the mobilization of young and first-time voters in Obama's 2008 campaign, the second by research questioning young people's technical proficiency (Bennett et al., 2008; Hargittai & Walejko, 2008). Yet both views of young citizens continue to gather support. The civic deficit story is buttressed with declining news consumption and political knowledge levels among younger demographics (Pew, 2007a,b; Wattenberg, 2006). At the same time, open source tools and social networking platforms create new opportunities for youth to form networks and take action to address many issues directly (Benkler, 2006; Coleman, 2007, 2008). Rather than trying to resolve these competing narratives as though they are competing for a single reality of citizenship, we propose that both are partly right in the sense that each describes different parts of a changing citizenship picture: The former accounting for the fragmentation of an old civic order, and the latter bringing emerging civic styles into focus. Various scholars have described historic changes in citizenship, along with ways in which related communication regimes affect civic engagement (e.g., Bennett, 1998; Schudson, 1998). The difficulty of interpreting inchoate civic forms may account for the current paradigm controversies over what counts as “civic” and “engagement,” particularly in the proliferation of online environments that appeal to youth. For example, an online discussion among leading scholars produced disagreement over whether participants in Harry Potter fan sites or World of Warcraft games were learning important civic skills (MacArthur Online Discussions, 2006). Before we can sort out the civic properties of popular online media, it may help to establish a baseline range of the conceptions of citizenship and civic engagement found in online environments more explicitly dedicated to civic engagement. These online civic environments include: political campaign and government sites, youth branches of interest organizations (e.g., National Rifle Association, Sierra Club), community youth organizations (e.g., Boys and Girls Clubs, YMCA), and civic engagement organizations that exist only in online form (e.g., YouthNoise, TakingITGlobal). With a few notable exceptions (e.g., Coleman, 2007, 2008), earlier work examining youth engagement online has generally failed to account for different conceptions of citizenship that may be communicated by different kinds of organizations. The result is that many studies have noted communication deficits in these environments (e.g., the limited uses of Web 2.0 features), but few have offered theoretical explanations for the unevenness of online communication with young citizens. Following a review of these studies, we propose a theoretical framework to map the youth civic Web in terms of the contrasting definitions of citizenship being communicated through opportunities to learn different civic skills. This framework is then applied to a sample of 90 U.S. youth civic sites to assess differences in how citizenship is constructed by opportunities to learn different civic skills. Images of online youth engagement The vast majority of American youth are online (Lenhart, Hitlin, & Madden, 2005, 2006). Although much of their involvement is entertainment-oriented, half the nation's 18- to 29-year-olds also sought political information in early stages of the 2008 campaign (Smith & Rainie, 2008). These younger citizens were significantly more likely than their elders to watch political video clips online, use social networking sites such as Myspace and Facebook for political purposes, and express opinions in online forums (Hargittai & Hinnant, 2008; Smith & Rainie, 2008). Although younger demographics clearly experience some forms of online engagement, the activities they engage in may not reflect earlier citizen communication habits (Madden 2006). For example, there are clear differences in the way different generations engage with political information online. Although most young people encounter some news online, most of them (65%) come across it casually or accidentally in the midst of other pursuits. In contrast, a majority of Internet users over 30 years of age (55%) actively seek out their news (Patterson, 2007). Moreover, the portion of people going without any news (from any source) is largest and rising most rapidly among the 18 to 24 age group (Pew, 2007b). Do these trends render young citizens out of touch politically? The conventional answer may be yes. But other views of information sourcing and sharing suggest that there are other ways to be connected politically (Jenkins, 2006; Pew, 2007a). These shifting foundations of communication create problems for interpreting existing research. For example, we know that young people who actively seek out information online generally report higher levels of offline civic engagement (Pasek, Kenski, Romer, & Jamieson, 2006; Shah, Cho, Eveland, & Kwak, 2005; Xenos & Moy, 2007). However, these findings may only be telling us that young people who display online citizenship orientations like their parents also behave offline like their parents. At the same time, such young people appear to be dwindling in number. What is missing here is a theoretical framework for observing and interpreting emerging forms of engagement that may not be captured by studies based on earlier conceptions of citizenship. Research on the youth civic Web remains largely descriptive. An emerging consensus has it that the lack of many Web 2.0 features diminishes the appeal of most formal civic engagement sites (Coleman 2007, 2008; European Commission, 2007; Montgomery, Gottleib-Robles, & Larson, 2004). An analysis of 73 U.S. youth civic sites found generally low levels of interactive features and active pedagogical styles, that is, efforts to convey civic skills (Bachen, Raphael, Lynn, McKee, & Philippi, 2008). A largely descriptive study of youth engagement sites in seven European nations showed that technology lags were common, even in the more developed democracies (European Commission, 2007). Other studies in the United States have found that online youth sites tend to offer little hyperlinking to other sites, few opportunities for interactivity, and that candidate sites (before Dean and Obama) offered few communication appeals to young voters (Bennett & Xenos, 2004, 2005; Xenos & Bennett, 2007; Xenos & Foot, 2008). Such findings may help explain some of the participation gap between social networking sites and civic sites. At the time of this writing, Facebook had 25 million users between the ages of 13 and 25 in the United States, making them the largest demographic bracket among the roughly 95 million American users (InsideFacebook.com, 2009). Even the most popular youth engagement sites pale by comparison: Idealist.org attracts 400,000 monthly visitors (compete.com), and YouthNoise (personal communication) and TakingITGlobal (Raynes-Goldie & Walker, 2008) have memberships in the range of 100,000–150,000. What is lacking is a theoretical explanation of why so many civic organizations develop online communication environments that limit the uses of popular digital technologies. Looking within this population of youth sites, it also becomes important to identify and compare communication environments that may be grounded in fundamentally different conceptions of citizenship. An interesting theoretical clue about why youth sites so often lack expressive technology affordances is offered by Coleman (2008), who analyzed the degrees of communication freedom offered to young users in a collection of sites in the United Kingdom. He found a division between activist (largely youth-built) sites, which gave young users autonomy in their actions, and institutional or governmental sites, which heavily managed the experiences of their users. In Coleman's reading, these differences did not reflect superficial design choices, but philosophical differences about whether young citizens are fully formed, or “citizens in training” in need of structured democratic education (Coleman, 2008). We expand on this distinction, and on the promising idea of civic pedagogies offered by Bachen and colleagues with a more formal classification of different civic learning environments based on a typology of the different conceptions of citizenship that characterize (and often limit) the communication design of many sites. Rethinking citizenship and engagement in the online generation Without a more systematic model of how different conceptions of citizenship and engagement become coded into civic communication technologies, it will remain difficult to make rigorous comparisons and interpret differences across the spectrum of online communities. The model developed in this section returns to our opening concern that there are different practical paradigms of citizenship in play, not just in the scholarly literature, but in the changing repertoires of civic practices in play in most postindustrial democratic societies. These civic paradigms, although not mutually exclusive, are often poorly integrated in schools, organizations, and online environments, and they have not been operationalized in research. The longest-running citizenship paradigm in the United States (with variants in many other democracies) may be termed a model of dutiful citizenship (DC) that dates from the progressive era of the last century (Schudson, 1998). The core characteristic of the DC style is that individuals participate in civic life through organized groups, from civic clubs to political parties, while becoming informed via the news, and generally engaging in public life out of a sense of personal duty. These defining characteristics of DC civic engagement are notably in decline among younger generations in the United States (Putnam, 2000) and elsewhere (Inglehart, 1997). As these civic styles begin to fade, other researches have identified new civic orientations emerging among the same younger demographic in many nations. These civic trends include the rise of more personally expressive cause-oriented politics based on lifestyle concerns such as consumer behaviors, and the emergence of direct action protest networks in a variety of local to global arenas (Bennett, 1998; Inglehart, 1997; Norris, 2002, 2003; Zukin, Keeter, Andolina, & Jenkins, 2006). The social origins of these changes have been attributed to social fragmentation and the restructuring of society, economy, and personal life related to globalization (Bennett, 1998; Giddens, 1991). The generational scope of these changes is truly impressive. Norris (2003) has charted parallel trends in 15 nations involving generational declines in what she terms citizen-oriented activism (primarily related to voting and elections) and rises in cause-oriented activism. However, we are not convinced that it makes sense to juxtapose the rise of personal causes with citizenship itself, as Norris does; cause activism strikes us as a perfectly valid, if different, citizen action style. Nor do we wish to contrast the idea of DC with “engaged citizenship” as Dalton (2009) does, as the practices of dutiful citizens seem to be “engaged” to us, even if expressed in different ways. We suggest a broader conceptual shift toward thinking about declines in the older model of DC, and the rise of another style that Bennett (2007, 2008) has termed actualizing citizenship (AC). This citizenship typology enables us to think about a generational shift away from taking cues as members of groups or out of regard for public authorities (opinion leaders, public officials, and journalists), and toward looser personal engagement with peer networks that pool (crowd source) information and organize civic action using social technologies that maximize individual expression. This scheme points to the growing importance of participatory media (Jenkins, 2006; Pew, 2007a) in the rise of a new civic paradigm as shown in Table 1. Table 1 Dutiful and Actualizing Styles of Civic Action and Communication . Civic Style . Communication Logic . Dutiful • Oriented around citizen input to government or formal public organizations, institutions, and campaigns • Primarily one-way consumption of managed civic information (news and political ads) • Rooted in responsibility and duty • When individual content production occurs, it is aimed at specific institutional targets (contacting elected officials, letters to newspapers) • Channeled through membership in defined social groups Actualizing • Open to many forms of creative civic input, ranging from government to consumer politics to global activism • Lines between content consumption and production blurred • Rooted in self actualization through social expression • Individual content production and sharing over peer networks that tie personal identity to engagement (which can occur in traditional political contexts such as viral video sharing in political campaigns) • Personal interests channeled through loosely tied networks . Civic Style . Communication Logic . Dutiful • Oriented around citizen input to government or formal public organizations, institutions, and campaigns • Primarily one-way consumption of managed civic information (news and political ads) • Rooted in responsibility and duty • When individual content production occurs, it is aimed at specific institutional targets (contacting elected officials, letters to newspapers) • Channeled through membership in defined social groups Actualizing • Open to many forms of creative civic input, ranging from government to consumer politics to global activism • Lines between content consumption and production blurred • Rooted in self actualization through social expression • Individual content production and sharing over peer networks that tie personal identity to engagement (which can occur in traditional political contexts such as viral video sharing in political campaigns) • Personal interests channeled through loosely tied networks Open in new tab Table 1 Dutiful and Actualizing Styles of Civic Action and Communication . Civic Style . Communication Logic . Dutiful • Oriented around citizen input to government or formal public organizations, institutions, and campaigns • Primarily one-way consumption of managed civic information (news and political ads) • Rooted in responsibility and duty • When individual content production occurs, it is aimed at specific institutional targets (contacting elected officials, letters to newspapers) • Channeled through membership in defined social groups Actualizing • Open to many forms of creative civic input, ranging from government to consumer politics to global activism • Lines between content consumption and production blurred • Rooted in self actualization through social expression • Individual content production and sharing over peer networks that tie personal identity to engagement (which can occur in traditional political contexts such as viral video sharing in political campaigns) • Personal interests channeled through loosely tied networks . Civic Style . Communication Logic . Dutiful • Oriented around citizen input to government or formal public organizations, institutions, and campaigns • Primarily one-way consumption of managed civic information (news and political ads) • Rooted in responsibility and duty • When individual content production occurs, it is aimed at specific institutional targets (contacting elected officials, letters to newspapers) • Channeled through membership in defined social groups Actualizing • Open to many forms of creative civic input, ranging from government to consumer politics to global activism • Lines between content consumption and production blurred • Rooted in self actualization through social expression • Individual content production and sharing over peer networks that tie personal identity to engagement (which can occur in traditional political contexts such as viral video sharing in political campaigns) • Personal interests channeled through loosely tied networks Open in new tab The two citizenship models defined in Table 1 are ideal types based on characteristic skills required of citizens, such as how to recognize, filter, and use different kinds of political information; appropriate forms of public expression; modes of affiliations with others in groups or networks; and characteristic forms of participation. We can use this framework to examine the mix of different civic skills being communicated in different settings from schools to Web sites, and locate and compare various civic environments along the AC–DC continuum. Thus, we propose thinking about Web environments as sites for learning various models of citizenship, and we can use formal measures of civic learning to categorize and compare the conceptions of citizenship being communicated in different sectors of the youth civic Web sphere. A starting point for thinking about the civic skill sets that define different paradigms of citizenship is to look at the schools as places where the most formal citizenship training takes place. Civic skills and styles of citizenship At first glance, the diversity of school environments seems to offer few civic learning goals to which all educators subscribe (Flanagan & Faison, 2001; Niemi & Junn, 1998). Indeed, an earlier review of the civic education literature revealed dozens of different desirable outcomes of civic learning (Bennett, Wells, & Rank, 2009). However, these many civic learning goals can be synthesized into a more compact set of learning categories that seem to define essential citizen skill sets. Our earlier analysis suggested that school civics curricula are still communicating models of DC (Bennett, 2008; Bennett, Wells, & Rank, 2009; see also Campbell, 2005, 2007; Langton & Jennings, 1968; McDivett et al., 2003; McIntosh et al., 2007; Metz & Youniss, 2005). However, the basic categories seemed general enough to allow adding other practices that are more consistent with AC engagement styles (Kahne & Middaugh, 2008; Kahne & Westheimer, 2006). The following overview explains how we produced this elaborated model of citizenship practices that can be used to observe various mixtures of AC and DC civic practices being communicated in settings from classrooms to online environments. Two helpful resources for reducing the voluminous literature on civic learning goals and outcomes are the Civic Mission of Schools report, released by CIRCLE and the Carnegie Corporation in 2003 (Gibson & Levine, 2003), and the Campaign for the Civic Mission of Schools (CCMS), an ongoing effort to implement the report's recommendations (CCMS, 2008). The scholars and practitioners who worked on the report compiled and agreed upon a set of 40 “civic competencies” necessary for effective citizenship. Putting aside civic dispositions such as trust, and critical analysis skills, which cannot be easily assessed on civic Web sites, these competencies are sort into four key categories: The Knowledge necessary to be an effective citizen. The Expression skills needed to communicate effectively. The skills needed for Joining Publics (groups or networks) that can emerge, coordinate, and organize around an issue or candidate. The skills needed to Take Action to address a specific issue or policy. Each of these skill categories contains various competencies thought important for the good citizen. As defined in most school curricula, it is clear that these skills skew heavily toward the DC citizenship model, which may be one reason that civic education often produces less than desirable results (Syvertsen, Flanagin, & Stout, 2007). Starting with the basic DC skills set, we can add AC skills to expand the repertoire of appropriate citizen engagement. Knowledge in most conventional (DC) thinking is variously defined to encompass: information about history, the Constitution, the founding fathers, wars, and other events (CIRCLE, 2006; Niemi & Chapman, 1998); understanding how government and democracy work (CIRCLE, 2006; Gibson, 2001); and identifying specific officeholders (McDevitt & Kiousis, 2004; Syvertsen et al., 2007), candidates, and positions on specific issues (Pasek, Feldman, Romer, & Jamieson, 2008). Expression typically includes discursive, cooperation, negotiation, and persuasion skills (Feldman, Pasek, Romer, & Jamieson, 2007; Gibson, 2001; Levine, 2008; McDevitt & Kiousis, 2004; Syvertsen et al., 2007); and communication tools (such as writing letters) that citizens may need to express themselves in public contexts (e.g., Student Voices, 2008). Knowing how to be effective group members includes organizing political events, running meetings, and finding consensus within groups (CCMS, 2008); leadership skills (Gibson, 2001; Westheimer & Kahne, 2004); experience with community groups (Andolina, Jenkins, Zukin, & Keeter, 2003); and understanding what groups do (CCMS, 2008; Torney-Purta, 2002). Taking civic action is a culminating point for engagement (e.g., Zukin et al., 2006) that typically includes: voting (e.g., Pasek et al., 2008) or developing positive intentions to vote (Hooghe, Kavadias, & Reeskens, 2006); understanding how to affiliate and support a political party or social movement (Torney-Purta, 2002); fundraising, campaigning, or intending to run to office (Torney-Purta, Lehman, Oswald, & Schulz 2001). Some attention is also paid to consumer actions (CCMS, 2008) and controversial activities such as protests or political graffiti (Torney-Purta et al., 2001). To create a more encompassing civic learning typology, we next added to each learning category a number of skills better suited to actualizing citizens (see Bennett, Wells, & Rank 2009 for a more detailed explanation of this). For example, we augmented the knowledge category beyond conventional notions of one-way authoritative transmission of information from sources such as the press, teachers, or other authorities, by adding the importance of opportunities for peer-to-peer knowledge sharing of the sort developed by Wikipedia. In this fashion, we expanded each category of civic skills to include more AC skills, resulting in the framework in Table 2. Each learning category can now be thought of as a continuum that runs from AC skills to DC skills, which may be observed in various combinations in different learning and engagement environments. The idea is not that any individual citizen falls neatly into one type or another, but that the two civic styles may combine into a broader repertoire of choices. For many older citizens, the civic repertoire may be weighted with more DC skills, and younger citizens may draw upon more AC styles. This framework yields the operational measures described in the next section to assess the citizenship styles being communicated in different online civic youth environments. Table 2 Dutiful (DC) and Actualizing (AC) Forms of Four Categories of Civic Learning . Knowledge . Expression . Join Publics . Take Action . Definition . Information that citizens should know . Training in effective public communication skills . Learning how to connect to others through networks and groups . Actions that engage citizens with specific public issues or campaigns . . DC . AC . DC . AC . DC . AC . DC . AC . Definition Information provided by authorities (e.g., teachers, officials, press) Information created and shared by peers Training for traditional forms of public address to institutions & authorities (e.g., letters to newspapers, petitions to public officials) Training for self-produced and distributed digital media (e.g., blogs, election videos, culture jams) Membership in traditional, hierarchical organizations Membership in peer-defined networks and groups Activities defined and managed by formal organizations or authorities Activities generated and endorsed by peers . Knowledge . Expression . Join Publics . Take Action . Definition . Information that citizens should know . Training in effective public communication skills . Learning how to connect to others through networks and groups . Actions that engage citizens with specific public issues or campaigns . . DC . AC . DC . AC . DC . AC . DC . AC . Definition Information provided by authorities (e.g., teachers, officials, press) Information created and shared by peers Training for traditional forms of public address to institutions & authorities (e.g., letters to newspapers, petitions to public officials) Training for self-produced and distributed digital media (e.g., blogs, election videos, culture jams) Membership in traditional, hierarchical organizations Membership in peer-defined networks and groups Activities defined and managed by formal organizations or authorities Activities generated and endorsed by peers Note: AC = actualizing citizenship; DC = dutiful citizenship. Open in new tab Table 2 Dutiful (DC) and Actualizing (AC) Forms of Four Categories of Civic Learning . Knowledge . Expression . Join Publics . Take Action . Definition . Information that citizens should know . Training in effective public communication skills . Learning how to connect to others through networks and groups . Actions that engage citizens with specific public issues or campaigns . . DC . AC . DC . AC . DC . AC . DC . AC . Definition Information provided by authorities (e.g., teachers, officials, press) Information created and shared by peers Training for traditional forms of public address to institutions & authorities (e.g., letters to newspapers, petitions to public officials) Training for self-produced and distributed digital media (e.g., blogs, election videos, culture jams) Membership in traditional, hierarchical organizations Membership in peer-defined networks and groups Activities defined and managed by formal organizations or authorities Activities generated and endorsed by peers . Knowledge . Expression . Join Publics . Take Action . Definition . Information that citizens should know . Training in effective public communication skills . Learning how to connect to others through networks and groups . Actions that engage citizens with specific public issues or campaigns . . DC . AC . DC . AC . DC . AC . DC . AC . Definition Information provided by authorities (e.g., teachers, officials, press) Information created and shared by peers Training for traditional forms of public address to institutions & authorities (e.g., letters to newspapers, petitions to public officials) Training for self-produced and distributed digital media (e.g., blogs, election videos, culture jams) Membership in traditional, hierarchical organizations Membership in peer-defined networks and groups Activities defined and managed by formal organizations or authorities Activities generated and endorsed by peers Note: AC = actualizing citizenship; DC = dutiful citizenship. Open in new tab Research design: Civic skills and the communication of citizenship in the youth civic Web As explained earlier, establishing a baseline for measuring how different styles of citizenship are communicated in online settings is simplified somewhat by starting with sites that are explicitly dedicated to civic engagement. However, this is by no means a homogeneous collection of Web environments. In order to see if our models of citizenship are distributed differently across different kinds of organizations, we cast a wide net (as described below) to include four general categories of sites that offer youth civic skills and engagement experiences: (a) traditional interest groups such as the Sierra Club that offer some forms of online engagement to youth, (b) brick-and-mortar community youth organizations such as the YMCA that have online presences, (c) government agencies and political campaigns that have some Web presence with youth messaging, ranging from the Peace Corps to BarackObama.com, and (d) online-only youth civic communities that have little or no offline presence, such as Do Something and Taking IT Global. All these sites include some variant of youth civic engagement in their mission statements, but most present few details about precisely what is meant by “civic,”“citizenship,” or “engagement.” The general research question here is: How are the varieties of citizenship and related indicators of civic skills learning communicated and distributed over different types of sites in the youth civic Web? Hypotheses Drawing on the work of Bennett, Coleman, Jenkins, and others discussed above, and with particular reference to the work on organizations and social technology of Bimber, Flanagin, and Stohl (2005), we expect that more traditional organizations will signal primarily to dutiful citizens through relatively limited offerings of interactive affordances for sharing knowledge, expressing views, creating groups or networks, or proposing action. In contrast, “online-only” organizations may promote more AC-style civic engagement, largely because they have fewer historical organizational identity and mission constraints in defining relations to their publics, and thus more freedom to deploy technologies that enable emerging AC styles of citizenship and action. We also expect that some learning is occurring among these organizations, even though they may have imperfect knowledge of how citizenship is changing for different groups. It seems likely that more flexible organizations such as election campaigns that seek to attract younger citizens may communicate a mix of AC and DC engagement opportunities, as contrasted, for example, with community youth organizations whose identity and funding often depend on continuing their historic missions of individual leadership and community service. These propositions can be formulated as general hypotheses: H1: Youth engagement organizations that exist primarily or only in online forms will display a disproportionate share of the actualizing AC learning opportunities in the youth civic Web sphere. H2: Conversely, organizations that exist primarily in conventional offline forms of membership-based, hierarchical organizations (such as traditional community youth organizations and interest groups with youth branches) should communicate opportunities to learn engagement skills via more dutiful DC forms of knowledge, expression, joining, and action. H3: Relatively flexible organizations such as political campaigns may show signs of adapting to changing civic environments by offering more AC engagement opportunities than conventional interest or community organizations, but fewer than found in online-only civic environments. More generally, and in line with