Exploring Linkages among Public Relations, Attribute Agenda-Building, Trust, and Corporate Reputation Mediated by Emotion

MISSING IMAGE

Material Information

Title:
Exploring Linkages among Public Relations, Attribute Agenda-Building, Trust, and Corporate Reputation Mediated by Emotion
Physical Description:
1 online resource (164 p.)
Language:
english
Creator:
Kim, Ji Young
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Mass Communication, Journalism and Communications
Committee Chair:
Kiousis, Spiro K
Committee Members:
Molleda, Juan Carlos
Ferguson, Mary Ann
Kraft, John

Subjects

Subjects / Keywords:
agenda -- attributes -- communication -- emotion -- issues -- reputation
Journalism and Communications -- Dissertations, Academic -- UF
Genre:
Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Grounded in the agenda-building perspective, the current study examined the role of issues in public relations communication strategies. The purpose of the study was 1) to demonstrate a causal relationship in agenda building, 2) to explore the role of affect in the relationship, and 3) to expand the scope of application of the concept into corporate communication. Particularly, a 2 x 2 between-subject factorial experiment was conducted with 225 participants. These participants were recruited from an online panel service with monetary compensation. Analysis of variance, regression, and a path analysis model were utilized to test the relationships. Some direct and indirect effects were found in the model supporting the mediating role of emotions in the relationship between the two independent variables (issue ownership and issue tone) and dependent variables (issue salience, trust, and corporate reputation). One of the main contributions of this study was to explore the predictors of corporate reputation in a corporate communication context. Theoretical and practical implications are presented in Chapter 5.
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.
Statement of Responsibility:
by Ji Young Kim.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Kiousis, Spiro K.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044463:00001


This item is only available as the following downloads:


Full Text



PAGE 1

EXPLORING LINKAGES AMONG PUBLIC RELATIONS, ATTRIBUTE AGENDABUILDING, TRUST, AND CORPORATE REPUTATION MEDIATED BY EMOTION By JI YOUNG KIM 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 2012 1

PAGE 2

2012 Ji Young Kim 2

PAGE 3

To my father, mo ther, and sister 3

PAGE 4

ACKNOWLEDGMENTS I am deeply grateful for the support and guidance of the many people who helped me complete work on my dissertation. Du ring a long journey, I was surrounded by several good mentors and suppo rters who kept me moving forward. Without their support, this work could not be done. First and foremost, I would like to expre ss my gratitude to my advisor, Dr. Spiro Kiousis, for his encouragement during the ent ire period of my doctoral study. When I first met Dr. Kiousis, he said that this resear ch institution in the Sunshine State would provide me with great opportuniti es to develop my career goals, and he was totally right. I have enjoyed meeting with my public rela tions students as an instructor and being involved with great research teams during the last four years. Also, I want to thank the members of my committee who hav e been extraordinary mentors throughout my doctoral study. Dr. Juan-Carlos Molleda has been a cheerful supporter for me, and I appreciate his great in sight and professional advice. Whenever I brought a question from my studies to his of fice, he was always willing to help and to discuss it with me. Also, his professionalism and passion for contributing to the field kept me constantly motivated. I couldnt have completed this dissertation without the help and ad vice of Dr. Mary Ann Ferguson. She has been very understa nding about what graduate students face during their final year of study. Her advice has helped me to manage my time and to enjoy the process. I am also very fortunate to have Dr. J ohn Kraft as my outs ide committee member. He gave lots of insightful and straightfo rward advice about my work. I have always 4

PAGE 5

enjoyed our meeting time which occurred in the most welcoming and cheerful atmosphere. I would like to express my deepest grat itude to Dr. Lynda Lee Kaid, who passed away last year. Thanks to the opportunity to join her research team in 2008, I developed interdisciplinary work expe riences. Her emphasis on collaborative learning processes taught me about the value of co llaboration throughout my study. And of course, I want to send my sincer e thanks to my professors at Syracuse University. I wouldnt be here right now without my Syracuse days. I appreciate all the encouragement I received from Professor St eve Masiclat, New Media Director, who helped me overcome the cultural and ac ademic challenges when I began my graduate study. And I will never forget my very firs t research class with Dr. Pam Shoemaker. I greatly appreciate her pati ence and thoughtful advice. Dr. Dennis Kindsey, Public Diplomacy Director, was a mentor who showed me ways to increase the value of life. And last but not least, Dr. Sung-Un Yang has been my mentor all through my studies. Dr. Yang and his family always made me feel like I was home when I was around them. I would like to thank my lovely fam ily for their unconditional love, support and encouragement. My family has always stood by me, so I could believe in myself and become a positive person. Since 2005, Ive been through a long journey as a foreign student who roamed around new majors and new places. My father, Hong Jin Kim, always gave full support for my decisions, an d encouraged me to try new things. From my mother, Hyun Sook Lim, I learned how to thank everyone for what I have even when I have a hard time or when I have hundreds of thi ngs to do. I also want to thank my little sister, Ji Yoon, for being a good friend. Her energy and wit always keeps me smiling. 5

PAGE 6

Also, I am so blessed and lucky to be surrounded by many brilliant and nice friends, whose support has inspired me a lot in my career. I would like to thank these amazing people: Sooyeon Hong, Hyunjin, Hyekyung, Joyce Jungwon, Jiyeon, Junghyun, Jiyoung Cha, Sooyeon Kim, Jooyun, Maria, Juliana Bumsub, Chunsik, Hyunji, Soo, Jungmin, Sunyoung, Wan, Dae-Hee, Jenny, Vanessa, David, Yeonsoo, Jinhyon, Hyejoon, Maguire friends (Moonhee, Jaejin, Soojin, and June-yung), Rajul, Hanna, Mari Luz, Eunhwa, Yoojin, Youngeun, Hyunsang, Kor ean community colleges, co-workers, and my students. I appreciate all who cheered me on every time I needed help and who made my time with them full of fun. Again, it would have been near ly impossible to accomplish this dissertation without all the help and support of many people. I sincerely thank all of them. 6

PAGE 7

TABLE OF CONTENTS Page ACKNOWLEDG MENTS..................................................................................................4LIST OF TABLES..........................................................................................................10LIST OF FI GURES ........................................................................................................11ABSTRACT ...................................................................................................................12CHAPTER 1 INTRODUC TION....................................................................................................132 LITERATURE REVIEW..........................................................................................21Agenda-Buildi ng Theo ry.........................................................................................21Agenda-Setting Theory....................................................................................21Agenda-Building Theory and Public Relations.................................................24Firstand Second-Level of Agen da Building and Agenda Setti ng....................26Substantive a ttribut es................................................................................28Affective at tribut es.....................................................................................30Compelling Argument and Priming E ffects.......................................................33Corporate Communication and Issues in Business................................................36Corporate Communication and Stakehol ders...................................................36Agenda Building in the Business Context.........................................................39The Role of Issues in Busi ness........................................................................43Definition of issue-ar ena...................................................................................44Issue Owner ship...............................................................................................46Emotional Mediating Factor in the Agenda-Building Process.................................48Trust as a Relationshi p Quality Ou tcome................................................................53Definition and Measur ement of Trust...............................................................53Corporate Communica tion and Tr ust...............................................................55Corporate R eputatio n..............................................................................................57Definition of Corpor ate Reputat ion...................................................................57Measurement of Corp orate Repu tation............................................................59Relationship between Message Exposur e and Corporate Reputation.............60Hypotheses and Resear ch Ques tions ....................................................................623 METHOD S..............................................................................................................70Experimental Design...............................................................................................70Participants .............................................................................................................72Procedur e...............................................................................................................73Pretest and Manipul ation C heck.............................................................................73Pretests Pr ocedure...........................................................................................73 7

PAGE 8

Stimuli 1: Issue Ownership...............................................................................74Stimuli 2: Tone of Issue Attr ibutes ....................................................................74Pretest Re sults.................................................................................................75Questionnaire Construc tion and Meas ures.............................................................75Issue Salie nce..................................................................................................76Attribute Sa lience.............................................................................................77Emotio n............................................................................................................77Priming E ffect...................................................................................................78Trust.................................................................................................................78Corporate R eputatio n.......................................................................................79Methodological Concerns.......................................................................................80Statistical Analys is..................................................................................................804 ANALYSES A ND RESULT S...................................................................................85Description of Collected Data.................................................................................85Descriptive St atistics...............................................................................................86Issue Ownership and Issue A ttributes Manipulatio n.........................................86Issue Salie nce..................................................................................................87Attribute Sa lience.............................................................................................87Priming E ffect...................................................................................................87Emotio n............................................................................................................87Trust.................................................................................................................88Corporate R eputatio n.......................................................................................88Evidence for Research Q uestions and Hy potheses................................................88Hypotheses Testing and Answer ing Research Questi ons................................88Additional Analysis: Path Analysis of Pr oposed M odel.....................................965 DISCUSSION AND CONCLUSIO NS....................................................................114Theoretical Imp lications ........................................................................................114Agenda-Building Theory.................................................................................115Corporate Communication and I ssues in Bu siness........................................120Mediating Factor in the A genda-Building Pr ocess..........................................122Trust as a Relationship Quality Outcome and Corporate Reputation.............124Practical Implic ations ...........................................................................................126Limitations and Suggested for Future Re search...................................................122Conclusion s..........................................................................................................134APPENDIX A STATEMENT OF INFORMED CON SENT............................................................135B EXPERIMENTAL STIMULI...................................................................................136C QUESTIONNA IRE................................................................................................141LIST OF REFE RENCES.............................................................................................148 8

PAGE 9

BIOGRAPHICAL SKETCH ..........................................................................................164 9

PAGE 10

LIST OF TABLES Table Page 3-1 Treatment condition combinat ions of independent variable s..............................823-2 Representation of each research cell.................................................................823-3 Meas ures............................................................................................................834-1 Demographi c profile s..........................................................................................994-2 Participants in each manipulation c ondition ......................................................1004-3 Frequency for each conditi on...........................................................................1004-4 Means and standard devia tions for m easures..................................................1014-5 Principle Axis Factor loading of emotion...........................................................1034-6 Two-way analysis of variance of iss ue ownership and issue tone on emotion.1044-7 Main effects of issue ownersh ip and issue tone on emotio n.............................1054-8 Pearson product-moment correlation coeffi cients.............................................1064-9 Two-way analysis of variance of iss ue ownership and issue tone on salience.1074-10 Main effects of issue ownersh ip and issue tone on salienc e............................1084-11 Direct and indirect effe cts on endogenous va riables ........................................1094-12 Summary of hypot heses testing.......................................................................110 10

