Content-Type: text/html Use of Minority Sources in News A.N. Mohamed [Edinboro University of Pennsylvania] & Anita Fleming-Rife Introduction Studies of media's handling of race issues invariably note the absence or misrepresentation of minority group perspectives.1 The oversight is generally a by-product of a "commonsense" approach to newsgathering that is based on identifiable values and practice codes whose legitimacy and fairness are taken for granted. In view of this, it is significant that some industry groups such as Gannett have instituted measures aimed at influencing the nature of coverage minorities receive.2 For some time during the 1980s, editors and reporters affiliated with Gannett papers were asked "to quote minority sources in every article they write"3 in order to achieve what's called a "mainstreaming" effect.4 The opinions of African American and other minority news sources were to be sought in connection with every subject in the news, not just on "ghettoized" race issues. This study will explore the possible effects on news that such an approach to news sources will entail. We will investigate how this approach to news might affect the definition and/or framing of social issues covered in the media. As many previous studies have shown, who the media tap as sources could provide insight into the prevailing social order5 at the same time that it could serve as a roadmap to how community consensus is achieved.6 Our analysis will be anchored in the theoretical frameworks of social conflict between groups vying for greater prominence in society's interpretive processes which, in turn, depend on access to news media. Newspapers from three chains (Gannett, Knight-Ridder and Scripps-Howard) will be compared for variations in how African American and other minority news sources are used. Different editorial policies with regard to use of news sources are expected to produce variations in the racial overtones of news stories. Media Selection of Sources and the role of Social Structure Conflict theories of 19th and early 20th century social thinkers offer useful frameworks for analyzing social structure and how news and the role of sources reflect on these structures. As Gans7 has indicated, much news is fundamentally about the tensions between power elites and those that may challenge their authority. In the mid 19th century, Karl Marx8 argued that the ideas disseminated in society are ideas of the economically powerful. As a way of outlining his materialist conception of history, Marx suggested that ideas could not flourish in any society unless they were accepted and/or tolerated by those who control the means of production. The German sociologist, Ferdinand Tonnies9 agreed that newspapers owned generally by socially privileged individuals, were much more likely to print information that supports, rather than challenges, the prevailing social order. These propositions, linking dominant ideas to economics, are extensions of August Comte's10 "organic conceptions of society." This analytic framework of society as organism, widely used as an explanatory tool by early sociologists, has clearly influenced contemporary communication scholars. Gerbner and Gross 11 for example, suggested that power elites use the media to perpetrate symbolic meanings of events and situations in a way that legitimizes the prevailing social order. Marcuse 12 observed that under the guise of public service, the media present a picture of the world that conforms to the established interests of economic and political elites. This concept of dominance of the many by the few was perhaps best captured by Thomas in her rhetorical question: How is it that a lot of people seem to act Against their own self-interest? In the United States, wealth and power are not evenly distributed; a very small number of people have a great deal and most everyone else has relatively little. Yet, the masses, with relatively little, work very hard to support this overall system which results in inequitable distribution of material wherewithal. Why is that? Or, more specifically, what role do the mass media play in helping to contain this pattern? 13 This view, linking society's dominant ideas to the interests of dominant groups, calls for an examination of the portrayal of social conflict in news and how the choice of news sources impacts the social dynamics borne of inter-group conflicts. In classical as well as contemporary sociological theory, social conflict has served as an analytic tool for studying the structure of society itself. Conflict theorists such as Marx see conflict as inevitable and inherent in all social transactions. It is viewed as a catalyst for historic social transformation as it has posed a perennial threat to the prevailing social order. "Conflict functionalists" such as Georg Simmel and Lewis Coser have argued that conflict is functional in that it serves to demarcate social boundaries between groups and brings about cohesion among those who need to identify with each other. 14 Contemporary communication scholars have shown that institutional ideologies of news organizations used to frame news are best revealed through coverage of social conflict. By ideology is meant the socially constructed interpretive codes that determine the meanings of situations or events to groups. As America's pragmatist thinkers noted, much of what is considered "social reality" consists of constructions of meanings borne of the subjective perceptions of situations by interacting individuals.15 Accordingly, the media's coverage of conflict situations offers an opportunity to discern the ability of dominant social groups to prevail in the interpretation of those situations. One important advantage for dominant groups in shaping meanings of events is the preponderance of political elites as news sources. Sigal16, noted that for major U.S. newspapers, "routine channels for news gathering constitute the mechanism for official dominance of national and foreign news." He determined that events under government control such as official press releases, press conferences, official proceedings, and "background" briefings, together accounted for 72 percent of the news channels for the New York Times and the Washington Post. About 14 years later, Brown et al.17 found that more than half of 5,248 sources cited in 846 front page stories were official government sources. And almost two decades after the Sigal study, Hansen 18 found that about 40 percent of daily reporting relied on official government sources. This preponderance of official sources is bound to affect not only the kind of information transmitted but also the interpretation of significant situations and events.19 As Berkowitz and Terkeust20 have argued, "what takes place in the public sphere