Gatekeeping and the Negativity Bias
Summary and Keywords
Research on media gatekeeping is focused on the factors leading to a distribution of information in media content that is systematically different from the “real world.” Early gatekeeping work examined editorial decisions, and emphasized the effect that a single editor’s preferences and beliefs could have on the content new consumers receive. The literature has gradually shifted to focus on more generalizable factors, however. These include organization-level assessments of newsworthiness and commercial/economic considerations; broader system-level factors including the impact of dominant ideologies and political and social norms; and common individual-level factors, including a range of cognitive and psychological biases.
The tendency for humans to prioritize negative over positive information is one such cognitive bias—and the growing literature on the negativity bias is discussed here as one example of a set of organization-, system-, and individual-level “gates” that have a systematic impact on news content. Negativity is just one example, however. Sensationalism, violence, geographic proximity, availability of visuals, prominence of celebrities—all of these tendencies in media content can and have been examined effectively using the gatekeeping metaphor. Some of this work is reviewed here, alongside some recent trends in gatekeeping work, including the “distributional” approach to gatekeeping, and the shift in gatekeeping brought on by the “new media” environment.
The aim of this review is to synthesize recent work in political communications focused on gatekeeping. The literature on media gatekeeping is vast, and there have already been several valuable reviews (e.g., Shoemaker & Vos, 2009). The initial purpose here is thus to outline the general approach and then to provide some illustrative examples of some organization-, system-, and individual-level factors associated with media gatekeeping. Sections then turn to related issues. A first section outlines an extension of individual-level analyses that draws a link between gatekeeping theory and work on the “negativity bias”—a salient issue in the study of political communication, and one that can be viewed in part as a version of media gatekeeping driven by our conscious and unconscious psychological tendencies. A concluding section discusses the implications of technological change for media gatekeeping. We begin, however, with a description of both the classic and more recent accounts of what exactly media gatekeeping is.
The classic media gatekeeping analysis is David Manning White’s (1950) case study of a wire editor at a small-town daily newspaper. The gatekeeping metaphor itself predates White’s work (Lewin, 1943), but White’s application of the idea to media content was powerful and, it turns out, very influential. The approach was simple: White catalogued the news stories provided by wire services, noted the ones that did (and did not) find their way into the newspaper, and then explored the editor’s reasons for including or excluding stories.
White’s work makes very clear the effect that a single editor’s preferences and beliefs can have on media content. The results were seminal, to be sure; but they likely came as a shock to no one. We all know that journalists and editors select content from a wide range of possible stories. We also suspect that their selections are systematically biased. Some editors and journalists will be interested in the economy, others in foreign affairs, others in local stories; some will seek out sensational stories, others will focus on local events, and so on. The end result is that the distribution of news content that most of us consume on a daily basis will be systematically different from the distribution of potential news content out there in the “real world.”
This difference, driven by a wide range of selection mechanisms, is the focus of media gatekeeping studies.1 Media gatekeeping is the information selection process whereby news editors, or journalists, or webmasters—or any other person or institution involved in the dissemination of information—makes decisions about which information is taken into account and which information is ignored. Every step of an information dissemination process involves a “gate” through which information passes, or does not. The task of gatekeeping research has been to understand how those gates function. In so doing, gatekeeping studies aim to describe the difference between “news” and “reality.”2
A Distributional Approach to Media Gatekeeping
Gatekeeping analyses are at their root focused on the processes that result in differences in distributions of information. There is on the one hand a distribution of information out there in the real world, RW. There is on the other hand a distribution of information in news content, M. Regardless of whether studies rely on case studies of individual editors or large-scale quantitative analyses, comparing RW and M, and understanding how and why they differ, are the central goals of media gatekeeping analyses.
Consider the top panel of Figure 1 (adapted from Soroka, 2012), which shows a hypothetical probability density function of some real-world phenomenon distributed across Dimension X. Dimension X could be any characteristic relevant to news selection. It could be sensationalism, or importance to the national economy, or violence in crime, or geographic proximity, or availability of visuals, or the prominence of movie stars. Regardless, imagine that all the information out there in the world is ranked along one of these dimensions, from low to high. The top panel of Figure 1 shows a hypothetical distribution of that information in the real world. There is some information to the left of this dimension, and some information to the right. Most information is somewhere in the middle.