PAGE 11

LIST OF FIGURES Figure Page 1-1 Compelling arguments: Attribut e effects on objec t salience...............................202-1 A theoretical framew ork......................................................................................672-2 Proposed comprehensiv e conceptual model......................................................682-3 Untested relationshi ps among va riables .............................................................694-2 Results of proposed m odel...............................................................................1124-3 Path analysis of the proposed model................................................................113 11

PAGE 12

12 Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy EXPLORING LINKAGES AMONG PUBLIC RELATIONS, ATTRIBUTE AGENDABUILDING, TRUST, AND CORPORATE RE PUTATION MEDIATED BY EMOTION By Ji Young Kim August 2012 Chair: Spiro Kiousis Major: Mass Communication One of the main contributions of this study was to explore the predictors of corporate reputation in a corporate co mmunication context. Grounded in the agendabuilding perspective, the current study exami ned the role of issues in public relations communication strategies. The purpose of the study was, 1) to demonstrate a causal relationship in agenda building, 2) to explore the role of affect in the relationship, and 3) to expand the scope of application of the co ncept into corporate communication. In particular, a 2 x 2 between-subject factor ial experiment was conducted with 225 participants. These participants were recr uited from an online panel service and received monetary compensation. Analysis of variance, regression, and a path analysis model were utilized to test the relationships. So me direct and indirect effects were found in the model, supporting the medi ating role of emotions in the relationship between the two independent variables (issue ownership and issue tone) and dependent variables (issue salience, trust, and cor porate reputation). Theoretical and practical implications are presented in Chapter 5.

PAGE 13

CHAPTER 1 INTRODUCTION Society is a collective entity consisting of multiple publics (stakeholders) who have different interests and needs (Heath, 2006). From an organizational standpoint, publics are a group or individuals who affect and/ or are affected by an organizations performance. Scholars have stated that a satisfactory balance of interests among diverse publics is essential for organizations to reach their goals (Harrison & Freeman, 1999). The increase of empowered publics due to the communication technology development and globalization has provid ed opportunities and challenges in the public relations field. For example, the role of publics has been emphasized in several public relations theoretical frameworks, such as relationship management or issue management (Luoma-aho & Vos, 2010). Focusi ng on long-term relati onship building, organizations, including corporations, have been involved in diverse social issues in their community and have been engaged in that public debate. Scholars have used the terms issue arenas, referring to the place w here diverse social factors interact with each other about community issues (Luoma-aho & Vos, 2010). The reactions of organizations towards the issues affect how these organizations are perceived by various publics (Luoma-aho & Vos, 2010). Miller (2010) found that issue advocacy practices affect the changes in approval ratings toward an industry. Meijer and Kleinnijenhuis (2006) also stated that organi zations competency in dealing with social issues affects the overall evaluat ions of various organizations by the public. In particular, scholars have explored the e ffects of issues on organizational reputation as one of the important outcomes of good rela tionships. The effect of issues has been widely explored in political communication. Sc holars have pointed out that when issues 13

PAGE 14

are primed, parties who owned the issue wo uld receive more attention and electoral gains (e.g., Kleinnijenhuis, Maurer, K epplinger, & Oegema, 2001; Petrocik, 1996; Petrocik, Benoit, & Hansen, 2003; Sheaf er & Weimann, 2005; Zaller, 1992). The impact of the media has focused greatly on the issue management perspective of public relations. In particular, scholars have explored the effects based on the theories of priming, framing, agenda setting, or agenda building (e.g., Kleinnijenhuis et al., 2001; Petrocik, 1996; Sheafer & Weimann, 20 05). Petrocik (1996) stated that framing helps publics recognize certain aspects of social problems while ignoring others, and framing affects subs equent public judgment and behavior. Sheafer and Weimann (2005) explored the role of media in Israeli election results grounded in the agenda-setting and agenda-building processes. Their findings indicate associations among the media, public opi nion, and associations bet ween the media and voting behaviors. Priming effects were supported show ing that the media hel ped voters to set their evaluation criteria for the polit icians (Sheafer & Weimann, 2005). To examine the role of issues in public relations, this study focused on the issue ownership concept. Exploring the priming effort s of political campaigns, Petrocik et al. (2003) stated that some political issues c an easily be attributed to certain parties. Publics used to believe that Democrats dealt well with such issues as social welfare while Republicans dealt well with such issues as taxes or size of government (Petrocik et al., 2003). The concepts assumption was that political parties competence in dealing with certain issues affects voters supportive behaviors toward the political parties when the issue is significant among voters (e.g., Belanger & Meguid, 2008; Cha, Song, & Kim, 2010; Elmelund-Praestekaer, 2011; Green & Hobolt, 2008). On the other hand, little 14

PAGE 15

empirical research has been conducted on the e ffects of issue ownership in corporate communication. Cha, Song, and Kim (2010) def ined issue ownership as the level of capability that a corporation has shown us all certain social issues. Meijer and Kleinnijenhuis (2006) indicat ed that issues in business news transferred into the mindset of the publics criter ia which was then used to evaluate the corporation involved in those issues. Issue ownership effects depend greatly on the role of the media and public relations efforts. Prior scholarship has emphasiz ed the role of media, indicating that the mass media can force public attention to ce rtain issues while ignoring others (e.g., Entman, 1993; Lang & Lang, 1966; Sheafer & Weimann, 2005). In particular, the agenda-setting hypothesis states that elements prominent in the mass medias picture of the world influence the pr ominence of those elements in the audiences picture (McCombs, Lopez-Escobar, & Llamas, 2000, p. 77). McCombs and Shaw (1972) found that the public would consider certain issues more prominently when those issues are portrayed by the media. For instance, Iy engar and Simon (1993) ascertained that the portrayal of the Gulf War crisis on TV news influenced the public in deciding what were the most important issues in the United States. Expanding on agenda setting, scholars hav e also suggested an agenda-building theory focusing on the role of information subsidies in building the media agenda and the public agenda (e.g., Kiousis, Mitrook, Wu, & Seltzer, 2006; Kiousis & Wu, 2008). Ohl, Pincus, Rimmer, and Harrison (1995) exam ined the relationship between information sources and media gatekeepers, and Turk (1986) found links between the agendas of state governments news releases and news media coverage. Providing information to 15

PAGE 16

the media and the public is one of the most important jobs for public relations professionals. This is accomplished through news releases, advertisements, campaigns, or speeches. Moreover, scholars have ex plored the linkages among policy agendas, media, and the public to define the role of public relations and to discuss public relations strategies (i.e., McKinnon, Tedesco, & Lauder 2001; Kiousis & Strmbck, 2010). The main idea of agenda building is the transf er of salience (in the form of objects or attributes) from one to the other. Scholars have distin guished the two levels of salience transfer: the salience of object (first-level) and the salience of attributes (second-level) (e.g., Kiousis, 2005; Lee, 2010). Attributes are defined as certain aspects of an object (Ghanem, 1997). When a political candidate is an object, his/her personality or position on the issues can be their attr ibutes. Researchers have shown evidence of the transfer of attribute salience (e.g., Golan & Wanta, 2001; Lee, 2010; McCombs & Shaw, 1972). The first level (object level) has been studied with regard to the strength of opinion, and the second level (attribute le vel) has been studied with regard to the direction of opinion (McCombs, 2004). Furt hermore, scholars have indicated that the two levels can also be linked. Kiou sis (2005) suggested t here are compellingarguments hypothesis explaini ng the linkages between attribute level of salience and object level of salience (Figure 1-1). For exam ple, evaluations by the public of political candidates differ between what the media portra ys as the family history of a political candidate and what the media portrays as the candidates policy issues. The compelling-arguments hypothesis leads this studys attention to priming effects and other consequences of the agenda-building process. Priming effects are measured by the importance of the issue and the perception of the organizations 16

PAGE 17

influence (Iyengar & Kinder, 1987). Grounded in agenda-building theor y, this study measures the priming effects on corporate re putation and trust. For example, reputation management becomes the central framework in corporate public relations (Hutton, Goodman, Alexander, & G enest, 2001). However, how and when reputation can be managed still has not been fully explored. Reput ation is defined as a perceptual representation of a companys past actions and future prospects that describes the firms overall appeal to all of its key c onstituents when compared with other leading rivals (Fombrun, 1996, p. 72). Wartick (1 992) also defined reputation as the aggregation of a single stakeholde r's perceptions of how well organizational responses are meeting the demands and expectations of many organizational stak eholders (p. 34). At first, the role of publ ic opinion was emphasized in the political communication arena; however, corporations have also become a social entities influencing and influenced by public opinion. The need for co mmunication management by corporations has increased the understanding of public opinion and the impr ovement of the quality of corporate-stakeholder relationships. Clark (2 000) stated that communication managers and business managers recognize the need to analyze multiple stakeholders (sometimes referred to as publics) to dev elop a sense of the needs and wants of those who are either critical to t he corporations existence or c apable of expressing significant concern (primary and secondary stakeholders ) (p. 374). J. Grunig, L. Grunig, and Ehling (1992) also emphasized the significance of organization-public relations, stating that the quality of the relationship mi ght prevent negative pub lic opinion about corporations. J. Grunig (1979) stated that cor porations would receive public attention by dealing with social issues since publics believe that corporations should be involved in 17

