University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA COLLEGE OF EDUCATION ONLINE MEDIA FRAMING OF THE WEST AFRICAN EBOLA VIRUS DISEASE (EVD) EPIDEMIC: A CONTENT ANALYSIS OF CNN, GLOBAL NEWS NETWORK (GNN) AND DAILY NATION BY ERNEST KOFI AGORDE (10550405) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE AWARD OF MHPIL COMMUNICATION STUDIES DEGREE DEPARTMENT OF COMMUNICATION STUDIES JULY 2017 University of Ghana http://ugspace.ug.edu.gh DECLARATION I hereby declare and confirm that this thesis is entirely my own work and was produced from research undertaken at the Department of Communication Studies, University of Ghana, Legon, under the supervision and guidance of Dr Gilbert Tietaah. All works by other scholars used in this study have been duly cited in-text and acknowledged in the bibliography section of the work. ……………………………………. ……………………………………. Ernest Kofi Agorde Dr. Gilbert Tietaah (Student) (Supervisor) Date:……………………………… Date:……………………………… i University of Ghana http://ugspace.ug.edu.gh ABSTRACT This study was aimed at identifying the dominant frames in the coverage of the West African Ebola Virus Disease (EBOLA) by CNN, Global News Network, Liberia and Daily Nation, Kenya. It was also designed to identify any relations between newsmakers and the issue- specific frames, examine ethical issues that emerged from the coverage and investigate the kinds of media representations of Africa in the coverage. A quantitative content analysis was carried out to study a total of 555 online news items by CNN, GNN and Daily Nation from the period 1st March, 2014 to 31st December, 2014 through a composite week sampling. The study found that “action”, “consequence” and “reassurance” frames were the major issue-specific frames used, with “action” being the most dominant. More than half of the stories were centred on disease statistics which included cases and casualty figures as well as conditions of Ebola patients. Majority of the stories were also found to be episodic in nature suggesting the predominance of coverage of events as incidences, rather than coverage of the disease holistically by situating it in context through the provision of background information. No differences were found in frames irrespective of the newsmaker involved. Newsmakers were framed in a similar way, with majority having action, consequence and reassurance frames being the most dominant. The media representation of Africa was generally neutral in the coverage, although CNN tended to be more stereotypical in its reporting on Africa. The media in this study did not embrace constructive journalism frames as alternative ways of reporting on Africa. Majority of the stories analysed recorded no issues of ethical concern. Finally, it was found that 63.5% of Daily Nation’s news were sourced from a non-African news agency while only 5.9% of GNN’s news came from its own journalists. The study recommends greater reliance on primary indigenous news sources such as African news agencies in the coverage of African issues, in order to avoid transferring any kinds of biases that may have been carried in the news of the foreign media houses. ii University of Ghana http://ugspace.ug.edu.gh DEDICATION I dedicate this work to the Almighty God, my family and all my friends for their overwhelming support that has seen me through the course of my study. Without my family especially, I would not have been able to come this far. iii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT Various people have contributed to my ability to successfully complete this course. My first gratitude goes to the Almighty God who has made my study a success. I wish to also acknowledge my supervisor, Dr. Gilbert Tietaah, whose valuable contributions and direction helped shape the work and enabled me to go through from the beginning to the end of this project. Thank you very much Dr. Tietaah and God richly bless you for all the support. I also want to acknowledge my mother, Miss Justine Ameka, my grandmother, Mrs. Isabella Ameka, my Uncles: Gabriel, Felix and Lawrence Ameka, my aunties, Salome and Regina Ameka, Mama Joan, my siblings, Fausta, Faustine, Bismark and Hilda, and my cousins for their financial and moral support throughout the duration of my study. Without you I know I would not have made it this far. May God richly bless all of you for the support. My appreciation also goes to all the lecturers of the Department of Communication Studies for contributing to the knowledge that I have acquired over the duration of this course, which is going to propel me to higher heights. May God continue to give you the wisdom to inspire many others. I also thank the administrative and support staff of the department, especially, Aunty Lizzy, Mr. Benson Osei Tutu, and Dame for all their help during my course of study. I want to also extend my appreciation to my MA and MPhil course mates for being part of this success story. Through the good and the bad, we have been able to successfully come to the end of the course and I know this is just the beginning of long lasting friendships and greater things to come. Finally, I want to thank all my friends who have supported me in various ways to enable me successfully complete my study. May God reward you abundantly for your support. iv University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION....................................................................................................................... i ABSTRACT .............................................................................................................................. ii DEDICATION........................................................................................................................ iii ACKNOWLEDGEMENT ...................................................................................................... iv TABLE OF CONTENTS ........................................................................................................ v LIST OF TABLES ............................................................................................................... viii LIST OF FIGURES ................................................................................................................ ix LIST OF ABBREVIATIONS ................................................................................................. x CHAPTER ONE: INTRODUCTION .................................................................................... 1 1.0 Background ................................................................................................................................... 1 1.1 The role of media in Health Communication ................................................................................ 2 1.2 Ebola and the History of Ebola Outbreaks .................................................................................... 4 1.3 Statement of the Problem ............................................................................................................ 5 1.4 Objectives of the Study ................................................................................................................. 7 1.5 Research Questions ...................................................................................................................... 7 1.6 Hypothesis Testing ........................................................................................................................ 8 1.7 Significance of the Study ............................................................................................................... 9 1.8 Background on Media Studied .................................................................................................... 10 1.8.1 Cable News Network (CNN) ................................................................................................. 10 1.8.2 Global News Network, Liberia (GNN)................................................................................... 10 1.8.3 Daily Nation, Kenya .............................................................................................................. 11 1.9 Scope of the Study ...................................................................................................................... 11 1.10 Organisation of the Study ......................................................................................................... 11 1.11 Chapter Summary ..................................................................................................................... 12 CHAPTER TWO: THEORETICAL FRAMEWORK ...................................................... 13 2.0 Introduction ................................................................................................................................ 13 2.1 Framing Theory ........................................................................................................................... 13 2.2 Chapter Summary ....................................................................................................................... 20 CHAPTER THREE: RELATED STUDIES ....................................................................... 21 3.0 Introduction ................................................................................................................................ 21 3.1 International media coverage of Ebola and related health epidemic outbreaks ....................... 22 3.2 Western media coverage of Ebola and related health epidemic outbreaks .............................. 29 3.3 African media coverage of Ebola and related health epidemic outbreaks ................................. 36 3.4 Asian media coverage of Ebola and related health epidemic outbreaks ................................... 39 v University of Ghana http://ugspace.ug.edu.gh 3.5 Chapter Summary ....................................................................................................................... 41 CHAPTER FOUR: METHODOLOGY .............................................................................. 42 4.0 Introduction ................................................................................................................................ 42 4.1 Research Design .......................................................................................................................... 42 4.2 Population ................................................................................................................................... 44 4.3 Sampling Technique .................................................................................................................... 45 4.4 Sample Size ................................................................................................................................. 46 4.5 Unit of Analysis ........................................................................................................................... 46 4.6 Data Collection Technique and Instrument ................................................................................ 47 4.7 Data Analysis ............................................................................................................................... 54 4.8 Intercoder Reliability ................................................................................................................... 55 4.9 Chapter Summary ....................................................................................................................... 55 CHAPTER FIVE: FINDINGS.............................................................................................. 57 5.0 Introduction ................................................................................................................................ 57 5.1 Distribution of Ebola News Coverage ......................................................................................... 57 5.2 Type of Articles ........................................................................................................................... 60 5.3 Subject Matter of Stories ............................................................................................................ 62 5.4 Source of Stories ......................................................................................................................... 64 5.5 Story Enhancement ..................................................................................................................... 66 5.6 Reflection of Enhancement on Story .......................................................................................... 69 5.7 News Makers/Story Actors ......................................................................................................... 70 5.8 Episodic and Thematic Framing .................................................................................................. 73 5.9 Causal Responsibility Framing ..................................................................................................... 75 5.10 Treatment Responsibility Framing ............................................................................................ 77 5.11 Issue-Specific Frames ................................................................................................................ 79 5.12 Media Representation .............................................................................................................. 81 5.13 Ethical Issues in Coverage ......................................................................................................... 84 5.14 Comparing News Makers and Coverage Frames ...................................................................... 86 5.15 Chapter Summary ..................................................................................................................... 89 CHAPTER SIX: ANALYSIS AND DISCUSSION OF FINDINGS.................................. 90 6.0 Introduction ................................................................................................................................ 90 6.1 Dominant Frames in the Coverage ............................................................................................. 90 6.1.1 Test of Hypothesis One: ..................................................................................................... 101 6.2 Issue-Specific Frames and News Makers Across media ............................................................ 102 6.3 Ethical Issues in the Coverage ................................................................................................... 105 6.4 Media Representation .............................................................................................................. 108 vi University of Ghana http://ugspace.ug.edu.gh 6.4.1 Test of Hypothesis Two: ..................................................................................................... 112 6.5 Chapter Summary ..................................................................................................................... 116 CHAPTER SEVEN: CONCLUSIONS AND RECOMMENDATIONS ........................ 117 7.0 Introduction .............................................................................................................................. 117 7.1 Summary of Findings ................................................................................................................. 117 7.2 Implications of Findings ............................................................................................................ 121 7.3 Conclusion ................................................................................................................................. 122 7.4 Limitations ................................................................................................................................. 123 7.5 Recommendations for further studies ...................................................................................... 123 7.6 Recommendations for Industry ................................................................................................ 124 7.7 Chapter Summary ..................................................................................................................... 124 REFERENCES ..................................................................................................................... 125 APPENDIX 1 ........................................................................................................................ 130 APPENDIX II ....................................................................................................................... 134 APPENDIX III ..................................................................................................................... 135 APPENDIX IV ..................................................................................................................... 137 vii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table Page Table 1: Monthly Distribution of Ebola News 58 Table 2: Comparing Monthly Distribution of Ebola News 59 Table 3: Comparing Types of Stories across Media 61 Table 4: Comparing Media and Subject Matter 63 Table 5: Comparing Media and Sources of News 65 Table 6: Comparing Enhancement and Media 68 Table 7: Comparing Media and Reflection of Enhancement on Story 70 Table 8: Newsmakers (Story Actors) 71 Table 9: Comparing Media and Newsmakers 72 Table 10: Comparing Episodic and Thematic Framing across Media 74 Table 11: Comparing Causal Responsibility Framing and Media 76 Table 12: Treatment Responsibility Framing 77 Table 13: Comparing Treatment Responsibility Framing and Media 78 Table 14: Issue-Specific Frames 79 Table 15: Comparing Issue-Specific Frames and Media 80 Table 16: Comparing Constructive Journalism Frames and Media 82 Table 17: Media Representation (Constructive Frames) across Media 84 Table 18: Ethical Issues in Coverage 85 Table 19: Ethical Issues in Coverage across Media 86 Table 15a: Comparing Issue-Specific Frames and Media 101 Table 17a: Media Representation (Constructive Frames) across Media 113 Table 20: Comparing Constructive Journalism Frames in the coverage of CNN and GNN 114 Table 20: Comparing Constructive Journalism Frames in the coverage of CNN and GNN 115 viii University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure Page Figure 1: Number of Ebola News Stories Analysed 58 Figure 2: Types of Stories 61 Figure 3: Subject Matter of Stories 62 Figure 4: Sources of Stories 64 Figure 5: Story Enhancement 66 Figure 6: Reflection of Enhancement on Story 69 Figure 7: Episodic and Thematic Framing 74 Figure 8: Causal Responsibility Framing 75 Figure 9: Constructive Journalism Frames 83 ix University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS Abbreviation Meaning CNN Cable News Network GNN Global News Network EBOLA Ebola Virus Disease PHEIC Public Health Emergency of International Concern US(A) United States (of America) WHO World Health Organisation UN United Nations CDC Centers for Disease Control and Prevention ECOWAS Economic Community of West African States x University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.0 Background The Ebola outbreak in West Africa was declared a Public Health Emergency of International Concern (PHEIC) by the World Health Organisation (WHO) on 8th August 2014, to commence a global fight for the reduction of the international spread of the outbreak. Reaching the status of a PHEIC meant that the outbreak constituted an “extraordinary event” and a public health risk to other states (World Health Organisation, 2016). Furthermore, WHO noted that: The possible consequences of further international spread are particularly serious in view of the virulence of the virus, the intensive community and health facility transmission patterns, and the weak health systems in the currently affected and most at-risk countries. A coordinated international response is deemed essential to stop and reverse the international spread of Ebola (WHO, 2016). The West African Ebola virus which broke out in Guinea in December 2013 has been identified as the largest, most complex and most persistent outbreak of the Ebola Virus Disease (EVD) since it was first identified in 1976. It is also the first case of the outbreak to have spread beyond East and Central Africa (Frieden et al., 2014; WHO 2017). The outbreak also constituted one of the most devastating news around the world in the second half of 2014 (Nwanne 2014). The epidemic, had claimed the lives of 11,323 people out of a total of 28,646 cases as at 27th March 2016. There are however over 10,000 Ebola survivors. The West African sub region was the worst hit with Liberia, Guinea and Sierra Leone recording the highest number of cases (28,610) and casualties (11,308). There were single cases each in Italy, Spain, the United Kingdom and Senegal. Nigeria had 20 cases and eight casualties, Mali had eight cases and six casualties, and the United States of America recorded four cases and one casualty (WHO, 2016a). 1 University of Ghana http://ugspace.ug.edu.gh The PHEIC status of the outbreak was declared ended on 29th March, 2016 after advice from the International Health Regulations Emergency Committee of the WHO after an assessment of the Ebola situation (WHO, 2016b). The international and local media played major roles in the dissemination of news on the Ebola outbreak. While local radio and television stations raised awareness about the disease and provided people with information on preventive measures, the international media tended to spread panic and fear in the world system by the nature of their coverage (Yusuf et al. 2015). 1.1 The role of media in Health Communication The role of the media in everyday life and its ability to influence what people think about and how they react to issues has long been established (Kiousis, 2011; Roberts & McCombs, 1994; McCombs, 2005; McCombs & Shaw, 1972). Kasperson et al. (1988) stated that the media’s repeated coverage of risk, draws public attention to it and away from other competing sources of concern. This means that how the media reports an issue would eventually affect how people form perceptions about that issue and their attitude towards it. The media has also been identified as having very important roles to play in public health education, risk, surveillance, and general health communication. Frewer et al. (2002) noted that individuals have at least two main sources of information regarding risks from which they base their judgements of the risk: the media and interpersonal networks. Furthermore, it can be argued that beyond the media, interpersonal networks constitute a form of two-step or multi- step flow of information which necessarily thrives on information from the media in the first instance. Therefore, the media are an inherent part of all health information seeking practices of people. Frewer et al. (2002) stated further that the media has a duty to disseminate information to warn the public about health concerns. Cullen (2014) also argued that the media effectively contributes to disease prevention, containment and control through informing the 2 University of Ghana http://ugspace.ug.edu.gh public with evidence-based facts. The media, through accurate and timely coverage of diseases, could lessen fear and stigma (Cullen, 2014). Online media has become a platform which generates massive amounts of information every day. As Cortesi and Gasser (2015: 1426) put it “the internet has become an important reservoir of information and site for its actual use”. Despite the importance of the platform, MacIntyre and Travaglia (2015) argued that in emergency health situations, the lack of oversight and loose gatekeeping nature of the platform, affords non-experts the opportunity to present themselves as experts and give information which may not be true. There is therefore the need to find out whether institutional media outlets behaved well on online platforms or whether they behaved unethically by relying on non-experts and using stereotypes. Also, despite the role that the media is supposed to play in health issues, research on media coverage of disease outbreaks such as AIDS, Ebola and SARS has revealed that the media had the potential of causing panic, perpetuating stigma and even contributing to the spread of the disease in question (Zhang & Matingwina 2016). This demonstrates the ‘double-edged sword’ nature of the media. That is, the media could contribute positively to health communication by helping people to make informed choices or they could contribute negatively, leading to people forming negative perceptions about health issues and negative attitudes towards those affected by the health issue involved. It is therefore prudent that we are interested in how the media cover issues like the Ebola epidemic. The forgoing arguments coupled with the claim by Yusuf et al. (2015) that the way some international media covered the West African Ebola epidemic contributed to spreading fear and panic into the world system, motivated the need for this study which sought to find out how an international media (CNN), a national media (GNN), and a regional (Daily Nation), 3 University of Ghana http://ugspace.ug.edu.gh covered (framed) the epidemic. The purpose is also to identify if the media had been used positively or negatively in the context of the latest outbreak as covered by these media outlets. 1.2 Ebola and the History of Ebola Outbreaks The Ebola Virus Disease (EVD), previously known as Ebola haemorrhagic fever, is a rare and deadly disease caused by an infection with a virus of the Filoviridae virus family (CDC, 2017). The natural reservoir host of Ebola virus remains unknown but researchers believe that the virus is animal-borne and the likely reservoir are bats. Five species of the Ebolavirus are identified, among which four - Ebola virus (Zaire ebolavirus), Sudan virus (Sudan ebolavirus), Taï Forest virus (Taï Forest ebolavirus, formerly Côte d’Ivoire ebolavirus), and Bundibugyo virus (Bundibugyo ebolavirus) - are known to cause disease in both humans and non-human primates (such as monkeys, gorillas and chimpanzees) while the fifth, Reston virus (Reston ebolavirus), has only caused diseases in nonhuman primates, but not in humans. The Bundibugyo ebolavirus, Zaire ebolavirus, and Sudan ebolavirus have been associated with large outbreaks in Africa (WHO, 2017; CDC, 2017). Ebola can be acquired through direct contact with the blood and body fluids, such as urine, saliva, and sweat of an infected person or someone who has died from the disease. Other modes of getting the disease are through objects such as needles, and syringes, contaminated by an infected or dead person’s body fluid, infected primate (apes and monkeys) or fruit bats, and through contact with the semen (through sexual intercourse) of someone who has recovered from the disease (CDC 2017). Ebola was discovered in 1976 with two simultaneous outbreaks: one occurred near the Ebola River in the Democratic Republic of the Congo, from which the virus takes its name. This was caused by the Zaire virus and was spread through close personal contact and the use of contaminated needles and syringes in hospitals/clinics. There were 318 cases and 280 deaths representing an 88% fatality rate. The other outbreak occurred at Nzara, Maridi and 4 University of Ghana http://ugspace.ug.edu.gh surrounding areas in what is now South Sudan. There were 284 cases and 53% casualties which included a lot of health personnel. Its spread was mainly through close personal contacts within hospitals and was caused by the Sudan virus. A third case occurred in England through laboratory infection of one person by a contaminated needle. It was also caused by the Sudan virus (CDC, 2016). Since 1976 when the first cases were identified, there have been 34 outbreaks of the Ebola virus aside from the West African Ebola epidemic. These produced 2,427 cases with 1,597 fatalities which is 65.8% of the cases. The West African outbreak, which was caused by the Ebola virus, resulted in 28,646 cases and 11,323 deaths. This means that there have been more cases and casualties in the West African outbreak than all the other outbreaks combined (WHO 2016). This explains why scholars (Nwanne, 2014; Frieden et al. 2014) and WHO describes it as the most devastating of all the outbreaks. It can also be observed that there are various strains of the virus which mutate differently in the in each outbreak hence the difficulty in addressing new outbreaks with vaccines from previous outbreaks. 1.3 Statement of the Problem Studies on Ebola have shown that the way the media reported the epidemic led to people forming negative perceptions about the virus (Yusuf et al. 2015). Ethical breaches on patients’ privacy (MacIntyre & Travaglia, 2015) and provision of misleading information (Pathak et al, 2015) were also found with media reports about the epidemic. MacIntyre and Travaglia (2015) concluded that disasters such as disease outbreaks lend themselves as occasions for ethical breaches in the publishing of health information. Such concerns flow from some empirical evidence about negative, stereotypical and sometimes unethical, reports and representations by Western media. These studies have tended to either examine how the media of different countries covered the outbreak of Ebola (Abeysinghe, 2016; Zhang & Matingwina, 2016; Yusuf et al., 2015) or compared how different media within a specific country covered the 5 University of Ghana http://ugspace.ug.edu.gh outbreak (Paul et al., 2016; Smith & Smith, 2016; Obukoadata & Abuah, 2014). While the findings of these studies are indicative, they raise the research curiosity about a possible association of the nature of framing of outbreak stories and such variables as proximity to incidence, cultural orientation, journalistic traditions and other stereotypical particularities. Furthermore, MacIntyre and Travaglia (2015) note that the internet serves as a critical source of rapid information. In the domain of health information and outbreak news reporting, this is an important attribute to be leveraged, particularly as audience statistics point to the fact that online sources have unsurprisingly, become the most patronized for news consumption, particularly among the youthful members of the population (McQuail, 2010; Cortesi & Gasser, 2015; Lenhart, 2015; Salman & Saad, 2015). Thus, online media has become a major, but little examined source of health information. Secondly, the internet is a platform that has the potential and ability to converge multimedia; that is, text, graphics, audio and video, in its stories. The convergence of these media attributes suggests a greater opportunity for audience engagement and comprehension, than their individual components could enable. This raises the question as to what extent the opportunities of online media for contributing to public awareness message presentation/dissemination being exploited? Notwithstanding these important factors, MacIntyre & Travaglia (2015) point out that the internet is an unregulated source of information, further raising the question of the extent to which the challenges of unregulated news sources were being avoided. This should raise both scholarly and policy research interest, with particular reference to health and infectious disease outbreak communication. Yet, there is little literature on studies that sought to compare an international media, (media outside Africa), regional media (media within Africa but removed from the outbreak region) and national media (media within an affected country) in terms of their online news coverage of the Ebola epidemic. This study sought to address this gap by 6 University of Ghana http://ugspace.ug.edu.gh comparing coverage of the West African Ebola epidemic by CNN (international media), Daily Nation (regional media) and GNN (national media). 1.4 Objectives of the Study The ultimate objective of this study was to find out how the international (CNN), national (GNN) and regional (Daily Nation) media outlets framed the West African Ebola epidemic. The specific objectives that underpinned the study were to: 1. identify the dominant frames used in the coverage of the West African Ebola Virus Disease (EVD) epidemic by CNN, GNN and Daily Nation. 2. identify whether the issue-specific frames differed based on newsmakers across the media in the coverage of the Ebola epidemic by CNN, GNN and Daily Nation. 3. examine any ethical issues arising from the coverage of the West Africa Ebola Virus Disease (EVD) epidemic. 4. investigate the media representations of Africa in the coverage of the West Africa Ebola Virus Disease (EVD) epidemic. 1.5 Research Questions To fulfil the objectives for this study, the following research questions were set: RQ1: What were the dominant frames in the coverage of the West Africa Ebola Virus Disease (EVD) epidemic by CNN, GNN and Daily Nation? RQ2: To what extent did the main issue-specific frames in the coverage of the West Africa Ebola Virus Disease (EVD) epidemic differ based on newsmakers across the media studied? RQ3: What ethical issues emerged in the coverage of the West Africa Ebola Virus Disease (EVD) epidemic? 7 University of Ghana http://ugspace.ug.edu.gh RQ4: What are the kinds of media representations received by Africa in the coverage of the West Africa Ebola Virus Disease (EVD) epidemic? 1.6 Hypothesis Testing In addition to the research questions set for this study, two hypotheses were tested. The hypotheses, the null hypotheses and the reasons for testing them are presented below. Hypothesis 1 H1: The differences in issue-specific frames used in coverage of the epidemic are significantly related to differences in media types. H0: There is no significant relationship between the issue-specific frames in the coverage of Ebola and media types. Shoemaker and Reese (1991) suggest that certain factors can influence media content and therefore determine how the media frame a particular issue. These factors include influences from individual media workers, media routines, and the ideology of that media. This could account for why the same issue would be reported differently by different media outlets. Since this study was comparing coverage of one public health epidemic (Ebola), in three different media outlets with different ideologies and routines, the study sought to ascertain the extent to which these differences produced differences in issue-specific frames. The purpose of Hypothesis 1, therefore, was to determine whether there was any significant relationship between the issue-specific frames and media outlets. The test would also determine whether the relationship, if any, was a strong or weak relationship. Hypothesis 2 H2: The kinds of media representation of Africa found in Ebola stories are significantly related to media outlets. 