The bottom panel of Figure 1 also shows a distribution of information. Here too, information is distributed across Dimension X, but in this case we see only the information that finds its way into media content. If the news selection process is entirely random, then M will be identical to RW. Put differently, if news content presents a perfectly representative selection of all the available information, our news stream (M) will look very similar to reality (RW). As has been noted, gatekeeping analyses are interested in the degree to which this is true—or, rather, the degree to which this is not true. Our news stream is not very similar to reality—it is systematically different. How is it different? Why is it different? These are questions that motivate gatekeeping analyses.
In Figure 1, the real world leans to the left side of Dimension X but media content leans to the right. This difference between M and RW implies a selection mechanism—a likelihood of story selection based on whether the information is to the left or right of Dimension X. In Figure 1, the selection mechanism that would convert the distribution in RW to the distribution in M is shown in the middle panel.3 In order to produce the right-leaning distribution in M from the left-leaning distribution in RW, we need a selection mechanism that pays more attention to information to the left of Dimension X; in this example, we need a selection mechanism that pays roughly twice as much attention to information on the left of X than it does to information on the right. This is the gatekeeping “function”—it is the selection mechanism that produces the representation of the real world that finds its way into media content.
The value of Figure 1 is largely theoretical—we cannot often directly observe entire distributions of information and calculate gatekeeping functions, of course (though see one effort in Soroka, 2012). But we may be well served to think about gatekeeping in this broadly generalizable way. For instance: there is an actual distribution of criminal acts, and there is a distribution of the criminal acts that find their way into news content. How do these distributions differ? Past work suggests that media coverage tends to give priority to more violent crime (Altheide, 1997). The relevant Dimension X in this case is violence. And the gatekeeping function for crime stories, due to a range of factors, appears to prioritize more rather than less violent crime stories.
An Example: Gatekeeping in U.S. Primaries
Meyrowitz’s (1994) study of the “almost invisible candidate” offers a particularly clear example of the ways in which news content offers a filtered, or “gated,” version of reality. The story in this case centers on Larry Agran, a candidate in the 1991 Democratic presidential primary. Objectively speaking, Agran seemed as capable a candidate as the other early participants. Meyrowitz argues that media coverage nevertheless systematically excluded Agran from the race. An early photo of four candidates offers a powerful illustration of the dynamic: Tsongas, Harkin, and Clinton appear in the New York Times, with Larry Agran (standing just to the right) cropped out of the picture.
Meyrowitz notes that, “With paradoxical logic, Agran was told by news media executives that he had not earned the right to media exposure, because, among other things, he had not received enough media exposure” (Meyrowitz, 1994, p. 94). This popularity-based narrowing of the field is in fact not unique to the 1991 primary season. The 2016 Republican primary features too many candidates to easily include in debates, at least in the way that televised debates are typically formatted. Fox News has accordingly set the following limit for the first debate: candidates must “place in the top 10 of an average of the five most recent national polls …”4 This decision led to a good deal of comment and debate, of course—it effectively removed the possibility that a lesser-known candidate could find an audience early in the primary season; it also excluded a number of quite viable and well-known contenders. Polling results were used in this instance as a way of defining newsworthiness, and only the most popular candidates could take part in the broadcast.
Of course, perceived popularity is likely a dimension on which many pieces of information—political and otherwise—are judged, and providing media consumers mainly with information in which they have already shown some interest is deeply problematic. How can we be fully informed when media are structured in a way that limits novel information? These accounts of media decisions in the context of U.S. party primaries make abundantly clear the potentially negative consequences of this form of gatekeeping. There are many other potential “gates,” however.
Factors Affecting News Selection (What Is Dimension X?)
What are the factors that drive news selection? Put differently, what information is most likely to find its way into mainstream news content? White’s (1950) account focuses on the biases of one individual editor; in that account, Dimension X is personal or editorial interest.
The real strength of gatekeeping analyses comes from the realization that there are systematic biases in news selection processes, however. There is variance in news selection across individual editors and journalists, to be sure. But many of the factors affecting news selection—indeed, many of the main factors affecting news selection—are common across individuals. They are institutional, or ideological, or sociological. The identification of these common factors—of the more generalizable Dimension Xs—has been the central contribution of work on media gatekeeping.
