FRAMING ANALYSIS OF THE NEW YORK TIMES AND THE GUARDIAN REGARDING MAD COW DISE ASE FROM 1986 TO 2002 By SEUNG HOON HAN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN MASS COMMUNICATION UNIVERSITY OF FLORIDA 2004
Copyright 2004 by Seung Hoon Han
ACKNOWLEDGMENTS I would like first and foremost to thank my committee chair, Dr. Spiro Kiousis, for his continued support of this thesis. I would also like to thank Dr. Michael Mitrook and Dr. Debbie Treise for their support. Finally, I would like to thank my wife, Bo Hyun; and my daughter, Ji Soo for their continued support of my research and my general well being iii
TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iii LIST OF TABLES .............................................................................................................vi ABSTRACT ......................................................................................................................vii CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................6 Mad Cow Disease.........................................................................................................6 BSE in the United States...............................................................................................8 Media Studies of BSE.................................................................................................10 Framing.......................................................................................................................11 Framing and Public Relations.............................................................................13 Phases of Crisis...........................................................................................................14 Information Sources....................................................................................................16 Tone of Media Coverage of Crisis..............................................................................19 Crisis and Framing Studies.........................................................................................20 Framing Analysis on Cross Cultural and International Setting..................................21 Research Questions and Hypotheses..........................................................................23 3 METHODOLOGY.....................................................................................................25 The Sample and Unit of Analysis...............................................................................25 Operational Definitions of Variables..........................................................................26 Issue.....................................................................................................................27 Source..................................................................................................................28 Tone of Media.....................................................................................................29 Inter-coder Reliability.................................................................................................29 Data Analysis..............................................................................................................30 4 FINDINGS..................................................................................................................31 Content Analysis.........................................................................................................31 Research Questions 1, 2, and 3............................................................................32 iv
Research Questions 4, 5, and 6............................................................................34 Hypotheses 1, 2, and 3.........................................................................................36 5 DISCUSSION AND CONCLUSIONS......................................................................48 Discussion...................................................................................................................48 RQ1: Are there Significant Changes in The Guardianâ€™s Dominant Issue Frame before and after March 21, 1996?....................................................................48 RQ2: Are there Significant Changes in The Guardianâ€™s Dominant Information Sources before and after March 21, 1996?......................................................49 RQ3: Are there Significant Changes in The Guardianâ€™s Tone of Media Coverage (Positive, Negative, and Neutral) before and after March 21, 1996?................................................................................................................49 RQ4: Are there Differences in the Observed Issue Frames between The New York Times and The Guardian from 1996 to 2002?........................................50 RQ5: Are there Differences in the Observed Information Sources between The New York Times and The Guardian from 1996 to 2002?................................51 RQ6: Are there Differences in the Observed Tone of Media toward the Issue between The New York Times and The Guardian from 1996 to 2002?...........51 H1: During the Pre-Crisis Phases of Each Country, the Observed Issue Frames of The New York Times will Show No Difference When Compared to those of The Guardian...............................................................................................52 H2: During the Pre-Crisis Phases of Each Country, the Observed Information Sources used by The New York Times will Show No Difference when Compared to those used by The Guardian......................................................53 H3: During the Pre-Crisis Phases of Each Country, the Observed Tone taken toward the Issue in The New York Times will Show No Difference when Compared to that taken in The Guardian........................................................53 Conclusions.................................................................................................................54 Limitations..................................................................................................................56 Future Research..........................................................................................................57 APPENDIX CODING PROTOCOL................................................................................58 LIST OF REFERENCES...................................................................................................61 BIOGRAPHICAL SKETCH.............................................................................................67 v
LIST OF TABLES Table page 4-1 The Guardian before and after 1996 on Media Issue Frame....................................38 4-2 The Guardian before and after 1996 on Media Sources..........................................39 4-3 The Guardian before and after 1996 in Tone of Media Coverage...........................40 4-4 The Guardian before and after 1996 in Tone of Media Coverage...........................40 4-5 Guardian before and after 1996 in Tone of Media Coverage (by Issue).................41 4-6 The New York Times and The Guardian after 1996 on Media Issues......................41 4-7 The New York Times and The Guardian after 1996 on Media Sources...................42 4-8 The New York Times and The Guardian after 1996 in Tone of Media Coverage....43 4-9 The New York Times and The Guardian after 1996 in Tone of Media Coverage (by News Source).....................................................................................43 4-10 The Guardian before and after 1996 in Tone of Media Coverage (by Issue)..........44 4-11 The New York Times and The Guardian before 1996 on Media Issues...................44 4-12 The New York Times and The Guardian before 1996 on Media Sources................45 4-13 The New York Times and The Guardian before 1996 in Tone of Media Coverage...................................................................................................................46 4-14 The New York Times and The Guardian before 1996 in Tone of Media Coverage (by News Source).....................................................................................................46 4-15 The Guardian before and before 1996 in Tone of Media Coverage (by Issue).......47 vi
Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts in Mass Communication FRAMING ANALYSIS OF THE NEW YORK TIMES AND THE GUARDIAN REGARDING MAD COW DISEASE FROM 1986 TO 2002 By Seung Hoon Han August 2004 Chair: Spiro K. Kiousis Cochair: Michael A. Mitrook Major Department: Journalism and Communications Since March 21, 1996, when the British Government officially acknowledged that a human had been infected by Mad Cow Disease, the issue of Mad Cow Disease has spread around the world. This study examined international newspaper coverage of Mad Cow Disease to compare media frames used in two countries: the United States which is just starting to deal with the issue, and the United Kingdom which is dealing with the after effects of the outbreak. For my study, a longitudinal content analysis was performed to examine coverage of the Mad Cow Disease issue in The New York Times and The Guardian. Two hundred and seventy five articles from The Guardian were analyzed from before the British governmentâ€™s announcement in 1996, as well as articles appearing in The Guardian after the announcement and articles appearing in The New York Times after the announcement. My study examined the dominant issue frames used in these stories, the primary information sources used in these stories, and valence of the coverage, by comparing vii
different media sources in different time periods. My study used a three-stage model of crisis communication in which â€œpre-crisisâ€ refers to the period in either the United States or the United Kingdom before any official government announcement regarding a human. The findings showed that the announcement in 1996 did not significantly change the media issue frame and news source selection in The Guardian, possibly because the crisis in 1996 was not truly unexpected by the British media. During the pre-crisis session, the British media already knew about the effects of the disease and may have been waiting only for the governmentâ€™s official acknowledgement. However, after the announcement, the tone of The Guardianâ€™s Mad Cow Disease articles turned decidedly negative. Through detailed analysis, I found that the major reason for this negativism might have been the British governmentâ€™s mishandled policy regarding Mad Cow Disease. Analysis showed that The New York Times used issue frames and source selection similar to those in The Guardian. However, The New York Times presented a more neutral tone than its British counterpart did, possibly because there had not been a comparable public relations blunder in the United States. Lastly, the pre-crisis stage in both countries showed similarities both in news source selection and tone of articles. Two conclusions can be drawn from these results. First, the major news outlets might have their own shared and standardized newsgathering processes. Second, my study demonstrates the theoretical validity of the three-stage model of crisis communication constructed from the experiences of public relations professionals. viii
CHAPTER 1 INTRODUCTION On March 21, 1996, the British Governmentâ€™s denial of human health risks related to bovine spongiform encephalopathy (BSE) changed sharply. British Health Secretary Stephen Dorrell was forced to make an emergency statement, saying that scientists had now found a possible connection between Mad Cow Disease, or Bovine Spongiform Encephalopathy (BSE), and its fatal human infection, so-called Creutzfeldt Jakob Disease (CJD). Since then, the issue of Mad Cow Disease has spread around the world. As of July 2002, there have been 134 definite or probable cases of variant CJD reported worldwide, mostly in European countries. Beyond European countries, even in Japan, more victims were discovered in August of 2002 (â€œWorld Watch,â€ 2002). Although the United States government has not yet confirmed a case of Mad Cow Disease, in Canada the largest beef exporter to the United States a positive case of Mad Cow Disease was officially acknowledged by the Canadian Food Inspection Agency (CFIA) in May 20, 2003 (Park, 2003 ). The link between BSE and CJD was discovered in 1996; however, there has still been no effective medicine developed against the disease. With an incubation period of up to 30 years, people who ate infected meat in the late 1980s might not yet be showing any signs of the disease, but eventually the number of people displaying symptoms could be huge (â€œBitter price,â€ 1996). With such fatal effects, this disease requires advance crisis planning. As Coombs and Holladay (2001) point out, â€œFor the preparation of a sound and effective crisis 1
2 response in media communication, analysis of a crisis situation is critical.â€ Previous media research concerning the BSE issue was mostly focused on the U.K., the origin of BSE, or countries in the European Union (E.U.) (Brookes, 1999, Lanska 1998, Rowe, Frewer, & Sjoberg, 2000). So far little research on Mad Cow Disease has compared media coverage in United States vs. the United Kingdom. The United States ranks as the worldâ€™s third largest beef exporter, as well as the largest beef consumer, by importing 1.5 million tons of beef each year (USDA, 2003). Considering the social and economic effect BSE has had on the United States, it is worth examining the different ways in which this international crisis has been covered by comparing national media coverage in the United States to national media coverage in the United Kingdom. According to Hofstede (1980), the United States and the United Kingdom are traditionally regarded as belonging to the same cultural category. However, with regard to the issue of Mad Cow Disease, these countries are currently in two different stages of the three-stage crisis model (Birch 1994, Mitchell 1986, Woodcock 1994). The three-stage crisis model comprises the pre-crisis phase and the post-crisis phase. The United Kingdom is currently in the post-crisis phase of the Mad Cow Disease crisis, as it has already gone through the pre-crisis and crisis phases. During the pre-crisis phase (from 1986 to 1996), the first case of Mad Cow Disease was reported, causing increased concerns in the British scientific community and the general public about the possibility of human infection. In the crisis phase (from 1996 to 2000), the British governmentâ€™s official acknowledgement of human infection shocked British society, and prompted countries in the EU to ban beef products imported from the U.K. The post-crisis phase began with publication of the official BSE report by the
3 British government in 2000. The government is still searching for the proper way to compensate victims, as well as for an actual cure for the disease. Meanwhile, in the United States, the issue is still in a â€œpre-crisisâ€ phase. There have been no officially reported cases; those thought to have been victims were shown to have been infected outside of the United States. However, the U.S. public seems to be concerned over a potential outbreak of this disease. One notable example occurred in 1996, when the Oprah Winfrey show discussed the Mad Cow controversy (Ramton & Stauber, 1997). During the show, Oprah stated that, â€œThis has just stopped me cold from eating another burger (p.19)!â€ Oprahâ€™s provocative segment on Mad Cow caused livestock traders on the floor of the Chicago Mercantile Exchange to sell off cattle futures, which fell a penny and a half per pound (the maximum allowable drop for a single dayâ€™s trading). To shed some light on how the media have portrayed this issue, my study investigated framing in a cross-national setting. Previous research on media frames focused primarily on how issues are presented and covered in the news (Entman, 1991; Semetko & Valkenburg, 2000). Through the framing process, journalists simplify an issue or concept into a certain image to advocate or criticize a certain view of an occasion or concern. In addition, the study of framing comparison can reveal, as Entman (1991) noted, â€œthe critical textual choices that framed the story.â€ Therefore, my study applied framing analysis to explore the different ways in which print media in the United States and the United Kingdom have portrayed the issue of Mad Cow Disease. My study focused on finding and comparing Mad Cow Disease in terms of how media outlets framed the issue, in terms of the sources used by different
4 media outlets, and in terms of the tone of media coverage in both the United States and the United Kingdom. Three main goals can be accomplished by comparing the two countries. First, the researcher can outline the issue attributes (governmental policy, economic consequences, public health concern, and science information) with which the media is concerned. Second, media sources can reveal those parties (government, interest groups, corporations, farmers, victims, general citizens, scientists, and media) that have had the greatest impact on media framing of the issue. Finally, the tone of media coverage, relative to the issue and the source, can indicate changes in the tone of media (neutral, positive, and negative) through longitudinal content analysis. My study used two methods of comparison. First, I explored how coverage in a British newspaper changed during the periods immediately before and immediately after the British governmentâ€™s acknowledgement in 1996 that a human had contracted Mad Cow Disease. Second, I compared differences between the United States and the United Kingdom in newspaper coverage of the same issue. A comparison of print media coverage in the U.S. and U.K. was made after the official announcement of BSE human infection in 1996. Then, by comparing the pre-crisis phases in both countries (the United Kingdom from 1985 to 1996 and the United States from 1996 to 2002), I compared whether there was any similarity between the targeted phase periods. This explored framing differences in issue, source, and tone of media regarding Mad Cow disease. Furthermore, by comparing different media and time spans, I examined the validity of the three-stage crisis model.