8 University of Ghana http://ugspace.ug.edu.gh H0: There is no significant relationship between the kinds of media representations of Africa in Ebola stories and media outlets. Ogunyemi (2011) stated that the most dominant perspective on representation of Africa in the western media was biased, crisis oriented, marginalized and contained negative stereotypes. He pointed out that the western media frames on Africa revolved around poverty, instability, disease, illiteracy and conflict. He further noted that western media constructed negative images and media representations of less developed countries. Since the current study was comparing the coverage of Ebola, a global health concern, by western and African media, and also interested in finding out how the media represented Africa in their coverage, this hypothesis was to determine whether, the differences, if any, that would be identified in the coverage were statistically significant. The test would also determine whether the relationship that may be identified was strong or weak. In order to test for relationships between western media and Africa media, Hypothesis 2 was tested twice. This was because there were two African media (GNN and Daily Nation) and one western media (CNN). The relationship between CNN and GNN was determined by Test A, and Test B was used to determine the relationship between CNN and Daily Nation. 1.7 Significance of the Study Various studies have been conducted on the coverage of the recent outbreak of the Ebola virus in the West African sub region. However, there is scant literature comparing coverage by international media (media outside the outbreak continent, that is, Africa), national media (media in the theatre of the disease) and regional media (an African media, non-affected and removed from the outbreak region). This study will fill this gap. The study also adds and contributes to existing knowledge on media coverage, specifically framing of public health epidemics in general and Ebola in particular. 9 University of Ghana http://ugspace.ug.edu.gh The study would also shed light on the similarities and differences in media representations of Africa in the coverage of Ebola by the different media outlets. The study will provide empirical evidence to guide policy and training interventions on health reporting and public health communication. 1.8 Background on Media Studied 1.8.1 Cable News Network (CNN) CNN is an American cable and satellite television news organisation founded in 1980 by the Time Warner Company. It is rated as the leading international news brand reaching four out of ten upscale consumers and decision makers in Europe, Middle East, Africa, Asia and Latin America (CNN, 2016). According to Shoemaker (1996) CNN’s strength lies in its ability to simultaneously transmit news from any part of many countries making it the quickest way “to get information about what’s happening around the world” (p.48). CNN also has the potential to set the agenda for other news outlets (Thussu, 2003). CNN, representing the western media, was chosen to find out how the western media covered the epidemic, in comparison with how the African media covered it. The frames used by CNN was relevant for finding out the kind of representation Africa received in the coverage on Ebola. 1.8.2 Global News Network, Liberia (GNN) Global News Network (GNN) newspaper is an English language newspaper in Liberia which reports political, sports, education, world, national, local, current affairs, business, science, technology, life style, health, fashion and general news. An initial online search on 25th May, 2016, to case the scene (Wimmer & Dominick, 2011) showed that GNN had a lot of news on Ebola as compared to other online platforms in Liberia. Liberia is one of the three countries that were heavily hit by the West Africa Ebola virus epidemic. It is selected to see how media in an affected country covered the outbreak. 10 University of Ghana http://ugspace.ug.edu.gh 1.8.3 Daily Nation, Kenya The Daily Nation is a brand of the Nation Media Group (NMG) of Kenya which was founded in 1959. As the leading multi-media house in the East African region, it has print, electronic as well as an internet media platform which attracts a regular readership claimed to be unparalleled in the region (http://www.nationmedia.com/who-we-are/). The Daily Nation is reported to have 170,000 circulations (Bloomberg News, 2014). Daily Nation was included in the study to find out how a leading African media removed from the epidemic’s immediate environment, covered the epidemic. 1.9 Scope of the Study The study sought to examine online framing of the West African Ebola epidemic. The Cable News Network (CNN), United States of America, Global News Network (GNN), Liberia and Daily Nation, Kenya were selected for the study. The three were selected in order to compare how a media outlet outside of Africa, a media outlet in one of the outbreak countries and a media outlet in Africa but removed from the outbreak region, framed their news on the Ebola epidemic online. 1.10 Organisation of the Study This study is organised into seven chapters. Chapter One introduced the study by providing a background of the entire study. The chapter also presented an overview of the Ebola virus and a history of Ebola outbreaks since 1976. It further explained the research problem, objectives of the study, research questions, significance of the study, organisation of the study and summary of the chapter. Chapter Two discussed the theoretical framework for the study. This was made up of the framing theory, supported by media representation, and stereotyping. Chapter Three reviewed literature on Ebola, studies on media coverage of public health epidemics and other diseases similar to Ebola, to the extent that these studies were relevant to the current study. The fourth 11 University of Ghana http://ugspace.ug.edu.gh chapter presented the methodology that was used in the study. Specifically, the research design, population, sampling technique, sample size, unit of analysis, content categories, data collection and analysis procedures were discussed. In Chapter Five, the findings of the study as related to the objectives and research questions are presented. Other findings not directly captured as research objectives or questions but which helped to provide perspectives for a comprehensive analysis of the coverage of the Ebola epidemic were also presented. Chapter Six presents a discussion of the findings in relation to the theory and reviewed literature. Chapter Seven concluded the work by presenting a summary of findings and practical implications of the findings. It also addressed the limitations of the study and made recommendations for industry and further studies. 1.11 Chapter Summary This chapter presented the background to the study. It gave an overview of the West African Ebola epidemic, highlighting the fact that the outbreak was the one with most cases (28,646) and casualties (11,323) since the Ebola virus was first discovered in 1976. The surveillance function of the media in health communication and in providing accurate and timely information during disease outbreaks, the objectives of the study, research questions and significance of the study were also presented. The background of media studied, the scope of the study and the organization of the study, sum up the topics discussed in the chapter. 12 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO THEORETICAL FRAMEWORK 2.0 Introduction This chapter presents the theoretical framework that underpins the study. Framing, stereotyping and media representation theories were examined with respect to the way they relate to the objectives of the study. Framing is the main theory for the study whereas stereotyping and media representation complement Framing, to help cater for the limitations in addressing some of the objectives and research questions. 2.1 Framing Theory Framing has its roots in sociology. Its foundation is often attributed to Erving Goffman, a sociologist, who held the view that interpretive designs constitute central elements of cultural belief systems (Littlejohn & Foss, 2009). He called the interpretive designs “frames” that we use in our daily lives to make sense of the world. Goffman noted that these frames help in the reduction of complex processes, interpreting and reconstructing reality. Goffman also held that meanings of things in the world were traditionally “conveyed through socialisation, creating a collective reality within a culture or society” (Littlejohn & Foss, 2009, p.408-409). However, in today’s world, the media delivers powerful frames of perception of the world that challenge and renegotiate lifeworld experiences (Littlejohn & Foss, 2009) Applied to media coverage, framing has acquired various conceptualizations due to its wide use in media and communications research. Littlejohn and Foss (2009) described framing to constitute how news stories are organised, defined and structured. They argued that even when journalists seek to be objective and balanced in their coverage, their reports give audiences cues as to how to understand the issues and direct them on which parts to focus on, which parts to ignore and how to evaluate the issues. 13 University of Ghana http://ugspace.ug.edu.gh According to Entman (1991), “to frame is to select some aspects of a perceived reality and make them more salient in a communication text, in such a way as to promote a particular problem, definition, causal interpretation, moral evaluation and or treatment recommendation for the item described” (p.52). It is a useful theoretical approach for analysing media content because frames highlight parts of information about an item that is the subject of a communication, so as to elevate their importance (Entman,1993). McQuail (2005) notes that in mass communication, it is assumed that the media do not only focus on the issues they cover but also the attributes (frames) of those issues which eventually influence how the public think about the issues and sometimes how they react. de Vreese (2005) argued that the media has become an important part of democracies and may shape public opinion through framing of issues and events in a certain way. To him, framing denotes how a communication source presents and defines an issue. He notes further that a frame refers to “an emphasis in salience of different aspects of a topic” (p.53). This suggests that framing or frames orient audiences to view a particular issue in the angle the media highlights. de Vreese (2005) identifies various typologies of frames. These include media frames, individual frames, issues-specific frames, and generic frames. Unlike issue-specific frames, which are pertinent to specific topics and events such as the internet, women’s movement and labour issues, generic frames transcend thematic limitations and can be employed in the study of different topics, and cultures (de Vreese, 2002). Generic frames include responsibility frames, and episodic frames and thematic frames. Iyengar (1991) suggests that news reports may be usefully analysed by their thematic or episodic content. Episodic framing depicts issues as particular instances or specific events, which lead to the exclusion of important details. Thematic framing, however, places events in a broader context of related events and takes the form of in-depth backgrounded reports. He stated that the standards and norms of news organisation and production reinforce episodic framing. However, because episodic framing 14 University of Ghana http://ugspace.ug.edu.gh “simplifies complex issues to the level of anecdotal evidence” (Iyengar, 1991, p.136-137), this induces a disorganised and isolated, rather than a general and contextual understanding of the issues involved. It is therefore important to investigate the generic frames because whether a news piece is episodic or thematic will affect the way people understand the issues presented (de Vreese, 2005). This study will analyse the episodic and thematic content of the online news by CNN, GNN and Daily Nation. Iyengar (1991) found that daily news coverage was biased towards episodic frames. In a disease situation like the Ebola epidemic, one typology of generic frames of interest is Iyengar’s (1991) responsibility frames. Responsibility frames include both causal and treatment reasonability. Causal responsibility addresses the source of the problem, that is, identifying someone or something as the cause, while treatment responsibility addresses the solution by identifying someone or something as the one to provide solutions to the problem that has been identified (Nitz & Ihlen, 2009). Iyengar (1991) found that individual attributions of cause and treatment responsibility tended to be associated with episodic framing, and thematic framing tended to attract societal attribution in terms of responsibility for cause and treatment. This study analysed the coverage by CNN, GNN and Daily Nation to identify the attributions of responsibility for cause and treatment of the Ebola outbreak. de Vreese (2005) has suggested two approaches of identifying frames in news. These are the inductive and deductive approaches. In the inductive approach, frames are studied in a way that enables the frames emerge from the data during the course of analysis. That is, the approach refrains from analysing news stories with a predefined news frame typology. The disadvantage of this approach is that, it can only be used to analyse small samples of data and it is difficult to replicate. The second approach is the deductive approach, which investigates frames that are defined and operationalized prior to the investigation. Scholars have recommended this 15 University of Ghana http://ugspace.ug.edu.gh approach for content analyses. The current study investigated the coverage of the West African Ebola Virus Disease epidemic using primarily the deductive approach with the framing typology defined and operationalised by Shih et al. (2008) and used to study print media framing of public health epidemics in the United States of America. Six predefined frames were investigated; namely: consequence, uncertainty, action, reassurance, conflict, and new evidence frames. The consequence frame referred to stories focused on the consequences of the diseases, such as human life (victims), social impact, or economic impact (cost). In addition to damages, it also included any phenomenon, political/social issues, events, or discussion generated by the occurrence or spread of diseases. The uncertainty frame referred to uncertainties in any aspect(s) of the epidemics including the cause, the cure, and possible spread. It also included the portrayal of the disease as something unknown that is in need of more exploration or examination by the experts or governments, and ultimately, the media. The action frame involved stories that stressed any action(s) against the disease, including prevention, potential solutions, or strategies. Reassurance frame stories expressed the idea that the public should not be worried about the effects of the disease. Stories that emphasized the readiness and/or successes of authorities in combating the disease were also included. Conflict frame stories focused on the difference in opinions as well as outright disagreements/arguments among news makers. It could be a debate about how to effectively combat the disease, disagreement about how diseases would evolve and how seriously it would 16 University of Ghana http://ugspace.ug.edu.gh affect people, or dispute over the appropriateness or legitimacy of actions. Conflict stories were constructed as antagonism between opposing opinions or stances. The new evidence frame referred to stories on new results/findings of research efforts or discovery of new evidence that helped advance the understanding of the disease or the ability to quell the disease. Included in this frame were discovery of new strains of the disease, new ways of transmitting/spreading, new methods to cure/treat/prevent the disease, and development of new medicine. In a bid to account for any new frames that might emerge from the data, an “other” category was added to the typology for the study. According to de Vreese (2005), the use of the deductive approach to studying frames, raises the question of which component of the news story (what), constitutes a frame. Entman (1993) suggested that frames can be examined and identified in the news by “the presence or absence of certain keywords, stock phrases, stereotyped images,…information and sentences that provide thematically reinforcing clusters of facts or judgements” (p.52). Shah et al. (2002) thinks frames can be identified in “choices about language, quotations, and relevant information” (p.367). To Gamson and Modigliani (1989) frames can be identified through framing devices, such as metaphors, exemplars, catch phrases, depictions and visual images. Tankard (2001) suggests 11 focal points that will help in identifying and measuring news frames. They are headlines, subheads, photos, photo captions, leads, source selection, quotes selection, pull quotes, logos, statistics and charts, and concluding statements and paragraphs. All the suggested elements of news that help in identifying frames were considered in analysing the frames in the coverage of the CNN, GNN and Daily Nation on the Ebola epidemic. Some scholars have also found that the way news is framed, determines its tone and therefore how the public perceives the issue being carried by the news. de Vreese (2004) states that the media can have a strong impact in constructing social reality and frames of reference for the 17 University of Ghana http://ugspace.ug.edu.gh audience. de Vresee (2005) again argues that the selection of frames by the media can determine the tone of coverage and has implications for how individuals and society perceive the issue at stake. The effects of these frames may be positive, negative or neutral based on how the issue is presented by the media. The frames identified in this coverage helped to know whether the coverage was positive or negative. Framing becomes even more important because of the implications of media representation and stereotyping. Media representation deals with the ways in which the media portray particular groups, communities, experiences, ideas, or topics from a particular ideological or value perspective. The type of stereotype that is associated with a group of people or ideas influences the kinds of representation they are given in the media. This means that with respect to the coverage of the Ebola epidemic, certain kinds of frames of reference to Africa, will lead to a type of media representation, which could eventually result in reinforcing existing stereotypes of Africa as poverty-ridden and disease-stricken (Ogunyemi, 2011). This study offers an opportunity to determine whether these charges of negative representation or stereotypes are necessarily exclusively or principally perpetrated by Western media. In other words, to what extent are stereotypical representations of Africa a product of geography or common association and membership? Stereotypes originated from social psychology. According to Wittenbrink et al. (1997), stereotypes are the beliefs about what a group is like and causal claims that relate group characteristics to one another. Baumeister and Finkel (2010) define a stereotype as a generalised belief about the characteristics of a group. They are about how people's thoughts, feelings, and behaviours are influenced by the actual, imagined, or implied presence of others. Stereotyping, therefore, represents the process of attributing the stereotypic characteristics to particular individuals only because of their membership in the group (Baumeister and Finkel, 18 University of Ghana http://ugspace.ug.edu.gh 2010). The stereotypic characteristics of a group constitute a causal chain of coherent account for why the groups are the way they are perceived to be. Brooks and Hebert (2006) state that in our consumption related society, much of what becomes regarded as important is based on the stories produced and disseminated by media institutions. They added further that much of what audiences know and care about is based on images, symbols and narratives in radio, television, film, music and other media. Media are therefore, central to what ultimately comes to represent our social reality. Ogunyemi (2011) states that the most dominant perspective on representation of Africa in the Western Media is biased, crisis oriented, marginalized and contains negative stereotypes. He further notes that western media, influenced by their ideological position or national interest considerations, will use their hegemonic power systematically to construct negative images and media representations of underprivileged others, especially the less developed countries. He points out that the hegemonic articulation of Africa in Western media revolves around poverty, instability, disease, illiteracy and conflict. To study the media representation and or stereotypes in the coverage of the West African Ebola epidemic, an adaptation of the constructive journalism frames used by Zhang and Matingwina (2016) for their study of the coverage of the West African Ebola epidemic by BBC and China Daily, will be used to investigate how CNN, GNN and Daily Nation narrated and represented Africa in their coverage of the Ebola epidemic. Constructive journalism offers alternatives to traditional negative stereotypes through the holistic coverage of issues from the perspective of people at the centre of the issues (Zhang and Matingwina 2016). News stories were studied to ascertain whether they were constructive, negative or neutral. Constructive stories reflected the themes of “accomplishments”, “hero” and “meaningful” narrative. These themes used positive psychological techniques and allowed for holistic coverage of Ebola stories. Negative news 19 University of Ghana http://ugspace.ug.edu.gh referred to news that narrated Ebola through stereotypical images of Africa to depict a sense of hopelessness and focused on issues such as fear and the use of the technique of ‘othering’ (blaming others) to blame the disease on people and their cultural practices (Zhang and Matingwina 2016). Aside from the technique of “othering”, negative news stories referred to stories that stressed on the portrayal of Africa as a place of violence, conflict, poverty, disease, corrupt governments and weak systems. Stories on Western superiority, negative cultural practices and portrayal of Africa as a single continent were also deemed negative. Neutral news included stories, which had neither the characteristics of constructive journalism, nor negative news; that is, news such as announcement of events or facts without comments. 2.2 Chapter Summary This chapter discussed the theoretical framework that underpinned the study. Framing theory was discussed bringing out its role in providing explanatory logic for the objectives of the study. Attributes of media representation and stereotyping were used to support the Framing theory for the purpose of studying the media representation of Africa by the CNN, GNN and Daily Nation. Episodic and thematic framing, responsibility frames and framing typology by Shih et al. (2008), specific to public health epidemics, were highlighted as attributes to drive the investigation of how CNN, GNN and Daily Nation covered the West African Ebola epidemic. Constructive Journalism frames were adopted to investigate media representation and stereotypes in the narrations of Ebola and Africa in the coverage of the three media outlets. 20 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE RELATED STUDIES 3.0 Introduction This chapter reviews related literature on the coverage of the West African Ebola Virus Disease (EVD) epidemic. The West African Ebola epidemic has received much scholarly attention, though from different ontological themes and within different geographical contexts and media traditions. A review of the literature produced at least 16 studies of direct bearing to the current study in terms of the subject matter of the West African EVD and coverage of other epidemics and disease outbreaks in as far as their assumptions, and implications relate to the current study. The preponderance of studies used content analysis (Zhang & Matingwina, 2016; Smith & Smith, 2016; Paul, Mahajan & Sahoo, 2016; Abeysinghe, 2016; Afzal et al., 2015; Washer, 2004; Nwanne, 2014; Shih, Wijaya & Brossard, 2008; El-Jardali et al., 2015; Basch, Basch & Redlener, 2014; and Cortiñas-Rovira, Pont-Sorribes & Alonso-Marcos, 2015). Only a couple employed other methods like audience surveys (Obukoadata & Abuah, 2014), or textual analysis (Abeysinghe & White, 2010) or mixed methods (El-Jardali et al., 2015) or experiment (Young et al., 2008). The majority of studies used national media (El-Jardali et al., 2015; Smith & Smith, 2016; Abeysinghe & White, 2010; Basch, Basch & Redlener, 2014; El-Jardali et al., 2015, Nwanne, 2014; Shih, Wijaya & Brossard, 2008; Cortiñas-Rovira, Pont-Sorribes & Alonso-Marcos, 2015; Washer, 2004; and Paul, Mahajan & Sahoo, 2016) and international media (Zhang & Matingwina, 2016; Abeysinghe, 2016; and Yusuf, Yahaya & Qabli, 2015). The studies ranged in temporal duration from four weeks to thirteen months. Theoretical underpinnings ranged from framing; social representation; constructive journalism, issue attention cycle, diffusion of innovation, to agenda setting. Most of the studies have been done on print media (Paul et al., 2016; Smith & Smith 2016; Abeysinghe, 2016; and Shih et al., 21 University of Ghana http://ugspace.ug.edu.gh 2008), with only a few on the internet and social media (Afzal et al. 2015, Yusuf et al., 2015; and MacIntyre & Travaglia, 2015). What seems outstanding is that there were similarities of the current coverage with how Ebola and other infectious diseases and epidemics were covered in the past. The cause of the epidemic and its spread, also formed part of the dominant issues in the coverage. In the following pages, these studies are briefly reviewed in order to draw out their points of interest to the current study, as well as demonstrate the extent to which they are similar to or differ from, each other. 3.1 International media coverage of Ebola and related health epidemic outbreaks Studies have been carried out on international media’s coverage of the West African Ebola outbreak, and other related diseases, key being one by Zhang and Matingwina (2016). Zhang and Matingwina (2016) examined the use of constructive journalism by the BBC and China Daily in the narrative on the 2013-15 outbreak of Ebola in West Africa. They noted that Ebola outbreaks had always received international media attention and the latest outbreak provided an opportunity and platform for both Western and non-Western media to narrate both Ebola and Africa. The study sought to specifically explore how different journalistic approaches were used to represent complex issues as well as the countries and people involved, to interrogate Western and Asian media journalistic styles employed and how they represented Africa, and to investigate whether alternative journalistic approaches such as constructive journalism could be recommended for reporting on Africa. Zhang and Matingwina’s (2016) study is relevant to the current study in its focus on examining dominant themes and representations of Africa in the coverage of Ebola by two foreign (non- African) media institutions. The current study, however, extends the research heuristic by comparing how western (CNN), national (Global News Network), and regional (Daily Nation) media covered the West African Ebola epidemic. 22 University of Ghana http://ugspace.ug.edu.gh The concept of constructive journalism was important in this study for finding out if the charge of negative, stereotypical coverage typically levelled against Western media was valid with respect to an outbreak declared to be a PHEIC in comparison with African media within or outside the country of outbreak. Constructive journalism offers alternatives to traditional negative stereotypes through the holistic coverage of issues from the perspective of people at the centre of the issues. The research questions that aided the Zhang and Matingwina (2016) study were to find out the slant of the Ebola stories by the BBC and China Daily, the dominant themes, whether new journalistic approaches emerged to cover crisis issues in Africa, and the possible implications of these for the narration of Ebola and the representation of Africa. Judgemental or selective sampling was used to select Ebola stories related to Africa for analysis. Thirty-two (32) stories each were selected from BBC and China Daily. Qualitative and quantitative content analysis techniques were employed for the study. The tone of coverage and salient themes of the two media were examined. The stories were categorised as constructive, negative or neutral in tone. Constructive stories reflected the constructive journalism themes of “accomplishments”, “hero” and “meaningful” narrative. These themes use positive psychological techniques and allow for holistic coverage of Ebola stories. Negative news referred to news that narrated Ebola through stereotypical images of Africa to depict a sense of hopelessness and focused on issues such as fear and the use of the technique of ‘othering’ (blaming others) to blame the disease on people and their cultural practices. Neutral news included stories which had neither the characteristics of constructive journalism nor negative news; that is, news such as announcement of events or facts without comments. Social representations theory, specifically previous studies on social representations of HIV AIDS and Ebola, were used to analyse the stories. 23 University of Ghana http://ugspace.ug.edu.gh Zhang and Matingwina (2016) found that the slant of stories by China Daily were more constructive (62.50%) than BBC (34.38%). Also, BBC’s stories were dominantly meaningful narrative stories, while China Daily had more hero narrative stories. BBC reported more negative (34.38%) and neutral (31.24%) stories than China Daily’s 18.75% each of negative and neutral stories. There were more stereotypes in BBC’s negative stories, dominated by the technique of ‘othering’ (blaming others), the cultural practices of affected countries, while fear mongering dominated the negative stories of China Daily. Despite this, BBC reduced the traditional Western media approach of negative news on Africa by employing some amount of constructive journalistic narratives. China Daily’s coverage, however, reinforced the use of alternative narratives by Chinese media on Africa. This was evident in the fact that China Daily had reported more positive news on Africa’s potential than the BBC. China Daily was also found to have reported more stories using the constructive journalism approach than the BBC. The dominant themes in the coverage by BBC and China Daily were the cause of the disease, the people featured, representation of Africa and the situation in Ebola-stricken countries. On the representation of Africa, the issues highlighted were focused on: the world being at risk of contracting the Ebola virus from Africa, local people making a difference in the fight against Ebola, the blame on African cultural practices and living conditions for the Ebola outbreak and the optimism of China Daily that Africa remains a continent of vast investment opportunities despite the outbreak. The study advances constructive journalism as the empowering approach in the representation of Ebola and Africa. This is due to constructive journalism’s focus on proffering possible solutions and further serving as a contra-narrative that will contribute positively to complex issues. The constructive journalism approach will contribute to effective health communication and help reshape the media discourse on Africa, they concluded. 