are struggles of meanings to make dominant one of several preferred interpretations of occurrences and issues." Accordingly, Hackett21 found that reports on Canadian political discourse mirrored very closely the views of the most dominant political figures. And Weaver and Elliott22 found that of 19 prominent local issues, 12 were transmitted by the media exactly as local authorities had framed them. What, therefore, are the implications of this dominance in pluralistic democratic society where the concerns of all constituent groups deserve to be heard? Media in Pluralistic Social Systems Homogeneous societies have more traditional social structures in which social differentiation by class, language, race, and lifestyle are largely nonexistent. The absence of social conflict also characterizes these communities as inter-group tensions are minimal, if at all. Olien and his colleagues,23 who have extensively studied community dynamics and the role of the media, report that media in this type of community operate essentially as extensions of community leaders in terms of defining and interpreting the community's affairs. As Olien et al. put it: . . .the small, homogeneous structure is typically without mechanisms for effectively challenging authority-no pluralism, no conflict abatement procedures, no organized challenge, and no conflict reporting about the issues.24 Pluralistic societies, on the other hand, are more complex and are characterized by significant variations in the range of demographic attributes of the population along the lines of race, ethnic background, religion, income, and education. These differences engender more than just cultural contrasts. Groups in pluralistic societies do not share common frames of reference for interpreting common environmental cues. Dissimilar life-experiences by ethnic, religious, or racial groups translate into an absence of a commonality of preferred interpretations of situations. Hence the higher likelihood for inter-group tensions and/or social conflict in pluralistic societies. Both interpersonal and mediated human communication are interpretive exercises in which the preferred meanings of dominant groups gain legitimation at the expense of weaker groups, according to social psychologists.25 The media are important catalysts for the power of the dominant groups as they help determine what Fish has called the interpretive group of prominence.26 Gans and others 27 have identified several factors that help preclude the views of weaker social groups from gaining prominence in the media for any significant periods of time. They include the combined effects of reporters' work routines, the principle of journalistic objectivity, and the social proximity of reporters to their main sources. Work routines shaped by constraints on time and resources compel reporters to rely on well-placed official sources in order to obtain the "most suitable" information, in the quickest time, from the fewest sources. Gans noted the high correlation between political and economic power and newsworthiness. Thus, while power leads to ready and easy access to the media, Gans says that "the powerless must resort to civil disturbances in order to obtain it." 28 As Sigal maintains, "News is not what happens, but what someone says has happened or will happen."29 And more often than not, those who define and frame news rank higher, rather than lower, in the social power hierarchy.30 This completes the cycle in which the preferred interpretations of the powerful remain dominant.31 For example, an analysis of Time and Newsweek's coverage of the "black agenda" during the 1980s showed that these two leading newsmagazines reflected the Reagan Administration's policies toward black America very closely and not at all that of the Urban League-a prominent black advocacy group.32 Just like the Reagan Administration, the highest priority "black agenda" issue for these newsmagazines concerned the "eradication of reverse discrimination disguised as affirmative action, quotas, and numerical goals," whereas the Urban League's top priority issues during that same period-eradication of "unemployment, hunger, and homelessness"-were ignored by the magazines.33 The media's bias toward the position of power groups manifests itself through devoting more space or time to the discussion of dominant perspectives. For example, Daniel and Allen34 noted that while Time magazine devoted only three paragraphs to what the authors described as a "landmark court decision" linking housing discrimination to public school desegregation, the magazine gave two complete pages to a discussion of the shortcomings of affirmative action. The authors argue that the disparity in treatment of the two issues is a reflection of the disparity in power; and that the Reagan Administration policies on civil rights were accorded greater prominence and legitimacy.35 In explaining the media's ability to mold public perceptions in line with the interests of the powerful, Parenti stated that: . . . the press can effectively direct our perceptions when. . .the message seems congruent with earlier notions about events. In this way the original implant is also a reinforcementof earlier perceptions. Seemingly distinct reports about diverseevents have a hidden continuity and a cumulative impact. . . designed to reinforce the dominan orthodoxy. 36 Analyses of media portrayals of social reality also points to the effects of socialization to specific professional values and approaches to journalism. Specific approaches and their concomitant values also help advance powerful groups' interpretations of reality and not that of weaker groups. In the U.S., news organizations have long been viewed as watchdogs that help keep the various branches of government accountable to the public. This view of the role of the media appears to contradict the holistic theory of society in which the media help protect the prevailing social order. But scholars like Olien, Donohue, and Tichenor 37 maintain that the more appropriate metaphor for the role of the media in the U.S. is that of a guard dog. As guard dogs, the media protect those at the vanguard of society against threats to the social order. Although, as Olien et al. explain, "Guard dog reporting in social conflict has a sharper focus on role incumbents than on power structure, since structures are what the guard dog role is conditioned to protect." 