5 In accordance with my research goals, chapter 2 reviews previous literature on framing, agenda setting, and crisis and media coverage. Chapter 2 also introduces research questions and hypotheses relating to the literature review. Chapter 3 provides methods for using the collected news content to measure framing. Results of the analysis are given in Chapter 4. Conclusions of my study, its limitations, and future research possibilities are summarized in Chapter 5.
CHAPTER 2 LITERATURE REVIEW Mad Cow Disease Bovine Spongiform Encephalopathy (BSE), better known as Mad Cow Disease, was identified in the United Kingdom by the Ministry of Agriculture, Fisheries and Food (MAFF) in November 1986. Researchers believed that BSE contaminated only cows with a protein derivative called â€˜Prionâ€™ and took 1 or 2 years to affect the brain, causing the cowâ€™s brain to degenerate into a form of â€œspongy Swiss cheese,â€ resulting in paralysis and death. They later found that the symptoms are similar in humans, and that it takes an average of 12.4 years for Prion to reach the human brain and cause death. This human variant of BSE is called Creutzfeldt-Jacob Disease (CJD) and 129 people in the U.K. were killed by CJD by 2003 (Kohn 2003). During the 1980s, most experts maintained that the BSE risk to humans was minimal. Dr. David Tyrrell, the first chairman of the Spongiform Encephalopathy Committee (SEAC), wrote publicly in 1988 that â€œany risk as a result of eating beef or beef products is minute. Thus we believe that there is no scientific reason for not eating British beef and that it can be eaten by everyoneâ€ (The BSE inquiry, 2000). In the early 1990s, while neither the scientific community nor the British Health Ministry distributed accurate information about BSE quickly or effectively enough, confirmed cases of BSE escalated, peaking in 1993. Some scientists began to claim that Mad Cow Disease might have been transmitted to humans. According to Lexis Nexis 6
7 database, the amount of British print media (for example, The Guardian, The Times, and The Independent) coverage relevant to the issue increased, but soon decreased after the British governmentâ€™s denial of human infection (Figure 2-1). In 1990, in an attempt to allay public fears, Agriculture Minister John Gunner posed in front of cameras with his daughter and a pair of burgers. He said, â€œMy wife eats beef, my children eat beef. That is everyoneâ€™s absolute protectionâ€ (Cannon, Connor, Hague & Nicholsonlord 1990, p.9). 040801201602001987198819891990199119921993YearArticle The Times The Guardian The Independent Figure 2-1. Number of Articles regarding Mad Cow Disease in British Print Media from 1987 to 1993 (Source: Lexis Nexis) Sir Richard Southwood, who had chaired the Working Party on BSE and its implications, took a more cautious, but similar, view. His report recognized that if his assumption proved wrong, â€œthe implications would be extremely seriousâ€ (The BSE inquiry 2000). These views suddenly shifted after March 1996. On March 20, 1996, British Health Secretary Stephen Dorell made a BSE announcement to the British House of Commons. He said the British government â€œhas concluded that the most likely explanation at present isâ€ that the 10 cases of CJD that had been identified in people under age 42 â€œare linked to exposure to BSEâ€ (Ramton & Stauber 1996, p.183). This official acknowledgement triggered the financial catastrophe
8 that farmers and the government had feared. In the days following the announcement, the meat market was, in the words of the International Meat Trade Association, collapsing like a house of cards (Hornsby, 1996). The crisis predominantly developed as a health crisis; however, Brookes (1999) found that the issue was complicated by the political and economical situations in Britain and the European Union. Public health concerns were eclipsed by explicit questions relating to the beef industry, politics, and the economy, resulting in a â€œbeef war.â€ In 2000, the British government published the official BSE Inquiry Report after conducting the inquiry from 1997 to 1999. The BSE inquiry, chaired by Appeal Court Judge Sir Nicholas Phillips, "reviewed the history of the emergence and identification of BSE and new variant CJD in the United Kingdom, and of the action taken in response to it up to 20 March 1996.â€ Attempting to gather only factual evidence, the inquiry took oral evidence from more than 300 witnesses, read more than 2,000 government files, and published more than 400 written statements. The hearings further investigated the evidence, addressed criticisms, and recalled witnesses, including former agricultural ministers Edwina Currie and Douglas Hogg (â€œBSE: The long search,â€ 2000). BSE in the United States To date, there have been no reported cases of BSE originating in either U. S. citizens or U. S. cattle. This does not mean, however, that cases of the disease have not been discovered in or near the U.S. On October, 2002, The New York Times reported a confirmed case of Mad Cow Disease in a 22-year-old British woman living in Florida (â€œNational Briefing Science and Health,â€ 2002). On November, 2002, The Wall Street Journal reported that a brain-wasting condition similar to Mad Cow Disease was spreading in North American elk and deer (Regaldo, 2002).
9 One of the most notable events related to Mad Cow Disease in the United States was from Oprah Winfreyâ€™s television show in 1996. After the show, beef industry commodity prices abruptly plummeted. James Reagan, the National Cattlemenâ€™s Director of Product Safety said, â€œAll of a sudden the price drops out of futures on beef.â€ Subsequently, a beef industry â€œfood disparagementâ€ lawsuit quickly drove Winfrey into silence (Ramton & Stauber, 1997). This case demonstrated the diseaseâ€™s potential effect on the economy and society in the United States. In November 2001, the Harvard Center for Risk Analysis, with support from the Department of Health and Human Services and the Department of Agriculture, published a report titled â€œEvaluation of the Potential for Bovine Spongiform Encephalopathy in the United States.â€ The report stated that BSE posed an "extremely low" risk to consumers and agriculture in the United States (Thomson, 2001). However, the analysis was not base on hard data, but solely on computer-generated â€œwhat ifâ€ scenarios. The results indicated that the risk of Mad Cow Disease was only a low risk if real life followed assumptions made in a computer model. â€œCould BSE show up in this country?â€ asked George Gray, acting director of the Harvard Center. â€œItâ€™s unlikely, but possibleâ€ (Webb, 2002). Most recently, in Canada, a close neighbor and the biggest beef supplier to the United States, a confirmed case of Mad Cow Disease was detected on May 20, 2003. A laboratory in the United Kingdom confirmed that a cow slaughtered in the Canadian province of Alberta in January 2003 had Mad Cow Disease. U.S. Secretary of Agriculture Ann M. Veneman announced the next day that imports of live cows, goats, sheep, and meat products from Canada had been temporarily banned as a precaution. "Information
10 suggests that . . . the possibility of transmission to animals in the United States is very low," said Veneman (â€œMad cow disease,â€ 2003). Media Studies of BSE Several media studies regarding the issue of BSE were conducted in European countries. Using a cross-national analysis approach, Rowe, Frewer, and Sjoberg (2000) examined how newspapers in Sweden and the United Kingdom characterized both the Chernobyl accident and BSE risks. They found that BSE was discussed using a greater number of characterizations in the U.K., while Chernobyl was reported using more characterizations in Sweden. This indicates that journalists reporting about issues that may be directly hazardous to their own countries tend to be alarmist rather than reassuring, and rarely use statistics to express degrees of risk. Brookes (1999) identified nationalistic characteristics in British newspaper coverage of the BSE issue. He found that, even during the coverage of this crisis, there were stereotypical representations of nationalism in news coverage such as international politics, sport and war. He concluded that the British news media played a pivotal role in encouraging a sense of nationalism among the British people. As the crisis developed from a health crisis to a political story about Britain and the European Union, public health concerns were replaced by questions of national identity relating to the beef industry and the economy, culminating in the so-called â€˜beef war.â€™ Therefore, there is a possibility that the national uniqueness used in each report on a crisis can be measured. In a public relations analysis, Verbeke, Ward, and Avermaete (2002) focused on the evaluations of publicity campaigns relative to regulations concerning identification and registration of bovine animals and compulsory labeling of beef and beef products in Belgium during September 2000. They examined consumer surveys before and after the
11 publicity event. While the mass media campaign scored well in terms of recall rate, informative content evaluation and restoration of consumer confidence in beef were not successful. However, little media research has been done involving a framing analysis of media coverage of Mad Cow Disease, and most studies have been limited to the United Kingdom and European countries. Framing Among many scholars, Gitlin is regarded as the first to introduce the concept of framing to media research. He described framing as â€œpersistent patterns of cognition, interpretation, and presentation, of selection, emphasis and exclusion by which symbol handlers routinely organize discourse, whether verbal or visualâ€ (Gitlin, 1980, p.7), and suggested that both journalists, who report the news, and audiences, who hear the reports, are affected by media frames in terms of organizing the world and perceived reality. With the help of framing techniques, journalists are able to arrange and deliver a large volume of information efficiently to the rest of the world (Gitlin, 1980). Pan and Kosicki (1993) conceived that a news media frame is a cognitive device used in the encoding, interpretation, and retrieval of information. They argued that â€œevery news story contains a dominant theme which functions as the central organizing ideaâ€ (p. 58). This dominant theme is the frame of the story. According to Pan and Kosickiâ€™s classification, one type of structure for news discourse is the thematic structure, in which stories contain hypothesis-testing features, or â€œissue storiesâ€ (p.60). A theme is presented or implied, while events are cited, sources are quoted, and propositions are pronounced. In this story type, events, sources and quotations all serve to support the hypothesis. Pan and Kosicki noted the basic units of thematic structure as consisting of a summary, including the headline and lead, as well as
12 the rest of the story. They concluded that â€œthe units of thematic structure only can result from analysis of an entire news articleâ€ (p.61). Entman (1993) argued that â€œframing is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and /or treatment recommendation for the item describedâ€ (p.52). Additionally, Entman identifies â€œsalienceâ€ as the key to understanding and describing â€œperceived realityâ€ rather than objective reality (p. 52). Entman (1991) provides another insight for framing analysis. He suggests that â€œnews frames consist of the key words, metaphors, concepts, symbols and visual imaginary in a news narrativeâ€ (p. 7). They simplify an issue or concept into a certain image and are manipulated by journalistic methods for advocating or criticizing a certain view of an occasion or concern. The basic elements of framing comprise sizing, magnifying or shrinking elements of the reality description in order to make the reality more or less salient. Therefore, we can identify media framing by observing its key words, metaphor usage, article size, and descriptive patterns. While researchers use various terms for defining the news mediaâ€™s framing or frame, there are some common points among these variations. Heo (2001) summarized the points as follows. First, framing must be related to journalistic decisions about what kind of news should be reported, and second, to which news media outlet covers an issue and how it reports the item. Therefore, we can summarize the function of frames in mass media issue coverage as a device of selection and description. These functions occur
13 throughout the whole journalistic work process, from inception in a journalistâ€™s mind to culmination in editing meetings. In summary, framing analysis is aimed at examining the underlying meaning and pattern of portraying certain issues or events in a particular way by the media under specific social contexts. The unit of analysis can be a whole news article. The socially constructed news is studied according to the sizing of analyzing elements and dominant forms of interpretations and arguments. Framing and Public Relations In this situation, public relations practitioners occupy positions ideally suited to frame issues in a way likely to advance both public and organizational interests. Their traditional roles as media and community liaisons offer opportunities for framing issues of interest, as do their less-recognized roles as lobbyists, negotiators, and environmental scanners. Miller and Reichert (2001) suggested the application of framing in the role of public relations. The authors stated the following: Our main thesis is that stakeholder attempts to frame issues interact with fundamental human values in ways that affect the relative attractiveness of policy options to the public and policy makers. Stakeholders articulate their positions and then monitor pubic responses to those articulations. If a stakeholderâ€™s articulation resonates positively with the public, then that group will intensify its efforts. On the other hand, when an articulation resonates negatively, the stakeholder group will change its articulation or withdraw from debate. We label our position â€œSpiral of Opportunity.â€ (pp. 108-109) One of the most significant studies in framing theory literature in public relations is from Hallahan (1999). He suggested the following: Framing is a potentially useful paradigm for examining the strategic creation of public relations messages and audience responses. Based on a literature review across disciplines, this article identifies 7 distinct types of framing applicable to public relations. These involve the framing of situations, attributes, choices, actions
14 issues, responsibility, and news. The premise of this article is that framing theory provides a potentially useful umbrella for examining what occurs in public relations. (p. 205) This article showed a concise and clear linkage with this theoretical perspective to the public relations scholar and practitioner. He also said, â€œFraming is a critical activity in the construction of social reality because it helps shape the perspectives through which people see the worldâ€ (p.206). Hallahan suggested framing could be the core of public relations: framing is not merely useful but is essential to public relations. In developing programs, public relations professionals fundamentally operate as frame strategists, who strive to determine how situations, attributes, choices, actions, issues, and responsibility should be posed to achieve favorable outcomes for clients. Framing decisions are perhaps the most important strategic choices made in a public relations effort. It is out of strategic framing that public relations communicators develop specific themes (i.e., key messages or arguments t hat might be considered by publics in the discussion of topics of mutual concern).â€ (p.235) Finally, Knight (1999) advised that framing could be a versatile tool in the field of public relations. If framing is used consistently both within and outside organizations, it represents a powerful mechanism through which public relations practitioners can mediate debates that are related to public policy. Phases of Crisis In categorizing the flow of social issues, political scientist Anthony Downs (1972) described what he called the â€œissue-attention cycleâ€; the rise and fall of an issue on the public agenda. As Downs observed, â€œEach of these problems suddenly leaps into prominence, remains there for a short time, and then â€“ though still largely unresolved â€“ gradually fades from the center of public attentionâ€ (p.38). The cycle contains five stages that vary in duration, but occur most often in the following sequence: pre-problem stage,
15 alarmed stage, realizing the cost of problem solution, decline of public interest stage, and post-problem stage. Lipsky and Olson (1976) applied this concept to describe the racial crisis in 1970s. They said that crises were highly cyclical, ending predictably through attrition of attention. Therefore, the riot-induced crisis was one in a series of highly visible concerns that appeared and disappeared when its dynamics were played out. Crisis as an issue has also attracted some of public relations researchersâ€™ interest. According to Grunig (1992), he stated â€œwhen conflict occurs, publics â€˜make an issueâ€™ out of the problemsâ€ (p.13). Heath (1990) said issue management as a crisis management includes the identification of and actions taken to affect issues. Because some issues can develop into crises, issues management contributes to crisis scanning (Gonzalez-Herrero & Pratt, 1996). Crisis management approaches provide several suggestions for outlining the crisis phases. Fink (1986) uses a medical illness metaphor to identify four stages in the crisis life cycle: Prodromal: Clues or hints that a potential crisis exists or begins to emerge. Acute or crisis breakout: A triggering event occurs along with the attendant damage. Chronic: The effects of the crisis remain as clean up work is still in progress. Resolution: There are some clear signals that the crisis is no longer a concern to the public â€“ it is over. Mitroff (1994) stressed a more prescriptive model. His model consists of five phases of crisis: Signal detection: Crisis warning signs should be identified as early as possible and acted on to prevent crisis.