24 University of Ghana http://ugspace.ug.edu.gh Similarly, Abeysinghe (2016) examined the print media’s narratives on border control in relation to Ebola coverage in the United Kingdom (UK), United States of America (US) and Australia. The study centred on the period from March 2014 to 30th April, 2015. A qualitative content analysis of the three most widely circulated print media of the USA (New York Times, Wall Street Journal, USA Today), UK (Daily Mail, The Sun, Daily Mirror), and Australia (The Australian, Daily Telegraph, Herald Sun) was done for themes and narratives. It was found that there was a difference in representation of the latest outbreak from the earlier narratives on Ebola outbreak in the past in a number of ways. Coverage was dominated by the domestic politics of border control and the theme of fear which resulted in the shift in focus from the issues surrounding the disease in affected countries. The coverage became a frame through which domestic political concerns were represented and fought through especially in the US. Specifically, issues were refocused by the media to concentrate on the domestic politics of the UK, US and Australia, however, providing less coverage on global health debates and effect on West Africa. The key themes found in the narratives centred on fear and contagion, Ebola and the West, the domestic politics of fear, the spread of Ebola, importers of the disease to the West, the depiction of Africa, and depictions of “uncivilized” practices of affected populations. These themes resonate with the findings of Zhang and Matingwina (2016) which also sought to find out dominant themes and representations of Africa. The theme of fear and contagion highlighted the ability of the disease to cross borders and populations through travel. The stories suggested that both the disease and public reaction towards the disease were undermanaged. In Australia, representations of fear revolved around party politics. In the UK and US, representations highlighted the issue as a government problem rather than a focus on the accounts of the public. The UK papers highlighted the absurdity of 25 University of Ghana http://ugspace.ug.edu.gh reactions from public fear. In the US, the representation of public fear was largely towards a critique of the public’s reaction as an overreaction. Furthermore, the domestic politics of border control in the US resulted in newspapers of different political alignments representing the issue in divergent ways. The theme of Ebola and the West revealed that global reaction to the disease, that is, concern and substantial mobilization around the event failed to materialise until the disease entered the West. Issues in the affected countries were underemphasized with only general discussions of Africa and Africans appearing, contrasted with specific and prolonged examination of cases in the West. The overall newspaper narrative portrayed Ebola as a Western problem; the ability of the disease to spread to the West. The narrative of concentration on the West further heightened when global attention to the disease rose, after the disease’s spread to the West. After cases were identified in the West, the West became the centre of the media narratives and the centre of representations of the disease. The theme of domestic politics of fear involved a vast number of reports on politics and political actions. Ebola was presented as a cause for reasonable concern, and domestic political actors harnessed the public fear as part of general political manoeuvring. Western media coverage of the outbreak had displaced the disease as a problem of the affected West African communities and rather focused on the impact in the West. The acknowledgement of the impact of the disease on West Africa was absent in most of the reports except in generalized ways. Essentially, even though the disease originated in West Africa, newspaper narration and representation focused on domestic issues of the US, UK and Australia. The theme of the spread of Ebola dwelled on the general representations of West Africa. These representations were mainly on the cultural factors of Africa that aided the spread of the epidemic in the region. The depictions of the agents of the spread of the disease also mirrored 26 University of Ghana http://ugspace.ug.edu.gh more domestic concerns. In the US, the theme centred on traditional domestic concerns such as fears around terrorism, issues surrounding race, and immigration as modes of spread. These narratives on the disease spreading to the West also dominated the findings of Zhang and Matingwina (2016) which was centred on narratives that suggest that the West is at risk of getting the disease from Africa Importers of the disease to the West was one of the themes of the narrative. It concerned issues of immigration and security pertaining to the spread of the epidemic to the UK, US and Australia. There were concerns about the spread of the disease through bioterrorism, hence the narrative around border control. In the depiction of Africa, the media represented the affected region in generalized terms and focused on particular cultural aspects of Ebola transmission. There was the depiction of the jungle and West Africans through reference to the spread of the virus through bush meat consumption, which was suggested as a mode of transmission to the West. Some cultural practices of affected West African populations were presented as aggravating the spread of the disease. Intimate contact with family through caring for the sick and preparation of bodies for burial were examples of these practices. The theme of depicting some practices of Africans as uncivilized was used as the point of “othering” (blaming others). The reports suggested that Ebola outbreaks will not be sustained in the West because such cultural factors do not exist. The narrations centred on the description of Africa’s jungle rather than contemporary urban space. The study concluded that the generalised depiction of Africa and African culture served to reinforce rather than negate the general tendency of the media’s reports on Africa. Another study on the Ebola outbreak was by Yusuf, Yahaya and Qabli (2015). Yusuf et al. (2015), summarised the comments of Africans and non-Africans about media reportage of the 27 University of Ghana http://ugspace.ug.edu.gh Ebola epidemic in West Africa. They gathered the comments from BBC, CNN, Metro News UK and The Lancet. They observed that the prolific and unbalanced nature of coverage of Ebola and broadcast by some media had thrown the world into unnecessary fear. Consistent with Smith and Smith (2016), they found that local media, notably television and radio, had raised awareness about the disease among the local people and offered some preventive measures. Information about the disease was carried by government (state) owned and private media through broadcasting programmes by professional medical and public health experts about the true causes, symptoms, modes of spread, and prevention and control measures. Yusuf et al. (2015), however, observed that the spread of misinformation from non- professionals led to the harsh treatment of Africans in the diaspora or on transit through their countries. This is corroborated by MacIntyre and Travaglia (2015) when they stated that emergency health situations like the Ebola epidemic affords people and non-experts the opportunity to present themselves as experts and give out information on the internet, which lacks oversight, that may entirely not be true. Biased and unbalanced coverage of the epidemic in the US also contributed in aggravating the emotional climate of fear They concluded that the media failed to provide enough awareness on how to prevent the outbreak in the first place, just like in the case of Paul et al. (2016). Since this study only examined comments on news articles, it is relevant to the current study which examines news coverage of the Ebola epidemic to find out how the media reported the epidemic which led to the creation of an atmosphere of fear and panic. In another study of international media coverage of Ebola, MacIntyre and Travaglia (2015) explored the effect of heightened vulnerability and reduced oversight on some of the less acknowledged aspects of ethics on the internet in the West African Ebola epidemic, and the need for a global accountability framework for medical ethics in social media. 28 University of Ghana http://ugspace.ug.edu.gh The ethical issues they identified were breach of patient’s privacy through public posting of names, explicit photographs and details about Ebola patients on the internet and posing with patients and survivors of the disease to take pictures for organisational promotions by some foreigners. They observed that whilst publications on medical issues are regulated, the internet, blogs, and social media are unregulated. Though unregulated, they served as critical sources for rapid information on the Ebola disease. To them, “it is unethical for public advice on medical practice be given by those without relevant expertise” (p.67). Since MacIntyre and Travaglia (2015) used 2 media, is it fair to attribute lack of ethics exclusively to their geo context and ideological bias? A comparison of CNN (Western), with GNN (local) and Daily Nation (African) would help answer research question three which sought to find out the ethical issues that emerged in the coverage of Ebola by these media outlets. 3.2 Western media coverage of Ebola and related health epidemic outbreaks Coverage of the West African Ebola outbreak by the Western media has also engaged the attention of scholars. A study by Basch, Basch, and Redlener (2014) examined the nature and extent of coverage about Ebola in the USA Today, The New York Times and The Wall Street Journal newspapers between 17th September, 2014 and 17th October, 2014. The study sought to know what aspects of the disease were communicated through the media since there was limited research on what aspects had been reported in the media. It was found that two major themes were dominant in these news items: coverage on Ebola cases in the US (39%, 117 news stories) and Ebola outbreak in Africa (33.6%,101 stories). The study also found an emphasis on the death tolls and cases in the USA because people were concerned about being infected with the Ebola virus disease. 29 University of Ghana http://ugspace.ug.edu.gh Basch et al (2014) concluded that confidence in the government’s ability to handle Ebola had reduced but coverage about the disease in popular circulated newspapers could help build public support for investment in disaster preparedness. They recommended that the media could do more to educate the public by reporting on how to prevent transmission of the disease from one person to another. In another study, Afzal et al., (2015) examined daily Ebola-related internet search and Twitter data in the United States during a six-week period ending Oct 31, 2014 to determine if news coverage was a significant factor in temporal patterns in Ebola related internet and Twitter data. It was found that each Ebola-related news video inspired tens of thousands of Ebola-related tweets and internet searches. Between 65 to 76 percent of the variance in all samples used in the study is described by the news media contagion model. This study goes to affirm again that media coverage could influence people’s attitudes towards things that are happening around them, and in this case Ebola. A similar study of the western media saw Shih, Wijaya and Brossard (2008) examine on how print media (New York Times) framed public health epidemics. Though not directly focused on the subject of Ebola, the study has conceptual/theoretical relevance for this study because it attempts to build an integrated theory of print news coverage of epidemic hazards by comparing media coverage of three health-related issues: avian flu, mad cow disease, and West Nile virus. It also sought to understand whether there were differences in media coverage of these epidemic diseases. They note that studies had shown that individuals used media coverage as a cognitive shortcut to make sense of complex issues. Looking for similarities or differences across diseases coverage for dominant themes, therefore, could help in understanding how people perceived these issues and the possible decision they might make. Also, similarity or difference 30 University of Ghana http://ugspace.ug.edu.gh in news coverage of these three diseases might provide information about the dynamics surrounding media coverage of epidemic hazards in general. Media framing and issue attention cycle were used as theoretical bases for the comparative analysis in the study. It focused on the analysis of the media’s use of dramatic, conflict, uncertainty, reassurance, consequence, action, and new evidence frames. The study found that “action” and “consequence” were the dominant frames used by journalists in the coverage on the three epidemic hazards. The consistency of these frames in the coverage across the three diseases suggests that journalists tend to use the same themes for stories of similar nature. It further suggests that journalists concentrate their attention on substantive aspects of epidemic hazards. Shih et al. (2008) note that any differences in the result of the study could be attributed to differences in diseases rather than differences in journalistic practices between different media outlets. Since the current study was analysing news from three different media outlets, it offers an opportunity to ascertain if geographical proximity and ideological orientation of media practices will make any difference in what Shih et al. (2008) assume therefore to be universal journalistic norms. Analyses for similarity and difference is also important because Shoemaker and Reese (1996) suggest that differences in news routines, organisational cultures, ideology, influence of outside sources and influence from individual media worker’s personal and professional backgrounds could affect the gathering and production of news which may result in differences in how the same news is presented by different media. Abeysinghe and White (2010) also studied media coverage and government documents through a textual analysis in Australia to uncover the social representations surrounding avian influenza implicit in these accounts. Durkheim’s concept of social representation underpinned the study. Narrative analysis was used to uncover the discourses implicit in the media and 31 University of Ghana http://ugspace.ug.edu.gh government documents and provide insight into broader socio-cultural notions of illness, infection and risk. It was found that both the media and government produced narratives situating the avian influenza through a collective memory of the catastrophic Spanish Influenza pandemic of 1918. The media emphasized threat, risk, and the calls by people for scientific and antiviral based solutions. Government sources were circumspect in their statements and did not amplify either risk nor solutions which cannot be delivered. The current study sought to study the representations (through frames) of Ebola in the media considering the fact that some studies (Yusuf et al., 2015) had shown that the way Ebola was represented elsewhere in the media, impacted on reactions of people towards it. Washer (2004) examined how the UK media framed the SARS disease which emerged in 2003. Even though this study was not about Ebola specifically, both SARS and Ebola share the common attributes of being epidemics. On the other hand, the fact that the SARS outbreak was first reported in Asia (China) and not Africa, should provide a basis for discovering the presence and peculiarity of stereotypes in the treatments of Africa in the coverage of the Ebola epidemic, if any. The study examined how the initial reporting of SARS was done, compared with other new infectious diseases that emerged in the past like HIV/AIDS and Ebola. Newspaper editions of Sunday, 16th March 2003 to Sunday, 13th April, 2003: the first four weeks of the outbreak were used. The sample included articles that mentioned SARS in the Sunday UK national newspapers only, that is, Independent on Sunday, Mail on Sunday, News of the World, Observer, People, Sunday Express, Sunday Mirror, Sunday Telegraph and Sunday Times. In all, 74 articles were identified and analysed together with all editorials and articles published on SARS in the British Medical Journal and The Lancet within the same period. 32 University of Ghana http://ugspace.ug.edu.gh Washer (2004) found that the SARS epidemic was presented as a dangerous threat to the UK. The initial reports presented the outbreak as a “killer” and “mysterious” virus that was spreading very fast. There were speculations about the origin of the disease in the reports by the second week of coverage with some papers suggesting it mutated from horses and pigs and this was compared with AIDS supposedly mutating from monkeys. Stories also centred on the fear of the outbreak becoming the next plague or the next influenza pandemic, like the Black Death, AIDS and Spanish Flu. Attention was focused in The Lancet, on western authorities (Western doctors and other experts) who were investigating the disease in collaboration with WHO and Chinese authorities. Also, heavily covered was how Western doctors (including British) and authorities were using their know-how to try and contain the disease. Similar to Abeysinghe’s (2016) finding that Africa did not receive specific attention but general mention in the coverage on Ebola, China, the centre of the outbreak of the SARS disease and Chinese authorities were referred to only in reference to Western scientists helping them in surveillance on the disease. The third week of the disease saw the shift in focus to reassurance of the potential of Western scientific biomedicine to successfully curb the outbreak. There was also a lot of attention on local Chinese corruption and inefficiency, and the effect of SARS on the local Chinese economy and the international economy. The narratives related to the cause and spread of the disease that became dominant in the reportage overtime were spitting, pollution, dirt, living close to animals, and comparison to previous epidemics. These related to the blame of some cultural and living conditions of Africans in the Ebola epidemic as causes and spread of the disease as found by Zhang and Matingwina (2016) and Abeysinghe (2016). The Lancet and British Medical Journal, however, focused on the search for the cause of SARS and how it was discovered by the Chinese 33 University of Ghana http://ugspace.ug.edu.gh authorities, similar to the dominance of causes of the Ebola epidemic in the coverage by BBC and Chinese Daily in Zhang and Matingwina’s (2016) study. There was focus on the disease’s mode of spread; that is, airborne, by droplet or by sewage system. Consistent with Zhang and Matingwina (2016) and Abeysinghe (2016), the technique of “othering” (we are not like them) was used to calm the fears of the populace that the threat was contained. This was done by placing responsibility for emergence and spread of the disease on the lifestyle of the Chinese. Comparing SARS reportage to previous infectious diseases, the SARS scare was found to have resonated with the attributions made to other infectious diseases throughout history. Unlike in the case of AIDS and Ebola which received little attention when they broke out, it was found that the speed at which the media unfolded the SARS episode, led to an immediate response from the international community. Consistent with coverage of previous epidemics was also the presentation of the disease as a threat to the human race on one hand, and immediately de- emphasizing the threat and giving reassurance of its containment on the other hand. Description of local Chinese in the disease outbreak region as passive, as was done in previous coverage on Ebola, was missing in the reportage on SARS because the Chinese in the SARS epidemic, were totally invisible in the reports. This phenomenon of backgrounding the people in the outbreak region and concentrating on the few Western cases and western authorities, played out in the coverage of the West African Ebola epidemic as found in studies by Zhang and Matingwina (2016) and Abeysinghe (2016). The Chinese at the centre of the outbreak were divided, with Hong Kong being regarded as Western while mainland China was regarded as “other”. Just like Africa and Africans were portrayed as disaster ridden in the coverage of the early outbreaks of Ebola, China and Chinese were also represented as a breeding ground for new infectious diseases. Consistent with previous coverage on infectious diseases, Western 34 University of Ghana http://ugspace.ug.edu.gh biomedicine was presented as the saviour in the SARS situation. It was concluded that the social representation of SARS resonated with past representations of infectious diseases throughout history. The study of SARS indicated that the representation was similar to how infectious diseases had been represented throughout history. The current Ebola study would enable a determination of whether there were some similarities in the reportage of the current Ebola outbreak and SAR as was compared in the UK study. In another study on the coverage public health epidemics, Cortiñas-Rovira, Pont-Sorribes & Alonso-Marcos (2015) content analysed the front pages and editorials of the El Pais, El Mundo, La Vangaurdia, El Periodico and Avui Spanish newspapers, to explore their handling of the swine flu outbreak of 2009 for a period of eight (8) weeks. The objective of the study was to find out the role played by the Spanish daily newspapers during the most critical period of the swine flu outbreak. The study found that El Mundo had the most ethical breaches on its front pages followed by Avui and El País, with La Vanguardia and El Periódico scoring better. The five newspapers were generally scaremongering and often even irresponsible in their front-page handling of the swine flu outbreak yet were prudent and responsible in editorials. Choices of words and images (scaremongering and sensationalist images) used in stories did not accurately reflect the reality. Editorials for all the newspapers in the sample however revealed better adherence to journalistic ethical principle. Differences were small, but according to the established parameters and criteria, the most ethical editorials were those of El Mundo, with El País and El Periódico showing the poorest scores. Also, front pages of papers that most infringed ethical codes were the newspapers with the most rigorous and objective editorials. 35 University of Ghana http://ugspace.ug.edu.gh Finally, on the Western media, Young et al. (2008), on the assumption that in an age of increasing globalization and discussion of the possibility of global pandemics, increasing rates of reporting of these events may influence public perception of risk, investigated the impact of high levels of media reporting on the perception of a frequently reported disease (avian flu) and an infrequently reported disease (yellow fever). A survey and experiments of McMaster University students were used to undertake the study. The results demonstrated that individuals considered infectious diseases that received a lot of media attention to be more severe and had higher disease-like status than diseases of comparable objective severity that received less media attention. They concluded that individuals would consider high media frequency disorders to be more serious and pose more of a threat than equally serious underrepresented infectious diseases. 3.3 African media coverage of Ebola and related health epidemic outbreaks African media coverage of the West African Ebola outbreak has also attracted some scholarship. Smith and Smith (2016) examined the nature and extent of coverage of Ebola in four widely circulated Nigerian newspapers: The Sun, The Vanguard, The Nation and The Punch, bringing out lessons learnt from Nigeria. The objectives of the study were to assess the role of the media in the reportage about the disease and the implication of the reportage in creating awareness and stopping the spread of the disease among the populace. A content analysis was done of the newspaper editions between the period of 20th July, 2014 (when the index case was reported in Nigeria) and 20th October, 2014, (when Nigeria was declared free of Ebola). A total of 1625 articles were analysed; made up of 69.4% (1127) straight news, 16.4% (267) features, 4.7% (76) opinions, 9.2% (149) editorials and 0.4% (16) interviews. 36 University of Ghana http://ugspace.ug.edu.gh It was found that August 2014 recorded the largest number of Ebola reports while coverage dropped in October 2014. Prior to the incidence of the disease in Nigeria, however, media reportage on the disease was done by only 0.1% of Nigerian newspapers. The common themes or frames identified in coverage on the Ebola Virus Disease were Ebola cases in Nigeria, which was consistent with Paul et al. (2016) finding of coverage of the epidemic in the Indian press, discrimination due to Ebola, and cure of Ebola through the use of salt water and kola. They also found that the media played a great role in curbing the spread of the disease by putting out information related to prevention. This is contrary to Paul et al.’s (2016) finding that the Indian media did less in sensitizing policy makers on prevention measures. Smith and Smith (2016), however, suggested that the media needed to do continuous and investigative reportage on disease outbreaks in order to help inform, educate and prepare the populace about infectious diseases and emerging public health issues. They also indicated that the media needed to use more expert sources so as to better educate the general public about infectious diseases and emerging health issues. They recommended that there should be continual coverage on the Ebola disease, for effective awareness, prevention and control of the disease despite the fact that the country had been able to contain the first outbreak of the disease in the country. In a related study, Nwanne (2014) conducted a content analysis of newspapers in Nigeria, using a random sample of news from the Sunday Vangaurd, Sunday Express, and Sunday Sun newspapers. The Agenda setting theory was used to underpin the study in order to meet the objective of determining the salience the international and local media put on the Ebola epidemic reportage. provided data for the study. 37 University of Ghana http://ugspace.ug.edu.gh He found that the media, both local (Nigerian) and international had discharged their surveillance function to the society by identifying dangers of Ebola and providing education about the disease, consistent with Yusuf et al. (2015) on the role of local media in the Ebola outbreak. Nwanne found that the media gave a lot of space in the print media for the discussion of the Ebola epidemic. Most of the stories on Ebola received lead treatment in newspapers and magazines. The reports centred particularly on the Ebola virus disease and what the Nigerian government did to tackle the disease. The local media in Nigeria, particularly Delta Broadcasting Service, were found to have translated all materials on the Ebola virus into local languages and dialects to ensure a wider reach of the message. Again, on the Nigerian media, Obukoadata and Abuah (2014) examined the surveillance function of the media within the context of the Ebola outbreak in Nigeria. They argued that the media were to be the drivers of surveillance, and with the added platform of social media, the media were supposed to lead in campaigns on disease prevention and control. Surveillance campaigns by the media on disease prevention and controls in Nigeria were evaluated to ascertain its extent of influence. The theoretical framework for the study was made up of diffusion of innovation, agenda setting and Lasswell’s communication model (functional paradigm of communication). While the preponderance of studies used content analysis, this study employed a survey of 600 respondents with a 32-item structured questionnaire, to evaluate what influence surveillance campaigns had on disease control and prevention in Nigeria during the Ebola outbreak. The study found a three-point influence of surveillance campaigns by the media on the society. They identified that: the surveillance function could result in negative consequences such as 38 University of Ghana http://ugspace.ug.edu.gh panic, if not well managed; the surveillance function could provide sufficient support for disease management as seen in the Ebola Virus Disease outbreak in Nigeria if properly managed and; an effective and well-thought-out surveillance campaign will correlate positively in controlling and preventing diseases such as Ebola. They concluded that the media in Nigeria lived up to their surveillance function responsibility by providing information on the disease to the public, consistent with Nwanne’s (2014) finding of the local Nigeria media’s provision of education about the disease to the public. They recommended that information on disease control and prevention must be strategically designed by the media and not left at the mercy of citizen journalists or purveyors of user generated content, since it could result in a discordance of information that would bring panic to the society. 3.4 Asian media coverage of Ebola and related health epidemic outbreaks In the Asian media context, Paul, Mahajan, and Sahoo (2016) examined how three major Indian English print media covered the outbreak of the 2014 Ebola. The study was carried out through a cross sectional content analysis of three newspapers spanning study 10 months from June 2014 to March 2015. A total of 967 news items comprising The Hindu news articles (443), The Times of India news articles (390) and The Telegraph news articles (134) were analysed. The reiterative process was used to understand the import of the articles. Unlike Zhang and Matingwina (2016), who found the dominant issues of BBC and Daily Nation’s coverage to be centred on the cause of the disease, the people featured, representation of Africa and situation in Ebola stricken countries, it was found that concerns about Ebola, especially articles about fear, caseload and death toll were the most dominant and major focus of stories (30.3%) followed by Ebola related activities, that is activities to curb the disease (28.9%) and Ebola as a disease, that is, facts about the disease (26.9%). There were only a few 39 University of Ghana http://ugspace.ug.edu.gh articles that sought to sensitize policy makers about the appropriate actions to take against the disease. Nearly half of Ebola news was from news agencies: AP (12%), AFP (10.1%), Asia News International (9.2%), and Indo-Asian News Service (6.7%). Also, half (50%) of news had their sources properly stated, with 10% of the news from anonymous sources. Most Ebola stories appeared in August and October 2014. Disease incidence, however, showed a rising trend and peaked in November 2014. Paul et al. (2016) observed that since India had no case of the Ebola Virus Disease, news articles should have focused on preventive strategies that citizens should take. Again, for the purpose of prevention and awareness, infographics could have been more useful to readers than mere pictures, but infographics were less utilized. Coverage could also have been used to highlight the preparedness of the Indian health system in anticipation of an outbreak. They recommended that authentic sources needed to be used and richer information on the diseases needed to be provided. In the same Asian context, El-Jardali et al. (2015) explored the role of the media in informing health policies in Lebanon, identifying factors influencing health reporting and investigating the role of evidence in health journalism and the quality of health reporting. A content analysis was conducted on Arabic, English and French newspapers published in Lebanon that tackled health related topics issues between January 2012 and December 2013, to assess the way the media reports on health-related issues and the quality of reporting. Findings showed that there was low quality of health reporting, limited use of evidence in health reporting and that only 39.8% of journalists relied on expert sources in their reportage on health issues. Factors that limited the use of evidence included time constraints, the complexity of 40 University of Ghana http://ugspace.ug.edu.gh research language, competition, lack of specialization in health journalism, limited skills and institutional capacity to access databases and the scarcity of local research. There was also a lack of specialized health journalists and sections dedicated to health in newspapers in Lebanon. They concluded by suggesting some strategies to improve the quality of health reporting. These were specialization of journalists in health reporting; research dissemination through media outlets, academic institutions, journalists’ education and research briefs; and forming a platform as a link between media, policy makers and researchers. 3.5 Chapter Summary This chapter was used to review literature on Ebola, coverage of Ebola, and coverage of other epidemics and infectious diseases which will letter aid the discussion of the current study. The dominant methodology used in most of the studies was content analysis (both quantitative and qualitative) with few studies employing audience survey, textual analysis, experiment and mixed methodology. The studies were mostly national (coverage by media within a particular country) and international (coverage by media from different countries). The studies ranged from a period of four weeks to thirteen months. Framing was the dominant theory in the studies with some studies also using theoretical frameworks comprising issue attention cycle and social representation theories among others. Majority of the studies were on print media while a few could be found on internet and social media. The next chapter would present the methodology that would be used to undertake the study. 41 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR METHODOLOGY 4.0 Introduction This study was inspired by five (5) key objectives, which are to: 1. identify the dominant frames employed by CNN, GNN and Daily Nation in their online framing of the West Africa Ebola epidemic which broke out in December 2013, 2. identify whether Issue-Specific frames differed based on newsmakers across media, 3. examine any ethical issues that might be found in the coverage, and 4. investigate the kinds of media representation Africa received in the coverage related to the disease. This chapter presents and explains the research methodology used in undertaking the study. The chapter specifically discusses the research design, population, sample size, sampling procedure, unit of analysis, data collection approaches and Intercoder reliability method used in the study. 