State-of-the-discipline reviews by Shoemaker and colleagues offer valuable summaries of the many different “gates” in gatekeeping research. The most recent of these, Shoemaker and Vos (2009), offers a detailed account of the literature that distinguishes between five levels of gatekeeping analysis: individual-level, communication routine–level, organizational-level, social institutional–level, and social system–level. Much of the related literature in each of these subfields is cited there, and the process will not be repeated here. Rather, a few illustrative examples are presented, collapsed into just three broad categories: organizational-level, system-level, and individual-level.
Organization-level factors here include a combination of what Shoemaker and Vos (2009) refer to as communication routine- and organizational-level analyses. There are advantages to distinguishing between the two categories, but there also is a good deal of overlap between them. Both focus on the nature of the daily news-producing environment—the typical processes by which journalists gather information, for instance, and the many characteristics of a news-making organization.
Berkowitz (1990) offers a valuable example of a gatekeeping analysis highlighting organization-level factors. His approach is not very different from White’s—the paper is based on in-person observation of decisions made in the newsroom of an Indianapolis network affiliate. Berkowitz’s results suggest a rather different story from White’s, however. In this television newsroom story selection “seemed to be a group process; content, therefore, was shaped by group dynamics” (p. 66); there were multiple gates, rather than a single editorial decision; and priorities and constraints that were particular to this organization (specific budgeting concerns, for instance) affected news decisions as well. In sum, Berkowitz’s gatekeeping account highlights the importance of the nature and structure of a news organization in media gatekeeping.
Shoemaker, Eichholz, Kim, and Wrigley (2001) offer a similar account. Their analysis of news coverage of 50 congressional bills points less to the importance of individual journalists’ priorities, and more to the impact of “routinized practices of journalism” (p. 235). Routinized practices are captured in this case through survey-based assessments of “newsworthiness,” by the political editors of 40 newspapers. Analyses suggest that the routinized component makes a much larger difference to story coverage than does a range of journalist-specific ideological, demographic, or experiential variables. This is evidence, the authors suggest, of the relative significance of a range of newsworthiness-related assessments inherent not in news journalists so much as in news organizations.
Of course, assessments of newsworthiness may adhere not so much to individual news organizations as to the practice of journalism as a whole. One organization’s decision-making structure, or the costs associated with moving one salaried journalist from one country to another—these are quite clearly related to a single new organization. Ideas about what constitutes “news” could be as well—different organizations may have different ideas about what is worth publishing, but there are also good reasons to believe that elements related to the economics of news-making, or definitions of newsworthiness, are common across news organizations. This is the possibility investigated in system-level analyses.
The consideration of system-level factors in gatekeeping analyses serves to link what are typically more narrowly conceived views of gatekeeping with much broader literatures on journalistic, political, and social norms. Indeed, much of the literature that could be regarded as system-level gatekeeping research does not self-define as analyses of gatekeeping at all. Consider work by Johnstone, Slawski, and Bowman (1972) focusing on the way in which journalists’ belief systems structure news selection and reporting. Consider also Gans’s (1979) path-breaking work on the impact of journalistic values and norms, or Herman and Chomsky’s (1988) “propaganda model” of mainstream news content. None of these studies is focused on gatekeeping per se, but all capture a range of broad, system-level factors—the impact of markets and advertisers, for instance, or of professional, social, and political norms—that structure the selection of news content.
The same is true for “new institutionalist” theory in political communications. Research in this area focuses on the ways in which institutions, defined broadly to include journalists’ practices, values, and routines, affect the production of news content. The seminal studies—Cook (1998) and Sparrow (1999)—deal not just with the impact of current institutions but also the timing and evolution of (and equilibrium in) those institutions, due to a range of sociological and economic factors. These new-institutionalist accounts focus on factors such as work routines, economic imperatives, and journalistic norms on news content.
