Citation
News Coverage and Online Discussions about the Middle East

Material Information

Title:
News Coverage and Online Discussions about the Middle East The Spiral of Silence in the Agenda-Setting Process
Creator:
Alkazemi, Mariam F
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (163 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Mass Communication
Journalism and Communications
Committee Chair:
WANTA,WAYNE M
Committee Co-Chair:
ARMSTRONG,CORY L
Committee Members:
KIOUSIS,SPIRO K
WOODS,PATRICIA JANICE
Graduation Date:
12/19/2014

Subjects

Subjects / Keywords:
Google ( jstor )
Islam ( jstor )
Journalism ( jstor )
Muslims ( jstor )
News media ( jstor )
Newspapers ( jstor )
Public opinion ( jstor )
Religion ( jstor )
Social media ( jstor )
Standard error ( jstor )
Journalism and Communications -- Dissertations, Academic -- UF
agenda-cutting -- agenda-setting -- newspapers -- social-media -- spiral-of-silence
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Mass Communication thesis, Ph.D.

Notes

Abstract:
The agenda setting effect describes the process by which citizens learn and prioritize the important issues of the day through exposure to news. The more salient a news topic in the media agenda, the more likely individuals will think it is important. The spiral of silence effect describes the process by which citizens gauge the majority opinion on issues prior to calculating the degree to which their opinion is congruent. This calculation helps individuals decide to voice or silence their views. This dissertation explores the agenda silencing effect, which explores the issues and tone of the traditional and social media agendas in two countries: the United States and Kuwait. Four social media accounts and four newspapers were selected in each country. A content analysis was employed to analyze the messages for tones and issues. The samples consisted of 2, 236 graphs from Kuwaiti newspapers and 6,089 graphs observations from U.S. newspapers. The selected social media accounts include those of politically outspoken personalities from the United States (N=136) and Kuwait (N=627). The results demonstrate that the agendas of social and traditional media can be similar when discussing some topics. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: WANTA,WAYNE M.
Local:
Co-adviser: ARMSTRONG,CORY L.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-12-31
Statement of Responsibility:
by Mariam F Alkazemi.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
12/31/2016
Resource Identifier:
974007263 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

NEWS COVERAGE AND ONLINE DISCUSSIONS ABOUT THE MIDDLE EAST: THE SPIRAL OF SILENCE IN THE AGENDA SETTING PROCESS By MARIAM FAISAL ALKAZEMI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

© 2014 Mariam Faisal Alkazemi

PAGE 3

To my biological and intellectual parents: Thank you for giving me a chance. To my siblings and friends: Thank you for helping me enjoy the ride. To the Muslim community and Islamic faith: Thank you for the ongoing inspiration.

PAGE 4

4 ACKNOWLEDGMENTS First, I would like to express my thanks to my wonderful committee members: Dr. Wayne Wanta, Dr. Spi ro Kiousis, Dr. Cory Armstrong and Dr. Patricia Woods. to thank Associate Dean Debbie Treise for her constant support, encouragement and concern for Kenneth Wald. Sec doctoral program was my decision and for supplying me with a list of jokes about the Ph.D. degree since I was old enough to realize that people called my father doctor. T May, Ghazi, Dalal, Saud, Bader and Ali are amazing people and I am lucky to be related. I am also extremely fortunate to have long term friends, including the Surles family, the Hostetler family, the Ma kover family, the Al Kharusi family, the Saymeh family, the Al Ibri family, the Gholami family and the Weber family. regularly. gle person who has ever volunteered to help me with inter coder reliability. Thank you all for putting up with me. These people include: Uncle Adnan, Hatem, At t a, Samer, Mahdi, Hadi, Sara, Yasmeen, Yousef and many, many more. Your help has been crucial to my growth. Thank you for your service. Finally, I would like to say a special thank you to two people who tolerate my incessant questions and peculiarities : m y friend, Sara , and my wonderful mentor, Dr. Wayne Wanta, whose patience is inspirational. Most people I had asked to be my mentor in my teenage years refused the offer, so I never imagined anyone would want to mentor me. Knowing you has been transformative. Thank you, Lord Wanta.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURE S ................................ ................................ ................................ ......................... 9 ABSTRACT ................................ ................................ ................................ ................................ ... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 12 Press Freedoms around the World ................................ ................................ .......................... 13 ................................ ................................ ................................ ......... 16 Medi a Ethics ................................ ................................ ................................ ........................... 18 Role of Media in International Conflict ................................ ................................ .................. 21 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 23 Agenda S ilencing ................................ ................................ ................................ .................... 23 Agenda Setting ................................ ................................ ................................ ........................ 24 Spiral of Silence ................................ ................................ ................................ ...................... 26 Comparing and Contrasting the Two Theories ................................ ................................ ....... 28 Levels of Analysis ................................ ................................ ................................ ........... 28 Definitions of Public Opinion ................................ ................................ ......................... 30 Account for Cognitive and Affective Processes ................................ .............................. 32 Application to Online Media ................................ ................................ ........................... 32 Differences by Issues ................................ ................................ ................................ ....... 33 Application Across Cultures ................................ ................................ ............................ 33 International Media ................................ ................................ ................................ ................. 34 Online Media ................................ ................................ ................................ .......................... 36 3 HYPOTHESES AND RESEARCH QUESTIONS ................................ ................................ 43 4 METHODS ................................ ................................ ................................ ............................. 47 Sample ................................ ................................ ................................ ................................ .... 49 Traditional Media ................................ ................................ ................................ ................... 49 Social Media ................................ ................................ ................................ ........................... 52 Coding Categories ................................ ................................ ................................ .................. 53 Generalizability ................................ ................................ ................................ ....................... 56 Issues ................................ ................................ ................................ ............................... 57 Nations ................................ ................................ ................................ ............................. 60

PAGE 6

6 Religions ................................ ................................ ................................ .......................... 65 Interpretation ................................ ................................ ................................ ................... 67 Inter Coder Reliability ................................ ................................ ................................ ............ 68 5 RESULTS ................................ ................................ ................................ ............................... 69 Results of Hypotheses ................................ ................................ ................................ ............. 69 Result s of Research Questions ................................ ................................ ................................ 86 6 DISCUSSION ................................ ................................ ................................ ....................... 118 Implications ................................ ................................ ................................ .......................... 128 Limitations ................................ ................................ ................................ ............................ 130 Future R esearch ................................ ................................ ................................ .................... 132 APPENDIX A INTER CODER RELIABILITY ................................ ................................ .......................... 133 B CODESHEET ................................ ................................ ................................ ....................... 145 LIST OF REFERENC ES ................................ ................................ ................................ ............. 156 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 163

PAGE 7

7 LIST OF TABLES Table page 5 1 Frequencies of appearance of issues in newspapers in the United States vs. Kuwait ....... 89 5 2 Frequencies of appearance of nations in newspapers in the United States vs. Kuwait ..... 90 5 3 Frequencies of appearance of religions in newspapers in the United States vs. Kuwait ... 91 5 4 Fr equencies of appearance of issues unique to Kuwaiti newspapers ................................ 91 5 5 Frequencies of appearance of issues unique to United States newspapers ........................ 92 5 6 Frequencies of Issues in Social Media in the USA vs. Kuwait ................................ ......... 92 5 7 Frequencies of Nations in Social Media in the USA vs. Kuwait ................................ ....... 93 5 8 Frequencies of Religions in Social Media in the USA vs. Kuwait ................................ .... 94 5 9 Frequency of Issues ................................ ..................... 95 5 10 Frequency of Issues Which Appear Only in U.S. Social Media ................................ ........ 95 5 11 Frequencies and Ranking of Issues According to Medium ................................ ............... 96 5 12 Frequencies and Ranking of Nations According to Medium ................................ ............. 97 5 13 Frequencies and Ranking of Religions According to Medium ................................ .......... 99 5 14 Order Correlation of Media for Issues. ................................ .............. 100 5 15 Order Correlation of Media for Nations. ................................ ............ 100 5 16 Order Correlation of Media for Religions. ................................ ......... 101 5 17 Frequency of Issues Per Ton e in Kuwaiti Newspapers ................................ ................... 102 5 18 Frequency of Nations Per Tone in Kuwaiti Newspapers ................................ ................. 103 5 19 Frequency of Religion Per Tone in Kuwaiti Newspaper ................................ ................. 105 5 20 Frequency of Unique Issues Per Tone in Kuwaiti Newspaper ................................ ........ 106 5 21 Frequency of Issues Per Ton e in Kuwaiti Social Media ................................ .................. 107 5 22 Frequency of Nations Per Tone in Kuwaiti Social Media ................................ ............... 108 5 23 Frequency of Religions Per Tone in Kuwaiti Social Media ................................ ............ 109

PAGE 8

8 5 24 Frequency of Unique Issues Per Tone in Kuwaiti Social Media ................................ ..... 110 5 25 Negative Tones of Issues by Media in Kuwait ................................ ................................ 111 5 26 Negative Tones of Nations by Media in Kuwait ................................ ............................. 112 5 27 Negative Tones of Religions by Media in Kuwait ................................ ........................... 113 5 28 Negative Tones of Unique Issues by Media in Kuwait ................................ ................... 114 5 29 Frequencies of the Objects per Tone in U.S. Social Media ................................ ............. 115 5 30 Tone Per Object in U.S. Newspap ers ................................ ................................ ............... 116 5 31 Negative Tones of Objects by Media in the United States ................................ .............. 117 6 1 Results of Predicted Relationships ................................ ................................ ................... 120 A 1 Inter Coder Reliability Results for Variables in the U.S. Social Media .......................... 133 A 2 Inter Coder Reliability Results fo r Variables in the Kuwaiti Social Media Account Coded by Three People ................................ ................................ ................................ .... 136 A 3 Inter coder Reliability Results for Variables in the Kuwaiti Social Media Account Coded by Three People ................................ ................................ ................................ .... 140

PAGE 9

9 LIST OF FIGURES Figure page 2 1 Agenda Setting Model ................................ ................................ ................................ ....... 42 2 2 Spiral of Silence Model ................................ ................................ ................................ ..... 42

PAGE 10

10 Abstract of Dissertation P resented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy NEWS COVERAGE AND ONLINE DISCUSSIONS ABOUT THE MIDDLE EAST: THE SPIRAL OF SILENCE IN THE AGENDA SETTING PROCESS By Mariam Faisal Alkazemi December 2014 Chair: Wayne Wanta Major: Mass Communication The agenda setting effect describes the process by which citizens learn and prioritize the important issues of the day through exposure to news. The more salient a news topic in the media agenda, the more likely individuals will think it is important. The spiral of silence effect describes the process by which citizens gauge the majority opinion on issues prior to calculating the degree to which their opinion is congruent. This calculation helps individuals decide to voice or silence their views. This dissertation explores the agenda silencing effect, which explores t he issues and tone of the traditional and social media agendas in two countries: the United States and Kuwait. Four social media accounts and four newspapers were selected in each country. A content analysis was employed to analyze the messages for tones and issues. The samples consisted of 2, 236 graphs from Kuwaiti newspapers and 6,089 graphs observations from U.S. newspapers. The selected social media accounts include those of politically outspoken personalities from the United States (N=136) and Kuwait (N=627). The results demonstrate that the degree of salience in newspapers can determine its degree of salience in social media. The

PAGE 11

11 results also show attribute diversity, where various issues in the media may be associated with multiple, and som etimes contradicting, attributes.

PAGE 12

12 CHAPTER 1 INTRODUCTION In the 1970s several influential theories of mass media were developed. Among them, were Noelle effect. Over time, th ese theories have been analyzed and mass communications scientists have been able to determine the contingent conditions in which the theories operate (e.g., Lasorsa, 1991; Wanta & Wu, 1992). However, the theories have largely been considered as two separ ate theories of mass communications. The agenda setting and spiral of silence theories have many similarities. For example, both explain the effect of the mass media. In the case of agenda setting, the theory would use the mass media to explain reasons t o discuss an idea ( Golan & Wanta, 2001) . In the case of the spiral of silence, the theory would use the mass media to explain self censorship (Hayes, 2007). While agenda setting requires a content analysis and survey component, the spiral of silence has largely relied on the survey data (Donsbach & Stevenson, 1984). The current study will apply a content analysis of traditional and online media to explore which conditions would result in the agenda setting effect and which would result in the spiral of s ilence effect. Further, the content analyzed data will be gathered from English language newspapers in the United States and Arabic language newspapers from Kuwait that report on Middle Eastern politics. A comparison of the issues reported in Kuwait and the United States will contribute to the concept of agenda silencing , which is the result of blurring the boundaries of agenda setting and the spiral of silence theories. By comparing media coverage of Middle Eastern politics, the current study will meas ure which issues are silenced in the U.S. media. The media agenda in Kuwait is interesting for several reasons. Kuwait is an Arabian Gulf state with a population of fewer than four million people of which foreign nationals

PAGE 13

13 outnumber native citizens. The Kuwaiti media industry was established by Levantine and Iraqi journalists under the encouragement of Kuwaiti politicians, and Kuwaiti media serves its international community. Further, Kuwait is the Arabian Gulf state with the most media freedoms accordin g to Freedom House (2012). Legal reforms have both enhanced and restricted the media in Kuwait as the development of democratic institutions have developed in Kuwait; yet, Kuwait is largely considered to be the most democratic of the Gulf States. Furthe r, Kuwait is one of the strongest allies to the United States in the Arab world. The importance of agenda silencing with regards to U.S. coverage of foreign news cannot be underestimated. Evidence suggests that the U.S. public is likely to consider a na tion to be important to U.S. interests as it receives more news coverage, and the public opinion is likely to be negative if the news coverage of the nation is negative (Wanta, Golan & Lee, 2004). Yet, both quality and quantity of foreign news in the United States is lacking because journalists often lack historical, cultural, religious, or geographical context, media are unable to show the impact of international developments on the lives of U.S. citizens, and/or media allocate limited space and reso urces to foreign news (Fahmy, 2007). In a democratic society , public opinion formation is vital to U.S. policy and theories such as agenda setting and the spiral of silence demonstrate the importance of the media on public opinion formation. Yet, U.S. me dia coverage of foreign news sometimes presents a perspective that differs from international perspectives (Fahmy, 2007). The purpose of this paper is to explore the degree to which international perspectives are silenced in the U.S. media. Press Freedoms aro und the World Freedom House (2013 ) is an organization which rates nations according to the press freedoms they afford. challenges to press freedoms presented themselves. For examp le, counterterrorism policies

PAGE 14

14 make it increasingly difficult for journalists to gain access to some information (Freedom, 2013). While the reports suggest that some journalists were detained or arrested for covering the Occupy Wall Street movement in earl y 2012, there are many laws which shield journalists from revealing sources (Freedom, 2013). While many states consider libel to be a criminal violation of the law, Colorado repealed criminal libel laws in 2012 and many others provide freedoms to journali sts (Freedom, 2013). Despite some challenges, the U.S. remains ranked in one of the 25 freest positions that a nation can be ranked in the world (Freedom, 2013). In Kuwait, Freedom House However, it s report expresses that several issues occurred in 2012 which demonstrated an increase of repressive governance. For example, media outlets that incited conflict were shut down regardless of whether it was sectarian or against the ruling family. During s ome Arab Spring riots in March 2012, more than 500 Twitter users were detained and 150 were charged for violating two articles of the Kuwaiti constitution by insulting the emir. Amin (2002) described some of the conditions that Arab journalists have had t o deal with. He explains that some journalists have economic pressures that many U.S. journalists face (Amin, 2002). However, they also face political and cultural pressures, including censorship, physical assault, threats and arrest (Amin, 2002). In so me cases, journalists have been detained, tortured or abducted (Amin, 2002). In other cases, journalists have had travel restrictions placed on them, had their passport withdrawn or were exiled (Amin, 2002). Despite such challenges, Amin (2002) argues th at journalists still have the power to promote social change and influence public opinion. The lack of media laws in the Gulf States contributes to the lack of academic research pertaining to the region, according to media law scholar who lived and condu cted research in the

PAGE 15

15 region: Matt Duffy (2013). Duffy (2013) suggests that there is a tendency of native Arabic speakers to avoid rigorous media scholarship, as well as an inaccessibility of the media law which may be not available in any other languages in some instances for researchers who may be more knowledgeable about media law due. Therefore, a variety of concerns may influence the amount of scholarship and news coverage that the Gulf States receive. In other words, some issues may be censored due t o a reluctance of natives to discuss them in addition to an inability for foreigners to access and/or understand them. In his analysis of media law in the Gulf states, Duffy (2013) traces censorship in the region to the deep cultural ties with Great Brita in as colonies or otherwise. He points out that an underdeveloped media law system may be a symptom of a greater problem which involves both corruption and an underdeveloped legal system (Duffy, 2013). Kuwait specifically states that freedom is guarante ed to both the media and those employing the scientific method for research purposes (Duffy, 2013). After demonstrating that Kuwait has the most freedoms of all the Gulf States, Duffy (2013) explains some of the similar press restrictions in the Gulf Stat es. For example, penal codes are applicable to those who defame others or criticize heads of states in most of the Gulf States (Duffy, 2013). Similarly, cybercrime laws were used during the Arab spring to penalize those who disseminated false news or cri ticized heads of states on Twitter (Duffy, 2013). Moreover, laws dealing with licensing of media outlets exist; however, some of these laws have not been updated in three decades despite changes in communication technologies (Duffy, 2013). An understand ing of media freedoms is crucial to the concept of agenda silencing because censorship may contribute to the fear of isolation that may exist. The current study will

PAGE 16

16 explore the issues in Kuwaiti and U.S. media. Newspaper journalists and social media use rs are not equal in the freedoms afforded to them in Kuwait and in the United States. The Gulf States are not the only state to suffer from a crisis in managing communication laws in an era of rapidly changing communication technologies (Duffy, 2013). All nations must balance rights of free speech with rights necessary for the maintenance of public order (Duffy, 2013). An understanding of the media landscape in the Arab world is necessary prior to an examination of the media agenda pertaining to the Middl e East. Arab journalist s value social change in the Middle East, according to survey data gathered by Pintak and Ginges (2008) which included more than 600 Arab journalists working for media distributed in the Arab world. Of these journalists, 27 percent were located in the Gulf States and Yemen, 26 percent were located in the Levant region of the Arab world (e.g., Lebanon and Syria), 25 percent were living in Egypt, 17 percent were located in North Africa (e.g., Morocco), 4 percen t were located in the United States or Europe, and approximately one percent did not specify their exact location within the Middle East ( Pintak & Ginges, 2008). The majority of the respondents were Muslim (88 percent), while seven percent reported being Pintak & Ginges, 2008). The journalists were asked questions pertaining to the function of journalists in society ( Pintak & Ginges, 2008). The journalists felt it was their duty to be involved in political reform (75 percent), educate public (66 percent), create news for social good (60 percent), give voice to the poor (58 percent), advocate for civic engagement (56 percent), promote development (53 percent), analyze issues (51 percent), advocate t he Palestinian cause (48 percent), transform society (44 percent), be an agent of change (43 percent), investigate government (40 percent),

PAGE 17

17 foster Arab culture (37 percent), promote spiritual values (33 percent), defend Arab causes (32 percent) and enhance Arab unity (29 percent) ( Pintak & Ginges, 2008). The journalists then reported ranked issues facing the Arab world according to importance. Political reform (46 percent), human rights (45 percent), poverty (42 percent), education (41 percent), Palestin e (34 percent), Iraq (29 percent), economy (28 percent), terrorism (28 percent), health (23 percent), globalization (18 percent), environment (15 percent) and other (4 percent) ( Pintak & Ginges, 2008). The greatest number of journalists (34 percent) rated the U.S. policy as the most threatening challenge facing the Arab world ( Pintak & Ginges, 2008). Other journalists felt that a lack of political change (32 percent), human rights abuses (23 percent), economy (20 percent), political instability (18 perc ent), terrorism (14 percent), globalization (6 percent) and other (2 percent) issues were the most threatening challenge facing the Arab world ( Pintak & Ginges, 2008). Survey data from Arab journalists reveals much information pertaining to the values of Arab journalists. However, analyses of media content in the Middle East can also reveal the values of Arab journalists. A content analysis was conducted and 3, 352 news items were gathered from Jordanian media outlets (Ali, 2006). The analyzed media mess ages came from a daily national newspaper, a small town weekly newspaper, a radio station and a television station (Ali, 2006). Ali (2006) found that news pertaining to international and domestic Jordanian politics were most prominent on radio and televis ion. In the capital city of Amman, the most popular newspaper issues included sports, business and culture (Ali, 2006). In the small town of Irbid, the most popular newspaper issues included culture, domestic politics and entertainment (Ali, 2006). If a study was a part of a greater study exploring news values around the world.

PAGE 18

18 Shoemaker and Cohen (2006) conducted both content analyses and focus groups in t en countries, including Australia, Chile, China, Germany, India, Israel, Jordon, Russia, South Africa and the United States. The scholars concluded that there are some universal truths pertaining to news around the world (Shoemaker & Cohen, 2006). One of these universal truths is that people want to be informed about socially deviant or significant events (Shoemaker & Cohen, 2006). Further, people do not agree about the prominence given to stories in their newspapers (Shoemaker & Cohen, 2006). This disag reement stems from a tendency for newspapers to avoid giving socially deviant and socially significant occurrences prominence (Shoemaker & Cohen, 2006). Shoemaker and Cohen (2006) conclude that such differences reveal that there is a distinction between t he concepts of news and newsworthiness. News is the product of journalistic activities, while news worthiness is an evaluation of what news is (Shoemaker & Cohen, 2006). This divide is universal and applies around the world. An understanding of newswort hiness in different cultures is important because the current study explores data collected from two places with different cultures. Media Ethics Some aspects of media ethics are also universal. For example, a reference for the pursuit of truth and object ivity was found in ethical codes for journalists in Muslim nations such as Algeria, Bangladesh, Malaysia, Morocco and Pakistan (Hafez, 2002). However, a similar requirement for truthful reporting exists in the journalistic codes in Western nations, such a s Finland, Germany, and Italy (Hafez, 2002). While the definition of objective and truthful reporting varies among nations, the idea of this ethical standard refers to the degree to which information is verified to be truthful and reported in such a way s uch that reality is not intentionally distorted (Hafez, 2002).

PAGE 19

19 Similarly, privacy is a part of ethical code for media practitioners in Muslim countries such as Turkey, Indonesia and Egypt (Hafez, 2002). Non Muslim countries such as Finland, Spain and the U ethics (Hafez, 2002). The right to privacy varies according to country, but is typically thought of as a desire to avoid reporting on the affairs of an individual unless it ha s greater implications for the public interest (Hafez, 2002). While some ethical standards for media industries may seem similar, there are some differences among the ethical values of some standards. For example, the journalism ethics codes of Iraq, Fra nce, Bangladesh and Kyrgyzstan do not mention a right to freedom of expression at all (Hafez, 2002). Some nations view freedom as a central value that can be limited when it interferes with certain areas of public life, such as politics, religion or cultu re (Hafez, 2002). These nations include Lebanon, Egypt, Kazakhstan, Pakistan and Saudi Arabia. Yet some other nations view freedom of expression as a value as long as it does not interfere with other rights (Hafez, 2002). These nations include Tunisia , Algeria, Spain, Germany and the United Kingdom (Hafez, 2002). Clearly, the differences presented in the ethical codes of media practitioners are not separated by whether or not the nation is primarily Muslim. The promotion of peaceful international affa irs is also a part of journalism ethics codes in Along the same li journalists that may create enemies for the state (Hafez, 2002). One finding that Hafez (2002) suggested may be unique to some of the Arab and Muslim world deals with religion. Fo r example, the codes of nations such as Egypt, Saudi Arabia,

PAGE 20

20 Pakistan and Malaysia suggest that journalists have a responsibility to promote traditions, as well as religious values and rituals (Hafez, 2002). One way such a responsibility could manifest it self could surface could be through a condemnation of speech that may offend either religious groups or sects within a religion (Hafez, 2002). Finally, some Arab nations like Egypt and Saudi Arabia suggest that there must not be a perversion of the Arabic language by the media (Hafez, 2002). An emphasis on preser ving religious traditions has been explored in the process of news selection for Al Jazeera (Cherribi, 2006). By exploring the political and economic powers of Qatar, Cherribi (2006) finds that political elites are not the only people who influence media frames in Al Jazeera. Rather, Cherribi (2006) suggests that religious elites who may have powerful positions in Islamic institutions also have a powerful role in influencing Al Jazeera coverage of issues. One particular issue which can be a focal point in exploring such a question is the French ban on veils (Cherribi, 2006). Cherribi (2006) suggested the use of a recently veiled anchorwoman to conduct an interview with the French Minister of Foreign Affairs was a public show of religiosity, which would then add immediate credibility to audiences across the Arab and Muslim world. Cherribi (2006) also demonstrated that a show in which a traditional Muslim and a Westernized Arab debated public issues on Al Jazeera . Yet, Cherribi (2006) argues that the tra ditional Muslim speaker frequently disrupted the Western Arab in such a way that hindered any kind of public debate. In other words, there is a tendency to promote religious and customary traditions by media in the Arab world that may not be easy to ident ify in the U.S. media. Therefore, a study of the Middle East media must examine a religious agenda of news.

PAGE 21

21 Role of Media in International Conflict Cultural considerations can affect journalists anywhere. Gatekeeping theory explains that there are differ ent stages along the news gathering processes in which information can either proceed to be distributed throughout society, or restricted (Watson, 1998). One such model suggests that there are five stages in gatekeeping of international news (Watson, 1998 ). First, an event occurs (Watson, 1998). Then, a foreign correspondent makes decisions about covering the event (Watson, 1998). Next, a regional bureau editor makes decisions relevant to the news article (Watson, 1998). Then, it is received to an age ncy central desk where (Watson, 1998). Finally, the individual editors can make decisions to spread or contain the news (Watson, 1998). Yet, it is important to acknowledge that often work related reasons are more likely to influence the process of gatekeeping than personal background of a reporter (Cassidy, 2006). J as important as journalistic t raining and the number of years on staff (Cassidy, 2006). When a journalist works for an organization, he/she must learn the organizational values of his employer (Watson, 1998). The editors are also influenced by the cultural climate of the institution w ithin the state (Watson, 1998). In other words, there are several steps along the news creation process where gatekeepers may restrict the movement of information through society. One example of this at work which concerns Kuwait is coverage of Operatio n Desert Storm (Watson, 1998). Since the media had to gain approval from military authorities prior to showing footage, only images and sounds that were approved were in the media (Watson, 1998).