4.1 Research Design The study employed the quantitative content analysis method to find out the dominant frames employed, differences in frames among the media outlets, whether frames differed depending on newsmaker, presence and nature of ethical issues in the coverage, and the kind of representation Africa received in media coverage of the Ebola epidemic. Due to its broad nature, content analysis has generated various definitions among scholars. To Wimmer and Dominic (2011), content analysis involves studying and analysing communication in a systematic, objective and quantitative manner for the purpose of measuring variables. Furthermore, it is used to describe a communication content, test hypotheses of message characteristics, compare media content to real world, assess the image of a particular group of people in the society and to establish a starting point for the study of media effects. Berelson (1952) defined content analysis as a technique used to objectively, systematically and quantitatively describe the manifest content of any communication. To 42 University of Ghana http://ugspace.ug.edu.gh Busch et al. (2005), content analysis involved the objective examination of the presence of certain phrases, themes, concepts, characters, words or any communicative language in a body of text. Content analysis was considered suitable for this study because the method is recommended for the study of media coverage (Wimmer & Dominick, 2011). In addition, as the literature (Chapter 3) showed, Zhang and Matingwina (2016), Paul et al. (2016), Smith and Smith (2016), and Abeysinghe (2016) who all sought to examine the coverage of the Ebola epidemic, with focus on identifying dominant themes and media representations employed the content analysis method studies. Shih et al (2008) who explored the dominant frames in print media coverage of public health epidemics in general, also used content analysis. There are two major ways a content analysis can be approached. These are through the qualitative and quantitative approaches. According to Wimmer and Dominick (2011), the quantitative approach helps provide accurate and precise representation, summary and report of results. Also, the quantitative approach helps the researcher to transform communication content into quantitative statistical data. When approached qualitatively, the analysis goes beyond quantifying data to analysing themes, patterns, and meanings that may be found explicitly or implicitly in a communicative text. It therefore allows “researchers to understand social reality in a subjective but scientific manner” (Zhang & Wildemuth 2009, p.1). Since the study was aimed at finding the extent (breadth) of the issues involved in the coverage, this study adopted the quantitative approach in accordance with the objectives and research questions. Also, Berelson (1952) and Macnamara (2005, p.3) suggested that quantitative content analysis serves purposes such as the description of message content, making inferences to producers of message content, make inferences to audiences of content and to predict the effects of content on audiences. Holsti (1969) suggested that quantitative content analysis helps researchers make inferences about the antecedents of communications, describe and make 43 University of Ghana http://ugspace.ug.edu.gh inferences about the characteristics of communications and make inferences about the consequences of communications. This means that some inferences about how the media covered the Ebola epidemic and how it may affect audiences can be drawn from the findings of this study, using a quantitative content analysis. Since the current study was expected to describe and compare the way the CNN, GNN and Daily Nation covered (framed) the Ebola outbreak, the quantitative approach was adopted. With the importance that online media has garnered as the new source of information for many people on both general news and health information (Cortesi & Gasser, 2015; McQuail, 2010; Lenhart, 2015; Salman & Saad, 2015) on the one hand, and the possibility of ethical breaches and provision of inaccurate information from non-expert sources in disease outbreak situations (MacIntyre and Travaglia, 2015) on the other hand, the online media was selected for the study in order to find out if any ethical issues occurred in the online coverage of Ebola by CNN, GNN and Daily Nation. Wimmer and Dominick (2003) identified ten stages in content analysis, namely: formulation of research questions or hypothesis, definition of population, selection of an appropriate sample from the population, selection and definition of unit of analysis, construction of the categories of content to be analysed, establishment of quantification system, training of coders and pilot study, coding the content according to established definitions, analysis of collected data and conclusion. These steps guided the methodology of the current study. 4.2 Population A population refers to an aggregate or totality of subjects or members that have a set of common characteristics of interest to a researcher (Polit and Hangler, 1999). The population for this study was the entire news coverage on Ebola epidemic by CNN, GNN and Daily Nation from the period of 1st March, 2014 to 31st December, 2014. 44 University of Ghana http://ugspace.ug.edu.gh The period 1st March 2014 to 31st December, 2014 was chosen because this period witnessed a lot of milestones about the disease. First, a pre-test to identify Ebola news showed that media coverage of the outbreak did not start until late March, 2014. More importantly, even though the index case of the disease occurred in December, 2013, it was on 22nd March, 2014 that the Ministry of Health of Guinea officially notified the World Health Organisation (WHO). Furthermore, the West African outbreak was only declared a Public Health Emergency of International Concern (PHEIC) on 8th August 2014 by the World Health Organisation, calling for global attention and support to fight the epidemic. It was also the period that cases of the disease spread into other parts of West Africa, the United Stated of America and Europe, making it a period that media attention was expected to heighten. Again, studies have shown that by November 2014, media reports had started declining hence; the decision to look at coverage up to 31st December, 2014. 4.3 Sampling Technique Krippendorff (2004) defined sampling as the reduction of the whole of available objects of analysis to a manageable corpus which is representative of the whole set (p. 86). He noted that different objects of analysis may require different sampling techniques which are appropriate and applicable to the object being analysed. Composite week sampling, also referred to as constructed week sampling, was used to select the days that represented the months in the study. This was employed because of its acknowledged relative superiority to both the random sample and consecutive day sample when dealing with media content (Wimmer & Dominick, 2011; citing Riffe, Aust & Lacy, 1993). Based on Wimmer and Dominic’s (2011) model, the days of the week were grouped into strata; for example, all Mondays in the month were in one strata, Tuesdays in another and the same for all the other days in the week. The dates of the days in the month were written on pieces of paper and folded. Since the online platform allows for news to be published every 45 University of Ghana http://ugspace.ug.edu.gh day of week and year, all the days, including Saturday, Sunday and holidays were included in the study. Seven days were then simple random sampled, one from each stratum (Sunday to Saturday) to construct the composite week. This technique was used to select a composite week for each of the months. 4.4 Sample Size A sample refers to the subset of a population that is selected for a study and which represents the larger population (Polit et al. 2001). In all, ten composite weeks were produced for the study, with one each representing a month from March to December, 2014. Since there are seven (7) days in a week and ten (10) months in the study, multiplying these two numbers, results in 70 days of news in the sample. Furthermore, since there are three news platforms involved in the study, this translates to 70 days’ publication of news by each of the media platforms making the total number of days of publications to be 210 days, that is, 70 days multiplied by 3 media platforms. In the sampling of the actual stories from the various media however, it was observed that some of the initial dates sampled in the composite week did not have stories. To address this problem, new dates were selected randomly from the remaining days in the stratum, until the days in that stratum were exhausted. If the days in the stratum were exhausted and there were still no stories, the day was labelled as having “No Story”. The composite weeks used for the current study are presented in Appendix III. 4.5 Unit of Analysis Unit of analysis refers to the elements that are examined in a study to enable the researcher make generalisations of all such units and to be able to explain the differences between them (Wimmer and Dominick, 2011). It could refer to a word, symbol, a theme or an entire article or story. This means that a unit of analysis is the smallest but most important element that is tested in a content analysis. The unit of analysis for this study were all the types of editorial 46 University of Ghana http://ugspace.ug.edu.gh content; that is, straight news, features, editorials, opinions and any other kind of news on the West Africa Ebola epidemic published on the days selected through the composite week sampling process described above. 4.6 Data Collection Technique and Instrument Barbie (1992, p.378) stated that content analysis involves reducing a wide variety of peculiar items of information to a limited set of attributes. This is done through a coding operation that is based on the logic of conceptualization and operationalization. Operationalization referred to “the mental process whereby fuzzy and imprecise concepts (notions) are made more specific and precise” (Barbie 2007, p.124). In other words, it refers “to the process through which we specify what we mean when we use particular terms in research” (Barbie 2007, p.124). Using the sampled dates from the composite week sample, news stories on the dates that were selected in the sample were accessed from the websites of the three media houses. The keyword “Ebola” was searched using the search bar on the website of CNN (http://edition.cnn.com), GNN (http://gnnliberia.com), and the Daily Nation (http://www.nation.co.ke). The headline, date and web address of all the stories on Ebola which fell within the period of 1st March 2014 and 31st December, 2014 on the dates selected in the composite week sampling were copied onto a Microsoft Office word document, so that they could be used to search for the stories to code. According to Krippendorff (2004) coding helps to capture the object of study in such a way that other researchers can reliably execute the same process and arrive at the same results. For this reason, the coding must be done in such a way to make it possible to re-search for its patterns. A coding guide with categories and frames informed by literature, objectives of the study, research questions and the theoretical framework, was developed to collect the data for analysis. 47 University of Ghana http://ugspace.ug.edu.gh The categories coded for in the study are operationalized below: Media: This category refers to the media that were studied; that is, CNN, GNN and Daily Nation Date of Story: This refers to the day, month and year in which the news story was published. Story Headline: Refers to the headline of the story being analysed. Type of News Story: Documents the genre or format in which the news story is presented. The types of news stories could be straight news, features, editorial, opinion and other. Straight News Story: Stories on the Ebola epidemic that were presented in an inverted pyramid style. These kinds of stories carry hard facts and answer the who, when, what, why, where and how questions. Features: Long analytical stories on Ebola and related issues that often has a personal slant and written in individual style painting a vivid picture in the reader’s mind. They are sometimes clearly labelled as such by the media that published them. Editorial: They are news stories that express the Editor or Editorial team’s opinion about Ebola. Opinion: They are the stories which express the ideas of people about Ebola and its related issues Other: Stories that cannot be clearly classified under any of the formats of news identified above. Subject matter of the story: refers to the central theme of the story under examination. Risk/Threat – stories that portray and emphasize on the disease ask a risk/threat. 48 University of Ghana http://ugspace.ug.edu.gh Awareness Creation/Education/Action – Stories about education, action, or awareness creation about the disease. Skills Training/Capacity Building – stories on skills training/capacity building of people on the disease. Donation/Fundraising – stories on fundraising and donation towards fighting the disease. Disease Statistics – stories about the new figures, successes, impact of the disease. Other: Stories that cannot be classified under any of the subject matter categories identified above. Source of Story – this explores where the stories were taken from. The sources looked at were Press Release; News Agency (African); News Agency (Non-African); Intermedia (stories from another media which is not an agency; E.g., CNN taking a story from BBC); in-house Journalist/Correspondent Initiated stories from beats, programmes, press conference, seminars etc, Individuals (private individual opinions); Other (stories sourced from other sources that were not listed above), and No Source stated (stories with no source information). Story Enhancement: this examines the kind of multimedia item that accompanied the news stories, since the online medium supports multimedia such as audio, picture and video accompanying the story. The options coded for were: Text only: referred to stories without any enhancement Video with Text: referred to stories that were dominantly video with anchor lead in Picture Stories: referred to photo stories or series of photos with some descriptive text Story with Video: referred to news stories accompanied with video(s) Story with Audio: referred to news stories accompanied with audio file(s) Story with Picture: referred to news stories accompanied with picture(s) 49 University of Ghana http://ugspace.ug.edu.gh Story with Picture and Video: referred to news stories accompanied with both picture(s) and video(s) Infographics: referred to visual images in the form of charts, diagrams, maps or tables used to provide information about Ebola. Story with Infographics: referred to stories that are accompanied with infographics Does the enhancement reflect the story: this was to examine whether the enhancements that come with the story gave more information or help to confirm the information carried by the story. The study set out to find out if the enhancement reflected the story: Fully – whether the enhancement complements, provides more information or supports the story fully. Partially - whether the enhancement complements, provides more information or supports the story partially. No – referred to stories that were accompanied by multimedia that does not relate to the story in anyway. No enhancement – referred to stories that were text only and had no multimedia accompanying them Story Actors (news makers): examined the news makers in the news stories. In other words, it identified the people who provided the news and or on whom the news stories were written. The newsmakers studied for were Victim/Affected (African) – referred to news about Ebola victims and or their families who were African, Victim/Affected (Non-African) – referred to news about Ebola victims and or their families who were non-African, Health Workers – referred to news stories that were on or about health workers such as nurses, doctors and other health care providers, International/Regional Organisations – news about organisations such as UN, ECOWAS, World Bank, and WHO, Pressure/Advocacy Groups and Organisations and 50 University of Ghana http://ugspace.ug.edu.gh NGOs – stories about groups such as Doctors Without Borders, Red Cross, and Samaritan Purse, Government Officials and Agencies – stories on Presidents, Ministers and heads of government agencies, Individual and Expert Opinions – news stories that were on individuals who do not belong to any of the earlier categories mentioned, some of whom are experts on the Ebola Virus Disease (EVD), Religious and Traditional Leaders – news on or by traditional and religious leaders such as chiefs and reverend ministers, and Other – which referred to news that cannot be categorised under any of the above described newsmakers. Episodic and Thematic Framing: this is a category of generic frame used to examine if the news stories were presented as Episodic – story without background information or Thematic – stories situated in a context with background information. Issue-Specific Frames: this examined the major issue-specific frames through which CNN, GNN and Daily Nation narrated Ebola. The frames were based on the framing typology of Shih et al. (2008). The frames studied for were: Action – action(s) against the disease e.g. potential solutions, prevention strategies (quarantines), provision of health interventions/facilities Conflict – differences in opinions, arguments/disagreements, antagonism between opposing opinions e.g. Whether the virus is airborne or not. Consequences – impact of the disease (economic, political, human lives, environmental, social) New Evidence – new findings/results, discovery of new methods to prevent/cure/treat the disease Reassurance – assurances of success against the disease, hope, allaying fears of the public, successes in the fight against the disease/efforts to rid of the disease 51 University of Ghana http://ugspace.ug.edu.gh Uncertainty – uncertainties about the disease (cause, cure, possible spread, portrayal as unknown) Other – stories that cannot be classified under any of the frames stated above Causal Responsibility Framing: this frame is another type of generic frame which sought to identify who or what the media assigned responsibility to as the cause or origin of the Ebola epidemic. The frames were Individual Responsibility, Institutional/Government Responsibility, International Donor Intervention, Global Action and None (where responsibility was not placed on anyone) and Other; that is, where it is not clear where to place the responsibility among the specific categories mentioned above. Treatment Responsibility Framing: just like its causal responsibility counterpart, this frame explored who has been identified in the stories on the Ebola epidemic as responsible for providing solution to the Ebola epidemic outbreak. The frames were Individual responsibility/Attitudinal change, Institutional/Government Action, International Donor Intervention, Global Action, Other, and None, were the options explored under this category. Constructive journalism framing (Media Representation): the purpose of this category was to help identify the kinds of media representations Africa received from the media in relation to the narrations about the Ebola epidemic. This constructive journalism frame was drawn from Zhang and Matingwina (2016) who used it to study how the BBC and China Daily represented Africa in their coverage of the West Africa Ebola outbreak. The codes under the negative news were drawn from literature. News stories will be studied to ascertain whether they are constructive, negative or neutral. Constructive stories reflected the themes of “accomplishments”, “hero” and “meaningful” narrative. These themes use positive psychological techniques and allow for holistic coverage of Ebola stories. The Hero narrative focuses on people who overcame 52 University of Ghana http://ugspace.ug.edu.gh numerous obstacles to change their own circumstances and those of the people around them, Accomplishment narrative highlights a problem and then narrate people’s journey towards finding solutions, and Meaningful narrative interrogates whether lessons were learned from problems which might be life-changing. Negative news referred to news that narrated Ebola through stereotypical images of Africa to depict a sense of hopelessness and focused on issues such as fear. Negative news stories would refer to stories that stressed on the portrayal of Africa as a place of violence, conflict, poverty, disease, and corruption governments and weak systems. Stories on Western superiority, negative cultural practices and portrayal of Africa as a single continent were also deemed negative. Neutral news included stories, which had neither the characteristics of constructive journalism, nor negative news, that is, news such as announcement of events or facts without comments. No Stereotype referred to stories that do not have any of the frames listed above. Ethical issues in the coverage of Ebola: This category examined the news stories to find out if there were any ethical issues. Specifically, the stories were analysed for issues of scare mongering (stories that caused worry, fear and panic), sensationalism (stories that contain facts but the language of presentation was shocking), breach of patient’s privacy (use of patient details: name, pictures, addresses), use of gory images and videos, other (ethical issues that cannot be easily classified under the ones mentioned above) and No ethical issues (stories that have none of the ethical issues stated). Before coding, the researcher and a colleague MPhil student who coded the data, went through the coding guide to agree on the definitions of the various categories in the guide. This was to ensure that the codes and categorisations were understood by both coders to mean the same 53 University of Ghana http://ugspace.ug.edu.gh thing so as to aid the collection of reliable data for the study. The coding was done by copying the web address of the story from the Microsoft Word document into the web browser which then loaded the story for coding. The stories were read and with the help of the coding guide (Appendix 1), the date of story, story headline, type of story, subject matter of story, source of story, story enhancement, reflection of enhancement of story, story actors (newsmakers), generic frames, causal and treatment responsibility frames, issue-specific frames, constructive journalism frame (media representation of Africa) and ethical issues in coverage were coded onto a coding sheet (Appendix II). News stories that just mentioned the disease without saying anything further about it and stories which had less than two paragraphs on the disease were eliminated to ensure there was news about the disease to analyse. Also, stories that turn out to be duplicates from the same media outlet upon a careful reading, were also eliminated to avoid coding the same story more than once. Since the purpose of the study was to compare online framing of Ebola epidemic by the CNN, GNN and Daily Nation quantitatively, the nominal scale of measurement was used. This is because it is the best measurement scale that fits this kind of data since it helps in providing counts or frequency distribution of the data, which is what this study sought to do. The question and accompanying variables as coded are presented in Appendix 1. 4.7 Data Analysis The information gathered and coded unto the coding sheet was entered into the Statistical Package for Social Sciences (SPSS) version 22 computer software for analysis. This software was used to analyse the results of the study. The results for the study are presented in tables, bars, and pie charts. Statistical tests and cross tabulations were done where necessary to effectively respond to the research questions raised in the study. Percentages and frequencies are used to describe the results obtained. Two hypotheses were tested using Chi-square and Cramer’s V values. The Chi-square was used to measure association between categorical 54 University of Ghana http://ugspace.ug.edu.gh variables while the Cramer’s V determined the strength of the association: that is, whether the association was strong or weak. These tests were done using the SPSS software. 4.8 Intercoder Reliability Inter-coder reliability refers “to levels of agreement among independent coders who code the same content using the same coding instrument” (Wimmer & Dominic, 2001, p.171). Intercoder reliability was ensured in this study by an Intercoder reliability test. The researcher and an independent coder who eventually coded the data for the study, conducted a pre-test coding of twenty percent (20%) of all the stories sampled for the study from each of the three media platforms. There were 305 stories sampled from CNN, 135 stories from GNN and 115 from Daily Nation. This means that 61 stories from CNN, 27 stories from GNN and 23 stories from Daily Nation were coded. This was done after both coders had agreed on the conceptualization and operational definitions of the categories and codes in the coding guide used for the study. This pre-test data was entered into http://dfreelon.org/utils/recalfront/recal2/ to test for reliability. The result of the test showed that there was an 88.3% agreement, equivalent of 0.84 of Cohen’s Kappa and 0.84 of Krippendorff's Alpha (nominal) reliability, among coders, which was higher than the average of 70% agreement suggested by Wimmer and Dominick (2006). 4.9 Chapter Summary This chapter presented the research design and the processes followed in the conduct of the study. It was indicated that the study was conducted using a quantitative content analysis to examine news online news of CNN, GNN and Daily Nation. The population, sampling technique, sample, unit of analysis, data collection instrument, method and analysis were all discussed. The study covers the period of 1st March, 2014 to 31st December, 2014 through a constructed/composite week sampling. In all 210 days of data are expected to be analysed. The data will be processed using the SPSS Software to generate statistical data for analysis. The 55 University of Ghana http://ugspace.ug.edu.gh inter-coder reliability method employed in the study was also presented. In the next chapter, the findings/data produced by these comprehensive processes are presented. 56 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE FINDINGS 5.0 Introduction This chapter presents the findings of the study. An objective of the study was to identify the dominant frames used in the coverage of the West African Ebola epidemic by the CNN, GNN and Daily Nation. The study was also designed to: identify whether issue-specific frames differed based on newsmakers across the media studied; examine any ethical issues that might be found in the coverage; and investigate the kinds of media representation Africa received in the coverage related to the disease. The data collected using the coding guide designed to address these objectives, was entered into the Statistical Package for Social Science (SPSS) software, after which descriptive results in the form of graphs, tables and figures were generated. The findings from this process are presented in this chapter. Cross tabulations and some statistical tests were also carried out to effectively address some of the research questions for the study. 5.1 Distribution of Ebola News Coverage The distribution of Ebola news coverage gives an overview of the coverage about the epidemic during the period of study. A total of 2,830 news stories on Ebola were accessed from the websites of the CNN, GNN and Daily Nation covering the period of 1st March, 2014 through 31st December, 2014. Close to two-thirds (65%) of the stories were sampled from CNN. GNN had 15% stories in the analysis and 20% of the news was sampled from Daily Nation. This shows a proportionally wide gap between stories published by the CNN and the two other media studied. 57 University of Ghana http://ugspace.ug.edu.gh FIGURE 1: Number of Ebola Stories Analysed 135 Frequency 115 305 0 50 100 150 200 250 300 350 Global News Network (GNN) Daily Nation Cable News Network (CNN) Through composite week sampling and the elimination of news items that only mentioned “Ebola” but did not say anything else about the epidemic that could be analysed, 555 stories were coded and analysed. From Figure 1, it is observed that more than half, that is, 55% (305) were from CNN, close to a quarter (24.6%) were from GNN and one fifth (20.7%; which translates to 115 stories) were selected from the Daily Nation. This indicates that CNN reported more stories on the epidemic than both GNN and Daily Nation which are African media outlets. It is understandable that based on proximity, GNN reported more news on the epidemic than the Daily Nation, since it is located within Liberia, the country most hit by the disease with the second highest number of cases (10,675) but with the highest number of deaths (4,809) (WHO, 2016a). TABLE 1: Monthly Distribution of Ebola News MONTHLY COVERAGE OF EBOLA MONTH Frequency Percent March 21 3.8 April 21 3.8 May 5 .9 June 13 2.3 July 43 7.7 August 85 15.3 September 83 15.0 October 202 36.4 November 55 9.9 December 27 4.9 Total 555 100.0 58 University of Ghana http://ugspace.ug.edu.gh The monthly distribution of Ebola news coverage (Table 1) indicated that October 2014, had the most coverage of news during the study period. It recorded over one third (36.4%) or 202 of the total number of stories analysed. August and September had the second significant number of stories. They each had close to one sixth (15%) of stories, that is, 85 and 83, stories respectively. Following these with less than 10 per cent of stories were November (9.9%), and July (7.7%). The least number of studies came in the month of May, where only 0.9% of stories were covered. TABLE 2: Comparing Monthly Distribution of News across Media KEY Most News Second Highest Least News MONTHLY DISTRIBUTION OF STORIES MEDIA MONTH Total CNN n (%) GNN n (%) Daily Nation n (%) March 7 (2.3%) 4(3.0%) 10(8.7%) 21(3.8%) April 11(3.6%) 5(3.7%) 5(4.3%) 21(3.8%) May 0(0.0%) 2(1.5%) 3(2.6%) 5(.9%) June 3(1.0%) 4(3.0%) 6(5.2%) 13(2.3%) July 19(6.2%) 9(6.7%) 15(13.0%) 43(7.7%) August 31(10.2%) 29(21.5%) 25(21.7%) 85(15.3%) September 34(11.1%) 30(22.2%) 19(16.5%) 83(15.0%) October 169(55.4%) 18(13.3%) 15(13.0%) 202(36.4%) November 21(6.9%) 24(17.8%) 10(8.7%) 55(9.9%) December 10(3.3%) 10(7.4%) 7(6.1%) 27(4.9%) TOTAL 305(100.0%) 135(100.0%) 115(100.0%) 555(100.0%) Comparing overall coverage among the media outlets under study, CNN recorded the highest coverage of news in six (April, July, August, September, October and December) of the ten months under study, as presented in Table 2. In addition to December (3.3%), where it recorded the same figure as CNN, GNN had the highest coverage of news among the three media outlets in November (17.8%). Daily Nation dominated Ebola news coverage in March (8.7%), May 59 University of Ghana http://ugspace.ug.edu.gh (2.6%) and June (5.2%). Overall, the coverage seemed to have started gradually in March, peaked in October and declined by the end of the December 2014. This was contrary to Smith and Smith (2016) who found in their study of the West Africa Ebola outbreak in four widely circulated Nigerian newspapers that most stories appeared in August 2014 and coverage declined in October 2014. It was consistent with Paul et al. (2016) who also studied the coverage of Ebola by three major Indian print media and found that most stories were covered in August and October 2014. It was also found that the diagnosis of the first case of Ebola in the United States, the case of the Liberian Thomas Eric Duncan, on 30th September, 2014, intensified media coverage of the disease. This may have contributed to October recording the highest number of stories (169), that is 55.4% of CNN’s overall coverage on the epidemic. Incidentally, the month of October also saw the diagnosis of three other cases in the USA. This is similar to Smith and Smith’s (2016) finding that Ebola stories increased in Nigeria after the index case was reported in Nigeria in July, 2014. Whereas proximity may be used to explain why GNN covered more news than Daily Nation, it is unclear why CNN covered the epidemic more than GNN which is at the heart of the epidemic, considering the fact that the online platform offers a 24-hour cycle of news and therefore makes it possible for as unlimited number of news to be posted within a day. 5.2 Type of Articles Type of articles gives an indication of how the news is presented. There are various ways that news can be presented, but in this study, these were analysed in terms of straight news, editorial, feature and opinion. 60 University of Ghana http://ugspace.ug.edu.gh FIGURE 2: Type of Articles 450 424 400 350 300 250 200 150 102 100 50 272 0 Straight News Feature Editorial Opinion Frequency The results showed (Figure 2) that more than three forth (76.4%); that is, 424 of the articles analysed on Ebola were straight news; a little over one sixth (18.4%) or 102 were features; and close to five percent (4.9%) or 27 were opinion pieces. There were only two (0.4%) editorials in the 555 articles analysed. The dominance of straight news in the coverage was consistent with the findings of Paul et al. (2016) who also studied the coverage of Ebola in three major Indian English print media. Straight news usually presents fact and less of the reporter’s interpretation and evaluation; which gives grounds for the conclusion that the media outlets preferred to cover the Ebola epidemic in hard news format. TABLE 3: Comparing Types of Articles Across Media KEY Most News Second Highest Least News MEDIA Type of News Total CNN n (%) GNN n (%) Daily Nation n (%) Straight News 191 130 103 424 62.6% 96.3% 89.6% 76.4% Feature 99 0 3 102 32.5% 0.0% 2.6% 18.4% Editorial 0 0 2 2 0.0% 0.0% 1.7% .4% Opinion 15 5 7 27 4.9% 3.7% 6.1% 4.9% TOTAL 305 135 115 555 100.0% 100.0% 100.0% 100.0% 61 University of Ghana http://ugspace.ug.edu.gh Comparing the types of news in the three media outlets as represented in Table 3, the result shows that almost all (96.3%) or 130 of GNN’s stories were straight news compared to 89.6% by Daily Nation and CNN’s 191 (62.6%) straight news stories. Close to one third (32.5%) or 99 of CNN’s overall stories were feature stories compared with three (2.6%) by Daily Nation and no story by GNN. CNN published more opinion pieces (15) that is, 4.9% of its overall stories in the study, followed by Daily Nation’s 6.1% and GNN’s five (3.7%) stories. The only two (0.4%) editorials found in the overall stories analysed were from Daily Nation which represented 1.7% of Daily Nation’s overall coverage. The opinion stories of CNN came from disease experts who were sharing their views on facts about the disease and the ways to tackle the epidemic. 5.3 Subject Matter of Stories The subject matter of the story refers to the central theme that the news story revolves around. In other words, the major idea carried by the story. FIGURE 3: Subject Matter of Stories Frequency Other 56 Disease Statistics 238 Donation / Funding 13 Skills Training / Capacity Building 5 Awareness Creation / Education / Action 176 Risk / Threat 67 0 50 100 150 200 250 The study found, as presented in Figure 3 that close to half (42.9%), representing 238 stories on Ebola were centred on disease statistics. About a third (31.7%) or 176 news stories of the total news analysed were focused on awareness creation, education and action to curb the epidemic. Reports on the epidemic with emphasis on its threat to humans and the risk 62 University of Ghana http://ugspace.ug.edu.gh associated with it was the next main subject matter of 12.1% (67) of the stories in the study. The “other” category of the subject matter dominated 10.1% (56) of all the stories analysed. This category of news centred on stories that did not have Ebola as the main focus of attention. A story like “CAF: Morocco will not host Nations Cup” came under this category of news. Stories on donation and funding (2.3%), and Skills training and capacity building (0.9%) were the least covered in the study. TABLE 4: Comparing Media and Subject Matter KEY Most News Second Highest Least News MEDIA SUBJECT MATTER Total CNN GNN Daily Nation Risk / Threat 42 5 20 67 13.8% 3.7% 17.4% 12.1% Awareness Creation / 84 67 25 176 Education / Action 27.5% 49.6% 21.7% 31.7% Skills Training / 2 0 3 5 Capacity Building .7% 0.0% 2.6% .9% Donation / Funding 2 9 2 13 .7% 6.7% 1.7% 2.3% Disease Statistics 140 40 58 238 45.9% 29.6% 50.4% 42.9% Other 35 14 7 56 11.5% 10.4% 6.1% 10.1% TOTAL 305 135 115 555 100.0% 100.0% 100.0% 100.0% Comparing the subject matter and media outlets as shown in Table 4, it was found that close to half (45.9%) or 140 of CNN’s stories were on Disease Statistics, similar to Daily Nation, which had half (50.4%) that is 58 stories, on disease statistics. Going by the stipulations of Paul et al. (2016) the Daily Nation, located in Kenya (outside outbreak) should have focused more attention on awareness creation/education/action which would have highlighted preventive strategies. On the other hand, the awareness creation/education/action theme was the dominant subject matter in GNN’s coverage, which was about half (49.6%), translating to 67 stories. Disease statistics was the second dominant subject matter in GNN’s coverage, with about thirty 63 University of Ghana http://ugspace.ug.edu.gh percent (29.6%) of stories, while the second dominant coverage of CNN and Daily Nation were (27.5%) and (21.7%) respectively on awareness creation, education and action about the disease. Risk and threat (13.2%) and “other” (11.5%), were the two other themes of significant attention in the coverage of CNN. Similarly, Daily Nation also had risk and threat stories (17.4%) and “Other” (6.1%) as the subject matter most covered behind disease statistics and awareness creation, education and action. In the case of GNN, “Other” (10.4%), and risk and threat were the 3rd and 4th important subject matter in the coverage. Donation and Funding featured more in GNN’s stories (6.7%) than Daily Nation (1.7%) and CNN (0.2%). Daily Nation (2.6%) and CNN (0.7%) were the only ones which had news on skills training and capacity building. 