Since the earliest days of mass media, researchers, social critics, politicians, and the general public have been concerned about the extent to which media representations reflect or deviate from “reality.” Over the years, a great deal of research and public debate have revolved around the kinds of images of the world that are created and disseminated by media, and how they compare to the “real world” as revealed by official statistics or other objective indicators.
Reality As Social Reality
The first problem research in this tradition must confront is the question of what constitutes “reality,” which is a topic of longstanding philosophical deliberation. Similarly complex is the question of how – and even whether – we can comprehend reality. Many theorists argue that humans construct what is perceived (and treated) as reality through social, cultural, and psychological mechanisms and structures. In a sense, this reflects Shakespeare’s notion in Hamlet that “There is nothing either good or bad, but thinking makes it so” (act II, scene ii). Berger & Luckman (1966) argued that reality is knowable only as a mediated phenomenon, and that this is always ultimately a social process. Through intersubjectivity, we share a sense of “everyday reality” with others, but this is socially and culturally constructed.
There is thus a fundamental debate over whether media and reality can be meaningfully compared; if reality is unknowable, then attempting to compare it to media is a futile enterprise (Schulz 1976). Given that, some researchers, rather than comparing news media to some standard of “reality,” have examined social, institutional, and psychological explanations for how journalists make decisions about what to cover (or not cover), and how (Donsbach 2004).
On the other hand, even if reality does not exist outside of human (cultural) construction and interpretation, and is constructed (rather than discovered) through human investigation and manipulation of symbol systems, many researchers still believe that the world represented in media can be compared to certain facts about life and society. From that perspective, the task is to determine the most useful and reliable indicators of how the media world deviates from observable structural parameters, or to compare mediated representations and unmediated experiences of the “same” event or phenomenon (Donsbach 2003).
For example, in a series of studies, Kepplinger compared the coverage of the oil supply in Germany with data on actual reserves (1979); media coverage of air and water quality with real-world biological measurements (1992); and the political activities of the German parliament with the coverage of politics by the emerging media in the postwar period (2002). Studies in this vein often show sharp disjunctures between “reality” and media coverage.
Media can also construct multiple (and often conflicting) realities. For example, comparative international analyses have shown how the US and “others” are presented by the press in different countries in the context of the “war on terror” and the war in Iraq (Nohrstedt & Ottosen 2005), or how media in Israel and Jordan each portrayed the peace process (Wolfsfeld et al. 2002).
On the other hand, the very “unreality” of media can also have a powerful impact on our sense of reality. As Fiske (1987, 21) argues, television is “an essentially realistic medium because of its ability to carry a socially convincing sense of the real.” In Brazil, a telenovela can provide interpretive frames that shape how viewers perceive contemporary political events (Porto 2005). As Pearson (2005, 406) notes, Mexicans are fond of saying “Life is like a telenovela,” to the point that “the line between the world and a world is often difficult to distinguish.”
Although many scholars concerned with the correspondence of media to reality tend to focus on news, for purposes of this discussion, the distinctions among news, scripted programs, and “reality television” are not especially relevant. At a more general level, media themselves can constitute a way of knowing (Chesebro 1984). Media representations are “real” in the sense that dreams, stories, legends, and rumors are real – they exist as phenomenological narratives and representations. That is, a statement about an event is not the event itself, but it is nonetheless itself an event.
Comparative measures of media and reality allow researchers to see how closely media stories reflect the facts of society and provide a basis for follow-up studies in different media, and/or in other societies, and/or over time. There is no expectation of any particular correspondence between reality and media reality; the key theoretical and empirical task is to illuminate specific and systematic discrepancies in order to better understand media institutions and to provide a basis for further inquiry into how media images inform our constructions of social reality.
Large portions of what we know (or think we know) are based not on first-hand experience, but on media representations of life, society, groups, and institutions. Researchers in the cultivation analysis tradition, for example, point out that most people have limited, if any, experience of places such as courtrooms, police stations, prisons, or hospitals, but that we have extensive and vivid images about what transpires in such locations, as well as about the sorts of people who work in them. Media provide us with a vast range of representations of things about which we have no direct knowledge, and these account for many of our “intersubjective” beliefs.