the earlier research reviewed above, we anticipate relatively few cases of pure AC communication environments online due to the preponderance of communication about DC citizenship in society. Such communication emanates largely from older generations who write about citizenship, design civic education policy and standards, and fund and manage most online communities. Thus, we expect to find a mix of AC and DC civic skills even in most of the online-only sites in our sample. Methods The sample We cast a wide net across the U.S. youth civic Web sphere (including international organizations with U.S. outreach), and then chose a selection of most-trafficked sites of different organization types. To assure identifying the broadest population of sites, and to establish continuity with other studies, we began with the sites found in earlier research. The list from Montgomery et al. (2004) resulted in 348 sites that passed an automated test of still being online (though many eventually proved to be inactive). From Bennett and Xenos (Bennett & Xenos, 2004, 2005; Xenos & Bennett, 2007) and Wells (2010) we drew an additional 70 sites. We then checked all the sites manually, eliminating 161 sites that were duplicates, or had not been active for a year or longer, or were no longer available. Next, to improve our sampling of major nonprofit organizations that do not often appear in studies of youth-only sites, we conducted Google searches using the names of the U.S.'s 100 biggest nonprofits (Clolery & Hrywna, 2006) paired with the search terms “youth,”“student,”“college,” and “social networking.” This yielded 22 additional sites. Finally, we conducted a search to identify sites pertaining to contemporary issues, and political or religious ideologies not uncovered by the above methods. Those searches included 54 key terms1 each combined with the youth-related search terms above. This yielded 88 further sites. Our initial catalog of living Web sites totaled 367. This population was then screened by undergraduate assistants and checked by the authors for sites (or sections of sites) having a primary focus on youth civic engagement.2 The 264 sites that met those criteria were then categorized by the authors according to their mission and the type of organization that created them. First, if a site was entirely online, without reference to offline activities, it was placed in an Online-Only category. Fifty-six sites fell in that category. A focus on a government organization, a candidate for office, or a political party, placed a site in a Government/Candidate category (28 sites). Organizations providing community or service involvement without explicit advocacy were placed in a Community/Service category (84 sites). Sites promoting advocacy for a cause or particular political interest group were placed in an Interest/Activist category (98 sites). For the most part, the clusters of sites were easy to sort out and fit well together. The one case where the members of a category displayed topical overlap involved Online-Only sites that displayed political, service, or activist agendas. However, our theoretical framework suggests that the distinction between conventional organizations and online-only organizations is critical for understanding underlying differences in conceptions of citizenship that affect the design of communication environments. The four site clusters were then adjusted to make sure that we included organizations that (a) represented local levels of large multibranch national organizations, or (b) might have eluded the mechanical search terms, but that offered clear opportunities for youth engagement (e.g., election campaign organizations or interest groups such as Sierra Club that are not youth-specific, but that have youth outreach).3 Finally, we used http://www.compete.com to obtain rough estimates of comparative traffic levels on the sites, and constructed a preliminary list of the most-trafficked sites in each category.4 Based on the size of the coding challenge, we cut off our sample at 90, with an oversample of 35 in the Online-Only category (in order to accommodate the great diversity of sites in this category), 15 in the Government/Candidate category (reflecting the smaller numbers and more limited youth focus in this category), and 20 each in the Community/Service and Interest/Activist categories.5 The proportional differences in subsample sizes were controlled for in all data analysis and reporting. Coding Coding was divided into two stages, each conducted by a different team of coders to avoid confounding the identification of whether a site page contained a civic learning opportunity (knowledge, expression, joining, and acting) with the civic style (actualizing or dutiful) that opportunity represented. The stages were a page selection process, in which one group of expert coders (including two of the authors) determined which pages contained general learning opportunities, and dutiful/actualizing coding, in which a different set of (blind) coders determined whether dutiful and/or actualizing forms of each learning goal were present on the pages selected. Page selection In the page selection process, three coders independently evaluated each site for the presence of any of the four learning goal categories. Navigating from the home page, the coders looked at each page linked from the main menu bars. The coders were thus limited to pages one link from the homepage, a choice that reduced the potential for randomness in searching differently designed Web sites, while capturing prominent features, and making the search through voluminous sites manageable. (Full code sheet available upon request.) Coders were instructed to select up to three pages for each learning category found on a site, to ensure that different (AC, DC, or both) forms of a particular civic skill were captured. The percent agreement across three coders for the presence of a learning category was 84.4% for all four combined, broken down to 83.9% for Knowledge, 82.1% for Expression, 82.0% for Joining Publics, and 89.7% for Take Action. Agreement on particular pages containing those goals was also very high, with some fall off from the first to the third page, as might be expected.6 The cases in which coders differed were resolved by consensus discussion to produce a common set of up to three pages per learning category. As Bachen et al. (2008) note, such pairwise percentage agreements are above the accepted level for this kind of analysis. Dutiful and actualizing coding A different team of four coders was trained to code the selected pages to determine whether the learning opportunities they contained reflected more dutiful or actualizing civic styles. For Knowledge the page features that presented users with information from one-way, authoritative sources such as site sponsors, news stories, public officials, or other external sources were coded as DC Knowledge, and opportunities for peer knowledge sharing were coded as AC.7 For Expression the pages offering communication training to produce content aimed at institutions or authorities, such as tips on how to write good letters or effectively telephone an elected official were coded DC; training in digital participatory media, such as how to create a podcast, design a video, or effectively communicate with a blog were coded AC. For Joining Publics the groups that were hierarchical, or created and defined by the site or an affiliated organization were coded DC; opportunities for users to define their own groups, or join groups or networks created by peers were coded AC. For Take Action, activities organized and managed by site sponsors or affiliated organizations and authorities were scored DC; peer generated actions suggested or reported upon by site users were scored AC. Reliability After the four coders were trained, a random 16-site subsample was selected for a reliability test. The reliabilities (pairwise percent agreement for four coders) were 100% for DC Knowledge (every site offering information provided some sort of DC Knowledge); 78% for AC Knowledge; 92% for DC Expression; 81% for AC Expression; 84% for DC Joining Publics; 94% for AC Joining Publics; 98% for DC Take Action; and 91% for AC Take Action.8 The overall reliability for all eight forms of learning was 91%. Once again, in line with the criteria proposed by Bachen et al. (2008), these exceed acceptable levels. Figure 1 illustrates how the coding process was applied to one particular page. This page is from http://dosomething.org, a site from the Online-Only category, and was selected in the page selection process as presenting Take Action learning because the site chose to name it “Projects” and because it clearly presented actions for users to take. In dutiful/actualizing coding, the coder determined that an actualizing opportunity to Take Action was present, because users could see actions which peers had taken, and suggest their own actions. Similarly, the coder determined that no dutiful action forms were present, as no action opportunities originated from site sponsors or external organizations or authorities. The page was coded as a Take Action learning opportunity, in an actualizing (AC) form. Figure 1 Open in new tabDownload slide Coding process example. Example of AC Take Action code from the Online-Only organization DoSomething.org. Figure 1 Open in new tabDownload slide Coding process example. Example of AC Take Action code from the Online-Only organization DoSomething.org. Results The sample of 90 sites produced a total of 255 civic learning opportunities, with 76% of those signaling the dutiful citizen style. As predicted, different types of organizations signal different models of citizenship, with the majority of actualizing learning opportunities (53%) occurring in the online-only site types, although those sites constituted only about one third of the total sample. Figure 2 illustrates these differences more clearly, showing that online-only youth sites have several notable qualities absent in sites produced by more conventional civic organizations: a greater balance of the four major categories of learning, a better mix of AC and DC learning opportunities, and by far the greatest opportunity to learn how to participate in peer-to-peer actualizing forms of knowledge sharing and public expression. Indeed, opportunities to learn any form of expression are scarce in conventional community, interest, and government organizations, which are aimed disproportionately at getting young people to engage with site-defined activities. In contrast, nearly 70% of the sites in the online-only group offer some expression training, and most of those provide either a mix of AC and DC expression skills or exclusively AC skills. It seems clear that conventional civic organizations overwhelmingly regard young people as subjects to be heavily guided, or as “citizens in training” who should be told what to do by authoritative figures (Coleman, 2008). Figure 2 Open in new tabDownload slide Distribution of learning opportunities. Figure 2 Open in new tabDownload slide Distribution of learning opportunities. Also in keeping with our predictions, the government and campaign category offered slightly greater opportunities for actualizing civic skills learning, particularly when compared to community organizations, and, to some extent, compared to interest/activist groups. This was driven largely by Clinton and Obama campaign affordances for sharing knowledge, blogging, and forming personal support groups. Figure 3 displays a portion of a page on the Clinton youth site Hillblazers titled “Spread the Word.” This page was selected in the category of Knowledge learning in the page selection process, and then coded for both dutiful and actualizing opportunities. On the DC side, it offers a link to information about Clinton provided by the site (“Learning more about Hillary–Hillary 101”)—typical for campaigns wishing to manage knowledge acquisition about their candidate as human and inspiring, and signaling a dutiful campaign-supporter relationship. At the same time, the heart of the page encourages supporters to tell Clinton's story in their own way (“You know better than we do what will get your friends and peers excited about Hillary…this is your opportunity to put your mark on this campaign”), a much more actualizing pattern. The signal sent is that learning might occur not only as a result of direct interaction with the information on the site, but also as a result of expression and sharing with peers. Figure 3 Open in new tabDownload slide “Spread the Word” page from Hillblazers. Example of Both actualizing (AC) and dutiful citizenship (DC) engagement with Knowledge skills: opportunities to gather information about the candidate both provided by the site, and by users sharing their own knowledge of the candidate. Figure 3 Open in new tabDownload slide “Spread the Word” page from Hillblazers. Example of Both actualizing (AC) and dutiful citizenship (DC) engagement with Knowledge skills: opportunities to gather information about the candidate both provided by the site, and by users sharing their own knowledge of the candidate. A striking similarity across all four categories of organizations that we did not fully anticipate is the tendency to offer highly managed (DC) opportunities for taking action, even in online-only youth communities. When it comes to enabling young citizens to act, it seems that site operators are generally not inclined to empower young people to create and promote their own activities. Perhaps this reflects clear action goals on the part of sponsoring organizations, but it may also reveal an implicit sense that young people are not capable of engineering their own effective action plans. The lack of advocacy training is a clear pattern across the sample. A typical site offering a core DC action skill is the voter registration page shown in Figure 4. Figure 4 Open in new tabDownload slide A core dutiful citizenship (DC) civic learning skill: registering to vote on declareyourself.com. Figure 4 Open in new tabDownload slide A core dutiful citizenship (DC) civic learning skill: registering to vote on declareyourself.com. Discussion One obvious conclusion from the data is that, as predicted in H1, conventional community and interest/activist organizations overwhelmingly reproduce their offline DC models of citizenship and civic engagement in their online sites. In one sense, it is not surprising that organizations believe that their online presence should reproduce their existing organizational identity. However, as many observers have noted, conventional civic organizations face shrinking memberships precisely because younger generations are not inclined to enter into formal membership relations (Bimber et al., 2005; Putnam, 2000). As predicted in H2, organizations that exist primarily as online communities have shed at least some conventional notions about proper (i.e., DC) citizenship, enabling them to communicate different kinds of relationships to user communities by offering them a mix of AC and DC civic skill sets. There are lessons here for both conventional organizations and for policymakers and funders looking for places to invest resources: The potential of online engagement to reach a broader population of younger citizens is barely being achieved because of rather rigid notions of who they are and how to communicate with them. Among conventional hierarchical civic organizations, it seems that some election campaigns (consistent with the prediction in H3) have begun to realize that potential to adjust their models of citizen–organization interaction to enable more AC participation in knowledge and content sharing and social networking, but it remains to be seen whether the war room command and control model of campaigning will ultimately stop yielding ground to autonomous social networking for fear of losing control of messages and organization. Another interesting implication of our findings about organizations defined by their conceptions of citizens is that while organizations that exist only or mostly online tend to offer the most balanced mix of knowledge and expression opportunities, they look much like other kinds of organizations when it comes to opportunities for joining publics and taking action. Moreover, joining and acting are so managed that they are often not even presented in terms of learning (just doing). Although some interest and activist groups do offer training kits on how to organize and act, such formal learning opportunities are rare. As a result, in order to include these key elements of civic engagement in our data, we had to lower the bar to adopt a “learning by doing” standard that included opportunities to join or act that were not accompanied by training in formal skills. This makes our findings generally consistent with the overall low levels of civic pedagogy reported by Bachen et al. (2008). At the same time, our finding that online-only sites differed systematically in their communication with AC citizens in two important learning categories (knowledge and expression) challenges the Bachen et al. (2008) finding that nonprofit sites offered richer and more interactive learning opportunities than either government or commercial sites. Our site typology shows that when sorted in specific terms of how organizations conceive of citizens, nonprofits are far from a uniform category. Conventional nonprofits (in the Community/Service and Interest/Activist site groups) offered the fewest actualizing (interactive learning) experiences. This theory-driven framework offers a useful starting point for thinking about how to identify and understand different implicit organizational models of citizenship as they are communicated through civic learning and engagement features on Web sites. One clear implication is that site owners may not be aware of (or able to transcend) the gaps between the often narrow citizenship ideals coded in online civic environments and the broader range of participatory media and social networking opportunities that young people routinely encounter in other environments such as Facebook. More personally creative and expressive opportunities for civic engagement can be offered easily on the civic Web if organizations can adjust their apparent needs to manage relationships with their publics. Indeed, the next study in this research project shows that user activity levels on various site features such as blogs and forums are significantly higher on sites that offer more opportunities to learn AC learning civic skills. We hope that our framework for thinking about how citizenship is communicated to young people is useful both for scholars seeking to understand the wellsprings of civic learning and engagement online and for practitioners who seek to develop more effective environments for youth participation. Acknowledgments The authors acknowledge the generous support of the John D. and Catherine T. MacArthur Foundation Digital Media and Learning Program that made this research possible. We also thank our coders: Jorunn Mjos, Melissa Aar, Daniel Weisbeck, and Lauren Snyder. Notes 1 " Key categories included political positions (e.g., “libertarian,”“socialist”), political issues (“gay rights,”“2nd amendment”), current issues of concern (“sustainability,”“Darfur,”“media literacy”), ethnicities (“African American,”“Latino”), and religions (“Christian,”“Muslim”). 2 " Our process involved looking first on the homepage, then on an “About” page, for evidence that the site was for youth (e.g., references to “youth,”“students,”“kids,” and any age ranges under 30) and that it involved some form of public engagement (e.g. “getting involved,”“improving one's community,”“speaking out,”“activism”). 3 " First, to include the kinds of local, community sites that most youth would be likely to interact with, we replaced the sites of national-level organizations in the Community/Service category with the sites of local branches, selected based on searches using randomly generated zip codes. For the Girl Scouts we drew the site of a Madison, Wisconsin chapter; for the Boys and Girls Clubs we drew the site of Metropolitan Denvers' clubs; and for 4H we drew the site of 4H in North Carolina. Second, even though they are not focused primarily on youth, we added the 2008 political campaign sites of McCain, Obama, and Clinton to assess any differences in the way they communicated engagement to young voters. Finally, we added to the sample a number of major interest organizations that eluded our search for organizations that had an explicit focus on youth, but that offered youth sections on their Web sites. This included the ACLU, NRA, and Sierra Club. 4 " The use of compete.com was simply a convenient way to roughly sort the sample. We do not claim that compete offers actual site traffic measures, but a rough indicator of comparative volumes that proved helpful in selecting sites without resorting to subjective criteria. For reference, for the 72 sites for which Compete data might be considered reasonably reliable, the sites averaged 21,547 monthly visits, with a high of almost 400,000 for idealist.org, and lows below 1,000 for several sites. 5 " The Online-Only category includes idealist.org, youthnoise.com, battlecry.com, declareyourself.com, rockthevote.org, takingitglobal.org, itsgettinghotinhere.org, dosomething.org. The government/campaign category includes libertarianrock.com, bostonyouthzone.com, peacecorps.gov, collegerepublicans.org, hillaryclinton.org, johnmccain.com, go.barackobama.com. Community organizations include Web sites for local chapters of Key Club, YMCA, Girl Scouts, Jewish Service, and 4-H. Interest group/activist organization sites included NRA, ACLU, Sierra Club, Feminist Campus, Out Proud, and Students for a Free Tibet. 6 " Following agreement that a learning goal was present on a site, agreement on the first page selected was as follows: Knowledge, 91.6%; Expression, 96.9%; Joining Publics, 97.8%; Take Action, 95.6%; with total first page agreement at 95%. Looking at the second most selected page, though slightly lower, the agreement is acceptable: Knowledge, 79.6%; Expression, 75.0%; Joining Publics, 89.7%; Take Action, 73.9%. Coders only selected three pages for single learning goal on a site if all coders agreed, following deliberation, on all three pages. 7 " Here is the coding instruction for Knowledge from the codebook. Positive scores on the (1) code were scored DC, and positive scores on the (2) code were scored AC:
Some sites want users to be able to find out more about issues or related matters when they are on the Web site. Some sites have pages that are specifically for this purpose. Looking at the pages selected from this site: " 1. Do any of the pages contain a listing of facts or background reports (on issues, problems, or how some political process works) produced by experts or by the operators of the site? " For example, some sites might have sections such as “what they never told you,” or “inform yourself on important issues,” or “what you should know about X (e.g., global warming)” " 2. Do any of the pages encourage users to share what they know about community or public issues or related matters with other users? " For example, some sites enable users to post their own reports and projects in order to share they have learned about a particular problem with other users. 8 " One coder's results in the reliability subsample were significantly different from others' on AC Take Action (that coder did not record any Take Action codes). Leaving these systematic errors out made the DC Take Action agreement 97%, and the AC Take Action agreement 100%. References Andolina , M. W. , Jenkins , K., Zukin , C., & Keeter , S. ( 2003 ). Habits from home, lessons from school: Influences on youth civic engagement . PS: Political Science & Politics , 36 ( 2 ), 275 – 280 . Google Scholar Crossref Search ADS WorldCat Bachen , C. , Raphael , C., Lynn , K. M., McKee , K., & Philippi , J. ( 2008 ). Civic engagement, pedagogy, and information technology on Web sites for youth . Political Communication , 25 ( 3 ), 290 – 310 . Google Scholar Crossref Search ADS WorldCat Benkler , J . ( 2006 ). The wealth of networks: How social production transforms markets and freedom. New Haven, CT: Yale University Press. Retrieved from http://www.benkler.org/wealth_of_networks/index.php/Download_PDFs_of_the_book. Bennett , W. L . ( 1998 ). The uncivic culture: Communication, identity, and the rise of lifestyle politics . PS-Political Science and Politics , 31 ( 4 ), 741 – 761 . Google Scholar Crossref Search ADS WorldCat Bennett , S. , Mator , K., & Kervin , L. ( 2008 ). The “digital natives” debate: A critical review of the evidence . British journal of educational technology , 39 ( 5 ), 775 – 786 . Google Scholar Crossref Search ADS WorldCat Bennett , W. L . ( 2007 ). Civic learning in changing democracies: Challenges for citizenship and civic education. In P. Dahlgren (Ed.), Young citizens and new media: Learning democratic engagement (pp. 59 – 77 ). New York, NY : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bennett , W. L . ( 2008 ). Changing citizenship in the digital age. In W. L. Bennett (Ed.), Civic life online (pp. 1 – 24 ). Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bennett , W.L. , Wells , C., & Rank , A. ( 2009 ). Young citizens and civic learning: Two paradigms of citizenship in the digital age . Citizenship Studies , 13 ( 2 ), 105 – 120 . Google Scholar Crossref Search ADS WorldCat Bennett , W. L. , & Xenos , M. ( 2005 ). Young voters and the Web of politics 2004: The youth political Web sphere comes of age. CIRCLE Working Paper 42, October 2005. Bennett , W. L. , & Xenos , M. ( 2004 ). Young voters and the Web of politics: Pathways to participation in the youth engagement and electoral campaign Web spheres. CIRCLE Working Paper 20, August 2004. Bimber , B. , Flanagin , A., & Stohl , C. ( 2005 ). Reconceptualizing collective action in the contemporary media environment . Communication Theory , 15 , 389 – 413 Google Scholar Crossref Search ADS WorldCat Campaign for the Civic Mission of the Schools (CCMS). ( 2008 ). Civic competencies. Retrieved from http://civicmissionofschools.org/site/resources/civiccompetencies.html. Campbell , D. E . ( 2005 ). Voice in the classroom: How an open classroom environment facilitates adolescents' civic development. CIRCLE Working Paper Series. Retrieved from http://www.civicyouth.org/PopUps/WorkingPapers/WP28campbell.pdf. Campbell , D. E . ( 2007 ). Sticking together: Classroom diversity and civic education . American Politics Research , 35 ( 1 ), 57 – 78 . Google Scholar Crossref Search ADS WorldCat CIRCLE. ( 2006 ). Civic education: What students are learning. Retrieved from http://www.civicyouth.org/index.php?s=iea. Clolery , P. , & Hrywna , M. ( 2006 , November 1). Revenues of NPOs soaring. Nonprofit Times. Retrieved from http://www.nptimes.com/pdf/NPTTop1002006.pdf. Coleman , S . ( 2007 ). From big brother to big brother: Two faces of interactive engagement. In P. Dahlgren (Ed.), Young citizens and new media: Learning democratic engagement . New York, NY : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Coleman , S . ( 2008 ). Doing IT for themselves: Management versus autonomy in youth e-citizenship. In W. L. Bennett (Ed.), Civic life online: Learning how digital media can engage youth (pp. 189 – 206 ). Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dalton , R. J . ( 2009 ). The good citizen: How a younger generation is reshaping American politics . Washington, DC : CQ Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC European Commission ( 2007 ). Young people, the Internet and civic participation. European Commission Sixth Framework Program, Report D6. Retrieved from http://www.civicweb.eu/content/blogcategory/3/7/. Feldman , L. , Pasek , J., Romer , D., & Jamieson , K. H. ( 2007 ). Identifying best practices in civic education: Lessons from the Student Voices program . American Journal of Education , 114 , 75 – 100 . Google Scholar Crossref Search ADS WorldCat Flanagan , C. A. , & Faison , N. ( 2001 ). Youth civic development: Implications of research for social policy and programs . Social Policy Report , 15 ( 1 ), 3 – 15 . OpenURL Placeholder Text WorldCat Ganz , M . ( 2009 , July 24). Remarks at Tufts University Symposium on “Obama administration's civic agenda.” Retrieved from http://www.c-spanarchives.org/program/id/210121. Gibson , C . ( 2001 ). From inspiration to participation: A review of perspectives on youth civic engagement. A Report from the Grantmaker Forum on Community & National Service and Carnegie Corporation. Retrieved from http://www.pacefunders.org/publications/pubs/Moving%20Youth%20report%20REV3.pdf. Gibson , C. , & Levine , P. ( 2003 ). Civic mission of schools. Carnegie Corporation of New York and CIRCLE. Retrieved from http://www.civicmissionofschools.org/site/campaign/cms_report.html. Hargittai , E. , & Hinnant , A. ( 2008 ). Digital inequality: Differences in young adults' use of the Internet . Communication Research , 35 ( 5 ), 602 – 621 . Google Scholar Crossref Search ADS WorldCat Hargittai , E. , & Walejko , G. ( 2008 ). The participation divide: Content creation and sharing in the digital age . Information, Communication and Society , 11 ( 2 ), 239 – 256 . Google Scholar Crossref Search ADS WorldCat Hooghe , M. , Kavadias , D., & Reeskens , T. ( 2006 , August 31). Intention to vote and future voting behavior: A multi-level analysis of adolescents in 27 countries. Paper at the 2006 Annual Meeting of the American Political Science Association, Philadelphia, PA. Retrieved from http://www.allacademic.com/meta/p152356_index.html. InsideFacebook.com ( 2009 , March 25). Number of US Facebook users over 35 nearly doubles in last 60 days. Retrieved from http://www.insidefacebook.com/2009/03/25/number-of-us-facebook-users-over-35-nearly-doubles-in-last-60-days/. Jenkins , H . ( 2006 ). Convergence culture: Where old and new media collide . New York, NY : NYU Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kahne , J. , & Middaugh , E. ( 2008 ). Democracy for some: The civic opportunity gap in high schools. CIRCLE Working Paper Series. Retrieved from www.civicyouth.org/PopUps/WorkingPapers/WP59Kahne.pdf. Kahne , J. , & Westheimer , J. ( 2006 ). The limits of political efficacy: Educating citizens for a democratic society . PS-Political Science and Politics , 39 ( 2 ), 289 – 296 . Google Scholar Crossref Search ADS WorldCat Langton , K. P. , & Jennings , M. K. ( 1968 ). Political socialization and the high school civics curriculum in the United States . American Political Science Review , 62 ( 3 ), 852 – 867 . Google Scholar Crossref Search ADS WorldCat Levine , P . ( 2008 ). A public voice for youth: The audience problem in digital media and civic education. In W. L. Bennett (Ed.), Civic life online: Learning how digital media can engage youth (pp. 119–138) Cambridge, MA : MIT Press , 119 – 138 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC MacArthur Online Discussions on Civic Engagement ( 2006 ). Retrieved from http://spotlight.macfound.org/resources/CivicEngagement-Online. Madden , M . ( 2006 ). Internet penetration and impact . Washington, DC : Pew Internet and American Life Project . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McDevitt , M. , & Kiousis , S. ( 2004 ). Education for deliberative democracy: The long-term influence of Kids Voting USA. CIRCLE Working Paper Series. Retrieved from http://www.civicyouth.org/PopUps/WorkingPapers/WP22McDevitt.pdf. McDevitt , M. , Kiousis , S., Wu , X., Losch , M., & Ripley , T. ( 2003 ). The civic bonding of school and family: How kids voting students enliven the domestic sphere. CIRCLE Working Paper Series. Retrieved from http://www.civicyouth.org/PopUps/WorkingPapers/WP07McDevitt.pdf. McIntosh , H , Berman , S. H., & Youniss , J. ( 2007 ). An interim report of the evaluation of a comprehensive high school civic engagement intervention in Hudson, MA. CIRCLE Working Paper Series. Retrieved from http://www.civicyouth.org/PopUps/WorkingPapers/WP58McIntosh.pdf. Metz , E. C. , & Youniss , J. ( 2005 ). Longitudinal gains in civic development through school-based required service . Political Psychology , 26 ( 3 ), 413 – 437 . Google Scholar Crossref Search ADS WorldCat Montgomery , K. , Gottleib-Robles , B., & Larson , G. O. ( 2004 ). Youth as e-citizens: Engaging the digital generation. Retrieved from http://www.centerforsocialmedia.org/ecitizens/index2.htm. Niemi , R. G. , & Chapman , C. ( 1998 ). The civic development of 9th- through 12th-grade students in the United States: 1996. Report for U.S. Department of Education, National Center for Education Statistics. Niemi , R. G. , & Junn , J. ( 1998 ). Civic education: What makes students learn? New Haven, CT : Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Norris , P . ( 2002 ). Democratic phoenix . New York, NY : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Norris , P . ( 2003 ). Young people & political activism: From the politics of loyalty to the politics of choice? Retrieved from http://ksghome.harvard.edu/~pnorris/Acrobat/COE.pdf. Palfrey , J. , & Gasser , U. ( 2008 ). Born digital: Understanding the first generation of digital natives . New York, NY : Basic Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pasek , J. , Feldman , L., Romer , D., & Jamieson , K. H. ( 2008 ). Schools as incubators of democratic participation: Building long-term political efficacy with civic education . Applied Developmental Science , 12 ( 1 ), 26 – 37 . Google Scholar Crossref Search ADS WorldCat Pasek , J. , Kenski , K., Romer , D., & Jamieson , K. H. ( 2006 ). America's youth and community engagement: How use of mass media is related to civic activity and political awareness in 14- to 22-year olds . Communication Research , 33 ( 3 ), 115 – 135 . Google Scholar Crossref Search ADS WorldCat Patterson , T. E . ( 2007 ). Young people and news. A report from the Joan Shorenstein Center on the Press, Politics, and Public Policy. Kennedy School of Government, Harvard University. Carnegie-Knight Task Force on the Future of Journalism Education. Pew Internet and Public Life Project ( 2007a ). Teens and social media: The use of social media gains a greater foothold in teen life as they embrace the conversational nature of interactive online media. Washington, DC. Retrieved from http://www.pewinternet.org/PPF/r/230/report_display.asp. Pew ( 2007b ). What Americans know: 1989–2007. Pew Research Center for the People and the Press. Retrieved from http://people-press.org/reports/pdf/319.pdf. Raynes-Goldie , K. , & Walker , L. ( 2008 ). Our space: Online civic engagement tools for youth. In W. L. Bennett (Ed.), Civic life online (pp. 161–188). Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schudson , M ., ( 1998 ). The good citizen: A history of American civic life . New York, NY : The Free Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Student Voices. ( 2008 ). Retrieved from http://student-voices.org/. Syvertsen , A. K. , Flanagin , C. A., & Stout , M. D. ( 2007 ). Best practices in civic education: Changes in students' civic outcome. CIRCLE Working Paper Series. Retrieved from http://www.civicyouth.org/PopUps/WorkingPapers/WP57Flanagan.pdf. Torney-Purta , J . ( 2002 ). The school's role in developing civic engagement: a study of adolescents in twenty-eight countries . Applied Developmental Science , 6 ( 4 ), 203 – 212 . Google Scholar Crossref Search ADS WorldCat Torney-Purta , J. , Lehman , R., Oswald , H., & Schulz , W. ( 2001 ). Citizenship and civic education in twenty-eight countries: Civic knowledge and engagement at age fourteen. International Association for the Evaluation of Educational Achievement. Retrieved from www.wam.umd.edu/~iea/. Wattenberg , M. P . ( 2006 ). Is voting for young people? London, England : Longman . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wells , C . ( 2010 ). Citizenship and communication in online youth civic engagement projects . Information, Communication and Society , 13 ( 3 ), 419 – 441 . Google Scholar Crossref Search ADS WorldCat Westheimer , J. , & Kahne , J. ( 2004 ). What kind of citizen? The politics of educating for democracy . American Educational Research Journal , 41 ( 2 ), 237 – 269 . Google Scholar Crossref Search ADS WorldCat Xenos , M. , & Bennett , W. L. ( 2007 ). The disconnection in online politics: The youth political Web sphere and US election sites, 2002–2004 . Information, Communication, and Society , 10 ( 4 ), 443 – 464 . Google Scholar Crossref Search ADS WorldCat Xenos , M. , & Foot , K. ( 2008 ). Not your father's Internet: The generation gap in online politics. In W. L. Bennett (Ed.), Civic life online (pp. 51 – 70 ). Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zukin , C. , Keeter , S., Andolina , M., & Jenkins , K. ( 2006 ). A new engagement? Political participation, civic life, and the changing American citizen . New York, NY : Oxford University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC © 2011 International Communication Association

Boczkowski, Pablo, J.;Peer,, Limor

doi: 10.1111/j.1460-2466.2011.01582.xpmid: N/A

Abstract This study examines 4 online news sites to compare stories that journalists display most prominently with stories consumers read most frequently. We find that journalists' chosen stories are “soft” with respect to subject matter but not story format, and that these choices diverge from consumers' choices, resulting in a choice gap. The study design makes important methodological contributions by using the story as the unit of analysis, operationalizing “soft news” in terms of subject matter as well as format, and considering the influence of journalists' and consumers' choices on each other. This article discusses the implications of the findings on such issues as the dynamics of agenda setting, the prospects for consumer-authored content, and the watchdog function of the media. There has been growing agreement among media scholars about a trend toward the “softening of news,” for lack of a better phrase (Baum, 2002; Bennett, 2003a; Patterson, 2000; Project for Excellence in Journalism, 1998; Schudson, 2003; Zaller, 2003). Research has often identified market forces as a critical driver of this trend (Baum, 2003; Delli Carpini & Williams, 2001; Hamilton, 2004; Kalb, 1998; Prior, 2007). The reduced competition enjoyed by American media organizations during most of the 20th century allowed them to combine strong economic performance with an editorial supply marked by a journalistic preference for hard news focused on public affairs topics (Fishman, 1980; Gans, 2004; Kaniss, 1991; Sigal, 1986; Tuchman, 1979). This situation persisted for decades, despite perception of a gap between this type of news and consumption patterns that expressed only moderate interest in it (Bogart, 1989). With recent increase in competition, companies have paid more attention to consumers' preferences and changed their supply of news accordingly (Klinenberg, 2005; McManus, 1994; Mindich, 2005; Underwood, 1993). However, scholars have debated whether these changes narrow the gap by giving consumers more of what they supposedly want, or create a new gap by underserving what could be seen as the main public for news—those highly interested in public affairs stories (Baum, 2002; Hamilton, 2004; Patterson, 2000, 2003; Prior, 2002). Current scholarship suffers from three limitations that prevent it from accurately characterizing the gap between journalists and news consumers, much less settling this debate. First, although most research relies on survey, circulation, and ratings data that measure consumer choices in the aggregate, consumers increasingly make their choices at the story level. Second, scholars disagree on what “soft news” means. Some focus on what stories are told, others on how they are told, and a third group on a combination of these two foci. Third, studies have inadequately addressed the influence of journalists' and consumers' choices on each other, thus potentially misrepresenting each group's choices. This article attempts to provide an analysis of the gap between journalists and news consumers in terms of their preference for hard or soft news. It does so by examining the character of the top news choices of journalists on four websites on the hard–soft dimension and by determining whether there is a gap between these choices and the top choices made by the consumers of these sites. Furthermore, this study is designed to overcome the three shortcomings identified above. First, the story is used as the unit of analysis to have a more fine-grained picture of consumers' choices. Second, the analysis discriminates between what stories are chosen and how they are told to have a more precise characterization of the object of study. Third, it contrasts all the stories collected with a subset of stories in which the mutual influence of journalists' and consumers' choices is likely to be less pronounced. We do this in the context of online news sites, which provide a fertile terrain to look at these issues because the Web is a highly competitive information environment that will likely have increased influence on the future of the industry. Results show a soft news supply in terms of the subject of news stories, but not in terms of how they are covered. They also show a sizable gap across the four sites between journalists' and consumers' choices when the subject of news is concerned, but a lesser and more variable gap regarding the story-telling format. The analysis presented in this article makes methodological and conceptual contributions. The methodological contributions center on overcoming the limitations mentioned above. The conceptual benefits derive from a better understanding of what a supply of soft news consists of, and of exactly how the news choices of journalists and consumers diverge. In turn, these benefits help make sense of transformations in the news industry and its publics, such as the dynamics of agenda setting, the prospects for consumer-authored content, and the watchdog function of the media in liberal democracies. Theoretical considerations There is a growing consensus among media scholars and analysts that the news has “gone soft.” There were already signs of this trend in studies that were published decades ago. For instance, the report The Age of Indifference (Times Mirror Center, 1990) maintains that “even some more traditional media have adopted a softer news focus in response to diminishing levels of interest in hard news” (p. 2). In their study of television news from 1972 to 1987, Scott and Gobetz (1992) argued that “there has been a slight tendency for television network news programs to increase the amount of soft news presented mostly during the last one-third of the newscast” (p. 406). The trend seems to have intensified recently, affecting both news products and how journalists experience their production. Thus, to Bennett (2003b), “what has happened to the news in the past twenty years is that it has shifted in the direction of soft news and sensationalism” (p. 131). Moreover, in the preface to the 25th anniversary edition of Deciding What's News, Gans (2004) reflected on that quarter-century and concluded that “the change in news content that exercises journalists the most is a decrease in hard … news” (p. xiv). Researchers have often regarded market dynamics as central in the evolution of this trend. During major part of the 20th century, American news organizations enjoyed relatively low levels of competition in their respective markets. This led to the coexistence of companies with strong economic performance, journalists with professional values marked by a preference for public affairs content, a style focused on issues and a disdain for their audiences (in Darnton's [1975, p. 176] apt formulation, journalists “really wrote for one another”), and a separation between the editorial and commercial logics of the enterprise. Lack of market competition allowed this mode of production to survive despite the perception of a gap between the interests of journalists and consumers. For instance, Bogart (1989) examines data on newspaper readers' ideal space allocation and interest in various topics and contrasts that with content analyses and editors' importance ratings. He reports that 49% of the stories in a typical newspaper are public affairs stories, whereas about 29% focus on nonpublic affairs content (Bogart, 1989, see Table 6.1, p. 192), but that “readers want more of certain kinds of content than they presently get,” including primarily nonpublic affairs topics such as health, nutrition, environment, home, religion, and travel (p. 294). Thus, concluding that there is “a discrepancy between editors' and readers' ratings of the same subject” (Bogart, 1989, p. 312). However, a growth in competition during recent decades, accelerated by the emergence of 24-hour cable television in the 1980s and the web in the 1990s, has been eroding the market position of traditional players. In turn, this has led to a stronger impact of a commercial logic on editorial pursuits and more attention to the audience's interests. Issue-based, public affairs news can be more expensive to produce and less commercially appealing than its personality-based, nonpublic affairs counterpart (i.e., draws an audience less attractive to advertisers, see Hamilton, 2004). Thus, the trend toward the softening of news has been seen as an attempt by media firms to counter their market decline and narrow the perceived gap between what they produce and what their audience seems to want.1 As Baum and Jamison (2006) have pointed out regarding political communication studies—although it applies more broadly to research on hard and soft news—“scholars have focused more on the supply of political information than on the nature of citizens' demands” (p. 947, emphasis in the original). However, some studies have used surveys of consumers' preferences, print circulation, broadcast ratings, and online-traffic data to ascertain whether softer news helps match the preferences of consumers or actually drives them away (Baum, 2002; Hamilton, 2004; Patterson, 2000; Prior, 2002; Rosenstiel et al., 2007). Two positions have emerged from these studies. One camp maintains that consumers want “softer” news, therefore the softening trend should help narrow the gap. For instance, commenting on his findings on website usage, Hamilton (2004) suggested that “the relative demand expressed for information on the Web shows patterns familiar from other media. Soft news and ‘news you can use’ often generate more interest than hard news” (p. 194). The opposing side contends that interest in “harder” content is paramount among consumers, thus a supply of softened news has created a new gap that undermines the market position of media firms. Thus, Patterson (2003) argued that “if we are to employ a standard based on the news that most consumers want and that most outlets provide, we would not want to have soft news in mind. We should be thinking in terms of hard news” (p. 140). The attempt to determine whether the softening of news might have narrowed a gap between the news choices of journalists and consumers or created one where none existed has been further complicated by a lack of agreement among scholars about the nature of this trend. At least three views can be distinguished in the literature. First, to scholars like Hamilton (2004), the trend entails a transformation in what news is reported: “I … use the term ‘soft news' to refer to programs with low levels of public affairs information and ‘hard news' to refer to shows with high levels of public affairs information” (p. 15). Second, to researchers like Baum (2007), it is about how the news is covered instead of what its subject is, so even public affairs stories can be softened: “Soft news outlets emphasize different aspects of political issues than do their hard news counterparts. In war, for instance … where hard news focuses on geopolitics, soft news focuses on body bags” (p. 107). Finally, to analysts like Gans (2003), the trend involves a change in the mix of what and how: The change in the hard/soft news ratio has taken on several forms … [It] can be seen clearly by the expansion of the “back of the book” sections … about subjects as varied as science and gardening … Another version of this change has come in program formats … For example, the historic one-hour or half-hour news documentary series … which was often devoted to investigative reporting about a single subject has been replaced by the semiweekly …“news magazines” that supply a mix of hard and soft news. (p. 29) The literature reviewed so far has advanced our understanding of the trend toward the softening of news and its consequences in terms of a possible gap between the choices of journalists and consumers. But it also has suffered from at least three limitations: (a) a relative mismatch between consumption data aggregated at the outlet or program level and consumer behavior that is increasingly disaggregated at the story level, (b) a lack of precision in the definition of the object of study, and (c) the negative effects of confounding factors in ascertaining the news choices made by journalists and consumers. First, in an increasingly competitive market where consumers have a growing array of unbundling techniques at their disposal—from remote controls to Really Simple Syndication (RSS) feeds—surveys of preferences or ratings and circulation data at the outlet or program level are too aggregate to assess the effectiveness of news supply on consumption choices often made at the story level.2Bogart (1989), who undertook one of the most sweeping attempts to “tackle the critical question of what content attracts [newspaper] readers” (p. 7), employed various methods but noted the difficulties of interpreting the resulting findings. Second, the lack of agreement about what the trend toward soft news consists of has made it difficult to have a precise characterization of the phenomenon. Is it about the subject of the stories, how they are covered or both? As different studies have defined the trend differently, this has also complicated a comparison of findings across them and subsequent theory development efforts. Third, in addition to issues of newsworthiness which derive from professional judgment, the choices that journalists make have also been increasingly influenced by value signals conveyed by consumers in their demand patterns. Today's journalists are more inclined to write for their audiences than Darnton and his colleagues did decades ago.3 By the same token, research on traditional (Graber, 1988) and new media (Sundar, Knobloch-Westerwick, & Hastall, 2007) has shown that consumers' choices are partly influenced by value signals embedded in the supply of stories. For instance, all else being equal, an item displayed on a newspaper's front page has a higher probability of being noticed and consumed than an item buried inside one of its sections. Situated within this literature on the softening of news, the goal of this article is to analyze the top news provided by journalists in four websites on the hard–soft dimension, and to determine whether there is a gap—in terms of subject and format—between these stories and the stories most often chosen by consumers on these respective sites. Moreover, the research approach adopted here aims to help solve, at least in part, the limitations depicted above. Methodology Sample Data were collected from four English language online news sites: chicagotribune.com (Chicago), seattlepi.newsource.com (Seattle), cnn.com (Cable News Network, CNN), and news.yahoo.com (Yahoo). Chicago and Seattle are two of the most visited websites of print newspapers in the United States, with almost 3 and 3.3 million unique users for November 2006, respectively (Newspaper Association of America, 2006).4 These sites are the online counterparts of the Chicago Tribune and the Seattle Post-Intelligencer, with an average daily circulation of about 567,000 and 128,000, respectively, for the period ending in September 2006 (Audit Bureau of Circulations, 2007).5CNN is the third most visited online news site, with a 2006 monthly average of almost 25 million monthly unique users, according to Nielsen measurements (Project for Excellence in Journalism, 2007). It is the online counterpart of the CNN, which had a median prime-time audience of 710,000 viewers during 2006 (Project for Excellence in Journalism, 2007). Yahoo is the most visited online news site, with a 2006 monthly average of over 28 million unique users, according to Nielsen measurements (Project for Excellence in Journalism, 2007). It is affiliated with Yahoo!, which runs the second largest web search engine with over 20% of market share for July 2007 (Nielsen//NetRatings, 2007). These sites were also selected because they exhibit variance in terms of parent media company and geographic scope. Research has shown that the behavior of the parent media company can influence that of its online counterpart (Boczkowski, 2004; Chan-Olmstead & Ha, 2003; Paulussen, 2004; Singer, 2004), thus we chose cases that represent print (Chicago and Seattle), broadcast (CNN), and online (Yahoo) firms. Because scholars have shown that geographic matters variously affect media production (Baisnee & Marchetti, 2006; Hannerz, 2004; Kaniss, 1991; MacGregor, 1997; Picard, 2002), we selected two cases with a national–global reach—CNN and Yahoo—and two local-metropolitan cases—Chicago and Seattle. We collected data on 21 randomly selected days—three for each day of the week—between April and June of 2007, for a total of three composite weeks. On each coding day, research assistants collected data at 10 a.m., 4 p.m., and 10 p.m. Central Time, yielding 63 distinct data collection shifts—21 days × 3 times per day—for each site. At each shift for each site, we identified the top 10 stories selected by journalists and by consumers, respectively. We focused on these stories because, from the universe of possible stories, these are the ones that are deemed most relevant by journalists and also garner most attention from consumers. Thus, they represent a suitable approximation of each group's key news preferences. Online stories were defined as text-based packages that included a headline; a story might, but need not, have multimedia features—such as video or graphics—or links to related stories. In cases in which there was a main story accompanied by one or more substories, only the main story was included. These criteria applied to all stories, regardless of whether they were breaking news, commentary, blogs, and so on. The stories selected by journalists were the 10 most prominently displayed stories on the homepage; depending on the page's format, these were either listed in vertical order or were identified by counting from left to right and from the top down in a grid-like manner.6 Editors had a role in selecting the stories that appeared on their respective homepages of these four sites. CNN, Chicago, and Seattle are extensions of “traditional” news organizations and their online operations include a significant presence of editors. Yahoo also employs human editors (Glaser, 2005). The stories selected by consumers were identified by examining the 10 most clicked stories on each site—as indicated, variously, by sites' “most viewed,”“most read,” or “most popular” lists; this information was publicly available on each site during our study.7 Across the 63 data collection shifts, each site yielded 1,260 stories—630 journalist-selected top stories and 630 consumer-selected top stories—for a total of 5,040 stories. The unit of analysis was the story. The following variables were coded8: intrasite news choice, intrasite news choice overlap, intersite convergent news choice, intersite convergent news choice overlap, what, and how.9 Analysis We examined news choices by journalists and consumers at two levels of analysis: within site (intrasite) and across sites (intersite). Within each site, we looked at 630 stories selected by journalists and 630 chosen by consumers. Across sites, we focused on stories that covered the same event and were selected by either journalists or consumers in more than one site in the same data collection shift. This analysis examined 539 such stories selected by journalists and 485 by consumers across the four sites. It enabled us to detect whether patterns of choice within one or more sites were also present across these sites. On the basis of the issues identified in Section “Theoretical Considerations,” to help tease out the potential influence of stories that might have been chosen by journalists because they were generating substantive site traffic among consumers, or by consumers because they were saliently displayed by journalists, we conducted two separate analyses at each level of analysis. The first analysis examined the top choices of journalists and consumers in each site and also the convergent choices of journalists and consumers across sites. This analysis provides a general view of the choices made by journalists and consumers. The second analysis looks at the stories exclusively chosen by either journalists or consumers in each site, but not by both. It also examines the convergent choices of journalists that were not shared by consumers and vice versa. That is, these are either the stories that journalists display prominently even though they have low levels of demand among consumers or stories with high levels of demand among consumers despite the fact that journalists do not display them prominently. Thus, this analysis affords a distilled picture of these news choices by focusing on the subset of stories that were selected by one group, but not by the other. Regarding the intrasite data, this second analysis examined 26% of the stories on CNN, 46% on Chicago, 60% on Seattle, and 67% on Yahoo. Concerning the intersite data, the second analysis looked at 58% of the convergent choices on the journalists' side that were not selected by consumers, and 53% of the convergent choices on the consumers' side that were not chosen by journalists. Results Patterns of choices within the sites We first examine choices regarding what subjects the stories are about and then move to analyze issues concerning how they are told. When all the stories are analyzed, in three of the four sites journalists select mostly nonpublic affairs news: 54% on CNN and 62% on Chicago and Seattle. Yahoo is the exception, but the proportion of this news is still high at 43% (Table 1). When only the stories with no overlap are considered, public affairs stories are the majority on CNN, Chicago, and Yahoo, but the share of nonpublic affairs news remains substantial on these three sites—36, 44, and 48%, respectively—and is still dominant on Seattle at 59%. Table 1 Intrasite Comparison of “What” of News Choices . Intrasite News Choice . Intrasite News Choice Overlap (Stories With No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 629 n = 165 n = 165  Public affairs 46 32 64 13  Nonpublic affairs 54 68 36 87 χ2 = 23.866, df = 1, p =.000 χ2 = 88.180, df = 1, p =.000 Chicago n = 630 n = 630 n = 291 n = 290  Public affairs 38 21 56 18  Nonpublic affairs 62 79 44 82 χ2 = 46.298, df = 1, p =.000 χ2 = 90.019, df = 1, p =.000 Seattle n = 630 n = 630 n = 381 n = 381  Public affairs 38 25 41 18  Nonpublic affairs 62 75 59 82 χ2 = 27.237, df = 1, p =.000 χ2 = 47.984, df = 1, p =.000 Yahoo n = 630 n = 630 n = 424 n = 424  Public affairs 57 44 52 33  Nonpublic affairs 43 56 48 67 χ2 = 21.349, df = 1, p =.000 χ2 = 32.410, df = 1, p =.000 . Intrasite News Choice . Intrasite News Choice Overlap (Stories With No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 629 n = 165 n = 165  Public affairs 46 32 64 13  Nonpublic affairs 54 68 36 87 χ2 = 23.866, df = 1, p =.000 χ2 = 88.180, df = 1, p =.000 Chicago n = 630 n = 630 n = 291 n = 290  Public affairs 38 21 56 18  Nonpublic affairs 62 79 44 82 χ2 = 46.298, df = 1, p =.000 χ2 = 90.019, df = 1, p =.000 Seattle n = 630 n = 630 n = 381 n = 381  Public affairs 38 25 41 18  Nonpublic affairs 62 75 59 82 χ2 = 27.237, df = 1, p =.000 χ2 = 47.984, df = 1, p =.000 Yahoo n = 630 n = 630 n = 424 n = 424  Public affairs 57 44 52 33  Nonpublic affairs 43 56 48 67 χ2 = 21.349, df = 1, p =.000 χ2 = 32.410, df = 1, p =.000 Note: CNN = Cable News Network. Open in new tab Table 1 Intrasite Comparison of “What” of News Choices . Intrasite News Choice . Intrasite News Choice Overlap (Stories With No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 629 n = 165 n = 165  Public affairs 46 32 64 13  Nonpublic affairs 54 68 36 87 χ2 = 23.866, df = 1, p =.000 χ2 = 88.180, df = 1, p =.000 Chicago n = 630 n = 630 n = 291 n = 290  Public affairs 38 21 56 