PAGE 18

responsible social actions. Findings of this current study would contribute to the following communication areas: public relation s strategies for business/organizations, corporate reputation and trust, and agendabuilding and agenda-sett ing theories. Trust was measured as one of the relation ship indicators. As suggested by Hon and J. Grunig (1999), trust has been consid ered one of the important public relations outcomes, measured by such multidimensio nal items as integr ity, dependability, or competence. In a business environment, corpor ate reputation is cruc ial when issues in business news become salient in the public mind, and these issues may act as a yardstick for evaluating those firms. For example, if a news story appears about an issue which the public perceives the organization to be handling successfully, that is, an owned issue, then the reputation of the organization will improve. If a new s story emerges about an issue which the public perceives the organization as being inc apable of handling, that is, a poorly owned issue, then the reputation of the organization will worsen (Me ijer & Kleinnijenhuis, 2006). This study explored how reputation and trust are affected by communication strategies in public relations focusing on the publics perception of issue ow nership and affective attributes of the messages. This study also examined the role of emotions in the agendabuilding process and effects. The study has two dimensions of attributes: substantive and affective dimensions (e.g., Golan & Wanta, 2001; McCombs & Shaw, 1972). Substantive attributes refer to the cognitive aspect of attributes, while affective attributes refer to the emotional (affective) aspect of attributes (e.g., Golan & Wanta, 2001). Frames are an example of the cognitive aspect of agenda building (agenda setting), and the tone 18

PAGE 19

(valence) has been used as an example of the affective aspect of agenda building (agenda setting) (e.g., Golan & Want a, 2001; McCombs & Shaw, 1972). Previously, scholars have stated that affect ive attributes would lead to emotional responses to a message and shape the over all composition of the object in the message (Schoenbach & Semetko, 1992; Sheafer, 2007). Coleman and Wu (2010) showed that the audiences emotions strongl y affected the overall evaluation of a political candidate by increasing the need fo r in-depth cognitive thinking about the candidate. Sheafer (2007) also found that t he evaluative tone of news media coverage affects voters political judgments. However, the role of the affective dim ension of attributes and emotions was less developed than the cognitive dimension in agenda-building studies. In addition to the valence (tone) of the message, this was widely used as the affective attribute, this study also aimed to see how emotions mediate the agenda-building effects. Emotional arousal measures how much an individual feels or is stimulated (or not stimulated). Hence, by linking issue ownership to the agenda-building theory, the purpose of this dissertation was, 1) to explore how the public relations practices of corporations are related to priming effect, trust, and cor porate reputation, and 2) to explore the role of emotion in the process and effect. In parti cular, using an experimental des ign, the current research manipulated issue ownership an d affective attributes of issues and explored the linkages with the corporat ions agenda-building outcome mediated by emotions. 19

PAGE 20

20 Figure 1-1. Compelling arguments: Attribute effects on object salience (Reprinted by permission from McCo mbs, Maxwell. 2004. Setting the agenda: The mass media and public opinion (Page 92, Box 6-1). Malden, MA: Polity)

PAGE 21

CHAPTER 2 LITERATURE REVIEW This chapter reviews previous literat ure related to the agenda-building theory, corporate communication and issu es in business, mediating factors in agenda building, and the consequences of the agendabuilding process. In par ticular, the consequences of agenda building have been explored in terms of the trust as a relationship quality outcome for corporate reputation. The linkages between t hese theoretical frameworks and each concept used in the current resear ch are explored. Fi gure 2-1 presents the theoretical framework of this study among public relations, emotional mediators, agenda salience, and public opinion. Agenda-Building Theory This section provides a theoretical overview of agenda building and agenda setting and presents the firstand second-level of the processes. In terms of the second-level of agenda building and agenda setting, this se ction discusses both substantive and affective attributes, and the linkages between the two levels, the so-called compelling arguments hypothesis. In addition, the pr iming effects of agenda building are also examined. Agenda-Setting Theory Since McCombs and Shaws (1972) propos al, the agenda-setting hypothesis has been widely investigated in various communica tion settings (e.g., Iyengar & Kinder, 1987; Kiousis & McDevitt, 2008; Kiousis et al., 2006; McCombs, 2004; McCombs & Reynolds, 2002). Conducting content analysis and surveys, McCombs and Shaw (1972) found a correlation on issue agendas between t he media and the public. Certain issues prominently presented in media coverage were also considered important issues by 21

PAGE 22

undecided voters (McCombs & Shaw, 1972). The main idea of the hypothesis is that elements prominent in the mass medias pict ure of the world infl uence the prominence of those elements in the audiences picture (McCombs et al., 2000, p. 77). Numerous previous studies have provided robust evi dences supporting this hypothesis (e.g., Kiousis & McDevitt, 2008; Kiousis et al., 2006; McCombs, 2004; McCombs & Reynolds, 2002). Dearing and Rogers (1996) noted that approximately 60% of the studies supported the agenda-setting hypothesis; they reviewed 112 empirical studies about the salience transfer between the media and the public agenda. The agenda-setting hypothesis is grounded in a tradition of the significant media effects perspective. The media choose and display certain news stories while ignoring others, and the news media influences what people know about the world (McCombs & Shaw, 1972). Lang and Lang (1966) emphasized the function of the media reporting that the mass media force attention to cert ain issues. They build up public images of political figures. They are constantly pres enting objects suggesting what individuals in the mass should think about, know about, have feelings about (p. 468). Cohen (1963) also stated that the press may not be successful much of the time in telling people what to think, but it is stunningly successful in telling its readers what to think about (p. 13). In political communication settings, sc holars hypothesized the influence of the mass media on public opinion and attitudes toward certain political issues and candidates. Iyengar and Simon (1993) found evi dence that the increase in television news coverage of the conflict in the Pers ian Gulf, between 1990 and 1991, influenced the political concerns of Americans about the Persian Gulf cris is such that it became the nation's most important problem. Sheafer (2007) also supported the agenda-setting 22

PAGE 23

linkage to media coverage, and public opinion showed that t he medias attention to the economy affects survey respondents awareness of the issue. The agenda-setting function of the media is also emphasized in international events. Wanta, Golan, and Lee (2004) stated that the need for international news has increased due to globalization fo llowing the end of the Cold War in 1991 and the rise of terrorism. The American publics concern for international events is higher than ever. Few individuals have direct experience with news event in foreign countries. For many, the sole source of information about world events is the press. Media coverage of international news then should play an impor tant agenda-setting function (Wanta et al., 2004, p. 367). Yang, Shin, Lee, and Wrigle y (2008) also emphasized the effects of mass media on a countrys reputation, along with the individuals first-hand experiences and interpersonal communication. Correlations have been found between forei gn news coverage and public opinion supporting the agenda-setting effe cts of international news coverage (e.g., Salwen & Matera, 1989; Semetko, Brzinski, Weaver, & Willnat, 1992; Wanta et al., 2004; Wanta & Hu, 1993). Conducting both content analysis and surveys, Salwen and Matera (1989) suggested that the amount of media cove rage about other nations can influence changes in what the public thinks about other nations over time. Semetko et al. (1992) also stated that a relationship exists between other nations visibility on TV news in the United States and U.S. public opinion about other nations. Sc holars have stated that the publics perception of other nations is affe cted by how those nations are portrayed in international news through diverse frames (S emetko et al., 1992). Moreover, Wanta and Hu (1993) indicated that certain issue ca tegories have stronger agenda-setting effects 23

PAGE 24

than other categories. The f our categories demonstrating strong agenda-setting effects were international conflict (involving the Un ited States), terrorism (involving the United States), crime/drugs, and military/nuclear arms (Wanta & Hu, 1993). Agenda-Building Theory and Public Relations Expanding from the agenda-se tting theory, communication scholars have also explored the antecedent fact ors of agenda-setting effects, and they have focused on how the media agenda can be set in the first place (e.g., Kiousis et al., 2006; Kiousis & Wu, 2008). Gandy (1982) addressed the need for broad conceptual understanding of agenda setting, suggesting that we go beyond agenda-setting to determine who sets the media agenda, how and for what purpose it is set, and with what impact on the distribution of power and values in society (p. 7). From a public relations perspective, the influence of external sources on the m edia and the public agenda is called agenda building, referring to sources interacti ons with gatekeepers, a give-and-take process in which sources seek to get their informa tion published and the press seeks to get that information from independent sources (Ohl et al., 1995, p. 91). Several external information sources affe ct media coverage and public opinion, directly or indirectly, including information subsidies, such as press conferences, news releases, or interviews. Miller (2010) expl ained that public relations professionals provide information subsidies to journalists to sway media attention, reducing journalists cost and efforts to gather information. In a congr essional election setting in Michigan, Kaid (1976) noted that journalis ts often use campaign announcements or candidate information provided by public relati ons professionals on their news stories. Turk (1986) also found evidence of the subsequent use of state governments news 24

PAGE 25

releases on media coverage. Gans (2003) stat ed that journalists respect their official sources, reporting what these sources tell them (p. 46). Similar to agenda-setting studies, the agenda-building effects were also investigated widely in political campaign co mmunications (e.g., Kaid, 1976; Kiousis et al., 2006; Kiousis & Wu, 2008), but t he agenda-setting and agenda-building concepts have also been applied to other communication settings in business or international events (e.g., Carroll & McCombs, 2003; Ohl et al., 1995). For example, Ohl et al. (1995) focused on the effectiveness of a companys press releases in a hostile corporate takeover context and found agenda-building e ffects from news releases on media coverage. Carroll and McCombs (2003) stat ed that the news media portrayals of corporations can influence how the public perceives the corporations. Agenda-building theory has m any meaningful implications in public relations. Lang and Lang (1981) declared that a genda building is useful to conceptualize the salience relationships among policymakers, the m edia and the public, and McKinnon et al. (2001) pointed out that agenda bu ilding emphasizes the role of public relations professionals in successful political candidacies. Kiousis and Strmbck (2010) also supported agenda-building effects on public opinion by studying the relationship between presidential public relations me ssages and job approval ratings. Comparing general or issue-specific job approval rati ngs, Kiousis and Strmbck (2010) showed different agenda-building effects across topics between two types of public relations messages. Kiousis et al. (2006) also explai ned the theoretical implications of agenda building in public relations: 25