Similar interests are intermittently evident in work that is focused more explicitly on gatekeeping. Relly, Zanger, and Fahmy (2015) explore gatekeeping by Iraqi journalists in part as a function of their commitment to or experience with democratic norms, for instance; Bennett’s (2004) account of gatekeeping includes market-driven factors alongside individual and organizational ones. This research signals the connection between media gatekeeping and a much broader body of system-level work. At the same time, this work also points toward a more traditional focus of gatekeeping analysis: the individual editor or journalist.
The original media gatekeeping study by White (1950), discussed previously, stands as a classic individual-level analysis of gatekeeping. There have been a good number of studies to follow White’s approach (e.g., Bleske, 1991). This work highlights, at an individual level, the broad range of considerations that go into the editorial and journalistic decisions structuring the content of mainstream news. We are able to observe systematic tendencies in a given editor’s interests and priorities—that is, we can observe that a given editor prioritizes news that is more geographically proximate, violent, visually interesting, critical (or not critical) of government, and so on.
Most individual-level gatekeeping work focuses on the preferences of editors and journalists. Where do these preferences come from? Some preferences are viewed as reflecting a range of organization- or system-level factors—journalists’ objectives, or audience interests, for instance. A limited but growing body of work has considered a range of psychological or cognitive factors underlying gatekeeping as well. Stocking and Gross’s (1989) How Do Journalists Think? offers one of the first arguments for the potential importance of “cognitive bias” in media gatekeeping. “Research has shown, for example, that people, even when instructed to be objective, tend to seek and select information in ways that confirm their initial beliefs. Other research has highlighted common distortions in memory, including a tendency to distort memories of an event to conform to new information about the event” (p. 4). In these and other ways, Stocking and Gross argue, cognitive biases matter to the ways in which journalists select and report the news. Donsbach’s (2004) work, “The Psychology of News Decisions,” reflects a similar focus; as does Henningham’s (1997) research exploring extroversion and neuroticism in Australian journalists.
Note that these examples focus on a range of broadly generalizable behaviors. These factors are cognitive or psychological, although they may overlap with organizational and systemic factors as well (i.e., Do extraverted individuals become journalists, or does the practice of journalism encourage extraversion?). It is in fact relatively rare for gatekeeping analyses to fall into just one of the three categories used here. A single “gate” may adhere primarily to an organization, or a system, or an individual, but gatekeeping is frequently described as a function of factors that spill across these categories. This is evident in the examples previously reviewed; it is also reflected in the connections between work on media gatekeeping and work on negativity biases.
The Negativity Bias
The prevalence of negative news content is one of the most frequently discussed—and frequently criticized—aspects of modern political communications. There is little doubt that media content is predominantly negative, but this negativity bias in news content is typically discussed independently from considerations of gatekeeping. There is no reason for this to be the case. Indeed, discussions of negativity biases reflect the same kinds of “gates” discussed previously. There is a good deal of work focused on the way in which journalist norms and procedures, or economic incentives, encourage increasing levels of negativity in news content, after all (e.g., Cappella & Jamieson, 1997; Patterson, 1994). Much of what the literature describes as driving negativity in news content falls quite easily into organization- and system-level media gatekeeping frameworks.
The same is true for work on negativity as a function of individual-level psychological and cognitive factors. Spurred on by a massive body of work on negativity biases in psychology (for a review, see Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001), there is a small but growing body of work that considers negativity in media content as a function of the tendency for both journalists and audiences to prioritize negative information over positive information (Shoemaker, 1996; Soroka, 2014). The “gate” in this instance is the human brain—or, at least, tendencies in human information processing and decision-making. There are a numbers of advantages—quite possibly evolutionary—to paying especially close attention to negative information. (The typical account emphasizes the fact that the potential consequences of a piece of negative information often outweigh the potential benefits of a piece of positive information.) The consequent gatekeeping effect is a systematic difference in the degree of negativity in the real world (RW) and in media content (M).