PAGE 22

22 International crises can be a very interesting case becaus e of the media effects that are unique to such situations. The CNN effect is one example of the balance of pressures in domestic and international policy after an international crisis (Livingston & Eachus, 1995). The idea behind the effect is the degree t o which U.S. foreign policy decisions are influenced by an increase in media coverage (Livingston & Eachus, 1995). Yet, critics of the CNN effect claim that it only operates when massive human rights are being violated and the U.S. is refusing military ac tion (Jakobsen, 2000). Rather, critics of the CNN effect suggest that very little attention to international conflict is covered by Western media (Jakobsen, 2000). Jakobsen (2000) explains that there are several stages in conflict, and some exist before v iolence occurs while others exist after violence has occurred. His article, which was published in the Journal of Peace Research , suggested that a focus on post violence conflict leads to an ineffective use of resources by media organizations and ineffective management of conflict by governments (Jakobsen, 2000). It follows then that multiple stakeholders can benefit from understanding the implications of the media silencing issues that deal with international conflict. Known for its political i nstability, the Middle East presents a particularly interesting case for exploring the concept of agenda setting. The current study will explore how the spiral of silence operates within the agenda setting process.

PAGE 23

23 CHAPTER 2 LITERATURE REVIEW Agenda Silencing A content analysis of mass communications literature suggested that the agenda setting effect of the mass media was among the most popular theories in the field (Bryant & Miron, 2004). McCombs and Shaw (1972) generated a theory that had so much explanatory power, that many scholars were able to use it to explain media effects despite developments in media technologies and industry. Although the question of the agenda setting effect has been explored with regards to differing media types (Wanta, 1997; Stromback &Kiousis, 2010), there is a gap in the literature exploring the role of social and online media in the agenda setting process. This gap is not limited to agenda setting literature; rather, there is a need for innovation with regards to the theoretical understanding of other effects of the mass media, including the spiral of silence. The purpose of this study is to merge agenda setting literature with spiral of silence literature to develop the concept of agenda silencing . The concept of a genda silencing can be distinguished from the concept of agenda cutting in several ways. 87). To demonstrate how this effect wor ks empirically , Colistra (2012) surveyed reporters to show that approximately 30 percent of reporters would avoid stories that may embarrass their associates or news organization. Fahmy (2010 ) suggested that while there are methodological problems to stud ying the agenda cutting effect, her examination of blogs in authoritarian press systems do reveal that they provide perspectives that are either removed or placed in a low priority in mainstream media. Rather than attempt to measure what is not stated, th e concept of agenda silencing posits that the degree to which an issue is salient in newspapers can moderate the degree to which it would appear in

PAGE 24

24 social media. The curre nt study contributes the relationship between the measurements of salience of an iss ue in newspapers , as created by McCombs and Shaw (1972), to examine the degree to which those Noelle Neumann (1974) would have considered hard core discuss such issues in social media . The rationale for this cannot be explained without exploring the agend a setting and spiral of silence theoretical areas separately. Agenda Setting The agenda setting function of the mass media has several layers. The first level of the agenda setting, also referred to as issue agenda setting, explains that as an issue gain s salience in the media, it is discussed more by the public (McCombs & Shaw, 1972). Analyses of media content contributed to the idea that certain political parties are more apt at handling or owning some issues, an idea referred to as issue ownership ( Br azeal & Benoit, 2008). Further analyses of media issues led to the idea of agenda diversity, which can be evaluated by measuring diversity of sources, content and the exposure of audience members through the use of media (Tan & Weaver, 2012). While issue agenda setting is most often used to understand contemporary issues, the agenda setting effect has contributed to historical research exploring As the agenda s etting scholarship continued to grow in breadth, it also grew in depth. A second level of the agenda setting function of the mass media was discovered. The second level of the agenda setting, also known as attribute agenda setting, demonstrates that obje cts (e.g. people, corporations or nations) in the media have certain attributes, which the public associates with the object as the attributes increase in salience (Ghanem, 1997). Some of these attributes are cognitive, while others are affective (Ghanem, 1997). Cognitive attributes, also referred to as substantive attributes, refer to some categories of topics relating to the media object (Ghanem, 1997; Kiousis, 2004). Affective attributes,

PAGE 25

25 sometimes referred to as valence, can be thought of as emotiona l characteristics of the media object (Ghanem, 1997; Kiousis, 2004). Again, the second level of agenda setting states that cognitive and affective attributes can be transferred through media salience. While Kiousis (2004) showed that salience can be operationalized by focusing on valence, prominence and attention , his factor analysis revealed that the two major components valence and visibility, which combines prominence and attention. Media salience has also been u sed to establish the theoretical concept of agenda building (Lang & Lang, 1983). As defined by Lang and Lang (1983), agenda building is the concept that interactions between the public, media and an actor can lead to conditions which alter public opinion. (Lang & Lang, 1983), agenda building has been applied to intercandidate agenda setting by conducting content analyses of press releases and other strategic, political co mmunications ( Kiousis & Shields, 2008). Regardless of whether the issue and attribute agenda building has been used to examine public or strategic messages from the mass media, the scholarship continues to expand in both breadth and depth. The third level of the agenda setting process suggests that the human memory stores information in cognitive networks. A study by Guo, Vu and McCombs (2012) shows that there is a correlation between content analysis of attributes of a media object and survey data about public opinion regarding the attributes of this media object. By using social network analysis, there is room for much scientific advancement on how these attributes are stored in the human memory. Since balance theory has been applied through the use of social network analysis (Wasserman and Faust, 1994), it is likely that future literature will explore questions such as the cognitive dissonance of varying attributes stored in cognitive networks. Regardless, the use of

PAGE 26

26 social network analysis is current ly being explored with regards to the advancement of the third level of agenda setting. All three levels of the agenda setting effect provide evidence for the transfer of salience of issues and attributes from the media to the public. Yet, developments in agenda setting research has yielded the idea of agenda cutting, which explores the degree to which topics are removed from the media agenda (Colistra, 2012). Agenda cutting can occur when an issue is mentioned once in the media and then replaced by anoth er issue or when an issue never appears in the media (Colistra, 2012). The current study explores the question of agenda silencing, which differs from agenda cutting. While agenda cutting states that an issue is removed from the media agenda, agenda sile ncing claims that the issue is still there. The current study will explore It is typical for media effects research to address questions that deal with the infl uence of mass media on interpersonal relationships ( Bryant & Miron, 2004). Yet, it is important to note that both agenda setting and the spiral of silence are macro societal level effects, which means they deal with cultural issues rather than interperson al ones. While agenda setting refers to the Neumann, 1977). Spiral of Silence According to the spiral of silence effect, the media provide a measure for individuals to calculate the degree to which their opinion is popular (Noelle Neumann, 1974). If one decides his/her opinion is unpopular, one opts to remain silent due to his/her fear of is olation (Noelle Neumann, 1974). While most people would acquiesce when confronted with the fear of isolation, some individuals prefer to express unpopular opinions (Noelle Neumann, 1977). These individuals may get used to isolation, or they may selective ly expose themselves to media

PAGE 27

27 content which is congruent with their opinions (Noelle Neumann, 1977). However, the majority of those faced with the fear of isolation due to having a minority opinion remain silent (Noelle Neumann, 1974). The theory was ad vanced by testing conditions of the public opinion climate. For communication was public or private (Donsbach, 1984). Similarly, minority perspectives were more likely to be expressed if an individual felt that popular opinion would change in the foreseeable future (Gylnn & McLeod, 1984). Individual were more likely to avoid expressing an opinion when the climate is hostile than when the climate is friendly, by employing exist strategies such as walking away (Hayes, 2007). In conditions where the whole country has less freedom to express opinions such as in the context of less democratic nations, political interest was found to be a moderating variable with regards to willingness to express minority perspectives (Willnat, 1996). Overall, researchers explored factors within a public opinion climate that would moderate the spiral of silence effect. Further, the spiral of silence theory withstood tests that app lied it to a wide range of issues. For example, the spiral of silence effect was found to be evident in situations where one feels certain about his/her perspective on an issue ( Matthes, Morrison, & Schemer, 2010). Specifically, issues relating to values were likely to result in polarizing discussions, where the spiral of silence effect would be weakened because individuals seem to have certain positions under these conditions ( Matthes, Morrison, & Schemer, 2010). In a study of homosexuality and the mil itary, the media not only helped individuals individuals learn to determine the popular opinions in society or segments of society

PAGE 28

28 (Gonzenbach, King & Jablonksi, 1999). Moreover, a study on willingness to express minority individuals regarding the O.J. Simpson trials revealed that race was a moderating variable for (Jeffres, Neuendorf, & Atkin, 1999). Therefore, evidence supports th e spiral of silence effect with regards to different issues and circumstances. To further understand the spiral of silence effect, researchers explored its role with varying personal qualities. For example, individuals were more likely to express their op inion with friends than strangers ( Crandall & Ayres, 2002) . Also, individuals who are typically anxious about communication were less likely to express an opinion than those who are more comfortable about communication ( Crandall & Ayres, 2002). Similarly , the spiral of silence was weakened as individuals reported more self efficacy and more political interest (Lasorsa, 1991). Despite differing issues and climates, the main methods employed to examine the spiral of silence effect consist of survey and exp erimental designs. The current study will add layers of complexity to the spiral of silence effect by exploring the role of media content. Comparing and Contrasting the Two Theories Levels of Analysis Both the agenda setting and spiral of silence are macr o level theories. Chaffee and Berger (1987) explained that communication research can employ four levels of analysis. The first level is the intra individual level, which deals with internal processes that individuals facilitate in communication activiti es. The second level is the interpersonal level, where communication occurs between two or more people but does not contain a large number of participants. The organizational level refers to large groups of people which pertain to communication. Finally , the macro or societal level deals with culture and how it pertains to communication.

PAGE 29

29 Agenda setting deals with the contribution of media to culture with the transfer of salience from the media to public agenda. On the converse, the spiral of silence d eals with the restriction of contributions of individuals within a culture who disagree with opinions expressed in the media. While both theories are examined at the macro level of analysis, researchers have examined the effects of the spiral of silence a nd agenda setting in other levels. For example, the interpersonal level of analysis has been examined in studies exploring both media effects. Wanta and Wu (1992) found that interpersonal communication supports the agenda setting effect when discussions r evolve around issues covered in the media. However, Wanta and Wu (1992) also found that issues not covered are also discussed in interpersonal settings. Similarly, the research of Yang and Stone (2003) found that individuals who do not use the media to g ain information about public affairs often talk to their friends, colleagues and family members who do use the media to learn about public affairs, thereby supporting the agenda setting effect of the media even when they do not rely on the media themselves . Clearly, the agenda setting effect has been examined in the interpersonal level of analysis. However, the spiral of silence has also explored in interpersonal settings. For example, Hayes (2007) outlined some strategies that individuals use when avoiding expressing their opinion in a hostile climate. The most popular of the strategies was reflec ting a question, followed by expressing uncertainty or ambivalence, which was followed by expressing indifference (Hayes, 2007). Some of the least popular strategies included saying nothing, walking away from a discussion or pretending to agree (Hayes, 20 07). Research dealing with opinion avoidance strategies in hostile opinion climates is one way that spiral of silence research has been conducted in the interpersonal level of analysis.

PAGE 30

30 Definitions of Public Opinion In addition, the two theories also expl ain how the mass media influence public opinion. However, the way the two theories define public opinion may differ. Scheufele and Moy (2000) demonstrate that there are two ways of defining public opinion, but that these ways of defining public opinion d o not necessarily compete with one another. Public opinion can be viewed as a collective social judgment that citizens decide upon after examining knowledge pertaining to the issue (Scheufele & Moy, 2000). In this view of public opinion, individuals are capable of engaging in the political processes (Scheufele & Moy, 2000). Along the same lines, this view of public opinion makes the public sphere a necessary part of a democracy and civic participation that promotes social change (Scheufele & Moy, 2000). The agenda setting theory defines public opinion in this way. However, the spiral of silence deals with public opinion as a way to create consensus and exerting social control (Scheufele & Moy, 2000). In this view, individuals are constantly faced with t he fear of isolation as they scan their environment to create a sense of social cohesion (Scheufele & Moy, 2000). According to Durkheim (1933), more evolved societies would demonstrate organic solidarity and allow individuals within the society freer move ment within society. Less evolved societies would demonstrate mechanical solidarity, whereby punishment occurs when social boundaries are crossed (Durkheim, 1933). The current study explores the concept of public opinion in Kuwait and the United States, which vary in terms of their social evolution. Despite this variance, Scheufele and Moy (2000) reconciled differences with regards to defining public opinion which would encourage the exploration of the spiral of silence in the agenda setting process with regards to issues relating to the Middle East. The first way the two ways of defining public opinion can be reconciled is by exploring communication as both an active and passive process (Scheufele & Moy, 2000). While

PAGE 31

31 communication may be more active in the agenda setting process, communication may be more passive in the spiral of silence (Scheufele & Moy, 2000). For example, the use of bumper stickers and campaign buttons can be explored as a passive method of expressing minority opinions with regards t o public issues (Scheufele & Moy, 2000). Spiral of silence research has also included wearing hats or t shirts with s logans on them or signing a petition (Perry & Gonzenbach, 2000). In other words exploring communication in the passive and active sense helps to decrease the differences between the definition of public opinion in spiral of silence and agenda setting. Secondly, the differences in definitions of public opinion can be bridged when one discussions (Scheufele & Moy, 2000). While individuals may be unconsciously scanning a social environment to estimate the degree to which their opinion would be accepted, fewer people are motivated to enter political discussions (Scheufele & Moy, 2000). Therefore, defining public opinion as an entity which exists regardless of whether or not it is explicitly expressed is crucial to an exploration of the spiral of silence within the agenda setting process. Third, Scheu fele and Moy (2000) argue that some definitions of public opinion are less inclusive. In one view of public opinion, only educated people who are engaged in civic participation are pertinent. However, expanding this category to include everyone is helpfu l in decreasing the differences in definitions of public opinion that lends itself to an examination of how the spiral of silence operates within the agenda setting process. A review of public opinion is important to review because the two theories being b lended to conceptualize agenda silencing define public opinion differently. In this study, issues covered

PAGE 32

3 2 and tones of coverage (affective attributes) will be examined closely to explore any instances of communication that is transferred in salience and t hat which is silenced. Account for Cognitive and Affective Processes Within the public opinion formation process, both theories suggest that cognitive and affective components exist. The second level of agenda setting explicitly states that cognitive and affective attributes are transferred from the media to the public agenda (Ghanem, 1997). an issue received, the placement of the coverage within a newspaper and the tone in which the issue was covered, added reliability to the idea that attributes are cognitive and affective. While most spiral of silence scholarship does not explicitly sta te it, this media effect also includes an affective and cognitive component. For example, Glynn and McLeod (1984) explored willingness to communicate minority opinion when the future majority opinion is expected to change. The rationale for their study w as to explore the degree to which fear of isolation would be reduced if the estimated future opinion was expected to change (Glynn & McLeod, 1984). The current study argues that the fear of isolation is an affective component and the opinion itself is a c ognitive component. Therefore, the spiral of silence research must explore both affective and cognitive attribute of minority opinions. Application to Online Media Another commonality between the mass communication theories involves the fact that scholars do not fully understand how they operate online. While some agenda setting scholars argue that online media do not have a specific media agenda, other scholars have shown that users gain access to the traditional media agenda in their online format (McCo mbs, 2005). For example, Roberts, Wanta and Dzwo (2002) found a correlation between the content of the New York Times and the content of electronic bulletin boards. Similarly, Liu and Fahmy (2011) found

PAGE 33

33 that individuals are less likely to feel threatened due to a fear of isolation online. Regardless, congruency of opinion with the perception of majority opinion is likely to predict willingness to communicate about an issue online. Differences by Issues Despite the ample similarities between the two the ori es, scholarship must acknowledge the differences that from the public and media agendas, agenda setting research deals with the transfer of salience from the media to the pub lic agenda. Yet, recent research pertaining to agenda setting acknowledges that some issues are either neglected or ignored by the media (Colistra, 2012). Some organizations, such as Media Tenor, have developed measurements for items that are cut off of the media agenda (Colistra, 2012). Looking at silence and speech as opposing sides of a spectrum expands our understanding of mass communication theories, and brings out more similarities between the spiral of silence and agenda setting effects. Applicati on A cross Cultures According to Wanta, King and McCombs (1995), both behavioral and attitudinal differences are to be expected among cultures. In their study of issue diversity, findings revealed that young people in Taiwan were more concerned with a grea ter variety of issues because they who were not very interested in democratic reforms in Hong Kong were willing to speak out regardless of the perception of m ajority opinion. Willnat suggested that a culture of political the process of democratization . revi ew of literature relating to international media is in order.

PAGE 34

34 International Media Social scientific development depends on a demand for knowledge about society. And a degree of independence which supports knowledge that is based on evidence, even if it do es not express support to segments of societies, which fund the creators of knowledge (Bourdieu, 1994). In the development of states, it is typical for knowledge to be institutionalized such that there is a conflict between dominant and dominating perspec tives (Bourdieu, 1994). The media are often viewed as a mass distributor of knowledge in society (Nisbet & Aufderheide, 2009). It follows then that the U.S. media often focuses on negative developments in international news, and rarely present perspectiv es of other nations (McPhail, 2010; Fahmy, 2007). Most international news around the world are reported for individuals working for U.S. or European wire, radio, television, magazine, photograph, and advertising services (McPhail, 2010). Thus, there is an unequal flow of information from Western nations to other parts of the world, including the Middle East (McPhail, 2010). would be emphasized during crises. After the Sep t. 11 th terrorist attacks, the media agenda contained more international content; however, the change did not last very long (Hatchen & Scotten, 2007). According to anthropologist Mary Douglas (1966), social pollution occurs when individuals within a grou p violate a rule of the group, when danger presses on the external boundary of a group, when members of a group are on the margins of it, and when there is a lack of internal consistency of the group. The lack of internal consistency of a group can occur when individuals that were legitimized within a bureaucratic system of a state use the system to advance their personal interests rather than to advance the interests of the public (Bourdieu, 1994). Social pollution can pose a threat to social cohesion, w hich Durkheim (1933) described could be encouraged in one of two ways: mechanical or organic solidarity. Mechanical

PAGE 35

35 solidarity refers to punitive measures that correct wrongdoings, while organic solidarity happens because individuals view themselves as a part of a society (Durkheim, 1933). The current study explores the degree to which group boundaries influence decisions of self censorship and willingness to speak out in online and traditional media i n Kuwait and the United States. Group boundaries between the West and the Middle East have been the focus of some scholarship. For example, Said (1978) argued that the West and the East seem to define one another based on a distorted view of the other as being different. Other scholars have considered provides many instances of organic solidarity for viewing international events through Western eyes. The U.S. media focus on domestic news rather than foreign news for many reasons. In addition to linguistic, cultural, religious and geographical barriers, it is more expensive for media organizations to o perate overseas (McPhail, 2010; Fahmy, 2007). Further, proximity of an event to their audience is one of the news values of journalists (Shoemaker & Cohen, 2006). However, a content analysis of thousands of news items from news media revealed that newswo rthiness alone does not determine what news is produced in the U.S. (Shoemaker & Cohen, 2006). In the United States, the media is characterized by a freedom to report on, opine about and even criticize government (Hatchen & Scotten, 2007). This Western mo del of the press has been criticized for having a sensationalist and consumerist bias (Hatchen & Scotten, 2007). One of the characteristics of Western media audiences is that they have become more interested in entertainment media than journalism partiall y due to a lack of credibility. After all, even the Western model of the press faces a degree of governmental pressures.

PAGE 36

36 However, the Western model of the press has many more freedoms than the developmental model of the press, which values local content b ecause of a belief that Western development journalism often sets out to give voice to local cultures and promote independence of former colonies, the media often leaders (Musa & Domatob, 2007, p. 322). In the process, opposing voices are silenced in those media systems. The current study will content analyze the media issues in newspapers in the United States and Kuwait as well as social media. In doing so, the current study will explore whether the traditional media is successful in telling social media users what to think about, which is the first level of agenda setting. Further, the present study will explore whether the traditional media is successful in telling social media users how to think about these issues. Finally, the present study will explore instances in which both issues and the attributes of issues are incongruent between social medi a users and traditional media. By examining the traditional and social media agenda, the current study explores the degree to which content is related to an agenda setting or spiral of silence effect. Online Media Agenda setting and the spiral of silence studies have been applied to online media (Roberts, Wanta, & Dzwo, 2002; McDevitt, Kiousis, & Wahl Jorgensen, 2003 ). For example, an agenda study revealed that The New York Times was more successful than Associated Press, Reuters, Time magazine and CNN suggested in providing information that users of electronic bulletin boards passed along electronically, as long as the issues were not controversial (Roberts, Wanta, & Dzwo, 2002). In this study, online media were content analyzed to fi nd an increase of

PAGE 37

37 salience in traditional media led to an increase in salience in online media with regards to some issues. In another study, an examination of the spiral of silence on computer mediated, interpersonal communication revealed that individu als discussed a controversial issue, such as abortion, in a more moderate tone than they reported their stance ( McDevitt, Kiousis, & Wahl Jorgensen, 2003). While this study employs a content analysis to examine the spiral of silence effect, it does so to c ompare self reported stance with tone of the messages. The purpose of this study will be to content analyze traditional media to explore the ways in which it affects social media discussions relating to the Middle East. Despite the fact that literature fo cusing on Middle Eastern media is scarce, it would be inaccurate to claim that political scientists studying the Middle East underestimate the importance of media. Rather, many political scientists have examined Middle Eastern newspapers, films, music, po litical cartoons and websites as evidence in order to answer some theoretical questions relating to the negotiation of power. The events of the Arab spring surfaced when addressing questions relating to the development of nations cannot be underestimated. i,1991, p. 53). For example, Khalidi (1991) pointed to themes in newspapers from Damascus, Beirut, Cairo and the Arab quarters of Istanbul to show that roots of Arab nationalism began prior to the First World War. Similarly, LeVine (1999) examined Arabi c newspapers based in Jaffa and Hebrew language fiction, films, and other media to provide a historical examination of how Arabs and Israelis identified with the emerging city of Tel Aviv prior to the British Mandate after the First

PAGE 38

38 World War. Along the s ame lines, Campos (2010) used excerpts of newspapers as evidence of tolerance and coexistence between Jews, Christians and Muslims in Palestine during the early 20 th century. Overall, scholars of Middle East Studies often select themes or excerpts from th e mass media to support an intellectual argument relating to history. Scholarship focusing on the Middle East has also relied on evidence from the mass media for analysis relating to political ethnography. For instance, Wedeen (1999) juxtaposed examples f the effects of President Hafiz al y, LeVine (2008) interviewed Muslim, heavy metal musicians from Morocco, Egypt, Israel/Palestine, Lebanon, Iran and Pakistan to understand how different individuals resolve identities that may seem contradictory. In doing so, LeVine (2008) uncovers storie s of people with varying degree of religiosity and differing political views. LeVine (2008) provides context that not only enriches the understanding of globalization with regards to music but also music as a coping mechanism for Muslim youth. Clearly, M iddle Eastern Studies includes examples from the media in political ethnography and history. As the events of the Arab spring unfolded, the deficiency in the literature relating to Middle Eastern media became apparent. Yet, scholars interested in the Ar ab spring compared developments and a measure of the emotional environment during the protests (Franklin, 2011, p. 17). In a study of the nations which use Twitt er the most, the Oxford Internet Institute found that Kuwait is an Arab country with a high penetration of Twitter and sends the most amount of messages between March 5 through March 13, 2012 ( Graham & Stephens, 2012). At this time,

PAGE 39

39 the Arab spring protes ts in Kuwait were active ( Freedom, 2013 ). The current study will content analyze social media from the United States and Kuwait to explore the congruence between traditional and social media in these nations. While Kuwait may seem like an unlikely compari son to the United States, Kuwait is being chosen as the case study for sev eral reasons. First, Kuwait has the freest press of the nations that make up the Arabian Gulf C ooperation Council (Freedom, 2013 ). Second, reforms to media law in Kuwait in 2006 lo osened laws permitting privately owned television stations and loosening the restrictions on obtaining newspaper licenses (Selvik, 2011). The result was a tripling of the number of newspapers in Kuwait (Selvik, 2011). Third, such developments in the medi a landscape influences the democratic elections for positions in a parliament that has veto power which it uses (Al Nashmi, Cleary, Molleda, & McAdams, 2010). While these freedoms s democracy is able to absorb political crises, such as sectarianism (Stephenson, 2011). Fourth, Kuwait is home to more foreign nationals than natives (World Factbook, 2013). associations were created between 1928 and 1 960, when politicians invited individuals from Iraq and the Levant to work as journalists (Alfraih, 1999). This history has created an international focus in Kuwaiti newspapers that still exists today. Therefore, Kuwait makes an interesting case study in comparing a diverse, international media to that of the United States. Further, Kuwait is one of many Arab nations impacted by the emergence of online media. hundre ds of Arabic websites gossip, music, e 120), its impact on Kuwait has been unique. For example, issues that are normally too controversial to be covered by traditional media were discussed on online blogs according to

PAGE 40

40 Kuwaiti media scholar, Ali Dashti (2009). Analyses of online political forums in the Middle East revealed that online users are not afraid to engage in political debate which can be critical of the United States and its foreign policy in the r egion (Al Nashmi et al., 2010). Being an affluent nation, Kuwaitis are eager to explore capabilities of online media, according to Al Nashmi et Such criticism has not gone unnoticed in the United States, where proactive measures were taken by the U.S. government to create positions of public diplomacy in which the Public diplomacy can be mediated, and scholars have found the degree to which Arab television viewers favor the United States is related to the perceived credibility of televised news content funded by the United States in the Arab world. (Fahmy, Wanta & Nisbet, 2012). S imilarly, an analysis of U.S. newspapers by showed that the public was more likely to view Muslims less negatively than the media portrayed with regards to the desire for peace, the similarities between Christians and Muslims, and the acceptance of various races in Islam (Bowe, Fahmy & Wanta, 2013). However, the public was more likely to view Muslims more negatively than the media portrayed with regards to questions dealing with gender equality and homosexuality (Bowe, Fahmy & Wanta, 2013). The authors suggest that more research is required to understand such subtleties in public opinion with regards to Islam (Bowe, Fahmy & Wanta, 2013). Though the relationship between public opinion and journalistic style is more ambiguous than clear, Vultee (2012) c demonstrate the degree to which American journalists are unfamiliar with Muslim culture (p. less dramatic and ne wsworthy

PAGE 41

41 to track the number of times it has been used since it first appeared in an Associate d Press article in February of 1979. His results show that the term appeared less frequently after the 2000s, when coverage of the Sept. 11 attacks resulted in a focus on international news and Islam in the United States media. While the coverage of Arab and Muslims societies by traditional media has been examined, online media analysts have been more concerned about the use of the medium to Berkman Center for Internet and Society which states that Arab bloggers discuss domestic political issues more than the United States and its foreign policy, an examination of how online media may challenge or compliment perspectives on mainstream media has not yet surfaced in the i ntellectual discussions regarding Middle Eastern media. In the last decade, this need for scholarship which examines multiple aspects of the media in the Middle East has resonated with several communication scientists. For instance, Duffy (2013) out lined the topics that media are prohibited from addressing according to media law of the Arabian Gulf states. In his analysis, Duffy (2013) also compared Arab media regulations to Political Rights that The underdevelopment of various aspects of the media system in the Arab world may produce an environment where some topics are silenced while others are not.