PAGE 26

This works to further our knowledge of the role of news releases and other public relations information subsidies to journalists in helping to shape both the media and public agendas. This knowle dge is also useful in furthering our understanding of other public relations theory areas as well. (Kiousis et al., 2006, p. 279) According to the scholars, an agenda-building per spective is valuable to other public relations theories, such as the conti ngency theory, relationship theory, and issue management theory (Kiousis et al., 2006). For example, public relations professionals affect the publics recognition of certain i ssues as problems or their awareness of certain organizations through salience formation. Consequently, these communication efforts can be utilized to affect the relati onship between an organization and its public, and to get the public more active in cert ain social issues (e.g., Hallahan, 2001; Ledingham, 2001; Reber & Cameron, 2003): Level one of agenda setting examined how individuals learn about the major issues of the day through medi a coverage an important process in political communication. This level emphasized the news medias role in generating awareness of iss ues that should concern the public. (Golan & Wanta, 2001, p. 258) Firstand Second-Level of Age nda Building and Agenda Setting In both the agenda-building and agendasetting frameworks, scholars have defined two levels of salience transfer: obj ect and attribute salience (e.g., Kiousis, Bantimaroudis, & Ban, 1999; Kiousis et al., 2006; McCombs & Reynolds, 2002; McCombs & Shaw, 1972; Weaver, Graber, McCombs, & Eyal, 1981). Referring to the first-level of agenda building and agenda setti ng, object salience transfer has been explored with issues or candidates; on the ot her hand, referring to the second-level of agenda building and agenda setting, attribute sa lience transfer has been explored with certain aspects of objects (e.g., Kiousis et al., 1999; Lee, 2010; McCombs & Shaw, 1993; Behr & Iyengar, 1985; Iy engar & Kinder, 1987; Iyengar, Peters, & Kinder, 1982). 26

PAGE 27

Ghanem (1997) defined attribute as the set of perspectives or frames that journalists and the public employ to thin k about each object (p. 5), and McCombs, Llamas, Lopez-Escobar, and Rey (1997) referred to the second-level of agenda setting as attributes salience of an object or the actor concerned. McCombs and Estrada (1997) asserted that these per spectives and frames called semantic devices draw attention to certain attributes and away fr om others (p. 246). Ev idence was found to support the fact that attribut es of an issue are transferred from the media to the public and from public relations information sources to the media and the public (e.g., Golan & Wanta, 2001; Kim, Scheufele, & Shanahan, 2002; Kiousis et al., 1999; Shaw & McCombs, 1977). Describing the emergence of the attribute level of agenda building and the agenda-setting theory, Kiousis (2005) stated that it has shi fted the focus of research away from investigating what topics news m edia cover to how they cover them (p. 4). Ghanem (1996) suggested four dimensions of second-level agenda building and agenda setting, subtopics, the framing mechani sm, affective elements, and cognitive elements. McCombs and Shaw (1972) also inve stigated the salience of attributes in addition to the salience of issues in terms of affective and cognitive attributes. Affect denotes a pro/con orientation, a feeling of liking or disliking something. Cognition, by contrast, denotes the individuals perception of the attitude object, his image or organized set of information and beliefs about a political object (McCombs & Shaw, 1972, p. 186). Previous scholars also defined the two dimensions of attributes as cognitive attributes (substantive attributes ) and affective attri butes (e.g., Carroll & McCombs, 2003; Kiousis et al., 2006; McCo mbs & Shaw, 1972). Substantive attributes, 27

PAGE 28

based on reasoning, were commonly explored in terms of frames of objects, and affective attributes were explored in terms of the evaluative t one of objects. Golan and Wanta (2001) explained that bot h cognitive and affective attri butes influence individuals evaluations of political candidates. Specifica lly, they stated that t he attribute level is essential to learning the char acteristics of political cand idates. In a business context, Carroll (2004) tested both t he firstand second-level of agenda setting, showing that media descriptions of the issues or attributes of the firms were associated with public perceptions about the firm regarding the issues. Carroll and McCombs (2003) also defined two levels of agenda setting: At the first level, agenda setting effects are on attention. At the second level, agenda setting effects are on compr ehension. These attribute agendasetting effects on public comprehension can be described in terms of two dimensions: substantive and evaluative (o r, cognitive and affective, if you will). (Carroll & McCombs, 2003, p. 38) Substantive attributes Cognitive or substantive attributes of agenda building and agenda setting deal with the logical aspect of the object, while affect ive attribute deals with the evaluative tone toward the object (Sheafer, 2007). Scholars have defined substantive dimension of attribute as the characteristics of news t hat help us cognitively structure news and discern among various topics (Kiousis et al., 1999, p. 417). Frames have been widely used in terms of the cognitive attribute of an object (McCombs et al., 2000). McCombs et al. (2000) distinguished tw o definitions of frames: In the case of the centra l theme, our concern is wit h the central focus of the picture. In cases of aspects, the frame distinguishes between what the picture included and what is outside, a use of the term very similar to the idea of framing in photography. (McCombs et al., 2000, p. 79) 28

PAGE 29

Tankard, Hendrickson, Silberman, Bliss, and Ghanem (1991) defined framing as the central organizing idea for news content that supplies a context and suggests what the issue is through the use of selection, empha sis, exclusion and elaboration (cited in Golan & Wanta, 2001, p. 248). An object ha s several aspects and one could focus on certain aspects while ignoring others through the framing process; the frame of an object affects peoples learning about soci al issues (Entman, 1993). Entman (1993) stated that the faming process can explain how people interp ret certain social issues and how people perceive the causes of those social issues. This substantive attribute of agenda-se tting influence on the mass media was explored widely concerning public opini on about candidate images, and various candidate characteristics were suggested as the sources that defined the image of political candidates. For example, McCo mbs et al. (1997) suggested ideology, qualification, and personality attributes to describe the traits of a specific candidate. Kiousis (2005) also defined several dimensions of candidates cognitive attributes, including intellectual ability, moral quality, and leadership ability. Both intellectual ability and moral quality are associated with a cand idates personal traits, and leadership ability is associated with a candi dates qualification (Kiousis, 2005). The attribute level of agenda setting and agenda building have al so been explored in a business context. Carroll and McCombs (2003) argued that the discussion of the transfer of salience between the media and the public is also well described in a business news context. Adopting traits from pr evious corporate reputation studies (e.g., Fombrun, 1998; Fombrun, Gardberg & Se ver, 2000), Carroll and McCombs (2003) 29

PAGE 30

suggested six substantive attributes of a corpor ation. The six key attr ibutes identified by Carroll and McCombs (2003) are as follows: Familiarity: Know ing the companies or its products well; Creating value: Producing high quality products, providing value for the money; Operational capability: Being well-run, efficient, and productive; Corporate citizenship: Caring ab out its employees and the community; Performance: Proven track record, good use of assets; Leadership/management: Having a CEO with vision, communicating values; Appeal: Being liked by stakeholders, being a good company to work for; Credibility: Being trustworthy, standing behind its practices (Carroll & McCombs, 2003, p. 40, adapted from Fombrun et al., 2000) Affective attributes In addition to the substantive (cognitive) di mension of attributes, the salience of the evaluative (affective) aspect of attri butes has also been considered an essential feature of the second level of agenda bui lding and agenda setting (e.g., McCombs & Shaw, 1972; Shrauger, 1967). Affective attr ibutes are defined as the emotional responses of the public (Kiousis et al., 1999). Deephouse (2000) defined the affective dimension of attribute as the overall evaluation of a fi rm presented in the media resulting from the stream of media stories about the firm (p. 1097) emphasizing the influence of the media in affecting corporate reputational evaluations. Sheafer (2007) explained that the evaluative tone of ne ws media coverage affects voters political judgment. In particular, Sheafer (2007) highlighted the significant role of negative affective attribute in influencing pu blic perception about issue importance: 30

PAGE 31

Information about negative developments c aptures our attention far more than information about positive deve lopments. A negative object attribute is, therefore, expect ed to increase object importance and accessibility on the public agenda. A positiv e tone, on the other hand, is not expected to have such an effe ct. (Sheafer, 2007, p. 23) Moreover, Schoenbach and Semetko (1992) f ound that a positive a tone in media coverage decreases public perception of iss ue importance since negative news catches the publics attention more naturally than either positive or neutral news. Explaining the significant effect of negative information, Sheafer (2007) also mentioned that the negat ive valence is associated with t he operational definition of the public agenda. To measure the importance of the pub lic agenda, researchers have asked the nature of the most important problems now facing the nation. Several researchers found significant linkages between an affective tone of news coverage and public opinion on candidate imag es, issue frames, or national image (e.g., Kepplinger, Donsbach, Brosius, & Staab, 1989; Kim & McCombs, 2007; S heafer, 2007; Wanta & Mikusova, 2010). In a longitudinal media content and public opinion analysis, Kepplinger et al. (1989) found that the medias evaluative assessments of a politician shifted public opinion about the politician. Kim and McComb s (2007) also explored the effects of affective attributes in gubernatorial and senatorial elections. Candidates positive or negative portrayals of the media coverage were related to public opinion about each candidate (Kim & McCombs, 2007). In an international news context, Wanta and Mikusova (2010) also supported the attribute level of agenda-setting effects. The findings showed that the tone of the news coverage affects whether the news readers assess the nation positively or negatively. 31

PAGE 32

As a critical element of persuasion, scholars have emphasized the role of affect in communication, in areas such as advertising or public relations. Affect refers to feeling, emotions, modes, or personalit ies (Izard, 1993), and scholars have explained that affect plays a significant role in affecting cogni tive thinking (e.g., Arnold, 1985; Dillard & Wilson, 1993). Batra ( 1986) defined affect as feeling toward a stimulus that leads to relative preferences toward that stimulus out of a class of similar stimuli (p. 54). From a psychological perspective, public evaluation on objects is always associated with some kind of affective evaluation (Morris, Squires, Taber, & Lodge, 2003). Focusing on affective priming effects, sc holars have stated that political terms are affective in nature, adding that it has been speculated that affect ive evaluations are stored in memory for all political concept s that have been repeatedly evaluated in the past, and that these affective responses are aut omatically elicited when the concept is activated in memory (Morris et al., 2003, p. 742). Scholars have explained that the affective attribute will direct the publics emotional responses and shape the overall evaluation of the message (e.g., Kious is et al., 1999; Lopez-Escobar, Llamas, McCombs, & Lennon, 1998b). The valence (or tone) of a message has been widely used to investigate the effect of an a ffective attribute of agenda building and agenda setting; however, the role of affective a ttributes has been developed in agenda building and agenda setting. Dealing with the affective dimension of communication messages, Lang, Dhillon, and Dong (1995) showed that arousing messages can increase the overall effectiveness of advertising, and Keller and Block (1996) found that emotionally arousing messages can affect the publics cognitive information processing. Other 32