18  Nonpublic affairs 62 79 44 82 χ2 = 46.298, df = 1, p =.000 χ2 = 90.019, df = 1, p =.000 Seattle n = 630 n = 630 n = 381 n = 381  Public affairs 38 25 41 18  Nonpublic affairs 62 75 59 82 χ2 = 27.237, df = 1, p =.000 χ2 = 47.984, df = 1, p =.000 Yahoo n = 630 n = 630 n = 424 n = 424  Public affairs 57 44 52 33  Nonpublic affairs 43 56 48 67 χ2 = 21.349, df = 1, p =.000 χ2 = 32.410, df = 1, p =.000 . Intrasite News Choice . Intrasite News Choice Overlap (Stories With No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 629 n = 165 n = 165  Public affairs 46 32 64 13  Nonpublic affairs 54 68 36 87 χ2 = 23.866, df = 1, p =.000 χ2 = 88.180, df = 1, p =.000 Chicago n = 630 n = 630 n = 291 n = 290  Public affairs 38 21 56 18  Nonpublic affairs 62 79 44 82 χ2 = 46.298, df = 1, p =.000 χ2 = 90.019, df = 1, p =.000 Seattle n = 630 n = 630 n = 381 n = 381  Public affairs 38 25 41 18  Nonpublic affairs 62 75 59 82 χ2 = 27.237, df = 1, p =.000 χ2 = 47.984, df = 1, p =.000 Yahoo n = 630 n = 630 n = 424 n = 424  Public affairs 57 44 52 33  Nonpublic affairs 43 56 48 67 χ2 = 21.349, df = 1, p =.000 χ2 = 32.410, df = 1, p =.000 Note: CNN = Cable News Network. Open in new tab A comparison of the choices made by journalists and consumers shows that when all stories are included in the analysis, on each one of the sites, journalists chose more public affairs stories than consumers (Table 1). The difference between the choices of journalists and consumers is 13 percentage points on Seattle and Yahoo, 14 percentage points on CNN, and 17 percentage points on Chicago. Chi-square tests of homogeneity show that the distribution of public affairs and nonpublic affairs stories is not the same for journalists and consumers (χ2 = 21.349–46.298, all significant at p =.000). When we look at the stories chosen exclusively by either journalists or consumers—excluding stories with overlap within the respective sites—differences between journalists' and consumers' share of public affairs stories increased on all sites: from 14 to 51 percentage points on CNN, from 17 to 38 percentage points on Chicago, from 13 to 23 percentage points on Seattle, and from 13 to 19 percentage points on Yahoo (Table 1). This widening of the gap between the choices of journalists and consumers is accompanied by increased χ2 values and remains significant (p =.000) on each one of the sites. On CNN, Chicago, and Seattle, this widening arises from increases in both journalists' share of public affairs stories and consumers' share of nonpublic affairs news. In Yahoo, the proportion of public affairs stories decreases for both journalists and consumers, with a more pronounced drop for the latter. Furthermore, looking across the sites, as the number of stories chosen exclusively by journalists and consumers decreased, the gap between them increased. For example, only 165 stories were chosen exclusively by journalists on CNN compared with 424 on Yahoo, and the growth in their respective gaps with consumer choices was 37 and 6 percentage points. In terms of how the stories are told, an analysis of the journalists' choices, either including all stories or only those without intrasite overlap, shows that straight news is a strong dominant option on all the sites. It ranges from 64 to 83% in the first approach and from 63 to 75% in the second (Table 2). When all the stories are analyzed in comparison with the choices of consumers, results show Chicago and Seattle journalists chose significantly more straight news than consumers: 77% versus 64% on Chicago and 64% versus 58% on Seattle (p =.000 and.034, respectively; Table 2). There are no significant differences on CNN and Yahoo. When only the stories without overlap are analyzed, the differences between journalists and consumers in the share of straight news on Chicago and Seattle rose from 13 to 30 percentage points and from 6 to 11 percentage points, respectively. The decrease in straight news among consumers was tied to an increase in commentary on Chicago and Seattle—from 15 to 29% and from 11 to 16%, respectively.10 Table 2 Intrasite Comparison of “How” of News Choices . Intrasite News Choice . Intrasite News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 630 n = 165 n = 165  Straight news 83 80 75 65  Feature style 14 17 17 30  Commentary 4 2 7 3  Alternative 0 1 1 2 χ2 = 1.536, df = 1, n.s.a χ2 = 3.673, df = 1, n.s.a Chicago n = 630 n = 630 n = 291 n = 290  Straight news 77 64 75 45  Feature style 18 20 20 24  Commentary 4 15 5 29  Alternative 0 1 0 1 χ2 = 28.116, df = 1, p =.000a χ2 = 52.260, df = 1, p =.000a Seattle n = 630 n = 630 n = 381 n = 381  Straight news 64 58 63 52  Feature style 26 29 25 30  Commentary 7 11 9 16  Alternative 2 2 3 2 χ2 = 8.652, df = 3, p =.034 χ2 = 12.727, df = 3, p =.005 Yahoo n = 630 n = 630 n = 424 n = 424  Straight news 75 78 68 73  Feature style 23 18 29 22  Commentary 1 2 2 2  Alternative 1 2 2 3 χ2 = 4.548, df = 3, n.s. χ2 = 5.449, df = 3, n.s. . Intrasite News Choice . Intrasite News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 630 n = 165 n = 165  Straight news 83 80 75 65  Feature style 14 17 17 30  Commentary 4 2 7 3  Alternative 0 1 1 2 χ2 = 1.536, df = 1, n.s.a χ2 = 3.673, df = 1, n.s.a Chicago n = 630 n = 630 n = 291 n = 290  Straight news 77 64 75 45  Feature style 18 20 20 24  Commentary 4 15 5 29  Alternative 0 1 0 1 χ2 = 28.116, df = 1, p =.000a χ2 = 52.260, df = 1, p =.000a Seattle n = 630 n = 630 n = 381 n = 381  Straight news 64 58 63 52  Feature style 26 29 25 30  Commentary 7 11 9 16  Alternative 2 2 3 2 χ2 = 8.652, df = 3, p =.034 χ2 = 12.727, df = 3, p =.005 Yahoo n = 630 n = 630 n = 424 n = 424  Straight news 75 78 68 73  Feature style 23 18 29 22  Commentary 1 2 2 2  Alternative 1 2 2 3 χ2 = 4.548, df = 3, n.s. χ2 = 5.449, df = 3, n.s. Note: CNN = cable news network. a To apply the χ2 test and eliminate expected cell frequencies <5; categories were collapsed to straight news and nonstraight news. Open in new tab Table 2 Intrasite Comparison of “How” of News Choices . Intrasite News Choice . Intrasite News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 630 n = 165 n = 165  Straight news 83 80 75 65  Feature style 14 17 17 30  Commentary 4 2 7 3  Alternative 0 1 1 2 χ2 = 1.536, df = 1, n.s.a χ2 = 3.673, df = 1, n.s.a Chicago n = 630 n = 630 n = 291 n = 290  Straight news 77 64 75 45  Feature style 18 20 20 24  Commentary 4 15 5 29  Alternative 0 1 0 1 χ2 = 28.116, df = 1, p =.000a χ2 = 52.260, df = 1, p =.000a Seattle n = 630 n = 630 n = 381 n = 381  Straight news 64 58 63 52  Feature style 26 29 25 30  Commentary 7 11 9 16  Alternative 2 2 3 2 χ2 = 8.652, df = 3, p =.034 χ2 = 12.727, df = 3, p =.005 Yahoo n = 630 n = 630 n = 424 n = 424  Straight news 75 78 68 73  Feature style 23 18 29 22  Commentary 1 2 2 2  Alternative 1 2 2 3 χ2 = 4.548, df = 3, n.s. χ2 = 5.449, df = 3, n.s. . Intrasite News Choice . Intrasite News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . CNN n = 630 n = 630 n = 165 n = 165  Straight news 83 80 75 65  Feature style 14 17 17 30  Commentary 4 2 7 3  Alternative 0 1 1 2 χ2 = 1.536, df = 1, n.s.a χ2 = 3.673, df = 1, n.s.a Chicago n = 630 n = 630 n = 291 n = 290  Straight news 77 64 75 45  Feature style 18 20 20 24  Commentary 4 15 5 29  Alternative 0 1 0 1 χ2 = 28.116, df = 1, p =.000a χ2 = 52.260, df = 1, p =.000a Seattle n = 630 n = 630 n = 381 n = 381  Straight news 64 58 63 52  Feature style 26 29 25 30  Commentary 7 11 9 16  Alternative 2 2 3 2 χ2 = 8.652, df = 3, p =.034 χ2 = 12.727, df = 3, p =.005 Yahoo n = 630 n = 630 n = 424 n = 424  Straight news 75 78 68 73  Feature style 23 18 29 22  Commentary 1 2 2 2  Alternative 1 2 2 3 χ2 = 4.548, df = 3, n.s. χ2 = 5.449, df = 3, n.s. Note: CNN = cable news network. a To apply the χ2 test and eliminate expected cell frequencies <5; categories were collapsed to straight news and nonstraight news. Open in new tab Patterns of choices across the sites Public affairs stories occupied the majority position of convergent news choices among journalists. But nonpublic affairs news still represented an important share of the top news stories selected by journalists, both when all stories were analyzed and when those with overlap were excluded—44 and 37%, respectively (Table 3). The picture is different regarding how the news gets told: straight news overshadowed the other options in stories with and without overlap with an 83 and 82% share, respectively. Table 3 Intersite Comparison of “What” and “How” of News Choices . Intersite Convergent News Choice . Intersite Convergent News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . . n = 539 . n = 485 . n = 311 . n = 255 . What  Public affairs 56 35 63 35  Nonpublic affairs 44 65 37 65 χ2 = 42.622, df = 1, p =.000 χ2 = 43.312, df = 1, p =.000 How  Straight news 83 80 82 78  Feature style 14 16 16 16  Commentary 2 4 2 5  Alternative 0 0 0 0 χ2 = 2.097, df = 1, n.s.a χ2 = 1.350, df = 1, n.s.a . Intersite Convergent News Choice . Intersite Convergent News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . . n = 539 . n = 485 . n = 311 . n = 255 . What  Public affairs 56 35 63 35  Nonpublic affairs 44 65 37 65 χ2 = 42.622, df = 1, p =.000 χ2 = 43.312, df = 1, p =.000 How  Straight news 83 80 82 78  Feature style 14 16 16 16  Commentary 2 4 2 5  Alternative 0 0 0 0 χ2 = 2.097, df = 1, n.s.a χ2 = 1.350, df = 1, n.s.a aTo apply the χ2 test and eliminate expected cell frequencies <5; categories were collapsed to straight news and nonstraight news. Open in new tab Table 3 Intersite Comparison of “What” and “How” of News Choices . Intersite Convergent News Choice . Intersite Convergent News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . . n = 539 . n = 485 . n = 311 . n = 255 . What  Public affairs 56 35 63 35  Nonpublic affairs 44 65 37 65 χ2 = 42.622, df = 1, p =.000 χ2 = 43.312, df = 1, p =.000 How  Straight news 83 80 82 78  Feature style 14 16 16 16  Commentary 2 4 2 5  Alternative 0 0 0 0 χ2 = 2.097, df = 1, n.s.a χ2 = 1.350, df = 1, n.s.a . Intersite Convergent News Choice . Intersite Convergent News Choice Overlap (Stories with No Overlap) . . Journalists (%) . Consumers (%) . Journalists (%) . Consumers (%) . . n = 539 . n = 485 . n = 311 . n = 255 . What  Public affairs 56 35 63 35  Nonpublic affairs 44 65 37 65 χ2 = 42.622, df = 1, p =.000 χ2 = 43.312, df = 1, p =.000 How  Straight news 83 80 82 78  Feature style 14 16 16 16  Commentary 2 4 2 5  Alternative 0 0 0 0 χ2 = 2.097, df = 1, n.s.a χ2 = 1.350, df = 1, n.s.a aTo apply the χ2 test and eliminate expected cell frequencies <5; categories were collapsed to straight news and nonstraight news. Open in new tab Regarding the subject of the stories, an analysis of all the stories shows that there is a significant 21-percentage-point difference between the choices made by journalists and consumers (p =.000). The choices of the former converge more on public affairs news and those of the latter more on nonpublic affairs stories (Table 3). After removing the stories with intersite overlap, the gap between the convergent choices of journalists and consumers increases from 21 to 28 percentage points. This gain results from changes on the journalists' side because overlapped stories were concerned mostly with nonpublic affairs subjects. Thus, removing them from the analysis altered the thematic distribution of journalists' convergent choices, but not that of consumers. Concerning how news are reported, there were no significant differences between stories that converged on the journalists' side and those that converged on the consumers' side, either before or after removing stories chosen by both. Discussion This study has aimed to ascertain the character of the top story choices made by journalists of online news sites on the hard–soft dimension and to determine whether there is a gap between these choices and the top choices made by the consumers of these sites. The analysis has shown that (a) journalists' choices on these sites are substantively “soft” in terms of what the stories are about, but not in terms of how they are told and (b) there is a gap between these choices and consumers' choices, that is, uniform and relatively strong in terms of the subject of the stories but variable and comparatively weaker in terms of how these stories are told. There is no established threshold after which the proportion of a particular type of content or format allows the analyst to characterize a news supply as “soft,” but we relied on two indicators to suggest that most top journalistic choices on these sites were marked by soft subject matters and hard story telling techniques. First, nonpublic affairs stories made up the majority on three of the four sites in the intrasite analysis of all stories, and reached a 55% combined average among all sites. After removal of the stories with overlap, these stories still constituted the majority on one site and a sizable minority on the remaining three, and reached a 47% combined average among all sites. In addition, the share of nonpublic affairs stories in the convergent choices of journalists was a minority, but still high considering that the focus was on the top news of the day. Second, the preeminence of content (i.e., what) over format (i.e., how) while measuring the extent of a “soft” news supply also becomes evident if we consider that in each case—both intrasite and intersite and before and after removal of overlapped stories—the highest share of public affairs subjects was lower than the lowest share of straight news formats. An examination of the contrast between journalists' and consumers' choices shows a sizable and uniform gap in terms of what stories are selected—which is consistent with studies that have shown reduced interest in public affairs among online news consumers (Althaus & Tewksbury, 2002; Hamilton, 2004; Schoenbach, de Waal, & Lauf, 2005; Tewksbury, 2003; Tewksbury & Althaus, 2000)—and a more variable situation in the case of how stories are told. In the analysis of stories' subject matter, three trends that illuminated the choice gap emerged after the removal of overlapped stories. First, in the intrasite analysis, on three of the four sites, journalists' choices moved in the direction of public affairs stories, and, on all sites, consumers' choices did so toward nonpublic affairs ones. Second and in relation to intrasite data, the smaller a postoverlapped subsample became, the larger the gap became. This suggests that perhaps when journalists have fewer stories to choose regardless of consumers' preferences they lean strongly toward public affairs news. Third, in the intersite analysis, the postoverlapped subsample of journalists' convergent choices moved in the direction of more public affairs stories, whereas the consumers' choices remained unchanged. This suggests that journalists' penchant for public affairs news might not cross over to consumers and that consumers' preference of nonpublic affairs news might be less dependent on journalists' choices. Regarding how stories are told, there was a significant gap on two sites and none in the intersite comparison. This signals greater variability in this dimension than in the case of the subject matter of stories. Differences between the what and how variables emerge not only when contrasting the uniformity of the former with the variability of the latter, but also when assessing their relative strengths. In all cases in which both what and how gaps were significant on a single site, the size of the gap in the first case was larger than that in the second. Additional analyses suggest that only a fraction of the stories placed at the top of the homepage by journalists stay on a given site throughout the day (i.e., appear in all three data collection times). The small amount of stories that stay on journalists' top 10 list longer are slightly more likely to be nonpublic affairs stories, but there is no evidence that these stories are more popular with consumers than public affairs news. Moreover, analyses of the interactions between the what and how variables revealed no significant evidence that consumers' choices mixed public affairs subject matters with nonstraight news formats. Concerning issues of geographic scope, the analysis yielded no major variance in terms of the stories' subject matter. The two global–national sites were at the two extremes of the gap continuum. However, there were differences in how the stories were told. The two local-metropolitan sites exhibited significant gaps in the how variable, partly driven by a lower prevalence of straight news chosen by consumers. Concerning issues of the type of media company, this study showed that Yahoo, a news aggregator, featured the highest share of public affairs news in the sample of all stories, the smallest gap in the what variable, and a strong hold on straight news. The highest share of public affairs news on Yahoo than on the other sites could result, at least in part, from the higher proportion of wire service copy—often more focused on public than on nonpublic affairs topics. The analytical approach adopted contributed greatly to the inquiry. First, using the story as the unit of analysis permitted a fine-grained depiction of the character of journalists' and consumers' choices. Second, discriminating between the subject matter and form of news has allowed us to address the multiple definitions of what soft news means and help settle empirical debates regarding its consequences for the gap between journalists and consumers. At least for the cases studied, a soft news supply and the gap with consumption patterns are strongly marked by the subjects covered in the news. Issues of story telling format are either absent or take a back seat. Third, that patterns existing when all stories were included in the analysis became magnified after removing stories with overlap underscores the heuristic power of this strategy to begin unpacking the influences of journalists' and consumers' choices on each other. As this study examines online news, it is important to consider the implications of two medium-specific factors for generalizability issues. First, although some initial studies have indicated a higher prevalence of public affairs stories in the consumption of traditional over online news, the more recent research has shown that topical preferences are stable across media (D’Haenens, Jankowski, & Heuvelman, 2004; Dutta-Bergman, 2004; Nguyen & Western, 2006; Pew Research Center for the People and the Press, 2007; Tewksbury, 2003, 2005). Second, there has been disagreement about whether the information about popularity levels of news stories on online sites affects consumption patterns and, if so, how important this effect is (Knobloch-Westerwick, Sharma, Hansen, & Alter, 2005; Sundar & Nass, 2001; Thorson, 2008). If such an effect existed and was strong, it could contribute to making the choice gap greater in online sites than in traditional media.11 It is worth considering four limitations of the research design. First, the researchers did not have access to their respective measurement processes utilized for calculating the most popular stories in the sites studied. If these processes differed markedly, it might have affected the data in a way that could have altered the results. But another interpretation is also feasible. If there was major variance in these processes that intra- and intersite patterns moved in similar directions could further strengthen the robustness of the findings. Second, headlines and leads on a homepage convey a better sense of the subject matter of a story than of its format. Although this discrepancy might have also affected the results, findings from a parallel ethnographic study of online news consumption conducted by the first author have also shown a preference for a straight news style (Boczkowski, 2010). Nevertheless, future studies should distinguish stories with various levels of information on the homepage about their story-telling techniques. Third, although the sample had variance in two important dimensions, it is comprised of only four sites which had a reputation for serious journalism and were based in the same country. Future research should examine sites of varying reputations and in different nations. Fourth, this study privileged order of appearance over other dimensions of story importance, such as use of visuals or headline size. Future scholarship on online news choices should try to take into account these dimensions. This study raises implications for understanding at least three sets of developments at the intersection between the choice gap and changes in the media industry and its broader social functions: namely, the potential consequences of an increasing supply of soft content for the news agenda, the growth of consumers' role as producers of widely circulated news and information, and the future of the media's watchdog function in liberal democracies. As for the evolution of large companies, such as those examined here, the growth in market competition, the increased attention paid to consumers' interests, and greater predisposition to cater to them than when the industry was dominated by a few organizations with monopoly or oligopoly power, provide a reasonable basis to believe that the supply of news might move toward decreasing the share of public affairs stories in an attempt to close the choice gap.12 These phenomena transformations might, in turn, call for conceptual changes. For instance, agenda-setting theory was premised on the idea that the media have a significant degree of autonomy in selecting what “are the most important topics of the moment” (McCombs, 2004, p. 20). However, if consumer behavior became a central factor shaping journalistic story selection, this premise would be challenged. This is consistent with Chaffee and Metzger's (2001, p. 375) suggestion that “the key problem for agenda-setting theory will change from what issues the media tell people to think about to what issues people tell the media they want to think about.” Scholars have recently theorized on the beneficial possibilities of consumers' growing involvement with the production of news (Benkler, 2006; Gillmor, 2004; Jenkins, 2006; Sunstein, 2006). Perhaps, nowhere has this idea been more forcefully argued than in Benkler's vision of a “networked public sphere,” which “allows all citizens to … no longer … be consumers and passive spectators. They can become creators and primary subjects. It is in this sense that the Internet democratizes” (Benkler, 2006, p. 272). However, the pervasiveness and strength of consumers' preferences for nonpublic affairs stories revealed in this study belie notions about a well-informed citizenry that often accompany this kind of argument. They also raise concerns about what might happen if consumers' choices shaped a large share of the media's content.13 Finally, one way in which media organizations contribute to the quality of democratic life is by playing a watchdog role over other powerful actors in society (Baker, 2002; Glasser & Ettema, 1998; Lazarsfeld & Merton, 1948; Waisbord, 2000). They do so mostly through public affairs stories, and their ability to be effective is dependent on spending considerable resources to gather information about wrongdoing. The prevalence of soft news in terms of subject matter, the possibility that this is a trend that might intensify in the direction of closing the gap with dominant consumption patterns, and the potential of strong consumer involvement in the production of news raise questions about the ability of media companies to play a strong watchdog role in the future. Taken together, these findings and trends point to potentially significant changes in the capacity of the media to positively affect the balance of power in society. Acknowledgments Research for this article has been supported by funding from the John S. and James L. Knight Foundation as part of the Carnegie-Knight Initiative on the Future of Journalism Education, the Innovation Fund from Northwestern University's School of Communication, and the Northwestern University's Media Management Center. We are grateful for the valuable contributions received from our research assistants: Dan Ellman, Elena Fields, Hilary Gifford, and Matt Weir. We thank Jim Ettema, Rich Gordon, Eszter Hargittai, Jay Hamilton, Bob Hariman, Eugenia Mitchelstein, Russ Neuman, Dan O’Keefe, Joe Turow, Jim Webster, and participants at the 2008 annual conference of the International Communication Association and the 2008 conference on the Future of Journalism at the Joan Shorenstein Center for Press, Politics, and Public Policy at Harvard University for helpful comments and advice on previous drafts. Notes 1 " This is not the only strategy pursued by news organizations to deal with a more competitive market and a perception of decrease market share. Other strategies include various forms of product differentiation, as it has been argued in scholarship on media fragmentation (Gandy, 2001; Jamieson & Cappella, 2008; Klinenberg, 2007; Neuman, 2001; Turow, 2005, 2006). However, production differentiation strategies need not be mutually exclusive with an overall softening of the news content—for instance, organizations have often softened their products in multiple, coexistent ways. 2 " We partly build on Tewksbury's (2003) strategy of using Nielsen//NetRatings page view consumption data, but we also extend it. Although his strategy moved toward analyzing stories, it did not examine them directly, but inferred their content from the information provided in their respective URLs—which also limited his sample selection. 3 " This predisposition seems to have intensified with the availability of data on sites' traffic in their respective newsrooms—including, for instance, at Yahoo News (Rosmarin, 2007). 4 " The Newspaper Association of America data combine the Seattle Times and Seattle Post-Intelligencer figures because the two companies have a joint operating agreement. Their web operations, however, are independent. 5 " The print edition of the Seattle Post-Intelligencer folded in spring 2009. 6 " Research on traditional media—for instance, Graber (1988)—has shown that visual cues such as headline size and location can influence consumption patterns. Because variance in homepage design of the sites studied for this project made comparison of such cues difficult, a decision was made to prioritize order of appearance in the operationalization of the top 10 journalists stories. 7 " During the period of study, in the four sites we studied, these stories were the most frequently clicked stories. These lists were updated every hour on Chicago, Seattle, and Yahoo, and every 20 minutes on CNN. Furthermore, whereas some news sites made available information on other measures of consumer choices, such as “most e-mailed” and “most recommended,” we focused on the most clicked stories because they seemed the most straightforward expression of consumers' preferences. 8 " Four trained research assistants coded the 5,040 stories. Intercoder agreement was assessed a subset of 5% of the data. Regular intercoder agreement levels averaged 94% and ranged from 88 to 100%. Cohen's κ intercoder agreement levels averaged 88% and ranged from 77 to 100%. Doubts that arose were subsequently consulted with the authors and resolved consensually. 9 " The variables and categories are defined as follows:
1. Intrasite news choice. There were two categories:
 (a) journalists: the top 10 stories that appeared on the homepage of each site in each shift and
 (b) consumers: the top 10 stories in the “most viewed,”“most read,” or “most popular” lists of each site in each shift.
 2. Intrasite news choice overlap. There were three categories:
 (a) overlap: stories that appear in both categories (a) and (b) of variable 1 in the same site during the same data collection shift,
 (b) no overlap—journalists: stories that appear in category (a) variable 1 but not in category (a) of variable 2, in the same site during the same shift, and
 (c) no overlap—consumers: stories that appear in category (b) variable 1 but not in category (a) of variable 2, in the same site during the same data collection shift.
 3. Intersite convergent news choice. There were three categories:
 (a) journalists: stories about the same event that are included in category (a) of variable 1 in more than one site in a given data collection shift,
 (b) consumers: stories about the same event that are included in category (b) of variable 1 in more than one site in a given data collection shift, and
 (c) no convergent choice: stories that do not fall under categories (a) or (b).
 4. Intersite convergent news choice overlap. Only stories that displayed convergence among either journalists or consumers—that is, only stories classified as (a) or (b) in variable 3—were coded for this variable. There were three categories:
 (a) overlap: stories included in either category (a) or (b) of variable 3 that are about an event that is also reported in a story appearing in the opposite category of variable 1 in at least one of the convergent choice sites in the same shift,
 (b) no overlap—journalists: stories included in category (a) variable 3 that are about an event that is not reported in a story appearing in category (b) of variable 1 in any of the convergent choice sites in the same data collection shift, and
 (c) no overlap—consumers: stories included in category (b) variable 3 that are about an event that is not reported in a story appearing in category (a) of variable 1 in any of the convergent choice sites in the same data collection shift.
 5. What. The main topic addressed in this story. There are two categories:
 (a) Public affairs stories: deal with news about politics, government, economics, business, international affairs, and “war on terror.” This includes stories about the activity of government, elected officials, political candidates; the economy and business developments; and events, happening in other countries, about the state or international organizations.
 (b) Nonpublic affairs stories: addressing subjects such issues as sports, crime, entertainment, technology, and weather. This includes stories about sports teams and events; criminal activity at the local, national or international level; visual and performing arts, and literature; medical, scientific, and technological matters; damage caused by disasters and large-scale accidents; and routine and nonroutine weather information.