5.4 Source of Stories The source of stories was aimed at identifying the person who gathered and put the information together to produce the news stories that were analysed in the study. Source of stories among other things, helps lend credibility to news stories and sometimes helps people guess the motive behind a piece of information. The sources’ inclinations and biases may also determine how the news is framed. The study sought to identify whether the news was originally from the media outlet being studied, intermedia sources, news agencies or any other sources that could be identified. Figure 4: Sources of Stories Sources of Stories 1 No Source 85 27 Journalist Initiated 297 37 News Agency - NON AFRICAN 84 18 Press Release 6 0 50 100 150 200 250 300 350 Frequency 64 University of Ghana http://ugspace.ug.edu.gh It was found, as presented in Figure 4 that more than half (53%), which is 297 stories, were journalist initiated. Journalist initiated means the story was done by a journalist of the media outlet being studied. Stories with no cited sources and News Agency (non-African) were second and third major sources of news with 85 stories (15.3%) and 84 stories (15.1%) respectively. All the other sources recorded lower than ten percent of stories each. They were Intermedia sources, 37 stories (6.7%), News Agency (African), 18 stories (3.2%) and Press Releases, six stories (1.1%). There was one story (0.2%) in the “Other” category which was sourced from Facebook. This finding is consistent with Paul et al. (2016) who also found in their study that most of the stories had their sources stated. TABLE 5: Comparing Media and Source of News KEY Most News Second Highest Least News MEDIA SOURCE OF NEWS Total CNN GNN Daily Nation Press Release 0(0.0%) 6(4.4%) 0(0.0%) 6(1.1%) News Agency (African) 4(1.3%) 13(9.6%) 1(.9%) 18(3.2%) News Agency (non-African) 2(.7%) 9(6.7%) 73(63.5%) 84(15.1%) Intermedia Sources 21(6.9%) 15(11.1%) 1(.9%) 37(6.7%) Journalist Initiated 262(85.9%) 8(5.9%) 27(23.5%) 297(53.5%) Individuals 13(4.3%) 8(5.9%) 6(5.2%) 27(4.9%) No Source 3(1.0%) 75(55.6%) 7(6.1%) 85(15.3%) Other 0(0.0%) 1(.7%) 0(0.0%) 1(.2%) TOTAL 305(100.0%) 135(100.0%) 115(100.0%) 555(100.0%) A comparative analysis of the media and sources of the stories in presented in Table 5 showed that whereas CNN had majority (85.9%) amounting to 262 stories coming from their journalists, more than half (55.6%) of GNN’s total stories, which is 75, had no source stated and close to two thirds (63.5%) or 73 stories out of 115 stories by Daily Nation were from a non-African News Agency; Agence France-Presse (AFP). The other sources of news for CNN 65 University of Ghana http://ugspace.ug.edu.gh were Intermedia sources (6.9%), Individuals (4.3%), News Agency (African) (1.3%), and News Agency (non-African) two stories (0.7%). There were three (1.0%) CNN stories without sources. The rest of Daily Nation’s stories originated from the media’s own journalists (journalist initiated) 27 (23.5%), no source (6.1%), individuals (5.2%), with intermedia and News Agency African producing one or 0.9% each of stories. GNN was the only media to have had news from all the sources identified in the study. The rest of their stories came from Intermedia Sources (11.1%) 15 stories, News Agency African (9.6%) 13 stories, nine (6.7%) stories from News Agency non-African, eight (5.9%) stories each from Journalist initiated stories and individuals, six (4.4%) stories form press releases and one (0.7%) story from the “other” category. 5.5 Story Enhancement Story enhancements complement the story by reinforcing the message, giving more detail about the story or adding more visual/pictorial illustration to the information in the story. Enhancements also help to attract readers to such stories as compared to long straight news stories without any enhancement. Since online is a multimedia medium, the study examined the news stories for video, audio, picture and infographic enhancements. The study specifically analysed the format in which the stories were presented. FIGURE 5: Story Enhancement Story Enhancement Types Infographics only 2 Text with infographics 2 Text with video and pictures 1 Story with picture 141 Story with video 229 Pictures Stories 2 Video with text 24 Text Only 154 0 50 100 150 200 250 Frequency 66 University of Ghana http://ugspace.ug.edu.gh It was found (Figure 5) that video enhancements, that is story with video were the most dominant, making over two fifths (41.3%) amounting to 229 stories of the total stories analysed. Text only, that is, stories with no enhancement were the second major type of stories, with over one fourth (27.7%) or 154 stories followed closely by stories with pictures which recorded 25.4% (141 stories) of the coverage analysed. There were 24 (4.3%) news stories which were video with a little text explanation as well as two (0.4%) each of picture with text, text with infographics, and infographics only. One story (0.2%) with a picture and video, was the story with the least enhancement in the study. From findings as illustrated in Table 6 comparing the media houses and types of story enhancements, it was found that about three fourth (74.8%) or 228 of all of CNN’s stories were accompanied by video enhancements. Stories with pictures (8.2%) was the second highest enhancement in CNN’s stories. There were 24 (7.9%) video with text stories and one (0.3%) story of all the stories analysed which contained both a video and picture. The video with text stories are stories that are dominantly video with details of the source, and a brief narration of what is contained in the video. There was 8.2% (25) text only stories. In the case of GNN, most (92.6%) or 125 of their total stories, did not have any enhancements. In other words, majority of their stories contained text only. 67 University of Ghana http://ugspace.ug.edu.gh TABLE 6: Comparing Media and Enhancement KEY Most News Second Highest Least News ENHANCEMENT MEDIA TYPE Total CNN n (%) GNN n (%) Daily Nation n (%) Text Only 25(8.2%) 125(92.6%) 4(3.5%) 154(27.7%) Video with text 24(7.9%) 0(0.0%) 0(0.0%) 24(4.3%) Pictures Stories 2(.7%) 0(0.0%) 0(0.0%) 2(.4%) Story with video 228(74.8%) 1(.7%) 0(0.0%) 229(41.3%) Story with picture 25(8.2%) 9(6.7%) 107(93.0%) 141(25.4%) Text with video and 1(.3%) 0(0.0%) 0(0.0%) 1(.2%) pictures Text with infographics 0(0.0%) 0(0.0%) 2(1.7%) 2(.4%) Infographics only 0(0.0%) 0(0.0%) 2(1.7%) 2(.4%) TOTAL 305(100.0%) 135(100.0%) 115(100.0%) 555(100.0%) It is not clear why they did not report the stories with pictures, considering they were at the heart of the outbreak. It was only nine (6.7%) of their stories that had pictures and one (0.7%) story that had a video accompanying it. One thing observed with the GNN stories that had pictures was that some of the pictures did not load, even though the caption of the picture was in the story indicating there was a picture. Majority (93.0%) or 107 of Daily Nation’s stories in the study were accompanied with pictures, four (3.5%) were text only, and two (1.7%) were accompanied with infographics. There were two (1.7%) stories that were infographics only in the stories by Daily Nation. The findings on story enhancements therefore reveal that the three media outlets had differences in the dominant enhancements that accompanied their stories. The GNN and Daily Nation had not made use of the advantage of the possibility of multimedia use online. The three media had also not used a lot of infographics; similar to what Paul et al. (2016) found on the study of Indian print media coverage of the Ebola outbreak. Infographics would be both 68 University of Ghana http://ugspace.ug.edu.gh attention-grabbing and informative for the purposes of awareness creation and prevention of Ebola. 5.6 Reflection of Enhancement on Story Sometimes the enhancements that accompany the news stories do not really reflect the information being given to the public and therefore defeat their purpose in the story. The reflection of the story was introduced to examine whether the enhancements reflect the story fully, partially or did not reflect the story at all. FIGURE 6: Reflection of Enhancement on Story Frequency 350 318 300 250 200 150 131 103 100 50 3 0 Fully Partially No No Enhancement The results presented in Figure 6 indicate that more than half (57.3%) of the enhancements which amounts to 318 stories, fully reflected the stories they accompany and close to one-fifth (18.6%), which is 103 stories, partially supported their stories. Only three stories (0.5%) had enhancements that did not reflect the accompanying stories. Almost a quarter (23.6%), that is 131 stories analysed did not have any enhancements. Findings of coverage among the individual media outlets presented in Table 7 shows that, majority (92.8%) which is 283 of CNN’s overall enhancements fully reflected the stories they accompanied. This means that they helped in adding more to what the text said or confirmed what was said in the news story. Less than seven percent (6.6%) of the remaining enhancements reflected partially on the stories they accompanied. 69 University of Ghana http://ugspace.ug.edu.gh TABLE 7: Comparing Media and Reflection of Enhancement on the Story REFLECTION MEDIA OF ENHANCEMENT Total CNN GNN Daily Nation Fully 283 6 29 318 92.8% 4.4% 25.2% 57.3% Partially 20 4 79 103 6.6% 3.0% 68.7% 18.6% No 0 0 3 3 0.0% 0.0% 2.6% .5% No Enhancement 2 125 4 131 .7% 92.6% 3.5% 23.6% TOTAL 305 135 115 555 100.0% 100.0% 100.0% 100.0% Partial reflection stories were ones that were accompanied by library tapes on Ebola. CNN stories that did not have enhancements were just two (0.7%) of their overall stories analysed. Unlike CNN, most of the stories by GNN (92.6%), that is 125 stories did not have any enhancements. Only six (4.4%) enhancements fully reflected the stories they accompanied, while 4 (3.0%) partially reflected the stories. Daily Nation also differed from both CNN and GNN in the reflection of the enhancement on the story. Over half, 68.7% (79) of Daily Nation’s enhancements partially reflected the story they accompanied while one fourth (25.2%) fully reflected the stories. Only three (2.6%) enhancements did not reflect the story, because they were not about Ebola. The number of Daily Nation stories without enhancements was four (3.5%). 5.7 News Makers/Story Actors Newsmakers sometimes serve as sources of news and the way they present and define issues determines the way news is framed (de Vreese 2005). With respect to the news makers or story actors, based on pre-test of data and literature, the study sought to find out whether news centred on victim/affected African, victim/affected non-African, health workers, international/regional organisations, pressure/advocacy groups and organisations/NGOs, 70 University of Ghana http://ugspace.ug.edu.gh government officials/agencies, individuals (including experts), religious/traditional leaders or other, that is, any other actors beyond the ones identified. TABLE 8: News Makers KEY Most News Second Highest Least News Responses NEWS MAKERS Frequency Percent Victim/Affected (African) 107 10.9% Victim/Affected (non-African) 112 11.5% Health Workers 136 13.9% International/Regional Organizations 161 16.5% Pressure/Advocacy Groups/NGOs 43 4.4% Government Officials/Agencies 256 26.2% Individual/Expert Opinions 126 12.9% Religious/Traditional Leaders 7 .7% Other 30 3.1% TOTAL 978 100.0% Percentages and totals are based on responses The responses for this question were multiple responses, that means the question allowed for more than one newsmaker to be identified within one story. This was necessary because preliminary look at the data showed that there were some stories that had more than one newsmaker. A maximum of three major news makers was coded for in a story. It was found per information in Table 8 that the total number of newsmakers involved in the 555 stories analysed were 978. The newsmakers featured most in the stories analysed were Government Officials, with over a quarter (26.2%) responses. International and Regional organisations with (16.5%) appearances, Health workers with (13.9%) responses, Individuals and Experts with 126 (12.9%) responses and Victim/infected (non-African) (11.5%) completed the top five newsmakers in the study. The least newsmakers in the stories analysed were Religious and Traditional leaders, who generated only seven (0.7%) of the news stories analysed. 71 University of Ghana http://ugspace.ug.edu.gh TABLE 9: Comparing Media and Newsmakers KEY Most News Second Highest Least News MEDIA NEWSMAKERS Total CNN GNN Daily Nation Victim/Affected (African) 62 24 21 107 12.7% 9.4% 9.1% Victim/Affected (non-African) 86 12 14 112 17.6% 4.7% 6.0% Health Workers 82 25 29 136 16.7% 9.8% 12.5% International/Regional 52 54 55 161 Organizations 10.6% 21.1% 23.7% Pressure/Advocacy 6 19 18 43 Groups/NGOs 1.2% 7.4% 7.8% Government 103 91 62 256 Officials/Agencies 21.0% 35.5% 26.7% Individual/Expert Opinions 79 25 22 126 16.1% 9.8% 9.5% Religious/Traditional Leaders 1 4 2 7 .2% 1.6% .9% Other 19 2 9 30 3.9% .8% 3.9% TOTAL 490 256 232 978 Percentages and totals are based on responses With respect to individual media outlets, as presented in Table 9, CNN’s coverage was dominated by government officials and agencies with 103 (21.0%) of stories, followed by 86 (17.6%) stories dominated by Victim/Affected (non-African), and health workers with 82 (16.7%) appearances. Other (3.9%), Pressure, Advocacy Groups and NGOs (1.2%), and Religious and Traditional leaders (0.2%) were the newsmakers with the least presence in CNN’s stories. The Government institution most involved in CNN’s coverage was the Centres for Disease Control and Prevention of the USA which was very instrumental in trying to find ways to curb the outbreak. Similar to CNN’s stories, Government Officials and Agencies were the most cited newsmakers in GNN’s stories. They represented 35.5% (19) stories from GNN’s 72 University of Ghana http://ugspace.ug.edu.gh coverage. International and regional organisations (21.1%), Individual and expert opinions, and health workers, which both recorded 9.8% stories made up top three newsmakers. The three newsmakers that contributed the least stories in GNN’s total stories were Victims/Affected (non-African), 24 stories, Religious and Traditional leaders, four stories, and “Other” which recorded two (0.8%) stories. Daily Nation, just like CNN and GNN had Government Officials and Agencies as its dominant newsmaker. They produced 26.7% (62) of the stories. The other top two newsmakers were Health Workers (12.5%) and International and Regional Organisations which had 23.7% or 55 news stories. The newsmaker that produced the least news in Daily Nation’s stories analysed was Religious and Traditional leaders, accounting for two (0.9%) stories. Some international organisations which made news in the coverage were the WHO, UN and ECOWAS who were making efforts to curb the outbreak. Some Pressure/advocacy groups and NGOs that were newsmakers are Doctors Without Borders, the Red Cross and Samaritan’s Purse, a US Christian humanitarian aid organisation. Individual newsmakers were mostly disease experts who shared their views about the outbreak. The individuals included Dr. Peter Piot, the director of London School of Hygiene and Tropical Medicine, who is credited with discovering the Ebola virus in 1978. 5.8 Episodic and Thematic Framing Episodic and Thematic frames were studied to find out whether news coverage on the Ebola epidemic by the media outlets studied, placed the stories within a certain context by giving them background information (thematic) or whether they were reported as episodes (episodic) or particular instances. 73 University of Ghana http://ugspace.ug.edu.gh FIGURE 7: Episodic and Thematic Framing Frequency 41.4%(230) 58.6%(325) Episodic Thematic Findings presented in Figure 7 show that close to three-fifths (58.6%), which translates to 325 news stories in the study were episodic in nature whilst the remaining 41.4% or 230 stories were thematic in nature. This means that majority of the news were reported without much background information. TABLE 10: Comparing Episodic and Thematic Framing across Media KEY Most News Least News FRAME MEDIA Total CNN GNN Daily Nation Episodic 216 95 14 325 70.8% 70.4% 12.2% 58.6% Thematic 89 40 101 230 29.2% 29.6% 87.8% 41.4% TOTAL 305 135 115 555 100.0% 100.0% 100.0% 100.0% Comparing framing type by media as indicated in Table 10, the study found that most of Daily Nation’s stories (87.8%) were thematic while GNN had 40 (29.6%) and CNN had 89 (29.2%) thematic stories. The episodic frame was dominated by CNN with 70.8% (216 stories) out of 305 stories whilst GNN had 95 (70.4%) stories and Daily Nation had 14 (12.2%) stories which were episodic. This means that more stories were reported as instances without background 74 University of Ghana http://ugspace.ug.edu.gh information or context. When stories are reported without context, they may not provide the broader picture of an issue. 5.9 Causal Responsibility Framing The causal responsibility frame sought to identify the persons or things that have been attributed as responsible for originating the problem. Figure 8 contains the details of this finding. FIGURE 8: Causal Responsibility Framing Causal Responsibility Framing 300 250 271 200 150 100 119 97 50 68 0 Frequency Individual Responsibility Institutional Responsibility None Other Figure 8 shows that in almost half, 48.8% (271) of the stories, nobody was positively identified as the cause of the Ebola epidemic. Most of the stories in this category only gave statistics of the Ebola epidemic without giving any interpretations. The activities of Government Institutions have been identified to contribute most to the Ebola epidemic outbreak with over a fifth (21.4%) of the stories. Stories such as lockdowns by Government which led to people hiding their sick family members from health workers and fleeing their towns were blamed for the increase in casualties. “Other”, which referred to source outside of the ones specifically mentioned, were also identified as causes of the Ebola epidemic, constituting close to twenty per cent (17.5%) of stories This category contained stories that framed animals like the fruit bat, monkeys and consumption of “bush meat” as the cause of the epidemic. Among this 75 University of Ghana http://ugspace.ug.edu.gh category were also news that suggested that accused International organisations and NGOs of coming to spread the disease through vaccination. Individual action and activities were also identified in over ten percent (12.3%) or 68 stories as the cause of the spread of the disease. Personal hygiene and some cultural practices such as bathing the dead before burial and African burial rites were identified as individual activities that could cause the disease to spread further since the virus was still active in the corpse and could be transmitted through any bodily fluid of the dead person. TABLE 11: Comparing Causal Responsibility Framing and Media KEY Most News Second Highest Least News CAUSAL MEDIA RESPONSIBILITY Total CNN GNN Daily Nation Individual Responsibility 32 18 18 68 10.5% 13.3% 15.7% 12.3% Institutional Responsibility 63 17 39 119 20.7% 12.6% 33.9% 21.4% None 164 73 34 271 53.8% 54.1% 29.6% 48.8% Other 46 27 24 97 15.1% 20.0% 20.9% 17.5% TOTAL 305 135 115 555 100.0% 100.0% 100.0% 100.0% Looking at the causal responsibility from the perspective of the media outlets (Table 11), it was found that GNN (54.1%) and CNN (53.8%) both had their highest causal responsibility frames to be “None” while Daily Nation had its highest causal responsibility frame to be Institutional responsibility, which accounted for 33.9% (39) of stories. The least causal responsibility frame recorded by Daily Nation and CNN was individual responsibility, which accounted for 15.7% (18) of Daily Nation’s total stories, and 10.5 (32) of CNN’s total stories. However, in the case of GNN the frame with the least stories was institutional responsibility (12.6%). 76 University of Ghana http://ugspace.ug.edu.gh 5.10 Treatment Responsibility Framing This examined the news stories for the people who had been identified or suggested in the news to be the ones responsible for intervening in the Ebola epidemic and the issues associated with it as captured in the news. TABLE 12: Treatment Responsibility Framing Responses TREATMENT RESPONSIBILITY Frequency Percentage Individual/Attitudinal Change 49 8.3% Institutional/Government Action 132 22.3% International Donor Intervention 74 12.5% Global Action 107 18.0% None 231 39.0% TOTAL 593 100.0% Percentages and totals are based on responses From Table 12, it could be observed that close to two fifths, 39.0% (231) of news stories did not attribute treatment responsibility to anyone. Stories of this nature just gave facts and figures about the Ebola epidemic without making any interpretation of those facts. They also included stories whose main subject matter were not Ebola. Institutional/Government action (22.3%) representing 132 stories, global action (18.0%) representing 107 stories and international donor intervention (12.5%) that is 74 stories, were the major entities that were identified as being responsible for taking action to fight the Ebola epidemic. Comparing the coverage across media types (Table 13), it was found that all the three media platforms had majority of their stories not attributing treatment responsibility to anyone. CNN had 44.2%, more than ten per cent news than GNN (33.8%) and Daily Nation (32.6%) in this category of treatment responsibility. In stories where treatment responsibility was attributed to someone, it was observed that the trend of attribution was not so different among the three media. 77 University of Ghana http://ugspace.ug.edu.gh TABLE 13: Comparing Treatment Responsibility Framing and Media KEY Most News Second Highest Least News TREATMENT MEDIA Total RESPONSIBILITY CNN GNN Daily Nation Individual / Attitudinal Change 19 18 12 49 6.1% 11.7% 9.3% Institutional / Government Action 73 33 26 132 23.5% 21.4% 20.2% International Global Intervention 28 26 20 74 9.0% 16.9% 15.5% Global Action 53 25 29 107 17.1% 16.2% 22.5% None 137 52 42 231 44.2% 33.8% 32.6% TOTAL 310 154 129 593 Percentages and totals are based on responses Institutional (Government) action was attributed by CNN (23.5%) and GNN (21.4%) as the main solution to the Ebola menace. Putting health institutions in shape to be able to treat Ebola patients and seeing to the enforcement of Ebola protocols, such as quarantines, to curtail the spread of the virus from a patient to the rest of his or her family were some of the actions stated. Even though Daily Nation (20.2%) had the least news which attributed responsibility to Institutional (Government) action, the difference is not very much from the CNN and GNN percentages of news. Daily Nation rather has “global action” as its main institution of treatment responsibility with 22.9% of stories, making Daily Nation the media with the highest percentage of news in that category. CNN had the second highest number of attributions of treatment responsibility to “global action” with 17.1% (53) news stories and GNN had the least with 16.2% of stories. Apart from the difference in the first and second attributions of treatment responsibility in this study, the order of ranking of the remaining attributions were the same. CNN, GNN and Daily Nation all had “international global intervention” and “individual attitudinal” change as their 78 University of Ghana http://ugspace.ug.edu.gh third and fourth choice of actors responsible for tackling the Ebola epidemic with just slight differences in percentage of stories. The overall effect of this is that no matter the type of media involved, it was framed that local governments and institutions should lead the way in tackling the Ebola epidemic before global support and international donors came in with their intervention. 5.11 Issue-Specific Frames This result examined the issue-specific frames in the online coverage of the Ebola Virus Disease epidemic. Shih et al.’s (2008) framing typology for the study of three public health epidemics was adopted for this study. They employed the frames to study the print media and indicated that the finding of differences in frames was due to differences in diseases. The purpose of the typology’s use in this study was to expand the scope of the frames to the online media and to also find out if there were differences or otherwise due to differences in media type. This finding was also based on multiple responses from the 555 news stories analysed. In all there were 796 manifestations of the various frames studied. TABLE 14: Issue-Specific Frames Responses ISSUE-SPECIFIC FRAMES Frequency Percent Action 357 44.8% Conflict 46 5.8% Consequence 182 22.9% New Evidence 18 2.3% Reassurance 100 12.6% Uncertainty 26 3.3% Other 67 8.4% TOTAL 796 100.0% Percentages and totals are based on responses It was found, as presented in Table 14. that close to half (44.7%) of the stories contained action frames. These were actions by various actors on how to deal with the disease. Consequence frame, with close to a quarter (22.5%), that is 182 stories, was the second Issue frame in the 79 University of Ghana http://ugspace.ug.edu.gh coverage. The frame included stories that recorded the social, economic, human casualties and all other effects of the disease on society. The third Issue frame in the coverage was the reassurance frame. This constituted 12.6% (100) stories in the coverage. The reassurance frame dwelled on stories that sought to reassure the public about the fact that despite the devastating effect of the virus, it could still be dealt with. “Other” is the next main issue frame in the coverage with close to nine percent (8.4%) of news. The “Other” frame contained stories that were not centrally on Ebola but had made some inference or attribution to Ebola. Conflict (5.8%), which was made up of stories on differences in opinion, arguments, disagreements, and antagonism between opposing opinions, uncertainty (3.3%), with stories about the uncertainties surrounding the disease, such as its cause, cure, possible spread and how it can be contained, and New Evidence frame (2.3%), which contained stories about new findings, discovery of new methods to prevent, treat and cure the disease, were the frames with the least news stories. This means that the media focused more on finding solutions to the outbreak, rather than dwelling on the other areas of the epidemic. TABLE 15: Comparing Issue-Specific Frames and Media KEY Most News Second Highest Least News ISSUE-SPECIFIC MEDIA FRAMES CNN GNN Daily Nation Total Action 179 95 83 357 43.9% 47.7% 43.9% Conflict 37 4 5 46 9.1% 2.0% 2.6% Consequence 89 41 52 182 21.8% 20.6% 27.5% New Evidence 10 4 4 18 2.5% 2.0% 2.1% Reassurance 38 40 22 100 9.3% 20.1% 11.6% Uncertainty 17 0 9 26 4.2% 0.0% 4.8% Other 38 15 14 67 9.3% 7.5% 7.4% TOTAL 408 199 189 796 80 University of Ghana http://ugspace.ug.edu.gh Comparing coverage across the media outlets in the study (Table 15), it was observed that all the three media outlets had action, consequence, and reassurance frames as their top three main issue-specific frames. CNN and GNN had conflict, uncertainty and new evidence frames, as their last three frames, consistent with overall coverage of the issue-specific frames. For Daily Nation, uncertainty frame preceded the conflict and new evidence frame in order of importance. 5.12 Media Representation The purpose of constructive journalism frames in this study was to help examine the kind of representation and narrations on Africa in relation to the Ebola epidemic. It was to identify whether the media in general and Western media (CNN) in particular would engage in negative stereotypical narratives on Africa in the treatment of a PHEIC like Ebola. The constructive journalism frames were grouped into constructive (positive), negative and neutral. Constructive stories reflect the themes of “accomplishments”, “hero” and “meaningful” narratives. These themes use positive psychological techniques and allow for holistic coverage of Ebola stories. Negative news referred to news that narrated Ebola through stereotypical images of Africa to depict a sense of hopelessness. They also included stories that stressed on the portrayal of Africa as a place of violence, conflict, poverty, disease, and corrupt governments and weak systems. Stories on Western superiority, negative cultural practices and portrayal of Africa as a single continent were also deemed negative. Neutral news included stories, which had neither the characteristics of constructive journalism, nor negative news, that is, news such as announcement of events or facts without comments. The study found per Table 16 that three out of four (77.7%) or 445 news stories were neutral in their representation and narration about African. This trend was the same in the coverage across individual media outlets. Hero and accomplishment narratives 3.6% each, African being portrayed as a place of poverty (3.4%), and reference to African Governments as corrupt leading to weak health systems (2.7%) were the major issues in the media representation and 81 University of Ghana http://ugspace.ug.edu.gh narrations around Africa in the Ebola epidemic coverage by three media outlets. The other issues found in the narration around Africa were Western superiority (2.2%), negative cultural practices (1.7%), accomplishment narrative (1.2%), disease, and “Other” with 1.0% stories each. The rest of the issues: Africa portrayed as a single country, othering, Africa presented as a place of violence and Africa presented as a place of conflict, received less than one per cent of stories. TABLE 16: Comparing Constructive Journalism Frames and Media CONSTRUCTIVE MEDIA JOURNALISM FRAMES Daily Nation n Total CNN n (%) GNN n (%) (%) Hero Narrative 12(3.8%) 3(2.1%) 6(4.9%) 21(3.6%) Accomplishment Narrative 3(.9%) 2(1.4%) 2(1.6%) 7(1.2%) Meaningful Narrative 6(1.9%) 7(4.9%) 8(6.6%) 21(3.6%) Violence 1(.3%) 0(0.0%) 0(0.0%) 1(.1%) Conflict 1(.3%) 0(0.0%) 0(0.0%) 1(.1%) Poverty 13(4.1%) 2(1.4%) 5(4.1%) 20(3.4%) Disease 4(1.2%) 0(0.0%) 2(1.6%) 6(1.0%) Corruption / Weak 8(2.5%) 2(1.4%) 6(4.9%) 16(2.7%) Systems Western Superiority 7(2.2%) 2(1.4%) 4(3.3%) 13(2.2%) Negative Cultural 5(1.6%) 0(0.0%) 5(4.1%) 10(1.7%) Practices Africa as a single country 2(.6%) 0(0.0%) 2(1.6%) 4(.6%) Othering 2(.6%) 0(0.0%) 0(0.0%) 2(.3%) Other 3(.9%) 1(.7%) 2(1.6%) 6(1.0%) Neutral 243(78.4%) 123(86.6%) 79(63.2%) 445(77.7%) TOTAL 310(100.0%) 142(100.0%) 121(100.0%) 573(100.0%) Percentages and totals are based on responses. Analysing these frames by comparing the frames with media in the study (Table 16), it was found that in the constructive frames; hero, accomplishment and meaningful narratives, CNN and Daily Nation dominated news on these frames. Daily Nation had the highest percentage of news in the all the three constructive frames (hero, accomplishment and hero narratives) and GNN had the second highest in both accomplishment and meaningful narratives. Apart from 82 University of Ghana http://ugspace.ug.edu.gh the hero narrative in which it came second, CNN had the least percentage of stories under the other constructive frames (accomplishment and meaningful narratives). Daily Nation and CNN dominated the negative and stereotypical frames through which Africa and Ebola were narrated in the coverage of the epidemic. It was only CNN that had stories on portrayal of Africa as a place of conflict, violence and the use of the technique of “othering”. The themes of Africa’s portrayal as a place of poverty, disease, a single continent, negative cultural practices and corrupt governments leading to weak systems, and the theme of Western Superiority, were dominated by Daily Nation, followed by CNN. GNN recorded the least negative stories in news on poverty, western superiority, corruption/weak systems and “other”. FIGURE 9: Constructive Journalism Frames Chart Title 77.7% Neutral 445 13.8% Negative 79 8.5% Constructive (Positive) 49 0 50 100 150 200 250 300 350 400 450 500 Percentage Frequency Looking at the media representation broadly from the point of whether it is constructive (positive), negative or neutral, the study found that the media in this study had been neutral in their narrations on Africa with reference to the Ebola epidemic coverage. Findings per Figure 9 show that over three out of every four stories (77.7%) were neutral, over one eighth (16.4%) of the stories were negative and close to ten percent (8.5%) were positive. 83 University of Ghana http://ugspace.ug.edu.gh TABLE 17: Media Representation (Constructive Frames) across Media KEY Most News Second Highest Least News Total MEDIA FRAMES CNN GNN Daily Nation Constructive 21 12 16 49 (8.5%) 6.8% 8.5% 13.2% Negative 46 7 26 79(13.8%) 14.8% 4.9% 21.5% Neutral 243 123 79 445(77.7%) 78.4% 86.6% 63.2% 310(100%) 142(100%) 121(100%) 573(100%) TOTAL Within specific media, as presented in Table 17 it was found that all the three media had most news on narration of Africa to be neutral. GNN had the highest percentage of neutral stories (86.6%), followed by CNN with 78.4% and Daily Nation with 63.2% of their total stories. For CNN and Daily Nation, negative news was their dominant news after the neutral news. Daily Nation had more negative (21.5%) stories than CNN’s negative news (14.8%). Daily Nation dominated the constructive frame with 13.2% of its total stories, followed by GNN with 8.5% and CNN who had 6.8% of constructive news in their total news analysed. 