PAGE 33

scholars have also emphasized the import ance of emotions in evoking rational information processing (e.g., Coleman & Wu, 2010; Marcus, Neuman, & MacKuen, 2000). Coleman and Wu (2010) show ed that emotional attributes of political candidates are essential to the formation of an overa ll impression of a candidates qualifications. In addition to valence (tone), another dimensi on exists of an affective attribute of messages, as suggested by previous scholars: emotional arousal (stimulating or not) (e.g., Clore, Schwarz, & Conway, 1994; Dunn & Schweitzer, 2005; Forgas, 1995; Gorn, Pham, & Sin, 2001; Schachter & Singer, 1962). According to Dunn & Schweitzer (2005), emotional arousal is disti nguished from valence (mood or tone), and its impact is greater than the valence on cognitive information-processing and overall evaluation of an object. The role of affect in cognitive and over all judgment of an object leads to the compelling argument and priming effect of agenda building and agenda setting. Taken from McCombss (2004, p. 92) book, Setting the Agenda: The Mass Media and Public Opinion Figure 1-1 in that book presents the e ffects of attribute on object salience to explain the compelling-argument hypothesis. Compelling Argument and Priming Effects McCombs (2004) explained that the firs t-level agenda setting is associated with the strength of an opinion, wh ile the second-level agenda se tting is associated with the direction of the opinion. However, scholars also argue that the st rength and the direction cannot be separated from each other in the agenda-building and agenda-setting processes (e.g., Lee, 2010; S heafer, 2007). Lee (2010) insisted that the direction of an opinion needs to be considered with the strength of that opinion. It would not be meaningful to gauge the opinions strength without considering its direction, especially 33

PAGE 34

when thinking of priming effect s, which generally measures how much to support or oppose something (Lee, 2010, p. 773). For example, when the media bring up unemployment as a subtopic of the economy, media coverage not only affects the importance of unemployment, but also affects the importance of the overall economy. Scholars have referred to this relationship as a compelling arguments hy pothesis between object and attribute level of agenda building and agenda setting (e.g., Kiousis, 2005; McCombs, 2004; McCombs & Ghanem, 2001; Severin & Tankard, 2001; Yioutas & Segvic, 2003). To deal with the attribute, McCombs (2004) explained that an object has numerous attributes, those characteristics and properties that fill ou t the picture of each object (p. 70). A certain attri bute of the media affects the a ccessibility of both the object and the attributes on the public (e.g., Ghanem, 1997; McCombs, 2004; McCombs & Ghanem, 2001). Severin and Tankard (2001) defined priming as the process in which the media attend to some issues and not others and thereby alter the standards by which people evaluate objects in the real world (p. 226), and Iyengar and Kinder (1987) found the relationships between agenda-setti ng effects and primin g. The priming function has been emphasized as a consequence of agenda-building and agendasetting effects, and priming is related to both firstand second-level of salience (McCombs, 2004). Emphasizing the role of attr ibute salience, previous sch olars have tried to explore the logical processes among affective attribute, cognitive attribute, and attitude strength (e.g., Golan & Wanta, 2001; Kiousis, 2005). From a psychological perspective, scholars have explained that the increased volume of thinking leads to a more plausible 34

PAGE 35

framework of the relationship between media agenda salience and public attitude strength (e.g., Abelson, 1995; Judd & Johnson, 1981; Tesser, 1978). Evidence shows a stronger effect of substantive attributes than affective attri butes on voters perception of political candidates (Golan & Wanta, 2001); however, several other scholars have also emphasized the role of affective (evaluative) attribute in affecting public opinion toward the object (candidates or issues) (e .g., Lee & Yoo, 2004; Sheafer, 2007). Sheafer (2007) addressed the role of t he evaluative aspect (both strength and direction) on the evaluation of the objects saying that the regular priming hypothesis ignores this affective component and theref ore does not fully explain how people use the primed issue when evaluat ing political actors and making electoral decisions (p. 34). In a regular priming effect, people usually make a judgment on a leader based on cognitive characteristics such as the l eaders performance associated with important issues. They then try to assess whether the leaders performance is good or bad in dealing with the issues (Sheafer, 2007). Howe ver, Sheafer (2007) argued that it is illogical for people to expend lots of cognitive effort to as sess a leaders performance. Instead, Sheafer (2007) suggested the concept of affective priming. When economic growth is primed, people will evaluate the president or the incumbent party based on the affective evaluation (positive or negative) they attach to the economic growth (for instance, is it growing and positive or dec lining and negative) (Sheafer, 2007, p. 26). Sheafer (2007) stated that the affective attributes (posit ive or negative) of information reduces the cost for people to make an eval uation about the object. Explaining attribute priming effects, Kim et al. ( 2002) stated that the priming effect assumed that the media impacted the publics evaluati on: Specifically, attribute priming hypothesizes that 35

PAGE 36

certain issue attributes emphasized in the media will become significant dimensions of issue evaluation among the public (pp. 1112). Kiousis (2003) also examined the association between media coverage and multip le public opinion based on priming and agenda setting. For example, while examining the effects of the Monica Lewinsky scandal in the late 1990s on presidential eval uations, Kiousis (2003) suggested that the media have different effects on cogniti ve and emotional public opinion. Corporate Communication a nd Issues in Business The following section provides a def inition of stakeholder and corporate communication, presents the agenda-building theory in a business context, and explores the role of issues in business in terms of the issue ownership theory. Corporate Communication and Stakeholders Developing theoretical frameworks around activism and issues management, public relations scholars have focused on active publics (Hallahan, 2001). Dewey (1927) defined a public as a group who recognizes problems, and Blumer (1966) defined a public as a group of people who ar e facing and engaging in issues. J. Grunig and Repper (1992) explained different stages to understand the characteristics of publics, such as stakeholders, public, or issue stages: An organization has a relationship with stakeholders when the behavior of the organization or of the stakehol der has consequences on the other. Public relations should do formative research to scan the environment and behavior of the organization to identify these consequences. Ongoing communication with these stakeholders hel ps to build a stable, long-term relationship that manages conflict that may occur in the relationship. (J. Grunig & Repper, 1992, p. 124) Stakeholders then become publics when they recognize a problem, and they enter the issue stage when they organize and create issues from the problem. The emphasis on activism and issue is grounded in the normative t heory perspective in public relations. In 36

PAGE 37

a normative theory, scholars value a two-way sy mmetrical model of public relations, and activism is the ultimate role of public relations (Hallahan, 2001). However, when considering corporate entities, the asymme trical model has also been suggested. Corporations build an unbalanc ed power relationship with an environmental group, a labor union, or a competing corporation si nce they have more resources and control power (Coleman, 1982). Including both active and inactive publics, this study considered publics and stakeholders as exchangeable terms, and followed a broad definition of a public as a potential audience of messages. Luoma-aho and Vos (2010) emphasized t he importance of understanding stakeholders (publics) in both corporat e communication and public relations areas. Decades ago, managers considered stakeholder s as obstacles to their business, and they ignored stakeholders in order to avoi d communication with them. This was one of the reasons that public relations had such a negative connotation in dealing with the media (Wartick, 1992). However, more recent research has suggested that public relations and corporate communication profe ssionals should consider the relationship with stakeholders as a valuable asset of their business, and should focus on building a long-term relationship rather than s hort-term performance outcomes. From an organizational standpoint a stakeholder is an indivi dual or a member of a group who affects and is affected by an organiza tions performance, and a satisfactory balance of interests among diverse stakeholders is essential to reaching organizational goals (Freeman, 1984a or 1984b; Harris on & Freeman, 1999). The impact of stakeholders has been discussed in a stakehol der theory or issue management model rooted in public relations (Luoma-aho & Vos, 2010). Relationship management scholars 37

PAGE 38

have suggested there is a critical impact required for building long-term relationship necessary to satisfy stakeholders, and, c onsequently, to affect the organizations reputation (e.g., Ledingham & Bruning, 2000a or 2000b; Wilson, 2000; J. Grunig & Huang, 2000). Not only not-for-profit organizations but also for-profit organizations and corporations are now involved with many social issues, such as global warming, public health, or human rights. These corporate so cial responsibility issues have become prominent concerns for corporations. A cor porate-stakeholder relationship perspective suggests that corporate social responsibilit y activities influence corporate agendas and become major sources of dialogue bet ween companies and their stakeholders (Bhattacharya, Korschun, & Sen, 2008; Clark, 2000). Studies defined this concept as a commitment to improve community well-being through discretionary business practices and contributions of corporate resources (Kot ler & Lee, 2004, p. 3). Bhattacharya et al. (2008) emphasized the significance of the ro le of corporate social responsibility activities in affecting company-favori ng outcomes, such as greater benefits to stakeholders and better quality of relationships. Organizations and corporations give thei r opinions on various community issues, engage in public debate, and deal with the issues that affect the reputation of the organization (Luoma-aho & Vos, 2010). For example, an energy company is involved in the debate on a new energy resource and it s economic or environmental consequences. The companys contributions to the issue deb ate, that is, intera cting with government agencies, environmental advocacy groups, inve stors, and customers, will affect the reputation of the energy company. 38

PAGE 39

Due to communication technology and globalization trends, corporate communication has become predictable. Mult iple stakeholders are involved in the corporate communication about issues, and the st akeholders interaction is critical both online and in traditional media settings. Hence, the role of public relations and corporate communication professionals should be broa dening into relationship management or issue management (Luom a-aho & Vos, 2010). Agenda Building in the Business Context The significant role played by public relations message strategies has been emphasized as affecting the media and t he public agenda, not only in political communication settings but also in corporat e communication (Cameron, Sallot, & Curtin, 1997; Curtin, 1999; Gandy, 1982). A variety of information subsidies exist, and news releases are a public relations effort from organizations to journalists. Carroll and McCombs (2003) stated that t he salience transfer is the key idea of agenda setting, so the process and the effect of agenda setting can fit well into business news context. As discussed in the Chapter 2, two levels of agenda building and agenda setting occur. The second level of agenda building and agenda setting is associated with the salience of the attributes of a ce rtain object. At the first level, agenda-building and agenda-setting effects concer n the attention of the m edia and the public, while at the second level, agenda-building and agenda-setting effects concern the comprehension of the issue or the object by the media and the publ ic. Also, attribute salience has been described in terms of tw o aspects: substantive and evaluative (or cognitive and affective) (Behr & Iyengar, 1985; Iyengar & Kinder, 1987; Iyengar, Peters, & Kinder, 1982; Lee, 2010; McCombs & Shaw, 1993). In terms of the first level 39