PAGE 42

42 Figure 2 1. Agenda S etting Model Figure 2 2. Spiral of Silence Model Media Object Affective attributes Cognitive attributes Public agenda Media Majority public opinion Fear of isolation; social desirability Decision to communicate opinion

PAGE 43

43 CHAPTER 3 HYPOTHESES AND RESEARCH QUESTIONS The current study explores the degree to which agenda setting occurs in Kuwait and in the United States. A content analysis of traditional and social media will be conducted in order to assess the degrees to which social media is congruent to traditional media both cognitively and affectively. As the academic literature suggests, the issues covered by traditional media are expected to differ across the two nations. H 1 : Traditional media in the United States will cover different issues than traditional me dia in Kuwait regarding news in the Middle East. By exploring which issues are present in Kuwaiti and U.S. newspapers when it comes to Middle East news, we can compare and contrast the media agenda in both nations. The current study substitutes a social m edia component for a survey component. Just as one would expect publics to differ cross nationally, one may expect social media in the two nations to discuss differing issues. H 2 : Social media in the United States will address different issues than social media in Kuwait regarding news in the Middle East. According to agenda setting theory, the news media transfers attributes from the media agenda to the public agenda. These attributes can be cognitive or affective. If the traditional newspapers are succ essful in setting the agenda for the social media users, then the agenda setting hypotheses will be supported, given that the present study substitutes social media for a survey component. H 3 : The high salient issues in Kuwaiti social media content will be congruent with the issues in traditional media content in Kuwait. H 4 : The high salient issues in U.S. social media content will be congruent with issues in traditional media content in the United States. H 5 : The high salient a ffective attributes in Kuwaiti social media content will be congruent with the affect ive traditional media content in Kuwait.

PAGE 44

44 H 6 : The high salient affective attributes in U.S. social media content will be congruent with the affec tive traditional media cont ent in the United States. If the evidence supports the agenda setting effect, then the previous four statements should be true. If there is incongruence between the affective and cognitive attributes of objects in the traditional and social media of Kuwait and the United States, then the agenda setting effect would not accurately describe the phenomenon. In the case that cognitive and affective attributes differ in traditional and social media, it would be appropriate to explore the possibility of the spira l of silence effect online. Literature suggests that approximately one fifth more or less of individuals with minority perspectives would voice their opinions regardless of circumstances ( Noelle Neumann, 1974 ). To account for such outspokenness, we would expect a minority of social media posts to be incongruent with the traditional media in both Kuwait and the United States. H 7 : Low salient issues from s ocial media content will be incongruent with the cognitive attributes from traditional media content in Kuwait. H 8 : Low salient issues from U.S. social media content will be incongruent with the cognitive attributes from traditional media content in the United States. H 9 : Low salient affective attributes from social media content will be incongruent with the affectiv e attributes from traditional media content in Kuwait. H 10 : Low salient affective attributes from U.S. social media content will be incongruent with the affective attributes from traditional media c ontent in the United States. Should the number of social media posts with cognitive and affective attributes that are incongruent with th ose of traditional media be fewer than 25 percent, then the spiral of silence would be supported online. If the social media posts with cognitive and affective attributes congruent with those of traditional media exceed 25 percent, then it could be important to explore similarities and differences between media agendas cross nationally because it is likely

PAGE 45

45 that most peopl interesting to compare differing agendas of mediated communication. Similarly, there are some reasons to assume that the media content in Kuwait will associate more cognitive attribut es with media objects than the United States. For example, the fact that more foreign nationals reside in Kuwait than locals would suggest that a greater focus would be placed on international news. The current study would explore the number of cognitive attributes associated with the Middle East in both Kuwaiti and U.S. newspapers as well as social media. Such an investigation would require a comparative analysis of how the issues, cognitive attributes and affective attributes in U.S. and Kuwaiti traditi onal and social media compare. It is likely that those who disagree with one medi um media agenda. Therefore, it would be interesting to compare what traditional and social media in the U.S. and individuals i n Kuwait have to say about one another as well. RQ 1 : Are issues covered by the traditional media in Kuwait related to the U.S. social media content that is incongruent with the U.S. traditional media? RQ 2 : Are affective attributes linked to issues covered by by the traditional media in Kuwait related to the U.S. social media content that is incongruent with the U.S. traditional media? RQ 3 : Are issues covered by the traditional media in U.S. related to the Kuwaiti social media content t hat is incongruent wi th the Kuwaiti traditional media? RQ 4 : Are affective attributes linked to issues covered by the traditional media in the U.S. related to the Kuwaiti social media content that is incongruent with the Kuwaiti traditional media? The four research questions li sted above should explain how the Kuwaiti and U.S. media agendas are different than one another. The issues, cognitive attributes and affective attributes should be compared and contrasted in order to gain a clearer picture of what kinds of information wo uld be available to users who have access to both media agendas as well as the kinds of

PAGE 46

46 information that are missing from those who have access to only one media agenda. Since the egards to the Middle East due to linguistic, religious and/or cultural differences, the agenda silencing effect would be media agenda may disagree with the media a genda of each nation individually as members of an outspoken minority.

PAGE 47

47 CHAPTER 4 METHODS Many mass communications theories have relied on content analysis to explain or predict a phenomenon. For example, the cultivation effect has relied on content analysis show that representations of homosexual couples on several television shows not only gave heterosexual couples an inaccurate idea of homosexual relationships, but also provided homosexual couples poor role models (Holz Ivory, Gibson & Ivory, 2009) . Exploring manufactured messages in the media is the premise of many theoretical frameworks within media effects research, including cultivation and agenda setting (Gerbner, 1988; McCombs, 2005). Content analysis is one of several quantitative methodolog ical approaches to the social sciences ( Neuendorf , 2002) . In the field of mass communications, it is especially important because it used to analyze the messages in the media ( Neuendorf , 2002) . Many theoretical frameworks have been employed through the u se of content analysis. The current study employs content analysis to a theoretical framework that often relies on survey or experimental research design. There are several benefits to using content analysis. Just as survey methods do not manipulate inde pendent variables that eliminate questions of generalizability resulting from the creation of an artificial manipulation, content analysis also analyzes messages that are natural occurring. Just as survey methods are concerned with questions of reliabilit y, content analysis requires inter coder reliability to calculate the degree to which coders agree with one another. Further, coding categories in content analysis are largely considered to be objective in the sense that more than one person should agree on the lack of subjectivity of the categories. Adequate training of coders is necessary to ensure that such reliability exists in content analysis.

PAGE 48

48 Like all social scientific methods, content analysis has its limitations. Since the messages are naturally occurring, questions of validity emerge in content analysis. Researchers must make sure that the data they gather from media messages address the theoretical question under scrutiny. The importance of coder training cannot be understated. After all, mas s communications researchers have identified two characteristics in messages: manifest and latent content. Manifest content is explicitly stated content that can be measured with ease compared to latent content ( Neuendorf , 2002). Latent content includes i mplicitly stated messages, which must be measured by creating indicators to test for their existence ( Neuendorf , 2002). Content analysis typically deals with manifest content; however, latent content is not completely missing from agenda setting research. It is not unusual to explore the agenda setting effect by employing a content analysis. After all, the first study to propose that an agenda setting function of the media employed content analysis as well as survey methods to show that the issues in the media are also in the public agenda (McCombs & Shaw, 1972). Studies exploring the role of newspapers in transferring issue salience from the media to the public agenda also employed a content analysis (Golan & Wanta, 2001). Thus, content analysis typical ly has explored media messages. While the aforementioned research may describe content analyses with mainly manifest content, agenda cutting examines messages for latent content as the issues ignored or neglected by the media are measured. Media Tenor is an organization reported to have found indicators which determine a method for measuring silence (Colistra, 2012); however, it is not clear what these indicato rs are. The current study measure s silence by comparing media coverage in two different nations : Kuwait and the United States.

PAGE 49

49 In both nations, economic conditions influence the issues that are reported. Colistra (2012) showed that at least one third of U.S. journalists avoid certain news stories that may not be in favor of the interest of their ne explored agenda cutting in her use of survey methods, the spiral of silence has typically relied on surveys and experiments rather than on content analysis (e.g., Jeffres, Neuendorf, & Atkin, 1999 ). Yet, there is one study which employed a content analysis in an exploration of the spiral of silence. In addition to employing an experimental design, McDevitt, Kiousis, and Wahl Jorgensen (2003) conducted a content analysis to examine the choices individuals made when discussing the issue of abortion online. Their findings suggest that there are multiple ways to discuss issues, and that individuals with extreme minority views may moderate their speech when communicating online without the nonverbal cues that are inherent to interpersonal communication. Sample A content analysis of traditional and social media in Kuwait and the United States was examined . Due to the need for consecutive news cov erage to determine media effects, the time frames included in this study begin s almost immediately after President Obama made a speech asking Congress to vote on a military strike in Syria: September 2013 . Three Arabic l anguage, Kuwaiti newspapers, three E nglis h language, U.S. newspapers and messages from TWitter are content analyzed. The unit of data collection will be an article for U.S. newspapers, the content on the front page cover of Kuwaiti newspapers and a tweet for social media. Traditional Media Two Kuwaiti newspapers have been selected for analysis. First, Al Qabas is a newspaper established in 1972 by four of the most eminent merchant families of Kuwait to represent the

PAGE 50

50 nsidered public opinion as oil wealth decreased their political power in Kuwait (Selvik, 2011). Next , Al sman, who targets a Shiite audience (p. 484). Thus, these newspapers would provide represent diverse editorial positions in Kuwait. Newspapers are well circulat ed in Kuwaiti society, where approximately 1.3 of the 2.6 million people living in Kuwait are expatriates, many of whom are Arab but not Kuwaiti (World Factbook, 2013). Furthermore, the literacy rates are over 90 percent for both males and females (World Factbook, 2013). Therefore, there are relatively fewer barriers to the accessibility of newspapers such as Al Qabas and Al Nahar compared to newspapers in some of the neighboring Arab countries where female illiteracy rates may be lower. Al Qabas ha s been daily newspapers in Kuwait for approximately four decades (Selvik, 2011). Yet, there are some discrepancies in reports of the circulation rates of Kuwaiti newspapers. While circulation statistics would demonstrate the degree to which the selected newspa pers are important, the selection criteria were not limited to circulation. Press Reference (2013) is one website which describes the history of newspapers around the world along with circulation data. Al Qabas is one of the most popular newspapers in Kuw ait, with its online version receiving 30,000 weekly hits in 2002 (Press, n.d.). Reports suggest that its paper version reached a circulation rate of 79,000 by 2001 which is a large decrease from its circulation rate of 120,000 prior to the Iraqi invasion of Kuwait in 1990 (Press, n.d.). By contrast, the circulation rate of Al Nahar was not listed, as it was established in 2007 and the page did not reflect any information from that time period in Kuwait (Press, n.d.).

PAGE 51

51 Having accumulated data from several so urces which include media corporations and Wikipedia, the World Association of Newspapers reports Al 2008 (World Association, 2009 ). However, circulation statistics for Al Nahar were unavailable (World Association, 200 9 ). Despite the discrepancies, few would argue that Al Qabas and Al Watan are popular newspapers in Kuwait. Unlike Al Qabas, which has existed for decades, Al Nahar was established in 2007 after media reform laws made it easier to gain licenses for newspa per ownership. Due to its relatively new status as a Kuwaiti newspaper, circulation rates for this newspaper are unavailable. While none of the newspapers selected for this study were ranked as having the highest circulation rates, they represent differi ng editorial positions since Al Qabas is funded by Sunni merchants, Al Watan is funded by members of the ruling family and Al Nahar is funded by a wealthy Shiite merchant. Funding sources are important according to Selvik (2011) because Kuwaiti media are used in the struggle for power in a political rivalry which exists between members of the ruling family. While the loosening of press restrictions in Kuwait allowed previously unheard voices to minority religious groups to contribute to public debate, Se access to financial and political resources also creates an uneven playing field for the exchange (p. 493). The fact that m ost newspapers which were established after 2006 support are more . Despite such setbacks to the democratization process, Selvik (2011) argues that the changes to the press law are significant because they signify social mobility which would allow newly

PAGE 52

52 wealthy merchants to own a newspaper. Overall, media ownership in Kuwait is an important factor for selection of a sample of newspapers. The question of media ownership also influences content in the United States. For example, news organizations that report on foreign affairs endure costs related to transportation, acc ommodation, and translation (Fahmy, 2009). On an even greater scale, economic conditions and world trade influence news coverage in the United States. In fact, U.S. media scholars have demonstrated that media coverage varies according to oil politics. F or instance, Wanta, Golan and Lee (2004) found that Kuwait and Saudi Arabia were outliners because they were viewed positively by individuals in the United States despite the little amount of press coverage they received. Along the same lines, Wanta and G olan (2009) found that the more involved a nation was in oil production, the more likely it was for its elections to receive media coverage. A content analysis of news relating to the Middle East is conducted with a sample from The New York Times and The Washington Post. Both the Times and the Post will be included because they are considered elite newspapers with a documented ability to set the media agenda for other news org anizations in the United States because of they do not focus on local coverage of communities but rather on the world at large (Martin Kratzer & Thorson, 2009). Social Media Social media were particularly important in the events of the Arab spring. The specific platform associated with each country varies. Reports on outspokennes s by nation show that Twitter is very important in Kuwait ( Graham & Stephens, 2012). In an analysis of intermedia agenda setting between political blogs and newspapers, Meraz (2011) points to the fact that partisan blogs are likely to cover the same issue s. Similarly, Culbertson (2007) showed that the more issues covered by a newspaper, the higher its quality was ranked. The current study will

PAGE 53

53 explore the degree to which social media users address issues that are congruent with traditional media. Twitter accounts of Kuwaiti activists known for their outspokenness on Twitter will be evaluated. These accounts will include a ma n and a wom a n. As of September 30, 2014, @ Salma_AlEssa has 38,200 followers and @bourashed has 168,000 followers . Keeping i n mind d 4 million people and that fewer than 2 million people are citizens of voting age, these Twitter accounts are very influential for Kuwait. Further, these Twitter accounts are kept by both individual and conse rvative users. In the United States, we will examine the following outspoken political commentators: The sample includes the conservative Ann Coulter (followed by 548,000 ) and t he liberal Rachel Maddow (followed by 2, 9 9 0 , 00 0 ). These are the accounts of ex ceptionally outspoken and influential individuals with regards to political matters in the United States . The social media accounts of men were excluded from the sample after an examination of the posts showed either no posts during the month of September 2013 or seemed to be managed by staff members. Coding Categories This study will explore three agendas: issues, nations and religions. Issues have been examined in studies such as Golan and Wanta (2004) which examined issues such as ed ucation, race and foreign policy when examining press coverage of George Bush and John McCain in the New Hampshire primaries. Similarly, perceptions of foreign nations were examined by Wanta, Golan and Lee (2004), when they found that negative coverage l eaves the public with negative perceptions of the nation. In his article that examined the past, present and future of agenda setting research, McCombs mentions that scholars would continue to explore agenda setting

PAGE 54

54 such media effects (p. 554). Therefore, there is a precedent and logic in agenda setting literature for the examination of multiple agendas, such as issues, nations and religions. In Oil Politics a nd the Gulf: Rulers and Merchants in Kuwait and Qatar , Jill Crystal (1995) pointed out that the merchants of Kuwait were vital to the development of Kuwait prior to discovering oil and the wealth generated as a result of this discovery. As the ruling fami ly of Kuwait gained wealth, they relied less on Kuwaiti merchants. Selvik (2011) went on to explain that Al Qabas is a newspaper that several merchant families created as a joint venture in an attempt to influence public opinion in Kuwait when they felt t heir power in Kuwaiti society was threatened. The current study analyzes messages for issues relating to merchants, tensions between merchants and the government, and the oil sector. Political issues within Kuwait include categories, such as the ruling f amily mentions (tone, appearance), national assembly (tone, appearance), minist ers ( tone, appearance ), prime ministers ( tone, appearance), election ( tone, appearance) , stateless people or bedoon (tone, appearance ). Political issues within Kuwait include: o il/petroleum (tone, appearance), merchan ts (tone, appearance ), socialism (tone, appearance), capitalism (tone, appearance), privatization (tone, appearance), and world trade (tone, appearance). Further, Crystal (1995) revealed that Shiites in Kuwait have a tradition of being loyal to the ruling family. She explains that Shiites showed their loyalty to the Kuwaiti ruling family revolution, some Kuwaiti Shiites Religious issues include Islam (general) (tone, appearance), Islam Sunnis (tone, appearance), Islam Shiites (tone, appearance), Islam Alawites (tone, appearance), Islam -

PAGE 55

55 Wahabis (tone, appearance), Islam Salafis (tone, appearance), Islam Ismaeli (tone, appearance), Islam Zaidi (tone, appearance), Islam Musli m Brotherhood (tone, appearance ), Judaism (tone, appearance), Zionism (tone, appearance), Christianity (general) (tone, appearance), Christian ity Catholic ( tone, appearance ), Christianity Coptic (tone, appe arance ), and Christianity Orthodox (tone, appearance). Finally, Crystal (1995) wrote her book a decade prior to the legislation allowing Kuwaiti women to vote and be elected for public office. While her research did not reveal this aspect of generated allegiances that could have been political and/or economic in nature. Therefore, I would like to code for media cont a general category for representations of other women. Outside of the Middle East, several development s have influenced the relationship of the United States with the Arab and Muslim world. For example, the newly canonized pope made the Middle East the focus of one of his speeches this year. Additionally, law suits exceeding $10 billion have been in cour based out of the United States. Further, political uprisings in the Middle East in the aftermath of the Arab spring have affected political relations between the two parts of the world. T hese are issues that must be examined in a content analysis. Nations analyzed in this study include d the 22 nations of the Arab league along with five other Middle Eastern and Asian nations: Kuwait (tone, appearance), Egypt (tone, appearance), Syria(tone, appearance), Saudi Arabia (tone, appearance), Tunisia (tone, appearance), Morocco (tone, appearance ), Sudan (tone, appearance), Lebanon (tone, appearance in headline, number

PAGE 56

56 of sentences per art icle), Iraq (tone, appearance ), Palestine (tone, appearan ce ), Algeria(tone, appearance), Djibouti (tone, appearance), Somalia (tone, appearance ), Libya (tone, appearance), Bahrain (tone, appearance), Oman (tone, appearance), Qatar (tone, appearance ), UAE (tone, appearance), Yemen (tone, appearance ), Jord an (tone, appearance ), Comoros (tone, appearance), Mauritania (tone, appearance), Iran (tone, appearance ), Pakistan (tone, appearance ), Israel (tone, appearance), Turkey (tone, appearance), and Afghanistan (tone, appearance). Generalizability A randomly constructed week (N=61) was generated and compared to the complete sample of September 2013 in order to understand the generalizability of the sample. A series of t tests were conducted to explore the mean differences between the data in the constructed week and the entire sample. The results indicate the number of graphs was greater in the constructed week (M=21.13, SD=9.83) than in the entire sample (M=20.43, SD=12.90); (difference of mean=0.69, standard error difference=1.74, t(357)=0.39, p=0.69). F or each article, every graph and headline was coded for tone. The results indicate that the tone of the headline was fewer positive in the constructed week (M=1.77, SD=0.56) than the entire sample (M=2.11, SD=5.67); (difference of mean= 0.34, standard erro r difference=0.72, t(356)= 0.46, p=0.64). The results also indicate that there were more graphs with a positive tone in the constructed week (M=2.83, SD=3.85) than in the entire sample (M=2.27, SD=3.36); (difference of mean=0.56, standard error difference =0.48, t(357)=1.15, p=0.24). The results also indicate that there were fewer graphs with a neutral tone in the constructed week (M=10.45, SD=6.54) than in the entire sample (M=11.08, SD=8.13); (difference of mean= 0.62, standard error difference=1.10, t(3 57)= 0.56, p=0.57). The results also indicate that there were more graphs with a negative tone in the constructed week (M=7.72, SD=5.03) than in the entire

PAGE 57

57 sample (M=7.09, SD=6.33); (difference of mean=0.63, standard error difference=0.86, t(357)=0.73, p= 0.46). A series of cognitive attributes were also coded for the numbers of graphs in which they appeared. These cognitive attributes were divided into two categories. One of these categories dealt with issues and the other category dealt with a nation or religion. Issues The issue of democracy appeared more in graphs of the constructed week (M=4.77, SD=5.90) than the entire sample (M=3.15, SD=4.31); (difference of mean=1.61, standard error difference=0.64, t(357)=2.49, p=0.01 ). The numbers of graphs mentioning Hezbollah or Hamas appeared fewer in the constructed week (M=0.18, SD=0.59) than in the entire sample (M=0.21, SD=0.60); (difference of mean= 0.03, standard error difference=0.08, t(357)= 0.36, p=0.71). The numbers of gra phs mentioning the West Bank or Gaza appeared fewer in the constructed week (M=0.04, SD=0.21) than the entire sample (M=0.03, SD=0.28); (difference of mean=0.01, standard error difference=0.03, t(357)=0.40, p=0.68). The issue of climate change appeared i n more graphs in the constructed week (M=0.08, SD=0.52) than in the entire sample (M=0.03, SD=0.27); (difference of mean=0.51, standard error difference=0.04, t(357)=1.11, p=0.26). Mentions of education appeared more in the constructed week (M=0.65, SD=1 .15) than in the entire sample (M=0.78, SD=2.81); (difference of mean= 0.13, standard error difference=0.36, t(357)= 0.36, p=0.71). Terrorism was mentioned in feweer graphs of the constructed week (M=1.23, SD=2.23) than the entire sample (M=1.37, SD=2.57 ); (difference of mean= 0.14, standard error difference= 0.35, t(357)= 0.42, p=0.67).

PAGE 58

58 References to children appeared in fewer graphs in the constructed week (M=0.34, SD=0.79) than in the entire sample (M=0.43, SD=1.43); (difference of mean= 0.08, standar d error difference= 0.18, t(357)= 0.44, p=0.65). fewer graphs in the constructed week (M=0.16, SD=0.81) than in the entire sample (M=0.21, SD=0.48); (difference of mean= 0.04, standard error difference= 0.10, t(357) =0.45 , p=0.65). References to history or historical figures appeared in more graphs in the constructed week (M=2.83, SD=3. 22) than in the entire sample (M=2.28, SD=3.13); (difference of mean=0.55, standard error difference= 0.44, t(357)=1.24, p=0.21). R eferences to sports appeared in more graphs in the constructed week (M=0.17, SD=1.09) than in the entire sample (M=0.04, SD=0.21); (difference of mean= 0.12, standard error difference= 0.14, t(357)= 0.91, p=0.36). References to the Arab League occurred in fewer graphs in the constructed week (M=0.09, SD=0.30) than in the entire sample (M=0.12, SD=0.47); (difference of mean= 0.02, standard error difference= 0.06, t(357)= 0 .46, p=0.64). More general references to Arabs that excluded the Arab League occurred fewer times in the constructed week (M=0.39, SD=0.71) than the entire sample (M=0.40, SD=0.86); (difference of mean= 0.01, standard error difference= 0.11, t(357)= 0.13, p=0.89). Health was mentioned in more graphs in the constructed week (M=0.60, SD=1.8 9) than the entire sample (M=0.39, SD=1.44); (difference of mean=0.21, standard error difference= 0.21, t(357)=0.98, p=0.32).

PAGE 59

59 Media were mentioned in fewer graphs in the constructed week (M=3.34, SD=3.65) than the entire sample (M=2.96, SD=3.70); (differe nce of mean=0.37, standard error difference= 0.52, t(357)=0.72, p=0.46). The Gulf Cooperation Council was mentioned in fewer graphs in the constructed week (M=0.09, SD=0.30) than the entire sample (M=0.14, SD=0.73); (difference of mean= 0.04, standard err or difference= 0.09, t(356)= 0.48, p=0.62 ). Intelligence was mentioned in more graphs in the constructed week (M=0.57, SD=1.37) than the entire sample (M=0.51, SD=1.34); (difference of mean=0.06, standard error difference= 0.18, t(357)=0.31, p=0.75). U.S. National Security was mentioned in fewer graphs in the constructed week (M=0.23, SD=0.64) than the entire sample (M=0.31, SD=0.83); (difference of mean= 0.08, standard error difference= 0.11, t(357)= 0.79, p=0.42). The Unite d Nations were mentioned i n fewer graphs in the constructed week (M=1.32, SD=2.27) than the entire sample (M=1.67, SD=3.28); (difference of mean= 0.34, standard error difference= 0.44, t(357)= 0.77, p=0.43). Traffic was mentioned in more graphs in the constructed week (M=0.03, SD= 0.17) than the entire sample (M=0.01, SD=0.12); (difference of mean=0.01, standard error difference= 0.01, t(357)=0.82, p=0.41). Mentions of Europe or the European Union were mentioned in more graphs in the constructed week (M=0.91, SD=2.09) than the enti re sample (M=0.42, SD=1.26); (difference of mean=0.49, standard error difference= 0.20, t(357)=2.43, p=0.01).

PAGE 60

60 Crime was mentioned in fewer graphs in the constructed week (M=0.48, SD=1.29) than the entire sample (M=0.67, SD=3.25); (difference of mean= 0.18 , standard error difference= 0.42, t(354)= 0.44, p=0.65). Protests were mentioned in fewer graphs in the constructed week (M=0.90, SD=2.49) than the entire sample (M=0.93, SD=1.98); (difference of mean= 0.03, standard error difference= 0.29, t(356)= 0.12, p=0.89). Oil and/or petroleum was mentioned in fewer graphs in the constructed week (M=0.09, SD=0.35) than the entire sample (M=0.11, SD=0.80); (difference of mean= 0.01, standard error difference= 0.10, t(357)= 0.18, p=0.85). Nations Kuwait was mentio ned in fewer graphs in the constructed week (M=0.01, SD=0.12) than the entire sample (M=0.02, SD=0.15); (difference of mean= 0.007, standard error difference= 0.02, t(357)= 0.34, p=0.73). Egypt was mentioned in fewer graphs in the constructed week (M=0.26, SD=0.94) than the entire sample (M=0.45, SD=1.82); (difference of mean= 0.19, standard error difference= 0.24, t(357)= 0.79, p=0.42 ). Syria was mentioned in fewer graphs in the constructed week (M=6.72, SD=7.03) than the entire sample (M=6.96, SD=7.44); (difference of mean= 0.23, standard error difference= 1.03, t(357)= 0.23, p=0.81). Saudi Arabia was mentioned in fewer graphs in the constructed week (M=0.14, SD=0.40) than the entire sample (M=0.27, SD=1.01); (difference of mean= 0.12, standard error difference= 0.13, t(357)= 0.94, p=0.34).