16 Probing and prevention: Analyzing crisis factors to reduce their potential harm. Damage containment: Outbreak and organization to try and minimize the spread of the crisis. Recovery: Try to get the organization back to normal operation. Learning: Reviewing and criticizing cases and accepting feedback. A three-stage model that encompasses the previous models has been recommended by a variety of crisis communication experts, including Birch (1994), Mitchell (1986) and Woodcock (1994). Its three stages are: Pre-crisis stage, which includes all aspects of crisis preparation: Prodrominal, signal detection and probing would be included in the pre-crisis stage. Crisis stage, which includes the actions taken to cope with the crisis or trigger event-damage containment: Crisis breakout are categorized in this stage. Post-crisis stage, which includes the period after the crisis has been resolved. Learning and resolution occur during this stage. This study uses the three-stage model not only because of its application parsimony and its inclusiveness, which can encompass previous crisis models, but also its similarity to the issue cycle. Information Sources Regarding the effect of information source, Shoemaker and Reese (1996) write that â€œInformation sources have a tremendous effect on mass media content, because journalists canâ€™t include in their news reports what they donâ€™t know sources may influence the news in subtle ways by providing the context within which all other information is evaluatedand by monopolizing the journalistsâ€™ time so that they donâ€™t have an opportunity to seek out sources with alternative viewsâ€ (p. 150). Gans (1979) defined sources as the people from whom journalists obtain news by interview or observation. Gans argued that while sources theoretically come from all
17 walks of life, they are often the elite and powerful who shape society, and â€œtheir recruitment and their access to journalists reflect the hierarchies of nation and societyâ€ (p.119). Bagdikian (1974) explained why journalists tended to depend on official sources. He asserted the mediaâ€™s dependence on routine sources could reflect issues of expediency, lack of initiative, or simply the absence of other sources. He also attributed the dependence on routine sources rather than individual investigative work to the growth of the wire services, and, more recently, in the diffusion of satellite communications that facilitate media re-broadcasting of events from most anywhere in the world without having to have specific reporters on the scene. McLeod and Hertog (1998) observed that journalists also have multiple incentives to use government officials as sources on stories, including the prestige they add to a story, their assumed objectivity, and their ready availability in the time-sensitive cycle of news production. As a result, Gans (1979) argued that heavy reliance by media on officials representing both government and private-sector entities presents a very specific (official) picture of society and its institutions. Terkildsen, Schnell and Ling (1998) argued that the reporterâ€™s role in framing is news reporting, not creating news. They stated that journalists act as intermediaries, not catalysts in the framing process. In turn, reporters point to pressure groups and government officials as the originators of issue frames. Turk (1986) emphasized the role of public relations practitioners as information subsidies to change the contents of media. Public relations practitioners use information subsidies to systematize their attempts, on behalf of organizations and institutions for which they work, in order
18 to influence media content and the opinions of those who rely upon the media for information. They are charged with the responsibility of providing â€˜officialâ€™ organizational versions of reality to the media in the hope that an organizationâ€™s view of what is real and important will be incorporated into media content (p.16). In this situation, the success of either government or interest groups depends on their ability to expand or contain an issue (Baumgartner & Jones 1993). Defining an issue over time and adapting those definitions to fit the changing political and social climate determines the playersâ€™ ability to inspire and mobilize support for their actions. To achieve this goal, interest groups must rely on mass media in order to have their interpretations of reality inserted into the public debate. However, due to a variety of constraints imposed by the media and society, these groups cannot be assured that the media will attune to their specific message, let alone promote it to the public. Even if the media picks up an interest groupâ€™s theme, the position advocated by the group may not be transmitted in its totality, the message may be placed in an unfavorable context, or, in the worst-case scenario, the interest groupâ€™s rhetoric may be substantially altered. In other words, once pressure groups put forth an issue perspective, the mediaâ€™s subsequent actions are beyond the scope of the pressure groupsâ€™ control. Even though the interest groups successfully deliver their message through the media, as Zaller (1992) suggested, ordinary citizens have limited attention and information-processing capabilities. They will take only a limited set of cognitions from the many stored in long-term memory while considering an issue. Cognitions that have been primed through recent mentions by a communicator and will be at the â€œtop of the headâ€that is, easily accessed from memory, and will be far more likely to influence opinion than unprimed, inaccessible cognition.
19 Therefore, as Davis (2000) noted, a wide range of organizations have begun adopting public relations as a means of achieving particular goals through media coverage. At the same time, media institutions, operating under tighter editorial budgets, have become more dependent on information supplied by external sources. The two trends have resulted in both the sudden growth of the professional public relations sector and changes to existing patterns of source access. Due to the recent trends, it is important to focus on source activity when seeking to explain the production of any media coverage (Schlesinger, 1990). Furthermore, it is particularly important when examining crisis or risk reporting because controversies about crisis or risk often involve exceptionally fierce source competition. Tone of Media Coverage of Crisis Discussing media coverage of crises, Dunwoody (1992) concluded: When it come to crisis coverage, it seems that the mass media can do nothing right. They are regularly accused of bias, sensationalism, inaccuracy, indifference and of being simplistic and polarized. If we believe the wealth of commentary that has spilled across the printed page since such landmark events as Love Canal and Three Mile Island in the late 1970s, then the mass media is â€“ in a word â€“ lousy at conveying appropriate notions of risk to general audiences (p.75). Brodie, Brady and Altman (1998) conducted a content analysis of media coverage of managed care. They found that coverage differed substantially depending on the media source, and that a change in information source can result in a change in the tone of the media (neutral, negative, and positive). Wiegman (1989) and colleagues evaluated the effect of news coverage on audience reaction to crisis. They found that readers who subscribed to newspapers with higher amounts of negative tone of coverage had a tendency to have more negative opinions of risk, thought the risks were more dangerous, and were more likely to gather more
20 information on the risk. Mazur (1990) reported that it is the quantity of research rather than the quality of research for risk reporting that produces concerns among the public. Frewer, Miles, and Marsh (2002) examined media tone during the risk-coverage of genetically modified food (GMO) in Britain in 1999. They compared the media tone data before, during, and after the event, and concluded that increased negative reporting on the part of media can amplify peopleâ€™s risk perception. Crisis and Framing Studies Crisis is one of the topics most frequently studied by many framing scholars. The popular definition of a crisis is â€œbig troubleâ€ in the form of a sudden and uncertain event that eventually induces massive media coverage. Therefore, a crisis can provide a good resource for media framing research. Entman (1991) analyzed U.S. newsmagazine coverage of both a Korean Airlines (KAL) plane shot down by a U.S.S.R. fighter in 1983 and an Iran Air Flight plane shot down by a U.S. naval ship in 1988. Through the analysis, he determined that the U.S. media had adopted a frame in reporting foreign events during the Cold War era. Currently, framing has entered the limelight for crisis communication studies in public relations because framing can be a valuable method for crisis situation analysis. Palenchar (2001) espoused the importance of framing in the study of crisis communication by stating that framing ultimately can help public relations practitioners understand how key stakeholders navigate through the information environment by better understanding the source of information that is used and how narrative elements frame the risk. One important question about framing in crisis is whether participant groups identified as first sources changed over time. Strong hegemony models suggest that one
21 of the ways in which contesting voices are silenced is by the persistent use of sources representing the dominant perspective (McLeod & Hertog, 1998; Olien et. al., 1989). For example, Noakes and Wilkins (2002) studied media coverage of the Palestinianâ€“Israeli conflict. Media showed Arafat as a representative negotiator for the Palestinians not only by presenting him as a main source of the Palestinian side, but also by characterizing the PLO as a legitimate representative of the Palestinian people, and the Palestinian desire for self-government as just. Gamson and his colleagues (e.g., Gamson & Lasch, 1983; Gamson & Modigliani, 1989) have extensively documented the assortment of frames defining such prominent public controversies as affirmative action, welfare, and nuclear power. To document taxonomy of frames for each issue, Gamson and colleagues first went to primary sources of elite political thought. Having identified the most important issue frames in these sources, Gamson and colleagues then charted the â€œpublic careerâ€ of each frame by tracing its changing prominence over the years in national news magazines, network news programs, editorial cartoons, and syndicated opinion columns. In many cases, abrupt changes in the framing of issues can be traced to specific events; for example, coverage of nuclear power shifted from a largely favorable â€œtechnological progressâ€ framing to an alarming â€œrunaway technologyâ€ frame following the Three Mile Island disaster (Gamson & Modigliani, 1989). In other cases, frame alterations are introduced by elites eager to reshape public debate in terms that favor their causes. Framing Analysis on Cross Cultural and International Setting The impact of culture in communication patterns is well accepted. Hofstede (1991) defined culture as "the software of the mind," indicating that culture is the totality of the learned behavior, the values, the ways of thinking, the attitudes, and the belief systems
22 shared among a people, or a nationality. As Reese (2001) proposed framing proposed frames as â€œorganizing principles that are socially shared and persistent over time, that work symbolically to meaningfully structure the social worldâ€ (p.11). And many framing analyses were done based on international and cultural comparison setting. For example, Zhang (1998) examined coverage of the return of Hong Kong sovereignty from the British to China in 1998. By comparing the content of Xinhua with the official Chinese news agency with that of Associated Press, an American source, he concluded different frames, found in the research, were from culture with a capitalist/democratic system, and culture with a centralized bureaucratic government and economy. Those differences were tended to be stressed when reporting foreign news. Semetko and Valkenburg find that in Europe the news can be episodic and still frame the government as responsible (2000). Other researchers have looked at frames more as categories of topics. For example Wanta and Hu identified 3 frames for foreign news: those that involve the United States and have a degree of conflict, those that have conflict but do not involve the United State and those that involve the United States but do not have conflict (1993). The approach most used in defining a media frame was describing it as a script or schema (Lodge and Hamil, 1986), a frame of reference (Scheufele, 1999) that the news events were presented in to help audiences interpret them (Entman, 1993). The framing typology for this research is based on Semetko and Valkenburgâ€™s (2000) news frame categories. They identified the prevalence of five news frames by integrating from earlier studies on framing and framing effects as follows: conflict frame, human interest frame, economic consequences frame, morality frame, and responsibility
23 frame. The researcher adopted these framing factors to the four frames from preliminary research. The first one was the â€œPublic Health Concern Frame,â€ which treated public concern and victimâ€™s tragedy, which included an emotional human interest frame. The second was the â€œScientific Information Frame,â€ which described disease symptoms and research based on scientific facts. â€œGovernmental Policy Frameâ€ was the third one; this includes the criticisms to the government policy and conflict frame between government and publics. Lastly, Economic Consequences Frameâ€ is adopted for describing economic concerns from BSE. Research Questions and Hypotheses For the purposes of this study, a longitudinal framing analysis was used to examine coverage of Mad Cow Disease issue in both The Guardian and The New York Times, the most prestigious papers in the United Kingdom and in the United States, respectively. This study analyzes news stories in The Guardian, comparing stories from before and after the official disclosure of Mad Cow Disease on March 21, 1996. The research questions are as follows: RQ1: Are there significant changes in The Guardianâ€™s dominant issue frame before and after March 21, 1996? RQ2: Are there significant changes in The Guardianâ€™s dominant information sources before and after March 21, 1996? RQ3: Are there significant changes in The Guardianâ€™s tone of media coverage (Positive, Negative, and Neutral) before and after March 21, 1996? After 1996, Mad Cow Disease gained huge media attention around the world. Comparisons are justified, therefore, between newspapers from two different countries: the coverage of the United Kingdomâ€™s The Guardian and the U.S.â€™s The New York Times. RQ4: Are there differences in the observed issue frames between The New York Times and The Guardian from 1996 to 2002?