38 By definition, therefore, the guard dog metaphor points to partiality to one interpretation of reality-that of the dominant social groups and not that of weaker ones. Perhaps a surprising source of journalistic bias toward dominant social perspectives is the principle of objectivity. An objective reporter who relies on news sources transmits the views of those sources without counterbalancing them with his or her own perspectives as journalistic professionalism dictates. This works to tip the values and perspectives contained in the news in favor of groups from whom the most influential and most numerous sources come. And as we've already seen, these are mainly the more dominant and more politically powerful social groups.39 Reporters' choices of news sources is influenced also by what Gans has called "social proximity". Early in the 20th century, Robert Park40 wrote that the level of communication that exists between groups is directly related to the "social distance" between them. The more similar individuals are in terms of race, class, and education, the shorter the social distance (social proximity), and therefore, the greater the level of communication. Likewise, Gans notes that "social proximity enables people of similar backgrounds and interests to make contact" while "obstructing" those who are different.41 Investigators of social-psychological influences in journalism suggest that members of racial minorities and lower-class poor are often not approached as news sources in part because of their "social distance" from reporters and editors. 42 the issue of social distance is perhaps best represented in the psychological separateness of people of different races. Consequently, real or perceived qualitative differences in the coverage of issues affecting different racial groups have been widely attributed to racism. "Modern Racism" and the Role of the Media Alexis de Tocqueville's American experience was marked by the impression made on him by the American Creed proclaiming "justice, freedom, and equality" that were uncommon in Europe and elsewhere during the 19th century.43 Yet he concluded that ingrained racial animosities "would plague the future as they had haunted the past."44 This fundamental paradox in the American character has been noted by social scientists who could not reconcile the egalitarian outlook of our collective self with the prejudiced and discriminatory attitudes of individuals.45 Although the racial climate in America has improved both socially and legally since the 1960s, individuals still are socialized in an atmosphere of racial animus.46 The openly racist legal and social codes of the past (especially in the South) have been supplanted by politically and socially more tenable attitudes that are nonetheless racist. While traditional racist sentiments such as segregation and legalized discrimination are no longer a part of mainstream media's fare, the literature suggests a preponderance of images and expressions whose collective connotative effect essentially add up to a racist media ideology.47 Separate analyses of media content48 have shown the existence of institutional frameworks of newsgathering practice that feed such tendencies as the proclivities for harboring hostilities toward black political and economic empowerment, for viewing blacks as too demanding, and for dismissing the idea that racial discrimination hampers black social progress.49 Some social scientists refer to this phenomenon as "modern racism,"50 others refer to it as "enlightened racism,"51 while still others call it "symbolic racism."52 McConahay53 has developed a scale for gauging the prevalence of modern racism. His instrument measures three basic characteristics: (1) a general aversion to blacks as a race because of stereotypical images formed through socialization, (2) a resistance to black political and economic demands such as those sought via legislative means, and (3) a denial that racial discrimination exists and that it contributes to an uneven playing field for blacks. McConahay's survey on modern racism elicited responses to statements such as the following: ù Over the past few years, blacks have gotten more economically than they deserve. ù Over the past few years, the government and news media have shown more respect for blacks than they deserve. ù Blacks are getting too demanding in their push for equal rights. Positive responses ("agree" and "strongly agree") to these statements are considered indices of modern racism. McConahay found that even after controlling for legitimate political conservatism, anti-black sentiments were still clearly discernible.54 Similar observations were made by Sears55 and Hacker.56 Indeed, the scope of modern racism gets even broader in Hacker's analysis of the status of race relations in the America of the 1990s: At the most visible level, growing numbers of white people are expressing misgivings over how black people are conducting themselves....More and more whites increasingly condemn blacks for casting their race as victims who have no control over their condition. White Americans are franker to admit their support for racial barriers. They describe themselves as bystanders who must watch as their country is held hostage by a demanding minority. In their view, the behavior of blacks is a major explanation for what ails America.57 One disturbing aspect of race-related attitudes is the tendency to generalize behavior characteristics in society's minority groups. Social psychologists note that individuals have a natural inclination to favor members of their own groups (the ingroup) than members of other groups (the outgroup).58 Consequently, perceptions of outgroup homogeneity becomes normal and routine just as ingroup heterogeneity is taken for granted.59 When some minorities misbehave, therefore, others in their group "are homogenized and assimilated to a negative stereotype by the ingroup whereas those in the ingroup (whites) see themselves as individuated members of a diverse group." 60 Studies suggest that media help perpetuate these sentiments in several ways. Entman, for example, noted that both local61 and national62 television news "promoted perceptions that exacerbate whites' racial antagonism"63 through their coverage of crime and political events involving blacks. His data showed that news of crime by blacks was mostly violent and/or involved drugs. Depictions of white criminals, by comparison, were more varied and included a significant and mitigating "white collar" component.64 Blacks were more likely to be shown in handcuffs or otherwise being physically restrained by police than were whites; and blacks were more likely to be shown in still mug shots than were whites-thus painting the picture of a more menacing criminal element.65 Furthermore, fewer pro-defense sound bites were included in news of black criminals than in news of whites.66 This supports an analysis of newspaper coverage of criminal court trials by Condit and Selzer67 who found that media routinely frame their coverage in a way that favors one side, thus cueing readers toward predictable conclusions. In coverage of political