PAGE 61

61 Tunisia was mentioned in more graphs in the constructed week (M=0.01, SD=0.12) than the entire sample (M=0.007, SD=0.08); (difference of mean=0.009, standard error difference= 0.01, t(3 57)=0.13, p=0.45). Morocco was mentioned in fewer graphs in the constructed week (M=0.00, SD=0.00) than the entire sample (M=0.01, SD=0.10); (difference of mean= 0.01, standard error difference= 0.01, t(357)= 0.78, p=0.43). Sudan was mentioned in fewer graphs in the constructed week (M=0.03, SD=0.17) than the entire sample (M=0.03, SD=0.23); (difference of mean= 0.004, standard error difference= 0.03, t(357)= 0.12, p=0.89). Lebanon was mentioned in more graphs in the constructed week (M=0.21, SD=0.66) t han the entire sample (M=0.20, SD=0.62); (difference of mean=0.008, standard error difference= 0.08, t(357)=0.09, p=0.92). Iraq was mentioned in more graphs in the constructed week (M=1.26, SD=2.53) than the entire sample (M=0.86, SD=2.23); (difference of mean=0.39, standard error difference= 0.32, t(357)=1.22, p=0.22). Palestine was mentioned in fewr graphs in the constructed week (M=0.34, SD=1.16) than the entire sample (M=0.25, SD=1.14); (difference of mean=0.09, standard error difference= 0.16, t(357)= 0.57, p=0.56). A lgeria was mentioned in more graphs in the constructed week (M=0.04, SD=0.38) than the entire sample (M=0.01, SD=0.18); (difference of mean=0.03, standard error difference= 0.03, t(357)=1.10, p=0.26).

PAGE 62

62 Djibouti was not mentioned in any graphs in the constructed week (M=0.00, SD=0.00) than the entire sample M=0.00, SD=0.00); thus, independent sample t tests were not computer for this variable. Somalia was mentioned in fewer graphs in the constructed week (M=0.14, SD=0.67) than the entir e sample (M=0.27, SD=1.69); (difference of mean= 0.12, standard error difference= 0.22, t(357)= 0.57, p=0.56). Libya was mentioned in fewer graphs in the constructed week (M=0.26, SD=1.03) than the entire sample (M=0.37, SD=1.38); (difference of mean= 0.1 1, standard error difference= 0.18, t(357)= 0.58, p=0.55). Bahrain was mentioned in more graphs in the constructed week (M=0.06, SD=0.30) than the entire sample (M=0.05, SD=0.35); (difference of mean=0.01, standard error difference= 0.04, t(357)=0.24, p=0. 80). Oman was mentioned in fewer graphs in the constructed week (M=0.00, SD=0.00) than the entire sample (M=0.01, SD=0.14); (difference of mean= 0.01, standard error difference= 0.01, t(357)= 0.74, p=0.46). Qatar was mentioned in fewer graphs in the con structed week (M=0.04, SD=0.21) than the entire sample (M=0.09, SD=0.51); (difference of mean= 0.04, standard error difference= 0.06, t(357)= 0.61, p=0.53). The United Arab Emirates was mentioned in fewer graphs in the constructed week (M=0.03, SD=0.17) than the entire sample (M=0.06, SD=0.26); (difference of mean= 0.02, standard error difference= 0.03, t(357)= 0.77, p=0.43).

PAGE 63

63 Yemen was mentioned in fewer graphs in the constructed week (M=0.00, SD=0.00) than the entire sample (M=0.03, SD=0.25); (difference of mean= 0.03, standard error difference= 0.03, t(357)= 1.15, p=0.25). Jordan was mentioned in fewer graphs in the constructed week (M=0.04, SD=0.21) than the entire sample (M=0.09, SD=0.31); (difference of mea n= 0.04, standard error difference= 0.04, t(357)= 1.12, p=0.26). Comoros was not mentioned in any graphs in the constructed week (M=0.00, SD=0.00) than the entire sample M=0.00, SD=0.00); thus, independent sample t tests were not computer for this variabl e. Mauritania was not mentioned in any graphs in the constructed week (M=0.00, SD=0.00) than the entire sample M=0.00, SD=0.00); thus, independent sample t tests were not computer for this variable. Iran was mentioned in more graphs in the constructed week (M=2.75, SD=5.97) than the entire sample (M=4.82, SD=3.70); (difference of mean=0.64, standard error difference= 0.70, t(357)=0.91, p=0.36). Pakistan was mentioned in fewer graphs in the constructed week (M=0.09, SD=0.39) than the entire sample (M=0.12, SD=0.77); (difference of mean= 0.02, standard error difference= 0.10, t(357)= 0.25, p=0.80). Israel was mentioned in more graphs in the constructed week (M=1.34, SD=3.72) than the entire sample (M=1.26, SD=3.61); (difference of mea n=0.08, standard error difference= 0.51, t(357)=0.16, p=0.87).

PAGE 64

64 Turkey was mentioned in fewer graphs in the constructed week (M=0.13, SD=0.34) than the entire sample (M=0.16, SD=0.44); (difference of mean= 0.02, standard error difference= 0.06, t(357)= 0.49 , p=0.61). Afghanistan was mentioned in fewer graphs in the constructed week (M=0.21, SD=0.58) than the entire sample (M=0.39, SD=1.5); (difference of mean= 0.17, standard error difference= 0.20, t(357)= 0.88, p=0.37). The United States was mentioned in fewer graphs in the constructed week (M=14.93, SD=9.49) than the entire sample (M=15.24, SD=12.82); (difference of mean= 0.30, standard error difference= 1.73, t(357)= 0.17, p=0.85). Russia was mentioned in fewer graphs in the constructed week (M=1.15, S D=1.72) than the entire sample (M=2.41, SD=5.62); (difference of mean= 1.30, standard error difference= 0.72, t(357)= 1.79, p=0.07). China was mentioned in fewer graphs in the constructed week (M=0.23, SD=0.49) than the entire sample (M=0.44, SD=2.24); (d ifference of mean= 0.21, standard error difference= 0.28, t(357)= 0.72, p=0.46). The United Kingdom was mentioned in fewer graphs in the constructed week (M=0.50, SD=1.20) than the entire sample (M=0.61, SD=1.25); (difference of mean= 0.10, standard error difference= 0.17, t(357)= 0.58, p=0.55). France was mentioned in more graphs in the constructed week (M=0.60, SD=2.20) than the entire sample (M=0.53, SD=1.92); (difference of mean=0.07, standard error difference= 0.27, t(357)=0.27, p=0. 78 ).

PAGE 65

65 Religions Religious purity was not mentioned in any graphs in the constructed week (M=0.00, SD=0.00) than the entire sample M=0.00, SD=0.00); thus, independent sample t tests were not computer for this variable. Religious impurity was mentioned in fewer graphs in the constructed week (M=0.0 0 , SD=0. 00 ) than the entire sample (M=0.003, SD=0.05); (difference of mean= 0.003 , standard error difference=0.0 07 , t(357)= 0. 4 5 , p=0. 65 ). References to religious or sectarian issues were mentioned in more graphs in the constr ucted week (M=0.39, SD=0.88) than the entire sample (M=0.27, SD=0.69); (difference of mean=0.11, standard error difference=0.11, t(357)=1.12, p=0.26). General references to Islam were mentioned in fewer graphs in the constructed week (M=0.72, SD=1.64) th an the entire sample (M=0.90, SD=1.93); (difference of mean= 0.18, standard error difference=0.26, t(357)= 0.68, p=0.49). References to Sunni Islam were mentioned in more graphs in the constructed week (M=0.26, SD=1.01) than the entire sample (M=0.174, S D=0.74); (difference of mean=0.08, standard error difference=0.11, t(357)=0.78, p=0.43). References to Shiite Islam were mentioned in more graphs in the constructed week (M=0.36, SD=1.47) than the entire sample (M=0.26, SD=1.16); (difference of mean=0.09 , standard error difference =0.17, t(357)=0.535, p=0.59). References to Alawi Islam mentioned in more graphs in the constructed week (M=0.06, SD=0.30) than the entire sample (M=0.03, SD=0.20); (difference of mean=0.02, standard error difference=0.03, t(35 7)=0.90, p=0.36).

PAGE 66

66 References to Wahabi Islam were not mentioned in any graphs in the constructed week (M=0.00, SD=0.00) than the entire sample M=0.00, SD=0.00); thus, independent sample t tests were not computer for this variable. References to Salafi Islam were mentioned in fewer graphs in the constructed week (M=0.00, SD=0.00) than the entire sample (M=0.007, SD=0.08); (difference of mean= 0.006, standard error difference=0.01, t(357)= 0.64, p=0.52). References to Ismaeli Islam were not mentioned in any graphs in the constructed week (M=0.00, SD=0.00) than the entire sample M=0.00, SD=0.00); thus, independent sample t tests were not computer for this variable. References to Zaidi Islam were not mentioned in any graphs in the constructed week (M=0.00 , SD=0.00) than the entire sample M=0.00, SD=0.00); thus, independent sample t tests were not computer for this variable. References to the Muslim Brotherhood were mentioned in more graphs in the constructed week (M=0.14, SD=0.51) than the entire sample (M=0.15, SD=0.74); (difference of mean= 0.006, standard error difference=0.09, t(357)= 0.06, p=0.94). References to the Nation of Islam were mentioned in fewer graphs in the constructed week (M=0.00, SD=0.00) than the entire sample (M=0.01, SD=0.11); (difference of mean= 0.01, standard error difference=0.01, t(357)= 0.90, p=0.36). General references to Judaism were mentioned in more graphs in the constructed week (M=0.52, SD=2.10) than the entire sample (M=0.29, SD=1.33); (difference of mean=0.23, standard error difference=0.20, t(357)=1.10, p= 0.26).

PAGE 67

67 Zionism were mentioned in fewer graphs in the constructed week ( M=0.01, SD=0.12) than the entire sample (M= 0. 0 1 , SD=0.15); (difference of mean= 0.0004, standard error difference=0.02, t(357)= 0.01, p=0.98). General references to Christianity were mentioned in fewer graphs in the constructed week (M=0.26, SD=0.65) tha n the entire sample (M= 0.26 , SD=1 .30 ); (difference of mean= 0. 002, standard error difference=0.17, t(357)= 0.01, p=0.98). References to Catholic Christianity were mentioned in fewer graphs in the constructed week (M=0.0 1 , SD=0. 12 ) than the entire sample (M=0.0 3 , SD=0. 32 ); (difference of mean= 0. 02 , standard error difference=0.0 4 , t(357 )= 0.49, p=0.62). References to Coptic Christianity were mentioned in fewer graphs in the constructed week (M=0.00, SD=0.00) than the entire sample (M=0.003, SD=0.05); (di fference of mean= 0.003, standard error difference=0.007, t(357)= 0.45, p=0.65). References to Orthodox Christianity were mentioned in fewer graphs in the constructed week (M=0.00, SD=0.00) than the entire sample (M=0.04, SD=0.45); (difference of mean= 0 .04, standard error difference=0.05, t(357)= 0.69, p=0. 49 ). Interpretation The results indicate that randomly constructed week is very similar to the entire sample. There is only one issue for which the randomly constructed week is statistically differe nt than that of the entire sample: democracy (p=0.01). One explanation for this is that the sample began almost immediately after President Obama asked the nation to vote on the U.S. involvement in the Syrian conflict. It is important to consider that th is issue is the only coded category appears differently in the randomly constructed week than in the entire sample. None of the Middle Eastern or United Nations Security Council nations for which the randomly constructed week is statistically different tha n that of the entire sample . No

PAGE 68

68 statistically significant differences were found for the two samples in the religion category of variables. The results suggest that data collected from The New York Times and The Washington Post are mostly generalizable. Riffe, Lacy and Fico (2005) explain a randomly constructed week as a sampling technique for daily newspapers. They cite research which shows that one or two constructed ructed week as an efficient technique to avoid including all daily newspapers in a sample, a constructed week was generated to show that the results are generalizable to the population from which the sample was gathered. Inter Coder Reliability At least two coders coded the social media and newspapers in Kuwait and the United States. Ten percent of the data were coded by two individuals, and these samples were selected randomly. To calculate the degree of agreement between the two coders, Krippe values were calculated for the different variables using software made available by Freelon could not be calculated. However, the number of obse rvations which agree with one another should reveal similarities. Overall, at least ten percent of the entire sample was coded by two coders. 1.0. The value s are provided in Appendix A. This was because some coders felt the style of a message included connotations relating to a specific issue, religion or nations. Different coders were trained to collect data again for these categories to use in future stud ies.

PAGE 69

69 CHAPTER 5 RESULTS Results of Hypothese s issues than traditional media in Kuwait regarding news in the Middle East . hypothesis, the frequ encies of the coded categories were evaluated. The coded categories can be divided into several categories, including issues, nations and religions. The coded issues include an adjusted list from Al issues which appear in satellite television in the Middle East based on historical contexts of the re gion. The frequencies of 13 categories appear in order of most to least salient in Table 5 1. The coded nations include Middle Eastern nations as well as the nations of the U nited Nations Security Council. The Middle Eastern nations consist of the 22 states that make up the Arab league in addition to Iran, Pakistan, Israel, Turkey, and Afghanistan. The United Nations Security Council nations include the United States, Russia , China, France, and the United Kingdom. Below, Table 5 2 shows the frequencies of the appearance of these nations. The coded religions include 16 varieties of three religions: Islam, Christianity and Judaism. The various categories within each religio n may refer to a spiritual or political sect. These include Islam (general), Sunni, Shiite, Alawite, Wahabi, Salafi, Ismaeli, Zaidi, Muslim Brotherhood, Nation of Islam, Judaism (genera), Zionism, Christianity (general), Catholic, Coptic, and Orthodox. Finally, two spiritual categories were included from the Muslim perspective, including purity and impurity. The results are shown below in Table 5 3. Then, there were some issues dealing with Kuwait that were unaddressed in the U.S. newspaper coverage of the Middle East. These issues were quite unique to Kuwait. They appear in the Table 5 4 below. Some of these issues include municipal governments, elections, specific

PAGE 70

70 ues were not shown in the U.S. media shows that they were silenced in the U.S. agenda of issues. Since only articles dealing with the Middle East were included in the sample of U.S. newspapers, the issues that appeared in the coverage of U.S. news whose ap pearance were unique appear in Tables 5 5 in order of most to least frequently appearing category. The issues include general references to Arabs that do not incl ude mentions of the Arab League and U.S. National Security . Overall, Tables 5 1, 5 2 and 5 3 show that the frequency of nations, issues and religion which appear in U.S. and Kuwaiti newspapers. When the issues from U.S. and Kuwaiti newspapers, shown in Table 5 Order Correlation, the correlation was signif icant (r= 0.60, p=0.02). In other words, evidence of agenda setting is shown as priority of issues in Kuwait and the United States appears in a similar order. Similarly, when the agenda of nations was tested across U.S. and Kuwaiti newspapers, the resul ts were statistically significant. The information shown in Table 5 2 was compared Order Correlation, and a correlation was found (r=0.62, p<0.001). Finally, the agenda of religions, shown in Table 5 3, were tested. The results w ere also statistically significant (r=0.42, p=0.08). Once again, evidence of agenda setting is shown. Regardless , Tables 5 4 and 5 5 shows that there are issues unique to newspapers in Kuwait and the United States. For example, the U.S. newspapers were m ore concerned with U.S. national security than Kuwaiti newspapers. Additionally, Kuwaiti newspapers were more concerned with local issues, such as municipal governance. Hence, the first hypothesis was partially supported.

PAGE 71

71 The second hypothesis states, issues than social media in Kuwait regarding news in the Middle East . test this hypothesis, the frequencies of the objects which appear in U.S. and Kuwaiti newspapers were placed into tables. T he coded nations include Middle Eastern nations as well as the nations of the United Nations Security Council. The Middle Eastern nations consist of the 22 states that make up the Arab league in addition to Iran, Pakistan, Israel, Turkey, and Afghanistan. The United Nations Security Council nations include the United States, Russia, China, France, and the United Kingdom. Below, Table 5 7 shows the frequencies of the appearance of these nations, which were not statistically different (r=0.13, p=0.46). T he coded religions include spiritual or political sects of Islam, Christianity and Juda i sm. These include Islam (general), Sunni, Shiite, Alawite, Wahabi, Salafi, Ismaeli, Zaidi, Muslim Brotherhood, Nation of Islam, Judaism (general), Zionism, Christianit y (general), Catholic, Coptic, and Orthodox. Moreover, two spiritual categories were included from the Muslim perspective, including purity and impurity. Table 5 8 shows these results in a table, which were not statistically different (r=0.15, p=0.56). The frequencies of the politica l issues in Kuwaiti and U.S. social media appear in Table 5 6 . Order Correlation, they did not appear to be significantly different (r=017, p=0.49). Two other tables were created to show t he issues that only Kuwaiti social media messages addressed (Table 5 10) and the issues addressed only by the U.S. social media users (Table 5 11) . In Kuwait, the posts dealt with some domestic political issues. Some differences were found among the soci al media by outspoken individuals in the two nations. While these tables cannot be compared statistically, it is clear that there are some

PAGE 72

72 issues that do not appear in social media posts from other nations. Therefore, the second hypothesis was partially s upported. high salient issues in Kuwaiti social media cont ent will be congruent with the issues of traditional media content in Kuwait . In order to test this hypothesis, nd traditional media were added together. This total was then sorted in descending order to reveal the high salient issues and low salient issues. The results were categorized in three tables. Table 5 1 1 shows the ranking of political issues, Table 5 1 2 shows the ranking of nations and Table 5 1 3 shows the ranking of religions. setting research, the number seven was used to pick the top seven low salient or high salient issues. However, sometimes the data did not lend itself to this rule and exceptions will be explained where they were made. Next, a rank order correlation was conducted on the three topical areas for which nations, strength of the ties between social media and newspapers in Kuwait. This is a typically used to test similarities between two agendas in agenda setting literature (Miller and Wanta, 1996; Kiousis, 2005). Overall correlations between the different media show the strength of ties among the different agendas. T able 5 1 4 shows the Rank Order Correlation values for issues, Table 5 1 5 shows the Rank Order Correlations for nations and Table 5 1 6 shows the Rank Order Correlation for religions. The results suggest that the different media have similar agendas regardless of the topic, albeit to different extents.

PAGE 73

73 For the purposes o f the third hypothesis, t he fr equencies were analyzed, and a threshold was developed to distinguish between high salient and low salient issues. Since the third hypothesis deals only with Kuwaiti media, f requencies of the occurrence of e ach category were summed for Kuwaiti newspapers and social media. Order Correlation was conducted to measure the similarity of the agendas of the two media. The seven issues which were considered to be high salient in Kuwaiti media are: media ( N = 4 99), democracy ( N = 267), education (N= 232), health ( N=199), crime (N=118), Islamic issues (N=87), GCC= (N Order correlation coefficient was positive, but not significant (r=0.42, p=0.33). The test was not significant because democ racy was the most salient issue in Kuwaiti newspapers, but media was the most salient issue in Kuwaiti social media. While education was the second most salient issue in Kuwaiti social media, it was the third most salient issue in Kuwaiti newspapers. Isl amic issues were the fourth most salient issue in Kuwaiti social media, but the sixth most salient issue in Kuwaiti newspapers. T he seven high salient nations in Kuwait are : Kuwait ( N =1,734), Syria ( N =292), U.S. ( N =291), Egypt ( N =203), Iran ( N =93), Russi a ( N =90) and Saudi Arabia ( N= 81). The Order correlation coefficient was not significant (r=0. 46, p=0.29). It was not significant because the most salient nations in Kuwaiti newspapers were: Kuwait, Syria, the U.S., Egypt, Iran, Saudi and Saudi Arabia respectively. However, Kuwaiti social media prioritized Saudi Arabia in third place after Kuwait and the United States. Egypt was the fourth category in Kuwaiti social media, while Iran was the seventh. Only two issues were high salient is a threshold of the frequencies which qualify a category to be a low salient issue was developed, only two religions occurred more than 50 times the maximum of the low salient category.

PAGE 74

74 These two catego ries are general references to Islam ( N= 81) and th e Muslim Brotherhood (N= 61). Since the statistical tests were not significant, it is evident that the issue agendas of Kuwaiti newspapers and social media are not similar . Thus, th e third hypothesis was not supported. high salient issues in U.S. social media content wil l be congruent with the traditional media content in the United States . In order to test this hypothesis, the high salient issues in U.S. media were determined. Then, the two media agendas were compared by a rank order correlation . The seven high salient issues in U.S. social media and newspaper agendas are: democ racy (N=940), m edia (N=927), history (N= 682), terrorism ( N=414), education (N= 246), crime ( N= 203) and health ( N= 199). The two agendas of issues were compared with a Spearman Rank Order C orrelation, and the results showed an insignificant relationship (r=0.18, p=0.69). It was not significant because the issue of democracy was most salient in the U.S. newspaper sample while media issues were most salient in the U.S. social media sample. While the second most salient issue in social media was education, the second most sa lient issue in U.S. newspapers dealt with the media. The seven high salient nations in U.S. social media and newspaper agendas are: the United States (N= 4,671), Syria ( N= 2,087), Russia ( N= 727), Iran ( N= 628), Israel ( N= 376), Iraq ( N= 260) and the United King dom ( N= 182). The two agendas of nations were compared with a Spearman Rank Order C orrelation, and the results showed a strong relationship (r=0.81, p=0.02). The statistical significance shows that U.S. social and traditional media agreed on the top three most salient nations: first the U.S., then Syria and finally Russia.

PAGE 75

75 Only four religions made up the high salient religion category, by exceeding the threshold for low salient categories. The high salient religions in U.S. social media and newspaper agen das are: general references to Judaism ( N= 88), Shiite Islam ( N= 80), general refer ences to Christianity (N= 79) and Sunni Islam ( N= 52). The two media agendas of religions were compared with a Spearman Rank Order Correlation, and the results were not significant (r=0.77, p=0.22). Both U.S. social media and newspapers mentioned Judaism the mos. However, Shiite Islam, Christianity and Sunni Islam were tied for second place in U.S. social media. Newspapers discussed Shiite Islam more than Christianity , which was more salient than Sunni Islam. Only one of the three tests was statistically significant. Although the two media agendas differed when dealing with political issues and religion, they were similar with regards to nations. Therefore, the fourth hypothesis is partially supported. The high salient affective attributes in Kuwaiti social media content will be congruent with the affective traditional media content in Kuwait . address this hypothesis, t ables were created to demonstrate the frequency of positive , negative and neutral coverage of issues. In order to understand the affective attributes associated with objects in the Kuwaiti media, Table 5 1 7 shows the tones of issues in Kuwaiti newspapers, Table 5 18 shows the tones of nations in Kuwaiti newspapers, Table 5 19 shows the tones of religions in Kuwaiti newspapers. Issues which only appeared in Kuwaiti newspapers are shown in Table 5 20 with their affective attributes. Then, issues in Kuwaiti social media are shown in Tables 5 21 , nations form Kuwaiti social media were shown in Table 5 22 and religions from Kuwaiti social media were shown in Table 5 23 .

PAGE 76

76 Some issues, nations and religions did not appear at all in Kuwaiti social media. These issues were removed from Table 5 21. These issues are: Arab League, Europe or t he European Union, Hezbollah/Ha mas, West Bank/Gaza, Global warming, and Intelligence . Some nations did not appear and were removed from Table 5 22, including: Comoros, Mauritani a, Afghanistan, Israel, Algeria, Djibouti, Somalia, Yemen, and France . Finally, some religions were removed from Table 5 23, including Coptic, Orthodox, Judaism, Zionism, Christianity, Alawi, Wahabi, Salafi, Ismaeli, Zaidi, Sunni, Purity, and Impurity . S ome issues did not appear in the Table 5 24, showing tones of issues unique to Kuwait, such as Vacation, Estejwab, and Dow . impactful on public opinion than positive or negati ve tones , all the negative tones o f the objects were collected and categorized by media. Table 5 25 shows the issues, Table 5 26 shows the nations and Table 5 27 shows the religions. This table is important because it shows the m ost and least frequent issues i n Kuwait . To answer the fifth research hypothesis, the number of negative attributes in Kuwaiti social media and newspapers by object were added. Then, the high salient affective attributes associated with issues were determined by ordering the is sues by rank. With regards to issues, high salient of the negative attributes in Kuwaiti media were associated with several issues. Included in this category are six of the categories coded for in data from both the U. S. and Kuwait: media (N= 123), democra cy ( N=91), education (N=77), health (N= 65), crime ( N= 63), and traffic (N =53). An additional five categories of issues unique to Kuwait were added to this list: government ( N= 129), housing ( N= 94), national assembly ( N= 92), ministers ( N= 57) and courts ( N= 53) . All 11 categories were combined and agendas of issues in Order

PAGE 77

77 Correlation, which showed that they are different (r= 0.04, p=0.89). The correlation was not significant because newspaper democracy, media, education, crime, health, housing, courts, ministers and then traffic. In social media, the order of the priorities was different with media being first, housing being second, g overnment being third, education fourth and traffic fifth. These differences demonstrate that different media can lead to agenda diversity, where different topics are salient depending on the medium. With regards to nations, the categories associated wit h the most negat ive attributes in Kuwaiti media were: Kuwait ( N= 472), Syria ( N= 128), Egypt ( N= 71) and the U.S. ( N= 69). There were only four nations that exceeded the threshold of the minorities. Rank Order Correlation showed that the affec tive attributes linked to objects are different in Kuwaiti social media and newspapers (r=0.20, p=0.80). It was not significant because Kuwaiti newspapers prioritized the high salient nations in the following order: Kuwait, Syria, Egypt and finally the Uni ted States. However, Kuwaiti social media showed a different order: Kuwait, then the United States, then Egypt and finally Syria. The differences in priority show agenda diversity by medium. Since religion did not appear more than 50 times in a negativ e tone, there were no high salient affective attributes in Kuwaiti media. The findings show that Kuwaiti traditional and social media differ with regards to the objects in the media associated with the most negative attributes. Hence, the fifth hypothesis was not supported.