PAGE 40

association, the salience of objects has been explored regarding political figures, institutions, corporations, and issues. While corporations can be associated wit h various attributes, little empirical research has been done to investigate t he relationship between the news media and public perceptions in corporate communicati on settings. Also, few studies have been conducted to establish the link between public relations and media coverage, and between public relations and public opini on grounded in the agenda-building theory. Public relations professionals provi de information subsidies, such as news releases, speeches, or press conferences, to shape media agendas and to affect public opinion (Curtin, 1999). These information subsidies have been commonly utilized by journalists to save cost to find informati on themselves (Miller, 2010). The influence of public relations on media content is critical in terms of public perception. Carroll and McCombs (2003) explained that much of the customers perception about the corporation is related to how the firm is described in the news media. The relationship between the firstand second-level of agenda building and agenda setting was also investigated propos ing a compelling arguments hypothesis (Figure 1-1). The compelling arguments hypothes is has been studied in an experimental design research to observe the causal association among variables in the agendabuilding and agenda-setting processes and effe cts. Kiousis (2005) stated that the compelling arguments hypothesis may be more appropriate for issues than political candidates, and Meijer and Klei nnijenhuis (2006) emphasized the effect of issues in business news on corporate reputation. Or ganizations and corporations deal with several issues, such as investment, profits, or service quality, as well as some societal 40

PAGE 41

issues, such as the environment, corrupti on, or energy. These issue-oriented media agendas affect how experts and others stake holders evaluate the co rporation, and the evaluation consequently affects their decisio ns about purchasing products or investing money. The effect of issue in business is also emphasized in the issue management domain of public relations (Botan & Taylor, 2004). Comparing data between media content analysis and a panel survey results, Meijer and Kleinnijenhuis ( 2006) provided empirical evidence on the second level of agenda-setting effects on public opinion. They observed issue ownership as the second level attribute of messages. In the busi ness arena, the effect on business has been explored in return on investm ent or reputation (e.g., Carr oll, 2004; Carroll & McCombs, 2003; Deephouse, 2000). Ohl et al. (1995), supporting se cond-level agenda building, found the effects of public relations effo rts in changing attribut e salience in media coverage. Zoch and Molleda (2006) also reviewed comm unication literature to examine the effects of organizations and t he media in building an agenda for publics. Focusing on the media relations function of public relations, Zoch and Molleda (2006) reviewed and interconnected concepts between framing, information subsidies, and the agendabuilding theory. Framing, as second-level age nda building, has functions to define social problems, to diagnose causes, to make judgments, and to suggest solutions. Practitioners try to fram e media, public, and policy agendas through information subsidies. Due to its impact on policy age nda, an organization or a corporation has been considered as an organizational polic y actor (Andsager & Smiley, 1998): Policy actors are entities such as government agencies, large corporations, elite professional organi zations and even citizen-activists 41

PAGE 42

who are outside the media but, because of their size and influence, also possess the ability to intervene in the production of news. Policy actors employ public information officers to communicate their frames. (Andsager & Smiley, 1998, p. 185) Lang and Lang (1981) suggested the agenda-building process as a reciprocal communication process through feedback betw een an organization and its publics. For example, not only political candidates but also voters can set issue agendas for a political campaign (Walters, Walters, & Gray, 1996). Interdepende ncy and reciprocity among diverse publics in the relationships em phasize the role of proactive information management, and practitioners prepare proac tive messages by scanning the environment and identifying various issues (Zoch & Molleda, 2006). The agenda-building hypothesis is also suggested in the international context. Several studies have focused on the role of international public rela tions in cultivating images of other nations (Kunczik, 1997, 2003; Zhang & Cameron, 2003). Exploring the relationships between international public relations messages, U. S. media coverage, and public opinion on other countries, Kiousis and Wu (2008) found that the increase in public relations counsel results in decreasin g bad news about those countries in the U.S. news media and in changing American public opinion. Zhang and Cameron (2003) also studied the impact of the Chin ese governments public relati ons campaign in U.S. media coverage about China. The goal of the camp aign was to present a new image of China to American publics through cultural events. Sino-U.S. relations have been shaped by public images and perceptions, and the images affect policy outcomes between the two nations (Isaacs, 1972). Zhang and Cameron (2003) showed that a similar pattern among major U.S. newspapers concerning cove rage of China created mostly negative news stories, and international public relations efforts had an effect in reducing negative 42

PAGE 43

coverage on China. Wilcox, Ault, and Agee (1989) stated that a national government tried to affect foreign publics perception toward the nation through lobbying and public information dissemination. This process is ca lled public diplomacy. According to the scholars, public diplomacy efforts have incorpor ated cultural aspects of foreign relations aimed at improving images of a country. K unczik (1997) also emphasized governments efforts to cultivate their national images for foreign publics. The Role of Issues in Business Corporations or business institutions have been considered as interest groups in the policy process (Berger, 2001), and they try to control issue agendas among the media, public, and policy through information subsidies (Schattschneider, 1960). The corporate agenda has been negle cted in the traditional agenda-setting process; however, scholars have increased their atte ntion to corporate agenda-building power by considering the sources influence on t he media (Berger, 2001). Gandy (1982) emphasized the role of information subsidies (e.g., news releases, spokespersons, satellite feeds, etc.) in affecting the m edia and policy agendas. Berger (2001) proposed an alternative agenda-setting model placing cor porate issues in t he center of the process. In a corporate model, corporate political actors play a powerful role in shaping federal policy agendas through lobbying, pub lic relations, advertising campaigns, or political debates (Domhoff, 1970; Domhoff, 1979; Gandy, 1982). Scholars have stated that agenda building can be successful when in formation subsidies are accepted by the media and policy professionals (Cameron et al., 1997). Highlighting the role of co rporation agenda building in th e public agenda, scholars have incorporated the idea of issues manag ement (Heath & Cousino, 1990). Renfro (1993) stated that corporations play a lar ger role in the public issues process than 43

PAGE 44

government (p. 2), and the role of business in public agenda building is enhanced when conflict issues are debated (Berger, 2001), and when the conflict is privatized (Schattschneider, 1960). Explaining framing of issues, Hallahan (1999) defined an issue as a dispute between two or more parties, us ually over the allocation of resources or the treatment or portrayal of groups in society (p. 217). I ssues are frequently constructed by interest groups who put the issues up for public discussion (Best, 1995). Through a process of agenda building, social problems (or issues) receive social acknowledgement, and public support for the i ssues is mobilized (Cobb & Elder, 1972). Miller (2010) suggested marketplace advocacy as a form of issue advocacy in the domain of an organizations image promot ion: Some researchers suggest that marketplace advocacy may be an effect ive means of both image building and influencing policy, because of its ability to persuade without seeming to do so (p. 85). The purpose of issue advocacy campaigns is to inform, educate, and persuade publics about how the business contributes to its society and community (Sethi, 1979), and through marketplace advocacy (a type of issue advocacy), business has been involved in promoting public issues such as health, energy, and environment. Miller (2010) described the benefits of marketplace advocacy for business as having both economic and psychological benefits (p. 89). Using the case study of the coal industry which has a great impact on the environment, Miller (2010) stated that marketplace advocacy results in greater approval for the advocated coal industry. Definition of issue-arena Emphasizing the role of the organization, Aula and Mantere (2008) viewed arenas in which an organizations reputation is created through interaction with the organizations publics. Even though the definit ion focused mainly on the role of 44

PAGE 45

organization, scholars have also stated that the role of the media cannot be ignored (e.g., Miel & Faris, 2008; Phillips & Y oung, 2009; Solis, 2009). Changes in media systems and technology affect the interaction between an organization and its public, and those changes affect the reputati on of the organization as well. Luoma-aho and Vos (2010) suggested severa l factors that influence the issuearena interaction: the number of actors involved, the amount of media visibility, and the intensity of public interest in the issue (e.g., Schattschneider, 1960). In a society, numerous issues and events occur, and it is impossible for all the issues to be debated simultaneously; hence, the issues compet e with each other to enter the media and public agenda (e.g., Berger, 2001; Berger, Hertog, & Park, 2002; Hilgartner & Bosk, 1988). This study aimed to observe the role of issue agenda in affecting public opinion on the issues and the organization that owned the issue. Previously, scholars argued that what the stakeholders perceived about an or ganization affects the images of the organization, and an organizations reputation is built on the assessments by different stakeholders (e.g., Bernstein, 1984; Bromley, 1993). The organization-centered mechanism has been shifted to the relationship standpoint emphasizing the role of stakeholders interacting in the issue-arena. As a response to shifts in the media landscape and business environment, issue management has bloomed since the 1970s, and scholars have focused more on reciprocal relationships than on a unidirectional one (Berger, 2001). Professionals also need to monitor media content and to follow dialogues between an organization and 45

PAGE 46

publics to better manage issues in which th ey are involved. Luoma-aho and Vos (2010) suggested the following axioms for fu ture directions in research: An issue-arena is a place where the publ ic debate about an issue is conducted. The interaction takes place in the traditional or virtual media. The actors can be active or passive, one could say on the stage or in the audience. The arena is dynamic; actors may be more or less active when time passes or leave entirely to go to another area. An organization can be active in multiple arenas, monitoring actions of other parties and/or actively engaging in the interaction. As the arenas are interrelated an organiza tion needs to coordinate and balance its communication strategies (Luom a-aho & Vos, 2010, p. 324). Issue Ownership In a business context, Meijer and Kleinnij enhuis (2006) emphasized the effects of issues on corporate reputation. The effect s of business news involving issues have been widely assumed, but little empirical resear ch has tested the effects. This research aimed to apply the concept of issue owner ship to the agenda-building process. The effect of issues has been widely studied in t he area of political communication to show how ideological and controversial issues affect the images and assessment of the involved political party. When the media emphasize the soci al welfare policy, publics would evaluate the conservative party more positively since they perceive that the issue is owned by the conservative party (Zaller, 1992). Scholars have noted that when the issue is primed in the public domain, issue-owned parties w ould receive more attention from the public and acquire electoral gains (e.g., Kleinnijenhuis et al., 2001; Petrocik, 1996; Petrocik et al., 2003; Sheafer & Weimann, 2005). 46