24 RQ5: Are there differences in the observed information sources between The New York Times and The Guardian from 1996 to 2002? RQ6: Are there differences in the observed tone of media toward the issue between The New York Times and The Guardian from 1996 to 2002? Based on Birch et alâ€™s concept of â€œThree Stages of a Crisis,â€ and Downsâ€™s â€œIssue-Attention Cycleâ€ (1972), hypotheses are formed on the assumption that there were similar patterns in the pre-crisis phases for both Britain (1986-1996) and the United States (1996-2002). Therefore, an additional 200 articles from The Guardian were sampled for a pre-crisis situation analysis (From November 1986 to January 1996) as well. The hypotheses are: H1: During the pre-crisis phases of each country, the observed issue frames of The New York Times will show no difference when compared to those of The Guardian. H2: During the pre-crisis phases of each country, the observed information sources used by The New York Times will show no difference when compared to those used by The Guardian. H3: During the pre-crisis phases of each country, the observed tone taken toward the issue in The New York Times will show no difference when compared to that taken in The Guardian.
CHAPTER 3 METHODOLOGY The Sample and Unit of Analysis The articles included in the study were published from 1986 to 2002 in The Guardian and from 1996 to 2002 in The New York Times. These newspapers are major gatekeepers in the United Kingdom and the United States, respectively, and their articles are often reprinted by regional media. The Guardian is one of the three most authoritative leading newspapers in the United Kingdom, along with The Times and The Independent. The Guardianâ€™s total circulation is about 403,999, which is 200,000 less than that of The Times, but The Guardian is famous for its high ethical standards of conscience, criticism, and liberalism, which make it the most popular newspaper for British intellectuals. Therefore, The Guardian is a good vehicle for recording coverage and criticism of the Mad Cow Disease crisis. The New York Times is considered one of the largest liberal newspapers and major sources of international news for U.S. citizens (Chomsky, 1989). Total weekday circulation is 1,159,954, which is less than only USA Today and The Wall Street Journal. It is also believed to be the most influential newspaper in the United States. The most important implication of the comparison pertains to the notion of the informed citizen. Comparing the coverage of two respected daily newspapers cultivates the notion of multiple perspectives and approaches. To arrive at the sample, a preliminary Lexis-Nexis search was conducted for all articles mentioning the phrases â€œBSEâ€ or â€œMad Cow Diseaseâ€ as key words in the 25
26 newspapers. In the case of The Guardian, from March 21, 1996 to December 31, 2002, a total of 1,233 articles were found. For framing analysis, the researcher did a systematic sampling of all the articles. By selecting every tenth article from chronologically listed articles, a total of 123 articles were sampled and analyzed. Unlike the selection interval for The Guardian before 1996, every fifth article from other groups of article was picked because they had much smaller sizes than The Guardian before 1996. Therefore, the researcher tried to keep a manageable sample size and proportional selection. In the case of The New York Times, the same technique was used. A total of 398 articles were found and every fifth article was selected. Therefore, the total sample number was 79. Lastly, 288 articles related to the issue were found in The Guardian from January 1, 1986 to March 20, 1996, and 57 articles were selected by choosing every fifth article. The unit of analysis was an article. An argument was defined as an assertion about BSE issues made either by the author of the article or the sources cited. Each argument was judged according to whether it portrays the topic negatively, positively or in a neutral way. The articles were also examined for the sources they used, in order to determine what types of organizations most affected the targeted media coverage. In addition, a quantitative analysis was conducted to determine what frames were commonly used to shape the audience's understanding of the BSE issue. Operational Definitions of Variables For the identification of important frames, the researcher had executed a preliminary analysis for two periods. The first was the period from March 21, 1996 to May 21, 1996, when the British government acknowledged human infection of BSE, and the second was the period from October 23, 2000 to December 23, 2000, when there was an official report of BSE by the British government. Each of these two periods showed
27 an exceptionally high amount of coverage. Eighteen articles from 1996 and 16 articles from 2000 were selected. Through this process, the researcher identified three distinct frames. Issue Data relating to the dominant issue was collected and categorized into four issue frames. The first was the â€œPublic Health Concernâ€ frame, which was related to public health problems and food safety concerns. This was a kind of human-interest frame that included an emotional angle to the presentation of victims and the disease. The second, the â€œScientific Informationâ€ frame, provided factual information based on scientific knowledge. Some examples of this would be the causes and effects of the disease, scientific research on BSE, and vaccine development. The third, the â€œGovernmental Policyâ€ frame, was concerned with government actions, political statements, and responsibility for the disease under government policy. It also included conflicts between government policy and various publics, including nongovernmental organizations (NGOs) and interest groups. Public cynicism and mistrust toward government policy were also included. The final issue frame was the â€œEconomic Consequencesâ€ frame, which reported how the disease might affect the economy of the home country as well as the outside world. Any article could deal with more than one issue. Therefore, the researcher chose the one issue from each article that seemed to dominate what the article was about. Dominant issue could be selected by its space size in articles, relevance to headline, and subjective judgments based on agreement with two coders.
28 Source For this content analysis, the researcher looked at two types of source categories. The categories consist of official sources and non-official sources. As Paletz and Entman (1981) suggest, the media source used in a story could affect the frame or theme that a journalist gives to that story. It was worthwhile to examine the tone of sources used in order to explore the differences between the two countries. The second category of media sources occurs within official sources, which can be divided into government and non-government sources. Government sources include various governmental departments such as health, science, agriculture, food, economy, and the President or Prime Minister. Scientific institutes supported by governments are also considered as official government sources. Non-government sources were another type of official source, which included corporations, interest groups (including NGOs), and scientific professionals independent from the government. Non-official sources included victims, families of victims, farmers and typical â€œman on the streetâ€ sources. Examples would be people who heard the governmentâ€™s announcement of Mad Cow Disease, people who witnessed the disease in a family member, and farmers whose cattle were destroyed either by disease or governmental order. After the initial official vs. non-official coding was done for the newspapers, the second round of coding was done with more specific types of sources in order to get more detailed results. The nine different categories comprise domestic government officials, foreign government officials, interest groups (including NGOs), corporations, media,
29 scientists, victims (including victim families), farmers (including ranchers) and general citizens. To determine the dominant source framing, only the first source quoted or paraphrased was coded for the purpose of this study. Tone of Media In order to identify media messages of the Mad Cow issue, the researcher categorized relevant articles according to whether the dominant message of the article was positive, negative, or neutral with each issue being assigned a score of +1 (positive), -1 (negative) or 0 (neutral or no coverage). The following words were used as guides during the coding process, but each decision was made by subjectively examining the entire article and not simply the presence or absence of key words. Positive messages included the words such as â€œpraise, beneficial, openness, optimistic, hope, clear, patience, eagerness,â€ and underlined governmental efforts to protect the public from the disease, provide compensation, and undertake other economic and environmental problem-solving efforts. Negative messages included the words such as â€œaccuse, disgraceful, late, angry, ignorant, alarm, incredulity, danger, shock, disaster, scare, loss, fail, puzzling.â€ A neutral message was one that did not criticize or support any side, but provided relevant sources to help the public understand the issue. Inter-coder Reliability After coding, an inter-coder reliability test was conducted to ensure the reliability of results. Inter-coder reliability indicates the level of agreement among independent coders who coded the same content using the same coding instrument (Wimmer &
30 Dominick, 1997). The following formula proposed by Holsti (1969) was applied to determine reliability for this study. Inter-coder Reliability = 2M/(N1 + N2) Where M represents the agreement of two coders' coding decisions and N1 and N2 are the total numbers of coding decisions made by the two coders. Data Analysis Since the goal of the researcher was to find and compare any differences in the frames between The Guardian and The New York Times, the researcher desired to use the most straightforward statistical method. Furthermore, most of the data was limited to nominal data. For these reasons, the research questions and hypotheses in this research were analyzed using the Chi-Square test. The researcher used the SPSS statistical program to perform these analyses.
CHAPTER 4 FINDINGS My study used agenda setting and framing theory to look at the differences in The New York Times and The Guardian in reporting Mad Cow Disease. Content analysis was performed to test differences in sources, issues, and attributes in these newspapersâ€™ articles on the crisis. Content Analysis An inter-coder reliability test was performed on a sample of 20 percent of the articles (80 out of 339 articles). Inter-coder reliability reached 90%. However, many of the articles that contained the key words â€œBSEâ€ and â€œMad Cow Diseaseâ€ were irrelevant to the issue. They were primarily anecdotal examples. For this reason, the author and the coder agreed to remove those cases. In the end, two hundred and seventy five articles were coded. By reducing the sampling size, however, the generalizability of the research might have been reduced. After the coding of articles was completed, the data were entered into SPSS. From there, differences of proportions tests were run on the data. Each research question and hypothesis was tested and conclusions were made regarding the content of the three types of newspaper articles: Articles appearing in The Guardian before the event in 1996, articles appearing in The Guardian after the event in 1996, and articles appearing in The New York Times after the event in 1996. 31
32 Research Questions 1, 2, and 3 The first set of three research questions sought to discover whether there were any changes in the contents of The Guardian regarding Mad Cow before and after the event in 1996. Research question 1 asked whether there were significant changes in The Guardianâ€™s dominant issue frame from before March 21, 1996, to after March 21, 1996. The amount of news coverage after the event more than tripled compared with the amount of news coverage before the event. However, there was not a statistically significant amount of change found for issue frame; data analysis yielded a Chi-square value of 6.457 (df=4, p= .167; Table 4-1). For both periods, â€œGovernment Policyâ€ and â€œPublic Health Concernâ€ were the most frequently discussed issues in The Guardian. These results indicate that the newspaper dealt with these issues even before the governmentâ€™s official acknowledgement of the disease. Even though no significant difference in issue frame was found between the periods before and after the crisis event, the percentage of stories containing scientific information decreased, interestingly, from 25.5 to 12.7%. One possible interpretation of this is that the facts of Mad Cow Disease had been officially denied by the British government before 1996. For this reason, the journalists in The Guardian, in order to heighten the credibility and objectivity of their articles in this uncertain situation, may have been willing to use more scientific professionalsâ€™ opinions. The next research question examined changes in information sources used by The Guardian before and after March 21, 1996. No significant difference was found ( 2 = 7.848; p= .550; Table 4-2). In both periods, â€œDomestic Governmentâ€ was the source
33 most frequently used in the newspaper. However, the percentage of stories using a governmental primary source showed a modest decrease after the crisis (42.6% before the crisis, and 32.1% after the crisis). Also, interest groups that did not exist prior to the crisis event in 1996 were used as sources by The Guardian after the government official announcement in 1996. The event in 1996 was a critical moment for the British government. The government had a chance to acknowledge its untruthful communication activities; it did not, and this decision probably cost the government some of its credibility. This may also have resulted in the governmentâ€™s influence regarding the issue decreasing. On the other hand, interest groups, including corporations and environmental NGOs, might have regarded the announcement as an opportunity to increase their own influence. For this reason, interest groups may have been more active after 1996. The last research question of the first set examined the tone of media coverage in The Guardian before and after the crisis. Results showed that a significantly more negative tone dominated after the event in 1996, yielding a Chi-square value of 12.075 (df=2, p=.002; Table 4-3). Before the crisis, the portions of negative-toned articles and neutral-toned articles were both 47.8%. After the crisis, the percentage of negative articles rose up to 72.7%, while the percentage of neutral articles declined (21.8%). One of the major negative changes in the tone of media occurred in the articles written about the domestic government; the change in tone for these articles showed a statistically significant difference, yielding a Chi-square value of 12.973 (df=2, p=.002; Table 4-4).