 6. How. The main story-telling technique. There are four categories:
 (a) straight news, which refers to a writing style that emphasizes the important facts and is often told in a nonpersonal, detached way. Often includes information about who, what, where, and when up at the top—sometimes after an anecdote;
 (b) feature style, alluding to stories that employ narrative or other literary devices and can include a more personal tone. It often has a beginning–middle–end narrative structure;
 (c) commentary, which includes all stories with a pronounced point of view about a topic or a product, such as columns, “op-eds,” and reviews; and
 (d) alternative formats, consisting of all other formats not mentioned above. 10 " Additional analyses show that most of these commentary stories are about movies on Chicago and about sports on Seattle. 11 " In addition, it should be noted that this research does not ascertain whether consumers utilize online news as a primary or secondary—that is, supplementary to the consumption of traditional media—source of information. Future research would benefit from an examination of this matter by complementing the content analysis data with methods such as surveys or interviews. 12 " Because this gap is only one factor that shapes the behavior of news organizations, this hypothetical scenario is contingent on an array of additional possible factors, including shifts in technology, ownership structure, regulatory environment, and public culture. 13 " One alternative is that consumers of these sites studied chose a large proportion of nonpublic affairs stories partly because these sites already furnished a relatively large share of public affairs stories. Thus their behavior might have been different if their choices had been made in the absence of a substantive dose of news about public affairs. Future research should elucidate the character of consumers' news choices in different circumstances of consumer-authored media. References Althaus , S. L. , & Tewksbury , D. ( 2002 ). Agenda setting and the “new” news: Patterns of issue importance among readers of the paper and online versions of the New York Times . Communication Research , 29 , 180 – 207 . Google Scholar Crossref Search ADS WorldCat Audit Bureau of Circulations. ( 2007 ). ABC U.S. Newspaper FAS-FAX Report, Arlington Heights, IL. Baisnee , O. , & Marchetti , D. ( 2006 ). The economy of just-in-time television newscasting: Journalistic production and professional excellence at Euronews . Ethnography , 7 , 99 – 123 . Google Scholar Crossref Search ADS WorldCat Baker , C. E . ( 2002 ). Media, markets and democracy . New York, NY : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Baum , M . ( 2002 ). Sex, lies, and war: How soft news brings foreign policy to the inattentive public . The American Political Science Review , 96 , 91 – 109 . Google Scholar Crossref Search ADS WorldCat Baum , M . ( 2003 ). Soft news goes to war: Public opinion and American foreign policy in the new media age . Princeton, NJ : Princeton University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Baum , M . ( 2007 ). Hard and soft news. In T. Schaefer & T. Birkland (Eds.), Encyclopedia of media and politics (pp. 106 – 107 ). Washington, DC : CQ Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Baum , M. , & Jamison , A. ( 2006 ). The Oprah effect: How soft news helps inattentive citizens vote consistently . The Journal of Politics , 68 , 946 – 959 . Google Scholar Crossref Search ADS WorldCat Benkler , Y . ( 2006 ). The wealth of networks: How social production transforms markets and freedom . New Haven, CT : Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bennett , W. L . ( 2003a ). News: The politics of illusion (5th ed.). New York, NY : Longman . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bennett , W. L . ( 2003b ). The burglar alarm that just keeps ringing: A response to Zaller . Political Communication , 20 , 131 – 138 . Google Scholar Crossref Search ADS WorldCat Boczkowski , P . ( 2004 ). Digitizing the news: Innovation in online newspapers . Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Boczkowski , P . ( 2010 ). News at Work: Imitation in an Age of Information Abundance . Chicago, IL : University of Chicago Press . Bogart , L . ( 1989 ). Press and public . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chaffee , S. H. , & Metzger , M. J. ( 2001 ). The end of mass communication? Mass Communication & Society , 4 , 365 – 379 . Google Scholar Crossref Search ADS WorldCat Chan-Olmsted , S.M. , & Ha , L. ( 2003 ). Internet business models for broadcasters: How television stations perceive and integrate the Internet . Journal of Broadcasting & Electronic Media , 47 ( 4 ), 597 – 617 . Google Scholar Crossref Search ADS WorldCat D’Haenens , L. , Jankowski , N., & Heuvelman , A. ( 2004 ). News in online and print newspapers: Differences in reader consumption and recall . New Media & Society , 6 , 363 – 382 . Google Scholar Crossref Search ADS WorldCat Darnton , R . ( 1975 ). Writing news and telling stories . Daedalus , 104 , 175 – 194 . OpenURL Placeholder Text WorldCat Delli Carpini , M. X. , & Williams , B. ( 2001 ). Let us infotain you: Politics in the new media environment. In W. L. Bennett & R. Entman (Eds.), Mediated politics (pp. 160 – 181 ). New York, NY : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dutta-Bergman , M. J . ( 2004 ). Complementarity in consumption of news types across traditional and new media . Journal of Broadcasting & Electronic Media , 48 ( 1 ), 41 – 60 . Google Scholar Crossref Search ADS WorldCat Fishman , M . ( 1980 ). Manufacturing the news . Austin, TX : University of Texas Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gandy , O . ( 2001 ). Dividing practices: Segmentation and targeting in the emerging public sphere. In W. L. Bennett & R. Entman (Eds.), Mediated politics (pp. 141 – 159 ). New York, NY : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gans , H . ( 2003 ). Democracy and the news . New York, NY : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gans , H . ( 2004 ). Deciding what's news: A study of CBS Evening News, NBC Nightly News, Newsweek, and Time (25th anniversary ed.). Evanston, IL : Northwestern University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gillmor , D . ( 2004 ). We the media: Grassroots journalism by the people, for the people . Sebastopol, CA : O’Reilly . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Glaser , M . (April 1, 2005 ). Inside Yahoo News: Aggregator brings RSS to the masses. OJR: The Online Journalism Review. http://www.ojr.org/ojr/stories/050331glaser/. Glasser , T. , & Ettema , J. ( 1998 ). Custodians of conscience . New York, NY : Columbia University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Graber , D . ( 1988 ). Processing the news: How people tame the information tide (2nd ed.). New York, NY : Longman . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hamilton , J . ( 2004 ). All the news that's fit to sell: How the market transforms information into news . Princeton, NJ : Princeton University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Hannerz , U . ( 2004 ). Foreign news: Exploring the world of foreign correspondents . Chicago, IL : University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jamieson , K. H. , & Cappella , J. ( 2008 ). Echo chamber: Rush Limbaugh and the conservative media establishment . New York, NY : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jenkins , H . ( 2006 ). Convergence culture: Where old and new media collide . New York : New York University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kalb , M . ( 1998 ). The rise of the “new news”: A case study of two root causes of the modern scandal coverage. Unpublished manuscript, Harvard University, Cambridge, MA. Kaniss , P . ( 1991 ). Making local news . Chicago, IL : University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Klinenberg , E . ( 2005 ). Convergence: News production in a digital age . The Annals of the American Academy of Political and Social Science , 597 , 48 – 64 . Google Scholar Crossref Search ADS WorldCat Klinenberg , E . ( 2007 ). Fighting for air: The battle to control America's media . New York, NY : Metropolitan Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Knobloch-Westerwick , S. , Sharma , N., Hansen , D. L., & Alter , S. ( 2005 ). Impact of popularity indications on readers' selective exposure to online news . Journal of Broadcasting & Electronic Media , 49 , 296 – 313 . Google Scholar Crossref Search ADS WorldCat Lazarsfeld , P. , & Merton , R. ( 1948 ). Mass communication, popular taste, and organized social action. In L. Bryson (Ed.), The communication of ideas (pp. 95 – 118 ). New York, NY : Harper . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC MacGregor , B . ( 1997 ). Live, direct and biased? Making television news in the satellite age . London, England : Arnold . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McCombs , M . ( 2004 ). Setting the agenda: The mass media and public opinion . Cambridge, England : Polity . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McManus , J . ( 1994 ). Market-driven journalism: Let the citizen beware? Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mindich , D . ( 2005 ). Tuned out: Why Americans under 40 don't follow the news . New York, NY : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Neuman , W. R . ( 2001 ). The impact of the new media. In W. L. Bennett & R. Entman (Eds.), Mediated politics (pp. 299 – 320 ). New York, NY : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Newspaper Association of America. ( 2006 ). NAA Newspaper Audience Database Report (Nadbase). Newspaper Association of America, Vienna, VA. Retrieved from http://www.marketingcharts.com/print/newspaper-sites-continue-to-expand-mediums-footprint-visits-near-all-time-high-2207/naa-nadbase-top-newspaper-websites-average-monthly-metrics-march-august-2007jpg. Nguyen , A. , & Western , M. ( 2006 ). The complementary relationship between the Internet and traditional mass media: The case of online news and information . Information Research - An International Electronic Journal , 11 ( 3 ). Retrieved from http://informationr.net/ir/11-3/paper259.html. OpenURL Placeholder Text WorldCat Nielsen//NetRatings. ( 2007 ). Nielsen/ NetRatings announces July U.S. search share rankings . New York Nielsen//NetRatings. Retrieved from http://www.nielsen-online.com/pr/pr_070820.pdf. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Patterson , T . ( 2000 ). Doing well and doing good: How soft news and critical journalism are shrinking the news audience and weakening democracy–and what news outlets can do about it. Unpublished manuscript, Harvard University, Cambridge, MA. Patterson , T . ( 2003 ). The search for a standard: Markets and media . Political Communication , 20 , 139 – 143 . Google Scholar Crossref Search ADS WorldCat Paulussen , S . ( 2004 ). Online news production in Flanders: How Flemish online journalists perceive and explore the Internet's potential . Journal of Computer-Mediated Communication , 9 ( 4 ). Retrieved from http://jcmc.indiana.edu/vol9/issue4/paulussen.html. OpenURL Placeholder Text WorldCat Pew Research Center for the People and the Press. ( 2007 ). Two decades of American news preferences, part 2: News interest across decades and “news eras.” Retrieved from http://pewresearch.org/pubs/574/two-decades-of-american-news-preferences. Picard , R . ( 2002 ). The economics and financing of media companies . New York, NY : Fordham University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Prior , M . ( 2002 ). Any good news in soft news? The impact of soft news preference on political knowledge . Political Communication , 20 , 149 – 171 . Google Scholar Crossref Search ADS WorldCat Prior , M . ( 2007 ). Post-broadcast democracy: How media choice increases inequality in political involvement and polarizes elections . Cambridge, England : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Project for Excellence in Journalism. ( 1998 ). Changing definitions of news, Pew Research Center, Washington, DC. Retrieved from http://www.journalism.org/node/442. Project for Excellence in Journalism. ( 2007 ). State of the news media 2007: An annual report on American journalism. Retrieved from http://www.stateofthenewsmedia.org/2007/narrative_cabletv_audience.asp?cat=2&media=6. Rosenstiel , T. , Just , M., Belt , T., Pertilla , A., Dean , W., & Chinni , D. ( 2007 ). We interrupt this newscast: How to improve local news and win ratings, too . New York, NY : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Rosmarin , R . ( 2007 ). The big news about Yahoo! Forbes.com. Retrieved from http://www.forbes.com/technology/2007/10/05/yahoo-news-moore-tech-internet-cx_rr_1005techyahoo.html. Schoenbach , K. , de Waal , E., & Lauf , E. ( 2005 ). Online and print newspapers: Their impact on the extent of the perceived public agenda . European Journal of Communication , 20 , 245 – 258 . Google Scholar Crossref Search ADS WorldCat Schudson , M . ( 2003 ). The sociology of news . New York, NY : W. W. Norton . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Scott , D. , & Gobetz , R. ( 1992 ). Hard news/soft news content of the national broadcast networks, 1972–1987 . Journalism Quarterly , 69 ( 2 ), 406 – 412 . Google Scholar Crossref Search ADS WorldCat Sigal , L . ( 1986 ). Sources make the news. In R. K. Manoff & M. Schudson (Eds.), Reading the news (pp. 9 – 37 ). New York, NY : Pantheon . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Singer , J . ( 2004 ). Strange bedfellows? The diffusion of convergence in four news organizations . Journalism Studies , 5 , 3 – 18 . Google Scholar Crossref Search ADS WorldCat Sundar , S. S. , Knobloch-Westerwick , S., & Hastall , M. ( 2007 ). News cues: Information scent and cognitive heuristics . Journal of the American Society for Information Science and Technology , 58 , 366 – 378 . Google Scholar Crossref Search ADS WorldCat Sundar , S. S. , & Nass , C. ( 2001 ). Conceptualizing sources in online news . Journal of Communication , 51 , 52 – 72 . Google Scholar Crossref Search ADS WorldCat Sunstein , C . ( 2006 ). Infotopia: How many minds produce knowledge . New York, NY : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tewksbury , D . ( 2003 ). What do Americans really want to know? Tracking the behavior of news readers on the Internet . Journal of Communication , 53 , 694 – 710 . Google Scholar Crossref Search ADS WorldCat Tewksbury , D . ( 2005 ). The seeds of audience fragmentation: Specialization in the use of online news sites . Journal of Broadcasting & Electronic Media , 49 ( 3 ), 332 – 348 . Google Scholar Crossref Search ADS WorldCat Tewksbury , D. , & Althaus , S. ( 2000 ). Differences in knowledge acquisition among readers of the paper and online version of a national newspaper . Journalism and Mass Communication Quarterly , 77 , 457 – 479 . Google Scholar Crossref Search ADS WorldCat Thorson , E . ( 2008 ). Changing patterns of news consumption and participation: News recommendation engines . Information, Communication & Society , 11 , 473 – 489 . Google Scholar Crossref Search ADS WorldCat Times Mirror Center. ( 1990 ). The age of indifference. Pew Research Center for the People & the Press. Washington, DC. Retrieved from http://people-press.org/1990/06/28/the-age-of-indifference/. Tuchman , G . ( 1979 ). Making news: A study in the construction of reality . New York, NY : The Free Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Turow , J . ( 2005 ). Audience construction and culture production: Marketing surveillance in the digital age . Annals of the American Academy of Political and Social Science , 597 , 103 – 121 . Google Scholar Crossref Search ADS WorldCat Turow , J . ( 2006 ). Niche envy: Marketing discrimination in the digital age . Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Underwood , D . ( 1993 ). When MBAs rule the newsroom: How marketers and managers are reshaping today's media . New York, NY : Columbia University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Waisbord , S . ( 2000 ). Watchdog journalism in South America: News, accountability, and democracy . New York, NY : Columbia University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zaller , J . ( 2003 ). A new standard of news quality: Burglar alarms for the monitorial citizen . Political Communication , 20 , 109 – 130 . Google Scholar Crossref Search ADS WorldCat © 2011 International Communication Association

Hoover, Stewart, M.;Coats, Curtis, D.

doi: 10.1111/j.1460-2466.2011.01583.xpmid: N/A

Abstract Findings from in-depth interviews with 19 Evangelical Protestant men regarding their ideas about masculinity, media, and religion are reported. The authors propose a critical test of a neoconservative view which laments a modern “crisis of masculinity” and which posits media as a villain in this crisis and religion, particularly conservative religion, as the savior. The study pursues a line of inquiry suggested by D. Gauntlett (2002) and S. M. Hoover (2006) with the hope of developing clearer ideas of how men make gendered narratives of self-understanding and presentation out of their media lives. These interviews contradicted the neoconservative claims, finding that media plays constructive roles in these men's senses of self in both masculine and religious terms. For most of the last decade, a quiet national debate has been building over men and masculinity. Some of this discourse has found its way into the public sphere, as in 2005 when Laura Bush announced a program to address the emerging gap between boys' and girls' educational attainment (NPR, 2005). Bush's project addressed one dimension of a perceived problem or even “crisis” of masculinity (Clare, 2000; NPR, 2007). This crisis is claimed to be rooted in decades-long trends where traditional roles and identities for men are thought to have come under stress (Archer, 2001; Clare, 2000). For many advocates of this view, there is a theme of “loss” in these relations, with men lamenting what they have lost in traditionally formed and sustained identity and life practice (NPR, 2007). For many such observers, the source of these shifting sands is feminism, both in the academy and in the broader culture. A set of arguments that we might call traditionalist or “essentialist” (Farrell, 2001; Macnamara, 2006; Mansfield, 2006) lament feminism as a project of cultural politics. Other voices have perceived an imbalance in these relations, but have seen it as an intellectual issue, arguing for the development of scholarly “men's studies” which could enter into dialogue with the more developed feminist traditions (Connell, 2003; Gavanas, 2004; Kimmel, 1995; Messner, 1997). Public perceptions and discourses have important implications for politics, power, and public policy, and much of the discourse of a “crisis” lies in such lay sensibilities. Over the course of the Bush Administration, a variety of voices expressed concerns well beyond boys' educational and life prospects, and there is evidence that this will not change with the Obama administration.1 A prominent force in this has been Dr. Don Eberly, who served in two key positions in the Bush Administration, first in the Office of Faith-Based Initiatives, and later as Acting Minister of Youth and Sport in Paul Bremer's interim Iraqi government (Berkowitz, 2005). Eberly's intellectual status in the neoconservative movement, though, stems from his own organization, The Civil Society Project, a think-tank devoted to the development of a neoconservative discourse on civil society and civic engagement.2 At the core of the conservative view is “virtue:” a focus on morality and moral culture. In the words of Gertrude Himmelfarb, “When we speak of the restoration of civil society it is a moral restoration we should seek” (Himmelfarb, 1995). For Eberly, fathers, masculinity, and manhood must be at the center of this. Lamenting the “… Republic of the Autonomous Self, where the individual is the only real sovereign, where ‘mediating’ structures have been leveled, and where rules proliferate, yet lack legitimacy,” Eberly identifies a collective need in the 21st century to build up the “good society” through revitalized civic consciousness and voluntary associations (Berkowitz, 2005). Eberly and other social critics identify the family as the central context through which the sentiments and habits of such civic vocation can be formed, though the family cannot go alone in their view. Voluntary associations are the primary medium through which civil society is realized, but they rely on the more fundamental unit of the family. Citing Edmund Burke's idea of families as “little platoons” and Toqueville's notion that families are the “first link” in a chain leading to civic engagement, Eberly argues that the family as a moral culture has suffered from the withdrawal of normative sources and influences in the culture, specifically “… parents, priests, and pedagogues …”. And the key parent in these regards, he contends, is the father. What he calls “father absence” is at the root of a variety of social ills, not least the withdrawal of families from their central moral role in crafting the good society. Eberly links his analysis to the broader discourse about “social capital” (cf. Putnam, 2001) in relation to a civil society in which religion plays a key role. Not-withstanding the reservations of some, religion is likely to have a stronger voice in the public square, both as a legitimate wellspring of personal values and as perhaps the richest source of renewed social capital in communities. (Eberly, 2002) Eberly argues that men and fathers have a specific responsibility to address issues of social capital in their own lives and in the lives of the next generation, including alienation and respect for/understanding of authority (Eberly, 1999a). Eberly is not specific about this, but feminism seems to lurk behind much of his critique. Lamenting the rise of identity politics, and decrying the idea that “… the personal is political …” he sees the specific problem of men to be the lack of social capital, traditional standpoints, and cultural anchors, leaving men without positive models. Two key elements of this argument are the roles of the media and of religion. Religion is seen as positive in the provision and support of what are seen as fundamental truths of male identity. Eberly refers to an imagined past where male identities were rooted in what he calls “God-centered masculinity”: Here, we find the man as he was meant to be, mirroring the true character of God. Here we find the paternal male who generates, not destroys life—the benevolent provider and defender, not the aggressor or predator. Here we discover the man who finds his strength and purpose as a father and friend, a protector and provider, and mentor and a moral example…” (1999b, p. 31) The media, in Eberly's view, play an unequivocally negative part in this drama. … many of the most corrupting viruses [in society] are now being borne along not by sinister politicians but by an entertainment and information media culture, and that this omnipresent culture is displacing the core social institutions that once shaped and molded the democratic citizen. (1999b, p. 9) To simplify this argument, civil society today depends on the cultivation of virtue and of the values of citizenship in the “little platoon” of the family. Fathers must play a particular role in this, but their role is under assault by a culture that has ceased to value the unique roles and values of maleness and masculinity. Social forces that surround the family can either support or undermine these values. Religion—as a moral space—plays a positive role, and the media a negative one. The present study We undertook a study of the role of media in relation to masculinity and the moral culture of the family through an interpretive, ethnographic study of media households. This paper reports on findings from in-depth interviews with 19 White, Evangelical Protestant men regarding their ideas about masculinity, media, and religion. The objective was to test some of the underlying assumptions of the neoconservative critique. Specifically, we looked at the following research questions: (a) What is masculine identity as articulated by these informants and what are its sources?, (b) What are their models and motivations of fatherhood, and what are their sources?, (c) What is the role of media?, and (d) What is the role of religion and/or spirituality in developing and supporting masculinity and fatherhood? Literature review The relevant literatures include: (a) feminist approaches to gender and representations of gender; (b) studies of those representations, particularly the persistence of masculinist symbols and values; and (c) media and identity. Contemporary work on men and masculinity looks beyond the concerns with gender stereotypes, which have dominated the field. As much of this work has been rooted in persuasive work in feminist analysis, it has been necessary to address commonalities and distinctions between scholarship on women and on men. Justad (2000) considered whether the two can be “balanced,” whether we can talk about one gender without talking about the other, and whether there remain issues specific to men and men's experience over and above those addressed through reference to the male “other.” He concluded that centered and focused work on men is overdue. In light of this, Connell (2003) called for the development of a “critical men's studies” to move beyond representations and “sex roles” to: (a) continued development of structuralist and poststructuralist approaches, (b) the recognition that there are “multiple masculinities,” and (c) ethnographic and interpretive work on the construction of masculinities in specific settings. As we noted earlier, some voices lament the role of feminism in setting the agenda in the area of gender relations. The scholarly literature in this area is divided in some significant ways. Some see the situation as a problem of lived experience (Farrell, 2001; Mansfield, 2006) rooted in the sense that there is an “essential” masculinity that has been denied by the influence of feminist critique and activism. Curiously, there is no clear sense in this literature of whether the “essential” nature of masculinity is biological or cultural in origin. A second important stream sees “masculinity” as a problem of cultural—and particularly media—representation. (Macnamara, 2006; Nathanson & Young, 2001). In the background is the issue of “hegemonic masculinity.” Broad literatures have addressed its form, nature, and constitution. These tend to agree that in common discourse, masculinity is still understood in “traditional” terms, though in a range of forms, from “hypermasculine” to “softer” valences. Connell's (2003) call for a “critical men's studies” would treat such ideas and symbols as resources, which men engage to “… make situationally specific choices from a cultural repertoire of masculine behaviour” (cf. Wetherell & Edley, 1999, p. 251). Hanke (1998) cautions, though, that the “idealized” masculine subjectivity that scholars might construct through such interpretations must still be understood to comprise “white, middle-class, heterosexual, professional-managerial men” (p. 186). The culturalist direction, which focuses on the construction of identities out of available repertoires of cultural resources—and which obviously draws on such important traditions as Butler's (1990) ideas about the ways genders are constituted—implies a role for the media, if indeed identity is understood as emerging when subjectivities take on symbolic resources available from a public inventory of possibilities (Gauntlett, 2002; Wetherell & Edley, 1999). Consalvo (2003), for example, persuasively argues that media scholars should look at how media “… construct masculinity in a particular time and place, and how these outlets differentiate between various forms of the masculine” (p. 29). Earlier traditions of media analysis stressed a more limited range of images and functions. As Craig (1992) notes, sex roles, male violence, and pornography were the most common topics (p. 4). Fejes (1992) argues that the original “sex role” perspective is now discredited, and that although media sex roles are stereotypes, there are important questions remaining about the implications of that fact (p. 9, 11). Most observers have agreed with Craig and with Hanke (1998), “… masculinity is what a culture expects of its men,” and that media images are representations at the center of that discourse (Craig, 1992, p. 3). A range of studies has addressed such representations. Examples include studies of the images of men and male gender in music (Denski & Sholle, 1992), in advertising (Barthel, 1992; Baudrillard, 1970), in prime-time television (Askew & Ross, 1990; Gunter, 1995), in war films and westerns (Donald, 1992), in children's television (Barcus, 1983), in news (Croteau & Hoynes, 1992), in sports (Sabo & Jansen, 1992), in youth cultures (Hearn & Malechi, 1992), and in film (Clark, 2002a; Kirkham & Thumin, 1995). These studies have generally found consistently “traditional” images of masculinity and manhood (Consalvo, 2003; Faludi, 2000). But, as Macnamara (2006) and Nathanson et al. (2003) argue, media representations of masculinity are at the same time in transition. Noting Craig's (1993) claim that “softer” and “reconstructed” male characters have begun to appear, Macnamara problematizes this change, arguing that such newer, more nontraditional male roles exist in relation to continuing traditionalism in role portrayals in such places as sports and situation comedies. These media, he observes, “… portray men as violent and aggressive thieves, murderers, wife and girlfriend beaters, sexual abusers, molesters, perverts, irresponsible, deadbeat dads and philanderers …” (p. 166). The larger question is whether media images merely represent masculinity or also constitute it. Saco (1992) notes that subjectivities—the constitutions of human subjects—emerge from particular and located discourses. She also argues that identities are composites of multiple subjectivities into which we are “hailed” (cf. Althusser) by culture. She concludes that through their interaction with emerging subjectivities, media do in fact construct gender, not just represent it (Saco, 1992, p. 25). Consalvo agrees, suggesting that media representations need to be understood in ideological terms. As she says, they are “… made to seem natural and inevitable, therefore becoming ideological and seeming ‘trans-historical’” (Consalvo, 2003, p. 30). The complexity of this question (“representation” vs. “constitution”) is addressed by Hanke (1992, 1998), who stresses that masculinity cannot be seen as a unitary, hegemonic force (and thus not supportive of any particular or focused definition of masculinity). Instead, there are alternative masculinities represented in the media. Even masculine narratives in media may lead to the destabilization of “masculinity” as a category “… which may allow male viewers to experience the ‘feminine’ pleasure that contradicts, if not deconstructs, the dominant ideology” (Hanke, 1998, p. 188). Consalvo (2003) observes, “… as with representations of women, there are multiple representations of men, and therefore multiple masculinities” (p. 29, see also Beynon, 2002). Connell is also concerned that the notion of masculinity not be seen as unitary or univocal. Connell (2000) reviews literatures, including the critical cultural work influenced by Foucault, and suggests that we think of gender “as … historically specific features of social life in modernity [that] constantly strikes difficulties around the division of labour and institutional practices” (pp. 21–22). Fiske (1997) stressed that media do produce both masculine and feminine subjects, and that gender overlaps with race, class, generation, and sexual preference; so the notion that there would be a unitary masculinity either represented or produced seems too simplistic. In light of such arguments, Hanke (1992) notes a particular problem with Cultural Studies' approach to the specific question of maleness and masculinity in these regards. Hanke notes that Fiske (and perhaps others in the field of Cultural Studies) believe that “… hegemonic ideology appears to have a great difficulty inserting itself into our everyday cultural experiences in a way that would define most people's commonsense understandings of the gender regime …. For Fiske, this is rooted in the idea that audiences can act with autonomy and produce subversive meanings from media texts” (p. 188). Hanke (1992) sees a larger issue here, though, wondering whether Cultural Studies expresses a certain “softness” on the issue of male domination, thereby making its understanding of the influence of ideologies of masculinity inadequate to the task of “… a critical cultural analysis of television and gender …” where issues of power and domination (central to commonplace understandings of masculinity) are central to the gender identities and relations presented (p. 197). Hanke argues that the complex and in some ways contradictory world of representations of masculinities in media has a coherence at its core, this “hegemonic masculinity.” Gauntlett (2002) provides an analysis that more directly addresses the ways that these mediated representations might relate to the way masculinity is understood, produced, and embodied. He argues that it is not a process flowing from media to effects. Taking on board Connell's call for a “critical approach,” Gauntlett turns to Giddens's (1991) notion of the reflexive project of the self, proposing “self-narrativising” as a method for capturing fluid and evolving identities around masculinity. There are several dimensions to this: (a) that the “late-modern” project of the self objectifies and routinizes ways of living and being that were previously less self-conscious; (b) “reflexivity,” or self-consciousness about the nature of social practice, with people conscious and calculating about their choices and actions; (c) an indirect role for the media which provides audiences with a view “behind the scenes” and autonomy over action, as is represented in media narratives; (d) a more direct role for the media as manifest sources of information about lifestyle, taste, sexuality, gender, family, domestic roles, etc., and as latent structuring influences, demonstrating through their narratives the idea of reflexive engagement with the project of the self; and (e) a further latent role of media in the implication that shared media consumption—like shared lifestyle choices—provides linkage to “communities of people who are ‘like us'—or people who, at least, have made similar choices” (p. 103). Hoover (2006) considers the methodological challenges of directly investigating the complex ways that media are articulated into such identity projects. Like Gauntlett, Hoover builds on the theoretical foundation of Giddens's work and Ricoeur's ideas about self-narratives to propose an approach he calls “plausible narratives of the self.” This is distinct from Gauntlett's approach in that it stresses constructivist ideas (cf. Hoover, Clark, & Alters, 2004) about the way knowledge is generated through the research process. As Hoover puts it, the notion of “plausibility”“… is rooted in the [symbolic] interactionist idea that these constructions of self involve a logic that they are fitted to certain contexts