news, Entman68 found that black leaders compared unfavourably with their white counterparts. Much more than whites, for example, blacks "complained of racial descrimination and often criticized government policy" for various inadequacies.69 Black leaders also tended to speak about a narrow range of issues affecting the black community, whereas white leaders spoke on matters of interest to the general public. Moreover, black leaders and activists in the news spoke in angry tones when registering their complaints about perceived injustices. Former New York City Mayor, David Dinkins (who is black), was concerned about this phenomenon enough to plead with the media "to seek out and cover those whose views may be less confrontational-even if we're not as loud, even if our words create less controversy."70 The mayor's concern is understandable, as the cumulative effect of these images is that they will likely deepen the resentment of blacks by whites.71 Although these depictions are not products of a conscious conspiracy, they may, nevertheless, construct a reality that is at variance with the truth. This was indeed the conclusion of a study in San Fransisco72 done in conjunction with four professional associations of minority journalists (NABJ, NAHJ, NAJA, AAJA).73 The study lamented how the "mainstream media's coverage of people of color is riddled with old stereotypes, offensive terminology, biased reporting and myopic interpretations."74 The nature of coverage racial minorities receive could thus be seen as a product of applying one frame of reference to make sense of another. This speaks directly to the frustration expressed by Roland Barthes with the "naturalness" with which newspapers "constantly dress up reality" without acknowledging the role of history in determining that reality.75 Likewise, Stuart Hall76 explained that the media's construction of the concept of race is determined by the ideology that informs the social consciousness of a given group at a given time. Here, ideology refers to the framework "which provides us with the means of 'making sense' of social relations."77 It is this media ideology that dictates the picture of reality about race relations that emerges, and which in turn, molds audience perceptions. To place the nature of this coverage in proper perspective, therefore, an examination of some aspects of organizational newsgathering practice is called for. Editorial Control and Use of News Sources A pivotal relationship in organizational structure is the one between ownership and content homogeneity and/or editorial autonomy. The overwhelming majority of U.S. daily newspapers (about 75 percent) are now owned by chains.78 And the trend toward group ownership has long generated interest in the implications of this control for news content.79 Organizations like Gannett and Knight-Ridder have gone to some length in trying to allay fears of content homogenization in their papers. Gannett, for example, often promoted itself as "A World of Different Voices Where Freedom Speaks;" while Knight-Ridder, after buying six more newspapers during the 1980s, tried to reassure readers: We don't own their opinions. We don't own their news columns. And quite frankly, we don't want to.80 But several content studies found evidence of homogenization among papers owned by the same chain. Glasser, Allen, and Blanks81 found that newspapers belonging to the Knight-Ridder group gave more similar news play to a specific story than other papers subscribing to the Knight-Ridder news service. Akhavan-Majid, Rife, and Gopinath82 found that editorials in Gannett papers "showed a high level of homogeneity and significant differences between the Gannett and non-Gannett papers." And Gaziano showed that newspapers belonging to smaller chains tended to be similar in their editorial endorsements of presidential candidates.83 But the most conspicuous attempt at influencing coverage is probably Gannett's policy of "mainstreaming" minority images in the news.84 The concept began in the mid 1980s when Gannett suggested that "someone of African-American, Hispanic, Asian, or Native American descent should always be quoted in everyday news coverage, such as stories about tax hikes or new school curricula."85 The aim was to avoid pigeonholing minorities by associating them only with certain kinds of stories. That some papers like the Detroit News had a special ombudsman to oversee the implementation of the mainstreaming idea attested to Gannett's commitment "to do the right thing." Because sources cited in news stories do indeed affect the nature of news, Hansen,86 suggested that "one way to encourage diversity in the marketplace of ideas is to increase the regular use of non-conventional sources for news." This assertion is especially important in light of conclusions by Lasorsa and Reese87 that much of the media's gatekeeping power derives from "their selection of sources, which largely determines the way stories are framed." The Hypotheses The foregoing discussion of the images in the media of African American and other minority sources of news, as well as Gannett's attempts at "mainstreaming" these sources, leads us to pose the following hypotheses: H1: Gannett's policy of source diversification will mean less reliance on traditional sources than would be the case with other chain newspapers. H2: Gannett papers are more likely than other chain newspapers to use minority sources evenly in stories about different subjects. H3: The thematic content of news will be influenced by the racial background of the sources cited. Method A content analysis of 15 newspapers from across the U.S. (five each from Gannett, Knight-Ridder, and Scripps-Howard) was undertaken. 88 The unit of analysis is the theme. We decided that since source messages come in thematic units, this would be the most appropriate unit of analysis. Therefore all themes within a story in which a source is quoted, directly or indirectly, were coded. Two classifications of sources were made: (a) sources were categorized in terms of occupation as elected official, public official, public figure, business leader, national civil rights leader, regional/local civil rights leader, community/religious activist, professional, ordinary citizen, and "other". Sources were also classified according to race as African American, Hispanic/Latino American, Asian American, Native American, and Caucasian. In order to test Entman's theory about qualitative differences in the thematic content of source information based on their race, stories were coded for their inclusion of such themes as anger, criticism of government policies/programs, advocacy of narrow interests, and complaints about government and society. Content was also coded for the presence of opposite themes as advocacy of general