34 Regarding issue frames, Table 4-3 shows that the tone of articles that dealt with government policy and economic concerns changed significantly and negatively. In the area of government policy, the percentage of overall negative articles increased from 40% to 72.4%, yielding a Chi-square value of 8.267 (df=2, p=.016; Table 4-4). The percentage of overall negative articles regarding economic concerns changed from 16.7% to 84% ( 2 = 12.392, df=2, p=.002; Table 4-5). This shows that government may have been the primary factor behind the change in the overall media tone. This negative coverage shows the failure of governmentâ€™s efforts at communication and trust building. Research Questions 4, 5, and 6 The second set of three research questions compared changes in the contents of The New York Times and The Guardian regarding Mad Cow Disease after the event in 1996. Research Question 4 examined the differences in The Guardianâ€™s and The New York Timesâ€™ respective dominant issue frames after March 21, 1996. The amount of news coverage in The Guardian (165 articles) was almost three times larger than that in The New York Times (63 articles). However, both papers showed no statistically significant difference in issue frame change, yielding a Chi-square value of 6.602 (df=4, p= .158; Table 4-6). Even though no statistically significant difference was found between the papers, â€œPublic Health Concernâ€ was the most common issue in stories in The New York Times (33.3%), while â€œScientific Informationâ€ (35.2%) and â€œGovernment Policyâ€ (35.2%) were the most common issues in stories in The Guardian. Research Question 5 examined the differences in news sources used in The Guardian and The New York Times after March 21, 1996. No significant difference was
35 found between the primary news sources used by the two newspapers ( 2 = 13.839; df=9; p= .128; Table 4-7). However, in The New York Times, â€œForeign Governmentâ€ (including British government) was the number one source (35.3%), while â€œDomestic Governmentâ€ (32.1%) was the most frequently used source in The Guardian. A peculiar finding was that The New York Times seemed more willing actively to adopt corporate sources (12.7% in The New York Times, and 5.5% in The Guardian), while The Guardian used â€œVictimâ€ (6.7%) as a primary source more than The New York Times did (1.6%). In the United Kingdom, The Guardian, since the issue was involved so directly with their audience, seemed to not have the luxury to use relatively indirect sources, such as corporations. Or, it may be possibly reasoned that public relations efforts made by U. S. corporations more active and effective than companies in the United Kingdom. Research question 6 measured differences between The Guardian and The New York Times in the tone of media coverage after the crisis in 1996. There were significant differences found in the tone of media coverage, yielding a Chi-square value of 14.784 (df=2, p=.001; Table 4-8). The New York Times was relatively balanced between negative and neutral stories (49.2% vs. 47.6%), while stories in The Guardian had a predominantly negative tone (79.5% vs. 21.8%). As far as the tone taken toward particular media sources, The Guardian covered the issue of â€œDomestic Governmentâ€ more negatively than The New York Times did, yielding a Chi-square value of 12.973 (df=2, p=.002; Table 4-9). Regarding issue frames, Table 4-10 shows that the use of negativity in stories covering â€œEconomic Concernâ€ was significantly different between The New York Times (46.7%) and that of The Guardian (84.0%). The coverage of
36 â€œGovernment Policy.â€ was close to significantly different between the two papers ( 2 = 5.914, df=2, p=.052; Table 4-10). The Guardian covered the issue more negatively in the issues of government and economics than its American counterpart. This is an indication that the United States government did a better job handling the crisis than the British government, it should be noted, though, that the U.S. government did not actually have to deal with cases of Mad Cow Disease spreading through the country in the same way that the British government did. Hypotheses 1, 2, and 3 The hypotheses in this content analysis predicted patterns of similarity between The New York Times after March 21, 1996 and The Guardian before March 21, 1996 (i.e., during each countryâ€™s pre-crisis period). Hypothesis 1 predicted a pattern of similarity in selected issues during the pre-crisis period between The Guardian and The New York Times. A statistically significant difference was found between media issues before the crisis in The Guardian and The New York Times ( 2 =12.136; df=4; p=.016; Table 4-11). Therefore, H1 cannot be accepted. In The New York Times, â€œPublic Health Concernâ€ was the number one issue (36.5%), while â€œGovernment Policyâ€ (35.2%) was The Guardianâ€™s most-discussed issue. In addition, The New York Times discussed â€œEconomic Concernâ€ (23.8%) more than The Guardian did (15.2%). There are two plausible reasons for why these differences may have resulted: First, suspected victims of the disease had been found only in British territory before 1996, and
37 second, the two governments used markedly different levels control regarding the issue. Until 2001, there had been no victims found in the United States; therefore, there was less reason for people in the United States to be upset than there was in the United Kingdom. The British governmentâ€™s MAFF (Ministry of Agriculture, Fisheries and Food) was the main institution that controlled and made decisions about agricultural issues. In the United States, though, both the government and interest groups for economic profits (for example, the National Cattlemenâ€™s Beef Association) are involved in making agricultural policy decisions. Hypothesis 2 predicted that there would be a pattern of similarity in media sources used between The Guardian and The New York Times. As the results indicated, no significant differences were found for media source selection between The Guardian and The New York Times, yielding Chi-square value of 11.499(df=9; p=.243; Table 4-12). For this reason, H2 is supported. The final hypothesis predicted that there would be a pattern of similarity in the tone of stories between The New York Times and The Guardian. No significant differences were found in this area before the crisis between The Guardian and The New York Times ( 2 =.645; df=4; p=.724). In The Guardian and The New York Times, negative attribute and neutral attribute were balanced. Additionally, positive attribute were quite infrequent (3.2%) in both papers. Therefore, H3 is supported. In a more detailed analysis, both newspapers showed similar article tones across news sources and issues. The only exception was covering â€œEconomic Concerns.â€ Table 4-15 shows that The New York Times covered the issue more negatively than The Guardian did (46.7% vs. 16.7%). The Chi-square value was 6.022 (df=2, p=.049).
38 Table 4-1. The Guardian before and after 1996 on Media Issue Frame The Guardian Before 1996 The Guardian After 1996 Total Public Count 13 46 69 Health % of Government 22.0% 78.0% 100% % within each paper 27.7% 27.9% 27.8% Scientific Count 12 21 33 Information % of Foreign Gov. 36.4% 63.6% 100% % within each paper 25.5% 12.7% 15.6% Government Count 15 58 73 Policy % of Interest Group 20.5% 79.5% 100% % within each paper 31.9% 35.2% 34.4% Economic Count 6 25 31 Concerns % of Corporation 19.4% 80.6% 100% % within each paper 12.8% 15.2% 14.6% Miscellaneous Count 1 15 16 % of Media 6.3% 93.8% 100% % within each paper 2.1% 9.1% 7.5% Total Count 47 165 212 % of Sources 22.2% 77.8% 100% 2 = 6.457; df=4; p= .167; n=212
39 Table 4-2. The Guardian before and after 1996 on Media Sources The Guardian Before 1996 The Guardian After 1996 Total Domestic Count 20 53 73 Government % of Government 27.4% 72.6 100% % within each paper 42.6% 32.1% Foreign Count 8 23 31 Government % of Foreign Gov. 25.8% 74.2% 100% % within each paper 17% 13.9% Interest Count 0 9 9 Group % of Interest Group 0% 100.0% 100% % within each paper 0% 5.5% Corporation Count 2 9 11 % of Corporation 18.2% 81.8% 100% % within each paper 4.3% 5.5% Media Count 1 4 5 % of Media 20.0% 80.0% 100% % within each paper 2.1% 2.4% Scientist Count 8 21 29 % of Scientist 27.6% 72.4% 100% % within each paper 17.0% 12.7% Victim Count 2 11 13 % of Victim 15.4% 84.6% 100% % within each paper 4.3% 6.7% Farmer Count 1 9 10 % of Farmer 10.0% 90.0% 100% % within each paper 2.1% 5.5% Citizen Count 1 1 2 % of Citizen 50.0% 50.0% 100% % within each paper 2.1% .6% Miscellaneous Count 4 25 29 % of Miscellaneous 13.8% 86.2% 100.0% % within each paper 8.5% 15.2% Total Count 47 165 212 2 = 7.848; df=9; p= .550; n=212
40 Table 4-3. The Guardian before and after 1996 in Tone of Media Coverage The Guardian Before 1996 The Guardian After 1996 Total Negative Count 22 120 142 % of Negative 15.5% 84.5% 100% % within each paper 46.8% 72.7% Neutral Count 22 36 58 % of Neutral 37.9% 62.1% 100% % within each paper 46.8% 21.8% Positive Count 3 9 12 % of Positive 25.0% 75.0% 100% % within each paper 6.4% 5.5% Total Count 47 165 212 2 = 12.075; df=2; p= .002; n=212 Table 4-4. The Guardian before and after 1996 in Tone of Media Coverage (by News Source) Negative Neutral Positive 2 P Domestic The Guardian Before 30%(6) 65%(13) 5%(1) 12.973 .002 Gov. The Guardian After 72%(38) 21%(11) 7%(4) Foreign The Guardian Before 25%(2) 63%(5) 12%(1) 2.395 .302 Gov. The Guardian After 57%(13) 35%(8) 8%(2) Corporation The Guardian Before 50%(1) 50%(1) 1.757 .415 The Guardian After 78%(7) 11%(1) 11%(1) Scientist The Guardian Before 75%(6) 25%(2) .871 .647 The Guardian After 71%(15) 19%(4) 10%(2) * df=2 *Interest group, Media, Victim, and Farmer were not computed, because some variable are missed in each 2-way table.