PAGE 78

78 The high salient affective attributes in U.S. social media content will be congruent with the affective traditional medi a To address this question, two tables were created. First, there is a table comparable to the two tables from the fifth hypothesis. It contains frequencies and Chi Squares of the objects per tone. The second table is a lit tle bit different however. The unit of analysis of the U.S. newspapers slightly differs. Each observation symbolizes an articles. Within that article, all of the graphs were counted and coded for objects and non directional tone. In order to obtain an e stimate of the tone per object, the total number of graphs with a specific tone (positive, negative or neutral) were divided by the total number of graphs. Then, the same was done for each object which appears (issues, nations, religions). Finally, the r atio for tone was divided by the ratio for the object. Following basic mathematical rules, the new ratio would give us the amount of tone (positive, negative or neutral) that appears per object (issues, nations and religions). Table 5 29 shows data from U.S. social media. However, since some categories did not appear at all in this table they were removed. These categories are: Globalization, Children's issues, Islamic Issues, Arab League, Health, UN, Traffic, Europe, Oil,Protests, West Bank/G aza, Gulf Countries Council , Kuwait, Egypt, Palestine, Algeria, Djibouti, Somalia, Libya, Bahrain, Oman, Qatar, UAE, Yemen, Saudi Arabia, Tunisia, Morocco, Sudan, Lebanon, Jordan, Comoros, Mauritania, Iran, Pakistan, Israel, Turkey, Afghanistan, UK, Franc e, Promotion, Corruption, National Assembly, Ministers, Prime Minister, Elections, Stateless People, Specific Law, Raqaba, Estejwab, Government, Courts, Municipality, Tribalism, Cost of Living, Aviation, Housing, Vacation, Energy, Water issues, Electricity , Dow, Borsa, Employment, Orthodox,

PAGE 79

79 Coptic, Zionism, Christianity, Islam (general), Sunni, Shiite, Alawi, Wahabi, Salafi, Ismaeli, Zaidi, Muslim Brotherhood, Purity, and Impurity. Table 5 30 shows data from U.S. newspapers. Once again, some categories d id not appear and were removed from the tables. These categories are: West Bank/Gaza, Globalization, Children's issues, Women's issues, Historical, Sports, Islamic Issues, Arab League, Health, Protests, Oil, Kuwait, Egypt, Saudi Arabia, Tunisia, Morocco, Sudan, Lebanon, Palestine, Algeria, Djibouti, Somalia, Libya, Traffic, Bahrain, Oman, Qatar, UAE, Yemen, Jordan, Comoros, Mauritania, Pakistan, Israel, China, France, Purity, Impurity, Islam (general), Sunni, Shiite, Alawi, Wahabi, Salafi, Ismaeli, Zaidi, Muslim Brotherhood, Judaism, Zionism, Christianity, Coptic, and Orthodox . Once again, s are more impactful on public opinion than positive or negative tones. Therefore, all the negative tone s of the objects were collected in a table and categorized by media. Table 5 31 displays this information. The sixth hypothesis will be addressed slightly differently than the fifth hypothesis due to differences in the units of data collected. Above, I d escribed the formula for finding the amount of negativity per graph. In order to gain a number that is similar to those used for data gathered in the U.S. social media, a process was developed. The process began by removing all the variables where absolut ely no mention occurred in either the U.S. social or traditional media. Next, the numbers which remained ranged between 0 and 35. Then, the average numbers of graphs per article, 20.43, was multiplied to the value which represents the negative tone per o bject. Finally, the results showed a range of frequencies that vary between 14.89 and 715.05. These values were summed with the frequencies of U.S.

PAGE 80

80 social media to determine whether they are high salient or low salient issues. The seven most frequently o ccurring issues, which had a negative tone, and had a frequency that is more than 50 times were high salient issues. The only five nations with enough affective attributes to qualify as high salient nations are: United Kingdom ( N= 715.05), Turkey ( N= 715.05) , Afghanistan ( N= 357.25), Iran ( N= 143.01), and United States ( N= 92.60). The affective attributes of the high salient issues in U.S. social media and newspapers were an inverse and insignificant relationship (r= 0.72, p=0.16). It was not significant beca use only the United States appeared in U.S. social media, whereas Turkey and the United Kingdom had the most negative affective attributes in U.S. newspapers followed by Afghanistan and Iran. In U.S. social media, no negative attributes were associated wi th any nation other than the United States. This suggests that some attributes associated with nations are being silenced in U.S. social media. The only religion to qualify as a high salient religion was Catholicism, which was not mentioned in U.S. social media. While there is not enough information to compare in a statistical test, the observation shows a difference between the two media. The lack of information also implies that a situation exists where one medium includes information that is silenced from another. The issues that qualified as high salient issues were: mentions of Europe and/or the European Union ( N= 715.05), intelligence ( N= 359.52), media ( N= 250.35), terrorism ( N= 181.76) and the United Nations ( N= 59.58). Order Correla tion showed that the attribute agendas were not similar (r= 0.10, p=0.87). It was not significant because Europe and/or the European Union was associated with the most negative attributes in U.S. newspapers, followed by intelligence and then media. Howe ver, the most negative attributes were associated

PAGE 81

81 to media, then terrorism and then intelligence in U.S. social media. These findings reveal evidence of agenda diversity depending on the medium. The sixth hypothesis suggested congruence between the affect ive attributes in U.S. social and traditional media. The results demonstrate that the agendas were different. Therefore, the sixth hypothesis was not supported. T salient issues from social media content will be incongruent with the issues from traditional media content in Kuwait . hypothesis, the total frequencies of issues from Kuwaiti traditional and social media were added. Issues, nations and religions that were mentioned below 50 times were considered low salient issues. The seven low salient issues in Kuwait are: terrorism ( N= N= 36), N= 21), intelligence ( N= 18), oil ( N= 15), history ( N= 11), and Hezbollah/Hamas ( N= 11). These agendas of these issues were compared for social and traditional media in Kuwait Order Correlation, which was inverse and not significant (r= 0.21, p=0.63). It was not significant because children issues were the most sa lient low salient issue in Kuwaiti social media, but terrorism was the most salient low salient issue in Kuwaiti newspapers. While historical references were the second most salient, low salient issue in ent, low salient In other words, different media show different agendas, thus demonstrating agenda diversity. The low salient nations from Kuwaiti social and traditional media include: the Unit ed Arab Emirates (N = 45 ), United Kingdom (N = 44 ), France ( N = 35 ), Iraq ( N =35 ), Jordan ( N = 33 ), Qatar ( N = 19), and Yemen ( N =1 7 ) . These low salient nations were compared with rank order correlation. The correlation coefficient were inverse and insignificant (r= 0.07,

PAGE 82

82 p=0.86). It was not significant because newspapers rated France, Iraq and the United Arab Emirates as the most salient nations in Kuwaiti newspapers. However, the most salient nation in Kuwaiti social media was the United Kingdom, followed by the Unite d Arab Emirates, which was followed by two nations tied at third place: Qatar and Jordan. Again, the order of the ranking of issues across media is not the same. Therefore, there is evidence of agenda diversity. The seven low salient religion categories are: Salafi Islam ( N= 7), Coptic Christianity ( N= 6), Shiite Islam ( N= 5), general references to Christianity ( N= 5), Sunni Islam ( N= 2), Catholicism ( N= 1), Alawite Islam ( N= 1). When the order of these religions were compared in social and traditional media, t he results were inverse and statistically insignificant (r= .40, p=0.37). It was not significant because the low salient religions in Kuwaiti newspapers appeared in the following order from most to least salient: Salafi Islam, Coptic Christianity, general Christianity, Shiite Islam, Sunni Islam and Catholicism. However, in Kuwaiti social media, Shiite Islam was the most prominent religion followed by Catholicism. All the other low salient religions were tied. Here, evidence of agenda diversity was reveale d. To summarize, the seventh hypothesis predicts that low salient issues in Kuwaiti newspapers and social media will be incongruent. The statistical tests were not significant, which suggests that the agendas are not similar. Therefore, the seventh hypot hesis was supported. Low salient issues from U.S. social media content will be incongruent with the issues from traditional media content in the United States . hypothesis, the same objects from the previous hyp othesis were tested; however, the agendas were from traditional and social media in the United States .

PAGE 83

83 Once again, the frequencies for U.S. social and traditional media were summed. The totals were ranked, and a threshold of 50 frequencies was the maximum value to qualify a category as a low salient category. N= 50), Gulf Countries council ( N= 43), Arab league ( N= 38), oil ( N= 35), global warming ( N= 11), and West Bank/Gaza ( N= 10). The two agendas of U.S. social and traditional media were compared Order Correlation. The results were inverse and not significant (r= 0.16, p=0.74). low salient issue in U.S. soci al media and newspapers. However, global warming was the second most salient low salient issue in U.S. social media, while the issue of the Gulf Cooperation Council was the second most salient low salient issue in U.S. newspapers. In U.S. social media, the Gulf Cooperation Council was tied at third place with the Arab League , West Bank/Gaza and oil. The Arab League was the third most salient low salient issue in U.S. newspapers and oil was the fourth. W hat makes these results even more interesting is the fact that four of the issues at third place did not appear once in U.S. social media. This suggest s that these issues were silenced in U.S. social media. The low salient nations were considered: Turkey ( N= 49) , Pakistan ( N= 37) , Jordan ( N= 29) , Qatar ( N= 27), the United Arab Emirates ( N= 18), Bahrain ( N= 16), and Sudan ( N= 11). However, since none of these nations appeared in U.S. social media, a statistical test was not conducted. The low salient religions wer e considered: the Muslim Brotherhood ( N= 46), general references to Islam ( N= 31), Orthodox Christianity ( N= 12), Catholicism ( N= 12), Alawite Islam ( N= 11), Zionism ( N= 5), and Salafi Islam ( N= 2). Order Correlation, the res ults were inverse and not statistically significant (r= 0.10, p=0.82 ). It

PAGE 84

84 was not significant because Catholicism was the only low salient religion to appear in U.S. social media, while the Muslim Brotherhood was the most salient low salient religion in U .S. newspaper articles dealing with the M iddle East. U.S. social media did not mention a religion other than Catholicism, which appeared once. However, after the Muslim Brotherhood, U.S. newspapers mentioned the following low salient religions : general I slam, Orthodox Christianity, Catholicism and Alawite Islam , Zionism and Salafi Islam respectively. Again, this suggests that some religions are silenced in social media though not newspapers. To summarize, low salient issues did not appear in a similar or der between U.S. social and traditional media. Therefore, the eighth hypothesis was supported. salient affective attributes from social media content will be incongruent with the affective attributes from traditional media content in Kuwait . In Kuwaiti newspapers and social media, six religious categories appeared in a negative tone fewer than 50 times. These categories are: general references to Islam ( N= 26), Muslim Brotherhood ( N= 20), Salafi Islam ( N= 5), Coptic Christianity ( N= 2), Christianity ( N= 1), and Shiite Islam ( N= 1). correlation between the two agendas (r=0.85, p=0.02). In other words, the order of the categories was simil ar. This is evidence of the agenda setting hypothesis. The seven low salient nations mentioned in a negative tone in Kuwaiti media are: Russia ( N= 32), Iran ( N= 20), the United Arab Emirates ( N= 16), Jordan ( N= 16), Saudi Arabia ( N= 15), Tunisia ( N= 14) and the United Kingdom ( N= Order Correlation revealed an inverse and statistically significant relationship (r= 0.88, p=0.007). Statistical

PAGE 85

85 significance means that the priority of the issues on the agenda is similar across Kuwai ti social and traditional media, which is evidence of the agenda setting hypothesis. The low salient issues combined issues that only appeared in Kuwait with issues analyzed in media from two nations. Once again, low salient issues appear a total of 50 or less tim es in the sample from Kuwait. These issues are: Islamic issues ( N= 42), Gulf Countries Council ( N= 23), protests ( N= N= 19), United Nations ( N= 12), intelligence ( N= 8), sports ( N= 7). These also include issues unique to Kuwait such as: aviation ( N= 40), corruption ( N= 26), municipality ( N= 25), and prime minister ( N= 23). These issues did not appear in a similar order in the two media (r= 0.18, p=0.60). Kuwaiti newspapers and social media agreed on the order of the top three issues: the G ulf Cooperation Council, followed by low salient issues appeared in Kuwaiti social media. This suggests that other low salient issues were silenced. The ninth hypothesis predicted incongruence amon g social and traditional media in Kuwait with regards to affective attributes associated with low salient issues. The findings suggest this may be true with regards to religions and nations, but it is not true when analyzing issues. Therefore, the ninth hypothesis is partially supported. Low salient affective attributes from U.S. social media content will be incongruent with the affective attributes from traditional media content in the United States . The steps for determi ning a low salient issue are similar to those outlined in the sixth hypothesis for high salient issues. However, low salient issues are defined as occurring 50 or less times.

PAGE 86

86 Four issues qualified as low salient issues: democracy ( N= 1), Hezbollah/Hamas ( N= 1), global warming ( N= 1), and education ( N= 6). However, since these issues were not associated with negative attributes in U.S. newspapers, no statistical tests were used to compare the two agendas. Only two nations qualified as low salient nations: Sy ria ( N= 46.63) and Iraq ( N= 1). Since there were only two categories, no statistical test was used to analyze this data. None of the religions qualified as a low salient religion. Religion was not discussed frequently in a negative tone in both media. The re is not enough information to make any conclusions here. The tenth category states that the agendas of low salient issues associated with negative attributes in U.S. traditional and social media will not be similar. Our findings show some degree of diff erences. Therefore, this hypothesis is supported. Results of Research Questions Are issues covered by the traditional media in Kuwait related to the U.S. social media content that is incongruent with the U.S. traditional media T he traditional media covered the Syrian conflict from multiple angles. The results show that there is very little lack of congruence between social and traditional media in the United States. Therefore, it is hard to make any conclu sions about the issues covered by traditional media in Kuwait and their relationship to U.S. social media. However, one of the two Kuwaiti social media accounts did mention U.S. entertainment and broadcast media. For example, Kuwaiti lady lose custody of her children because of bad behavior in social networking applications! #kuwait #cnn @nbcnightlynews @jayleno broadcast media content may be more congruent with Kuwaiti social media content than U.S.

PAGE 87

87 newspapers content. However, the current analysis does not include any samples from broadcast media. traditional media in Kuwait related to U.S. social media content t hat is incongruent with the U.S. from both U.S. social media and Kuwaiti traditional media. One such instance is clear. Alqabas covered the Syrian conflict w ith major headlines and images of the Syrian Free Army preparing to launch a morter towards Deir Ezzor. The headlines . the article explains that a military strike against Syria is awaiting a green light from the U.S. Congress. In U.S. social media, one such example of the Syrian conflict appearing includes a anyone not on Such a social media message includes messages about Obama and Congress, which both appear in the Kuwaiti social media message. What is interesting about t his message is that the U.S. social media message is critical of the Obama administration. In the Kuwaiti newspaper, the issue is covered in a neutral tone. related to the Kuwaiti social media content that is incongruent with the Kuwaiti traditional media? One area of incongruence appears with regards to immigration in the U.S. social media. ars for Fatal Gang

PAGE 88

88 issue appeared in the U.S. social media, it did not appear in the Kuwaiti traditional media sample. ctive attributes linked to objects covered by traditional media in the U.S. related to the Kuwaiti social media content that is incongruent with media and U.S. trad itional media, it is hard to make any kind of conclusion about the affective attributes linked to the issues covered by U.S. social media as they relate to Kuwaiti social media.

PAGE 89

89 Table 5 1. Frequencies of appearance of issues in newspapers in the United States vs. Kuwait Issue USA Kuwait Democracy 939 257 Media 884 234 Historical figures 681 5 Terrorism 411 48 Education 235 200 Children Issues 128 13 Health 118 177 Hezbollah /Hamas 63 11 Sports 53 23 Women Issues 49 32 Oil 35 14 Arab league 38 10 Global Warming 9 4 *p= 0.02, Spearman rank correlation= 0.60

PAGE 90

90 Table 5 2. Frequencies of appearance of nations in newspapers in the United States vs. Kuwait Nation USA Kuwait Kuwait 7 1474 Egypt 135 185 Syria 2074 282 Saudi Arabia 81 53 Tunisia 2 13 Morocco 3 0 Sudan 11 5 Lebanon 61 22 Iraq 259 35 Palestine 75 9 Algeria 4 1 Djibouti 0 0 Somalia 82 2 Libya 111 2 Bahrain 16 8 Oman 4 2 Qatar 27 15 UAE 18 35 Yemen 11 17 Jordan 29 29 Comoros 0 0 Mauritania 0 1 Iran 628 92 Pakistan 37 2 Israel 376 9 Turkey 48 16 Afghanistan 117 3 USA 4542 250 Russia 721 88 China 131 3 United Kingdom 182 32 France 158 35 *p<0.001, Spearman rank correlation=0.62

PAGE 91

91 Table 5 3. Frequencies of appearance of religions in newspapers in the United States vs. Kuwait Religions USA Kuwait Islam (general) 3 1 50 Sunni 52 2 Shiite 80 3 Alawaite 11 1 Wahabis 0 0 Salafis 2 7 Ismaeli 0 0 Zaidi 0 0 Muslim Brot herhood 46 55 Nation of Islam 4 Judaism (general) 87 0 Zionism 5 0 Christianity (general) 79 5 Catholic 11 0 Coptic 1 6 Orthodox 12 0 Imp urity 1 0 Purity 0 0 *p=0.08, Spearman rank correlation=0.42 Table 5 4. Frequencies of appearance of issues unique to Kuwaiti newspapers Category Kuwait Government 334 Corruption 27 Elections 33 Municipal governance 52 Ministers 246 Cost of Living 9 Financial markets 31 Court 117 Specific Laws 37 Water issues 18 Electricity 33 Prime Minister 111 National assembly 286 Privatization 3 Estejwab 13 Consumer loans 21 Stateless Peple 11 E mployment 40

PAGE 92

92 T able 5 5 . Frequencies of appearance of issues unique to United States newspapers Category USA Arab 122 National Security 95 Table 5 6. Frequencies of Issues in Social Media in the USA vs. Kuwait Issue USA Kuwait Media 43 265 Sports 1 34 Education 11 32 Crime 4 13 Health 0 22 Islamic issues 0 19 Democracy 1 10 Children Issues 0 8 Historical figures 1 6 GCC 0 6 Intelligence 5 0 Global Warming 2 0 Women Issues 1 4 Terrorism 3 1 Hezbollah /Hamas 1 0 West Bank/Gaza 0 0 Arab league 0 0

PAGE 93

93 Table 5 7. Frequencies of Nations in Social Media in the USA vs. Kuwait Nation USA Kuwait Kuwait 0 260 USA 129 41 Saudi Arabia 0 28 Egypt 0 18 Syria 13 10 United Kingdom 0 12 UAE 0 10 Lebanon 0 7 Bahrain 0 7 Russia 6 2 Oman 0 4 Qatar 0 4 Jordan 0 4 Palestine 0 3 Sudan 0 3 Turkey 0 2 Pakistan 0 2 China 1 1 Iraq 1 0 Libya 0 1 Iran 0 1 Tunisia 0 1 Morocco 0 1 Algeria 0 0 Djibouti 0 0 Somalia 0 0 Comoros 0 0 Mauritania 0 0 Israel 0 0 France 0 0 Afghanistan 0 0 Yemen 0 0 Order Correlation=0.13

PAGE 94

94 Table 5 8. Frequencies of Religions in Social Media in the USA vs. Kuwait Religion USA Kuwait Islam (general) 0 31 Sunni 0 0 Shiite 0 2 Alawaite 0 0 Wahabis 0 0 Salafis 0 0 Ismaeli 0 0 Zaidi 0 0 Muslim Brotherhood 0 6 Nation of Islam Judaism (general) 1 0 Zionism 0 0 Christianity (general) 0 0 Catholic 1 1 Coptic 0 0 Orthodox 0 0 Impurity 0 0 Purity 0 0 *p=0.56, Spearman Rank Order Correlation=0.15

PAGE 95

95 Table 5 9 . Frequency of Issues Unique to Social Media Issue Kuwait Government 33 Corruption 20 Elections 15 Municipal governance 10 Ministers 8 Cost of Living 5 Financial M arkets 4 Court 3 Specific Laws 2 Water Issues 2 Electricity 1 Prime Minister 1 National A ssembly 1 P rivatization 1 Estejwab 0 Consumer L oans 0 Table 5 1 0 . Freque ncy of Issues Which Appear Only in U.S. Social Media Category US Politics 48 Immigration 10 Healthcare 7 Economics 5 Abortion 1 Marijuana 1

PAGE 96

96 Table 5 1 1 . Frequencies and Ranking of Issues According to Medium Issue U.S. Social Media Kuwait Social Media U.S. Newspapers Kuwait Newspapers Total Ranking Media 43 265 884 234 1426 1 Democracy 1 10 939 257 1207 2 Historical figures 1 6 681 5 693 3 Education 11 32 235 200 478 4 Terrorism 3 1 411 48 463 5 Crime 4 13 199 105 321 6 Health 0 22 118 177 317 7 Intelligence 5 0 153 18 176 8 Children Issues 0 8 128 13 149 9 Sports 1 34 53 23 111 10 GCC 0 6 43 55 104 11 Islamic issues 0 19 68 87 12 Women Issues 1 4 49 32 86 13 Hezbollah /Hamas 1 0 63 11 75 14 Oil 0 1 35 14 50 15 Arab league 0 0 38 10 48 16 Global Warming 2 0 9 4 15 17 West Bank/Gaza 0 0 10 0 10 18 Notes: When adding the Kuwaiti newspaper and social media columns, the frequencies for high salient issues in Kuwait are: media (N=499), democracy (N=267), education (N=232), health (N=199), crime (N=118), Islamic issues (N=87), and GCC (N=61). When adding the U.S. newspaper and social media columns, the frequencies for high salient issues in the United States are: democracy (N=940), media (N=927), history (N=682), terrorism (N=414), educatio n (N=246), crime (N=203), and health (N=199). When adding the Kuwaiti newspaper and social media columns, the frequencies for low salient intelligence (N=18), oil (N= 15), history (N=11), and Hezbollah/Hamas (N=11) . When adding the U.S newspaper and social media columns, the frequencies for low salient League (N=38), oil (N=35), glo bal warming (N=11), West Bank/Gaza (N=10).

PAGE 97

97 Table 5 1 2 . Frequencies and Ranking of Nation s According to Medium Notes: When adding the Kuwaiti newspaper and social media columns, the frequencies for high salient nations in Kuwait are: Kuwait (N=1,734), Syria (N=292), United States (N=291), Egypt (N=203), Iran (N=93), Russia (N=90), and Saudi Arabia (N=81). Nation U.S. Social Media Kuwait Social Media U.S. Newspapers Kuwait Newspapers Total Ranking USA 129 41 4542 250 4962 1 Syria 13 10 2074 282 2379 2 Kuwait 0 260 7 1474 1741 3 Russia 6 2 721 88 817 4 Iran 0 1 628 92 721 5 Israel 0 0 376 9 385 6 Egypt 0 18 135 185 338 7 Iraq 1 0 259 35 295 8 United Kingdom 0 12 182 32 226 9 France 0 0 158 35 193 10 Saudi Arabia 0 28 81 53 162 11 China 1 1 131 3 136 12 Afghanistan 0 0 117 3 120 13 Libya 0 1 111 2 114 14 Lebanon 0 7 61 22 90 15 Palestine 0 3 75 9 87 16 Somalia 0 0 82 2 84 17 Turkey 0 2 48 16 66 18 UAE 0 10 18 35 63 19 Jordan 0 4 29 29 62 20 Qatar 0 4 27 15 46 21 Pakistan 0 2 37 2 41 22 Bahrain 0 7 16 8 31 23 Yemen 0 0 11 17 28 24 Sudan 0 3 11 5 19 25 Tunisia 0 1 2 13 16 26 Oman 0 4 4 2 10 27 Algeria 0 0 4 1 5 28 Morocco 0 1 3 0 4 29 Mauritania 0 0 0 1 1 30 Comoros 0 0 0 0 0 31 Djibouti 0 0 0 0 0 11

PAGE 98

98 When adding the U.S. newspaper and social media columns, the frequencies for the high salient nations in the U.S. are: United States (N=4,671), Syria (N=2,087), Russia (N=727), Iran (N=628), Israel (N=376), Iraq (N=260). When adding the Ku waiti newspaper and social media columns, the frequencies for low salient nations i n Kuwait are: United Arab Emirates (N=45), United Kingdom (N=44), France (N=35), Iraq (N=35), Jordan (N=33), Qatar (N=19), and Yemen (N=17). When adding the U.S newspaper and social media columns, the frequencies for low salient nations in United States are: Turkey (N=49), Pakistan (N=37), Jordan (N=29), Qatar (N=27), United Arab Emirates (N=18), Bahrain (N=16) and Sudan (N=11).

PAGE 99

99 Table 5 13 . Frequencies and Ranking of Religions According to Medium Religion U.S. Social Media K uwait Social Media U.S. Newspapers Kuwait Newspapers Total Ranking Islam (general) 0 31 31 50 112 1 Muslim Brotherhood 0 6 46 55 107 2 Judaism (general) 1 0 87 0 88 3 Shiite 0 2 80 3 85 4 Christianity (general) 0 0 79 5 84 5 Sunni 0 0 52 2 54 6 Catholic 1 1 11 0 13 7 Orthodox 0 0 12 0 12 8 Alawaite 0 0 11 1 12 8 Salafis 0 0 2 7 9 10 Coptic 0 0 1 6 7 11 Zionism 0 0 5 0 5 12 Nation of Islam 4 4 13 Impurity 0 0 1 0 1 14 Purity 0 0 0 0 0 15 Wahabis 0 0 0 0 0 15 Ismaeli 0 0 0 0 0 15 Zaidi 0 0 0 0 0 15 Notes: When adding the Kuwaiti newspaper and social media columns, the frequencies for high salient religions in Kuwait are: general Islam (N=81) and Muslim Brotherhood (N=61). When adding the U.S. newspapers and social media columns, the frequencies for the salient religions in the United States are: Judaism (N=88), Shiite Islam (N=80), general Christianity (N=79), and Sunni Islam (N=52). When adding the Kuwaiti newspaper and social media columns, the frequencies for low salient religions in Kuwait are: Salafi Islam (N=7), Coptic Christianity (N=6), Shiite Islam (N=5), general Christianity (N=5), Sunni Islam (N=2), Catholicism (N=1) and Alawite Islam (N=1). . When adding the U .S newspaper and social media columns, the frequencies for low salient religions in United States are: Muslim Brotherhood (N=46), general Islam (N=31), Orthodox Christianity (N=12), Catholicism (N=12), Alawite Islam (N=11), Zionism (N=5), and Salafi Islam (N=2).