PAGE 47

Scholars examined the effects of news on pub lic opinion toward a political party and subsequent public behavior through election outcomes (Kleinnijenhuis et al., 2001). Issue news benefits the issueowned party when the media covers the issue in a positive light, while issue news benefits the opposite party (not the party owning the issue) when the media covers the issue in a negative light (Kleinnijenhuis et al., 2001). Petrocik (1996) stated that issue problems are recogniz ed by the news media through the framing and priming process, and the issues set the criter ia for voters when they evaluate a party. Issue-owned parties are selected based on the perceived competency of the party in dealing with certain problem issues (Petrocik, 1996). Petrocik et al. (2003) explored the priming effort s of political campaigns in presidential election cases. Some issues, such as social welfare, hav e been owned by Democrats; issues, such as the size of the government or taxes, have been owned by Republicans; and issues, such as the economy or security, have not been owned by a single party (Petrocik et al., 2003). Petrocik et al. (2003) noted that i ssue ownership of vague issues is based on how much credit or blame is given to each party. Sheafer and Weimann (2005) pointed out the e ffective role of the media in an Israeli election context by summarizing the agenda-setting and agenda-building processes as follows: Associations between reality and media attention, associations between the media and public opinion, linkages to voting behavior, and priming effects on actual voting results. The results indicate that the media help voters to set their evaluation criteria (Sheaf er & Weimann, 2005). According to an issue-ownership perspective, whether or not an organization owned an issue would determine that organizations influenc e on public perception. 47

PAGE 48

Budge and Farlie (1983) argued that a political party takes advantage when the election campaign promotes issues that the party has emphasized. The assumption of the effects is that voters would like to vote for a candidate and/or party when the candidate and/or party is perceived competent in hand ling certain salient issues (Belanger & Meguid, 2008; Cha, Song, & Kim, 2010; Elmelund-Praestekaer, 2011; Green & Hobolt, 2008). On the topic of corporate issue ownersh ip, Cha, Song, and Kim (2010) measured issue ownership by a corporations capabili ty to solve certain issues. Meijer and Kleinnijenhuis (2006) pointed out that when the issue in business news transfers its salience to the publics mind, the issue salience becomes a yardstick to evaluate the corporation, and salience affects the reputation of the corporation: If there is news about an issue that t he public perceives the organization to be handling successfully, that is, an owned issue, then the r eputation of the organization will improve. If there is news about an issue that the public regards the organization as being incapable of handling, that is, a poorly owned issue, then the reput ation of the organization will worsen. (Meijer & Kleinnijenhuis, 2006, p. 545) To explore the effects of issue ownership scholars have generated a cross-sectional study based on the agenda-setting hypothesis and have showed that the amount of news about business issues involved affects the salience of an issue and the reputation of the corporation (Meijer & Kleinnijenhuis, 2006). Scholars have suggested considering the issue ownership theory in te rms of the substantiv e dimension of the attribute level of agenda setting (Carroll & McCombs, 2003). Emotional Mediating Factor in the Agenda-Building Process This section provides definitions and measurement of the mediator in this study, emotional arousal, also known as a contingent condition for agenda-setting effects. Wu 48

PAGE 49

and Coleman (2009) found that negative info rmation has more power to transfer the medias agenda of candidate attributes to th e public (p. 775). Previous studies have showed that the agenda-build ing and agenda-setting effe cts would depend on the degree of some other variables, such as an i ssues obtrusiveness, si ze of effect, time, level of media exposure, or respondents educational level (e.g., Dearing & Rogers, 1996; Kim & McCombs, 2007; Wanta, 1997; Weaver, 1977; Winter & Eyal, 1981; Zucker, 1978; Yagade & Dozier, 1990). Yagade and Dozier (1990) explored agendasetting effects between two types of issues concrete (drug abuse, energy) and abstr act issues (nuclear arms race, federal budget deficit) and found that the effects are greater with concrete issues than with abstract issues. Kim and McCombs (2007) also showed that the media effects on public opinion are greater for heavy medi a users than light users. Among the various factors, the effects of affective and emotional factors have been less developed compared to those of cogniti ve factors. Hence, this study mainly focused on the effects of emotional arousal to explore the role of affective attributes in agenda-building research, and trust also plays a pivotal role in the business reputation study. Previous studies have suggested a critical ro le for affective attributes in leading emotional responses to a message and in shapi ng the overall composition of the story (e.g., Kiousis et al., 1999; Lopez-Escobar, Llamas, & McCombs, 1998a). Using the tone of a story, studies have demons trated how the tone affects the overall evaluations of the object (issues or political figures) in the messages. When a news story covers a political candidate positively, the audience would ev aluate the political ca ndidate favorably. 49

PAGE 50

Scholars have focused on the role of affe ct in cognitive and overall thinking about an object (e.g., Arnold, 1985; Batra, 1986; D illard & Wilson, 1993). Affect was defined as feeling toward a stimulus that leads to re lative preferences toward that stimulus out of a class of similar stimuli (Batra, 1986, p. 54). Scholars have found supportive evidence that explains the effect s of affective attributes on overall evaluation of objects (e.g., Kiousis et al., 1999; Lopez-Escobar et al., 1998b). Emotion is complex and dy namic in nature (e.g., Dunn & Schweitzer, 2005; Smith & Ellsworth, 1985). Compared to mood ( positive or negative feelings), Dunn and Schweitzer (2005) indicated that emotion has multiple appraisal cues. For example, negative feelings contain anger or fear, and positive feelings contain happiness or satisfaction. According to Dunn and Schweitz er (2005), these different appraisals are related to different subsequent behaviors or judgments: Anger is characterized by high ot her-person control, sadness by high situational control, and guilt by hi gh personal control. That is, when assessing a negative situation, people typically feel angry if they perceive another person to be responsible, sad if they perceive nonhuman factors (e.g., illness or natural disaster) to be responsible, and guilt if they perceive themselves to be responsible. (Dunn & Schweitzer, 2005, p. 738) Also, the attribution process of emotion is unclear, and there is always a chance to misattribute the causes of feeling (Dunn & Schweitzer, 2005). When multiple incidents (or people) are interrelated with someones angry feelings, he/she can blame a wrong incident (or person) for his/her angry feeling. Smith and Ellsworth (1985) proposed the following dimensions of appraisal: pleasantness, anticipated effort, ce rtainty, attentional activity, self-other responsibility/control, situational control, and emotion. These dimensions of appraisal vary depending on the types of appraisal. 50

PAGE 51

Scholars have traditionally suggested two dimensions of emotion pleasantness and arousal (e.g., Dunn & Schweitzer, 2005; Sm ith & Ellsworth, 1985). The valence of a message has been widely used to explore the role of affective attributes in agenda building and agenda setting; however, the role of affect still has been underdeveloped in agenda building and agenda setting. Scholars have stated that emotionally arousing messages can influence cognitive information processing by evoking rational thinking (e.g., Coleman & Wu, 2010; Keller & Block, 1996; Marcus et al., 2000). Lang et al. (1995) explored the effects of arousing messages on the overall effectiveness of advertising. Coleman and Wu (2010) also showed that emotional attri butes of a political candidate affect the overall impression of the candidates qualifications. Comparing the effects between affective and cognitive a ssessments, Coleman and Wu (2010) found that an audiences emotional assessment towa rd a political candidate has greater influence on the overall evaluat ion of the candidate than c ognitive assessment, such as a candidates qualifications. These studies have argued that emotions can evoke indepth and cognitive information processing about an object (Coleman & Wu, 2010; Marcus et al., 2000). Dunn and Schweitzer (2005) also found t hat emotions can enhance the level of cognitive thinking to ma ke a judgment toward an object. Sheafer (2007) also addressed the idea that the af fective priming effect should be explored further. Emphasizing theory-based tests to under stand why and when the agenda-setting effects occur, Miller (2007) suggested a mo re proximal factor of mediated agendasetting effects, relevance determined by a ffect, inference, and personal importance. Affect determines the relevance through the emotional reactions a news story arouses 51

PAGE 52

(Miller, 2007, p. 692). Miller ( 2007) stated that the valenc e of emotion (positive or negative) affects judgments toward objects, emotional arousal is generally related to the prominence judgments, and s pecific negative emotions mediate the effects: The most likely negative emotion to mediate agenda setting is anxiety or fear. Specifically, if a news story arouses anxiety or fear, people may pay more attention to the specific content of the story, cognitively elaborate on the information, and come to a conclusi on that the issue is an important one for the country. (M iller, 2007, p. 693) The affective attributes were measured mainly by two dimensions: valence (positive or negative) and arousal (stimulating or not) (e.g., Clore et al., 1994; Forgas, 1995; Gorn et al., 2001; Schachter & Singer, 1962). Compared to val ence, emotions are shorter in duration and more intense (Dunn & Schweitzer, 2005, p. 737). From a social psychology perspective, scholars have explai ned emotions as the level of arousal distinguished from valence (positive or negative) (e.g., Dunn & Schweitzer, 2005; Forgas, 1995; Hirt, Levine, McDonald, Melton, & Martin, 1997). Emotional arousal is distinguished from valence (m ood or tone), and its impact is greater than the valence on the cognitive information processing and overall evaluation of an object (Dunn & Schweitzer, 2005). In Millers (2007) study, emotion was measured by asking which participants felt angry, sad, proud, hopeful, happy, and afraid while reading the stimuli story. Furthermore, this study explores several emotional conditions such as anger, pride, guilt phase, grat itude, happiness, and sadness, as suggested by Dunn and Schweitzer (2005). For example, anger is characterized by high other-person control, sadness by high situational control, and guilt by high personal control (Dunn & Schweitzer, 2005, p. 8). Pride and guilt are self-oriented and happiness and sadness are situational judgments (Dunn & Schweitzer 2005). Emotions in each dimension were suggested as follows: 52