41 Table 4-5. Guardian before and after 1996 in Tone of Media Coverage (by Issue) Negative Neutral Positive 2 P Public The Guardian Before 62%(8) 38%(5) 1.085 .297 Health The Guardian After 76%(35) 24%(11) Scientific The Guardian Before 50%(6) 42%(5) 8%(1) 3.136 .208 Info. The Guardian After 71%(15) 14%(3) 14%(3) Gov. The Guardian Before 40%(6) 60%(9) 8.267 .016 Policy The Guardian After 72%(42) 22%(13) 6%(3) Economic The Guardian Before 17%(1) 50%(3) 33%(2) 12.932 .002 Concerns The Guardian After 84%(21) 4%(1) 12%(3) MiscellanThe Guardian Before 100%(1) 1.067 1.000 Eous The Guardian After 47%(7) 53%(8) * df=2 Table 4-6. The New York Times and The Guardian after 1996 on Media Issues The New York Times The Guardian After 1996 Total Public Count 23 46 69 Health % of Government 33.3% 66.7% 100% % within each paper 36.5% 27.9% 30.3% Scientific Count 4 21 25 Information % of Foreign Gov. 25.0% 79.5% 100% % within each paper 23.8% 35.2% 32.0% Government Count 15 25 40 Policy % of Interest Group 20.5% 79.5% 100% % within each paper 23.8% 35.2% 32.0% Economic Count 15 6 21 Concerns % of Corporation 37.5% 62.5% 100% % within each paper 9.5% 9.1% 9.2% Miscellaneous Count 6 15 21 % of Media 28.6% 71.4% 100% % within each paper 9.5% 9.1% 9.2% Total Count 63 165 228 % of Sources 27.6% 72.4% 100% 2 = 6.602; df=4; p= .158; n=228
42 Table 4-7. The New York Times and The Guardian after 1996 on Media Sources The New York Times The Guardian After 1996 Total Domestic Count 16 53 69 Government % of Government 23.2% 76.8 100% % within each paper 25.4% 32.1% Foreign Count 18 23 41 Government % of Foreign Gov. 43.9% 56.1% 100% % within each paper 28.6% 13.9% Interest Count 3 9 12 Group % of Interest Group 25.0% 75.0% 100% % within each paper 4.8% 5.5% Corporation Count 8 9 17 % of Corporation 47.1% 52.9% 100% % within each paper 12.7% 5.5% Media Count 1 4 5 % of Media 20.0% 80.0% 100% % within each paper 1.6% 2.4% Scientist Count 5 21 26 % of Scientist 19.2% 80.8% 100% % within each paper 7.9% 12.7% Victim Count 1 11 12 % of Victim 8.3% 91.7% 100% % within each paper 1.6% 6.7% Farmer Count 2 9 11 % of Farmer 18.2% 81.8% 100% % within each paper 3.2% 5.5% Citizen Count 1 1 2 % of Citizen 50.0% 50.0% 100% % within each paper 1.6% .6% Miscellaneous Count 8 25 33 % of Miscellaneous 24.2% 75.8% 100.0% % within each paper 12.7% 15.7% Total Count 63 165 228 2 = 13.839; df=9; p= .128; n=228
43 Table 4-8. The New York Times and The Guardian after 1996 in Tone of Media Coverage The New York Times The Guardian After 1996 Total Negative Count 31 120 151 % of Negative 20.5% 79.5% 100% % within each paper 49.2% 72.7% Neutral Count 30 36 66 % of Neutral 45.5% 54.5% 100% % within each paper 47.6% 21.8% Positive Count 2 9 11 % of Positive 18.2% 81.8% 100% % within each paper 3.2% 5.5% Total Count 63 165 228 2 = 14.784; df=2; p=.001 Table 4-9. The New York Times and The Guardian after 1996 in Tone of Media Coverage (by News Source) Negative Neutral Positive 2 P Domestic The NY times 29%(5) 65%(11) 6%(1) 13.313 .001 Gov. The Guardian After 72%(38) 21%(11) 7%(4) Foreign The NY times 39%(7) 61%(11) 3.719 .156 Gov. The Guardian After 57%(13) 35%(8) 8%(2) Interest The NY times 67%(2) 33%(1) .800 .371 Group The Guardian After 89%(8) 11%(1) Corporation The NY times 50%(4) 38%(3) 12%(1) 1.765 .414 The Guardian After 78%(7) 11%(1) 11%(1) Scientist The NY times 60%(3) 20%(1) 20%(1) .462 .794 The Guardian After 71%(15) 19%(4) 10%(2) Victim The NY times 100%(1) .364 .546 The Guardian After 73%(8) 27%(3) * df=2 *Media, and Farmer were not computed, because some variable are missed in each 2-way table.
44 Table 4-10. The Guardian before and after 1996 in Tone of Media Coverage (by Issue) Negative Neutral Positive 2 P Public The NY times 52%(12) 43%(10) 5%(1) 5.216 .074 Health The Guardian After 76%(35) 24%(11) Scientific The NY times 25%(1) 50%(2) 25%(1) 3.516 .172 Info. The Guardian After 71%(15) 14%(3) 15%(3) Gov. The NY times 47%(7) 53%(8) 5.914 .052 Policy The Guardian After 73%(42) 22%(13) 5%(3) Economic The NY times 47%(7) 53%(8) 13.807 .001 Concerns The Guardian After 84%(21) 4%(1) 12%(3) MiscellanThe NY times 67%(4) 33%(2) .687 .407 Eous The Guardian After 47%(7) 53%(8) * df=2 Table 4-11. The New York Times and The Guardian before 1996 on Media Issues The New York Times The Guardian Before 1996 Total Public Count 23 13 36 Health % of Government 63.9% 36.1% 100% % within each paper 36.5% 27.7% 32.7% Scientific Count 4 12 16 Information % of Foreign Gov. 25.0% 75.0% 100% % within each paper 6.3% 25.5% 14.5% Government Count 15 15 30 Policy % of Interest Group 50.0% 50.0% 100% % within each paper 23.8% 31.9% 27.3% Economic Count 15 6 21 Concerns % of Corporation 71.4% 28.6% 100% % within each paper 23.8% 12.8% 19.1% Miscellaneous Count 6 1 7 % of Media 85.7% 14.3% 100% % within each paper 9.5% 2.1% 6.4% Total Count 63 47 110 % of Sources 57.3% 42.7% 100% 2 = 12.136; df=4; p= .016; n=110
45 Table 4-12. The New York Times and The Guardian before 1996 on Media Sources The New York Times The Guardian Before 1996 Total Domestic Count 18 20 35 Government % of Government 27.4% 72.6 100% % within each paper 25.4% 42.6% Foreign Count 18 8 26 Government % of Foreign Gov. 69.2% 30.8% 100% % within each paper 28.6% 17.0% Interest Count 3 0 3 Group % of Interest Group 100.0% 0% 100% % within each paper 4.8% 0% Corporation Count 8 2 10 % of Corporation 80.0% 20.0% 100% % within each paper 12.7% 4.3% Media Count 1 1 2 % of Media 50.0% 50.0% 100% % within each paper 1.6% 2.1% Scientist Count 5 8 13 % of Scientist 38.5% 61.5% 100% % within each paper 7.9% 17.0% Victim Count 1 2 3 % of Victim 33.3% 66.7% 100% % within each paper 1.6% 4.3% Farmer Count 2 1 3 % of Farmer 66.7% 33.3% 100% % within each paper 3.2% 2.1% Citizen Count 1 1 2 % of Citizen 50.0% 50.0% 100% % within each paper 7.3% 8.5% Miscellaneous Count 8 4 12 % of Miscellaneous 66.7% 33.3% 100.0% % within each paper 12.7% 8.5% Total Count 63 47 110 2 = 11.499; df=9; p= .243; n=110
46 Table 4-13. The New York Times and The Guardian before 1996 in Tone of Media Coverage The New York Times The Guardian Before 1996 Total Negative Count 31 22 53 % of Negative 58.5% 41.5% 100% % within each paper 49.2% 46.8% Neutral Count 30 22 52 % of Neutral 57.5% 42.3% 100% % within each paper 47.6% 46.8% Positive Count 2 3 5 % of Positive 40% 60% 100% % within each paper 3.2% 6.4% Total Count 63 47 110 2 =.645; df=2; p=.724; n=110 Table 4-14. The New York Times and The Guardian before 1996 in Tone of Media Coverage (by News Source) Negative Neutral Positive 2 P Domestic The NY times 27%(5) 68%(11) 6%(1) .823 .663 Gov. The Guardian Before 30%(6) 65%(13) 5%(1) Foreign The NY times 39%(7) 61%(11) 2.256 .278 Gov. The Guardian Before 25%(2) 63%(5) 12%(1) Corporation The NY times 50%(4) 38%(3) 12%(1) 1.875 .392 The Guardian Before 50%(1) 50%(1) Scientist The NY times 60%(3) 20%(1) 20%(1) 1.733 .420 The Guardian Before 75%(6) 25%(2) * df=2 *Interest group, Media, Victim, and Farmer were not computed, because some variable are missed in each 2-way table.
47 Table 4-15. The Guardian before and before 1996 in Tone of Media Coverage (by Issue) Negative Neutral Positive 2 P Public The NY times 52%(12) 43%(10) 2%(1) .746 .688 Health The Guardian Before 62%(8) 38%(5) Scientific The NY times 25%(1) 50%(2) 25%(1) 1.143 .565 Info. The Guardian Before 50%(6) 42%(5) 8%(1) Gov. The NY times 47%(7) 53%(8) .136 .713 Policy The Guardian Before 40%(6) 60%(9) Economic The NY times 47%(7) 53%(8) 6.022 .049 Concerns The Guardian Before 17%(1) 50%(3) 33%(2) MiscellanThe NY times 67%(4) 33%(2) .467 .495 Eous The Guardian Before 100%(1) * df=2
CHAPTER 5 DISCUSSION AND CONCLUSIONS Discussion RQ1: Are there Significant Changes in The Guardianâ€™s Dominant Issue Frame before and after March 21, 1996? The results showed no significant relationship in the issue frames in The Guardian before and after the crisis. This implies the media was already discussing the issue before the crisis. It was in 1982 when Mad Cow Disease became an important social issue in the United Kingdom. Since that time, The Guardian had consistently discussed the disease in view of government policy along with their concern for public health. They already regarded this disease as a critical social, economical, and health issue. Therefore, before the governmentâ€™s acknowledgement of human infection from Mad Cow Disease in 1996, The Guardian seems to have established its own frame regarding the issue. The results also showed that the number of articles after the crisis in 1996 was more than triple the number of articles before the crisis. The British government had been criticized for its information policy of being silent in order to minimize the issue of Mad Cow Disease. While the governmentâ€™s policy of silence seemed to succeed at first, it may have helped to induce explosive media coverage after the official acknowledgement in 1996. This incident showed a typical case of mistreated crisis management. Another feature found in this analysis was a decrease in the amount of scientific information included in Mad Cow stories after 1996. Before 1996, the British 48
49 government had denied the seriousness of the disease and provided limited related information media. Therefore, The Guardian seemed to take advantage of the kind of objectivity with which scientific information is credited in order to heighten its credibility. RQ2: Are there Significant Changes in The Guardianâ€™s Dominant Information Sources before and after March 21, 1996? As with Research Question 1, there were no significant changes found in The Guardianâ€™s source selections when comparing coverage from before and after the crisis. There was a slight change in source diversity (e.g. the decrease of Governmentâ€™s source), but not at a significant level. This shows that the governmentâ€™s announcement did not spur reporters to go out and find different news sources. This can be explained in two ways. First, even though there was some decrease in frequency of use of government sources, the total frequency of articles after the crisis in 1996 more than doubled the frequency from before the crisis (20 articles vs. 53 articles). Therefore, there was not much difference between the two periods. Second, if the change were significant, there might be a possibility that the government had lost control over the issue and that its sources were replaced by other sources, such as interest groups. In actuality, though, this immobility of news source could suggest that the accused organization was a continuous media target whether it lost its credibility or not. It also might show the situation that went against the British government, which had tried to keep in low profile and lost control of its message. RQ3: Are there Significant Changes in The Guardianâ€™s Tone of Media Coverage (Positive, Negative, and Neutral) before and after March 21, 1996? Unlike the first two research questions, there were significant changes in the tone used by The Guardian before and after the crisis. The crisis in 1996 broke the balance of
50 negative and neutral tone of articles. After the government provided journalists with information regarding the disease that it had previously denied them, the government was roundly attacked for having kept the information from them in the first place. This picture became clearer with detailed analysis. In articles in which the media used sources, only news articles using government sources significantly increased their negative tones. It can be taken from this that The Guardian lost its neutral view as well as its faith in the British government. In addition, news regarding government policy and economic concern also became more negatively changed. We can then infer that, prior to the crisis, The Guardian was able to appreciate the impending threat posed by Mad Cow Disease. However, after the crisis in 1996, while The Guardianâ€™s issue framing and news source selection showed little difference, its criticism toward the government became more serious and more negative. RQ4: Are there Differences in the Observed Issue Frames between The New York Times and The Guardian from 1996 to 2002? The researcher intended to explore whether there was any difference in issue framing, under the assumption that the media approach would be different regarding local issues and international issues. However, no statistically significant difference was found between The New York Times and The Guardian. Certainly, though, The Guardian paid attention more to Mad Cow Disease as a local issue (165 articles) than The New York Times did as an international issue (63 articles). One of the major reasons for this similarity might come from the idea of a standardized and elaborated journalistic format. Both The New York Times and The Guardian are leading national papers in their own countries and both are famous for their
51 high credibility and public trust. Therefore, it is natural to assume that their stories would include reasonably similar news features. RQ5: Are there Differences in the Observed Information Sources between The New York Times and The Guardian from 1996 to 2002? As discussed in Research Question 4, it is safe not to expect significant differences in journalistic format between the two papers. The result implies that even though this crisis could be interpreted as either international or domestic, both papers used similar means to determine their sources. The New York Times broke somewhat with The Guardian by making use of corporate sources in their articles (The New York Times: 12.7%, The Guardian: 5.5%). The Guardian, meanwhile, differed from The New York Times by finding sources among victims of the disease (The New York Times: 1.6%, The Guardian: 6.7%). However, the impact of these articles was too minimal to have made any major difference. RQ6: Are there Differences in the Observed Tone of Media toward the Issue between The New York Times and The Guardian from 1996 to 2002? Interestingly, even though the selected issues and sources showed similarities, the tone of media coverage used by each paper differed significantly. The major reasons for this difference may be similar to the explanation of the results for Research Question 3. In the United States, unlike the United Kingdom, there were no Mad Cow disease cases found, and the U.S. government had not been criticized for a policy of silence like that of the British government. Itemized analysis showed more clear reasons. Among the news sources, â€œDomestic Governmentâ€ was the only significantly and negatively covered by The Guardian. Furthermore, â€œGovernment Policyâ€ was the most highly criticized issue in The Guardian.