PAGE 100

100 Table 5 14 Order Correlation of Media for Issues . U.S. Social Media Kuwait Social Media U.S. Newspapers Kuwait Newspapers U.S. Social Media .190 .565 * .330 p= .449 p= .018 p= .181 N= 18 N= 17 N= 18 Kuwait Social Media .190 .558 * .754 ** p= .449 p= .020 p= .000 N= 18 N= 17 N= 18 U.S. Newspape r s .565 * .558 * .627 ** p= .018 p= .020 p= .007 N= 17 N= 17 N= 17 Kuwait Newspapers .330 .754 ** .627 ** p= .181 p< .00 1 p= .007 N= 18 N= 18 N= 17 **. Correlation is significant at the 0.01 level (2 tailed). *. Correlation is significant at the 0.05 level (2 tailed). Table 5 15 Order Correlation of Media for Nations. U.S. Social Media Kuwait Social Media U.S. Newspapers Kuwait Newspapers U.S. Social Media .135 .573 ** .395 * p= .461 p= .001 p= ..025 N= 32 N= 32 N= 32 Kuwait Social Media .135 .196 .601 ** p= .461 p= .283 p< .00 1 N= 32 N= 32 N= 32 U.S. Newspapesr .573 ** .196 .623 ** p= .001 p= .283 p< . .001 N= 32 N= 32 N= 32 Kuwait Newspapers .395 * .601 ** .623 ** p= .025 p<.001 p< . 001 N= 32 N= 32 N= 32 **. Correlation is significant at the 0.01 level (2 tailed). *. Correlation is significant at the 0.05 level (2 tailed).

PAGE 101

101 Table 5 16 . Order Correlation of Media for Religions. U.S. Social Media Kuwait Social Media U.S. Newspapers Kuwait Newspapers U.S. Social Media .150 .319 .323 p= .565 p= .212 p= . 206 N= 17 N= 17 N= 17 Kuwait Social Media .150 .416 .492 * p= .565 p= .096 p= . 045 N= 17 N= 17 N= 17 U.S. Newspaper s .319 .416 .425 p= . 212 p= . 096 p= .089 N= 17 N= 17 N= 17 Kuwait Newspapers .323 .492 * .425 p= . 206 p= . 045 p= .089 N= 17 N= 17 N= 17 **. Correlation is significant at the 0.01 level (2 tailed). *. Correlation is significant at the 0.05 level (2 tailed).

PAGE 102

102 Table 5 1 7 . Frequency of Issues Per Tone in Kuwaiti Newspapers Issue Negative Percent Neutral Percent Positive Percent Democracy 88 12.7 119 10.7 50 11.6 Hezbollah /Hamas 4 0.6 7 0.6 0 0 West Bank/Gaza 0 0 0 0 0 0 Global warming 2 0.3 1 0.1 1 0.2 Education 67 9.7 98 8.8 35 8.1 Terrorism 19 2.7 26 2.3 3 0.7 Globalization 0 0 2 0.2 0 0 Children's issues 10 1.4 3 0.3 0 0 Women's issues 18 2.6 10 0.9 4 0.9 Historical 0 0 3 0.3 2 0.5 Sports 2 0.3 15 1.4 6 1.4 Islamic Issues 37 5.3 21 1.9 10 2.3 Arab League 3 0.4 7 0.6 0 0 Health 59 8.5 82 7.4 36 8.3 Media 73 10.5 114 10.3 47 10.9 GCC 23 3.3 23 2.1 9 2.1 Intelligence 8 1.2 10 0.9 0 0 UN 11 1.6 26 2.3 2 0.5 Traffic 44 6.3 30 2.7 12 2.8 Europe 3 0.4 10 0.9 2 0.5 Crime 61 8.8 39 3.5 5 1.2 Protests 21 3 9 0.8 0 0 Oil 4 0.6 9 0.8 1 0.2

PAGE 103

103 Table 5 18. Frequency of Nations Per Tone in Kuwaiti Newspapers Nation Negative Percent Neutral Percent Positive Percent Kuwait 419 60.4 711 64.1 344 79.6 Egypt 64 9.2 90 8.1 31 7.2 Syria 124 17.9 144 13 14 3.2 Saudi Arabia 8 1.2 22 2 23 5.3 Tunisia 13 1.9 0 0 0 0 Morocco 0 0 0 0 0 0 Sudan 5 0.7 0 0 0 0 Lebanon 8 1.2 11 1 3 0.7 Iraq 10 1.4 11 1 14 3.2 Palestine 4 0.6 3 0.3 2 0.5 Algeria 1 0.1 0 0 0 0 Djibouti 0 0 0 0 0 0 Somalia 1 0.1 1 0.1 0 0 Libya 1 0.1 1 0.1 0 0 Bahrain 5 0.7 1 0.1 2 0.5 Oman 1 0.1 0 0 1 0.2 Qatar 1 0.1 8 0.7 6 1.4 UAE 14 2 17 1.5 4 0.9 Yemen 1 0.1 11 1 4 1.2 Jordan 16 2.3 9 0.8 4 0.9 Comoros 0 0 0 0 0 0 Mauritania 1 0 0 0 0 0 Iran 20 2.9 57 5.1 15 3.5 Pakistan 1 0.1 1 0.1 0 0 Israel 7 1 2 0.2 0 0 Turkey 5 0.7 11 1 0 0

PAGE 104

104 Table 5 18. Continued. Afghanistan 1 0.1 1 0.1 1 0.2 USA 60 8.6 157 14.1 33 7.6 China 0 0 3 0.3 0 0 UK 7 1 19 1.7 6 1.4 France 10 1.4 16 1.4 9 2.1 Russia 32 4.6 48 4.3 8 1.9

PAGE 105

105 Table 5 1 9 . Frequency of Religion Per Tone in Kuwaiti Newspaper Religion Negative Percent Neutral Percent Positive Percent Purity 0 0 0 0 0 0 Impurity 0 0 0 0 0 0 Islam (gene ra l) 21 3 16 1.4 13 3 Sunni 0 0 0 0 2 0.5 Shiite 1 0.1 0 0 2 0.5 Alawi 0 0 1 0.1 0 0 Wahabi 0 0 0 0 0 0 Salafi 5 0.7 2 0.2 0 0 Ismaeli 0 0 0 0 0 0 Zaidi 0 0 0 0 0 0 Muslim Brotherhood 18 2.6 36 3.2 1 0.2 Judaism 0 0 0 0 0 0 Zionism 0 0 0 0 0 0 Christianity 1 0.1 2 0.2 2 0.5 Catholic 0 0 0 0 0 0 Coptic 2 0.3 2 0.2 2 0.5 Orthodox 0 0 0 0 0 0

PAGE 106

106 Table 5 20 . Frequency of Unique Issues Per Tone in Kuwaiti Newspaper Issue Negative Percent Neutral Percent Positive Percent Water issues 5 0.7 9 0.8 4 0.9 Electricity 9 1.3 17 1.5 7 1.6 Dow 0 0 0 0 0 0 Borsa 10 1.4 11 1 10 2.3 Employment 5 0.7 26 2.3 9 2.1 Cost of Living 4 0.6 4 0.4 1 0.2 Aviation 20 2.9 22 2 7 1.6 Housing 47 6.8 37 3.3 20 4.6 Vacation 0 0 0 0 0 0 Energy 1 0.1 3 0.3 1 0.2 Promotion 0 0 2 0.2 0 0 Corruption 13 1.9 7 0.6 7 1.6 National Assembly 92 13.3 135 12.2 59 13.7 Ministers 54 7.8 132 11.9 60 13.9 Prime Minister 23 3.3 49 4.4 39 9 Elections 15 2.2 16 1.4 2 0.5 Stateless People 1 0.1 9 0.8 1 0.2 Speicific Law 18 2.6 10 0.9 9 2.1 Raqaba 9 1.3 8 0.7 8 1.9 Estejwab 9 1.3 4 0.4 0 0 Government 108 15.6 157 14.1 69 16 Courts 51 7.3 49 4.4 17 3.9 Municipality 25 3.6 15 1.4 12 2.8 Tribalism 1 0.1 2 0.2 1 0.2

PAGE 107

107 Table 5 21 . Frequency of Issues Per Tone in Kuwaiti Social Media Issue Negative Percent Neutral Percent Positive Percent Democracy 3 2.7 5 2.2 2 1.3 Education 10 9.1 14 6.1 8 5.3 Terrorism 0 0 1 0.4 0 0 Globalization 0 0 10 4.3 13 8.6 Children's issues 2 1.8 3 1.3 3 2 Women's issues 1 0.9 1 0.4 2 1.3 Historical 2 1.8 1 0.4 3 2 Sports 5 4.5 17 7.4 12 7.9 Islamic Issues 5 4.5 7 3 7 4.6 Health 6 5.5 9 3.9 7 4.6 Media 50 45.5 118 51.3 97 64.2 GCC 0 0 6 2.6 0 0 UN 1 0.9 5 2.2 2 1.3 Traffic 9 8.2 4 1.7 3 2 Oil 0 0 1 0.4 0 0 Crime 2 1.8 11 4.8 0 0

PAGE 108

108 Table 5 22 . Frequency of Nations Per Tone in Kuwaiti Social Media Nation Negative Percent Neutral Percent Positive Percent Kuwait 53 48.2 131 57 76 50.3 Egypt 7 6.4 8 3.5 3 2 Syria 4 3.6 6 2.6 0 0 Saudi Arabia 7 6.4 13 5.7 8 5.3 Tunisia 1 0.9 0 0 0 0 Morocco 0 0 1 0.4 0 0 Sudan 1 0.9 1 0.4 1 0.7 Lebanon 2 1.8 4 1.7 1 0.7 Iraq 0 0 0 0 0 0 Palestine 0 2 1 Libya 1 0.9 0 0 0 0 Bahrain 1 0.9 2 0.9 4 2.6 Oman 1 0.9 0 0 3 2 Qatar 1 0.9 0 0 3 2 UAE 2 1.8 4 1.7 4 2.6 Jordan 0 0 2 0.9 2 1.3 Iran 0 0 1 0.4 0 0 Pakistan 1 0.9 1 0.4 0 0 Turkey 1 0.9 1 0.4 0 0 USA 9 8.2 14 6.1 18 11.9 Russia 0 0 2 0.9 0 0 UK 5 4.5 7 3 0 0 China 0 0 1 0.4 0 0

PAGE 109

109 Table 5 2 3 . Frequency of Religions Per Tone in Kuwaiti Social Media Religion Negative Percent Neutral Percent Positive Percent Islam (general) 5 4.5 16 7 10 6.6 Shiite 0 0 1 0.4 1 0.7 Muslim Brotherhood 2 1.8 2 0.9 2 1.3 Catholic 0 0 0 0 1 0.7

PAGE 110

110 Table 5 2 4 . Frequency of Unique Issues Per Tone in Kuwaiti Social Media Issue Negative Percent Neutral Percent Positive Percent Water issues 0 0 2 0.9 0 0 Electricity 0 0 1 0.4 0 0 Borsa 0 0 1 0.4 0 0 Employment 5 0.7 26 2.3 9 2.1 Cost of Living 3 2.7 1 0.4 1 0.7 Aviation 20 2.9 22 2 7 1.6 Housing 47 6.8 37 3.3 20 4.6 Energy 1 0.1 3 0.3 1 0.2 Promotion 0 0 2 0.2 0 0 Corruption 13 1.9 7 0.6 7 1.6 National Assembly 0 0 1 0.4 0 0 Ministers 3 2.7 4 1.7 1 0.7 Prime Minister 0 0 0 0 1 0.7 Elections 2 1.8 12 5.2 1 0.7 Stateless People 0 0 2 0.9 0 0 Specific Law 0 0 2 0.9 0 0 Raqaba 9 1.3 8 0.7 8 1.9 Government 21 19.1 12 5.2 0 0 Courts 2 1.8 1 0.4 0 0 Municipality 0 0 10 4.3 0 0 Tribalism 1 0.1 2 0.2 1 0.2

PAGE 111

111 Table 5 2 5 . Negative Tones of Issue s by Media in Kuwait Issue Newspapers Social Media Democracy 88 3 Hezbollah /Ha mas 4 0 West Bank/Gaza 0 0 Global warming 2 0 Education 67 10 Terrorism 19 0 Globalization 0 0 Children's issues 10 2 Women's issues 18 1 Historical 0 2 Sports 2 5 Islamic Issues 37 5 Arab League 3 0 Health 59 6 Media 73 50 GCC 23 0 Intelligence 8 0 UN 11 1 Traffic 44 9 Europe 3 0 Crime 61 2 Protests 21 Oil 4 0

PAGE 112

112 Table 5 2 6 . Negative Tones of Nations by Media in Kuwait Issue Newspapers Social Media Kuwait 419 53 Egypt 64 7 Syria 124 4 Saudi Arabia 8 7 Tunisia 13 1 Morocco 0 0 Sudan 5 1 Lebanon 8 2 Iraq 10 0 Palestine 4 0 Algeria 1 0 Djibouti 0 0 Somalia 1 0 Libya 1 1 Bahrain 5 1 Oman 1 1 Qatar 1 1 UAE 14 2 Yemen 1 0 Jordan 16 0 Comoros 0 0 Mauritania 1 0 Iran 20 0 Pakistan 1 1 Israel 7 0 Turkey 5 1 Afghanistan 1 0 USA 60 9 Russia 32 0 China 0 0 UK 7 5 France 10 0

PAGE 113

113 Table 5 2 7 . Negative Tones of Religions by Media in Kuwait Issue Newspapers Social Media Purity 0 0 Impurity 0 0 Islam (gene r a l) 21 5 Sunni 0 0 Shiite 1 0 Alawi 0 0 Wahabi 0 0 Salafi 5 0 Ismaeli 0 0 Zaidi 0 0 Muslim Brotherhood 18 2 Judaism 0 0 Zionism 0 0 Christianity 1 0 Catholic 0 0 Coptic 2 0 Orthodox 0 0

PAGE 114

114 Table 5 2 8 . Negative Tones of Unique Issues by Media in Kuwait Issue Newspapers Social Media Water issues 5 0 Electricity 9 0 Dow 0 0 Borsa 10 0 Employment 5 5 Cost of Living 4 3 Aviation 20 20 Housing 47 47 Vacation 0 0 Energy 1 1 Promotion 0 0 Corruption 13 13 National Assembly 92 0 Ministers 54 3 Prime Minister 23 0 Elections 15 2 Stateless People 1 0 Specific Law 18 0 Raqaba 9 9 Estejwab 9 0 Government 108 21 Courts 51 2 Municipality 25 0 Tribalism 1 1

PAGE 115

115 Tabl e 5 2 9 . Frequencies of the Objects per Tone in U.S. Social Media Issue Negative Percent Neutral Percent Positive Percent Democracy 1 1.5 0 0 0 0 Hezbollah /Hamas 1 1.5 0 0 0 0 Global warming 1 1.5 1 2.3 0 0 Education 6 9 4 9.1 1 4.2 Terrorism 3 4.5 0 0 0 0 Women's issues 0 0 1 2.3 0 0 Historical 0 0 0 0 1 4.2 Sports 0 0 1 2.3 0 0 Media 12 17.9 16 36.4 15 62.5 Intelligence 2 3 3 6.8 0 0 Crime 3 4.5 1 2.3 0 0 Syria 9 13.4 3 6.8 1 4.2 Iraq 1 1.5 0 0 0 0 USA 64 95.5 42 95.5 23 95.8 Russia 6 9 0 0 0 0 China 0 0 1 2.3 0 0 Judaism 0 0 1 2.3 0 0 Catholic 0 0 1 2.3 0 0

PAGE 116

116 Table 5 30 . Tone Per Object in U.S. Newspapers Object Negative Neutral Positive Education 0 0 4 Terrorism 8.75 1.25 0 Health 0 0 3 Media 11.6667 1.66667 0 GCC 0 0 2 Intelligence 17.5 2.5 0 UN 2.91667 0.41667 0 Traffic 0 0 1 Europe 35 5 0 Egypt 0 0 19 Syria 1.84211 0.26316 0 Iran 7 1 0 Israel 0 0 1 Turkey 35 5 2 Afghanistan 17.5 2.5 25 USA 1.4 0.2 48 Russia 0.72917 0.10417 0 China 0 0 1 UK 35 5 0 Christianity 0 0 1 Catholic 35 5 0 U.S. National Security 0 0 12

PAGE 117

117 Table 5 3 1 . Negative Tones of Objects by Media in the United States Issue Social Media Newspapers Democracy 1 0 Hezbollah /Ha mas 1 0 Global warming 1 0 Education 6 0 Terrorism 3 1.25 Media 12 1.66667 GCC 0 0 Intelligence 2 2.5 UN 0 0.41667 Europe 0 5 Crime 3 0 Syria 9 0.26316 Iraq 1 0 Iran 0 1 Turkey 0 5 Afghanistan 0 2.5 USA 64 0.2 Russia 6 0.10417 UK 0 5 Catholic 0 5 Water issues 0 N/A Electricity 0 N/A

PAGE 118

118 CHAPTER 6 DISCUSSION setting role of the mass media Similarly, Noelle ng the effect of media on the expression of unpopular opinions. This study merges both these mass media effects to explore the agenda silencing effect. Understanding agenda silencing particularly with regards to news involving the Middle East is particula rly important. Reporters Without Border (2014) shows that most journalists that were killed in 2014 were working in the Middle East, and Fahmy (2009) showed that U.S. reporters often lack cultural, religious and linguistic context when covering internatio nal news. Exposure to perspectives that do not appear in the U.S. media or in traditional media provide valuable insight to developing an understanding of the region. Just as U.S. media professionals and researchers can improve their understanding of th e Middle East by understanding the agenda silencing effect, the United States can be better understood by Middle Eastern nations and media professionals. For example, a basic understanding of a free and independent press may be difficult to understand by journalists and media researchers who have no experience working in such an environment. The agenda silencing effect can help Middle Easterners understand the United States more effectively by exploring less salient issues and negative attributes associat ed with objects in U.S. newspapers and messages posted by outspoken social media users. Measurements of issues and affective attributes are common in studying two traditional areas of mass communications research: agenda setting and the spiral of silence. By comparing the issues and affective attributes of content from two U.S. and two Kuwaiti newspapers, the

PAGE 119

119 current study explores the areas of relative silence. Furthermore, social media accounts of four politically outspoken individuals were included to explore the relationship between the agendas in traditional and social media. The findings of this resear ch show the limited scope of news coverage of the Middle East in American newspapers and among politically outspoken social media users. While the U.S. media may set the agenda for newspapers in a partially free media system such as that of Kuwait, this s tudy shows that there is a disconnect between politically outspoken social media users and newspapers in that environment. To summarize, this study demonstrates that any public opinion research with data collected only from traditional media may miss issu es that are important to average citizens of a country with a partially free media system. The inclusion of social media accounts also allows for an explication of agenda diversity. Agenda diversity has been utilized in the past to explore international n ews coverage in U.S. newspapers. In an analysis of opinion editorial content of The New York Times with regards to the Israeli Palestinian conflict, Golan and Wanta (2004) found that guest columnists added unique content that is different from the contribu array of perspective, online media, such as blogs, have been utilized by political campaigns to address a wide range of issues than newscasts and advertisements on television (Sweetser, Golan & Wanta, 2008). Since blogs, the rapidly changing field of mass communications provid es abundant possibilities for research of the effects of the expanding capabilities of online media. The current study explores the differences between issues covered in newspapers and social media in two countries: the United States and Kuwait.

PAGE 120

120 Table 6 .1 summarizes the findings of this study. Table 6 1. Results of Predicted Relationships Hypothesis Predicted Relationship Results 1 Issues will differ in traditional media from U.S. and Kuwait . Partially s upported 2 Issues will differ in social media from U.S. and Kuwait . Partially s upported 3 High salient issues in social media and traditional media in Kuwait will be congruent. Not s upported 4 High salient issues in social media and traditional media in the U.S. will be congruent . Partially s upported 5 Affective attributes of high salient traditional and social media will be congruent. Not s upported 6 Low salient of affective attributes from U.S. traditional and social media will be congruent. Not supported 7 Low salient issues from social media and traditional media in Kuwait will differ. Supported 8 Low salient issues from social and traditional media in the U.S. will differ. S upported 9 Low salient of affective social and traditional media will be congruent. Partially s upported 10 Low salient of affective attributes from U.S. social and traditional media will be congruent. Supported

PAGE 121

121 In general, the findings sho w that Kuwait and U.S. have different issues in their newspaper agendas. Evidence suggests that social media will discuss similar topics that are most salient in newspapers, which supports the agenda setting hypothesis. Support for agenda silencing appear s when issues are absent from newspapers. The evidence presented in this study suggests that the negative attributes of objects that appear in traditional and social media differ. This is true for both the traditional and social media in Kuwait and the U nited States. The first hypothesis stated that traditional media in Kuwait and the U.S. would address different issues. The results show a difference in the salience of issues in both places. For example, terrorism appeared in 411 graphs in U.S. newspape rs but in 48 graphs in Kuwaiti newspapers. Along the same lines, references to media appeared 884 times in the U.S. and 234 such as general references to Arabs and an emphasis of U.S. national security. Along the same lines, Kuwaiti newspapers appeared more concerned with municipal elections than U.S. newspaper coverage of the region. One explanation for such results is the concept of newsworthiness being a unive rsal concept, as described by Shoemaker and Cohen (2006). Proximity is one component of newsworthiness that becomes evident when comparing news content from two different locations. In this case, it was especially pronounced because the two locations are different countries with differing media systems, cultures and languages. Newsworthiness is determined, in part, by the proximity of the events to readers. Data analyzed in this study show that U.S. papers tried to show the significance of stories to th e U.S. Yet, it is also important to consider the evidence of the agenda setting effect. When the three agendas (issues, nations, religions) were compared in Kuwaiti and U.S. newspapers,

PAGE 122

122 statistically significant similarities were found in the priority of agendas. This finding is consistent with been explored closely in scholarship examining cultural imperialism in nations with a developmental model of the press. T he second hypothesis stated that the social media in the U.S. would address different issues than social media in Kuwait regarding news in the Middle East. Table 5 5 shows the issues found in U.S. and Kuwaiti social media. Table 5 in social media from both nations, and Table 5 7 does the same with religions. Then, the three agendas were compared for social media from the two nations. The priority of the issues were different (r=0.17, p=0.49). The priorities of the nations was not s imilar across U.S. and Kuwaiti social media (r=0.13, p=0.46). The priorities of religions was not similar either (r=0.15, p=0.56). Therefore there is no evidence of agenda setting here. Politically outspoken Kuwaiti social media users discussed domestic p olitical issues, shown in Table 5 9, such as corruption and municipal elections. As table 5 10 demonstrates, references to immigration, healthcare reform and abortion are found in the accounts of politically o utspoken U.S. social media users. One possi ble explanation for this is the concept of selective exposure. Himelboim (2014) conducted social network analysis on Twitter messages posted by talk show hosts in the United rticularly pronounced among conservatives and Republicans (p. 194). This concept was shown once again individuals gain access to international news from another nation is underutilized. Their findings

PAGE 123

123 reveal that online journalism has yet to connect news articles from around the world through hyperlinks. The third hypothesis states that high salient issues in social media will be congruent with issues in traditio nal media content in Kuwait. The results showed that the ranked order of the issues in Kuwaiti social and traditional media were not similar with regards to nations and religions, but different with regards to issues. These findings sugg est that there may be a difference between the priorities of issues in the two agendas. These findings are consistent with the work of Wanta and Foote (1994) which determined that there are some variables that influence the amount of coverage an issue receives which go beyo nd news coverage. They conducted a time series analysis which explained the concept of agenda surfing lity to influence the news media and therefore public agendas as a third party, a phenomenon referred to as agenda building. Wanta and Foote (1994) showed that some issues lend themselves to more agenda surfing than other issues. While the news media and president both react to some events, such as a sensational crime, the president can keep issues relating to international crises on the media agenda even as time goes on by continuing to emphasize it even after the issues may be considered less newswort hy due to the passage of time . The fourth hypothesis states that high salient issues in social media will be congruent with issues in traditional media content in Kuwait. The results show that the most frequently appearing issues and religions appear wi th different degrees of salience; however, the results also show that the nations seem to appear in a similar order. In other words, the nations discussed in

PAGE 124

124 Kuwaiti social and traditional media show evidence of agenda setting. However, this is not true when the topical area s involve religion and political issues. The fifth hypothesis states that high salient issues in social media will be similar with issues in the traditional media of the United States. When the five most frequent l y mentioned issues were computed , the ranking of the issues in the U.S. traditional media agenda was found to be different than those of the U.S. social media agenda. This could be the result of examining Twitter accounts of politically outspoken individ uals, who may view the traditional media as an adversary. The sixth hypothesis states that high salient affective attributes in social media content will be congruent with those of traditional media in the United States. The hypothesis was not supported, which shows that agenda silencing can occur at the second level when the issue frequently appears in U.S. newspapers. The seventh hypothesis states that low salient issues from social media content will be incongruent with the issues from traditional media co ntent in Kuwait. When tested, the agendas were not statistically significant. This means that they are different. This is interesting because it suggests that there are similar agendas in some topical area of the news, though not all. Wanta, King a nd McCombs (1995) found different sets of demographics and motivations for individuals interested in a wide array of issues in the United States and Taiwan. Here, the work of Wanta and Wu (1992) is important because it demonstrated that interpersonal comm unication could increase the priority of issues in the public agenda. One such explanation can be that the social media content producers have a personal agenda that may compete with the press when the issues are not major issues. This would be consisten t with

PAGE 125

125 step flow of information from television networks to the general public by a group of people they called early recognizers. Personality traits of early recognizers are important to consider. Applying this concept to social media is useful because news journalists on social media are expected to inject some opinion , a ccording to Lasorsa, Lewis and Holton (2012). In other words, newspapers may only set the agenda for social media in certain circumstance s. The eight h hypothesis states that low salient issues from social media content will be incongruent with the issues of traditional media content in the United States. Th is hypothesis was supported . In ot her words, the low salient issue s were differe nt between social media and traditional media in the United States . A study conducted by Wanta , Stephenson, Turk and McCombs (1989) found that that the president can be successful at setting the agenda in certain circumstances. One such circumstance was the personality of a president, as a president who sees the media as an adversary is unlikely t o be influenced by it. It is possible that politically outspoken individuals who may see traditional media as an adversary when an issue does is not very frequently appearing. The ninth hypothesis states that low salient affective attributes from social media content will be incongruent with those of traditional media in Kuwait. The results showed that there is a moderate and statistically significant relationship for religions (r=0.85, p=0.02) and issues (r= 0.88, p=0.007), but not nations (r= 0.18, p= 0.60) . Here, it is interesting that the correlation is negative with regards to issues but positive for religions . In other words, newspapers and social media messages discuss different topics negatively . While the religions show evidence of the