and certain ‘others’ (such as university-based researchers) for whom they must ‘make sense’” (Hoover, 2006, p. 94). Hoover suggests that these “narratives” are the object of interpretive research inquiries into audience experience with media. At the same time, though, … The purpose of the interviews is not to “collect” these narratives as such. The interviews collect discourses, interactions among informants and interactions between informants and interviewers. The agenda of the interview is to talk about media, home, family, religion, spirituality, and values. What results… is…elements of the larger, more comprehensive narratives of self through which these informants make sense of themselves for themselves and for others. (Hoover, 2006, p. 94) By these means, Hoover and Gauntlett propose that richer, more descriptive, and more relevant accounts can be made of the ways that women and men make sense of masculine identity through media practices. Hoover argues that such an approach is particularly relevant for studying religious ideas and discourses in media cultures. Religion and spirituality, he argues, are necessarily complex, nuanced, and layered, and it is only through careful, reflexive interpretation, and analysis that accounts can be made. The questions under consideration here—media, religion, and masculine identity—involve a layering of complexity on complexity, underscoring the need for careful, iterative, and constructivist methods. So, although there remain significant issues to be resolved in Hoover's and Gauntlett's approaches, this study proceeds along the lines they propose, hoping both to develop new insights into the questions here, and to further refine theory and method. Religion is not a major focus of the research reported here, in spite of its centrality to the conservative view of masculinity. Valuable recent work in the field of religious studies understands religion in the same sort of layered, nuanced way that we look at relations between masculinity and media. Important voices in religion theory suggest that we look at religion in daily life as a practice of meaning-making (Roof, 1999), and that it is more important to understand religion “as achieved” through these practices than “as ascribed” (Warner, 1993). Literatures in the emerging field of media and religion research have shown that religion and media interact in important ways in identity and meaning at the household level (Hoover et al., 2003) and among specific audiences (cf. Clark, 2002b). In these regards, religion and media can be seen to condition each other's “reception” at the household level (Hoover, 2006), something that we see in some detail in these findings in relation to these Evangelical households. Approach and method To date, in-depth interviews have been conducted with 30 households representing both “sides” of the Protestant-political “divide” in contemporary American religious politics: Evangelicals and Nonevangelicals (cf. Hunter, 1992).3 Nineteen of these families were identified as Evangelical Christian. Of these, all were White. All respondents were married at the time of the interview, and all were heterosexual. Forty-three percent of the families reported an annual income of more than $45,000. Eighty-one percent of the Evangelical men were between 30 and 50 years of age. All but one had children living at home, and only three Evangelical families had children older than 10 years at the time of the interviews. Seven of these Evangelical families lived in the American South, 10 in the West, and two in the Pacific Northwest. Respondents were recruited using a social network sampling method, which involved recruiting from the social networks of contacts in various parts of the United States. Respondents were asked to be involved in a study concerning the relationships among media, religion, masculinity, and civic engagement. The methods employed here involved what Bernard (2002) called “purposive sampling,” a sampling method that depends upon researchers' judgments to select respondents who fit the particular purposes of the study—in this case, White, Protestant men (p. 183). The point here was not to establish external validity, but to find key informants and critical cases. Thus, there was no attempt to create an unbiased sample; rather, this sample was “intentionally biased to get answers to questions of practical importance” regarding constructions and negotiations of masculinity in mediated and religious environments (Bernard, 2002, p. 184). The end, then, was to produce idiographic knowledge, not nomothetic knowledge, knowledge that applies “the richness of the particular … to other cases of the same problem in the larger culture.” (Lindlof, 1995, p. 57) Interviews were conducted in each household in two segments, an initial interview of around 2 hours in length with all the adults in the family, based on a semistructured interview agenda. These interviews were transcribed and analyzed, and specific follow-up questions were devised for individuals, again using a semistructured interview agenda. A second interview was then conducted with the male adult in the household. These were also around 2 hours in length. All interviews were transcribed and analyzed as well with the intention of elucidating the life narratives of manhood, fatherhood and masculinity that emerged from these conversations. Our analytic approach is rooted in constructivism and in Hoover's notion of the “plausible narrative of the self.” We conceive of the unit of analysis to be these “narratives,” which we interrogate along the dimensions that are most relevant to the questions under study. Our approach iterates through our narrative field material in a way described by LeCompte and Schensul (1999). “It starts with the laborious and time-consuming tasks of pulling apart fieldnote narratives … and then trying to figure out why they fit together as they do” (p. 69). From there, we move to a process consistent with Glasser and Straus's “constant comparison method” where we constantly compare “the similarities and differences in behaviors, settings, actors, and other dimensions of cultural life” (LeCompte and Schensul, 1999, p. 76). Finally, we begin to identify the most significant dimensions of these narratives, again through constant comparison. Following LeCompte and Schensul (1999), The stream of behavior of language is recorded and then separated into discrete concepts using constant comparison. The items are then ‘chunked’ into categories. Subsequent steps link the categories into concepts or theoretical constructs that, in turn, permit selection or development of theories that the researcher can use to explain what was observed in the field. (pp. 76–77) The overall scope of the study was, as we have said, occasioned by the neoconservative claims about masculinity, religion, and media. In keeping with that critique, the interviews looked at the question of media, and how media contribute to ideas of masculinity, fatherhood, and civic engagement in relation to these families' and men's religious and spiritual values and ideas. Within this large field of inquiry, our agenda here is to report on specific learnings around media, media reception, and the way that media are integrated into White Evangelical identity. Our investigations in the 19 Evangelical households in our sample can be said to constitute a “critical test” of Eberly's (1999) argument that self-conscious and morally focused religious culture—particularly White, Protestant, middle-class culture—should support the development of positive male identities directed at virtuous civic engagement. There is extensive evidence that American Evangelicalism understands that it should be such a place (cf. Roof, 1999). Evangelicalism is further identified with specific and focused approaches to masculine identity, which stress a variety of kinds of traditionalism (Edgell, 2006; Wilcox, 2004). It should also be noted that, although these questions are relevant across other ethnic, racial, and class lines, the focus of this study is on the particular neoconservative view articulated by Eberly and others, which, while aimed at a broader audience, is articulated (if only implicitly) and largely received along particular ethnic, racial, and class lines, that is, White and middle class. This is not to deny that similar claims about manhood and masculinity can be found in the African American and Hispanic communities. The implicit (and clearly problematic) idea that the White, middle-class context is somehow generic in terms of domestic values and ideals points to its centrality as the kind of “critical test” we had in mind for this research. The conservative critique is significant for its role in contemporary politics, as has been evident in recent U.S. campaigns and in the prominent advocacy of some of these ideas by figures including President Obama. To look first at places where its core prescriptions seemed to be aimed is an obvious first step. Very helpful subsequent work that extends this is obviously indicated and will be undertaken by ourselves and—we anticipate—by others. It is further worth saying that the relevance of these issues outside the U.S. context should not be assumed, but provocative implications might well emerge from analysis and interpretation with other national contexts in mind. Media as constructive resources for “plausible narratives of the self” “Positive” media as resource for identity Our interviews demonstrate that these Evangelical men are both reflexive about their own media practices and conversant with the dominant public scripts about media (cf. Hoover et al., 2004). That is, they critique media generally, seeing them as sources of a variety of social ills and as sources of a perceived “crisis of masculinity.” Yet, at the same time, all of them are able to pinpoint positive examples of masculinity in relation to media that align with their religious ideals and that help them understand themselves as men. In fact, significantly, they use examples from media to articulate their values about gender as readily as they used Biblical references or church teachings for this. As one respondent put it, referring to a specific recent discussion with youth at his church, “Our common language was movie language”—not Biblical text. This example suggests that these Evangelical men do not simply use media in instrumental ways alone, or that they are simply “guilty pleasures” for them. Rather, media function for them as they do for many others in society: They become a common source of shared language and shared experience that are integrated into interactions and potentially into life narratives and identity constructions as well. For example, one respondent, John, a student at an Evangelical seminary, found the character “Jack” on the television series, Lost, representing positive values of leadership. Jack, he said, “… actually does take up a head position. He actually tries to lead people … not just the women, the men as well. So he actually does show what leadership is about, caring for others, taking care, thinking of them first before himself.” Another respondent, Jasper, an Evangelical church pastor and new father, identified with the children's host, Mr. Rogers. He said, “I think he is a guy … who's humble and confident…. I think when you think about a man dealing with children, I think he is a great model for that because he relates to children in a way that they can receive and understand. Yet, at the same time, he's very much an adult. He doesn't have to be childish or immature.” These Evangelical men identify with a range of media, and we have found that these identifications often articulate with their vocations or life course. Given the variety, if not diversity, of media messages, this should not come as a surprise. Yet, beyond this level of identification, a common set of themes has emerged in our study among men who identify as Evangelical Christian and who espouse essentialist and/or masculinist ideas about gender. These themes focus on male leadership in the home and in society, male courage and strength, and male provision and protection. Media examples of these themes, as expressed by men in our study, include “Jack” on Lost, “Andy Taylor” on The Andy Griffith Show, “Mr. Ingalls” on Little House on the Prairie, and “Luke Skywalker” in the Star Wars films. However, one of the most prominent role models representing the embodiment of this set of themes is the actor/director Mel Gibson, particularly his turn as William Wallace in Braveheart. Clark, a pastor at a nondenominational, Evangelical church, best sums up this sentiment. Positive qualities of manhood. I think Mel Gibson has done a pretty good job. In the Lethal Weapon series … when we meet his character in Lethal Weapon 1, his wife has died and he's sort of this guy out there on the fringe of exploding himself … because his wife is dead…. But then as you go on in Lethal Weapon 4, he gets married and finds his new love and they have a baby and all this, and he becomes kind of a whole person again. So, in one sense that character helps us understand that man in and of himself will not carry on a race. There needs to be the counterpart. There needs to be the woman to complete him, to help him, but then also to produce children and begin the cycle of life all over again. But then just to use Mel Gibson again, you go to Braveheart, and you see a man who has a cause. He has a passion; he has something he is willing to die for. And I think that's probably—if you’re going to say what's the core difference between a man and a woman, a woman by and large will be protecting her children, and she may be willing to die to save her children. But she's going to try to do everything she can to protect her family whereas a man, he may be willing to die for a cause that he just firmly believes in, and that's kind of the Braveheart mentality. “Negative” media as resource for identity In addition to incorporating media positively into their narrative constructions of gender, the men in our sample also draw on negative male stereotypes in media. Although these negative references to media reinforce the received “public scripts” about the damaging effects of media, these references also work constructively in men's identity narratives. Naming particular media examples as “negative” provides boundary markers for identity; they represent “male others” from which men want to distance themselves. One example of this is in readings of the gay characters in Will and Grace. One respondent, Ned, focused on the gay characters in this show in contrast to the character, “Mr. Ingalls,” in Little House on the Prairie, who embodied for him a vision of the ideal man. I: We’ve talked about Mr. Ingalls as a media character that represents positive qualities of manhood. Can you think of a particular media character that's the opposite, that represents negative qualities, what you would perceive as negative qualities of manhood? N: Yeah, anytime I watch any of those homosexual type shows where there's—I'm trying to think of that stupid one, um, I can't remember the name of it because I never watch it. I: Will and Grace? N: Will and Grace. I: So that would be a representative of… Is it just the homosexuality, or are there a whole bunch of other qualities that are wrapped up in that that make that a negative? N: Both. He acts like a woman about half of the time, the way he talks. He relates to his girl friend better than he does a guy friend. I: How about the characteristics that you ascribed to Michael Landon's character in Little House—honesty, hardworking, strength and things like that? Do you see those characteristics in Will or not? N: I don't get the feeling of strength at all. I don't know about honesty. I haven't watched it enough. Every once in a while, I’ll flip it on and I’ll flip channels. So, I just catch a blurb of it. Definitely, it's a 180-degree difference from Mr. Ingalls. I don't know how to describe that, but it's just different. It is not surprising that an Evangelical, heterosexual man would pinpoint representations of homosexuality as a central “sin” of the entertainment media. However, it is surprising that this has not come up all that often in our interviews with Evangelical men. In fact, homophobic and/or heterosexist readings of media have been few in comparison to those critiquing the entertainment industry's portrayal of fatherhood. Respondents typically lament that fathers are depicted as socially challenged imbeciles juxtaposed against intelligent, hardworking, strong mothers. Among our informants, the most commonly cited examples of this stereotype are Homer Simpson, “Jim” in According to Jim, “Raymond” in Everybody Loves Raymond, “Tim Taylor” in Home Improvement, and “Al Bundy” in Married with Children. For example, Harold, an Evangelical seminary student, said, “You’ve kind of gone from Father Knows Best to Bill Cosby but then from there to modern shows where the father is the doofus, you know?” Also, Clark, who we heard from above: Well, gee, negative qualities of manhood…. Tim “the Tool Man” Taylor, he was always messing up something, breaking something, blowing up something, but yet he had little kernels of wisdom occasionally. Usually he stole those from Wilson across the fence and then couldn't quite say it right, so still that attitude that the wife has all the answers. She's the smart one, and the guys are just around so we have somebody to laugh at. In summary, then, it is clear that the Evangelical men in our study engage media reflexively. They incorporate what they perceive as “good” and “bad” media into their senses of self. As these excerpts from interviews illustrate, media resources and the values attached to them become the language through which men can articulate religious ideals of masculinity (e.g., what they call “headship”—cf. Wilcox, 2004) and essentialist understandings of masculinity (e.g., courage and strength as “naturally” male characteristics).4 Thus, rather than seeing media as unequivocally antagonistic, it is more appropriate to understand media as constructive in masculine identity narratives. In response to the neoconservative critique, then, it would be more productive to evaluate the ways that men engage media in their constructions of a gendered self than to think about media only in terms of “positive” or “negative” portrayals. Such an approach would examine how men attach value to media products and how that valuation allows them to both construct boundaries of self and to situate their narratives within what they perceive to be legitimated, social (gendered) spaces. Scripts about media In spite of the complexity that emerges in our interviews, most of the men in our study still express a negative view of media. Thus, it is not only important to evaluate the role of media in gender constructions, but also the role of what Hoover et al. (2003) call “accounts of” media in these constructions. For example, as Larry, a seminary student and youth pastor put it, “… I think most of [media] is destructive. I mean I think that … most of the messages that are being portrayed through movies and TV shows are far from Biblical, if even moral, if even truthful.” This view follows a common public script about the media in both mainstream and in Evangelical Christian discourses. It should not be surprising that middle class, White Christian men hold this view of media because it reinforces a longstanding class critique of media as lowbrow culture and a longstanding religious critique of media as morally bankrupt (cf. Fore, 1990; Meyers, 1989). These ideas are strong and consistent among religiously identified viewers across the religious spectrum (Hendershot, 2004; Hoover, 1988, 2006). Yet Larry, and most of the other men in our study also saw media as playing a positive role in their lives. First, they found satisfaction and meaning in media, and, second, they engaged with media to maintain cultural currency or relevance. It is too simplistic to dismiss this love/hate relationship with media as a contradiction within narratives of self, or as a rupture of them. Rather, it is best to understand their narratives as incorporating an “account of” media where these narratives can be said to express a desire to belong to a putative “common culture” by connecting with a common class—and gender—discourse (Hoover, 2006). At the same time, these accounts also reveal a desire to belong to a particular religious culture by connecting with a moral critique of the media that they see as specific to that culture. These accounts also work as a process of distinction in Bourdieu's sense: An effort to distinguish oneself from a “modern” culture that is perceived to be morally permissive and lax and, largely, anti-Christian. Thus, embodying this “common” critique of media is a way to locate oneself within legitimated social and moral space, a way to identify oneself as a particular type of moral person, a particular type of Christian, a particular type of man, etc. In addition to this, the engagement with media, that is the enjoyment and meaning taken from it, is not necessarily (or only) a contradiction of self, simply a guilty pleasure, or simply an effort to situate oneself in moral, social space; rather, it must be seen as a way to negotiate within this moral space, a way to distinguish oneself within a social space. For example, Larry, in addition to lamenting the state of mainstream media, also leveled a critique against religious media and demanded a Christian ethic of mainstream media engagement. Christian media, he said, tended to be “made by some cheesy Christian.” Thus, although his family engaged with Christian media, particularly Christian music and children's media, he advocated a “holistic” approach that did not “shut out” mainstream media. He said: I just think that as [the kids] get older we’re not seeking out what Christian movie's coming out now so we can go see it because it's going to be safe for our kids. Our filter is more, what can we be looking at for our kids to watch and for us to watch that's going to be edifying to us in some way? And so if that's our filter then it totally opens up the door beyond just Christian being the only media [that] we’re going to allow into our house … There's just a piece of that [Christian media] that we really rebel against because it feels like a watered down weak Christianity … So it's kind of walking that tightrope between utilizing Christian media a lot because it represents our values but that's not what shapes us as Christians. It is clear that Larry did not simply engage media as a “guilty pleasure”; rather, he saw this engagement with media as necessary and valuable to his identity and social relations. Thus, on the one hand, he expressed the common cultural and religious critique of mainstream media, thereby locating himself within a particular middle class, Evangelical Christian social space, although on the other hand, he critiqued what he perceived to be an older, “modern” isolation from media embedded in this Evangelical Christian social space. What had resulted from this isolationist critique—in his view—was a mediocre Christian subculture that he viewed with “disgust.” The nuances within his discourse thus suggest a desire to belong to a particular moral, religious social space, yet also a desire to move within and to change—though not necessarily subvert—this space. Looking at narratives about media in this way moves the discourse beyond contradictions or ruptures of self. Instead, lodging “accounts of” media within social discourses and social spaces of power encourages analysis of the ways in which conflicting narratives “work” to situate a person in social space and within dynamics of power. Conclusions and discussion Our conversations have led to a significant challenge to the critique we set out to test. In that view—and indeed in a wide range of perspectives on the relationship between media and gender—media are seen as playing a consistently negative role. Their dominant representations of men and masculinity are claimed to convey either traditionalist senses of what Hanke calls “hegemonic masculinity” or to carry models and representations that run directly counter to positive or normative roles for men. Some critics further identify the individualism and commercialism of media as stressing values that contradict positive “masculine” values of responsibility, virtue, and character (Eberly, 1999). The religious valence of these critiques suggests, as we have said, that religion—and we can presume that by this, they mean conservative religion—should be a kind of antidote to both the negative media influences and to the broader “crisis of masculinity.” If this were true, we might have expected these men's narratives of media in relation to identity to be consistent with these critiques. Indeed, many of our informants did verbalize “accounts of” media that echoed these larger received discourses in the culture. In fact, they articulated such received discourses along two dimensions. They talked about media in ways that were consistent with generally accepted ideas among educators and cultural elites that certain kinds of media and media consumption are simply “bad.” And they also talked about media in ways that were consistent with the more focused, moralistic critiques found among religious leaders and groups. But, what they did with media in their own behaviors and in their accounts of self contradicted the sense that media categorically contradict positive and constructive practices of self and identity. As we have seen, and as Gauntlett (2002) predicted (though he was silent on the question of religion), these informants have used media to build identities that are normatively “positive.” They do this through a reflexive engagement with media and a constructive process of articulating ideas and symbols derived from the media into senses of who they are. Their appropriations have a kind of elasticity. They can take on board what they see to be positive role models and positive representations and identify with them directly. At the same time, they can constructively use representations that they consider to be “negative” as ways of being distinct through their own codings of them. Further, they can use their particular combination of these decodings as a way of claiming an identity that is mapped onto both the context of the broader “common culture,” and the more particular contexts of the religious culture with which they also identify. Thus, it appears that the media are a salient resource for self instead of a direct challenge to it, at least for these informants. And, in spite of the speculations to the contrary, the media, as these informants articulate them can be complementary (not necessarily contradictory) to religion. These men pick and choose media images and messages that “fit” their (often religious) understanding of the world and their gendered place within it. These insights serve an evolving understanding of the relationship between media and religion as well. Just as the media have been shown to be articulated into the processes whereby gendered identities are formed and maintained in contemporary life, so they can be said to contribute to religious identities. More significantly, perhaps, are the implications that this process has for religion and for religious authority. The neoconservative critique implies that religiously inflected ways of framing and interpreting media can ultimately contest and confront their putative negative influences. These interviews show that the processes and the relationships are much more subtle, layered, and nuanced than that. In fact, the media can be seen to play a culturally definitive role in relation to religious and gender identities at the same time that cultural authorities and audiences in their homes share a common critique of those same media. In spite of those critiques, the way that mediated images, symbols, and practices are used by audiences mean that the media do, in important ways, have the “last word.” Notes 1 " This is evidenced by comments made by President Obama, for example, his Father's Day speech at the Apostolic Church in Chicago, June 15, 2008 (http://my.barackobama.com). 2 " Eberly also founded the National Fatherhood Initiative (http://www.fatherhood.org/about/organization-history). 3 " Seventeen other families have been interviewed for this study as well. These include five Jewish families, six Roman Catholic families, and six families with no religious affiliation or self-identified as “spiritual.” 4 " More research is needed to determine whether this relationship to media is the same with non-Evangelical Christian men. Also, it should be noted that at least one Evangelical Christian male in our study did not express the religious ideal of headship or essentialist characteristics of gender in his narrative, though he did still use the language of media to express his narrative of gender, for example, describing positive male qualities expressed in Mr. Rogers' Neighborhood. References Askew , S. , & Ross , C. ( 1990 ). Boys don't cry: Boys and sexism in education . Milton Keynes, England : Open University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Barcus , F. E . ( 1983 ). Images of life on children's television: Sex roles, minorities, and families . New York, NY : Praeger . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Barthel , D . ( 1992 ). When men put on appearances: Advertising and the social construction of masculinity. In S. Craig (Ed.), Men, masculinity, and the media (pp. 137 – 153 ). Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Baudrillard , J . ( 1970 ). The consumer society: Myths and structures . Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Berkowitz , B . ( 2005 ). Don Eberly's Conservative Civil Society. In MediaTransparency.org. Retrieved from http://www.mediatransparency.org/personprofile.php?personID=124. Bernard , H. R . ( 2002 ). Research methods in anthropology: Qualitative and quantitative approaches . New York, NY : AltaMira . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Beynon , J . ( 2002 ). Masculinities and culture . Buckingham, England : Open University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Butler , J . ( 1990 ). Gender trouble: Feminism and the subversion of identity. London, England: Routledge. Clare , A . ( 2000 ). On men: Masculinity in crisis . London, England : Chatto and Windus . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Clark , J. M . ( 2002a ). Faludi, fight club, and phallic masculinity: Exploring the emasculating economics of patriarchy . The Journal of Men's Studies , 11 ( 1 ), 65 – 76 . Google Scholar Crossref Search ADS WorldCat Clark , L. S . ( 2002b ). From angels to aliens: Teenagers, the media, and the supernatural . New York, NY : Oxford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Connell , R.W . ( 2000 ). The men and the boys . Sydney, Australia : Allen & Unwin . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Connell , R. W . ( 2003 ). Masculinities, change, and conflict in global society: Thinking about the future of men's studies . The Journal of Men's Studies , 11 ( 3 ), 249 – 266 . Google Scholar Crossref Search ADS WorldCat Consalvo , M . ( 2003 ). The monsters next door: Media constructions of boys and masculinity . Feminist Media Studies , 3 ( 1 ), 27 – 45 . Google Scholar Crossref Search ADS WorldCat Craig , S . ( 1992 ). Considering men and the media: Introduction. In S. Craig (Ed.), Men, masculinity, and the media (pp. 1 – 7 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Craig , S . ( 1993 ). Selling masculinities, selling femininities: Multiple genders and the economics of television . The Mid-Atlantic Almanac , 2 , 15 – 27 . OpenURL Placeholder Text WorldCat Croteau , D. , & Hoynes , W. ( 1992 ). Men and the news media: The male presence and its effect. In S. Craig (Ed.), Men, masculinity, and the media (pp. 154 – 168 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Denski , S. , & Sholle , D. ( 1992 ). Metal men & glamour boys: Gender performance in heavy metal. In S. Craig (Ed.), Men, masculinity, and the media (pp. 41 – 61 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Donald , R. R . ( 1992 ). Masculinity and machismo in Hollywood's war films. In S. Craig (Ed.), Men, masculinity, and the media (pp. 124 – 136 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Eberly , D . ( 1999a ). Families, fathers, and the making of democratic citizens. Civil Society Project Essays on Civil Society, 99 (1) . Retrieved from http://www.civilsocietyproject.org/_files/FathersDemocraticCitizens.pdf. Eberly , D . (Ed.). ( 1999b ). The faith factor in fatherhood: Renewing the sacred vocation of fatherhood . New York, NY : Lexington Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Eberly , D . ( 2002 ). The soul of civil society . New York, NY : Lexington Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Edgell , P . ( 2006 ). Religion and family in a changing society . Princeton, NJ : Princeton University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Farrell , W . ( 2001 ). The myth of male power . New York, NY : Penguin . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fejes , F . ( 1992 ). Masculinity as fact: A review of empirical mass communication research on masculinity. In S. Craig (Ed.), Men, masculinity, and the media (pp. 9 – 22 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Fiske , J . ( 1997 ). Understanding popular culture . London, England : Unwin-Hyman Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fore , W. F . ( 1990 ). Mythmakers: Gospel, culture, and the media . New York, NY : Friendship Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gauntlett , D . ( 2002 ). Media, gender and identity: An introduction . London, England : Routledge . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Gavanas , A . ( 2004 ). Domesticating masculinity and masculinizing domesticity in contemporary U.S. fatherhood politics . Social Politics: International Studies in Gender, State & Society , 11 ( 2 ), 247 – 266 . Google Scholar Crossref Search ADS WorldCat Giddens , A . ( 1991 ). Modernity and self-identity: self and society in the late modern age. Stanford, CA: Stanford UP. Gunter , B . ( 1995 ). Television and gender representation . London, England : John Libbey . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hanke , R . ( 1992 ). Redesigning men: Hegemonic masculinity in transition. In S. Craig (Ed.), Men, masculinity, and the media (pp. 185 – 198 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Hanke , R . ( 1998 ). Theorizing masculinity with/in the media . Communication Theory , 8 ( 2 ): 183 – 203 . Google Scholar Crossref Search ADS WorldCat Hearn , J. , & Melechi , A. ( 1992 ). The transatlantic gaze: Masculinities, youth, and the American imaginary. In S. Craig (Ed.), Men, masculinity, and the media (pp. 215 – 232 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Hendershot , H . ( 2004 ). Shaking the world for Jesus . Chicago, IL : University of Chicago Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Himmelfarb , G . ( 1995 ). The demoralization of society from Victorian virtues to modern values . New York, NY : Alfred A. Knopf . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hoover , S. M . ( 2006 ). Religion in the media age . London, England : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hoover , S. M. , Clark , L. S., & Alters , D. F. ( 2004 ). Media, home, and family . New York, NY : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hunter , J. D . ( 1992 ). Culture wars: The struggle to control the family, art, education, law, and politics in America . New York, NY : Basic Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Justad , M. J . ( 2000 ). Women's studies and men's studies: Friends or foes? The Journal of Men's Studies , 8 ( 3 ), 401 – 406 . Google Scholar Crossref Search ADS WorldCat Kimmel , M. S . (Ed.). ( 1995 ). The politics of manhood: Profeminist men respond to the Mythopoetic men's movement (and the Mythopoetic leaders answer) . Philadelphia, PA : Temple University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kirkham , P. , & Thumin , J. ( 1995 ). Me Jane: Masculinity, movies, and women . New York, NY : St. Martin's Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC LeCompte , M. , & Schensul , J. ( 1999 ). Designing and constructing ethnographic research . Walnut Creek, CA : Alta Mira Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lindlof , T. R . ( 1995 ). Qualitative communication research methods . Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Macnamara , J. R . ( 2006 ). Media and male identity: The making and remaking of men . New York, NY : Palgrave-Macmillan . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Mansfield , H. C . ( 2006 ). Manliness . New Haven, CT : Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Messner , M. A . ( 1997 ). Politics of masculinities: Men in movements . Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Meyers , K. A . ( 1989 ). All God's children and blue suede shoes: Christians and popular culture . Grand Rapids, MI : Crossway Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC National Public Radio. ( 2005 , February 9). Laura Bush: Putting boys in the spotlight. All Things Considered, NPR. National Public Radio. ( 2007 , July 2). Masculinity in the twenty-first century. Talk of the Nation, NPR. Obama , B . ( 2008 ). Father's day speech. Retrieved from http://my.barackobama.com/page/community/post/stateupdates/gG5nFK. Putnam , R . ( 2001 ). Bowling alone: The collapse and revival of American community . New York, NY : Simon and Schuster . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Roof , W. C . ( 1999 ). Spiritual marketplace: Baby Boomers and the remaking of American religion . Princeton, NJ : Princeton University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sabo , R. , & Jansen , S. C. ( 1992 ). Images of men in sport media: The social reproduction of gender order. In S. Craig (Ed.), Men, masculinity, and the media (pp. 169 – 184 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Saco , D . ( 1992 ). Masculinity as signs: Poststructuralist feminist approaches to the study of gender. In S. Craig (Ed.), Men, masculinity, and the media (pp. 23 – 39 ). Newbury Park, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Warner , S. R . (1993, March). Work in progress toward a new paradigm for the sociological study of religion in the United States . American Journal of Sociology , 98 ( 5 ), pp. 1044 – 1093 . Crossref Search ADS WorldCat Wetherell , M. , & Edley , N. ( 1999 ). Negotiating hegemonic masculinity: Imaginary positions and psycho-discursive practices . Feminism & Psychology , 9 , 335 – 356 . Google Scholar Crossref Search ADS WorldCat Wilcox , W. B . ( 2004 ). Soft patriarchs, new men: How Christianity shapes fathers and husbands . Chicago, IL : University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2011 International Communication Association