52 We can understand from the results that The Guardian and The New York Times used similar issue frames and issue selections but the different tone of media toward governmental news source and policy might come from mishandled public relations by the British government. H1: During the Pre-Crisis Phases of Each Country, the Observed Issue Frames of The New York Times will Show No Difference When Compared to those of The Guardian. As the results show, a statistically significant difference was found in issue frames between the two pre-crisis sessions (The Guardian before 1996 and The New York Time from 1996 to 2002). Unlike The New York Times, The Guardian seems to have been preoccupied with the issue of â€œGovernment Policy.â€ This implies that, as the previous results showed, The Guardian was aware of the seriousness of the disease; The Guardian also seems to have had more interest in the disease and a more suspicious view of the government than The New York Times. This may have been because the British governmentâ€™s MAFF (Ministry of Agriculture, Fisheries and Food) had been more directly involved with the Mad Cow Disease issue than any United States governmental agency did. In the United States, the NCBA (National Cattlemenâ€™s Beef Association), a non-governmental interest group, has taken a more active role against the issue than the U.S. government has, while the Food and Drug Administration (FDA), a U.S. counterpart of MAFF, has taken surprisingly little action regarding the issue. For example, from 1985 to 2001, only 6 press releases about the issue were found on the FDAâ€™s website archive. In the United States, as the coverage of the issue makes clear, the disease had not been a part of the news agenda before 1996, and the amount of coverage, therefore, was
53 relatively low. These reasons might have kept The New York Times from focusing on the issue more seriously and critically. H2: During the Pre-Crisis Phases of Each Country, the Observed Information Sources used by The New York Times will Show No Difference when Compared to those used by The Guardian. The results show no significant difference in the source selection between the two pre-crisis sessions. Unlike the differentiated issue framing, source selection seemed to follow a standardized newsgathering process, because none of the three sets of comparison in this research showed significant differences in source selections. If this is true, it might be interesting in the future to compare news sources of major newspapers like The New York Times with those of more tabloid newspapers regarding a crisis. H3: During the Pre-Crisis Phases of Each Country, the Observed Tone taken toward the Issue in The New York Times will Show No Difference when Compared to that taken in The Guardian. The most interesting result is that there was no significant difference in article tone regarding Mad Cow disease between The Guardian before 1996 and The New York Times after 1996. Even though there were some differences in issue selection, the overall tone of the articles seemed similar. This could mean that each newspaper was among the most trusted and prestigious newspapers in its country; each one, then, might have its own norm of criticism toward the issue. Furthermore, each period of gathering articles was more than five years, long enough to dilute any immediate effect of the 1996 crisis. In summary, two of the three hypotheses that were based on the 3-stage issue cycle model were supported. Even though there were situational differences between the two countries (for example, actual victims from the disease and government control level toward the issue), those two pre-crisis situations showed similarity. The cycle model was
54 originally suggested by practitioners based on their experience in industrial field. Through the analysis, this empirical model may have chance to show its theoretical validity. Conclusions The researcher examined how the coverage in The Guardian regarding Mad Cow Disease changed before and after the British governmentâ€™s acknowledgement of human infection from Mad Cow disease in 1996. The researcher assumed the British governmentâ€™s confession was a critical turning point for the event, because that announcement directly affected the collapse of the beef industry in the United Kingdom and commerce with E.U. countries. Also, since that event, news media in the United States have begun to be interested in the issue. Furthermore, the author explored similarities between the coverage of The Guardian and The New York Times regarding the issue in the targeted phase periods. The findings showed that the announcement in 1996 did not change the media issue frame and news source selection in The Guardian, because the crisis in 1996 was not a truly unexpected crisis for the British media. During the pre-crisis session (before 1996), the media already knew about the effects of the disease and waited only for the British governmentâ€™s official acknowledgement. However, after the acknowledgement in 1996, the tone of The Guardianâ€™s articles regarding Mad Cow disease turned decidedly negative and, in many cases, expressed outright hostility. Through detailed analysis, the researcher found that perhaps the major reason for this negativism was the British governmentâ€™s mishandled policy regarding the Mad Cow disease problem. Therefore, it could be inferred that the governmentâ€™s concealment of information and denial of human infection caused an explosive increase
55 of coverage with a negative tone in The Guardian after the governmentâ€™s announcement in 1996. Although on the surface, the announcement could be viewed as transparent communication on the part of the British government, it really seemed to represent an involuntary acknowledgement prompted by public pressure. On the other hand, The New York Times showed similar issue frame and source selection as The Guardian. However, The New York Times presented a more neutral tone than its British counterpart, because there had been no such public relations blunder in the United States. There had been no proof at that point that Mad Cow Disease really existed in the United States. Last, two of the three hypotheses were supported. The pre-crisis stage in both countries showed similarities both in news source selection and tone of articles. These can be inferred two ways. First, the major news outlets might have their own shared and standardized news gathering processes. If the comparison had been executed between different kinds of media, the results might be quite different. Second, it proves the theoretical validity of the 3-stage model constructed from the experiences of public relations professionals. In summary, this research tried to compare and contrast issue framing, news source selection and the tone of media under different time periods and different media nationalities. Overall, the main reason for any change in media coverage was the British governmentâ€™s mishandled communications to the public. As Gonzalez-Herrero and Pratt (1996) suggested, this could have been an example of an organizationâ€™s rapid reaction to accusations being leveled by the media. The British government had been criticized severely by the British media and subsequently reacted to these accusations by disclosing
56 the existence of humans who had been infected. Through its official acknowledgement of the disease, the government lost its control not only of the message, but also of the framing of the message and the framing of the crisis situation, thus losing the ability to determine the tone of the persuasive exchange. This research provided the starting point of public relations case research. The United States and the United Kingdom both showed typical pre-crisis stage situations of their time, and an analysis of news articles showed that the British government committed a public relations blunder. Now is the time to explore what may have caused the results found in this research: press releases and other media clips. Limitations There are limitations in this thesis. First, since this study drew bivariate comparison between the periods and newspapers, the simultaneous effects of external variables were not determined and controlled. Another main limitation is the lack of analysis of information subsidies. For a more thorough investigation, press release content about the issue should have been analyzed and compared with media coverage. The accuracy of the Lexis/Nexis database was another concern. There were some problems in claiming that what was included in these sources was truly representative of the newspapers being analyzed. As noted previously, the screening process used to reduce the sample size might have reduced the generalizability of the research. Finally, the coders analyzed each article by hand. Invariably, human-coded content analysis relies on subjective judgments. Any time subjective judgments are used in content analysis, the overall accuracy of the study may be called into some question.
57 Future Research This research examined The Guardian, which underwent the entire three-stage crisis process, and The New York Times, which had only gone through the pre-crisis stage. However, in January 2004 the Food and Drug Administration announced that the United States was no longer an exception to the now-global Mad Cow Disease. During January and February, The New York Times ran 67 articles regarding the disease. Now would be an opportune time to construct a truly parallel comparison between the United States and the United Kingdom throughout the whole crisis process. Based on these results, further research is available for examining public relations strategies and their effects. There have been previous attempts to utilize agenda setting theory and framing in public relations research (e.g. Hallahan, 1999). Further research in this area can utilize press release materials from the two relevant government departments: the FDA in the United States and the Ministry of Agriculture and Fisheries in the United Kingdom. With this research, we can not only analyze the messages of each government department, but also measure the level of effects of their efforts.
APPENDIX CODING PROTOCOL Fill out one sheet per article. Analyze all of the stories given to you. Article Number: For each coding sheet, a sequential integer should be entered in the blank to the right of the page. Start with the number â€ for the first article coded, continue with â€, and so forth. Date: Enter the date of the article being coded on the line to the right of the page in a month/day/year format. Coder: Enter the appropriate number for who you are. Enter â€ for Seung Hoon Han and â€ for the second coder. The main coding table: If an article of the New York Times is based on an official source from the British Government, it shows the characteristic of (X_112). Then the coder should mark â€œYesâ€ Column on row X_112. In case of the Guardian article, the column â€œYesâ€ is divided with â€œBâ€ and â€œA.â€ â€œBâ€ represents â€œBeforeâ€ which means this article was published before the mad cow crisis on March 21, 1996. â€œAâ€ represents â€œAfter the Crisis.â€ Therefore, if a Guardian article contains â€œPublic Health Concern Issueâ€ and published in 1990, the coder should mark â€œYesâ€ column on row â€œX_231.â€ Information Source: The source is the document or person providing information for the article. Article often have more than one source. You will just be coding the first source you come to. You will be coding for two different categories of sources: the official source and the specific sources relating to the crisis. o Official sources consist with government and non-government sources. Government sources include various governmental departments and the President or Prime Minister. Scientific institutes supported by governments are also included. Non-government sources are another type of official source, which include corporations, interest groups (including NGOs), and scientific professionals independent from the government. o Non-official sources include victims, a victimâ€™s family, farmer and typical â€œman on the streetâ€ sources. o The coding process for official / non-official sources is followed by the nine different categories coding include: domestic government officials, foreign government officials, interest groups (including NGOs), 58
59 corporations, media, scientists, victims (including victim families), farmers (including ranchers) and general citizens. Dominant Issues: Select the appropriate issue frame in articles. One article may contain more than one issue. Therefore, to code the dominant issue framing, the researcher selects only one issue, the one most extensively represented in the article. o Public Health Concern Frame: This is related to public health problems and food safety concerns. This is a kind of human-interest frame that includes an emotional angle to the presentation of victims and the disease. o Scientific Information Frame: This provides factual information based on scientific knowledge, some examples of which would be, the causes and effects of the disease, scientific research on BSE, and vaccine development. o Governmental Policy Frame: This frame is concerned with the issue of government actions, political statements, and responsibility for the disease under government policy. It also includes conflicts between government policy and various publics including NGOs and interest groups. Public cynicism and mistrust toward government policy are also included. o Economic Consequences Frame: This frame describes how the disease might affect the economy of the home country and world. Tone of Media: The overall balance of media coverage for each issue. o The positive messages include the words â€œpraise, beneficial, openness, optimistic, hope, clear, patience, eagerness,â€ and underline governmental efforts to protect the public from the disease, provide compensation, and undertake other economic and environmental problem-solving efforts. o The negative messages include the words â€œaccuse, disgraceful, late, angry, ignorant, alarm, incredulity, danger, shock, disaster, scare, loss, fail, puzzling.â€ o A neutral message is one that does not criticize or support any side and provides relevant sources to help the public understand the issue.