PAGE 126

126 agenda setting hypothesis, differences with issues show ev idence of attribute diversity and agenda silencing. The tenth hypothesis states that low salient affective attributes from social media will be incongruent with those of traditional media in the United Sta tes. The results show that there are differences . One explanation for this could deal with the fact that more media freedoms are afford in the United States , which c ould allow social media users to discuss similar issues differently without fear of retr ibution. Lasorsa and Wanta (1990) found that interpersonal and personal experiences take away from the effect the media have on us. In Noelle politically outspoken social media accounts consist of hard core pe rsonalities . Just as Brosius and Weimann (1996) found the personality traits of an early recognizer, it could be fascinating to explore the personalities of those to whom the spiral of silence does not apply. Therefore, the concept of agenda silencing re lies on differentiating between people who use the media and are affected and those who use the media and are not affected. In the case that one may read both newspapers and social media of politically outspoken personalities, agenda silencing is likely to occur when one realizes the differences between the tones attached to low salient issues in the media. Since low salient issues by definition are not the most frequently appearing issues, a consumer of news may experience cognitive dissonance because o f multiple, contradictory affective attributes. While previous studies have discussed evidence of agenda diversity, the current study reveals evidence of attribute diversity. Agenda diversity refers to the effect found in news consumers who access multiple sources of news. When asked about issues they perceive to be important, they list a wider range of issues with which they are concerned. Wanta, King and McCombs (1995) found that there

PAGE 127

127 may be differences in the demographic that exhibits this effect most pronouncedly across cultures and nations. From the idea of agenda diversity, the concept of attribute diversity is born. Attribute diversity refers to the effect when news consumers access information which associated different, and sometimes conflicti ng, attributes to an object or issue. For example, the current study found that a nation , such as India , may be described in positive or neutral terms in U.S. news but negatively in U.S. social media. The first research question asks whether issues covered by Kuwaiti traditional media are related to U.S. social media, where it disagrees with U.S. traditional media. Rojas (2010) suggested that some online users engage in correctional behavior online. The findings of this do not show a huge degree of incongruence between U.S. social media and U.S newspapers. One explanation for this may be that the social media sample selected f or this study may not include enough correctional behavior. It could be that the politically outspoken social media accounts chosen for this analysis do not specialize in the Middle East. Rather, they seem to act as a critique of the U.S. governmental ac tions due to the watchdog function of the press. However, it is important to note that the Kuwaiti social media did reference some U.S. broadcast entertainment media. One reason broadcast media in the U.S. may be more relevant to international social medi a is a lack of access linguistically. While broadcast media have been used by some individuals interested in learning the English language ( Nan & Mingfang, 2009 ), newspapers may require greater linguistic abilities. The second research question asks if affective attributes linked to issues covered by traditional media in Kuwait related to U.S. social media content that is incongruent with the U.S. traditional media. The third research question asks if issues covered by traditional media in the

PAGE 128

128 U.S. are related to Kuwaiti social media content that is incongruent with the Kuwaiti traditional media. Due to a lack of overlapping data, there was not enough information to address these questions. The fourth research question asks about how affective attributes linked to objects covered by U.S. newspapers are related to Kuwaiti social media, where it is incongruent with Kuwaiti traditional media. The results demonstrate that there are not enough similarities between Kuwaiti social media and U.S. news papers to come with any conclusions. In addition to linguistic issues which may make the content of U.S. newspapers less accessible to politically outspoken Kuwaiti social media users, self censorship may also play a role here. Alkazemi and Wanta (2014) found that Kuwaiti newspapers with both conservative and liberal editorial stances published negative messages in political cartoons. While conservative newspapers concentrated on topics relating to Kuwait, liberal newspapers focused on international issu es. The findings of Alkazemi and Wanta (2014) shows the importance of considering the willingness of social media users to voice disagreement with U.S. newspaper content online. Additionally, it is important to consider that the exposure of the Kuwaiti social media users to the U.S. is not limited to media. In some social media messages, one social media user at Millennia. Therefore, first hand exposure to the United States may decrease some of the disagreement with U.S. traditional media because there is a possibility of agreeing with the U.S. newspaper analyses. Implications These findings have implications for the analytical study of the Middle East. Edward Arab and Muslim world, but none in the Arab or Muslim world for studying the United States (p.

PAGE 129

129 324). While the possibilities for understanding the United States increases as the internet decreases barriers to access of the news and information regarding the United States, the study indicates that very few social media users follow the newspaper of the opposite nation. These findings are pertinent to journalists co vering international news about the Middle East because they are in a unique position to make insightful contributions that will increase mutual understanding of international issues to both U.S. audiences as well as audiences abroad. While journalists in the United States have a history of exposing corruption and are protected by shield laws that ensure the freedom of the press, such functions and protections are scarce in the Arab world (Duffy, 2013). The legal systems in many Arab countries discourage journalists be as beneficial as it is to focus on conflict in domestic affairs. While covering conflict can be a challenging task in the Middle East, it is n ot impossible. To approach such a task, the findings of this dissertation would suggest that journalists should follow outspoken social media accounts to understand the way in which conflict is discussed as well as the most important political conflicts i n the opinions of these social media users. Journalists may be surprised to find that there are some conflicts with which U.S. audiences can relate. Public diplomacy, which can be described as the mediation of messages from governments directly to public s of other nations, is another area in which the findings of this study may be applicable. In the social media accounts, citizens explicitly list political concerns they believe their nations are facing. In some instances, they provide links to internati onal organizations that address legal reform that may combat corruption. These concerns can be particularly useful for those specializing in the United States to introduce how such issues have

PAGE 130

130 been dealt with government transparency, public records and jo urnalist shield laws to expose such corruption. Further, public diplomacy experts can understand the priority of political concerns in the Arab world and understand where and how disagreement with the priority of U.S. foreign policy in the region occurs. In addition, policy practitioners may find these findings relevant as well. Experts in the Middle East have long known that there can be a discrepancy between the attitudes by public officials and other members of th e public in emerging democracies and/or developing countries ( Musa & Domatob , 2007). The field of Middle East studies has shown that ethnographers, political scientists and historians specializing in the region use the mass media to provide evidence of at titudes. The current study suggests that it may be useful to examine the attitudes of politically outspoken social media users in the region to understand how U.S. foreign policy can be interpreted in the region. In some cases, explanations of factually inaccurate statements can be made to correct misperceptions in the region. Finally, these findings have implications for mass communications research. It is not unusual for mass communications research to show the strength of the ties between two agenda s. The current study contributes by collecting primary data from another nation. In the age of big data, it is possible to collect a formidable archive of newspapers from various nations around the world. The possibilities for analysis of international media and its effects are endless. Limitations Although this study makes several contributions , the current study contains some limitations. For example, only four politically outspoken social media accounts were selected to be included in the sample. To detract from this limitation, several hundred of Twitter posts were analyzed. Furthermore, the traditional news media included in this study were limited to four

PAGE 131

131 newspapers from two countries. Therefore, a randomly constructed week was created to mitigat e some of the risks by showing the generalizability of the newspaper coverage. Due to the events in the time period from which the sample was selected, the replicability of this study may be limited. At the end of August in 2013, President Obama asked Con gress to vote on the U.S. involvement in Syria. Since the media provide information to citizens interested in participating in the democratic process, this was an extremely important time for journalists to cover the Middle East extensively. The fact tha t President Obama set a precedent by asking the American people to vote on a military matter would suggest that media coverage of this event may be quite different than coverage of Syria in 2014 when Obama used his authority as president to strike targets in Syria. Since the analyses included statistical tests to show the correlation between the rank of issues in the traditional and social media agendas, the results are simply correlational. In other words, none of the analyses of this study determine caus ation. Regardless, the analyses are consistent with a body of agenda setting studies that have utilized correlational tests (Kiousis, 2011; Sweetser, Golan and Wanta, 2008). Despite these limitations, this study contributes to the body of agenda setting research by exploring silence of the media about low salient issues. While most agenda setting research focuses on what is being communicated, the current study focuses on information that is not being communicated in one media system by comparing data gathered from two media systems. To do so, the concept of age nda silencing blends two traditional mass communications theoretical areas: agenda setting and the spiral of silence. By exploring agenda silencing, public opinion research can examine the two functions of the mass media that McCombs (2005)

PAGE 132

132 explained are integral to agenda Future R esearch In order to further explicate the concept of agenda silencing and agenda diversity, future research may explore the concept of censorship across differing cultures. Further, an analysis of the time it takes for the traditional media to influence the social media agenda could be conducted to understand the effect of one agenda on another. These analyses may include some elements of the spi ral of silence, such as measuring the amount of opinion included in the online messages in addition to tone. Finally, an analysis of the personalities of politically outspoken people by culture could prove to be interesting.

PAGE 133

133 APPENDIX A INTER CODER RELIABILITY Table A 1 . Inter C oder Reliability Results for Var iables in the U.S. Social Media Krippendorff's Alpha N Agreements N Disagreements N Cases Is this post 0.825397 16 1 17 Type of post 0.907042 16 1 17 Is it a retweet? 1 17 0 17 Link 0.76259 14 3 17 Tone 0.821138 15 2 17 Democracy N/A 17 0 17 globalization N/A 17 0 17 Lebanon/Syria N/A 17 0 17 Hezbollah/Hamas 1 17 0 17 West Bank/Gaza N/A 17 0 17 Global Warming N/A 17 0 17 Al Qaeda N/A 17 0 17 Education 1 17 0 17 Terrorism N/A 17 0 17 Darfur N/A 17 0 17 Children Issues N/A 17 0 17 Women Issues N/A 17 0 17 Historical figures N/A 17 0 17 Heroism/Sports N/A 17 0 17 US Foreign Policy 1 17 0 17 Islamic issues N/A 17 0 17 Iran's nuclear N/A 17 0 17 Iraq War 1 17 0 17 Arab league N/A 17 0 17 Health N/A 17 0 17 Media 0.879121 16 1 17 Politics 1 17 0 17 Economics 0.645161 16 1 17 Abortion N/A 17 0 17 Marijuana N/A 17 0 17 Obamacare N/A 17 0 17 Immigration N/A 17 0 17 Crime 1 17 0 17 GCC N/A 17 0 17 Intelligence 1 17 0 17 United Nations N/A 17 0 17 Traffic N/A 17 0 17 Europe/European Union N/A 17 0 17 Gas or oil N/A 17 0 17 Housing N/A 17 0 17

PAGE 134

134 Table A 1 . Continued Krippendorff's Alpha N Agreements N Disagreements N Cases Aviation N/A 17 0 17 Raqaba N/A 17 0 17 Vacation N/A 17 0 17 Energy N/A 17 0 17 Promotion N/A 17 0 17 Gold and Currency N/A 17 0 17 Corruption N/A 17 0 17 Merchants N/A 17 0 17 Socialism N/A 17 0 17 Capitalism N/A 17 0 17 P rivatization N/A 17 0 17 World Trade N/A 17 0 17 Retirement N/A 17 0 17 Loans N/A 17 0 17 Water issues N/A 17 0 17 Electricity N/A 17 0 17 Dow N/A 17 0 17 Money markets N/A 17 0 17 Borsa N/A 17 0 17 Employment N/A 17 0 17 Cost of Living N/A 17 0 17 Estejwab N/A 17 0 17 Government N/A 17 0 17 Court N/A 17 0 17 Municipality N/A 17 0 17 Specific Law N/A 17 0 17 National Assembly N/A 17 0 17 Ministers N/A 17 0 17 Prime Ministers N/A 17 0 17 Elections N/A 17 0 17 Stateless people N/A 17 0 17 Kuwait N/A 17 0 17 Egypt N/A 17 0 17 Syria 1 17 0 17 Saudi Arabia N/A 17 0 17 Tunisia N/A 17 0 17 Morocco N/A 17 0 17 Sudan N/A 17 0 17 Lebanon N/A 17 0 17 Iraq 1 17 0 17

PAGE 135

135 Table A 1 . Continued Krippendorff's Alpha N Agreements N Disagreements N Cases Palestine N/A 17 0 17 Algeria N/A 17 0 17 Djibouti N/A 17 0 17 Somalia N/A 17 0 17 Libya N/A 17 0 17 Bahrain N/A 17 0 17 Oman N/A 17 0 17 Qatar N/A 17 0 17 UAE N/A 17 0 17 Yemen N/A 17 0 17 Jordan N/A 17 0 17 Comoros N/A 17 0 17 Mauritania N/A 17 0 17 Iran N/A 17 0 17 Pakistan N/A 17 0 17 Israel N/A 17 0 17 Turkey N/A 17 0 17 Afghanistan N/A 17 0 17 USA 0 16 1 17 Russia N/A 17 0 17 China 0 16 1 17 United Kingdom N/A 17 0 17 France N/A 17 0 17 Purity N/A 17 0 17 Sunnis N/A 17 0 17 Shiite N/A 17 0 17 Alawaites N/A 17 0 17 Wahabis N/A 17 0 17 Salafis N/A 17 0 17 Ismaeli N/A 17 0 17 Zaidi N/A 17 0 17 M uslim B r ot h erhood N/A 17 0 17 Judaism (general) N/A 17 0 17 Zionism N/A 17 0 17 Christianity (general) N/A 17 0 17 Catholic N/A 17 0 17 Coptic N/A 17 0 17 Orthodox N/A 17 0 17 Impurity N/A 17 0 17 Islam (general) N/A 17 0 17

PAGE 136

136 Table A 2: Inter C oder Reliability Results for Variables in t he Kuwaiti Social Media Account Coded by Three People Krippendorff's Alpha Average Pairwise percent agreement Is this post 0.96 98.74 Type of post 0.915 94.96 Is it a retweet? 0.871 96.226 Link 0.981 98.74 Tone 1 100 Democracy N/A 100 Globalization N/A 100 Lebanon/Syria N/A 100 Hezbollah /Hamas N/A 100 West Bank/Gaza N/A 100 Global Warming N/A 100 Education 1 100 Terrorism N/A 100 Children Issues N/A 100 Women Issues 1 100 Historical figures 1 100 Heroism/Sports 1 100 Islamic issues 1 100 Arab league N/A 100 Health 1 100 Media 9.925 96.226 Politics N/A 100 Economics N/A 100 Abortion N/A 100 Marijuana N/A 100 Obamacare N/A 100 Immigration N/A 100 Crime 1 100 GCC 1 100 Intelligence N/A 100 United Nations N/A 100 Traffic 1 100 Europe/European Union N/A 100

PAGE 137

137 Table A 2: Continued Krippendorff's Alpha Average Pairwise percent agreement Gas or oil N/A 100 Raqaba N/A 17 Aviation N/A 17 Housing N/A 17 Vacation N/A 100 Energy N/A 100 Promotion N/A 100 Gold and currency N/A 100 Corruption N/A 100 Merchants N/A 100 Socialism N/A 100 Capitalism N/A 100 Privitization N/A 100 World Trade 1 100 Retirement N/A 100 Loans N/A 100 Water issues N/A 100 Electricity N/A 100 Dow N/A 100 Money markets N/A 100 Borsa N/A 100 Employment N/A 100 Cost of Living N/A 100 Estejwab N/A 100 Government N/A 100 Court N/A 100 Municipality 1 100 Specific Law N/A 100 National Assembly N/A 100 M inisters 1 100 Prime M inisters N/A 100 Elections 1 100 Stateless People N/A 100 Kuwait 0.415 93.71

PAGE 138

138 Table A 2: Continued Krippendorff's Alpha Average Pairwise percent agreement Egypt 1 100 Syria N/A 100 Saudi Arabia 1 100 Tunisia N/A 100 Morocco N/A 100 Sudan N/A 100 Lebanon N/A 100 Iraq 1 N/A Palestine N/A 100 Algeria N/A 100 Djibouti N/A 100 Somalia N/A 100 Libya 1 100 Bahrain N/A 100 Oman N/A 100 Qatar N/A 100 UAE 1 100 Yemen N/A 100 Jordan N/A 100 Comoros N/A 100 Mauritania N/A 100 Iran N/A 100 Pakistan 1 100 Israel N/A 100 Turkey 1 100 Afghanistan N/A 100 USA 0.881 97.48 Russia N/A 100 China 1 100 United Kingdom 1 100 France N/A 100 Purity N/A 100 Sunnis N/A 100 Shiite N/A 100

PAGE 139

139 Table A 2: Continued Krippendorff's Alpha Average Pairwise percent agreement Alawaites N/A 100 Wahabis N/A 100 Salafis N/A 100 Ismaeli N/A 100 Zaidi N/A 100 Muslim B r ot h erhood 0.795 98.74 Judaism (general) N/A 100 Zionism N/A 100 Christianity (general) N/A 100 Catholic N/A 100 Coptic N/A 100 Orthodox N/A 100 Impurity N/A 100 Islam (general) 0.795 98.74

PAGE 140

1 40 Table A 3. Inter coder Reliability Results for Variables in t he Kuwaiti Social Media Account Coded by Three People Krippendorff's Alpha Albania NA Angola 0 Argentina NA Arab General 1 Armenia 1 Australia 1 Austria NA Bangladesh NA Belgium NA Bosnia NA Brazil NA Burundi NA Canada 0.943 Canary Islands NA Columbia NA Congo NA CorrectionsCount 1 Correct Tone 1 Croatia NA Cuba NA Czech Republic NA Denmark NA Eritrea NA Ethiopia NA Finland NA Georgia NA Germany 1 Ghana NA Graph Count2 0.999 Graph Count 0.999 Greece NA Grenada NA Haiti NA Headline 0.517 HeadlineTone 0.517 Health 0.774

PAGE 141

141 Table A 3. Continued Hungary NA Iceland NA India NA Indonesia NA Ireland NA Italy 1 Japan NA Kazakhstan NA Kenya 0 Kosovo 1 Kuwait NA Kyrgzstan NA Laos NA Lithuania 1 Luxembourg NA Mali 0 Men's Names 0.941 Mexico NA Nation of Islam NA National Security 1 Negative Tone 0.735 Netherlands NA Neutral Tone 0.991 New Zealand NA Nicaragua NA Nigeria NA North Korea 1 Norway NA Oil Petroleum 1 Panama NA Phillipines NA Poland NA Portugal NA Positive Tone 0.98 Protests 1 Puerto Rico NA Religious Sectarian 0.85 Romania NA Rwanda NA

PAGE 142

142 Table A 3. Continued Senegal NA Serbia NA Singapore NA Slovakia NA South Africa NA South Korea NA Spain NA Sweden NA Switzerland NA Tajikistan NA Tanzania 0 Thailand NA Turkmenistan NA Uganda NA Ukraine NA Uzbekistan NA Vatican NA Venezuela NA Vietnam NA Women's Names per Article 0.876 Yugoslavia NA Egypt 1 Syria 0.999 Saudi Arabia 1 Tunisia NA Morocco NA Sudan 1 Lebanon 1 Iraq 0.986 Palestine 1 Algeria NA Djibouti NA Somalia NA Libya 1 Bahrain 1 Oman NA Qatar 1

PAGE 143

143 Table A 3. Continued UAE 1 Yemen NA Jordan 1 Comoros NA Mauritania NA Iran 1 Pakistan NA Israel 1 Turkey 1 Afghanistan 1 USA 1 Russia 1 China 1 United Kingdom 1 France 1 Purity NA Sunnis 1 Shiite 1 Alawaites 1 Wahabis NA Salafis NA Ismaeli NA Zaidi NA Muslim Brotherhood NA Judaism (general) NA Zionism NA Christianity (general) NA Catholic NA Coptic NA Orthodox NA Impurity NA Islam (general) 1 Democracy 0.84 Hezbollah/Hamas 1 West Bank/Gaza NA Global Warming NA Education 1 Terrorism 1 Children Issues 0.998

PAGE 144

144 Table A 3. Continued Women Issues 1 Historical figures 0.987 Heroism/Sports 1 Arab league 1 Media 0.881 Crime 1 GCC 1 Intelligence 1 United Nations 1 Traffic 0 Europe/European Union 0.997

PAGE 145

145 APPENDIX B CODE SHEET Publication Please pick the number associated with the publication you are coding: 1. Washington Post 2. New York Times 3. Al Qabas 4. Al Nahar 5. Twitter @Salma_AlEssa 6. Twitter @ Bourashed 7. Twitter Rachel Maddow 8. Twitter Ann Coulter Newspapers Date: DDMMYYYY of the day article/Twitter message was published. Tone: 9. N egative: if the message was written using words with negative connotations. 10. Positive: if the message was written using words with positive connotations. 11. Neutral: if the message was neither positive nor negative. Nations: 12. Kuwait a) Present when explicitly listed. b) Present when capital city is mentioned: Kuwait City. c) Other cities and downs should be examined with Google to determine national reference. d) Present when the writing style in Arabic utilizes Kuwaiti vernacular or phrases . e) Present when Kuwaiti emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. f) Otherwise, absent. 13. Egypt a) Present when explicitly listed. b) Present when capital city is mentioned: Cairo . c) Other cities and downs should be examined with Google to determine national reference. d) Present when the writing style in Arabic utilizes Egyptian vernacular or phrases. e) Present when Egyptians emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. f) Othe rwise, absent. 14. Syria a) Present when explicitly listed. b) Present when capital city is mentioned: Damascus . c) Other cities and downs should be examined with Google to determine national reference. d) Present when the writing style in Arabic utilizes Syrian vernacula r or phrases.

PAGE 146

146 e) Present when references to rebel or terrorist organizations operating in Syria may be present, such as the Free Syrian Army or the Islamic State of Iraq and Syria. f) Present when Syrian emissaries, companies, historical figures or artists are n amed. i) Google must be utilized to look up names. g) Otherwise, absent. 15. Saudi Arabia a) Present when explicitly listed. b) Present when capital city is men tioned: Riyadh . c) Other cities and downs should be examined with Google to determine national reference (i.e., Mecca, Jeddah, etc.) . d) Present when the writing style in Arabic utilizes Saudi Arabian vernacular or phrases . e) Present when Saudi emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. f) Otherwise, abs ent. 16. Tunisia a) Present when explicitly listed. b) Present when capital city is mentioned: Tunis . c) Present when the writing s tyle in Arabic utilizes Tunisian vernacular or phrases . d) Other cities and downs should be examined with Google to determine national refere nce. e) Present when Tunisian emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. f) Otherwise, absent. 17. Morocco a) Present when explicitly listed. b) Present when capital city is mentioned: Rabat . c) Present when the writing style in Arabic utilizes Moroccan vernacular or phrases. d) Other cities and downs should be examined with Google to determine national reference. e) Present when Moroccan emissaries, companies, historical figures or artists are named. i) Google must be uti lized to look up names. f) Otherwise, absent. 18. Sudan a) Present when explicitly listed , regardless of South Sudan or North Sudan . b) Present when capital cities of South Sudan (Juba) or North Sudan (Khartoum) are present . c) Other cities and downs should be examined with Google to determine national reference (i.e., Darfur) . d) Present when Sudanese emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. e) Otherwise, absent. 19. Lebanon a) Present when explicitly listed. b) Present w hen capital city is mentioned: Beirut . c) Present when the writing style in Arabic utilizes Lebanese vernacular or phrases. d) Other cities and downs should be examined with Google to determine national reference (i.e., be careful to discern between Tripoli in L ibya and Lebanon). e) Present when Lebanese emissaries, companies, historical figures or artists are named.

PAGE 147

147 i) Google must be utilized to look up names. f) Otherwise, absent. 20. Iraq a) Present when explicitly listed. b) Present when capital city is mentioned: Baghdad . c) Present when the writing style in Arabic utilizes Iraqi vernacular or phrases. d) Other cities and downs should be examined with Google to determine national reference (i.e., Najaf, Karbala, Basra, Fallujah) . e) Present when references to rebel or terrorist orga nizations operating in Iraq is present, such as the Islamic State of Iraq and Syria. f) Present when Iraqi emissaries, companies, historical figures or artists are named (i.e., . i) Google must be utilized to look up names. g) Otherwise, absent. 21. Palestine a) Present when explicitly listed as the Palestinian Territories . b) Present when the writing style in Arabic utilizes Palestinian vernacular or phrases. c) Other cities and downs should be examined with Google to determine national reference. d) Absent when references to Hamas were made, as there is a separate category for that . e) Absent when references to West Bank and Gaza were made, as there is a separate category for that. f) Present when Palestinians emissar ies, companies, histori cal figures or artists are named . i) Google must be utilized to look up names. 22. Algeria a) Present when explicitly listed. b) Present when capita l city is mentioned: Algiers . c) Other cities and downs should be examined with Google to determine n ational reference. d) Present when Algerian emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. e) Otherwise, absent. 23. Djibouti a) Present when explicitly listed. b) Present when capital city is mentioned: Djibouti. c) Other cities and downs should be examined with Google to determine n ational reference. d) emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. e) Otherwise, absent. 24. Somalia a) Present whe n explicitly listed, even in the context of (Somali American). b) Present when capita l city is mentioned: Mogadishu . c) Other cities and downs should be examined with Google to determine n ational reference. d) Present when references to rebel or terrorist organizations operating in Somalia, such as Al Shabab. e) Present when Somalian emissaries, companies, historic al figures or artists are named (police officers) i) Google must be utilized to look up names.