5.13 Ethical Issues in Coverage Based on previous studies on disease coverage in general and public health epidemics in particular, it was identified that disease coverage is usually accompanied by some ethical issues which may affect the patients involved. It was also noted that the internet as a medium of information on health issues made this problem worse. This study therefore set out to examine the ethical issues that emerged in the coverage on the Ebola epidemic. Specifically, the study examined news stories for scare mongering, sensationalism, breach of patient’s privacy, the use of gory images and “other” for any other ethical issue that may be identified. Stories that did not contain any of these issues were coded to have “No Ethical issues”. 84 University of Ghana http://ugspace.ug.edu.gh TABLE 18: Ethical Issues in Coverage Responses EHTICAL ISSUES Frequency Percent Scare Mongering 26 4.6% Sensationalism 43 7.6% Breach of Patient's Privacy 5 .9% Gory Images 1 .2% Other 32 5.7% No Ethical Issue 458 81.1% TOTAL 565 100.0% The findings presented in Table 18 indicate that most of the stories (81.1%) did not contain any ethical issues that usually are identified with coverage on epidemics and disease outbreaks. Sensationalism (7.6%) was the major ethical issue involved in the Ebola epidemic coverage by the three media in this study. Sensational stories usually contain facts about the epidemic, which are presented in shocking language. Other (5.7%), which represented stories which could not clearly be categorised under any of the specific ethical issues listed, was the third ethical issue with the most stories. Some of the stories here were on the human trial of Ebola vaccines where the details of those to be involved in the trial were put out. Because they were not patients, such stories were categorised as “Other” instead of breach of patient’s privacy. Scare mongering (4.6%), in which stories are presented with exaggerations that eventually cause fear and panic, was the next dominant issue. Breach of Patient’s privacy and use of gory images, recorded below one per cent of coverage each to complete the list of ethical issues in the coverage on the epidemic. Consistent with the overall coverage figures, as in Table 19, all the three media platforms had majority of their stories having no ethical issues. It was only CNN that had stories on the “other” category of ethical issues. GNN and Daily Nation had no stories on use of gory images and other categories and there was no story on breach of patient’s privacy by Daily Nation. The 85 University of Ghana http://ugspace.ug.edu.gh rest of Daily Nation’s storied were on Sensationalism (6.0%) and scare mongering which had 4 (3.2%) of stories. TABLE 19: Ethical Issues in coverage across Media KEY Most News Second Highest Least News MEDIA ETHICAL ISSUES Total CNN GNN Daily Nation Scare Mongering 21 1 4 26 6.8% .7% 3.4% Sensationalism 30 6 7 43 9.6% 4.3% 6.0% Breach of Patient's Privacy 2 3 0 5 .6% 2.2% 0.0% Gory Images 1 0 0 1 .3% 0.0% 0.0% Other 32 0 0 32 10.3% 0.0% 0.0% No Ethical Issue 225 128 105 458 72.3% 92.8% 90.5% TOTAL 311 138 116 565 For GNN the stories were on sensationalism (4.3%), breach of patient’s privacy (2.2%) and scare mongering (0.7%). CNN had stories on all ethical issues that were studied. The rest of their stories aside from the ones on “no ethical issue” and “other” were sensationalism 30 (9.6%) stories, scare mongering 21 (6.8%) stories, and breach of patient’s privacy and use of gory images which had less than one per cent of stories each. 5.14 Comparing News Makers and Coverage Frames In order to examine the frames used by the various media for specific newsmakers identified in the study, a cross tabulation of media, newsmakers and coverage frames was done (see Appendix IV). This helped to address objective three and answer research question two. It was also necessary to ascertain the assertion by Yusuf et al. (2015) that before the outbreak reached other parts of the world, especially America, the headlines and frames in the Western media were negative, portraying the outbreak as impossible to tackle. Some of the headlines were “No 86 University of Ghana http://ugspace.ug.edu.gh cure for Ebola” and “no vaccination against it”. However, when the outbreak extended to the USA, the headlines and frames changed into hopeful ones, suggesting there is hope in tackling the outbreak. Knowing how the various newsmakers were framed would help address this issue. It was found that action frame was the dominant frame in CNN’s coverage on victim/affected African newsmakers with 40.4% of news, with consequence (28.1%) and reassurance (12.4%) frames being the second and third major frames respectively. Similarly, for victim/affected non-African newsmakers, action frame dominated with 41.4%, followed by consequence (29.7%) as the second dominant frame. The third dominant frame in CNN’s coverage on victim/affected (non-African) was “other” with 11.7% of other news frame while reassurance frame was fourth with 8.1% of CNN’s reassurance framed stories. Following the trend of victims/affected African, action (45.9%), consequence (28.7%), and reassurance (9.0%) were the three major frames of CNN’s frames on health workers. Furthermore, action, consequence and reassurance frames were the top three frames on international/regional organisations, pressure/advocacy groups and NGOs, and government officials / agencies as newsmakers in CNN’s coverage. Individual/expert opinion, and “other” newsmakers had action and consequence as their top two dominant frames but “other” as their third dominant frame, unlike the other newsmakers. The only one story on religious/traditional leaders as newsmakers fell under the consequence frame in CNN’s coverage. GNN’s coverage had similarities in frames on various newsmakers as pertained in CNN’s coverage. Action, consequence and reassurance frames dominated frames on newsmakers and in most cases “other” was the fourth dominant frame. The action, consequence and reassurance frames were the three top frames in GNN’s coverage on victim/affected African, victim/affected non-African, health workers, and pressure/advocacy groups and NGO 87 University of Ghana http://ugspace.ug.edu.gh newsmakers. International/regional organisations, and government officials/agencies also had action frame as their dominant frame but reassurance and consequence frames changing position as second and third dominant frames, different from the other newsmakers mentioned earlier. Individual/expert opinion had action (50.0%), other (20%) and consequence (16.7%) as the top three frames by GNN. Religious/traditional leaders had only four news items, three (75%) being action and one (25%) “other” frame. “Other” newsmakers had consequence (25%), reassurance (25%) and other frame (50%) in GNN’s coverage. There were no stories with uncertainty frames by GNN for any of the newsmakers. The frames of Daily Nation on newsmakers followed the trends of CNN and GNN. Action consequence and reassurance dominated coverage on the various newsmakers. Victim/affected (African), victim/ affected (non-African), health workers, international/regional organisations, pressure/advocacy groups/NGOs, and government officials/agencies all had their first, second and third dominant frames to be action, consequence and reassurance respectively. Individual/expert opinion and “other” newsmakers also had their first and second dominant frames to be action and consequence but differed from the others in their third dominant frame. Reassurance and other with 8.6% each of news were the third dominant frames for individual / expert opinion newsmakers while the “other” frame was the third dominant frame for “other” newsmakers in the coverage by Daily Nation. From the findings it can be concluded that apart from the percentage of news which differed slightly, there was not much difference in the frames employed by the various news outlets in their coverage on the newsmakers identified in the study. Also, the action, consequence and reassurance frames dominated the coverage on majority of the newsmakers. 88 University of Ghana http://ugspace.ug.edu.gh 5.15 Chapter Summary This chapter presented the findings of the study. Generally, the study found that straight news dominated the coverage by CNN, GNN and Daily Nation on the Ebola epidemic. The subject matter of the stories was mainly on disease statistics and most of these stories were journalist- initiated stories. Government officials and agencies were the major news makers in the stories. News in this study mostly captured events and therefore there was not much background information to most of the stories. This means the coverage was more episodic in nature. The three media narrated their Ebola stories around action, consequence and reassurance frames. The three media did not assign causal and treatment responsibility of the disease to anyone in majority of their stories, however, when responsibility was assigned, all of them assigned responsibility of cause and treatment to government/institutional action. Finally, the narrations and representations of Africa in relation to the Ebola virus disease by the CNN, GNN and Daily Nation was largely neutral and there were no major ethical issues in the framing of the Ebola epidemic. 89 University of Ghana http://ugspace.ug.edu.gh CHAPTER SIX ANALYSIS AND DISCUSSION OF FINDINGS 6.0 Introduction This study conducted a quantitative content analysis of online news by an international media (CNN), national media (GNN) and African media (Daily Nation) on the West African Ebola epidemic which broke out in Guinea in December 2013. The study sought to: find out the dominant frames in the coverage of the epidemic; examine the extent to which differences and similarities in issue-specific frames across the media platforms depended on newsmakers; examine the ethical issues arising from the coverage; and investigate the kinds of media representation Africa received in the coverage through the narratives on Africa in relation to the epidemic. The study covered the period of 1st March, 2014 to 31st December, 2014. This chapter discusses the key findings, which help address the objectives and answer the research questions raised for the study. The discussion draws inferences and links from previous studies reviewed and the theoretical framework that underpinned the study. The discussion is done along the research questions of the study. 6.1 Dominant Frames in the Coverage The study examined coverage to identify the issue-specific frames, episodic and thematic frames, and responsibility (causal and treatment) frames employed in the online media coverage of the West Africa Ebola outbreak. This was to respond to the first objective and, flowing from it, the first research question on the dominant frames used in the coverage. The issue-specific frames studied were based on the frame typology of Shih et al. (2008). It was found that, collectively across the three media platforms, the action frame, which chronicled actions that were taken to contain the outbreak, dominated with 44.7% of news. Action frames included stories about what was being done to curb the Ebola epidemic. Actions such as donations to the affected countries, quarantining of communities and deployment of health 90 University of Ghana http://ugspace.ug.edu.gh personnel, for instance, by the US to the affected regions to support the logistics and human resource needs of the countries were contained in the frame. This was consistent with the findings of both Shih et al. (2008) and Nwanne (2014) about action as the most dominant frame in their studies of the US print media and Nigerian media respectively. The finding, however, was contrary to Zhang and Matingwina (2016) and Paul et al. (2016) who found cause of the disease and cases of Ebola as dominant frames in their study of BBC and China Daily, and English Indian media respectively. Furthermore, it was contrary to Smith and Smith (2016) who also found Ebola cases and Abeysinghe (2016) who found the theme of fear and contagion to be the most dominant frames in their study of Nigerian, and US, UK and Australian print media respectively. Finally, it was inconsistent with Basch et al (2014) who found emphasis on death tolls as the dominant theme in their study of the US print media. Furthermore, similar to Shih et al. (2008), this study found the consequence frame as the second main issue-specific frame in the coverage. This frame constituted 22.5% of all the stories in the study. This frame recounts the consequences or impact (social, cultural, political or economic) of the Ebola epidemic on the affected people, communities and countries. Loss of lives was one consequence of the outbreak that was very much emphasized. There were also stories on the economic impact of the outbreak on Liberia, Guinea and Sierra Leone, who were the most hit. Consistent with this finding were the findings of the situation in Ebola-stricken countries by Zhang and Matingwina (2016), and death tolls and cases by Basch et al. (2014). The third main issue-specific frame was the reassurance frame which had 12.6% (100) of the stories. The reassurance frames involved stories that promised and assured people of hope of overcoming the epidemic. In CNN’s reportage this frame was very dominant in October stories when cases of the disease were diagnosed in the US. Citizens demanded action from the 91 University of Ghana http://ugspace.ug.edu.gh governments and through government officials and agencies, citizens were assured of measures to stop the disease from spreading into the larger population. Other, conflict, uncertainty, and new evidence were the frames that received the least coverage among the issue-specific frames studied. One major source of conflict frame was about the mode of transmission of the virus from one person to another. Experts disagreed on whether the virus was airborne or not that is, whether it could be transmitted by air or not. There were also disagreements about whether mass quarantines and lockdowns were the way to go or not, and whether Ebola is infectious or contagious. Some experts argued that the virus was contagious (spreads easily from person to person), while others say it is infectious (takes only a small amount to cause an infection). On the specific media, coverage among the various media showed (Table 15) that all the media were consistent with the overall coverage with action, consequence and reassurance being their top three dominant frames. In the case of CNN, the “other” frame also had the same amount of news as the reassurance frame. The three media outlets were not so consistent with the remaining frames. In order of percentages, “conflict”, “uncertainty”, and “new evidence” was the framing pattern of the rest of CNN’s coverage on the Ebola epidemic. In GNN’s coverage, “other”, “conflict”, and “new evidence” with the same percentage of news, and “uncertainty” frame with no stories, were the last four frames. Daily Nation was also consistent with “other” as its fourth dominant frame but differed from CNN and GNN in the order of its last three frames. “Uncertainty”, “conflict” and “new evidence” frames were the last three frames in Daily Nation’s coverage. This means that overall, there was little difference in the framing of Ebola by CNN, GNN and Daily Nation, especially when they had similarities in their top three dominant frames. 92 University of Ghana http://ugspace.ug.edu.gh The coverage of the Ebola epidemic by CNN, GNN and Daily Nation was therefore predominantly about the activities that were being carried out in order to control the spread of the disease, following with the effects of the virus in terms of economic, social, political and cultural impact and then reassuring audiences that there was hope in the fight against the virus. This approach is not out of place since it was important that the people know about the measures all the stakeholders were putting in place to ensure that the disease was put in check and reassure them to be able to will go about their livelihood and activities without any fears. Some of the actions reported in the coverage were building of health facilities, quarantining of communities and people who were suspected of having being exposed to the virus and provision of medical supplies to help health workers protect themselves. It was not every action carried out to combat the disease that was perceived as positive by the people and the community. An action such as a lockdown by the government of Sierra Leone to enable health volunteers go and talk to people door-to-door in a bid to halt the spread of the virus was criticised by Medecins Sans Frontieres (Doctors Without Borders), a health charity NGO which was instrumental in the fight against the virus. They held such an action as something that would drive people underground and jeopardise the trust between the people and healthcare providers. This situation they note, could lead to concealment of cases and push the sick away from health systems. Also, the fact that the consequence frame was the second main issue- specific frame suggested that as the media provided the opportunity for people to get to know what stakeholders were doing to address the virus (action), the people were also being made to know about the impact of the virus on individuals and society as a whole. This would make people see the need to comply with the necessary authorities in the bid to curb the virus. The economic impact of the low patronage of bush meat, which was regarded as one of the modes of contracting the disease in the most heavily affected countries, was widely reported. The loss of lives of both ordinary people and health workers in the fight against the disease also 93 University of Ghana http://ugspace.ug.edu.gh dominated the consequence frame. Reassurance being the third main issue frame and coming just after the consequence frame goes further to make the point that despite the debilitating effects of the virus on people and the society, there was hope in fighting and eradicating the virus. If the conflict and uncertainty frames, which were the fifth and sixth major frames, had been the main issue-specific frames, it would have sent a signal to people that the authorities were not doing much to curb the outbreak and therefore there would have been increased fear and panic in the society. From the dominant issue-specific frames identified in the study, it is evident that the stories could not have created much fear and panic, contrary to conclusions by Yusuf et al. (2015). The implications of these for theory is that Shih et al.’s (2008) typology of issue-specific frames developed for the study of epidemic hazards under framing theory, and adopted for this study, produced similar findings as the original study which found action and consequence as the dominant frames in the coverage of public health epidemic hazards. This implies that the framing of public health epidemic hazards, in this case Ebola, did not differ based on differences in media type and different journalistic practices or geographical location. This further confirms the assertion by Shih et al. (2008) that journalists tend to use the same themes for stories of similar nature. This means that the research heuristic of the framing of public health epidemics using Shih et al.’s (2008) framing typology has been extended from its application to print media to the online media platforms of CNN, GNN and Daily Nation. This study also examined some generic frames in the coverage of Ebola by CNN, GNN and Daily Nation. The generic frames studied were responsibility frames, that is, causal and treatment responsibility, and episodic and thematic frames. The findings from this study on the episodic and thematic framing of issues indicated that close to three out of every five, that is, 58.6% of the stories were episodic whiles the remaining 41.4% 94 University of Ghana http://ugspace.ug.edu.gh stories were thematic. This finding is consistent with Iyengar’s (1991) finding that daily news coverage was biased towards episodic frames and his assertion that the standards and norms of news organisation and production reinforce episodic framing. It must be noted that online news is a daily news format. The episodic framed stories presented issues and events as disconnected instances, which could sometimes make it difficult for people to connect the issues for deeper understanding. However, the advantage of this is that it helps in creating awareness which is an important function in outbreak communication. Thematic framing of news situated the news within a context, providing the necessary background needed for easy understanding and linking stories with previous stories, relevant historical and socio-economic contexts that would help clarify issues better. This approach helps to provide an in-depth analysis and clarification on an issue. The thematic framed stories in this study were mostly backgrounded with how the Ebola outbreak started in December 2013, the current state of the disease in terms of statistics on cases and casualties, and how it has been tackled so far. With respect to the specific media outlets in the study, it was found that CNN and GNN were consistent and similar in having most of their stories presented in episodic manner to the overall coverage. Daily Nation differed from the two by having majority of its stories (87.8%) to be thematically framed. Considering there had not been any cases of Ebola in Kenya, where the Daily Nation is located, it was surprising why they rather gave detailed information about most of their stories than CNN and GNN which had cases and therefore needed to provide their audiences with as much information about the virus. GNN, which was operating within the country of the outbreak in Liberia, had only close to a third (29.6%) of their total stories thematically framed, similar to CNN which also had only 29.2% thematic stories. This means over 70% each of stories from CNN and GNN were reported without contextual basis which could lead to a non-contextualised understanding of the issues in those stories. 95 University of Ghana http://ugspace.ug.edu.gh Causal responsibility framing in this study was predominantly not attributed to anyone. Almost half, 48.8% of the stories, did not attribute responsibility of the cause or origin of Ebola to anyone. In other words, the media were neutral in who was to blame for the disease outbreak. The findings on the subject matter of stories presented in Table 4 (Chapter 5) could help explain this better. It was found that the dominant subject matter of stories in the coverage was disease statistics, with 42.9% of stories. Disease statistics stories gave the figures about the cases and casualties of the Ebola outbreak as well as updates on patients without any interpretations. This means that in many cases there was no mention of causal agents in the reports. Some disease statistics stories however attributed causal responsibility; hence, the percentage of news in the dominant subject matter and dominant causal responsibility frame were not the same. When causal responsibility was attributed however, institutional and government actions were blamed for the most cause and spread of the Ebola epidemic. Some activities of government or governmental institutions that were referred to were poor border control, lack of provision of the necessary health facilities to fight the outbreak, as well as forced quarantines and lockdowns by governments in affected countries, which was criticised as an action that could lead to people hiding relatives with the virus. “Other” which mostly bordered on stories that framed animals and animal products such as monkeys, fruit bats and bush meat as the origin, cause and spread of the disease, was the second entity that was attributed in terms of responsibility. This was followed by individual responsibility, which suggested that some activities of individuals were also to blame for the spread of the Ebola virus. Poor hygiene and cultural practices such as bathing and touching the dead before burial were identified as some of the causes of the spread of the virus. 96 University of Ghana http://ugspace.ug.edu.gh In causal responsibility framing (Table 11), the different media outlets differed in how each of them framed who or what was causally responsible for blame in the Ebola outbreak. For CNN and GNN, no one was attributed to as responsible for causal responsibility in most of their stories, just as in the case of the overall causal responsibility framing. They however differed in the other frames that identified entities as the origin or cause of the virus. While CNN framed institutions and government agencies as the main cause, GNN had “other” as its dominant causal frame after “none”. The reference to government institutions as the main cause in CNN’s coverage was likely due to the fact that when the outbreak extended to the United States, Republican politicians and some other citizens blamed the Obama government for poor border control which allowed for someone like Eric Thomas Duncan, the first person to be diagnosed of the disease in the US, to enter the country. This was similar to the finding by Abeysinghe (2016) who found that the moment the first case of Ebola got to the US, Ebola became a domestic issue of border control, a phenomenon which shifted the focus of coverage from affected countries to the few cases in the US. The disease became a government problem than a focus on the accounts of the public. This could also account for the biased and unbalanced coverage in the US media, suggested by Yusuf et al. (2015). GNN may have framed “other” which contained news that sought to blame the spread of the outbreak on animals such as monkeys and fruit bats, and the meat of these bush animals because of the reality of the presence of these animals within the forests of the affected countries. “Other” and individual responsibility were the last two entities attributed for causal responsibility in the coverage by CNN. In the case of individual responsibility, because the cultural practices identified as the main mode of transmission of the disease were not very common in the American context, that could have accounted for why it was the least attributed to in terms of causal responsibility. Daily Nation framed government institutional action and 97 University of Ghana http://ugspace.ug.edu.gh activities as the most responsible for the origin and cause of the spread of the Ebola virus. None was the second dominant, followed by other and individual responsibility. It could be argued that the geographical location of the media involved could have played a role in who was identified as culpable in causal responsibility framing. On who was responsible for finding solution to the Ebola menace, as presented in Table 12, the media again did not attribute responsibility to anyone in almost forty per cent (39.0%) of their stories. In cases where treatment responsibility was attributed however, similar to the case of casual responsibility, the overall media coverage identified government action as the main way of tackling the disease. Activities such as the provision of health facilities and resources, education and creating awareness about the disease, instituting effective border controls and ensuring the implementation of preventive measures were placed at the doorstep of governments and their agencies and institutions. President Ellen Johnson Sirleaf of Liberia and other leaders in the affected countries had stated on various platforms that the disease was a national health problem which had dire consequences for the socio-economic life of the people, which she was resolved to tackle. Global action frame was the next attributed with treatment responsibility in the coverage. This recognised the fact that the outbreak was beyond just national concerns but something that could spread to other parts of the world, hence the need for both local and foreign cooperation to eliminate the disease. The UN and WHO were involved in this call for global action in the fight against the disease. International donor intervention was exclusively also framed as one of the ways to tackle the outbreak. This may be based on the assumption that the interventions needed to tackle the epidemic were beyond the remit of the local governments and therefore must be provided by foreign donors, especially developed countries. 98 University of Ghana http://ugspace.ug.edu.gh Lastly, individual actions and activities were also believed to have a role in curbing the spread of the virus. Maintaining good hygiene, washing of hands, treatment of water with chlorine before use and avoiding personal contact with infected persons were advocated for individuals to adopt in the fight against the disease. Families were also advised to desist from contact with their dead relatives before burial to avoid contracting the disease. They were also urged to report casualties of the Ebola virus to the health workers for burial. Finding that individual attitudinal change was the least attributed to in the treatment responsibility frame was surprising since the major ways of contracting the diseases were coming into contact with the bodily fluids of infected person or contaminated objects from infected persons. This was prevalent through the cultural practices and rites related to burial and consuming an infected animal. Stories should therefore have advocated more for behaviour change and taking precautions against the disease. Governments and international organisations could provide all the human resources and health facilities needed, but it will take individual attitudinal change to enable the people receive the benefits that come from those facilities, hence, individual attitudinal change should have received a lot more prominence in the coverage than has been found. Treatment responsibility framing within media was not different from the overall treatment framing. All three media outlets, consistent with the overall coverage, attributed treatment responsibility to no one (none) in majority of their stories. While CNN and GNN had institutional/governmental action as their main entity with the responsibility to act and curb the outbreak, Daily Nation called for global action in the fight against the epidemic. Considering the global and international outlook of CNN, one may have thought that their framing of treatment responsibility would be more global oriented, however, it was not. What may have accounted for this frame of institutional/government action was that after the first case of the virus was diagnosed in the US, there were various calls on the Obama government to act so as 99 University of Ghana http://ugspace.ug.edu.gh to prevent more cases of the disease. All the three media outlets differed in their next two frames on treatment responsibility framing with CNN having global action and international global intervention, GNN with international global intervention and global action, and Daily Nation’s next two being institutional/government action and international global intervention. Individual attitudinal change was the last frame by each of the three media. Considering the fact that governments and their institutions were identified as responsible in both causal and treatment responsibility framing, it is logical that someone who creates a problem must help solve it but it is also strange that government was attributed as the main entity causally responsible for the epidemic, when it had been reported that the main mode of transmission of the disease was through contact with the bodily fluids of an infected person. The findings on episodic and thematic framing, and the responsibility frames were contrary to Iyengar (1991) who found that individual causal and treatment responsibility tended to be associated with episodic framing, while thematic framing was associated with societal causal and treatment responsibility frames. In this study, the dominant frame was episodic while both causal and treatment responsibility were attributed to institutional responsibility. Overall, there were similarities and differences in the dominant frames that were identified in the coverage, from the issue-specific frames, and the two generic frames that were examined in this study. These similarities and differences were based on the order of importance that was attributed to the frames by the various media outlets. The similarities and differences observed also suggested that geographical proximity and ideological orientations of the media could lead to differences in frames by different media on the same issue, such that the importance attached to an aspect of an issue could vary among different media. These differences and similarities may also be based on the difference in routines, and journalistic approaches in the gathering and presentation of the news as suggested by Shoemaker and Reese (1991). 