Potter, W., James

doi: 10.1111/j.1460-2466.2011.01586.xpmid: N/A

Abstract Scholars have generated evidence of a wide variety of mass media effects over almost 9 decades of research. Although each of these effects has been defined in a relatively clear manner, there has been much less conceptualizing about what constitutes a mass media effect in general. Rarely have scholars provided a formal definition of mass media effect, instead opting to provide a definition in either an ostensive or primitive manner. In this essay, a conceptualization of “mass media effect” at the most general level is synthesized from this previous definitional work. The proposed conceptualization posits 4 general kinds of mass media effects: gradual long-term change in magnitude, reinforcement, immediate shift, and short-term fluctuation change. Since the scholarly study of mass media effects began in the 1920s (Lowry & DeFleur, 1988), researchers have generated a very large effects literature, with some estimates as high as 4,000 published studies (Potter & Riddle, 2007). This literature has documented a great many effects. Although most of these individual effects present scholars with a clear conceptualization of that particular effect, there is comparatively little work that focuses attention on what a “media effect” means at a more general level of conceptualization. That is, we have clear definitions for many specific effects (such as agenda setting, cultivation, third person, and so on), but we lack a clear conceptualization about the idea of “mass media effect” in general. Given the length of time we have now been studying mass media effects and the size of the empirical literature, it is possible to abstract some organizing principles that tend to capture the essence exhibited across all the specific effects. This set of organizing principles could then serve as a conceptualization of mass media effect in general. The term “mass media effects” continues to be used widely despite the fact that many media scholars struggle with the term “effect.” For example, Wright (1986) points out that “it should be apparent that we speak of the ‘effects’ of mass communication only as a convenient expression” (p. 11). And Morley (1992) says that “‘effects’ is thus a shorthand, and inadequate, way of marking the point where audiences differentially read and make sense of messages” (p. 86). Traudt (2005) argues that “the word effect is something of a misnomer, used by many because of its economy and convenience as a label encompassing a complex and varied field of theory and research” (p. 11). In the introduction to their Handbook of Media Process and Effects, Nabi and Oliver (2009) assert that the most important conceptual issue facing the field of mass media studies is: “What do we mean by ‘media effects'?” (p. 2). They argue that media scholars have paid little attention to this fundamental question for two reasons. One reason is because junior scholars are under pressure to pursue grants and to publish a large number of studies so that they develop habits to replicate past research methods and topics “often at the expense of the more sophisticated, theoretically rich work that we value as a discipline” (p. 2). A second reason is that the scholarly field has been primarily motivated to identify specific harmful effects along with the factors that would increase the probability of those harmful effects occurring. Wartella and Reeves (1985) have also made this observation when they pointed out that as each new mass medium has developed, the first question the public asked has been about how the medium might influence behavior in a negative manner. The scholarly field has been much less motivated to develop conceptualizations about the nature of mass media effect in general—one that would allow for positive and neutral effects in addition to the focus on negative effects and shift the focus to what they all have in common. To the two reasons for why scholars have not produced more conceptual work on the nature of “mass media effect,” I add a third reason—the complexity of the phenomenon. This complexity is revealed in the different ways that reviewers of this literature have struggled to organize their presentations. An analysis of the major reviews of the media effects literature suggests that there are nine conceptual issues raised about the nature of mass media effect (see Table 1). Perhaps, the most prevalent of these issues is the concern about type of effect, which almost all reviewers acknowledge (Basil, 1992; Grossberg, Wartella, & Whitney, 1998; Jeffres, 1997; Lazarsfeld, 1948; McLeod & Reeves, 1980; Roberts & Maccoby, 1985; Weiss, 1969). However, the reviewers vary considerably about what the effect types are. For example, Grossberg et al. (1998) say there are three types: cognitive, affective, and conative; Lazarsfeld (1948) argued for four types: knowledge, attitudes, opinions, and behavior; Bryant and Zillmann (2009) say there are five types: behavioral, attitudinal, cognitive, emotional, and physiological. There is also a difference of opinion on each of the other eight issues listed in Table 1. Table 1 Definitional Considerations “Type of effect issue: Do the effects show up as behaviors only, or do they also show up in variety of other ways such as cognitions, attitudes, beliefs, affects, physiology, etc.? (Basil, 1992; Grossberg et al., 1998; Jeffres, 1997; Lazarsfeld, 1948; McLeod & Reeves, 1980; Roberts & Maccoby, 1985; Weiss, 1969) Level of effect issue: Does the effect take place at the microlevel, the macrolevel, or both? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Change Issue: Does the effect show up as an alteration or a stabilization of something already existing? (Grossberg et al., 1998; Hovland, 1954; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Influence issue: Must the media have a direct influence or can it have an indirect influence? (McLeod & Reeves, 1980; Roberts & Maccoby, 1985) Pervasiveness issue: Do the media exert effects on everyone or is their influence conditional, such that different people are influenced in different ways? (Baran & Davis, 2000; Grossberg et al., 1998; Littlejohn, 1999; McLeod & Reeves, 1980; McQuail, 2005; Roberts & Maccoby, 1985; Severin & Tankard, 2001; Wartella & Reeves, 1985; Weimann, 2000) Type of media stimulus issue: Do the media exert a diffuse-general influence on the culture because of their form and the nature of their channels of transmission or is the influence keyed to content-specific factors in their messages? (Basil, 1992; McLeod & Reeves, 1980; Perse, 2001) Intentionality issue: Should scholars be concerned with effects that were not intended by senders of receivers or only intended effects? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Perse, 200; Weiss, 1969) Timing of effect issue: Should scholars be concerned with effects that take a long time to show up and difficult to trace to media influence or only those effects that occur during exposures? (Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Lazarsfeld, 1948; Perse, 2001; Roberts & Maccoby, 1985; Weiss, 1969) Measurability issue: Should scholars be concerned with latent effects or only those manifest effects that can be more easily observed? (Grossberg et al., 1998; Hovland, 1954; Weiss, 1969) “Type of effect issue: Do the effects show up as behaviors only, or do they also show up in variety of other ways such as cognitions, attitudes, beliefs, affects, physiology, etc.? (Basil, 1992; Grossberg et al., 1998; Jeffres, 1997; Lazarsfeld, 1948; McLeod & Reeves, 1980; Roberts & Maccoby, 1985; Weiss, 1969) Level of effect issue: Does the effect take place at the microlevel, the macrolevel, or both? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Change Issue: Does the effect show up as an alteration or a stabilization of something already existing? (Grossberg et al., 1998; Hovland, 1954; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Influence issue: Must the media have a direct influence or can it have an indirect influence? (McLeod & Reeves, 1980; Roberts & Maccoby, 1985) Pervasiveness issue: Do the media exert effects on everyone or is their influence conditional, such that different people are influenced in different ways? (Baran & Davis, 2000; Grossberg et al., 1998; Littlejohn, 1999; McLeod & Reeves, 1980; McQuail, 2005; Roberts & Maccoby, 1985; Severin & Tankard, 2001; Wartella & Reeves, 1985; Weimann, 2000) Type of media stimulus issue: Do the media exert a diffuse-general influence on the culture because of their form and the nature of their channels of transmission or is the influence keyed to content-specific factors in their messages? (Basil, 1992; McLeod & Reeves, 1980; Perse, 2001) Intentionality issue: Should scholars be concerned with effects that were not intended by senders of receivers or only intended effects? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Perse, 200; Weiss, 1969) Timing of effect issue: Should scholars be concerned with effects that take a long time to show up and difficult to trace to media influence or only those effects that occur during exposures? (Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Lazarsfeld, 1948; Perse, 2001; Roberts & Maccoby, 1985; Weiss, 1969) Measurability issue: Should scholars be concerned with latent effects or only those manifest effects that can be more easily observed? (Grossberg et al., 1998; Hovland, 1954; Weiss, 1969) Open in new tab Table 1 Definitional Considerations “Type of effect issue: Do the effects show up as behaviors only, or do they also show up in variety of other ways such as cognitions, attitudes, beliefs, affects, physiology, etc.? (Basil, 1992; Grossberg et al., 1998; Jeffres, 1997; Lazarsfeld, 1948; McLeod & Reeves, 1980; Roberts & Maccoby, 1985; Weiss, 1969) Level of effect issue: Does the effect take place at the microlevel, the macrolevel, or both? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Change Issue: Does the effect show up as an alteration or a stabilization of something already existing? (Grossberg et al., 1998; Hovland, 1954; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Influence issue: Must the media have a direct influence or can it have an indirect influence? (McLeod & Reeves, 1980; Roberts & Maccoby, 1985) Pervasiveness issue: Do the media exert effects on everyone or is their influence conditional, such that different people are influenced in different ways? (Baran & Davis, 2000; Grossberg et al., 1998; Littlejohn, 1999; McLeod & Reeves, 1980; McQuail, 2005; Roberts & Maccoby, 1985; Severin & Tankard, 2001; Wartella & Reeves, 1985; Weimann, 2000) Type of media stimulus issue: Do the media exert a diffuse-general influence on the culture because of their form and the nature of their channels of transmission or is the influence keyed to content-specific factors in their messages? (Basil, 1992; McLeod & Reeves, 1980; Perse, 2001) Intentionality issue: Should scholars be concerned with effects that were not intended by senders of receivers or only intended effects? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Perse, 200; Weiss, 1969) Timing of effect issue: Should scholars be concerned with effects that take a long time to show up and difficult to trace to media influence or only those effects that occur during exposures? (Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Lazarsfeld, 1948; Perse, 2001; Roberts & Maccoby, 1985; Weiss, 1969) Measurability issue: Should scholars be concerned with latent effects or only those manifest effects that can be more easily observed? (Grossberg et al., 1998; Hovland, 1954; Weiss, 1969) “Type of effect issue: Do the effects show up as behaviors only, or do they also show up in variety of other ways such as cognitions, attitudes, beliefs, affects, physiology, etc.? (Basil, 1992; Grossberg et al., 1998; Jeffres, 1997; Lazarsfeld, 1948; McLeod & Reeves, 1980; Roberts & Maccoby, 1985; Weiss, 1969) Level of effect issue: Does the effect take place at the microlevel, the macrolevel, or both? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Change Issue: Does the effect show up as an alteration or a stabilization of something already existing? (Grossberg et al., 1998; Hovland, 1954; McLeod & Reeves, 1980; Perse, 2001; Weiss, 1969) Influence issue: Must the media have a direct influence or can it have an indirect influence? (McLeod & Reeves, 1980; Roberts & Maccoby, 1985) Pervasiveness issue: Do the media exert effects on everyone or is their influence conditional, such that different people are influenced in different ways? (Baran & Davis, 2000; Grossberg et al., 1998; Littlejohn, 1999; McLeod & Reeves, 1980; McQuail, 2005; Roberts & Maccoby, 1985; Severin & Tankard, 2001; Wartella & Reeves, 1985; Weimann, 2000) Type of media stimulus issue: Do the media exert a diffuse-general influence on the culture because of their form and the nature of their channels of transmission or is the influence keyed to content-specific factors in their messages? (Basil, 1992; McLeod & Reeves, 1980; Perse, 2001) Intentionality issue: Should scholars be concerned with effects that were not intended by senders of receivers or only intended effects? (Basil, 1992; Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Perse, 200; Weiss, 1969) Timing of effect issue: Should scholars be concerned with effects that take a long time to show up and difficult to trace to media influence or only those effects that occur during exposures? (Grossberg et al., 1998; Hovland, 1954; Jeffres, 1997; Lazarsfeld, 1948; Perse, 2001; Roberts & Maccoby, 1985; Weiss, 1969) Measurability issue: Should scholars be concerned with latent effects or only those manifest effects that can be more easily observed? (Grossberg et al., 1998; Hovland, 1954; Weiss, 1969) Open in new tab It is understandable why there is a range of opinion on each of these nine conceptual issues. As different scholars approach the examination of this complex phenomenon, they focus on different parts of it. For example, psychologists focus on effects on individuals, whereas sociologists focus on effects on larger aggregates of people and economists focus on effects on resources rather than people. Laboratory experimenters focus on short-term effects immediately after exposure to particular messages, whereas ethnographers focus on effects that take much longer to accumulate and manifest themselves in more subtle ways. This focusing on one part of the phenomenon helps scholars reduce the complexity to a point where the task of designing a research study becomes more manageable. However, it is also important that scholars periodically think about the bigger picture of our growing understanding about the phenomenon in full. The continual clarification of a field's central construct is an essential scholarly task. Metaphorically, the task of building a scholarly field is like assembling puzzle pieces when we (a) must create the pieces, and (b) do not know what the big picture of the assembled pieces will look like. Instead we proceed by creating new pieces and trying different configurations of assembly until we begin converging on a vision of a big picture at which time the assembly (and the creating of the missing pieces) proceeds more efficiently. Mass media effects scholars have created a great many pieces to the puzzle of mass media effects; however, we have spent much less effort on thinking about the big picture and speculating about how the pieces could be assembled to reveal that bigger picture. The purpose of this essay is to address the bigger picture by first reviewing how mass media scholars have conceptualized the idea of effect through the way they have presented definitions. Then I will synthesize a conceptualization of mass media effect by assembling the ideas in those definitions along with the issues that scholars have been struggling with over the nine decades the field has existed. Forms of defining mass media effect Definitions of mass media effect can be seen in review articles, edited books, and textbooks. At times, scholars will clearly present a definition for the concept, but most often the scholars' definitions for the concept have to be inferred from the way those scholars present the concept by critiquing it, by challenging assumptions about it, or by displaying examples of it. The analysis of these definitions is organized in this essay according to one of three methods that scholars have used to present the definitions in their own writings. These three methods—ostensive, primitive, and formal—are illuminated and critiqued below. Ostensive method The ostensive method of defining something is to present examples. An ostensive form of definition is typically used by scholars who publish reviews of the mass media effects literature (Basil, 1992; Bryant & Thompson, 2002; Bryant & Zillmann, 2002; Grossberg et al., 1998; Harris, 2004; Hovland, 1954; Iyengar, 1997; Jeffres, 1997; Lazarsfeld, 1948; Littlejohn, 1999; Lowry & DeFleur, 1988; McDonald, 2004; McGuire, 1986; McLeod & Reeves, 1980; McQuail, 2005; Perse, 2001; Roberts & Maccoby, 1985; Severin & Tankard, 2001; Sparks, 2006; Weiss, 1969). Rather than providing a formal definition with classification rules, these scholars present a list of things they regard as instances of mass media effects and leave it to the reader to infer what all of these examples have in common. The ostensive method of defining things works extremely well in everyday life. When we are toddlers, a good deal of language acquisition comes from the ostensive method of defining things. Parents typically teach their children what many things are by simply pointing out examples and saying the word—for example, “dog.” When children see enough examples, they will understand what a dog is. They infer what the classification rules are without having to learn a formal definition. The ostensive method of defining things also works well in the scholarly realm up to a point. When a research field is new and looking for boundaries, structure, and meaning, scholars take comfort in the concreteness of examples, so the ostensive type of definition is valuable. But beyond a certain point, the presentation of more examples loses its value and we need to think beyond the individual examples and infer patterns. Through trial and error these inferred patterns become more useful as they develop into classification rules that help us decide what new examples should be included or excluded. Also over time, these inferred patterns provide us with a clearer conceptualization of the nature of the phenomenon we are trying to define. Primitive method The primitive method of defining something assumes that others share substantially the same meaning, so there is no reason to articulate a formal definition. However, the danger with the primitive method is that other people might not share the same definition, and this is especially the case when the definition has several elements that vary in the degree to which they are shared. With mass media effects, the assumed shared meaning is that the mass media exert a pervasive causal influence that shows up in a variety of significant effects. For example, Perse (2001) claims that the primary assumption of mass media scholars has been that the “media and their content have significant and substantial effects” (p. 3). McQuail (2005) says “the entire study of mass communication is based on the assumption that the media have significant effects” (p. 456). This assumption shows up most clearly in the challenges to it. Perhaps, the most prevalent challenge has been in the form of a debate over whether the media exert an influence and if so how strong that influence is. Many scholars write about the historical disagreement on this point (Baran & Davis, 2000; Grossberg et al., 1998; Littlejohn, 1999; McQuail, 2005; Severin & Tankard, 2001; Wartella & Reeves, 1985; Weimann, 2000). For example, McQuail (2005) explains that the early history of thinking about mass media effects was a debate about whether the media exerted a powerful or no effect and that we have evolved to a point where scholars generally assume that the audience negotiates meaning. Although the media messages are important as stimulating those negotiations, the audience members' personal interpretations are even more important. What is most interesting about the writings on this debate is that scholars have assumed that other scholars have accepted this assumption about media effects. Therefore, the assumption is more than the media exert an influence; the assumption is also that other scholars accept this assumption. This is what reveals most strongly that these scholars are approaching the idea of media effects in a primitive manner, that is, rather than providing a conceptualization, media scholars assume a conceptualization that already exists and it is widely shared. Other challenges to this assumption take the form of additions to and subtractions from the assumption. This too is evidence of a primitive approach to defining mass media effects, because these scholars present their addition or subtraction without providing a clear conceptualization of the core to which they are making the addition or subtraction. That is, rather than provide a formal definition of mass media effects, scholars will either challenge a feature and argue for its subtraction or argue that an important feature is missing and suggest an addition. An example of a subtraction is provided by Wright (1986) who contends that causation should be eliminated from the assumed definition. He warns that mass media influence is not simple or direct when he asserts that it does not cause any “of the social phenomena in which mass communication is implicated. From our sociological perspective, an analytical model of direct cause and effect is inappropriate and unproductive.” He adds, “we seek to understand the kinds of contributions that mass communication and interpersonal communication together make to important social phenomena, including both socially differentiated and commonly shared knowledge, beliefs, and behavior” (p. 163). As for additions to the assumed definition, Webster and Phalen (1997) argued that too often “the traditional form of the effects question asks, what do media do to people. But this is not the only way that media effects can be conceived” (p. 188). They add that social institutions and the media themselves are targets of influence. Implied in this addition is an assumption that the mass media influence only individuals and not larger social aggregates. Other examples of additions are provided by Traudt (2005) who calls for more attention to individual differences when he asserts that “audience members are unique in that they bring a personal set of filters to the mass media experience and that these differences are a function of prior life experiences” (p. 11) and these must be taken into consideration when examining media effects. Traudt also adds that often what researchers claim as effects are temporary and therefore call into question whether an effect occurred or not when he says, “The results from key and influential studies never demonstrate in absolute terms a total and lasting relationship between the exposure to media content and consequent attitude or behavioral changes” (p. 11). Looking across these primitive forms of definition, it appears that mass media scholars believe that there is a shared assumption that defines effects as outcomes among individuals as a result of pervasive exposure to media messages, usually negative. This assumption begs for a conceptualization that could address questions like the following: What qualifies as an outcome? Must an outcome be a change or can it be a reinforcement of something already existing? What is the timing of effects, that is, must it be observed during the media exposure or can it be observed much later? If it can be observed much later, how can it be attributed to media influence? Does it need to be observable in order to be considered an effect? These are important questions if we are concerned about conceptualizing mass media effects in general and trying to understand big picture patterns that span across the wide variety of individual effects we have identified so far. Examples of formal definitions There are a few examples where scholars provide a formal definition of mass media effect. These definitions address the assumption about media influence and elaborate this central idea by providing details on some of the nine definitional issues from Table 1. These articulations serve to convey a more complete sense of the nature of the phenomenon of mass media effect as well as provide a sense of classification rules. For example, Jeffres (1997) in Mass Media Effects argues that before defining effects, he first needs to define mass media, communication, encoding, channel, and decoding process. Although he provides a formal definition for each of these concepts, he concludes his definitional section by saying, “The issue of what constitutes effects is difficult to resolve because it means different things to different people. ‘Effects’ may refer to the relationship between encoding and decoding activities within mass communication…. ‘Effects’ may link mass communication processes with other systems and the larger society…. Media effects may refer to the impact of media ‘as a whole.’” (pp. 4–5). Bryant and Zillmann (2009) say that “when scholars talk about media effects, they are considering the social or psychological changes that occur in consumers of the media message systems—or in their social milieu or cultural values—as a result of being exposed to, processing, or acting on those mediated messages. Five classes of media effects on individuals are often considered: behavioral, attitudinal, cognitive, emotional, and physiological” (p. 13). Arguably the clearest example of a formal definition of mass media effects is offered by McQuail (2005). In his glossary to the fifth edition of McQuail's Mass Communication Theory, McQuail defines effects of media as “the consequences or outcomes of the working of, or exposure to, mass media whether or not intended. They can be sought at different levels of social analysis. There are many types of effect, but it is usual to distinguish at least between effects that are behaviourial, attitudinal (or affective) and cognitive. Effects are distinct from ‘effectiveness’, which relates to the efficiency of achieving a given communicative objective” (p. 554). Thus, he deals explicitly with the definitional issues of intentionality, level, and type. Now that we are close to a century of empirical work on mass media effects, it is time to move beyond assuming that there is a shared definition for mass media effect in general. It is time to move beyond ostensive definitions and try to develop a formal definition that would provide a more clearly articulated conceptual frame for all the specific media effects. Toward a conceptualization of mass media effect The task of conceptualizing “mass media effect” requires synthesis. Moreover, I began this task by analyzing the literature to identify the key definitional ideas. The next step is to decide which of those many ideas to employ in assembling a conceptualization that captures the essence of the phenomenon. There are two strategies that can be used to select elements—a prescriptive strategy and a descriptive strategy. The prescriptive strategy directs the synthesizer to select a subset of definitional elements from the literature using some criterion, such as utility. The scholar filters through the ideas to select only those that meet the criteria. In contrast, the descriptive strategy tries to include as many elements as possible in order to illuminate the full range of ideas existing in the literature. The prescriptive strategy is most useful when a general conceptualization already exists and a scholar believes that conceptualization contains faulty elements; the scholar then presents a vision different than the existing conceptualization and argues for an alternative conceptualization that subtracts out the elements she regards as faulty and adds other elements she thinks were missing from the older conceptualization. Thus, she prescribes an alternative conceptualization that is argued to be superior to the older conceptualization. Because I am not challenging an existing conceptualization, but rather the lack of one, and because I want this conceptualization to be as inclusive of all perspectives as possible, I will use the descriptive strategy to select definitional elements. The purpose of this conceptualization then is to incorporate as much of the thinking about the phenomenon as possible and to organize it in a way so as to make a set of clear statements that reveal its essential nature as well as its boundaries. As for its essential nature, the central question is: What kinds of general effects are there and how are they similar and different from one another? As for its perimeter, the central question is: What are the classification rules that can be used to distinguish between mass media effect and noneffect? The definition Building from the main themes in the mass media effects literature, I propose the following formal definition: A mass media effect is a change in an outcome within a person or social entity that is due to mass media influence following exposure to a mass media message or series of messages. Let us examine each of this definition's key terms. It makes most sense to examine these key ideas in this order: social entity, outcome, change, mass media, influence, and exposure. Social entity refers to both an informal collection as well as a formal organization of individuals. Informal collections are typically ad hoc groups that happen to form during a media exposure then dissipate, such as audiences in movie theaters. Formal groups typically have a history, social norms, and structure. This would include audiences for massively multiplayer online computer games that form around mass media messages; it also would include groups that form for nonmedia purposes, such as institutions (especially the family, criminal justice system, educational system, religion, and the political system including elections), cultures, and the public. This definitional element addresses the issue of level, that is, it suggests a range of choices about the unit of measurement (at what level should researchers gather evidence for an effect) and unit of analysis (can the evidence be aggregated and how should it be contextualized to look for patterns of effects). Writing about the phenomenon of communication in general, Chaffee and Berger (1987) said that researchers have typically used one of four levels of analysis: intraindividual, interpersonal, network or organizational, and macroscopic societal levels. They argued that this distinction across levels is important because each level attracts a different kind of researcher, suggests different goals for the research, and relies on different theories as well as different methods. More recently, McLeod, Kosicki, and McLeod (2010) point out that the “number of levels distinguished by social scientists varies from two (e.g., macro/micro) to many. One common scheme used in sociology is a three-level system: macrostructural (social institutions and patterns of social behavior), meso-interactional (relationships among individuals interacting with others), and microindividual (psychological processes and properties of human experience” (pp. 188–189). Clearly, the issue of level is important. With mass media effects, it appears useful to make a distinction across three levels: individual, aggregate of individuals, and human creations. The individual level uses the person as the unit of measurement and unit of analysis. Aggregates of individuals are collections both informal and formal (i.e., task oriented work groups in organizations); measures at this level would include group dynamics, group cohesion, public opinion, public mood, and the like. Human creations