interest issues, supportive and/or constructive comments about government, and statements of general solidarity with the American people. In order to improve reliability of coding these thematic variables, each of them was coded dichotomously. Thus, if sources made explicit references to their disposition (e.g., anger, complaint) toward particular policies or made their feelings about issues clear through context, a value of "1" was recorded. If no direct reference to a source's disposition were made, or if the context was vague and unclear, a value of "0" (zero) was recorded. The themes were further coded according to broad topical categories as (1) social, (2) legal, (3) political, and (4) economic. Five nationally significant stories were selected for analysis. The stories include three that had a strong racial element and two that had no racial overtones at all. The stories, which span a period of six years from the late 1980s to the mid 1990s are: (1) President Ronald Reagan's veto of a civil rights law passed by Congress in March 1988, (2) The Iran-Contra story after the indictment of Colonel Oliver North in March 1988, (3) The Supreme Court's abrogation in January 1989 of a Richmond (Virginia) Affirmative Action ordinance that reserved 35% of city contracts for minority bidders, (4) The debate over, and the passage of, the North American Free Trade Agreement (NAFTA) in November 1993, and (5) the story of a South Carolina mother who drowned her two young sons and then blamed a fictitious black male carjacker for their abduction in November 1994. One week's worth of news about each of these stories was coded. Inter-coder reliability coefficients (on all variables) between three coders were .84 at the beginning of coding and .86 toward the end of coding. Findings The content analysis produced an N of 2733 themes. Of these, the race of the news source could be clearly identified either through direct references in the stories, through photographs, or through knowledge of the sources by coders in 2159 cases, or 78.9% of the time. Minorities comprised 11.7% of identifiable sources for all the stories. Caucasians comprised 88.3% of the sources. Of the three newspaper chains, Gannett's papers used a slightly higher proportion of black sources (13.5%) than either Knight-Ridder (9.2%) or Scripps-Howard (7.3%) (Chi Sq= 63.53, p=.0001; see Table 1). But when the proportion of all minority sources is considered, there is little difference between the three chains, with Gannett using minority sources 16.1% of the time, compared to 13.2% for Knight-Ridder, and 16.5% for Scripps-Howard. Table 1 Race of News Source by Newspaper Chain* Black Hispanic Caucasian Total Gannett N 89 13 551 657 Col% 13.5 2 83.9 Knight- N 88 19 828 954 Ridder Col.% 9.2 2 86.8 Scripps N 40 44 460 550 Col% 7.3 8 83.5 Total N 217 76 1839 2155 Col% 10 3.5 85.1 Chi Sq=63.53 df=10 p=.000 * Percentages do not add up to a 100 due to rounding. Table 2 Kind of News Source Cited by Newspaper Chain Public/elected officials civil rights leader local/regional activist professional public figure business leader ordinary citizen other Total Gannett N= 380 raw % 48 col % 27 18 2 42 44 5 44 123 15 29 78 10 39 20 2 17 115 14 29 11 1 20 789 100 29 Knight/Ridder 635 52 45 16 1 37 35 3 35 216 18 52 49 4 24 57 4 48 180 15 45 16 1 29 1204 100 44 Scripps- Howard 386 52 27 9 1 21 19 2 19 77 10 4 74 10 37 40 5 34 102 14 25 27 4 50 734 100 27 TOTAL 1401 43 98 416 201 117 397 54 2727 Cramer's V = .138; p= .000 To test our first hypothesis, that Gannett newspapers will depend less on traditional sources of news, a Chi Square test of the relationship between chain ownership and type of news source was computed (see Table 2). The data shows that overall, Gannett papers relied less on elected and/or other government officials than did either Knight-Ridder and Scripps-Howard newspapers. While a majority of Gannett sources (about 52%) were neither elected nor other government officials, a majority of sources quoted by both Knight-Ridder and Scripps papers were government and/or elected officials (about 53% each). With N=2155, Cramer's V is significant at p=.000. This shows that Gannett did indeed depart from the news media tradition of relying heavily on "powerful" sources, who are most directly responsible for instituting and executing public policies and programs. It's worth noting that Gannett also has relied more heavily on sources that were identified as civil rights leaders and as local or regional activists. Sixty-two or 7.8% of all sources used by Gannett were civil rights, or other local activists while Scripps-Howard used only 28 or 3.8% of these sources and Knight-Ridder papers used 51 or only 4.2 of these sources. Cramer's V is significant at the p=.000 level and the hypothesis is therefore supported. The second hypothesis seeks to compare the three chains on how they use minority sources across different kinds of stories. It was not possible to run statistical comparisons to test this hypothesis because of insufficient frequencies in some data cells. However, we may state that the data (see Table 3) presents inconclusive evidence in this case. The hypothesis states that Gannett papers are more likely than other chain papers to use minority sources in different kinds of stories. The data, however, shows that none of the Gannett papers in our study used a minority source (Black, Hispanic, Asian, or Native American) in three of the five stories. Only one native American source was cited in the fourth story, and only two Table 3 News source of five stories by race and newspaper chain Gannett Knight-Ridder Scripps-Howard Black Hisp White Black Hisp White Black Hisp White STORY Richmond N= 20 - 31 25 2 48 17 - 34 Affirm/action Raw% 39 - 71 33 2.6 63 33.3 - 66.7 Col% 22.5 - 7.8 28.4 10.5 5.8 42.5 - 7.4 Reagan veto N= 4 - 63 5 - 106 2 2 50 Raw% 6 - 94 4.5 - 95.5 3.7 3.7 92.6 Col% 4.5 - 11.4 5.7 - 12.8 5 4.5 11 Susan Smith N= 49 - 91 53 - 160 9 6 81 Story Raw% 35 - 65 25 - 75 9.3 6.2 83.5 Col% 55 - 16.5 60 - 19 22.5 13.6 17.6 NAFTA N= 16 13 858 5 16 389 10 34 217 Raw% 5.6 4.5 89.6 1.2 3.8 91.3 3.8 12.8 82 Col% 18 100 46.7 5.7 84.2 47 25 77.3 47.2 Iran/Contra N= - - 109 - 1 125 2 2 78 Raw% - - 98.2 - 8 98.4 2.4 2.4 93 Col% - - 19.7 - 53 15 5 4.5 17 Totals N= 89 13 551 88 19 828 40 44 460 Raw% 13.5 2 83.9 9.2 2 86.8 7.3 8 83.5 Asian American sources were cited in the fifth story. Knight-Ridder papers did not cite Hispanic, Asian and Native American sources in two of the five stories, and no Blacks were cited in one of the stories. Scripps-Howard papers did not cite any Asian American sources in three of the stories. Hispanics were not cited in at least one story, and Native Americans were not cited in four of the five stories. Overall, Gannett and Knight-Ridder