100 University of Ghana http://ugspace.ug.edu.gh 6.1.1 Test of Hypothesis One: The hypothesis tested was: H1: The differences in issue-specific frames used in coverage of the epidemic are significantly related to differences in media outlets. TABLE 15a: Comparing Issue-Specific Frames and Media ISSUE-SPECIFIC MEDIA FRAMES Total CNN GNN Daily Nation Action 179 95 83 357 43.9% 47.7% 43.9% Conflict 37 4 5 46 9.1% 2.0% 2.6% Consequence 89 41 52 182 21.8% 20.6% 27.5% New Evidence 10 4 4 18 2.5% 2.0% 2.1% Reassurance 38 40 22 100 9.3% 20.1% 11.6% Uncertainty 17 0 9 26 4.2% 0.0% 4.8% Other 38 15 14 67 9.3% 7.5% 7.4% TOTAL 408 199 189 796 Results are based on non-empty rows and columns in each innermost subtable X2 = 61.724 df = 14 p = 0.00 The findings on issue-specific frames (Table 15a) indicated that “action”, “consequence” and “reassurance” were the dominant frames of the seven frames. Also, coverage among the three media outlets found “action”, “consequence” and “reassurance” frames to be the top three frames for each of the media respectively. This was consistent with Shih et al. (2008) who used the same frames to study three public health epidemic hazards: avian flu, mad cow disease and West Nile virus in the New York Times. Shih et al. (2008) suggested that any differences in the result of their study was due to differences in disease rather than differences in journalistic practices, since they studied only one media. The results of the current study also suggest that differences in media or journalistic practices does not result in differences in frames since all 101 University of Ghana http://ugspace.ug.edu.gh the three media outlets framed the issues in a similar way. The hypothesis was tested to find out whether there was any significant relationship between the differences, in issue-specific frames in the coverage of Ebola and media type, since the results of the study show that difference in media does not affect how public health epidemics are framed. The level of significance of 0.05 was set for this test, while the null hypothesis was: H0: There is no significant relationship between the issue-specific frames in the coverage of Ebola and media types. A Chi-square value of 61.724 was obtained, with a probability of 0.00. The probability value is less than the significant value of 0.05. This means that the null hypothesis that there is no significant relationship between the issue-specific frames and media types, is rejected. The research hypothesis (H1) is therefore significant, indicating a relationship exists between the issue-specific frames in the coverage of Ebola and media type. The Cramer’s V value, which measures the strength of relationship (association) identified by the Chi-square test, was 0.19. This suggests that the relationship identified between the frames and media types is not very strong. A look at the Table 15a however does not show a difference in the major issue-specific frames in the coverage by the media, since all of them have “action”, “consequence” and “reassurance” as the top three frames with just differences in frequencies and percentages. This could have explained why the strength of relationship is not very strong. 6.2 Issue-Specific Frames and News Makers Across media Yusuf et al. (2015) asserted that there was unbalanced and biased coverage of the outbreak by many US media houses which contributed to increasing fear among the US population. They also found that even though a lot of people in the West African sub-region had contracted the disease and about 50% of them had died by August, the Western media, especially the US had not paid attention to the epidemic until after September, 2014 when the disease spread to the US. This was found to be true in this study since it was found that before September, there 102 University of Ghana http://ugspace.ug.edu.gh were 71 stories in the study from the CNN, and 34 stories were found in the sample of September, the month Ebola got to America. However, October, the month immediately after the disease got to America, the number of stories jumped to 169, the highest number of stories in the study and highest the for CNN. This figure is bigger than the total number of stories analysed from GNN and Daily Nation respectively. Yusuf et al. (2015) also noted that the spread of the disease from West Africa to the US caused a sudden change in the headlines, from headlines that portrayed a situation of hopelessness “No Cure for Ebola”, to that of hope “Experimental drug likely saved Ebola patients” which they indicated was biased and unbalanced coverage. The second objective and research question of this study which sought to identify whether the issue-specific frames differed based on newsmaker across the media studied, was supposed to confirm or deny assertion. This question was therefore to help ascertain the issue-specific frames that were used in the coverage for the various newsmakers by the three media outlets, so as to conclude whether depending on the newsmaker, especially victim/affected (African) and victim/affected (non-African) the frames differed. The victim/affected (African) referred to newsmakers who were African Ebola victims or their relatives and victim/affected (non-African) were newsmakers who were non- African Ebola victims or their relations. From the table presented in Appendix IV and findings at Chapter 5.14, it would be observed that in the coverage of CNN, framing of the various news makers did not differ a lot except in terms of percentages and in only a few instances when the top three frames were different. The top three dominant issue-specific frames on victim/affected (African), health workers, international/regional organisations, and government officials/agencies were “action”, “consequence” and “reassurance” frames. 103 University of Ghana http://ugspace.ug.edu.gh Victim/affected (non-African), and Pressure/advocacy groups/NGOs had action, consequence as their top two issue-specific frames but differed in their third frame. Pressure/advocacy groups/NGOs had reassurance frame having the same percentage as the consequence frame to be the second highest frame while uncertainty was its third major issue-specific frame. Individual/expert opinion newsmakers had action, conflict and other as their dominant frames, and “Other” which referred to newsmakers different from the ones listed had action, consequence and “other” frame as their dominant frame. Religious and traditional leaders had only one story, which was on consequence frame. For GNN, there were variations in the way the top three frames appeared but similar to CNN the top three frames in most cases were action, consequence and reassurance frames. Health workers, and pressure/advocacy groups and NGOs, newsmakers had action, consequence, and reassurance as their top three dominant frames. Individual/expert opinion newsmakers had action, other and consequence frames as their dominant frames. Government officials/agencies and international/regional organisations also had action, reassurance and consequence as their top three dominant frames with the difference in frames being the reassurance and consequences frames changing places. For victims/affected (African) and victims/affected (non-African) the consequence frame was the most dominant followed by action and reassurance respectively and in the case of Victims/affected non-African, the “other” also constitutes their third major frame since it had the same percentage of news as the reassurance frame. Other newsmakers had “other”, reassurance and consequence as their dominant frames. Religious and traditional leaders had stories on only action and other frames and the dominant frame was action. Daily Nation followed the trend of having action, consequence and reassurance as dominant frames on most newsmakers in the coverage. Apart from religious/traditional leaders, and 104 University of Ghana http://ugspace.ug.edu.gh “other” newsmakers, who had stories only on “action” and “other”; and “action”, “consequence” and “other” frames respectively, all the other newsmakers: victims/affected (African), victim/affected non-African, health workers, international and religious organisations, pressure/advocacy groups and NGOs, government officials and agencies, and individual/expert, all had action, consequence and reassurance as their top three dominant frames. This means that there was little difference in frames irrespective of the newsmaker involved. In the nutshell, on the question of whether frames differed depending on the newsmaker involved, it could be concluded that headlines may be negative or positive, but the issues they carried were not different irrespective of the newsmaker involved, particularly irrespective of if the newsmaker was African or non-African, that is victims and relatives of the people affected with the Ebola. There would have been differences in the frames, favourable to the non-African newsmakers, but it was not so in this coverage. Both African and non-African newsmakers received the same kind of framing, which were largely action, consequence and reassurance framing except for variations in percentages and frequencies. CNN, which is a Western media that would have varied its frames for African and non-African newsmakers per the assertion by Yusuf et al. (2015), also had the same frames for both African and non-African newsmakers. 6.3 Ethical Issues in the Coverage An objective of the study as reflected, in research question RQ3, was to find out the ethical issues that came up in the coverage of the Ebola epidemic by the CNN, GNN and the Daily Nation. As presented in Table 18, findings revealed that about four out of every five stories (81.1%) did not record any form of ethical issue. Of the stories that recorded ethical issues, sensational stories (sensationalism) recorded the highest of 7.6% followed by other, which contained stories that could not be classified under any of the listed ethical issues studied, with 105 University of Ghana http://ugspace.ug.edu.gh 5.7% stories, and scare mongering which recorded 4.6% stories. Breach of patient’s privacy (0.9%) and use of gory image in just one story (0.2%) were the issues with the least stories. Most of the sensationalist stories (30) were found in CNN’s stories with 7 in Daily Nation’s stories and 6 in GNN’s stories. It was the same with scare mongering where CNN had 21 stories, Daily Nation 4 stories and GNN with 1 story. Since the first case of Ebola was diagnosed in the United States on 30th September, 2014, the media attention had heightened and a lot of attention was placed on possible cases of the outbreak in the country. Sensational stories centred on the Liberian, Thomas Eric Duncan, who became the first person to be diagnosed of Ebola in the US. It was presented as though he knew he had the disease before heading to the US to deliberately infect people. This was also because he had answered ‘no’ to a question of whether he had taken care of an Ebola patient prior to the trip, which turned out to be a lie. Wilson, Shoichet & Yan, 2014 presented that Before leaving Liberia, Duncan also answered no to a question about whether he had touched the body of someone who died in an area affected by the disease, said Binyah Kesselly, board chairman of the Liberia Airport Authority (Wilson, Shoichet & Yan, 2014) One significant finding when it came to ethical issues was that the media only published details about the health workers, journalists and patients whose status significantly became very important in the history of the recent outbreak. One of such persons was Thomas Eric Duncan, who was the first person to be diagnosed with the disease in the US. It was also observed that in the case of the health workers and journalists, until they were tested positive of the disease, their identities were not given in the news stories. This was contrary to MacIntyre and Travaglia (2015) who found the use of explicit pictures of Ebola patients, publishing of details about patients and pictures of survivors being used for promotions by foreign companies. There was only one case of the use of gory image in the stories that were analysed. Pictures and videos used to enhance and tell the stories were adequately censored by covering faces of victims and reflected the stories they were used with. This was also in contrast with Cortiñas-Rovira et al.’s 106 University of Ghana http://ugspace.ug.edu.gh (2015) finding of the use of sensationalist and scaremongering images which did not present a true picture of stories by Spanish newspapers in their reportage of the 2009 swine flu outbreak. In terms of the pictures and enhancements presenting the true picture of the story in which they were used, it was found in this study that 57.3% of the 555 stories analysed, had their enhancements fully reflecting their stories, 18.6% partially reflecting the stories and only 0.5% enhancements did not reflect their stories. The rest of the stories (23.6%) did not have any enhancements (See Table 7 & Figure 6). This means that the majority of the pictures reflected the stories unlike in the case of Cortiñas-Rovira et al.’s (2015) finding. It was also found that majority of stories by GNN, 125 (92.6%) stories were text only. A check on the website of GNN, however, revealed that non-Ebola news during that period had pictures. It could be argued that this may be due to the fact that being at the centre of the epidemic, the use of text only could be to minimize exposing citizens to the gory details of the epidemic which they were already witnessing. They could have, however, used infographics for the purpose of creating awareness and disease prevention. Finally, one possible ethical issue in disaster situations such as the Ebola epidemic, as highlighted by MacIntyre and Travaglia (2015), was the use of non-expert sources. In this study however, as presented in Table 5, the findings suggest that the sources of news used were sources that could be easily verified. It was only 15.3% of stories that did not have their sources stated, showing that about 3 out of every 4 stories had their sources stated. This means that the media examined in this study did not appeal to non-expert sources in their coverage of the Ebola epidemic. In general, it can be concluded that the ethical issues in the framing of Ebola by the CNN, GNN and Daily Nation were minimal. This finding is contrary to MacIntyre and Travalgia’s (2015) finding which indicated that during disasters such as the Ebola epidemic, people ignore ethics 107 University of Ghana http://ugspace.ug.edu.gh and publish information that they would not have ordinarily published on the internet. Also, no explicit pictures of patients were posted online by the media. It also contradicts MacIntyre and Travalgia’s (2015) finding about posting of explicit pictures of victims online. The overall effect of these findings on ethical issues is that ethical issues are not limited to the Western media context since there were cases of sensational, scare mongering and breach of patient’s privacy stories in the coverage of GNN and Daily Nation, which are African media, though very minimal. 6.4 Media Representation One of the objectives and the fourth research question for this study was to find out the way Africa was represented by the media in the coverage through their narrations around Ebola. In the broader light, the study sought to find out if the media, especially the Western Media, in their representation of Africa, were constructive or continued to represent Africa through negative stereotypical images. In essence it was to find out whether news on Africa was constructive (positive), negative or neutral. Particularly, constructive or positive news referred to news that reflected the themes of “accomplishments”, “hero” and “meaningful” narratives. These themes used positive psychological techniques and allowed for holistic coverage of Ebola. Negative news on the other hand referred to news that narrated Ebola through stereotypical images of Africa to depict a sense of hopelessness. Stories that stressed on the portrayal of Africa as a place of violence, conflict, poverty, disease, and corrupt governments leading to weak systems were classified as negative stories. Furthermore, stories on Western superiority, negative cultural practices and portrayal of Africa as a single continent were also deemed negative. Neutral news referred stories that had neither the characteristics of constructive news, nor negative news. News such as announcement of events or facts without comments were deemed to be neutral. 108 University of Ghana http://ugspace.ug.edu.gh Generally, the framing of Ebola by CNN, GNN and Daily Nation, and the narrations around Africa presented a neutral representation of Africa. Over three out of four stories (77.7%) representations on Africa were neutral. There was, however, more negative news than constructive (positive news). There was only 8.5% constructive news about Africa, compared to 13.8% negative stories on Africa. This was contrary to Zhang and Matingwina (2016) who found constructive stories to have dominated the coverage. Analysing the coverage within the various media, consistent with the overall coverage, CNN, GNN and Daily Nation had most of their stories representing Africa in a neutral way. On constructive (positive) and negative representations however, Daily Nation had over one fifth of its stories (21.5%) as negative and 13.2% constructive (positive) news, CNN followed with 14.8% negative stories and 6.8% positive news, and GNN had 4.9% negative stories and 8.5% positive news. This means that it was only GNN that had constructive (positive) news as its second dominant frame in terms of representation. And unlike Zhan and Matingwina who found BBC, representing the Western Media, to have reduced their traditional approach of negative news on Africa, CNN had more negative news than their constructive news. Considering 63.5% of Daily Nation’s stories came from a non-African news agency, the Agence France-Presse (see Table 5), it could be that the representation in their stories were based on the perspective of the news agency and not Daily Nation’s, therefore, indicating that it was based on Western media ideology. On the constructive frames, CNN had 12 hero narrative stories, three accomplishment narrative stories and six meaningful narrative stories. GNN’s hero narrative stories were three, with two accomplishment narrative stories and seven meaningful narrative stories. Daily Nation’s constructive frames consisted of hero narrative (six), accomplishment narrative (two), and meaningful narrative (eight) stories. CNN has stories under all the negative story frames. As 109 University of Ghana http://ugspace.ug.edu.gh presented in Table 16, the major negative stories were reference to Africa as a place of poverty and corrupt governments which results in weak systems. Stories always pointed to the fact that the outbreak had occurred in one of the poorest regions of the African continent and Guinea for instance was referred to as one of the poorest nations in the world. On weak systems, reference was made to the health systems of Africa in general and the outbreak region in particular, indicating that they had weak health systems which were as a result of corrupt governments. This similar to Washer (2004), who found in the coverage of SARS by the UK media that there was a lot of attention on local Chinese corruption and inefficiency. There were stories which also portrayed Africa as a place of violence, conflict, disease and as a single continent. Liberia’s recent civil war was referred to in the coverage on frames that referenced Africa as a place of conflict and violence. References to the outbreak, presented it as though the outbreak was in the whole of Africa by generic reference to Africa and not to West Africa or the specific countries which were hit by the outbreak. Western superiority was also highlighted especially by reference to what the Centres for Disease Control and Prevention (CDC) of America, American soldiers and NGOs like the Doctors Without Borders, who were all foreigners, were doing in the fight against the epidemic. This was consistent with how Western biomedicine was presented as the saviour in the SARS coverage by the UK media in the findings of Washer (2004). The preparation of vaccines for the disease and reference to facilities at the CDC for containment of the disease were some major activities of Western health experts used by the media to portray Western superiority. Local health workers were presented as giving support to the foreign health workers and that the foreigners were the ones with the know-how and their intervention would help contain the disease. Some African cultural practices were also referred to as negative in some of the coverage. The cultural practices referred to were burial rites that involved the bathing of the dead person and 110 University of Ghana http://ugspace.ug.edu.gh touching the dead in the Ebola affected countries. These rites were said to have been the major modes of the spread of the disease. The cultural practices of the people and weak health systems were used as a technique of “othering” in the coverage in relation to the Ebola narration on Africa by CNN, similar to the coverage of SARS by UK media in the findings of Washer (2004). The “other” frame also contained negative news that was not captured in the frames listed. Some of the other frames made reference to African’s jungle by imputing the cause of Ebola to bush meat, fruit bats and monkeys. These findings were similar to those found by Abeysinghe (2016) in the coverage of Ebola in the US, UK and Australian print media. Even though majority of the cases of Ebola occurred in the three West African countries of Liberia, Sierra Leone and Guinea, CNN indulged in an over concentration on the few cases in America since the disease spread there through Eric Thomas Duncan. This goes to prove the disproportionate news coverage on Ebola which was skewed towards the few issues in America. This was consistent with Abeysinghe (2016) who found that coverage in the three most widely circulated print media of the USA, UK and Australia on the West Africa Ebola epidemic was dominated by the domestic politics of border control and the theme of fear such that it resulted in the shift in focus of issues surrounding the disease in the affected countries. It was also similar to Washer (2004) who found that even though China was the centre of the 2003 SARs outbreak, China and Chinese authorities were only referenced when Western scientists were helping them in surveillance on the disease. This could explain Yusuf et al.’s (2015) suggestion of disproportionate news coverage by Western media. GNN’s negative stories were on poverty, corrupt governments and systems, western superiority and “other”. Poverty, disease, corrupt governments leading to weak systems, negative cultural practices, portrayal of Africa as a single continent and “other” were the negative stories by Daily Nation. This indicates that unlike in the case of Zhang and Matingwina (2016), who 111 University of Ghana http://ugspace.ug.edu.gh found the major themes of negative stories to differ in the case of BBC and China Daily, the themes that dominated the negative news by CNN, GNN and Daily Nation were similar to some extent and bordered around poverty and corrupt governments in all the three media. From media representation studies perspective, these findings are consistent with the assertion by Ogunyemi (2011) that the hegemonic articulation of Africa in the Western Media revolves around poverty, disease and conflict. It can therefore be concluded that the representation of Africa in the coverage of the Ebola epidemic by CNN, GNN and Daily Nation was generally neutral. There were no major differences in the approach of CNN, GNN and Daily Nation in their representation of Africa in their narrations on Ebola. The use of constructive journalism approaches (positive news) in the Ebola coverage by the CNN, GNN and Daily Nation was minimal and recorded the lowest percentage of coverage on Africa in the study. Since negative news, which carried news that represented Africa through stereotypical images was found in the coverage of all the three media outlets, it can be argued that it is not only the western media that engages in stereotyping Africa in their reports. However, GNN had more constructive (positive) news on Africa than rest of the media. The CNN and Daily Nation recorded more negative than constructive (positive) news on Africa. Considering majority of Daily Nation’s stories were sourced from Agence France-Presse (AFP), a Western ideological news agency, it is suggested that the West, even though largely neutral in their representation of Africa in the Ebola coverage, had still not adopted the constructive journalism approach in the coverage on Africa since they recorded more negative news than constructive news. 6.4.1 Test of Hypothesis Two: The hypothesis tested was: H2: The kinds of media representations of Africa in Ebola stories are significantly related to media type. 112 University of Ghana http://ugspace.ug.edu.gh Media representation of Africa was found (Table 17a) to be generally neutral (77.7%). Negative stories (13.8%), however, dominated the rest of the stories than constructive (positive) stories (8.5%). Within coverage by the individual media outlets, it was found that only GNN had more (8.5%) constructive (positive) stories, about twice its 4.9% negative stories in their coverage on Africa with respect to Ebola. CNN had 14.8% negative stories, which is a little over twice the 6.8% constructive stories in their coverage. In Daily Nation’s coverage there was also more negative stories (21.5%) than constructive stories (13.2%). TABLE 17a: Media Representation (Constructive Frames) across Media MEDIA FRAMES Total CNN GNN Daily Nation Constructive 21 12 16 49 (8.5%) 6.8% 8.5% 13.2% Negative 46 7 26 79(13.8%) 14.8% 4.9% 21.5% Neutral 243 123 79 445(77.7%) 78.4% 86.6% 63.2% TOTAL 310(100%) 142(100%) 121(100%) 573(100%) This means that CNN and Daily Nation both had more negative news than constructive (positive) news. The hypothesis was therefore aimed at determining whether there was any significant relationship between media type and media representation of Africa as found in Ebola stories. The level of significance set for the study was 0.05. The null hypothesis for this test was: H0: There is no significant relationship between the kinds of media representations of Africa in Ebola stories and media type. Since there was one western media (CNN) and two African media (GNN and Daily Nation), two Chi-Square tests were done: the first one (Test A) tested the relationship between CNN and GNN, and the second (Test B) tested the relationship between CNN and Daily Nation. This was 113 University of Ghana http://ugspace.ug.edu.gh to ensure a clear picture of relationship was obtained between individual media instead of lumping the two Africa media together against one western media. Table 20: Comparing Constructive Journalism Frames in the coverage of CNN and GNN Constructive MEDIA TOTAL Journalism Frames CNN GNN Constructive 21 12 23 (Positive) 6.80% 8.50% 46 7 53 Negative 14.80% 4.90% 243 123 366 Neutral 78.40% 86.60% TOTAL 310(100%) 142 (100%) 452 X2 = 17.946 df = 3 p = 0.00 In Test A (Table 20), a Chi-square value of 17.946 was obtained, with a probability of 0.00. Since the probability value is less than the significant value of 0.05, the null hypothesis is rejected. The hypothesis (H2) is therefore significant, indicating that a significant relationship exists between the kinds of media representation Africa received in Ebola stories, and the CNN and GNN. The results in Table 20 show that both CNN and GNN were neutral in their representation of Africa. However, there were differences in their constructive and negative frames on Africa. GNN had more constructive frames on Africa than CNN. The negative stories in CNN’s coverage were about three times (14.80%) more than GNN’s negative stories (4.90%). This supports the test result that differences exist in the frames based on media. The Cramer’s V value, which measures the strength of the relationship, was 0.19. This means that even though a significant relationship exists between how Africa is represented in Ebola stories, and the CNN and GNN, this relationship is only 19% and therefore not very strong. The second test (Test B) for Hypothesis 2 was a comparison of CNN and Daily Nation. The result of this test is presented in Table 21. 114 University of Ghana http://ugspace.ug.edu.gh Table 21: Comparing Constructive Journalism Frames in the coverage of CNN and Daily Nation Constructive MEDIA TOTAL Journalism Frames CNN Daily Nation 37 Constructive 21 16 (Positive) 6.8% 13.2% 46 26 72 Negative 14.8% 21.5% 243 79 322 Neutral 78.4% 65.3% TOTAL 310(100%) 121(100%) 431 X2 = 14.090 df = 3 p = 0.003 The Chi-square for constructive journalism frames in the coverage of CNN and Daily Nation was 14.090 with a probability of 0.003. Since the probability value is less than the significant value, there is a statistically significant relationship between the constructive frames, and CNN and Daily Nation. The null hypothesis (H0) that there was no significant relationship between the kinds of media representations of Africa in Ebola stories and media types was rejected. Table 21, however shows that there was no difference in CNN and Daily Nation’s media representation of Africa in Ebola stories except the differences in frequencies and percentages. Both media outlets were largely neutral in their representation of Africa. They both also had more negative than positive stories on Africa. The strength of the relationship per the Cramer’s V value of 0.1 indicated that the relationship between the frames and media type was only 10 percent, therefore, not very strong. The findings on CNN’s media representation of Africa were consistent with Ogunyemi’s (2011) finding that western media tended to represent Africa through negative stereotypical images on poverty, disease and conflict, and Zhang and Matingwina’s (2016) finding of more negative news in BBC’s stories on Ebola, compared to China Daily’s. By the same stretch of Ogunyemi’s (2011) analysis, it would have been expected that since GNN and Daily Nation, were African, they would both have had more constructive news that negative news in their coverage, but as indicated earlier, that was not the case. Daily Nation was also found to have more negative than positive news in their 115 University of Ghana http://ugspace.ug.edu.gh representation of Africa. The fact that majority of Daily Nation’s stories (63.5%) were sourced from a western ideological news agency (AFP), could have accounted for why Daily Nation also had more negative than constructive (positive) news. In the nutshell, results for both tests showed that there were significant relationships between the kinds of media representations Africa received in Ebola stories and the media types, however, these relationships were not very strong. The findings of the study on media representation, however, suggested that GNN and Daily Nation were similar in their kinds of representation of Africa by having more negative than positive news in their coverage, while only GNN had more positive more negative news in their coverage. 6.5 Chapter Summary This chapter discussed the findings of the study with reference to theory and the literature that was reviewed for the study. The discussion was aimed at fulfilling the objectives of the study and answering the research questions raised. The dominant frames identified in the study were action, consequence and reassurance frames. The frames were not different across the media platforms that were involved in the study. Irrespective of the newsmaker involved, the frames used by the different media were similar. Ethical issues were neutral in the coverage. Media representation of Africa was largely neutral but CNN still had more negative stories on Africa than positive, indicating their continual focus on the negative stereotypical images of Africa. 116 University of Ghana http://ugspace.ug.edu.gh CHAPTER SEVEN CONCLUSION AND RECOMMENDATIONS 7.0 Introduction This study sought to identify the dominant frames in the online media coverage of the West African Ebola Virus Disease (EVD) epidemic which broke out in Guinea in December 2013. The study was further designed to identify whether the frames in the coverage differed or were the same irrespective of the newsmakers, examine any ethical issues that may emerge from the coverage and finally, investigate the kind of media representation Africa received from the CNN, GNN and Daily Nation in their coverage of the Ebola epidemic. News on Ebola between 1st March 2014 and 31st December 2014 were studied. The summary of key findings and practical implications of these findings, limitations, conclusion and recommendations for further study and industry are presented in this chapter. 