60 Coding Sheet A. Article Number _____________ID B. Date ___/___/___ DATE C. Coder 1)Seung Hoon Han 2)#2_________ CODER Newspaper Source Specific Source Yes TSS Domestic Gov (X_111) Foreign Gov (X_112) Interest Group (X_113) Corporation (X_114) Media (X_115) Official (X_11) Scientist (X_116) Victims (X_121) Farmers (X_122) Non-Official (X_12) Citizen (X_123) The NY Times (X_1) Miscellaneous (X_101) B A B A Domestic Gov (X_211) Foreign Gov (X_212) Interest Group (X_213) Corporation (X_214) Media (X_215) Official Source (X_21) Scientist (X_216) Victims (X_221) Farmers (X_222) Non-Official (X_22) Citizen (X_223) The Guardian (X_2) Miscellaneous (X_201) Newspaper Dominant Issues Yes TSS Public Health Concern (X_131) Scientific Information (X_132) Governmental Policy (X_133) Economic Consequence (X_134) The NY Times (X_1) Miscellaneous (X_102) B A B A Public Health Concern (X_231) Scientific Information (X_232) Governmental Policy (X_233) Economic Consequence (X_134) The Guardian (X_2) Miscellaneous (X_202) Newspaper Tone of Media Yes TSS Negative (X_141) Neutral (X_142) The NY Times (X_1) Positive (X_143) B A B A Negative (X_241) Neutral (X_242) The Guardian (X_2) Positive (X_243)
LIST OF REFERENCES Bagdikian, B.H. (1974). Congress and the Media: Partners in Propaganda. Columbia Journalism Review, 12. Baumgartner, F. R., & Jones, B. D. (1993). Agendas and Instability in American Politics. Chicago: The University of Chicago Press. Birch, J. (1994). New factors in crisis planning and response. Public Relations Quarterly, 39(4), 31-34. Bitter price of short cuts; we are all guilty about beef: farmers most of all. (1996, March 21). The Guardian, p.18. Brodie M., Brady L.A., & Altman D.E. (1998). Media Coverage of Managed Care: Is There a Negative Bias? Health Affairs 17(1), 9-25. Brookes, R. (1999). Newspapers and national identity: the BSE/CJD and the British press. Media Culture & Society, 21(2), 247-264. BSE: The long search for the facts. (2000, August 29). BBC News. Retrieved May 18, 2003, from http://news.bbc.co.uk Cannon G., Connor S., Hague H., & Nicholsonlord D. (1990, May 20). Mad Cows and Englishmen. Independent, p.19. Chomsky, N. (1989). "The United States and Indochina: Far from an aberration," in Douglas Allen and Ngo Vinh Long (Eds), Coming to Terms: Indochina, the United States and the War. Boulder: Westview Press. Coombs, W. T., & Holladay, S.J. (2001). An extended examination of the crisis situations: A fusion of the relational management and symbolic approaches. Journal of Public Relations Research, 13 (4), 321. Davis, A. (2000). Public relations, news production and changing patterns of source access in the British national media. Media Culture & Society 22(1), 39-60. Downs, A. (1972). Up and down with ecology: The â€œIssue-Attention Cycle.â€ The Public Interest, 28, 38-50. 61
62 Dunwoody, S. (1992). The media and public perceptions of risk: How journalists frame risk stories. In D. Bromley and K. Segerson (Eds.), The Social Response to Environmental Risk. Boston: Kluwer Academics. Entman, R.M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51-58. Entman, R.M. (1991). Framing U.S. coverage of international news: Contrasts in narratives of the KAL and Iran incidents. Journal of Communication, 41(4), 6-27. Fink, S. (1986). Crisis management: Planning for the inevitable. New York: AMACOM Frewer, L.J., Miles, S., & Marsh, R. (2002). The media and genetically modified foods: Evidence in support of social amplification of risk. Risk Analysis 22(4), 701-711. Gamson, W.A., & Lasch, K.E. (1983). The political culture of social welfare policy. In S.E. Spiro, & E. Yuchtman-Yaar (Eds.), Evaluating the welfare state: Social and political perspective (p.397-415). Academic Press, New York. Gamson, W.A., & Modigiliani, A. (1989). Media discourse and public opinion on nuclear power: A constructionist approach. American Journal of Sociology 95(1), 1-37. Gans, H.J. (1979, 1980). Deciding What's News. New York: Random House, Vintage Books. Gitlin, T. (1980). The Whole World Is Watching: Mass Media in the Making and Unmaking of the New Left (p. 7). Berkeley, CA: University of California Press. Gonzalez-Herrero, A., & Pratt, C. B. (1996). An integrated symmetrical model of crisis-communications management. Journal of Public Relations Research, 8(2), 79-106. Grunig, J.E. (1992). Communication, public relations, and effective organizations: An overview of the book. In J.E. Grunig (Ed.), Excellence in public relations and communication management (pp.1-30). Hillsdale, NJ: Lawrence Erlbaum. Hallahan, K. (1999). â€œSeven models of framing: Implications for public relations.â€ Journal of Public Relations Research 11(3), 205-242. Heath, R.L. (1990). Corporate issues management: Theoretical underpinnings and research foundations. In J.E. Grunig & L.A. Grunig (Eds.), Public relations research annual (Vol.2, pp.29-66). Hillsdale, NJ: Lawrence Erlbaum. Heo, K. J. (2000). Newspapersâ€™ Coverage of North Korean Leader. Paper presented at the Association for Education in Journalism & Mass Communication Convention, Miami Beach, FL. Hofstede, G. (1991). Cultures and organizations: Software of the mind. London: McGraw-Hill UK.
63 Hofstede, G. (1980). Cultureâ€™s consequences: International differences in work-related values. Newbury Park, CA: Sage. Holsti, O. (1969). Content analysis for the social sciences and humanities, Reading, MA: Addison-Wesley. Hornsby, M. (1996, April 11). EU postpones decision on beef byproducts. The Times. Retrieved May 13, 2003, from http://www.the-time.co.uk Knight, M.G. (1999). Getting past the impasse: Framing as a tool for public relations. Public Relations Review, 25 (3), 381-398. Kohn, C. (2003, June 9). Scientists predict swift end to epidemic. Pain & Central Nervous System Week, p.2. Lanska, D.J. (1998). The mad cow problem in the U.K.: Risk perceptions, risk management, and health policy development. Journal of Public Health Policy, 19(2), 160-183. Lipsky, M., & Olson, D.J. (1976). The processing of Racial Crisis in America, Politics & Society 6 (1), 79-103. Lodge, M.G. & Hanmil, R. (1986). A partisan schema for political information processing, American Political Science Review, 80, 505-520. Mad cow disease reported in Canada. (2003, May 21). CNN News. Retrieved May 21, 2003, from http://www.cnn.com Mazur, A., & Hall, G.S. (1990). Effects of social-influence and measured exposure level on response to radon. Sociological Inquiry 60(3). 274-284 McLeod, D.M., & Hertog, J.K. (1998). Social control and the mass media's role in the regulation of protest groups: The Communicative Acts Perspective. In D. Demers and K. Viswanath (Eds.), Mass Media, Social Control and Social Change. Ames, IA: Iowa State University Press. Miller, M. M., & Riechert, B. P. (2001). The Spiral of Opportunity and Frame Resonance: Mapping the Issue Cycle in News and Public Discourse. In: Framing Public Life: Perspectives on Media and Our Understanding of the Social World. Mitchell, T. H. (1986). Coping with a corporate crisis. Canadian Business Review, 13, 17-20. Mitroff, I.A. (1994). Crisis management and environmentalism: A natural fit. California Management Review, 36(2), 101-113. National briefing science and health: Woman has human form of Mad Cow Disease, (2002, October 18). The New York Times, p. A21.
64 Noakes, J.A., & Wilkins, K.G. (2002). Shifting frames of the Palestinian movement in U.S. news. Media Culture & Society, 24(5), 649-672. Olien, C.N., Tichenor, P.J., & Donohue, G.A. (1989). "Media and Protest." In L.A. Gruning (Ed.), Environmental Activism Revisited: The Changing Nature of Communication Through Organizational Public Relations, Special Interest Groups and the Mass Media (pp. 22-39). Troy, OH: North American Association for Environmental Education. Palenchar, M.J. (2001). Media coverage of risk events: A framing comparison of two fatal manufacturing accidents. Paper presented at the annual convention of the Association for Education in Journalism & Mass Communication, Washington, D.C. Paletz, D.L., & Entman, R.M. (1981). Media Power Politics. New York: Free Press. Pan, Z., & Kosicki, G.M. (1993). Framing analysis: An approach to news discourse. Political Communication, 10, 6175. Park, A. (2003, June 2). Plagued Canada, Mad cow disease strikes Americaâ€™s northern neighbor. Time. Retrieved May 20, 2003, from http://www.time.com Ramton, S., & Stauber, J (1997). Mad Cow U.S.A. Monroe, Maine: Common Courage Press. Reese, S.D. (2001). â€œPrologue-Framing Public Life: A Bridging Model for Media Research) In S.D. Reese, H.H. Gandy, & A.E. Gant (Eds.), Framing Public Life (pp. 7-32). Mahwah, New Jersey: Lawrence Erlbaum Associates. Regaldo, A. (2002, November 5). Deer disease stirs huntersâ€™ concern over food safety. The Wall Street Journal, D3. Regaldo, A. (2002, September 27). World watchâ€”Europe/Africa: Briefly. The Wall Street Journal, A12. Rowe, G., Frewer, L., & Sjoberg, L. (2000). Newspaper reporting of hazards in the UK and Sweden. Public Understanding Science, 9(1), 59-78. Sachsman, D.B. (1976). Public relations influence on coverage of environment in San Francisco. Journalism Quarterly, 53 (1), 54-60. Scheufele , D.A. (1999). Framing as a theory of media effects. Journal of communication, 49(1), 103-122 Schlesinger, P. (1990). Rethinking the sociology of journalism: Source strategies and the limits of media-centrism. In D. Demers and K. Viswanath (Eds.), Mass Media, Social Control and Social Change (p.61-83). Ames, IA: Iowa State University Press.
65 Semetko, H.A., & Valkenburg, P.M. (2000). Framing European politics: a content anlysis of press and television news. Journal of Communication, Spring, 93-109. Shoemaker, P.J., & Reese, S.D. (1996). Mediating the message: theories of influences on mass media content. New York: Longman. Terkildsen, N., Schnell, F.I., & Ling, C. (1998). Interest groups, the media, and policy debate formation: an analysis of message structure, rhetoric, and source Cues. Political Communication 15(1), 45-61. The BSE Inquiry (2000, October). The report of BSE inquiry: The inquiry into the BSE and variant CJD in the United Kingdom. Retrieved April 20, 2003, from http://www.bseinquiry.gov.uk Thomson, T.G. (2001, November 30). Regarding Release of Harvard BSE Risk Assessment. Message posted to http://www.hhs.gov/news/press/2001pres/200111 30b.html United States Department of Agriculture, Foreign Agricultural Service (April 28, 2003), Total Beef Trade Forecast at Record Level for 2003. Retrieved from May 18, 2003, from http://www.fas.usda.gov/ Verbeke,W., Ward, R.W., & Avermaete, T. (2002). Evaluation of publicity measuring to the EU beef labeling system in Belgium. Food Policy, 27 (4), 339-353. Wanta, W. & Hu, Y. (1993). The agenda-setting effects of international news coverage: An examination of differing news frames. International Journal of Public Opinion Research, 5, 250-264. Webb. D. (2002). How now, mad cow? Assessing the disease risk for Americans. Environmental Nutrition, 25(3), p.1. Wiegman, O., & Gutteling, J.M. (1989). Newspaper coverage of the technological and environmental hazards and the reactions of readers. Journalism quarterly 66(4). 844-852. Wimmer, D. R., & Dominick, R. J. (1997). Mass Media Research: An Introduction, Belmont, CA, Wadsworth Publishing Company. Woodcock, C. (1994, November 24). How to face crisis with confidence. Marketing, pp.8-10. World Watchâ€”Asia/Pacific: Japan reports new mad cow case. (2002, August 23). The Wall Street Journal, A6. Zaller, J.R. (1992). The nature and Origins of Mass Opinion. New York: Cambridge University Press.
66 Zhang, Y. (1998). Covering the Hong Kong transition: A content analysis of the news stories by Chinaâ€™s Xinhua news agency and the Associated Press of the United States between May 1 and August 31, 1998. Unpublished Doctoral Dissertation, University of Kentucky, Lexington.
BIOGRAPHICAL SKETCH Seung Hoon Han was born and raised in Seoul, South Korea. He completed his B.A. from the Hanyang University in advertising and public relations in 1995. After spending a 5 year career in a public relations agency in South Korea, he began work on his masterâ€™s degree at the University of Florida. He graduated from the University of Florida in August 2004. 67