papers used considerably more minority sources than Scripps-Howard papers (102, 107, and 88 respectively). The third hypothesis is meant to test Entman's theory that there are predictable content themes that are more reliably associated with minority sources than non-minority sources of news. These include advocacy of narrow interests, complaints about their conditions in society, anger about real or perceived social injustices and public policies, and general criticism of the prevailing social order. Entman also argued that minority news sources were cited in circumstances in which they were less likely than other sources to advocate general interest themes or support the prevailing social order. The data in Table 4 show the relationship between race of source and six content themes: anger, complaints, criticism, narrow interest, general interest, and support for the prevailing social order. (Data on Asian and native American sources were eliminated from the table because of the small numbers of sources from these two groups.) The largest statistical differences between the news sources of various racial groups is in the presence of complaints and criticism themes in their accounts. African American sources, for example, had complaint themes in their comments 53% of the time compared to 13% for Hispanic sources and only 10% for white sources (Kendall's Tau B = -.314, N=2133, p=.000). A majority of black sources also criticized the system (58.5%) compared to 32% of Hispanic sources and only 17% of white sources (Kendall's Tau B=-.279, N=2132, p=.000). A majority of black sources spoke about narrow interests (63%) in their accounts to news media as opposed to 25% of Hispanic sources and 26% of white sources. (Kendall's Tau B= -.196, N=2132, p=.000). A majority of both black and Hispanic sources (71% and 60.5%, respectively) did not speak to general interest themes compared to a majority of white sources (55.5%) who did (Kendall's Tau B = .190, N=2133, p=.000). Table 4 Race of news source by content themes in three newspaper chain Anger Complaint Criticism Narrow General Support interest interest yes no yes no yes no yes no yes no yes no Black N 99 118 115 102 127 90 137 80 45 172 34 183 Raw% 46 54 53 47 58.5 41.5 63 37 21 79 16 84 Hispanic N 16 60 10 66 24 52 19 57 30 46 22 54 Raw% 21 79 13 87 32 68 25 75 39.5 60.5 29 71 White N 313 1526 182 1658 306 1533 474 1365 1021 819 793 1047 Raw% 17 83 10 90 17 83 26 74 55.5 44.5 43 57 Kendall's Tau B -174 -314 -279 -196 .190 .148 Significance level .000 .000 .000 .000 .000 .000 N 2132 2133 2132 2132 2132 2132 Significantly larger majorities of minority sources (84% for blacks; 71% for Hispanics) failed to offer support to the system as opposed to 57% of whites who did not (Kendall's Tau B = .148, N=2133, p=.000). Although anger was not registered by a majority of sources of any race, it is noteworthy that black sources registered their anger much more frequently than did other news sources. While 46% of black sources expressed anger or said that they were angry about something, only 21% of Hispanic sources and 17% of white sources did the same (Kendall's Tau B = -.174, N=2132, p= .000). Our third hypothesis is therefore supported. Conclusions As Olien and his colleagues 89 have noted, who the media tap as news sources can provide insight into the prevailing social order. Potentially, it could also affect whether community consensus is achieved or whether social acrimony is stirred. Accordingly, conflict theorists assert that societal groups representing different interests compete constantly for legitimation of their views and for a wider acceptance of their interpretations of social reality. 90 As Berkowitz and Terkeurst put it, the ultimate objective of special interest groups within society "is to protect and strengthen their social position and power through interpretations that facilitate acceptance of their preferred meanings"91 of events and situations. This is accomplished in large measure through mass media interpretations of those situations and events.92 Consequently, patterns of news source use by media organizations becomes critical to our understanding of consensus building processes as well as the dynamics of inter-group conflict. In this study, two conclusions can be drawn from how 15 newspapers belonging to the three largest newspaper chains in the country have used news sources from racial minority groups: 1) Newspapers belonging to all three chains continue to associate minority sources with only certain types of stories - namely, stories with a strong racial/ethnic component. Gannett papers, for example, did not quote Hispanic sources in any of the stories except in coverage of NAFTA. Both Knight-Ridder and Scripps-Howard papers quoted Hispanic sources in connection with NAFTA significantly more frequently than with any other story (84% and 77% of total number of quotations, respectively). African-American sources were used in the greatest proportions by all chains in relation with only race-related stories. Eighty-two percent of the quotations of African-American sources by Gannett newspapers came with the three stories on race relations-the Richmond affirmative action story, the Reagan veto of a civil rights bill, and the Susan Smith story out of South Carolina. Knight-Ridder papers associated black sources with "race" stories even more - with 94% of quotations of blacks coming with the three "race" stories. Gannett's policy of seeking minority sources for all news stories in its papers, initiated in the mid-1980s by former Chairman of Gannett, Al Neuharth, has not been successful in achieving its objectives. This was an experiment aimed at diversifying the views about the issues and events addressed in news stories as well as rectifying the long-standing practice of relying on a small "elite" pool of sources for most stories. Implementation of this policy, however, proved difficult as newsrooms put up resistance to this revolutionary approach to news gathering and news writing. Many reporters and editors considered it impracticable, arguing that it disrupted normal work routines and made it difficult to meet editorial deadlines. Others opposed the policy in principle - saying that the requirement to speak to minority sources amounted to a social engineering of news content. In the end, however, the policy did not survive the departure of the man whose brainchild it was -- Al Neuharth. 