Analyses Of Media Content
Systematically coded, quantitative content analysis has been frequently employed to illuminate how media construct different aspects of reality. An online bibliography of content analyses listed in Communication Abstracts between 1990 and 1997 features 428 separate entries (Neuendorf 2000).
These studies cover an immense range of topics. Studies comparing reality and media reality have examined issues as diverse as the portrayal of persons over 50 in television commercials; alcohol and tobacco use in daytime soap operas; news coverage of infectious diseases; women scientists in popular magazines; sex and contraception in prime-time programs; art and artists on network television news; television’s messages about the environment; the image of journalists on prime-time television; and hundreds more.
The number of areas in which the “real world” and the media world can be compared is virtually boundless. Entering the phrase “media portrayals of” into the Google search engine will produce tens of thousands of hits, with links to a broad range of articles and sites that examine media representations. Researchers have explored the correspondence between reality and media reality in relation to dozens of wide-ranging topics, including girls and women, ethnic minorities, weight loss surgery, bipolar disorder, terrorism, hate crimes, sex in the workplace, sports, suicide, poverty, aging, and many more.
In the 1930s and 1940s, many content analyses were conducted on then burgeoning forms of popular culture, including movies, radio, song lyrics, and magazines. The technique was applied to television almost immediately after it emerged. Early studies by Smythe (1954) and Head (1954) established basic parameters for examining television’s representations of demography (gender, age, class, race, occupations) and violence that other studies would emulate for decades to come. Smythe (1954) analyzed a week of New York television programs in 1951, 1952, and 1953 and Head (1954) studied a 13-week sample of 1952 network programs. It is noteworthy that Smythe began his report with the caveat that “Reality is too elusive a concept to be pinned down definitively” (p. 143).
Both of these seminal studies found that the demography of the television world diverged sharply from the real world. On television, there were twice as many male characters as female characters, and males tended to be older than females. Adults were vastly overrepresented in the television world; at that time, over half of the US population, but only a quarter of the TV population, was younger than 20 or older than 50.
Most TV characters were white Americans. “American Negroes” accounted for 2 percent of the TV world. Non-Americans were mostly English, Italian, and French; there were no Jews, Africans, Indians, or Asians other than Chinese, who represented 0.2 percent of the TV population (compared to 22 percent of the world’s population at that time).
Violence on television occurred at a rate of 6.2 acts per hour (Smythe 1954), and was far more frequent on children’s programs (22.4 acts per hour for “children’s drama” and 36.6 per hour for “children’s comedy drama”). Both studies found that the most common program type was crime drama.
These early studies found that upper- and upper-middle-class occupations were greatly over-represented, as were certain occupations at the lower and higher ends of the employment scale. Over half of television characters, compared to about 10 percent of the US population, were professionals, managers, service workers, and private household workers. The latter reflect the dominance of upper-class characters, who typically (in the media reality of the times) had servants at home. Occupations such as operatives, craftsmen, and farmers were virtually invisible in the TV world, but they accounted for nearly 50 percent of the actual workforce. In terms of specific occupations, teachers were the “cleanest, kindest, and fairest,” while scientists were “the least honest, least kind, and most unfair.” Lawyers were the “dirtiest” of all occupational types (Smythe 1954, 155).
Dozens of studies conducted in the intervening decades have confirmed and replicated the portraits of the TV world drawn by Smythe and Head, especially with regard to gender, class, and violence. One notable exception is that the number of African-American characters has increased over time.
Other studies have continued to focus on television, given the medium’s dominant role as the most widely shared storyteller of contemporary culture. One of the most sustained investigations of media reality was George Gerbner’s Cultural Indicators Project. Starting in 1967 and continuing into the late 1990s and beyond, an annual week-long sample of prime-time and weekend daytime US network broadcast programming was systematically coded for hundreds of aspects of the world as portrayed on television and the people who live in that world. The project accumulated data on thousands of programs and tens of thousands of characters over more than 30 years.