would include formal organizations, institutions, and culture; measures at this level would include things like openness (willingness to respond to new information and points of view) and rigidity (fixedness of structures). Thus, this conceptualization uses the basic structure of Chaffee and Berger, but collapses their distinction between interpersonal and organizational into one category called aggregate of individuals. With the mass media rather than communication in general, the Chaffee and Berger distinction between interpersonal and networks seems less important because the media essentially influence individuals who then influence others through their conversations, actions, organizational rules, and so forth. Thus, the media influence on nonexposed people either in interpersonal or organizational interactions is indirect, but still important to study. Outcome requires the consideration of type of effect. Given the concerns for types of effects in the literature, it seems useful to list six types: cognitions, attitudes, beliefs, affects, physiology, and behaviors. Although these six types have come largely from studying effects on individuals, they would seem to have utility as a structure for thinking about effects at other levels of measurement. All of these except for physiology have outcome analogs in the larger social entities. For example with the public, cognitions show up as public knowledge; attitudes show up in public opinion polls as evaluations of public figures and issues; beliefs show up as aggregate public beliefs; affects show up as public mood or public emotions such as grief, fear, or happiness; and behavior shows up as public habits of voting, product purchases, exposure patterns, and so forth. This definitional element addresses the issues of type of effect and level. Change can be in the magnitude or weight of an outcome. Typically, change is regarded as a shift in magnitude of a measure, and when there is no shift, change is not regarded as occurring. However, that type of conceptualization disregards reinforcement effects where there is no shift in magnitude, but where there is a shift in weight. For example, a series of media messages might serve to reinforce a person's existing attitude; thus it appears that no change has taken place because the attitude is the same; however, the weight of the attitude has increased, making it more difficult to alter in the future. This suggests three properties of an effect: kind, magnitude, and weight. To illustrate these properties think of a person's voting preference during a political campaign that is heavily covered in the media through news and ads. At one point in time, the person might favor candidate X, but then over the course of a week shift to favoring candidate Y; this is a change in kind, because there is a shift from one attitude to a different attitude. Or the person might change from a mildly supportive attitude toward candidate X to a strongly supportive attitude toward candidate X; this is a change in magnitude of the attitude from mild to strong. Or a person might have a strongly supportive attitude based on acceptance of a significant other's strongly supportive attitude, but then add reason after reason to support that strong attitude; this is a change in weight of the strongly supportive attitude making it harder and harder to effect a shift in either kind or magnitude. This definitional element addresses the issue of the nature of change. Mass media is a term, like media effect, that scholars have struggled with over the years (see Potter 2009a for a critical analysis of definitions of mass media). Sorting through the definitional issues, I proposed that the “mass media” be defined as organizations that use technological channels to distribute messages for the purpose of attracting an increasingly large audience and conditioning those audiences for repeated exposures so as to increase one's resources such that the enterprise is at least self supporting. This definition has three key classification rules of technological channel, purpose of using the channel, and economic viability. Each of these three qualifications is necessary, but none of them is sufficient (see Potter 2009a for a description of how this definition was developed). Notice that the definition does not focus on channels themselves, because with digitization of messages, the differences across traditional channels (i.e., film, recordings, television, computers) is blurring to a point where they are losing their discriminating power; thus, a list of channels no longer serves as a good ostensive form of definition because each channel can be used in both a mass as well as a nonmass manner. The definition focuses less on channels themselves and more on how they are used. The emphasis is on the sender. In order to be a mass medium, the sender is typically an organization (but can also be an individual) and the sender's main intention is to condition audiences into a ritualistic mode of exposure, that is, the mass media are much less interested in coaxing people into one exposure than they are in trying to get people into a position where they will repeatedly and regularly be exposed to their messages. When an organization uses a technological channel of communication to create and maintain an audience, it is a mass medium. Thus, mass media want to preserve their audiences so that they can maintain their revenue streams and amortize their high initial costs of attracting the audience the first time over the course of repeated exposures. Also, the organization strives to generate resources through subscription fees for access to messages, outright sales of messages, or advertising revenue in order to at least cover all expenses and therefore be self-sustaining or to make a profit. This definition allows for nonprofits to qualify as mass media as long as they can continue to generate enough resources to sustain themselves. Thus, this definition would classify a person's Facebook page as nonmass while classifying Facebook itself as mass media. Influence of the mass media is conceptualized broadly as including both direct and indirect, long term as well as short term, and conscious as well as unconscious. The mass media exert a continual influence on individuals through direct exposure to their messages. The mass media also exert a continual influence on individuals indirectly through other people, groups, and institutions that have been influenced by the mass media. Thus, the challenge in considering mass media influence lies in determining the degree and nature of their influence as it is exerted through a constellation of many nonmedia factors such as a person's states, traits, motives, interpretations of meaning under conditions when individuals are active processors of messages and meaning as well as when they are passive acceptors of media conditioning (see Potter 2009b). Causation is too simple a concept to apply to mass media effect. Causation requires the cause to precede the effect in time, whereas mass media influence is cyclical, thus acknowledging a reciprocal structure where a factor of influence may precede an outcome then in turn be itself influenced by the outcome. Also, causation requires the ruling out of “third variables” beyond the causal and effect variables. But with mass media effects, there are always multiple variables in constant interaction. Therefore, the purpose of effects research is not to establish causation, but rather to calibrate the degree of mass media influence acting in concert with many other variables. Exposure to media influence can be either direct or indirect. Direct influence occurs when people physically encounter a media message. Indirect is when people do not physically encounter a message, but when elements of that message are transmitted to them through other people (through conversations, way of dress, way of acting, etc.), objects (encounters with media advertised products), or procedures (rules of institutions that have been shaped by mass media). Also, exposure can take place in an unconscious as well as a conscious state. Thus, people do not need to be paying attention—or even be aware—of a particular media message or media messages in general in order for them to be exposed to the influential elements in those messages. Furthermore, the outcome need not be intended by either the sender or the receiver. These definitional elements address the issues of nature of influence, pervasiveness, type of media stimulus, and intentionality. This definition is based on two assumptions. The first assumption is that mass media exposures are constant, and thereby media influence is constant. Patterns of thinking, feeling, and behaving are continuously being created, altered, and conditioned by the continual stream of media exposures. Therefore, media effects are ongoing and in a constant state of flux. Some effects occur immediately during a media exposure session, whereas others take a while to show up. Some effects are temporary and dissipate after a few seconds, whereas other effects once they are manifested last a lifetime. This assumption addresses the issue of timing. The second assumption is that some media effects are manifested, whereas others are latent. Latency does not mean that effects do not exist, only that there are effects that cannot be observed at a given point in time. These latent effects in this conceptualization are referred to as baseline effects. This assumption addresses the issue of measurability. As can be seen from the above definition, all the positions on each of the nine conceptual issues are included. Thus, this conceptualization is very broad and does not rule out any perspective in the literature. Implications of this conceptualization There are some interesting implications that follow from this proposed conceptualization of mass media effect in general. First, this conceptualization suggests that there are four general kinds of mass media effects. Second, the conceptualization provides some clear classification rules for what constitutes a mass media effect. A third implication deals with theory and how this conceptualization can be used as an organizing device to structure our understanding of the many theories that have been offered for various parts of the overall phenomenon of mass media effect. A fourth implication concerns how the data from empirical research need to be analyzed to make a case for an effect. Kinds of effects Given the properties and assumptions outlined above, the mass media can exert four possible patterns of influence: gradual long-term change in magnitude, gradual long-term nonchange in magnitude, but change in weight (reinforcement), immediate shift, and short-term fluctuation change. Thus, there are four general kinds, because all mass media effects follow one of these four patterns. These four patterns are illustrated in Figure 1: (a) gradual alteration of a baseline, (b) reinforcement of baseline, (c) sudden alteration of the baseline, and (d) sudden fluctuation from the baseline with a return to the baseline. Let us examine these patterns in a bit more detail. Figure 1 Open in new tabDownload slide Types of media influence patterns. (a) Long-term alteration of baseline; (b) reinforcement of baseline effect; (c) immediate shift of baseline effect; (d) short-term fluctuation. Figure 1 Open in new tabDownload slide Types of media influence patterns. (a) Long-term alteration of baseline; (b) reinforcement of baseline effect; (c) immediate shift of baseline effect; (d) short-term fluctuation. The idea of baseline is key to this conceptualization. At any given point in time, the baseline indicates the relative magnitude of the effect. The baseline reflects the degree of influence not just from the mass media, but also from a constellation of other factors about a person and his/her environment. Thus, the baseline indicates that the media influence does not act alone, but is embedded in a complex of other factors that interact to determine the direction and degree of magnitude of an effect. It serves as an essential context for interpreting the meaning of short-term fluctuation effects. Thus, the idea of baseline helps us move beyond the debate over whether the media in general exert a weak or powerful influence and serves to shift our focus into more productive activities where we calibrate the differential degree of media influence over a wide range of specific effects and across a variety of people. With a long-term change type of effect, the messages from the mass media gradually alter a person's baseline. Figure 1a illustrates this pattern. The line in the figure represents a person's baseline on a particular effect. Over time there is a slow gradual upward slope that indicates an increasing degree of the effect. An example of this would be a cultivation effect where over time a person is more likely to believe the world is a mean and violent place. In contrast with a long-term reinforcement type of effect, the media influence serves to maintain the status quo with the particular effect (see Figure 1b). There is no slope to the baseline—it is flat. An example of this would be when a person continually exposes herself to the same political point of view in magazines, newspapers, television, and the Internet. Her political attitude experiences greater and greater reinforcement, that is, it becomes more and more fixed and hence much more difficult to alter. With a reinforcement effect, the change is in the weight rather than in the magnitude. With an immediate shift type of effect, the media influence serves to alter something in a person during an exposure or shortly after a particular exposure, and that alteration lasts for a relatively long time (see Figure 1c). That alteration may be fairly minor. However, there are times when the degree of change might be fairly dramatic. An example of a dramatic immediate change effect might be when a teenager watches a movie about an attractive person in a particular career—say a heart surgeon—and the teenager decides she wants to be a heart surgeon, talks about this career choice continually, and alters her study habits to earn the grades necessary to go to college and medical school. With a short-term fluctuation change type of effect, the media trigger a fluctuation off the baseline during the exposure or shortly after. The change is short-lived and the person returns to the baseline level quickly (see Figure 1d). This is a fairly prevalent finding in a lot of studies of public information/attitude campaigns and with most laboratory experiments. Researchers find a spike up in knowledge, attitude change, or behavioral intention as a result of exposure to some media material, but this change is not observed in subsequent measurement periods beyond a few days after the exposure. Baselines differ in terms of slope and elasticity. Slope refers to angle (an upward slope indicates a generally increasing level of an effect whereas a downward slope indicates a generally decreasing level of an effect) and degree (a sharp angle reflects a relatively large degree of change in effects level, whereas a flat slope reflects a continuing level in the baseline). Elasticity reflects how entrenched the baseline is. Over time a baseline that has been reinforced continually by the same kind of media messages will become highly entrenched (more weighty) making it less and less likely that there will be fluctuations off the baseline, and when there are fluctuations, those fluctuations are smaller and smaller over time. Fluctuations have three properties: duration, magnitude, and direction. The duration refers to how long the fluctuation lasts before returning to the baseline. Magnitude refers to how far the fluctuation deviates from the baseline. And direction refers to whether the fluctuation moves upward (thus representing an increase in the magnitude) or downward (thus representing a decrease in magnitude). Thus, the baseline provides the essential context for interpreting fluctuation scores. Notice the dotted line in each of the four graphs in Figure 1. These dotted lines represent the manifestation level. In the first graph (a), notice that the baseline remains below the manifestation level. This indicates that the degree of the effect has not reached a level where there are spontaneous observables. By this I mean that the research participants exhibit something that clearly indicates a change that can be attributed to media influence. In two of the other three graphics, there are examples of the baseline breaking above the manifestation level; with those three patterns we have clear manifestations of a media influenced effect. Should we limit our conceptualization of media effects to only those effects where manifestations occur? I would answer no; we should also be sensitive to what occurs below the manifestation level—what can be referred to as baseline effects. Returning to Figure 1a, notice that the baseline has a positive slope which indicates a gradual long-term change. The line does not move above the manifestation level, but something is happening that indicates media influence. For example, let us say a young girl exposes herself to lots of print messages on a particular topic. Over time these exposures gradually increase her reading skills and increase her interest in that topic as her knowledge base grows. Her baseline moves close to the manifestation level. Then one day she picks up an article on the topic, and begins telling all her friends about what she has just learned (this activity takes place above the manifestation level because it is spontaneous and easy to observe her knowledge, attitudes, and emotions as she exhibits them to her friends). However, is it accurate to conclude that this one exposure to the article is responsible for the manifestation? That conclusion is too simple and undervalues the complexity of the media influence process. We must account for the long-term media influence that allowed her to practice her reading skills and grow her interest in this topic. The magnitude of the manifestation level is a combination of the initial level on the baseline and the magnitude of the fluctuation itself. Contrast this with a young boy who did not have this pattern of practicing his reading skills or growing his interest on this topic; his baseline would be far below the manifestation level. If he were to read the same article, he is not likely to manifest the same indicators as did the girl; however, the boy could still have been influenced by his exposure to the article (change in level) although he did not manifest that effect. Baseline effects are changes that take place below the manifestation level. It is likely that most of the mass media influence is in the form of baseline effects rather than manifest effects. If we do not pay attention to baseline effects, we will limit our ability to understand media influence. Baseline effects present a more difficult measurement challenge than do manifestation effects. This challenge parallels the challenge of social scientists who wanted to break through the limits imposed by behaviorism; it was much more difficult to measure attitudes, cognitive states, and other internal hypothetical constructs and make strong cases for their validity compared to behaviors. We are indeed fortunate that social scientists have accepted the challenge of measuring hypothetical constructs. Now it is time for mass media effects scholars to rise to the challenge of more clearly documenting baseline effects in addition to manifesting effects. At this point I must offer a caveat. The graphics in Figure 1 are presented as metaphorical illustrations of the ideas of baseline, fluctuations, and manifestation level. They are not intended to be literal representations. It is unlikely that we will get to a point where we will have developed measurement tools that are capable of generating accurate plots of the magnitude of an effect with high precision. However, these graphics can be useful in guiding our thinking about effect patterns, the need to consider constellations of influence as baselines, and the critical importance of interpreting fluctuations as movements from baselines. Classification rules The conceptualization developed in this essay is very broad. However, we need to be careful not to be overly broad and thus lead us to classify everything possible as a mass media effect; instead, we need classification rules that can also rule certain things out. The proposed conceptualization suggests two classification rules. One classification rule is that there must be a clear outcome that evidences change. Recall that the change can be in magnitude or weight. If there is not a change in either magnitude or weight, there is no mass media effect. The other classification rule is that the media must be demonstrated to have exerted an influence. This influence can be direct or indirect. The direct influence is a fairly obvious one to document. For example, in a two-group experiment where one group is exposed to a media message as a stimulus and the second group is not (the control group), we look for a greater degree of evidence for X effect as a dependent variable across the two groups. If the treatment group exhibits more X than the control group we conclude that the media treatment influenced the difference. If there is no difference in X across the two groups, then we must conclude that there was no direct media influence exerted. The indirect effect is more challenging to document, because it relies less on empirical evidence and more on a reasoned argument. Organize theory development Over the last several decades, scholars have developed a vast array of theories to suggest different mass media effects and to explain how the media exert their influence in bringing about these various effects. For example, Potter and Riddle (2007) conducted an analysis of mass media effects literature published in 16 scholarly journals and found 144 different theories in the 336 articles that featured a theory. How do we organize so many theories? One way is to ask if these theories deal with 144 different mass media effects or if perhaps they all deal with the same effect, but present 144 different competing explanations for the set of factors that bring it about. Or perhaps it is a combination of the two. Until we can apply an organizational scheme to the landscape of theories we cannot know the extent to which we are “inventing the same wheel” over and over or even how many “wheels” we have. The proposed conceptualization of mass media effects can be used as a template for organizing the theories. To illustrate the way this can be accomplished, let us consider eight of the most prominent theories of mass media effects: agenda setting, cognitive capacity, cultivation, framing, media priming, social cognitive learning, third person, and uses and gratifications. When we analyze these theories purely for their effects, we can see that cultivation, agenda setting, social cognitive learning, media priming, and framing all deal with the long-term effect on cognitions, beliefs, attitudes, and behaviors. Third-person theory is a subset dealing with the special condition of conflicting first person and third-person beliefs developed over the long term. Uses and gratifications as an effects theory deals with media usage. Cognitive capacity as an effects theory focuses only on how the human mind processes different amounts of information at a given time during media exposures. So these eight theories deal with essentially three conceptualizations of effects: (a) long-term shaping of cognitions, beliefs, attitudes, and behaviors which are interdependent; (2) long-term shaping of media exposure habits and those habits influence on immediate decisions; and (3) cognitive processing of information during media exposures. There is a good deal of overlap in those conceptualizations. Some are more general (such as cultivation) than others (such as agenda setting) which in turn are more general than others (such as third-person effect). Given the size of the empirical literature as well as the number of theories for mass media effects, the field is in danger of fragmentation where further efforts contribute more to clutter than to systematic advancement of explanation. To avoid this danger we need efforts to organize our past work and present a clearer picture of how our scholarly literature is building strong and clear answers to questions such as: How many types of media effects are there? What is the relative prevalence of different kinds of media effects in society? and Which units of analysis are most susceptible to which effects? The beginning point for such an analysis is a general conceptualization of mass media effect. The general conceptualization presented in this essay can be used to determine the extent to which theories overlap or are nested by categorizing theories according to baseline reinforcement, baseline alteration, temporary fluctuation, or long-term fluctuation. Theories can be further compared according to type of effect: cognitive, attitudinal, belief, affective, physiological, behavioral, or some combination. For theories that are found to focus on the same effect, we can then examine how they differ in their explanatory mechanism, that is, specifying which factors are responsible for bringing about their focal effect. This procedure will result in identifying theories that need to be tested together to determine which explanatory mechanism is most useful—thus eliminating the need for the lesser explanatory systems and directing resources in the direction of the stronger explanatory system. In short, a general conceptual definition of mass media effects provides not only a more clear foundation for the scholarly field, it also provides a means of mapping the scholarship and orienting the field toward a more efficient use of its limited scholarly resources and thus accelerating the movement toward providing compelling answers to the most important questions about our focal phenomenon. Design of empirical research This conceptualization of mass media effect makes a distinction between “effect” and “prevalence of an effect.” This might at first seem like a subtle distinction, but it is a rather important one, especially when we see how the two are continually confounded in typical analyses that focus on mean differences across groups. To illustrate, think of an experiment with 50 participants in a treatment condition and 50 participants in a control condition. Now let us imagine two different scenarios for results. First, let us say that one person in the treatment condition has a high score on the effects outcome variable, but when that score is averaged together with the outcome scores for the other 49 participants in the treatment condition, the resulting mean score is no different (statistically) than the mean score on the outcome variable in the control group. In this case, the experimenter concludes the treatment had no effect, but this is a faulty conclusion because the treatment did have an influence, albeit on only one participant. Second, let's say that 15 people in the treatment condition exhibited a very high outcome score and another 15 people exhibited a very low outcome score, but the average mean for the treatment group was no different (statistically) from the mean of the control group. In this case, the experimenter concludes that the treatment had no effect; again this is a faulty conclusion. Given the proposed conceptualization, there were effects observed in both scenarios described above, but because the effects were not prevalent enough in the first scenario, the experimenter concluded there was no effect. In the second scenario the effect was not prevalent enough in only one direction, so the experimenter concluded that there was no effect. Because prevalence is confounded with effect, the experimenter drew a faulty conclusion from his/her results. This proposed conceptualization of “effect” requires a sharp focus on the unit of analysis (i.e., the individual, the group, the institution, etc.) then looking for evidence of a change (in magnitude or weight) over some period of time—at minimum before and after a particular exposure to a media message. The statistical procedure of analysis of variance is too crude a tool to illuminate effects because it aggregates outcome scores into group means, and by so doing it elevates the importance of prevalence beyond the importance of change. In contrast, the proposed conceptualization requires the computation of change scores as the outcome measure then using another statistical procedure such as multiple regression to explain the change variation. With this proposed conceptualization of mass media effects, the examination of media influence is focused on changes in the unit of analysis over time rather than in computing averages across grouped units at one point in time. Conclusion This essay addresses the need to develop a conceptualization of mass media effect in general. The synthesized conceptualization incorporates a full range of elements across nine definitional issues of type of effect, level, change, influence, pervasiveness, media stimulus, intentionality, time, and measurability. This proposed conceptualization of mass media effects can be useful in three areas. It can clarify the essence of mass media effects by focusing attention primarily on four general kinds of effects. It can provide classification rules that serve as a formal definition. It can serve as a template for helping to organize the great many of theories that have been developed over the years to explain mass media effects. And it can shift the research focus away from comparing group means and onto analyzing change and then explaining the role that mass media exert in bringing about that change. References Baran , S. J. , & Davis , D. K. ( 2000 ). Mass communication theory: Foundations, ferment, and future . Belmont, CA : Wadsworth/Thomson . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Basil , M. D . ( 1992 , May). A ‘new world’ of media effects. Paper presented at the 42nd Annual Conference of the International Communication Association, Miami, Florida. Bryant , J. , & Thompson , S. ( 2002 ). Fundamentals of media effects . Boston, MA : McGraw Hill . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bryant , J. , & Zillmann , D. ( 2009 ). A retrospective and prospective look at media effects. In R. L. Nabi & M. B. Oliver (Eds.). The Sage handbook of media processes and effects (pp. 9 – 17 ). Los Angeles, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chaffee , S. H. , & Berger , C. R. ( 1987 ). What communication scientists do. In C. R. Berger, & S. H. Chaffee (Eds.), Handbook of communication science (pp. 99 – 122 ). Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Grossberg , L. , Wartella , E., & Whitney , D. C. ( 1998 ). Mediamaking: Mass media in a popular culture . Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Harris , R. J . ( 2004 ). A cognitive psychology of mass communication (4th ed.). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hovland , C. I . ( 1954 ). Effects of the mass media of communication. In G. Lindzey (Ed.), Handbook of social psychology (pp. 1062 – 1103 ). Reading, MA : Addison-Wesley . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Iyengar , S . ( 1997 ). Overview: The effects of news on the audience. In S. Iyengar & B. Reeves, (Eds.) Do the media govern? (pp. 211 – 216 ). Thousand Oaks, CA : Sage Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jeffres , L. W . ( 1997 ). Mass media effects (2nd ed.). Prospect Heights, IL : Waveland . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lazarsfeld , P. F . ( 1948 ). Communication research and the social psychologist. In W. Dennis (Ed.), Current trends in social psychology (pp. 218 – 273 ). Pittsburgh, PA : University of Pittsburgh Press . Littlejohn , S. W . ( 1999 ). Theories of human communication (6th ed.). Belmont, CA : Wadsworth . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McDonald , D. G . ( 2004 ). Twentieth century media effect research. In J. D. H. Downing, D. McQuail, P. Schlesinger & E. Wartella (Eds.), The Sage handbook of media studies (pp. 183 – 200 ). Thousand Oaks, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC McGuire , W. J . ( 1986 ). The myth of massive media impact: Savaging and salvagings. In G. Comstock (Ed.), Public communication and behavior (Vol. 1, pp. 173 – 257 ). Orlando, FL : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McLeod , J. M. , Kosicki , G. M., & McLeod , D. M. ( 2010 ). Levels of analysis and communication science. In C. R. Berger, M. E. Roloff, & D. R. Roskos-Ewoldsen (Eds.), The handbook of communication science (2nd ed., pp. 183 – 200 ). Los Angeles, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC McLeod , J. M. , & Reeves , B. ( 1980 ). On the nature of media effects. In S. B. Withey & R. P. Abeles (Eds.), Television and social behavior: Beyond violence and children (pp. 17 – 54 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McQuail , D . ( 2005 ). McQuail's mass communication theory (5th ed.). London, England : Sage Publications . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Morley , D . ( 1992 ). Television, audiences, and cultural studies . London, England : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nabi , R. L. , & Oliver , M. B. ( 2009 ). Introduction. In R. L. Nabi & M. B. Oliver (Eds.). The Sage handbook of media processes and effects (pp. 1 – 5 ). Los Angeles, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Perse , E. M . ( 2001 ). Media effects and society . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Potter , W. J . ( 2009a ). Arguing for a general framework for mass media scholarship . Thousand Oaks, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Potter , W. J . ( 2009b , November). Conceptualizing mass media effect. Paper presented to the Communication and Social Cognition Division of the National Communication Association, Chicago, IL. Potter , W. J. , & Riddle , K. ( 2007 ). A content analysis of the media effects literature . Journalism & Mass Communication Quarterly , 84 , 90 – 104 . Google Scholar Crossref Search ADS WorldCat Roberts , D. F. , & Maccoby , N. ( 1985 ). Effects of mass communication. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology (Vol. 2, 3rd ed., pp. 539 – 599 ). New York, NY : Random House . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Severin , W. J. , & Tankard , J. W. , Jr. ( 2001 ). Communication theories: Origins, methods, and uses in the mass media (5th ed.). New York, NY : Longman . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sparks , G. G . ( 2006 ). Media effects research: A basic overview (2nd ed.). Belmont, CA : Thomson Wadsworth . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wartella , E. , & Reeves , B. ( 1985 ). Historical trends in research on children and the media: 1900–1960 . Journal of Communication , 35 ( 2 ), 118 – 133 . Google Scholar Crossref Search ADS WorldCat Weimann , G . ( 2000 ). Communicating unreality: Modern media and the reconstruction of reality . Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Weiss , W . ( 1969 ). Effects of the mass media of communication. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology (Vol. 2, 2nd ed., pp. 77 – 195 ). Reading, MA : Addison-Wesley . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2011 International Communication Association