2) The status quo, on the other hand, clearly presents problems. The current practice of citing minority sources only in connection with "race" stories influences how news consumers come to associate racial minorities with certain unpopular dispositions toward some social and political issues as Entman93 has previously shown. Our data confirms Entman's hypothesis that minority sources are more likely to speak in angry tones, to complain about their condition in society, to criticize public policies and programs, and to speak against the prevailing social order or social system. Although this is explained in part by the media's association of minorities with stories about certain kinds of issues, the effect is nonetheless detrimental to society as a whole - as it helps feed the climate of racial tension that still pervades the lives of millions of Americans. References 1) Carolyn Martindale, The White Press and Black America. (N.Y.: Greenwood Press, 1986); Christopher P. Campbell, Race, Myth and the News. (Thousand Oaks: Sage, 1995); Kirk A. Johnson, "Black and White in Boston," Columbia Journalism Review. May/June 1987,pp. 50-52;Carol Bradley Shirley, "L.A. Stories: Where Have You Been?" Columbia Journalism Review. July/August 1992, pp. 25-26. 2) Valarie Basheda "When should you quote minority sources?" American Journalism Review. October 1995. 3) Basheda, "When should you quote minority sources?" p. 12. 4) Basheda, "When should you quote minority sources?" 5) Herbert J. Gans, Deciding What's News: A Study of CBS Evening News, NBC Nightly News, Newsweek, and Time. (N.Y.: Pantheon Books, 1979). 6) Clarice N. Olien, George Donohue, and Phillip Tichenor, "Conflict, consensus, and public opinion," pp. 301-322 in Theodore Glasser and Howard E. Sypher (eds.), Public Opinion and the Communication of Consent. (N.Y.: The Guilford Press, 1995). 7) Gans, Deciding What's News. 8) Karl Marx and F. Engeles, German Ideology. (London: Lawrence & Wishart, 1936). 9) Ferdinand Tonnies, On Sociology: Pure, Applied, and Empirical. (Chicago: University of Chicago Press, 1923). 10) August Comte, System of Positive Polity. Vol. 2. (London: Longmans Green & Company, 1852). 11) George Gerbner and Larry Gross, "Living with television: the violence profile." Journal of Communication 26: 173-199, 1976. 12) Herbert Marcuse, One Dimensional Man. (London: Routledge & Kegan Paul, 1964). 13) Sari Thomas, "Mass Media and the Social Order." Pp. 611- 627 in Gary Gumpert and Robert Cathcart (eds.) INTER MEDIA 3rd ed.(New York: Oxford University Press, 1986). 14) Lewis Coser, The Functions of Social Conflict. (Glencoe, Il: The Free Press, 1956); Georg Simmel, Conflict: The Web of Group Affiliations. (N.Y.: The Free Press, 1955 [1908]). 15) Herbert Blumer, Symbolic Interaction: Perspectives and Method. (Englewood Cliffs, N.J.: Prentice Hall, 1969). 16) Leon V. Sigal, Reporters and Officials: The Organization and Politics of Newsmaking.(Lexington, MA.: D.C. Heath, 1973). 17) Jane D. Brown, Carl R. Bybee, Stanley T. Wearden, and Dulcie M. Straughan, "Invisible power: newspaper news sources and the limits of diversity," Journalism Quarterly. 64: 45-54 (Spring, 1987). 18) Kathleen A. Hansen, "Source diversity and newspaper enterprise journalism." Journalism Quarterly. 68(3): 474-482, 1991. 19) Teun A. van Dijk, Communicating Racism: Ethnic Prejudice in Thought and Talk. (London: Sage, 1987). 20) Dan Berkowitz and James V. TerKeurst, "Community as interpretive community: rethinking the journalist-source relationship," Journal of Communication. 49(3) Summer 1999, pp. 125-136. 21) Andrew Hacker, Two nations: black and white, separate, hostile, and unequal. (N.Y.: Scribner, 1992). 22) David Weaver, and S.N. Elliott, "Who sets the agenda for the media? A study of local agenda-building." Journalism Quarterly, 62(1): 87-94. (1985). 23) Olien, Donohue, and Tichenor, "Conflict, consensus, and public opinion." 24) Olien, Donohue, and Tichenor, "Conflict, consensus, and public opinion," p. 25) D.B. Hindman, "Community newspapers, community structural pluralism, and local conflict with non local groups," Journalism & Mass Communication Quarterly. 73 (4): 1996, pp. 708-721. 26) Stanley Fish, Is there a text in this class? The authority of interpretive communities. (Cambridge, MA.: Harvard University Press, 1980). 27) Gans, Deciding What's News.; Doris J. Giago and Bill Huntzicker,"Tips in covering and portraying American Indians," in The American Indian and the Media. Tim Giago (ed.) (Minneapolis: National Conference on Christians and Jews; Minnesota-Dakota Region, 1991;13-15). 28) Gans, Deciding What's News. 29) Leon V. Sigal, "Sources make the news," pp. 9-37 in Robert Manoff and Michael Schudson (eds.), Reading the News. (N.Y.: Pantheon Books, 1986). 30) Robert A. Hackett, "A hierarchy of access: aspects of source bias in Canadian TV news," Journalism Quarterly 62(2), summer 1985, pp. 256-265, 277. 31) Jack L. Daniel and Anita L. Allen, "Newsmagazines, Public Policy, and the Black Agenda." Pp. 23-45, in Geneva Smitherman-Donaldson and Teun A. van Dijk (eds.), Discourse and Discrimination (Detroit: Wayne State University Press, 1988). 32) Daniel and Allen, "Newsmagazines, public policy, and the Black Agenda." 33) Daniel and Allen, "Newsmagazines, public policy, and the Black agenda." 34) Daniel and Allen, "Newsmagazines, public policy, and the Black agenda." 35) Daniel and Allen, "Newspapers, public policy, and the Black agenda." 36) Michael Parenti, Inventing Reality (N.Y.: St. Martin's Press, 1986). 37) Olien, Donohue, and Tichenor, "Conflict, consensus, and public opinion." 38) Olien, Donohue, and Tichenor, "Conflict, consensus, and public opinion." 39) Teun van Dijk, , Elite Discourse and Racism. (London: Sage, 1993); Sigal, "Sources make the news"; 40) Robert Park and E. Burgess, "Introduction to the Science of Sociology. (Chicago: University of Chicago Press, 1921). 41) Gans, Deciding What's News. 42) Gans, Deciding What's News. 43) Alexis de Tocqueville, Democracy in America, II Trans. Henry Reeves. (N.Y: Alfred A. Knopf, 1946). 44) Quoted in Andrew Hacker, Two nations: black and white, separate, hostile, and unequal.(N.Y.: Scribner, 1992), p. 215. 45) John F. Dovidio and Samuel L. Gaertner, "Prejudice, discrimination and racism: historical trends and contemporary approaches." pp. 1-34 in J.F. Dovidio and S.L. Gaertner (eds.) Prejudice, discrimination and racism. (Orlando: Academic Press, 1986). 46) James M. Jones, "Racism: a cultural analysis of the problem." pp. 279-314 in J.F. Dovidio and S.L Gaertner (eds.), Prejudice, discrimnination and racism. 47) Robert M. Entman, "Modern racism and the images of Blacks in local television news," Critical Studies in Mass Communication. 7:332-345, 1990; Robert Entman, "Blacks in the news: television, modern racism and cultural change." Journalism Quarterly.69(2): 341-361, 1992; Robert Entman, "Representation and reality in the portrayal of blacks on network television news," Journalism Quarterly. 71(3): 509-520, 1994; Campbell, Race, Myth and the News; Hemant Shah and Michael C. Thornton, "Racial ideology in U.S. mainstream news magazines coverage of Black-Latino interaction, 1980-1992." Critical Studies in Mass Communication. 