Consider an account of negativity in news content that is more firmly rooted in ideas about media gatekeeping. What accounts for negativity in news content? Past work (cited previously) suggests a number of factors. System-level gates include conceptions within journalism about what constitutes “news,” ideas (as well as evidence) about what audiences are interested in, and sometimes-fierce competition for audiences among commercial media outlets. Organization-level gates include the priorities of particular news outlets (sometimes conditioned by system-level factors) such as an interest in local crime reports, or a belief that an audience is particularly interested in stores that are sensationalized or rich in colorful visuals. Individual-level factors include the preferences of individual editors or journalists, but also the more general human tendency to prioritize negative information over positive information. Of course, these individual-level factors are part of what drives system- and organizational-level ones and vice versa. And the end consequence of this multilevel gatekeeping is a distribution of information in media content that is systematically more negative than reality.
Gatekeeping and Technological Change
Are there reasons to believe that any of this will change alongside media technology? There clearly are technological elements relevant to gatekeeping, even in the more traditional organizational, systemic and individual-level approaches. The factors that editors regard as important to newsworthiness (i.e., good images), the structure of a newsmaking organization (i.e., geared toward weekly, or daily, or hourly output), the number of “gates” in a news production process (i.e., at a small newspaper versus a national broadcaster) are all ways in which the nature of communications technologies have been seen as important to gatekeeping processes. (Also consider formatting considerations in Meyrowitz’s account of Larry Agran, cited previously.)
That said, recent technological change has led to significant shifts in how news can be disseminated, and indeed in who can produce it. One consequent topic of interest has been the way in which news content might shift in light of the ready availability of high-quality photos and videos from almost anywhere—Livingston and Bennett (2003) argue that new technology has facilitated a rise in “event-driven” news content, for instance. Bro and Wallberg (2014) consider the ways in which the use of Facebook in mainstream media changes who has control over gatekeeping; Singer (2014) makes a similar argument regarding the role of reader-produced content (i.e., discussion) on digital news websites. Williams and Delli Carpini (2004) make the broader argument that the new media environment—in which more and more people are able to distribute news independently—increasingly reduces the power of traditional gatekeepers.
Whether new technologies, and especially social media, are fundamentally changing the nature of gatekeeping is unclear. There still is gatekeeping, after all—and a range of organizational, systemic, and individual-level factors that drive decisions about the information that is disseminated, or ignored. It may be that the increasing volume of user-provided content decreases the importance of journalist norms and news-organizational factors, while increasing the relevance of individual-level psychological tendencies. This seems likely, though it is pure conjecture at this stage. Insofar as information passes through a series of gates to reach a large audience, media gatekeeping is as relevant a theory as it was in 1950.
In short, it is the still the case that the distribution of information in media content (regardless of medium) is different from the distribution of information in reality. Understanding the nature and magnitude of this difference—and by implication, the tendencies in news selection within organizations and individuals—is the central objective of work on media gatekeeping.
Altheide, D. L. (1997). The news media, the problem frame, and the production of fear. Sociological Quarterly, 38, 647–668.Find this resource:
Barzilai-Nahon, K. (2008). Toward a theory of network gatekeeping: A framework for exploring information control. Journal of the American Society for Information Science and Technology, 59(9), 1493–1512.Find this resource:
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psychology, 5(4), 323–370.Find this resource:
Bennett, W. L. (2004). Gatekeeping and press-government relations: A multigated model of news construction. In L. L. Kaid (Ed.), Handbook of political communication research (pp. 283–313). Mahwah, NJ: Lawrence Erlbaum.Find this resource:
Berkowitz, D. (1990). Refining the gatekeeping metaphor for local television news. Journal of Broadcasting & Electronic Media, 34(1), 55–68.Find this resource:
Bleske, G. L. (1991). Ms. Gates takes over. Newspaper Research Journal, 12(4), 88–97.Find this resource:
Bro, P., & Wallberg, F. (2014). Digital gatekeeping. Digital Journalism, 2(3), 446–454.Find this resource:
Cappella, J. N., & Jamieson, K. H. (1997). Spiral of cynicism: The press and the public good. New York: Oxford University Press.Find this resource:
Cook, T. E. (1998). Governing with the news: The news media as a political institution. Chicago: University of Chicago Press.Find this resource:
Donohue, G. A., Tichenor, P. J., & Olien, C. N. (1972). Gatekeeping: Mass media systems and information control. In F. G. Klein & P. J. Tichenor (Eds.), Current perspectives in mass communication research (pp. 41–70). Beverly Hills, CA: SAGE.Find this resource:
Donsbach, W. (2004). Psychology of news decisions: Factors behind journalists’ professional behavior. Journalism, 5(2), 131–157.Find this resource:
Gans, H. J. (1979). Deciding what’s news: A study of CBS Evening News, NBC Nightly News, Newsweek, and Time. New York: Pantheon Books.Find this resource:
Henningham, J. (1997). The journalist’s personality: An exploratory study. Journalism and Mass Communication Quarterly, 74(3), 615–624.Find this resource:
Herman, E. S., & Chomsky, N. (1988). Manufacturing consent: The political economy of the mass media. New York: Pantheon Books.Find this resource:
Johnstone, J. W. C., Slawski, E. J., & Bowman, W. W. (1972). The professional values of American newsmen. Public Opinion Quarterly, 36(4), 522–540.Find this resource:
Lewin, K. (1943). Forces behind food habits and methods of change. Bulletin of the National Research Council, 108, 35–65.Find this resource:
Livingston, S., & Bennett, W. L. (2003). Gatekeeping, indexing, and live-event news: Is technology altering the construction of news? Political Communication, 20(4), 363–380.Find this resource:
Meyrowitz, J. (1994). The (almost) invisible candidate: a case study in news judgement as political censorship. In M. Aldridge & N. Hewitt (Eds.), Controlling Broadcasting: Access Policy and Practice in North American and Europe (pp. 93–107). London: Manchester University Press.Find this resource:
Patterson, T. E. (1994). Out of order. New York: Vintage Books.Find this resource:
Relly, J. E., Zanger, M., & Fahmy, S. (2015). Democratic norms and forces of gatekeeping A study of influences on Iraqi journalists’ attitudes toward government information access. Journalism & Mass Communication Quarterly, 92(2), 346–373.Find this resource:
Shoemaker, P. J. (1996). Hardwired for news: Using biological and cultural evolution to explain the surveillance function. Journal of Communication, 46(3), 32–47.Find this resource:
Shoemaker, P. J., Eichholz, M., Kim, E., & Wrigley, B. (2001). Individual and routine forces in gatekeeping. Journalism & Mass Communication Quarterly, 78(2), 233–246.Find this resource:
Shoemaker, P. J., & Vos, T. P. (2009). Gatekeeping theory. New York: Routledge.Find this resource:
Singer, J. B. (2014). User-generated visibility: Secondary gatekeeping in a shared media space. New Media & Society, 16(1), 55–73.Find this resource:
Soroka, S. N. (2012). The gatekeeping function: Distributions of information in media and the real world. Journal of Politics, 74(02), 514–528.Find this resource:
Soroka, S. N. (2014). Negativity in Democratic politics: Causes and consequences. New York: Cambridge University Press.Find this resource:
Sparrow, B. H. (1999). Uncertain guardians: The news media as a political institution. Baltimore: Johns Hopkins University Press.Find this resource:
Stocking, S. H., & Gross, P. H. (1989). How do journalists think? A proposal for the study of cognitive bias in newsmaking. Bloomington, IN: ERIC Clearinghouse on Reading and Communication Skills, Smith Research Center, Indiana University.Find this resource:
White, D. M. (1950). The “gate keeper”: A case study in the selection of news. Journalism Quarterly, 17(4), 383.Find this resource:
Williams, B. A., & Carpini, M. X. D. (2004). Monica and Bill all the time and everywhere: The collapse of gatekeeping and agenda setting in the new media environment. American Behavioral Scientist, 47(9), 1208–1230.Find this resource:
(1.) Note that although my focus here is on gatekeeping in mass media, there is a good deal of work on gatekeeping that is not focused on media. For a particularly useful cross-disciplinary review see Barzilai-Nahon (2008).
(2.) It is worth distinguishing between work that focuses on the selection of one event or another and work interested in the selective framing of a single event. Some gatekeeping work views selection and framing in tandem (e.g., Donohue, Tichenor, & Olien, 1972). The somewhat more parsimonious view, however, and the one adopted here, is that gatekeeping theory focuses on the selection mechanism, and that issue framing is a different, albeit sometimes related, issue.
(3.) Note that mathematically, the selection mechanism is identified by dividing M by RW.