Many of the project’s findings echo those of Smythe and Head despite the passage of time. Although the percentage of women in the TV world did increase somewhat over the years (from 27 percent of the TV world in the 1970s to 35 percent in the 1990s), males continued to outnumber females. Daytime serials and game shows are more balanced, but females are especially under-represented in children’s and news programs. Women are twice as likely to play the role of wife as men are to play the role of husband. Women age faster than do men, and are more likely to be shown as “evil” as they age. Older people appear far less on television than an accurate representation of reality would require. People over 65 account for over 12 percent of the actual US population but less than 3 percent in the television world, and older women are especially scarce on television.
Poor and working-class people continue to be nearly invisible on television, appearing in less than one-tenth of their actual population share, while middle-class characters are over-represented, as are professionals (doctors, lawyers, judges, business moguls, among others). White males are consistently over-represented. Villains and “bad guys” are disproportionately from the lower classes and more likely to be represented as people of color or mentally ill. The percentage of African-Americans in network prime-time programs has increased (but only among males), roughly matching their corresponding percentage in the US population (about 12 percent). Asian/Pacific characters account for less than half, and Latino characters are less than one-third, of their real proportion of the US population, while Native Americans are nearly invisible.
Media reality is violent. Between 60 and 70 percent of the network programs in each weekly sample contained violence, with 4 to 6 acts of violence per hour. The small year-toyear fluctuations in these data show no clear pattern or tendency, and taken as a whole the patterns seem highly consistent over the decades. Among major characters, 40 percent commit violence and 43 percent are victims. On children’s programs, over 80 percent of males and two-thirds of females are involved in violence. Fewer than 2 percent of characters are shown as having any physical disability, and just over 1 percent are portrayed as mentally ill, but over 70 percent characters who are portrayed as mentally ill commit violence. (In reality, mental illness does not predict violence.)
Another large-scale study of television content in the US, the National Television Violence Study, examined 10,000 hours of programming between 1994 and 1997, and found many parallel patterns, with about 60 percent of prime-time programs featuring violence, with no major differences between broadcast and cable fare. Children’s shows were even more likely to contain violence, with an average of 14 episodes of violence per hour, compared to six in other programs. Beyond the sheer frequency with which violence is encountered in the media world, the reality of media violence bears little resemblance to the reality of violence. Nearly six out of ten violent incidents do not depict any pain, and about half depict no harm; close to 90 percent show no blood or gore.
Stories of crime and violence dominate news coverage as well as fictional programs, and the coverage does not match real-world crime patterns. Murder accounts for a disproportionate amount of both local and national news; murder suspects represent 0.13 percent of all those arrested, but 25 percent of all suspects in the news. From 1970 to 2000, almost a quarter of all stories were crime-related, although corporate crime typically receives relatively little attention.
The amount of crime coverage in television news does not reflect actual crime rates; editorial decisions and judgments of news value determine coverage, locally and nationally. Also, television network news over-represents white victims and under-represents African-American victims. African-Americans are more likely to be shown as perpetrators of crime and less likely than whites to be portrayed as police officers.
Researchers have pointed out many other significant discrepancies between the reality of crime statistics and television’s depictions of crime and violence. Perpetrators of crime are apprehended and convicted far more often on television than in reality. Women on television are more likely than men to be victims of homicide (the reverse is true in reality), and women on television are three times as likely to commit crime as are women in reality. Homicides account for 79 percent of the crimes in the television world, compared to only 0.01 percent of actual crimes (Brown 2001). Conversely, nearly 70 percent of actual crimes are theft or robbery, but these are only 5 percent of the crimes portrayed on television.
Accounting For Media Reality
Many varied factors account for these patterns, but the central explanations are commercial and cultural. Even the earliest content analysts of television pointed to the commercial context of programming as the major explanation for the media reality they found. Commercial media content is designed to feel familiar, to reproduce formulas, and to gratify common audience expectations; a program that strays too far from the mold would be jarring to the audience. Commercial media have always thrived on imitating the successful; fear of losing the audience drives programming decisions. The patterns described here also have deep cultural and historical roots that predate modern media; current media are not the source of these images, but television in particular has mass-produced and mass-distributed them to an unprecedented degree. All eras and cultures have relied on stories to express and represent both reality and ideology, but never before has any society produced and consumed as many stories as we do now.