11:141-161, 1994; Jane Rhodes, "The visibility of race and media history." Critical Studies in Mass Communication. 10(2): 184-190, 1993. 48) James S. Ettema, "Press rites and race relations: a study of mass- mediated ritual." Critical Studies in Mass Communication. 7(4): 309-331, 1990; Entman, "Blacks in the news;" Entman, "Modern racism and the images of blacks in local television news." 49) Entman, "Representation and reality in the portrayal of blacks on network television news;" Campbell, Race, Myth & the News. 50) John B. McConahay, "Modern racism, ambivalence, and the modern racism scale," pp. 91-125 in J. Dovidio and S. Gaertner (eds.), Prejudice, Discrimination, and Racism. (Orlando: Academic Press, 1986). 51) S.Jhally and J. Lewis, Enlightened racism: the Cosby Show, audiences, and the myth of the American dream. (Boulder, CO.: Westview Press, 1992). 52) D.O.Sears, "Symbolic racism," pp. 53-84, in P.A. Katz and D.A. Taylor (eds.), Eliminating Racism. (N.Y.: Plenum, 1988); Christopher Campbell, Race, Myth, and the News. (Thousand Oaks, CA.: Sage, 1995). 53) McConahay, "Modern racism." 54) McConahay, "Modern racism." 55) Sears, "Symbolic racism." 56) Hacker, Two nations: black and white. 57) Hacker, Two nations: black and white. p. 210. 58) Patricia W. Linville and Gregory W. Fischer, "Stereotyping and perceived distribution of social characteristics: an application to ingroup-outgroup perception." pp. 165-208 in J.F. Dovidio and S.L. Gaertner, (eds.), Prejudice, discrimination and racism. 59) Linville and Fischer, "Stereotyping and perceived distributions of social characteristics." 60) Entman, "Blacks in the news," p. 345. 61) Entman, "Blacks in the news." 62) Entman, "Representation and reality." 63) Entman, "Representation and reality," p. 510. 64) Entman, "Representation and reality," p. 512. 65) Entman, "Representation and reality," p. 513. 66) Entman, "Representation and reality," p. 513. 67) Celeste M. Condit and J. Ann Selzer, "The rhetoric of objectivity in the newspaper coverage of a murder trial." Critical Studies in Mass Communication. 2(3): 197-216, 1985. 68) Entman, "Representation and reality." 69) Entman, "representation and reality," p. 515. 70) Andy Court, "Can the press do the right thing?" p. 27. 71) Kirk A. Johnson, "Black and white in Boston;" Entman, "Representation and reality." 72) M.L Stein, "Racial stereotyping and the media." Editor & Publisher. Aug. 6, 1994, pp. 12-13. 73) These are: The National Association of Black Journalists, the Asian American Journalists Association, National Association of Hispanic Journalists, and the Native American Journalists Association. 74) Stein, "Racial stereotyping and the media," p. 12. 75) Roland Barthes, Mythologies. (N.Y.: Hill & Wang, 1972 [1957]). 76) Stuart Hall, "The whites of their eyes: racist ideologies and the media." pp. 18-22 in Gail Dines and Jean M. Humez (eds.), Gender, Race and Class in Media. (Thousand Oaks: Sage, 1995). 77) Hall, "The whites of their eyes," p. 19. 78) Ronald T. Farrar, Mass Communication: An Introduction to the Field. (Madison: Brown & Benchmark, 1996). 79) Ben Bagdikian, "Democracy and the media," pp. 176-184 in The Media Monopoly. (Boston: Beacon Press, 1983); James S. Featherstone, "Duplication and newspaper content in contrasting ownership situations," Journalism Quarterly 55: 549-554, 1978; Ronald Wagenberg and Walter Soderlund, "The influence of chain ownership on editorial comment in Canada," Journalism Quarterly. 52: 93-98, 1975. 80) Columbia Journalism Review. March/April 1987, p. 17. 81) Theodore Glasser, David S. Allen, and S. Elizabeth Blanks, "The influence of chain ownership on news play: a case study." Journalism Quarterly. 66(3):607-614, 1989. 82) Roya Akhavan-Majid Anita Rife, and Sheila Gopinath, "Chain ownership and editorial independence: a case study of Gannett newspapers." Journalism Quarterly. 68(1/2): 59-66, 1991. 83) Cecile Gaziano, Chain newspaper homogeneity and presidential endorsements, 1972-1988," Journalism Quarterly 66(4):836-845, 1989. 84) Basheda, "When should you quote minority sources?" 85) Basheda, "When should you quote minority sources?" 86) Hansen, "Source diversity and newspaper enterprise journalism." 87) Dominic Lasorsa and Stephen Reese, "News source use in the crash of 1987: a study of four national media." Journalism Quarterly. 67(1): 60-71, 1990. 88) The papers used in this study are: The Clarion-Ledger (of Jackson, Ms.), Detroit News, Cincinnati Enquirer, and Tucson Citizen (all owned by Gannett); Miami Herald, Philadelphia Inquirer, San Jose Mercury News, Fort Worth Star Telegram, Detroit Free Press (all owned by Knight-Ridder); Birmingham Post Herald, Cincinnati Post, Rocky Mountain News, Alburqueque Tribune, and Ventura County Star (all owned by Scripps-Howard). 89) Olien, Donohue, and Tichenor, "Conflict, consensus, and public opinion." 90) Douglas Blanks Hindman, "Community newspapers, community structural pluralism, and local conflict with non-local groups." Journalism & Mass Communication Quarterly 73(3): 708-721, 1996. 91) Berkowitz and Terkeust, ""Community as interpretive community: rethinking the journalist-source relationship." 92) Olien, Donohue, and Tichenor, "Conflict, consensus, and public opinion"; Douglas Blanks Hindman, Robert Littlefield, Ann Preston, and Dennis Neumann, "Structural pluralism, ethnic pluralism, and community newspapers." Journalism & Mass Communication Quarterly 76(2): 250-263, 1999. 93) Entman, "Representation and reality in the portrayal of blacks on network television news." 51) E. Hynds and C. Martin, "How non-daily editors describe status and function of editorial pages," Journalism Quarterly. 56(2): 318-323, 1979; R.W. Budd, "Attention score: a device for measuring news 'play'," Journalism Quarterly 41(2): 259-262, 1964. Shannon Rossi Martin, "Proximity of event as factor in selection of news sources," Journalism Quarterly 65(4) winter 1988, pp. 986-989, 1043. Dominic L. Lasorsa and Stephen D. Reese, "News source use in the crash of 1987: a study of four national media, Journalism Quarterly 67(1): spring 1990, pp. 60-71. Kathleen A. Hansen, "Source diversity and newspaper enterprise journalism." Journalism Quarterly 68(3): autumn 1991, pp. 474-482 Patricia Aufderheide, "Controversy and the newspaper's public: the case of Tongues Untied" Journalism Quarterly 71(3): autumn 1994, pp.499-508. Gertrude J. Robinson, "Making news and manufacturing consent: the journalistic narrative and its audience," pp. 348-369 in Theodore Glasser and Howard E. Sypher (eds.), Public Opinion and the Communication of Consent. (N.Y.: The Guilford Press, 1995). R. J. Rummel, Libertarianism and international violence," Journal of Conflict Resolution 27(1), 1983, pp. 27-71.