7.1 Summary of Findings The findings showed that even though the online platform presents the advantage of a 24-hour cycle of news to all media irrespective of size, the GNN and Daily Nation did not publish as much news on Ebola as CNN, to the extent that the October 2014 sample of CNN news analysed alone was more than the total news sampled from both GNN and Daily Nation. It was found that October 2014 had the highest number of news items in the coverage. This was suspected to be due to the fact that it was the end of September, 2014 that cases of the disease got to the US leading to a sharp increase of about three times the coverage by CNN in October. The least number of stories were recorded in May 2014. The findings also revealed that the Ebola epidemic coverage was reported largely as straight news stories. Disease statistics, which presented statistics about Ebola cases and casualties and also the conditions of Ebola patients, dominated coverage as the main subject matter. Even though awareness creation/education was the second dominant subject matter, the dominance 117 University of Ghana http://ugspace.ug.edu.gh of disease statistics meant the media did not prioritise the severity of the outbreak and therefore did not focus more on preventive strategies. Daily Nation’s coverage revealed that half of their stories were also centred on disease statistics. Located in a country where there was no case of the disease, it would have been better they focused on preventive strategies, that is education and awareness creation about the disease, than concentrating on the cases and casualties of the disease. This was also something recommended by Paul et al. (2016) in their study of Indian media’s coverage of the Ebola epidemic. Majority of the stories in the study were found to have stated their sources and most of these sources could be verified. Most of the news was gathered and published by the media’s own journalists. It was also found that majority of Daily Nation’s stories were sourced from the Agence France-Presse (AFP) news agency. Apart from the fact that the agency had an office in Kenya, it was not clear why there was so much reliance on news from that agency alone. Furthermore, CNN was the only media found to have explored the numerous advantages of the multimedia use in their stories. CNN used videos and pictures in most of their news. Daily Nation also used some pictures and infographics in their stories, but majority of GNN’s stories were text only. Unlike in the case of Cortiñas-Rovira et al. (2015), the enhancements used by the media houses in this study reflected the stories they accompanied. Some of the videos and pictures that accompanied these stories contained information about symptoms and how to wear personal protective equipment. GNN could have contributed more to awareness creation by enhancing their stories with pictures, video or audio items that provided facts about the disease and ways to curb it. All the media could also have used more infographics than just ordinary pictures in order to create more awareness. Government officials and agencies, international/regional organisations and health workers were the major news makers in the stories analysed. This was largely because governments and 118 University of Ghana http://ugspace.ug.edu.gh their agencies, both local and foreign, international organisations such as the UN, W.H.O and ECOWAS, and health workers were very much involved in trying to find solutions to the outbreak. Majority of news in the Ebola epidemic coverage was reported as disconnected episodes (episodic) without much background information. It was only Daily Nation which had more thematic framed stories among the media outlets studied. By implication, people may not get a clear picture about some of the stories which they do not have any background or previous knowledge about and this could affect the conclusion they draw out of such stories. “Action”, “consequence” and “reassurance” frames were the most used issue-specific frames of coverage with “action” being the most dominant. Whilst the media’s audiences were provided with information about what the various stakeholders were doing to curb the disease (action), they were also provided with the devastating effects of the epidemic (consequence), consistent with the finding of disease statistics as the main subject matter of the stories. No matter how bad the consequences of the epidemic were, people were reassured that there was hope of successfully curbing the disease (reassurance). A significant relationship was found between the differences in issue-specific frames and media type. This means that the media involved, be it CNN, GNN or Daily Nation, was likely to influence the kind of issue-specific used in the coverage of the Ebola epidemic. The strength of this relationship was however found to be weak. Responsibility framing in the study revealed that majority of the stories of CNN and GNN did not attribute causal responsibility to anyone, while Daily Nation attributed causal responsibility to institution/government action. Most stories by all the three media did not attribute treatment responsibility to no one. Causal responsibility was the origin or cause of the Ebola epidemic, while treatment responsibility was focused on who was to make interventions to solve the 119 University of Ghana http://ugspace.ug.edu.gh Ebola problem. The possible explanation is that majority of the stories were on disease statistics and they only presented facts about the cases and casualties of the disease without any interpretations. When responsibility was attributed, CNN and GNN attributed both causal and treatment responsibility to government and their institutions as the cause of the epidemic and were also responsible for fixing the problem. Daily Nation also attributed treatment responsibility to government and their institutions to act on curbing the disease. Findings on media representation of Africa in the narrations on the coverage of Ebola showed that the media had been neutral in their representation of Africa. There were only a few instances of constructive journalism frames in the study, even though constructive journalism was advanced as the alternate and progressive way for media to report issues holistically on Africa from the perspective of Zhang and Matingwina (2016). The CNN has not turned away from the Western media’s practice of reporting on Africa through negative stereotypical images, a conclusion arrived at based on the fact that it had more negative than constructive stories on Africa. Daily Nation also had more negative stories about Africa than constructive stories. However, since most of Daily Nation’s stories were sourced from a non-African news agency, it was not clear if the frames were those of Daily Nation or the news agency. The kinds of media representation of Africa received in the coverage was found to be significantly related to media type. This means a specific media outlet would likely influence the kinds of media representations Africa received in the coverage of the Ebola epidemic. This association was also found to be weak. Ethical issues in the coverage on the Ebola outbreak were generally neutral, since most of the stories recorded no ethical issues. This implies that the media involved were mindful of the already devastating effect of the disease and were careful of the kind of pictures as well as videos that are attached to their stories so as not to escalate emotions through the kind of reports 120 University of Ghana http://ugspace.ug.edu.gh they gave. The media only reported full details of those whose cases marked significant milestones in the current outbreak’s history and even so, did not reveal their details until the people were tested positive for the disease. No differences were found in the frames used by the different media outlets irrespective of the newsmaker involved. The news makers were framed in a similar way, with action, consequence and reassurance frames being the dominant frames in most cases. Lastly, the dominant frames from the different media outlets did not differ much. They were all consistent with the overall findings on the issue-specific frames, having action, consequence and reassurance as the dominant frames but differed in the order of the remaining frames: conflict, new evidence, uncertainty and other frames. These differences were identified to be based possibly on differences in journalistic practices of news gathering and production. 7.2 Implications of Findings The findings generally reveal that the media in this study used the media positively to inform and create awareness about the Ebola epidemic, based on the findings on the subject matter which showed that education, awareness and action were the dominant subject matter of the stories after disease statistics. This study adds to framing literature by finding action and consequence as the dominant frames in the coverage of Ebola, consistent with Shish et al.’s (2008) study on public health epidemics which found the two frames as the most dominant. This implies that differences in media outlets and different journalistic practices did not result in differences in framing of public health epidemic hazards. It also confirms Iyengar’s (1991) finding that daily news reports were biased towards episodic frames and that the standards and norms of news organisation and production reinforce episodic framing. 121 University of Ghana http://ugspace.ug.edu.gh It also reveals that Daily Nation and CNN dwelled on the stereotypic images in their coverage on representation of Africa. Considering Daily Nation is African but most of their stories were sourced from a Western ideological news agency (Agence France-Presse), it was concluded that the Western media continues to dwell on stereotypes in their representation of Africa. 7.3 Conclusion This study generally examined the online framing of the West Africa Ebola epidemic through a content analysis of the CNN, GNN and Daily Nation. The conclusions that can be drawn from this study also help in addressing some of the questions asked in the first chapter of the work. Firstly, online framing of the epidemic revealed that the media does not differ in their framing of similar issues, based on the finding that the action, consequence and reassurance frames dominated the coverage across all the three media platforms. This indicates that media types, journalistic practices and geographical location did not result in any differences in how the media framed the disease. Secondly, media have not embraced the use of constructive journalism frames, which have been identified as contra-narrative frames on African in media representation circles since it was minimally used in the stories on the Ebola epidemic. Negative media representation and stereotyping still remains an issue in media coverage on Africa. Though it was minimal, it remains a fact that coverage of disasters on Africa always presents an opportunity for the media to revisit the stereotypes that are associated with the continent. Thirdly, based on findings about the fact that apart from disease statistics (42.9%), stories on awareness creation/education/action (31.7%) represented the second dominant subject matter of stories, it can be concluded that online media in the coverage of the Ebola epidemic had the potential of creating awareness among those who were exposed to their messages. As to if they really did create the necessary awareness, could be explored in future studies through audience studies. Finally, considering the sources of news as examined in this study, unlike in the case of MacIntyre and Travaglia (2015), the problem of the use of unregulated news sources did not occur. Most stories 122 University of Ghana http://ugspace.ug.edu.gh originated from in house journalists, with the others coming from intermedia sources, news agencies, individual experts and press releases. 7.4 Limitations The major limitation of this study was the inability of the researcher to access the editors of any of the media outlets selected for the story. Due to this, it was not possible to get the views of these editors about the editorial choices that went into decisions about how they covered the Ebola epidemic. This would have helped explain some of the findings in this study better. 7.5 Recommendations for further studies From the findings of this study, it is recommended that a mixed method study should be done on the coverage of Ebola by GNN and Daily Nation and should include interviews of editors of these mediums on their editorial policies and decisions in the coverage. This was due to the fact that majority of GNN’s Ebola stories were found to be text only and Daily Nation sourced most stories from Agence France-Presse (AFP). The qualitative part will also help to explore the coverage holistically, thereby giving more details about the coverage beyond the quantitative data provided by this study. This study examined coverage by an international (CNN), national (GNN) and African (Daily Nation) media. CNN and GNN were situated in affected countries, while Daily Nation was far removed from the outbreak zone. While, Guinea, where GNN is located, was heavily hit, the US, where CNN is located, had only a few cases. It is recommended that a study should compare coverage of media from an affected West African country and media from neighbouring West Africa countries to ascertain how different they would frame the epidemic that devastated the region for over two years. There have been studies which sought to examine how the media covered the epidemic, but there is hardly any study from the perspective of media audiences. A future study can explore audience perception about media coverage of the Ebola outbreak. 123 University of Ghana http://ugspace.ug.edu.gh 7.6 Recommendations for Industry Flowing from the first recommendation and the finding that 63.5% (73) stories of Daily Nation’s total news were sourced from a non-African news agency (AFP), it is recommended that African media organisations should rely more on primary indigenous news sources such as the African news agencies in order to avoid transferring any kinds of biases that may have been carried in the news of the foreign media houses. The media in this study especially GNN and Daily Nation did not make full use of the advantages of online media which allows for a 24-hour news cycle and the opportunity to publish as many news items as possible. Consequently, it is recommended that online media should make use of this opportunity to be able to reach more people with information about health-related issues, considering also that a lot of people are now relying on online platforms for their news. Lastly, the advantage of online platform which allows for the use multimedia must also be utilised by media houses. The findings also showed that GNN and Daily Nation did not use much multimedia in their stories. The use of infographics is recommended preferably for the purposes of awareness creation by the media. 7.7 Chapter Summary This chapter presented the summary of major findings, implications of the findings, the limitations of the study, conclusion and recommendations for both further studies and for the media industry. 124 University of Ghana http://ugspace.ug.edu.gh REFERENCES Abeysinghe, S. (2016). Third World Quarterly Ebola at the borders: newspaper representations and the politics of border control. Third World Quarterly, 37(3), 452–467. https://doi.org/10.1080/01436597.2015.1111753 Abeysinghe, S., & White, K. (2010). Framing disease: The avian influenza pandemic in Australia. Health Sociology Review, 19(3), p369-381. Afzal T. S, B. G., Bliss, N., Brown, S., and Epinoza, B. (2015). Mass Media and the Contagion of Fear: The Case of Ebola in America. PLoS ONE 10(6): e0129179. doi: 10.137/journal.pone.0129179 Baumeister, R.F., and Finkel, E.J., (Eds.) (2010). Advanced Social Psychology. New York: Oxford University Press Basch, C. H., Basch, C. E., and Redlener, I. (2014). Coverage of the Ebola Virus Disease Epidemic in Three Widely Circulated United States Newspapers: Implications for Preparedness and Prevention. Health Promotion Perspectives, 2014, 4(2), 247-251 Babbie, E. (1992). The Practice of Social Science Research (6th Ed.). Belmont: Wadsworth. Babbie, E. (2007). The Practice of Social Research (11th Ed.). Belmont: Thomson Learning. Berelson, B. (1952). Content Analysis in Communication Research. New York: Free Press. Busch C., De Maret P. S., Flynn T., Kellum R., Le, Brad Meyers S., Saunders M., White R., & Palmquist M. (2005). Content Analysis. Colorado State University. Retrieved from http://writing.colostate.edu/guides/research/content/ Brooks, D., & Hébert, L. (2006). Gender, Race, and Media Representation. In B. J. Dow & J. T. Wood (Eds.). The SAGE Handbook of Gender and Communication, doi: http://dx.doi.org/10.4135/9781412976053.n16 CDC 2016. (2017, March 20). Outbreaks Chronology: Ebola Virus Disease. Retrieved from https://www.cdc.gov/vhf/ebola/outbreaks/history/chronology.html CDC 2017. (2017, March 19). About Ebola Virus Disease. Retrieved from https://www.cdc.gov/vhf/ebola/about.html CNN (2016, July 10). New Global Survey Ranks CNN as top international news brand. Retrieved from http://cnnpressroom.blogs.cnn.com/2014/10/16/new-global-survey- ranks-cnn-as-top-international-news-brand/ Cortesi, S., & Gasser, U. (2015). Youth online and news: A phenomenological view on diversity. International Journal of Communication, 9(1), 1425–1448. Cortiñas-Rovira S., Pont-Sorribes, C., & Alonso-Marcos, F. (2015). Simulating and Dissimulating News: Spanish Media Coverage of the Swine Flu Virus. Journal of Contingencies and Crisis Management, 23(3), 159-168. 125 University of Ghana http://ugspace.ug.edu.gh Cullen, T. (2014). A tale of two epidemics: Media reporting on Ebola. The Conversation. Retrieved from https://theconversation.com/a-tale-of-two-epidemics-media-reporting- on-ebola-34803 De Vreese, C. H. (2002). Framing Europe. Television news and European integration. Amsterdam: Aksant Academic Publishers. De Vreese, C. H. (2004). The effects of frames in political television news on audience perceptions of routine political news. Journalism and Mass Communication Quarterly, 81, 36–52. De Vreese, C. H. de (2005). News framing: Theory and Typology. Information Design Journal + Document Design, 13(1), 51–62. https://doi.org/0.1075/idjdd.13.1.06vre El-Jardali F., Bou. K. L., Bawab, L., Kdouh, O., El-Sayed F., Rachidi, H., et al. (2015). Health Reporting in Print Media in Lebanon: Evidence, Quality and Role in Informing Policymaking. PLoS ONE 10(8): e0136435. Doi:10.1371/journal. pone.0136435 Entman, R. B. (1991). Framing US coverage of international news: Contrasts in narratives of the KAL and Iran air incidents. Journal of Communication, 41, 6–27. Entman, R. B. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43, 51–58. Freelon, D., (nd). ReCal2: Reliability for 2 Coders. Retrieved from http://dfreelon.org/utils/recalfront/recal2/ Frewer L, Miles S, & Marsh, R. (2002). The media and genetically modified foods: evidence in support of social amplification of risk. Risk Anal 22:701–711 Frieden, Thomas, R., Inger, D., Beth P. B., M.D., Kenyon, T., & Nichol, S., (2014). Ebola 2014 — New Challenges, New Global Response and Responsibility. The New England Journal of Medicine. 317(13), 1177-1180. Global News Network. (2016, July 10). Retrieved from http://newspaperhunt.com/papers/Liberia/global-news-network.php Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear power: A constructionist approach. American Journal of Sociology, 95, 1–37. Holsti, Ol. R. (1969). “Content Analysis.” Content Analysis for the Social Sciences and Humanities. Boston: Addison-Wesley, 597–692. Iyengar, S. (1991). Is anyone responsible? How television frames political issues. Chicago, IL: University of Chicago Press. Kasperson, R. E., Renn, O., Slovic, P., Brown, H. S., Emel, J., Goble, R., Kasperson, J. X. and Ratick, S. (1988), The Social Amplification of Risk: A Conceptual Framework. Risk Analysis, 8: 177–187. doi:10.1111/j.1539-6924.1988.tb01168.x 126 University of Ghana http://ugspace.ug.edu.gh Kenyan Media company linked to Kenyatta offers free daily paper. (2016, July 10). Retrieved from http://www.bloomberg.com/news/articles/2014-07-01/kenyan-media-company- linked-to-kenyatta-offers-free-daily-paper Kiousis, S. (2011). Agenda-setting and attitudes: exploring the impact of media salience on perceived salience and public attitude strength of us presidential candidates from 1984 to 2004. Journalism Studies, 12(3), 359-374. Krippendorff, K. H. (2004). Content Analysis: An Introduction to Its Methodology. (2nd Ed.) Thousand Oaks California: SAGE Publications Inc. Lenhart, A. (2015). Teens, social media and technology overview 2015: Smartphones facilitate shifts in communication landscape for teens. Pew Research Center, (April), 1–47. https://doi.org/10.1016/j.chb.2015.08.026 Littlejohn, S. W., & Foss, K. A. (2009). Encyclopaedia of Communication Theory. https://doi.org/10.1007/s13398-014-0173-7.2 MacIntyre, C. R., & Travaglia, J. F. (2015). Heightened Vulnerability, Reduced Oversight, and Ethical Breaches on the Internet in the West African Ebola Epidemic. The American Journal of Bioethics, 15(4), 65-68, DOI: 10.1080/15265161.2015.1010017 Macnamara, J. (2005). Media Content Analysis: Its Uses, Benefits and Best Practice Methodology, Asia Pacific Public Relations Journal. 6(1), 1-34. McCombs, M. (2005). A look at agenda-setting: Past, present and future. Journalism studies, 6(4), 543-557. McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public opinion quarterly, 36(2), 176-187. McQuail, D. (2005). McQuail’s Mass Communication Theory. (5th Ed). Sage Publications: London. McQuail, D. (2010). Mass Communication Theory (6th Ed.). London: Sage. Nitz, M., & Ihlen, Ø. (2009). Transatlantic perspectives on the U.S. 2004 election: The case of Norway. Nordicom Review, 30(1), 113–129. Nwanne, Ben U. (2014). Media Reportage of the Ebola Crisis: Lessons from Nigeria. Global Journal of Arts Humanities and Social Sciences, 2(10), pp.30-38 Obukoadata, P. O., & Abuah, F. A. (2014). Media Surveillance Function Within the Context of the Ebola Outbreak in Nigeria: Influences and Perceptual Frames. International Journal of Scientific Research and Innovative Technology, 1(5), 53–66. Retrieved from http://www.ijsrit.com/uploaded_all_files/2097007039_c4.pdf Ogunyemi, O. (2011). Representation of Africa Online: Sourcing Practice and Frames of Reference. Journal of Black Studies, 42(3), 457– 478 127 University of Ghana http://ugspace.ug.edu.gh Pathak R., Poudel, D.R., Karmacharya, P., Pathak, A., Aryal, M.R., Mahmood, M., et al. (2015). YouTube as a source of information on Ebola virus disease. North Am J Med Sci., 7, 306-9. Paul, S., Mahajan, P. B., & Sahoo, J. (2016). Global killer Ebola: How was it handled by the Indian press? Indian Journal of Community Health, 28(2), 125–131. Polit, D., Beck, C. & Hungler, B. (2001). Essentials of Nursing Research: Methods, Appraisal and Utilisation (5th Ed.). Philadelphia: JB Lippincott Polit, D.F. & Hungler, B.P. (1999). Nursing Research: Principles and Methods (6th Ed.). Philadelphia: JB Lippincott Riffe, D, Aust, C., & Lacy, S. (1993). The effectiveness of random, consecutive, and constructed week sampling in newspaper content analysis. Journalism Quarterly. 70(1), 133-139 Roberts, M., & McCombs, M. (1994). Agenda setting and political advertising: Origins of the news agenda. Political Communication, 11, 249–262. Salman, A., & Saad, S. (2015). Online Political Participation: A Study of Youth Usage of New Media. Mediterranean Journal of Social Sciences, 6(4). https://doi.org/10.5901/mjss.2015.v6n4s3p88 Shah, D., Watts, M. D., Domke, D., & Fan, D. (2002). News framing and cueing of issue regimes. Explaining Clinton’s public approval in spite of scandal. Public Opinion Quarterly, 66, 339–370. Shih T., Wijaya, R., & Brossard, D., (2008). Media Coverage of Public Health Epidemics: Linking Framing and Issue Attention Cycle toward an Integrated Theory of Print News Coverage of Epidemics, Mass Communication and Society, 11(2), 141-160 Shoemaker, P. J., & Reese, S. D. (1996). Mediating the Message: Theories of Influences on Mass Media Content. https://doi.org/10.1177/1326365X14540245 Smith, S. S., & Smith, S. S. (2016). Media coverage of the Ebola virus disease in four widely circulated Nigerian newspapers: lessons from Nigeria. Health Promotion Perspectives, 6(2), 92–5. https://doi.org/10.15171/hpp.2016.16 Tankard, J. W. (2001). The empirical approach to the study of media framing. in S. D. Reese, O. H. Gandy & A. E. Grant (Eds.), Framing public life (pp. 95–106). Mahwah, NJ: Lawrence Erlbaum. The Nation Media Group. (2016, July 10). Who we are. Retrieved from http://www.nationmedia.com/who-we-are/ Thussu, D. K. (2003). Live TV and bloodless deaths: war, infotainment and 24/7 news. Dies / Freedman, 117-132. 128 University of Ghana http://ugspace.ug.edu.gh Washer, P. (2004). Representations of SARS in the British newspapers. Social Science and Medicine, 59(12), 2561–2571. https://doi.org/10.1016/j.socscimed.2004.03.038 Wilson, J., Shoichet, C. E., and Yan, H. (2014, October 3). Liberia may prosecute Ebola victim. Retrieved from http://edition.cnn.com/2014/10/02/health/ebola-us/index.html Wimmer, R. D., & Dominick, J. R. (2003). Mass media research: An Introduction (7th Ed.). Belmont, CA: Thompson-Wadsworth. Wimmer, R. D., & Dominick, J. R. (2006). Mass media research: An Introduction (8th Ed.). Belmont, CA: Thompson-Wadsworth. Wimmer, R. D., & Dominick, J. R. (2011). Mass media research: An introduction (9th Ed.). Belmont, CA: Thompson Wadsworth Wittenbrink, B., Judd, C. M., & Park, B. (1997). Evidence for racial prejudice at the implicit level and its relationship with questionnaire measures. Journal of Personality and Social Psychology, 72, 262–274. World Health Organisation. (2016, July 13). Statement on the 1st meeting of the IHR Emergency Committee on the 2014 Ebola outbreak in West Africa. Retrieved from http://www.who.int/mediacentre/news/statements/2014/ebola-20140808/en/ World Health Organisation. (2016a, May 30). Ebola Situation Report - 30 March 2016. Retrieved from http://apps.who.int/ebola/current-situation/ebola-situation-report-30- march-2016. World Health Organisation. (2016b, March 29). Statement on the 9th meeting of the IHR Emergency Committee regarding the Ebola outbreak in West Africa. Retrieved from http://www.who.int/mediacentre/news/statements/2016/end-of-ebola-pheic/en/ World Health Organisation. (2017, March 20). About Ebola. Retrieved from http://www.who.int/mediacentre/factsheets/fs103/en/ Young M.E., Norman, G.R., & Humphreys K.R. (2008). Medicine in the Popular Press: The Influence of the Media on Perceptions of Disease. PLoS ONE 3(10): e3552. doi: 10.1371/journal.pone.0003552 Yusuf, I., Yahaya, S., Qabli, S. (2015). Role of Media in Portraying Ebola in and outside Africa. Journal of Tropical Diseases & Public Health 3(152). DOI:10.4172/2329-891X.1000152 Zhang, Y., & Matingwina, S. (2016). A new representation of Africa? The use of constructive journalism in the narration of Ebola by China Daily and the BBC. African Journalism Studies, 37(3), 19–40. https://doi.org/10.1080/23743670.2016.1209224 Zhang, Y., & Wildemuth, B. M. (2009). Qualitative Analysis of Content. In B., Wildemuth (Ed). Applications of Social Research Methods to Questions in Information and Library Science, 222-231. Westport, CT: Libraries Unlimited. 129 University of Ghana http://ugspace.ug.edu.gh APPENDIX 1 CODING GUIDE FOR EBOLA NEWS COVERAGE 1. Media 1. Cable News Network (CNN) 2. Global News Network (GNN, Liberia) 3. Daily Nation (Kenya) 2. Date of Story ------------------------------------------------------------------------ 3. Story Headline --------------------------------------------------------------------- 4. Type of Story 1. Straight news story 2. Features 3. Editorial 4. Opinion 5. Other 5. Subject matter of stories 1. Risk / Threat – stories that emphasize on the disease as a risk / threat 2. Action / Awareness Creation / Education – about the disease 3. Skills Training / Capacity Building – stories on skills training / capacity building on the disease e.g. Nurses, Doctors, health workers 4. Donation / Funding – fundraising and donation towards fighting the diseases 5. Disease Statistics – stories about the new figures, facts, disease status, patient status, successes, impact of the disease, 6. Other 6. Source of Stories 1. Press Release 2. News Agency AFRICAN (GNA etc) 3. News Agency NON-AFRICAN (AP, AFP, Reuters etc) 4. Intermedia – stories from another media house which is not an agency (e.g. BBC taking a story from CNN) 5. Journalist Initiated / correspondent – from programmes, press conference, seminar, workshop, donation etc 6. Individuals – private people’s opinions, letters 7. No Source Stated – stories with no source indicated 8. Other 7. Story Enhancement 1. Text only – a full text story with no enhancement 2. Video with text – a full video story with anchor lead in 3. Pictures Stories – Photo stories (series of photos with descriptions) 130 University of Ghana http://ugspace.ug.edu.gh 4. Story with video – A full text story with video imbedded 5. Story with audio – a full text story with audio imbedded 6. Story with picture – a full text story with pictures 8. Does the enhancement reflect the story? 1. Fully – enhancement fully reflects / supports the story 2. Partially - enhancement partially reflects / supports the story 3. No - enhancement does not reflect / support the story 4. No enhancement – refers to text only stories 9. News Makers (Story Actors) 1. Victim / Affected African 2. Victim / Affected Non-African 3. Health Workers – Doctors, Nurses 4. International / Regional Organisations – e.g. UN, WHO, World Bank 5. Pressure / Advocacy Group & Organisations / NGOs – Doctors Without Borders 6. Government Officials / Agencies – Presidents, Ministers etc 7. Individual / Expert Opinions – Disease experts / private individuals 8. Religious / Traditional Leaders – chiefs, reverend ministers 9. Others 10. Episodic and Thematic Framing 1. Episodic – story of events with little background information 2. Thematic – stories situated in a context with enough background information 11. Issue-Specific frames (Code all that apply) 1. Action – action(s) against the disease 2. Conflict – differences in opinions, arguments/disagreements, antagonism between opposing opinions 3. Consequences – impact of the disease (economic, political, human lives, environmental, social) 4. New Evidence – new findings / results, discovery of new methods to prevent / cure / treat the disease 5. Reassurance – assurances of success against the disease, hope, allaying fears of the public, successes in the fight against the disease 6. Uncertainty – uncertainties about the disease (cause, cure, possible spread, portrayal as unknown), difficulties in containing the disease 7. Other 12. Causal Responsibility Framing 1. Individual Responsibility – individual’s actions as causes/spread of the disease 2. Institutional / Government Responsibility– government and other institutions actions as cause / spread of the disease 3. None – no one / nothing is identified as responsible for causing the disease 4. Other 131 University of Ghana http://ugspace.ug.edu.gh 13. Treatment Responsibility Framing (Code all that apply) 1. Individual Responsibility / Attitudinal change – individuals are the ones to solve the problem 2. Institutional / Governmental Action – government and institutions are responsible for solving the problem 3. International Donor Intervention – donors and international community are responsible for solving the problem. 4. Global Action – both local and international help needed to solve problem. 5. None – no one / nothing is identified as responsible for solving the problem 14. CONSTRUCTIVE JOURNALISM FRAMES (Media representation of Africa / Reference / description to/of Africa) A. Constructive 1. Hero Narrative - focuses on people who overcame numerous obstacles to change their own circumstances and those of the people around them 2. Accomplishment Narrative - highlight a problem and then narrate people’s journey towards finding solutions 3. Meaningful Narrative - interrogates whether lessons were learned from problems which might be life-changing 4. Negative - stories that narrated the Ebola story through stereotypical images of Africa to depict a sense of hopelessness, focused on the theme of fear 5. Violence – Africa portrayed as a place of violence 6. Conflict - Africa portrayed as a place of conflict 7. Poverty - Africa portrayed as a place of poverty 8. Disease - Africa portrayed as a place of disease 9. Corruption (in African Governments) / Weak Systems - Africa portrayed as a place of weak systems and corrupt governments. 10. Western Superiority – Western experts / doctors and institutions portrayed as superior to African ones and the ones who can solve the problem 11. Negative Cultural Practices – Reference to African cultural practices as negative Africa as a single country – Africa portrayed as a single country 12. ‘Othering’ – presenting Africans as different from others hence the Ebola outbreak 13. Other B. Neutral – Stories which lacked the characteristics of constructive and negative news e.g. announcements of events; facts without comments 14. Neutral 15. Ethical issues in the coverage of Ebola 1. Scare mongering – stories that cause worry, fear, panic 2. Sensationalism – shocking stories / language 3. Breach of patient’s privacy – use of patient details: name, picture etc. 4. Gory images / videos 5. Other 6. No ethical issue 132 University of Ghana http://ugspace.ug.edu.gh APPENDIX II CODING SHEET CODING SHEET S/N Q1 Q2(D-M-Y) Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 12 Q13 Q14 Q15 134 University of Ghana http://ugspace.ug.edu.gh APPENDIX III COMPOSITE WEEK ANALYSED FOR THE EBOLA NEWS COVERAGE STUDY CNN (1ST MARCH – 31ST DECEMBER 2014) MONTH SUN MON TUES WED THURS FRI SAT March 30 24 25 No story 27 No Story 22 April No Story 14 1 23 17 11 12 May No Story No story No story No story No story No story No Story June 23 26 July 28 16 31 4 26 August 17 25 19 27 28 8 23 September 7 22 16 3 18 5 6 October 5 6 28 15 30 3 4 November 9 17 11 12 6 21 29 December No Story 8 2 10 No Story 5 No Story GNN (1ST MARCH – 31ST DECEMBER 2014) MONTH SUN MON TUES WED THURS FRI SAT March No Story 24 No Story 26 27 28 No Story April No Story No Story 8 9 24 No Story 5 May 4 No Story No Story 28 No Story No Story No Story June No Story 16 17 No Story 12 No Story 14 July 27 21 29 23 31 25 26 August 17 25 19 27 28 8 23 September 7 22 16 3 18 5 13 October 5 6 28 1 30 3 4 November 16 17 11 12 06 21 22 December 7 8 2 17 25 5 6 135 University of Ghana http://ugspace.ug.edu.gh DAILY NATION (1ST MARCH – 31ST DECEMBER 2014) MONTH SUN MON TUES WED THU FRI SAT March 23 31 25 No Story 27 28 22 April 20 14 15 No Story 3 11 No Story May No Story No Story 27 14 15 No Story No Story June 22 No Story 24 18 19 13 28 July 27 28 29 30 24 25 26 August 17 25 19 27 28 8 23 September 7 22 16 10 18 5 13 October 5 6 28 15 30 3 4 November 23 No Story 11 12 No Story 7 8 December 7 8 2 10 25 5 6 136 University of Ghana http://ugspace.ug.edu.gh APPENDIX IV COMPARING NEWS MAKERS AND ISSUE-SPECIFIC FRAMES ACROSS MEDIA NAM FRAMES E OF New MEDI NEWS Actio Confli Consequen Evidenc Reassuran Uncertain Tota A MAKERS n ct ce e ce ty Other l CNN Victim / 36 4 25 3 11 1 9 89 Affected - 40.4 10.1 AFRICAN 4.5% 28.1% 3.4% 12.4% 1.1% % % Victim / 46 5 33 1 9 4 13 111 Affected- Non- 41.4 11.74.5% 29.7% .9% 8.1% 3.6% African % % Health 56 10 35 2 11 5 3 122 Workers 45.9 8.2% 28.7% 1.6% 9.0% 4.1% 2.5% % Internationa 37 2 28 5 9 2 2 85 l / Regional Organizatio 43.5 2.4% 32.9% 5.9% 10.6% 2.4% 2.4% ns % Pressure / 6 0 2 0 2 1 0 11 Advocacy Groups / 54.5 0.0% 18.2% 0.0% 18.2% 9.1% 0.0% NGOs % Government 74 16 32 1 23 8 9 163 Officials / 45.4 Agencies 9.8% 19.6% .6% 14.1% 4.9% 5.5% % Individual / 42 16 13 2 4 5 14 96 Expert 43.8 14.6 Opinions 16.7% 13.5% 2.1% 4.2% 5.2% % % Religious / 0 0 1 0 0 0 0 1 Traditional Leaders 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0% Other 10 1 5 1 1 2 3 23 43.5 13.0 4.3% 21.7% 4.3% 4.3% 8.7% % % TOTAL 307 54 174 15 70 28 53 701 GNN Victim / 13 0 15 1 7 0 1 37 Affected - 35.1 AFRICAN 0.0% 40.5% 2.7% 18.9% 0.0% 2.7% % Victim / 5 0 8 0 1 0 1 15 Affected- Non- 33.3 0.0% 53.3% 0.0% 6.7% 0.0% 6.7% African % Health 14 0 11 0 8 0 3 36 Workers 38.9 0.0% 30.6% 0.0% 22.2% 0.0% 8.3% % Internationa 40 3 14 0 22 0 6 85 l / Regional Organizatio 47.1 3.5% 16.5% 0.0% 25.9% 0.0% 7.1% ns % 16 0 7 1 5 0 1 30 137 University of Ghana http://ugspace.ug.edu.gh Pressure / Advocacy 53.3 0.0% 23.3% 3.3% 16.7% 0.0% 3.3% Groups / % NGOs Government 70 4 27 4 30 0 6 141 Officials / 49.6 Agencies 2.8% 19.1% 2.8% 21.3% 0.0% 4.3% % Individual / 15 1 5 0 3 0 6 30 Expert 50.0 20.0 Opinions 3.3% 16.7% 0.0% 10.0% 0.0% % % Religious / 3 0 0 0 0 0 1 4 Traditional 75.0 25.0 Leaders 0.0% 0.0% 0.0% 0.0% 0.0% % % Other 0 0 1 0 1 0 2 4 50.0 0.0% 0.0% 25.0% 0.0% 25.0% 0.0% % TOTAL 176 8 88 6 77 0 27 382 Daily Victim / 16 0 12 0 7 0 1 36 Nation Affected - 44.4 AFRICAN 0.0% 33.3% 0.0% 19.4% 0.0% 2.8% % Victim / 11 1 8 0 6 1 0 27 Affected- Non- 40.7 3.7% 29.6% 0.0% 22.2% 3.7% 0.0% African % Health 22 1 19 1 9 2 0 54 Workers 40.7 1.9% 35.2% 1.9% 16.7% 3.7% 0.0% % Internationa 44 4 28 2 13 6 4 101 l / Regional Organizatio 43.6 4.0% 27.7% 2.0% 12.9% 5.9% 4.0% ns % Pressure / 15 3 13 0 5 4 0 40 Advocacy Groups / 37.5 7.5% 32.5% 0.0% 12.5% 10.0% 0.0% NGOs % Government 49 4 34 3 13 4 3 110 Officials / 44.5 Agencies 3.6% 30.9% 2.7% 11.8% 3.6% 2.7% % Individual / 17 2 8 1 3 1 3 35 Expert 48.6 Opinions 5.7% 22.9% 2.9% 8.6% 2.9% 8.6% % Religious / 1 0 0 0 0 0 1 2 Traditional 50.0 50.0 Leaders 0.0% 0.0% 0.0% 0.0% 0.0% % % Other 4 0 4 0 0 0 3 11 36.4 27.3 0.0% 36.4% 0.0% 0.0% 0.0% % % TOTAL 179 15 126 7 56 18 15 416 Percentages and totals are based on responses. 138