PAGE 148

148 f) Otherwise, absent. 25. Libya a) Present when explicitly listed. b) Present when capital city is mentioned: Tripoli. c) Present when the writing style in Arabic utilizes Libyan vernacular or phrases. d) Other cities and downs should be examined with Google to determine n ational reference (i.e., Ben Ghazi. e) emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. 26. Bahrain a) Present when explicitly listed. b) Present when capital city is mentioned: Manama. c) Present when the writing style in Arabic utilizes Bahraini ver nacular or phrases. d) Other cities and downs should be examined with Google to determine n ational reference. e) Present when Bahraini emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names, (i.e., royalty and di ssidents). f) Otherwise, absent. 27. Oman a) Present when explicitly listed. b) Present when capital city is mentioned: Muscat. c) Present when the writing style in Arabic utilizes Omani vernacular or phrases (i.e, the Sultanate). d) Other cities and downs should be examine d with Google to determine n ational reference (i.e., Salala h ). e) Present when Omani emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names, (i.e., Sultan Qaboos). f) Otherwise, absent. 28. Qatar a) Present when explicitly listed. b) Present when capital city is mentioned: Doha . c) Present when the writing style in Arabic utilizes Qatari vernacular or phrases. d) Other cities and downs should be examined with Google to determine n ational reference. e) Present when Qatari emis saries, companies, historical figures or artists are named. i) Google must be utilized to look up names. f) Otherwise, absent. 29. United Arab Emirates a) Present when explicitly listed. b) Present when capital city is mentioned: Abu Dhabi. c) Present when the writing style in Arabic utilizes U.A.E vernacular or phrases. d) Other cities and downs should be examined with Google to determine n ational reference (i.e., Dubai). e) Present when U.A.E emissaries, companies, historical figures or artists are named, including education (i.e ., branches of international universities in the U.A.E). i) Google must be utilized to look up names. f) Otherwise, absent. 30. Yemen

PAGE 149

149 a) Present when explicitly listed. b) c) Present when the writing style in Arabic utilizes Ye meni vernacular or phrases. d) Other cities and downs should be examined with Google to determine n ational reference. e) Present when U.A.E emissaries, companies, historical figures or artists are named, (i.e., civil war, use of drones). i) Google must be utilized to look up names. f) Otherwise, absent. 31. Jordan a) Present when explicitly listed. b) Present when capital city is mentioned: Amman. c) Present when the writing style in Arabic utilizes Jordanian vernacular or phrases. d) Other cities and downs should be examined with Google to determine n ational reference. e) Present when Jordanian emissaries, companies, historical figures or artists are named, including education (i.e., royalty). i) Google must be utilized to look up names. f) Otherwise, absent. 32. Comoros a) Present when explicitly listed. b) Present when capital city is mentioned: Moroni. c) Other cities and downs should be examined with Google to determine n ational reference. d) i) Google must be utilized to look up names. e) Otherwise, absent. 33. Mauritania a) Present when explicitly listed. b) Present when capital city is mentioned: Nouakchott. c) Other cities and downs should be examined with Google to determine n ational reference. d) ompanies, historical figures or artists are named. i) Google must be utilized to look up names. e) Otherwise, absent. 34. Iran a) Present when explicitly listed. b) Present when capital city is mentioned: Tehran. c) Other cities and downs should be examined with Google to de termine n ational reference (i.e., Mashhad, Shiraz, Isfahan). d) Present when Iranian emissaries, companies, historical figures or artists are named, (i.e., Rouhani, Zarif, Khamenei, Supreme Leader, Shah). i) Google must be utilized to look up names. e) Otherwise, absent. 35. Pakistan a) Present when explicitly listed. b) Present when capital city is mentioned: Islamabad . c) Other cities and downs should be examined with Google to determine n ational reference (i.e., Lahore).

PAGE 150

150 d) Present when Pakistani emissaries, companies, historic al figures or artists are named (i.e., the 1970s war and references to East and West Pakistan) i) Google must be utilized to look up names. e) Otherwise, absent. 36. Israel a) Present when explicitly listed. b) Present when capita l city is mentioned: Jerusalem . c) Other citi es and downs should be examined with Google to determine n ational reference (i.e., Tel Aviv, Golan Heights , kibbutz ). d) Present when Israeli emissaries, companies, historical figures or artists are named. i) Google must be utilized to look up names. e) Otherwise, absent. 37. Turkey a) Present when explicitly listed. b) Present when capital city is mentioned: Ankara. c) Other cities and downs should be examined with Google to determine n ational reference (i.e., Istanbul). d) Present when Israeli emissaries, companies, historical fi gures or artists are named. i) Google must be utilized to look up names. e) Otherwise, absent. 38. Afghanistan a) Present when explicitly listed. b) Present when capita l city is mentioned: Kabul . c) Other cities and downs should be examined with Google to determine n ational reference. d) Present whe n Afghan i emissaries, companies, historical figures or artists are named (i.e., Karzai) . i) Google must be utilized to look up names. e) Otherwise, absent. Religions: 39. Islam (general) a) Present if there are references to Islam/Muslim, the Koran, mosques, fasting, or conversion. b) Present if there are references to Arabic language words that may have an Islamic connotation, such as describing someone as having died in religious ways (i.e., Islamist, Islamophobia) . c) Otherwise, absent. d) Names should be inserted in Google to see a possible connection to this category. 40. Islam Sunnis a) Present if explicitly stated. b) Otherwise, absent. c) Names should be inserted in Google to see a possible connection to this category. 41. Islam Shiites a) Present if explicitly stated.

PAGE 151

151 b) Present if Shiite symbols are used, such as Shiite saints (Hassan/Hussain), or the term c) Otherwise, absent. d) Names should be inserted in Google to see a possible connection to this category. 42. Islam Alawites a) Present if explicitly stated. b) Otherwise, absent. c) Names should be inserted in Google to see a possible connection to this category. 43. Islam Wahabis a) Present if explicitly stated. b) Otherwise absent. c) Names should be inserted in Google to see a possible connection to this category. 44. Islam Salaf is a) Present if explicitly stated. b) Otherwise absent. c) Names should be inserted in Google to see a possible connection to this category. 45. Islam Ismaeli a) Present if explicitly stated. b) Otherwise absent. c) Names should be inserted in Google to see a possible connecti on to this category. 46. Islam Zaidi a) Present if explicitly stated. b) Otherwise absent. c) Names should be inserted in Google to see a possible connection to this category. 47. Islam Muslim Brotherhood a) Present if explicitly stated. b) Present where references to former Eg yptian president, Mohammad Morsi, exist. c) Otherwise absent. d) Names should be inserted in Google to see a possible connection to this category. 48. Judaism (general) a) Present if explicitly stated. b) Present where references to Jewish New Year , Rosh Hashanah, rabbinical scholars, rabbi, or other religious references exist. c) Otherwise absent. d) Names should be inserted in Google to see a possible connection to this category. 49. Zionism a) Present if explicitly stated. b) Otherwise absent. c) Names should be inserted in Google to see a possible connection to this category. 50. Christianity (general) a) Present if explicitly stated. b) c) Otherwise absent. d) Names should be inserted in Google to see a possible connect ion to this category. 51. Christianity Catholic a) Present if explicitly stated.

PAGE 152

152 b) Present if references to the Pope exist. c) Present when references to the Vatican do not merely describe the geographic location. d) Otherwise absent. e) Names should be inserted in Google to see a possible connection to this category. 52. Christianity Coptic a) Present if explicitly stated. b) Otherwise absent. c) Names should be inserted in Google to see a possible connection to this category. 53. Christianity Orthodox a) Present if explicitly stated. b) Present when describing various Orthodox followers, such as Syrian nuns. c) Otherwise absent. d) Names should be inserted in Google to see a possible connection to this category. 54. Purity a) Present if there are references to Islamic purity rituals, including ablutio n. b) Present if there are references to variations of the Arabic word for spiritual purity: tahara. c) Otherwise, absent. Issues: 55. Media a) If a medium is specified (i.e., video, newspaper, movie, book, etc.), then present. b) If the name of newspapers, social media platforms, movies, books, songs, or other media formats are explicitly stated, then present. c) If journalists or editors are mentioned, then present (i.e., letters to the editor). d) If sources are mentioned, then present. This would include statements such a The e) f) Unfamiliar names should be entered into Google to determine if they are newspapers frm other countries. g) If statements h) Otherwise, absent. 56. Democracy a) b) If references to legislative bodies with powers to vote or veto were used, present. c) Otherwise, absent. 57. Historical figures a) If references to former presidents, political parties or world leaders were used, present. (i.e., Nazis, the Iranian Shah or the Soviet Union ). b) If references to artists from previous time periods were used, present. c) If references to religious figures from other time periods were used, present. d) If a date prior to 2011 appears, then present. e) If re ferences to previous world events appear, then present (i.e, the Korean War, Vietnam). f) Otherwise, absent. 58. Education

PAGE 153

153 a) If names of universities appear, present. b) If professors are referenced because their work is acknowledged, present. c) If students are mentione d, present. d) If references to bodies governing education, present. e) Otherwise, absent. 59. Terrorism a) b) c) their own category. However, present when these organizations are described as terrorist organizations, which they often are. d) Otherw ise, absent. 60. Crime a) Present if references to robberies, shootings, gang related activity occurs. b) category present. c) Otherwise, absent. 61. Health a) Present if any references to hospi tals appear. b) Present if any references to diseases appear. c) Otherwise, absent. 62. Intelligence a) b) Present when references to the Central Intelligence Agency is references (e.g., CIA). c) Otherwise, absent. 63. National Security a) Present if references to U.S. national security is made only. b) c) Otherwise, absent. 64. Children Issues a) b) Present if a n article deals with children (i.e., how to talk to your kids about war). c) Otherwise, absent. 65. Sports a) Present if references to sporting teams, events, or paraphernalia are mentioned. b) Otherwise, absent. 66. Gulf Countries Council a) b) c) d) Present when references to the Gulf are made in Arabic. e) Otherwise, absent. 67. Islamic issues a) Present when spec ific activities pertaining to Islam are mentioned.

PAGE 154

154 b) c) Present when references to any of the five pillars of Islam are made, (i.e., prayer, zakat (charity), religious pilgrimage, faith in a monotheistic deity). d) Otherwise, absent. 68. Women issues a) b) Present in articles dealing with the number of women in leadership positions. c) Present when women are writing articles or being interviewed about their role as a woman. d) Otherwise, absent. 69. Arab League a) Present when references to Lakhder Brahimi were made. b) Present when references to the Arab League organization were made. c) Oth erwise, absent. 70. Global warming a) b) Present when issues pertaining to this topic are referenced. c) Otherwise, absent. 71. West Bank/Gaza a) ely. b) Not present when references to Palestinians are made, as another category describes this issue. c) Otherwise, absent. 72. Hezbollah/Hamas a) b) Present when individuals names are listed after an association with th e organization is made. c) Present when d) Otherwise, absent. 73. Oil/petroleum a) b) Otherwise, absent. 74. Merchants a) If explicitly stated, present. b) Otherwise, absent. 75. Socialism a) If explicitly stated, present. b) Otherwise, absent. 76. Capitalism a) If explicitly stated, present. b) Otherwise, absent. 77. Privatization a) If explicitly stated, present. b) If references to selling government owned business appear, present c) Otherwise, absent.

PAGE 155

155 78. World trade a) If explicitly stated, present. b) If references to international businesses appear, then present. c) If advertorial references are made in social media to international businesses, then present. d) Otherwise, absent. 79. National assembly a) resent. b) If references are made to Kuwaiti parliamentarians, then present. c) If references are made to the Speaker of the House, then present. d) Search names on Google to ensure that references are clear. e) Otherwise, absent. 80. Ministers a) If references to a specific minister are made, then present. b) If ministers are referred to by name, then present. c) Absent when a ministry is mentioned without reference to the minster. d) Otherwise, absent 81. Prime Minister a) If the Kuwaiti Prime Minister is mentioned by name or position, the n present. b) Otherwise, absent. 82. Elections a) If references to elections or voting are made, then present. b) Present if there is a reference to parliamentary elections in Britain, U.S. Congress, c) Advertisements for voting for a specific issue or candidate qualify as present. d) Otherwise, absent. 83. Bedoun (stateless people) a) If stateless people are explicitly mentioned, then present. b) If citizenship application processes for stateless people are mentioned, then present. c) Otherwise , absent. Twitter 84. Retweeted? 1= no, 2=yes 85. Number of hashtags? Count 86. Link 1= no, 2=yes 87. Tone: a) 1=negative b) 2=neutral c) 3=positive

PAGE 156

156 LIST OF REFERENCES Alfraih, S. A. (1999). Hona AlKuwait . Kuwait: Dar Alqurtas Lelnashr. Ali, M . I. T. ( 2006 ). news in Jordan? In News a round the World: Content Practitioners and the Public , 239 258. Alkazemi, M. F., & Wanta, W. (2014). Kuwaiti political cartoons during the Arab Spring: Agenda setting and self censorship. Journalism , 1464884914533072. Al Nashmi , E., Cleary, J., Molleda, J. C., & McAdams, M. (2010). Internet political discussions in the Arab world: A look at online forums from Kuwait, Saudi Arabia, Egypt and Jordan. International Communication Gazette , 72 (8), 719 738. Al Obaidi, J. A. (2011). Bro Based on Historical Contexts: Instructional Media Lessons. International Journal Of Instructional Media , 38 (2), 159 168. Amin, H. (2002). Freedom as a value in Arab media: Perceptions and att itudes among journalists. Political Communication , 19 (2), 125 135. Bowe, B. J., Fahmy, S., & Wanta, W. (2013). Missing religion: Second level agenda setting and Islam in American newspapers. International Communication Gazette . Bourdieu, P. (1999). Rethink ing the state: Genesis and structure of the bureaucratic field. State/culture: state formation after the cultural turn , 53 75. Brosius, H. B., & Weimann, G. (1996). Who Sets the Agenda Agenda Setting as a Two Step Flow. Communication Research , 23 (5), 561 580. Bryant, J., & Miron, D. (2004). Theory and research in mass communication. Journal of communication , 54 (4), 662 704. Brazeal, L. M., & Benoit, W. L. (2008). Issue Ownership in Congressional Campaign Television Spots. Communication Quarterly , 56 (1) , 17 28. Campos, M. (2010). Ottoman Brothers: Muslims, Christians, and Jews in Early Twentieth Century Palestine . Stanford University Press. Cassidy, W. P. (2006). Gatekeeping similar for online, print journalists. Newspaper Research Journal , 27 (2), 6. Cha ng, T. K., Himelboim, I., & Dong, D. (2009). Open Global Networks, Closed International Flows World System and Political Economy of Hyperlinks in Cyberspace. International Communication Gazette , 71 (3), 137 159. Cherribi, S. (2006). From Baghdad to Paris Al Jazeera and the Veil. The Harvard International Journal of Press/Politics , 11 (2), 121 138.

PAGE 157

157 Chaffee, S. H., & Berger, C. R. (1987). What communication scientists do. Handbook of communication science , 99 122. Colistra, R. (2012). Shaping and Cutting the Me Agenda and Frame Building and Agenda Cutting Influences. Journalism & Communication Monographs , 14 (2), 85 146. Crandall, H. M., & Ayres, J. (2002). Communication apprehension and the spiral of silence. Journa l of the North west Communication Association , 31 , 27 39. Culbertson, H. (2007). Agenda diversity indicates newspaper quality. Newspaper Research Journal , 28 (1), 40. Dashti, A. A. (2009). The role of online journalism in political disputes in Kuwait. Journal of Arab & Muslim Media Research , 2 (1 2), 91 112. Donsbach, W. & Stevenson, R. L. (1984). Challenges, problems and empirical evidence of the theory of spiral of silence. Paper presented at the Conference of the International Communication Associati on, San Francisco, CA. Duffy, M. (2013, August). Arab media regulations: Identifying restraints on freedom of the press in laws of six Arabian Peninsula countries . Communication Law and Policy Division Association for educators of journalism and mass communications, Washington, DC. Durkheim, E. (1933). The division of labor. Trans. G. Simpson. New York: Macmillan . Fahmy, S. (2009). How Could So Much Produce So Lit tle?. International media communication in a global age , 147. empirical study. In 11th international symposium on online journalism (pp. 23 24). Fahmy, S., Wanta, W., & Nisbet, E. C. (2012). Mediated public diplomacy: Satellite TV news in the Arab world and perception effects. International Communication Gazette , 74 (8), 728 749. Franklin, S. (2011). Sunrise on the nile. Columbia Journalism Review , 49 (6), 17. F reedom Hou se. (201 3 ). Kuwait . Retrieved from http://www.freedomhouse.org/report/freedom press/2013/kuwait Freedom House. (2013). United States. Retreived from http://www.freedomhouse.org/report/freedom press/2013/united states Gerbner, G. (1998). Cultivation analys is: An overview. Mass Communication & Society, 1 (3), 175 194.

PAGE 158

158 Ghanem, S. (1997). Filling in the tapestry: The second level of agenda setting. Communication and democracy: Exploring the intellectual frontiers in agenda setting theory , 3 14. Glynn, C. J., & McLeod, J. M. (1984). Public opinion du jour: An examination of the spiral of silence. Public Opinion Quarterly , 48 (4), 731 740. Golan, G., & Wanta, W. (2001). Second Level Agenda Setting in the New Hampshire Primary: A Comparison of Coverage in Three New spapers and Public Perceptions of candidates. Journalism and Mass Communication Quarterly, 78 (2), 247 259. Golan, G., & Wanta, W. (2004). Guest Columns Add Diversity To NY Times' Op Ed Pages. Newspaper Research Journal , 25 (2) , 70 82 . Gonzenbach, W. J., Kin g, C., & Jablonski, P. (1999). Homosexuals and the military: An analysis of the spiral of silence. Howard Journal of Communication , 10 (4), 281 296. Graham , M., & Stephens, M. (Producer). (2012, June 19). A Geography of Twitter [Print Photo]. Retrieved fro m http://www.oii.ox.ac.uk/vis/?id=4fe09570 Hafez, K. (2002). Journalism ethics revisited: A comparison of ethics codes in Europe, North Africa, the Middle East, and Muslim Asia. Political Communication , 19 (2), 225 250. Hatchen, W. A., & Scotton, J. F. (20 07). The world news prism: Global information in a Satellite Age. Hayes, A. F. (2007). Exploring the forms of self censorship: On the spiral of silence and the use of opinion expression avoidance strategies. Journal of Communication, 57 (4), 785 802. Hime lboim, I. (2014). Political Television Hosts on Twitter: Examining Patterns of Interconnectivity and Self Exposure in Twitter Political Talk Networks. Journal of Broadcasting & Electronic Media , 58 (1), 76 96. Holz Ivory, A., Gibson, R., & Ivory, J. D. (2009). Gendered relationships on television: Portrayals of same sex and heterosexual couples. Mass Communication and Society , 12 (2), 170 192. Jakobsen, P. V. (2000). Focus on the CNN effect misses the point: the real media impact on conflict management is invisible and indirect. Journal of Peace Research , 37 (2), 131 143. Jeffres, L.W., Neuendorf, K.A. & Atkin, D. (1999). Spiral of silence: Expression opinions when the climate of opinion is unambiguous. Political Communication, 16( 2), 115 13 1. Johnson, T. J., Wanta, W., Byrd, J. T., & Lee, C. (1995). Exploring FDR'S relationship with the press: A historical agenda setting study. Political Communication , 12 (2), 157 172. Khalidi, R. (1991). Ottomanism and Arabism in Syria before 1914: a reasses sment. The origins of Arab nationalism , 51 72 .

PAGE 159

159 Kiousis, S. (2004). Explicating media salience: A factor analysis of New York Times issue coverage during the 2000 US presidential election. Journal of Communication , 54 (1), 71 87. Kiousis, S. (2011). Agenda s etting and A ttitudes : 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. Kiousis, S., & Shields, A. (2008). Intercandidate agenda setting in presidential elections: Issue and attribute agendas in the 2004 campaign. Public Relations Review , 34 (4), 325 330. Kuttner, R. (2007). The Race Newspapers have a bright future as print digital hybrids after all -But they'd better hurry. Columbia Journalism Review , (R), 24. Lang, G. E., & Lang, K. (1983). The battle for public opinion: The president, the press, and the polls during Watergate (p. 305). New York: Columbia University Press. Lasorsa, D. L. (1991). Political outspokenness: Fact ors working against the spiral of silence. Journalism & Mass Communication Quarterly , 68 (1 2), 131 140. Lasorsa, D. L., Lewis, S. C., & Holton, A. E. (2012). Normalizing Twitter: Journalism practice in an emerging communication space. Journalism Studies , 1 3 (1), 19 36. Lasorsa, D. L., & Wanta, W. (1990). Effects of personal, interpersonal and media experiences on issue saliences. Journalism & Mass Communication Quarterly , 67 (4), 804 813. Liu, X., & Fahmy, S. (2011). Exploring the spiral of silence in the vir willingness to express personal opinions in online versus offline settings. Journal of Media and Communication Studies , 3 (2), 45 57. Livingston, S., & Eachus, T. (1995). Humanitarian crises and US foreign policy: Somalia and the CN N effect reconsidered. Political Communication , 12 (4), 413 429. LeVine, M. (2005). Overthrowing Geography: Jaffa, Tel Aviv, and the Struggle for Palestine, 1880 1948 . University of California Pr. LeVine, M. (2008). Heavy metal Islam: rock, resistance, and the struggle for the soul of Islam . Random House Digital, Inc. Matthes, J., Morrison, K. R., & Schemer, C. (2010). A spiral of silence for some: Attitude certainty and the expression of political minority opinions. Communication Research , 37 (6), 774 800. M artin Kratzer, R., & Thorson, E. (2009). Patterns of News Quality. International Media Communication in a Global Age , 160. McCombs, M. (1999). Personal involvement with issues on the public agenda. International Journal of Public Opinion Research , 11 (2), 1 52 168.

PAGE 160

160 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. McDevitt, M., Kiou sis, S., & Wahl Jorgensen, K. (2003). Spiral of moderation: Opinion expression in computer mediated discussion. International Journal of Public Opinion Research , 15 (4), 454 470. McPhail, T. L. (2010). Global communication: Theories, stakeholders, and trend s . Wiley. com. Meraz, S. (2011). Using time series analysis to measure intermedia agenda setting influence in traditional media and political blog networks. Journalism & Mass Communication Quarterly , 88 (1), 176 194. Nan, Y., & Mingfang , Z. (2009). Using VOA Special English to Improve Advanced English Learners' Productive Use of High Frequency Words. In English Teaching Forum , 47( 3 ) , 26. Neuendorf, K. A. (2002). The content analysis guidebook . Sage. Nisbet, M. C., & Aufderheide, P. (20 09). Documentary film: towards a research agenda on forms, functions and impacts. Mass Communication and Society. 12 (4), 450 456. Noelle Neumann, E. (1974). The spiral of silence: A theory of public opinion. Journal of Communication, 24 , 43 51. Noelle N eumann, E. (1977). Turbulences in the climate of opinion: Methodological applications of the spiral of silence theory. Public Opinion Quarterly, 77 (4), 143 158. Perry, S. D., & Gonzenbach, W. J. (2000). Inhibiting speech through exemplar distribution: Ca n we predict a spiral of silence?. Journal of Broadcasting & Electronic Media , 44 (2), 268 281. Pintak, L., & Ginges, J. (2008). The mission of Arab journalism: Creating change in a time of turmoil. The International Journal of Press/Politics , 13 (3), 193 22 7. Press reference. (n.d). http://www.pressreference.com/Gu Ku/Kuwait.html http://en.rsf.org/press freedom barometer journalists killed.html?annee=2014 Riffe, D., Lacy, S., & Fico, F. G. (2005). Analyzing media messages. Using quantitative content analysis in research. Taylor & Francis Group, New York. Roberts, M., Wanta, W., & Dzwo, T. H. D. (2002). Agenda setting and issue salience online. Communication Research , 29 (4), 452 465.

PAGE 161

161 How perceptions of media and media effects shape political behaviors. International Journal of Public Opinion Research , 22 (3 ), 343 363 . Said, E. W. (1978). Orientalism . New York: Vintage Books. Selvik, K. (2011). Elite Rivalry in a Semi Democracy: The Kuw aiti Press Scene. Middle Eastern Studies , 47 (3), 477 496. Scheufle, D. A., & Moy, P. (2000). Twenty five years of the spiral of silence: A conceptual review and empirical outlook. International Journal of Public Opinion Research , 12 (1), 3 28. Shoemaker, P. J., & Cohen, A. A. (20 06 ). News a round the World: Content Practitioners and the Public . Routledge. Stephenson, L. (2011, Apr. 11). Ahistorical Kuwaiti Sectarianism. The Middle East Channel . Received http://mideast.foreignpolicy.com/posts/2011/04/29/ahi storical_kuwaiti_sectarianism Strömbäck, J., & Kiousis, S. (2010). A new look at agenda setting effects comparing the predictive power of overall political news consumption and specific news media consumption across different media channels and media types . Journal of Communication , 60 (2), 271 292. Sweetser, K. D., Golan, G. J., & Wanta, W. (2008). Intermedia agenda setting in television, advertising, and blogs during the 2004 election. Mass Communication & Society , 11 (2), 197 216. Takeshita, T. (2006). Cur rent critical problems in agenda setting research. International Journal of Public Opinion Research , 18 (3), 275 296. Tan, Y., & Weaver, D. H. (2012). A genda diversity and agenda setting from 1 956 to 2004: What are the trends over time?. Journalism Studies , (ahead of print), 1 17. Wanta, W., & Foote, J. (1994). The president news media relationship: A time series analysis of agenda setting. Journal of Broadcasting & Electronic Media , 38 (4), 437 448. Wanta, W., Golan, G., & Lee, C. (2004). Agenda setting and international news: Media influence on public perceptions of foreign nations. Journalism & Mass Communication Quarterly , 81 (2), 364 377. Wanta, W., & Golan, G. J. (2009). Coverage of Foreign Elections in the United States. International media communication in a global age , 109. Wanta, W., King, P. T., & McCombs, M. E. (1995). A comparison of factors influencing issue diversity in the US and Taiwan. International Journal of Public Opinion Research , 7 (4), 353 365.

PAGE 162

162 Wanta, W. & Wu, Y. (1992). Interpersonal Com munication and the Agenda Setting Process. Journalism & Mass Communication Quarterly, 69 (4), 847 855. Wasserman, S., & Faust, K. (1994). Social Network Analysis: Methods And Applications (Structural Analysis In The Social Sciences) Author: Stanley Wasserm an. Watson, J. (1998). Media communication . Palgrave Macmillan. Wedeen, L. (1999). Ambiguities of domination: Politics, rhetoric, and symbols in contemporary Syria . University of Chicago Press. W orld Association of Newspapers (2013). Retrieved http://www. warc.com/Pages/Store/PDF/Newspaper_trends_sample_Kuwait.pdf World Factbook. (2013). https://www.cia.gov/library/publications/the world factbook/geos/ku.html Vultee , F. (2012). A PALEONTOLOGY OF STYLE: The evolution of the Middle East in the AP Stylebook 1977 2010. Journalism Practice , 6 (4), 450 464. the Middle East. Journal of Co mmunication Inquiry , 30 (4), 319 336. Yang, J., & Stone, G. (2003). The powerful role of interpersonal communication in agenda setting. Mass Communication and Society , 6 (1), 57 74.

PAGE 163

163 BIOGRAPHICAL SKETCH Mariam F. Alkazemi has attended George Washington University (2007, B.A., Journalism), Michigan State University (2009, M.A., Advertising and Public Relations) and University of Florida (2014, Ph.D., Mass Communications). Her publications appear in Journal of Relig ion, Media & Digital Culture and Journalism: Theory, Practice and Criticism . At the University of Florida, she has been the instructor of record for four courses: Applied Fact Finding, Mass Media & You, Journalism Studies and World Communication Systems. She was the recipient of the following awards: 2013 College of Journalism and Communications Graduate Student Research Award, 2014 College of Journalism and Communications Graduate Student Teacher Award, 2014 Graduate School Graduate Teaching Assistant A ward . None of these achievements would have been possible were it not for her outstanding mentor, Dr. Wayne Wanta.