The stories and images of the media reflect popular ideological and commercial values, including the glorification of youth culture, particular intersections of race, class, and gender, the valorization of certain occupations, ritualistic struggles between good and evil, and so on. As with any cultural or industrial product, television stories reflect the values, priorities, and needs of those who produce them. This explains why so many aspects of media reality appear to have changed little in more than half a century of research. Slow, gradual changes in cultural reality do come to be reflected in media reality (and vice versa), but without meaningful change in the dominant institutional commercial structures, significant changes in media reality are unlikely to be seen.
- Berger, P. L., & Luckmann, T. (1966). The social construction of reality: A treatise in the sociology of knowledge. Garden City, NY: Doubleday.
- Brown, N. J. (2001). A comparison of fictional television crime and crime index statistics. Communication Research Reports, 18(2), 192–199.
- Center for Communication and Social Policy (1998). National television violence study, vol. III. Thousand Oaks, CA: Sage.
- Chesebro, J. W. (1984). The media reality: Epistemological functions of media in cultural systems. Critical Studies in Mass Communication, 1, 111–130.
- Donsbach, W. (2003). Objectivity in reporting. In D. H. Johnston (ed.), Encyclopedia of international media and communication. San Diego, CA: Academic Press, pp. 383–391.
- Donsbach, W. (2004). Psychology of news decisions: Factors behind journalists’ professional behavior. Journalism, 5(2), 131–157.
- Fiske, J. (1987). Television culture. London: Routledge.
- Gerbner, G. (1972). Violence and television drama: Trends and symbolic functions. In G. A. Comstock & E. Rubinstein (eds.), Television and social behavior, vol. 1, Content and control. Washington, DC: US Government Printing Office, pp. 28–187.
- Gerbner, G. (1994). Women and minorities on TV: A study in casting and fate. Media Development, 41(2), 38–44.
- Gerbner, G., Gross, L., Jackson-Beeck, M., Jeffries-Fox, S., & Signorielli, N. (1978). Cultural indicators: Violence profile no. 9. Journal of Communication, 28(3), 176–207.
- Head, S. W. (1954). Content analysis of television drama programs. Quarterly of Film, Radio, and Television, 9(2), 175–194.
- Kepplinger, H. M. (1979). Creating a crisis: German mass media and oil supply in 1973/74. Public Opinion Quarterly, 43, 285–296.
- Kepplinger, H. M. (1992). Artificial horizons: How the press presented and how the population received technology in Germany from 1965–1986. In S. Rothman (ed.), The mass media in liberal democratic society. New York: Paragon House, pp. 147–176.
- Kepplinger, H. M. (2002). Mediatization of politics: Theory and data. Journal of Communication, 52(4), 972–986.
- Neuendorf, K. (2000). The content analysis guidebook online. At academic.csuohio.edu/ kneuendorf/content/bibs/comabsbib.htm, accessed August 8, 2006.
- Nohrstedt, S. A., & Ottosen, R. (eds.) (2005). Global war: Local views: Media images of the Iraq war. Gothenburg: Nordicom.
- Pearson, R. C. (2005). Fact or fiction? Narrative and reality in the Mexican telenovela. Television and New Media, 6(4), 400–406.
- Porto, M. P. (2005). Political controversies in Brazilian TV fiction: Viewers’ interpretations of the telenovela Terra Nostra. Television and New Media, 6(4), 342–359.
- Schulz, W. (1976). Die Konstruktion von Realität in den Nachrichtenmedien: Analyse der aktuellen Berichterstattung. Freiburg and Munich: Alber.
- Signorielli, N. (1984). The demography of the television world. In G. Melischek, K. E. Rosengren, & J. Stappers (eds.), Cultural indicators: An international symposium. Vienna: Austrian Academy of Sciences Press, pp. 137–157.
- Smythe, D. W. (1954). Reality as presented by television. Public Opinion Quarterly, 18, 143–156.
- Wolfsfeld, G., Khouri, R., & Peri, Y. (2002). News about the other in Jordan and Israel: Does peace make a difference? Political Communication, 19, 189–210.