PAGE 53

Anger: angry, mad, irritated Gratitude: appreciative grateful, thankful Guilt: guilty, remorseful, sorry Pride: proud, self-fulfilled Happiness: joyful, happy, elated Sadness: gloomy, sad, upset (Dunn & Schweitzer, 2005) This study explored the effects of em otion in the agenda-bu ilding process in addition to the valence (tone) of a message. The influence of affective attributes on cognitive and overall evaluation of an object would expand the understanding of the compelling-arguments hypothesis Moreover, this study aimed to expand the knowledge in the corporate communication context. Morris et al. (2003) showed t hat evaluations of certain objects are always somewhat associated with affective evaluation, and their study explored the affective pr iming effects in political contexts. However, affective priming effects have not been fully explored, particularly in corporate communication. Trust as a Relationship Quality Outcome This section reviews the relationship persp ective of public relations. Particularly, based on a corporate communication perspective this study reviews the concept of trust, one of relationship quality outcome variables. Definition and Measurement of Trust Emphasizing the need for theor y development in public relations, Ferguson (1984) suggested a relationship between an organizati on and its public as the central unit of a public relations study. From a relationshi p perspective, Cutlip, Center, and Broom (1994) viewed public relations as the m anagement function that establishes and maintains mutually beneficial relationshi ps between an organization and the publics on whom its success or failure depends (p. 2). Scholars have stated that organizations should develop a long-term relationship with t heir public to achieve their organizational 53

PAGE 54

goals effectively (e.g., Dozier, L. Grunig, & J. Grunig, 1995; Hon, 1997; Huang, 1999). The organization-public relationship concept has been largely studied by scholars to define the characteristics of effective pub lic relations and to emphasize the managerial role of public relations (e.g., J. Grunig, 2006; L. Grunig, J. Grunig, & Dozier, 2002; J. Grunig & White, 1992; Ledingham & Bruning, 1998). Huang (19 98) defined the concept as the degree that the organi zation and its publics trust one another, agree on who has rightful power to influence, experience, experience satisfaction with each other, and commit oneself to one another (p. 12). Ki and Hon (2009) also stated that studi es of organization-public relationships include three stages: (a) antecedents of rela tionships, (b) relationship maintenance strategies, and (c) relationship quality out comes (p. 1). Relationship maintenance strategies mean any organizatio nal behavioral efforts that attempt to establish, cultivate, and sustain relationships with strategic public s (Ki & Hon, 2009, p. 5). Ki and Hon (2009) suggested six dim ensions of relationship outcomes: access, positivity, openness, sharing of tasks, networking, and assur ances. Scholars have poi nted out that these relationship cultivation strategies can be us ed to make better re lationship outcomes. Antecedents of relationships mean collective social norms or environmental factors that affect relationship building, and consequences of the rela tionship mean outcomes of relationship such as trust (e.g., Hon & J. Grunig, 1999; Ki & Hon, 2009). In measuring the effectiveness of public relations efforts, scholars have suggest ed relationship quality outcomes including control mutu ality, satisfaction, trust, and commitment (e.g., Hon & J. Grunig, 1999; Yang & J. Grunig, 2005). 54

PAGE 55

Ledingham and Bruning (1998, 2000a, 2000b ) suggested the concept of an organization-public relationship operationalizing the concept with five dimensions including trust, openness, involvement, invest ment, and commitment. Hon and J. Grunig (1999) suggested a multiple-item scale for measuring relationship quality outcomes: control mutuality, satisfaction, and commitmen t. Trust is one of the relationship quality outcomes that act as a predict or of public attitudes toward an organization, and scholars have suggested a multidimensio nal construct to measure the trust concept: integrity, dependability, and competence (Hon & J. Grunig, 1999). Thes e constructs refer to whether or not an organization is fair and just whether or not an organization will do what it says it will do, and whether an organizati on has the ability to do what it says it will do. As Yang and J. Grunig (2005) menti oned, trust (relationship quality outcomes) can influence subsequent judgments of an organization such as organizational reputation. Corporate Communication and Trust Seltzer and Zhang (2011) applied the organi zation-public relations framework to political communication and stated that polit ical parties could get favorable public attitudes from trust, which is one of the re lationship quality outcomes. Trust also can play an essential role in the relations hip-building process for other types of organizations, such as business firms. Building a good relationship with publics has become one of the most im portant goals of an organization due to its positive consequences (Swift, 2001; Yang & J. Grunig, 2005). Swift (2001) considered trust as the main domain of business: Central to the notion of accountabi lity and to contemporary social accounting practice is the concept of trust. Accountabilit y is based upon a distrust of corporate management, whereas corporate reput ation building is 55

PAGE 56

about strategically seeking to establish trust in stakeholder relationships in order to negate formal accountability requirements. (Swift, 2001, p. 16) From an organization-public relationship perspective, scholars have suggested antecedents and consequences of relations hip (McKnight & Choudhury, 2001; McKnight, Choudhury, & Kacmar, 2002a, 2002b). The communication process is the factor that links relationship dynamics. Walton (1969) stated that the dynamics of the organization can best be understood by underst anding its systems of communication (p. 109). Antecedents of relationships are the fa ctors that cause relationship formation, including public perceptions, motives, needs, and behaviors, and consequences are relationship outcomes, such as goal achiev ement (Cutlip et al., 1994). J. Grunig and Huang (2000) also explained the antecedents and consequences of organization-public relationships, and emphasized communication strategies that link the relationship blocks. According to scholars, social and cultural norms, collective perceptions and expectations, need for resources, and legal/ voluntary necessity were suggested as antecedents of the relationship; goal ac hievement, dependency/loss of autonomy, and routine and institut ionalized behavior were suggested as consequences of the relationship. The concept of relationships wa s explained as the pro perties of exchanges and communication activities (J. Grunig & Huang, 2000). As one of the relationship quality outcomes, this study expl ored trust as a critical factor to evaluate public relations communication strategies (e.g., King-Casa, Tomlin, Anen, Camerer, Quartz, & Montague, 2005; Swift, 2001; Y ang & J. Grunig, 2005). 56

PAGE 57

Corporate Reputation This section provides a definition of corporate reputation, presents the measurement of corporate r eputation, and explores the relationship between message exposure and corporate reputation. Definition of Cor porate Reputation Corporate image has been a domain of pub lic relations research. Image is something built in peoples mind and public relations professionals aim to improve their organizations image by sending out communication messages to the public (Olasky, 1987). J. Grunig (2003) cr iticized use of image in public relations due to its ambiguity and negative connotation. People use image inte rchangeably with the related terms of message, reputation, perception, or attitude, and people consider that the word image means something is manipulated, that is making a good corporate image could mean deceiving the public about the reality of the corporation (J. Grunig, 2003). Also, corporate image refers to only superficial sym bolic meaning of corporation based on the identity built through logos, trademark, or name of the corpor ation (Olins, 1978; Goffman, 1959). Impression m anagement or symbolic communication activities (e.g., advertising) have been emphasized as the role of public relations practitioners (J. Grunig, 2003). However, behavioral relationship has been getting more attention since scholars have focused on long-term relationship build ing (Ferguson, 1984; J. Grunig & White, 1992). J. Grunig (2003) indicat ed that cognitions attributed to an object are accumulated in peoples mind to create a long-term memory, and that me mory is used when people think (evaluate) about the object. Corporate reputation is an example of this long-term schema that represents behavioral relationships (J. Grunig, 2003). 57

PAGE 58

Kiousis and Wu (2008) stated that public relations scholars have frequently looked at agenda-building and a genda-setting within the domains of corporate imagebuilding or social issue promotion (p. 70). Wartick (1992) def ined the corporate reputation as the aggregation of a single stakeholders perceptions of how well organizational responses are meeting the demands and expectations of many organizational stakeholders (p 34). A stakeholder is defined as any group or individual who can affect or is affected by the achi evement of the organizations objectives (Freeman, 1984, p. 64, cited in Wartick, 1992). Milgrom (1981) showed that corporate reputation is related to favorable public asse ssments; that is, a corporation has a better reputation when the corporation is viewed mo re favorably than other competitors. Corporate reputation is the publics perception about the corporations effectiveness, and multiple public (stakeholde rs) are involved in the process (Wartick, 1992; Winfrey, 1989). Multiple pu blic or stakeholders exist in the relationship between an organization and the public, and they would have a different perception about the corporate reputation. The definition of a stakeholder woul d affect the evaluation of corporate reputation. Gotsi and Wilson (2001) defined corpor ate reputation as follows: A stakeholders overall eval uation of a company over time. This evaluation is based on the stakeholders direct experiences with the company, any other form of communication and symbolism that provides information about the firms actions and/or a com parison with the actions of other leading rivals. (Gotsi & Wilson, 2001, p. 29) Summarizing common elements of corporat e reputation, Gotsi and Wilson (2001) explained the characteristics of the concept as follows: 1) the concept is dynamic, 2) it needs time to build, 3) corporate reputation and corporate image are related, and 4) a company can have different reputations held by multiple stakeholders (e.g., Fombrun, 1996; Gray & Balmer, 1998). 58

PAGE 59

Several key factors affect corporate reput ation: company size, product quality, or prices. As agenda-setting and agenda-buildi ng scholars have emphasized, media exposure also plays a critical role in bui lding corporate reputat ion. Focusing on the business-media interaction, scholars have shown a correlation between the media and corporate reputation (Wartick, 1992). Wartick (1992) show ed that the tone of media exposure is associated with the change in corporate reputation. Measurement of Corporate Reputation Hutton et al. (2001) suggested reputation management as a driving philosophy of corporate public relations. Co rporate reputation or image is an intangible concept that cannot be managed directly; however, scholars hav e stated that it can direct publics attitude and behavior toward a corporation. M easuring corporate reputation is not an easy process, and the factors that go into developing corporate reputation are still not understood fully. Hutton et al. (2 001) indicated that corporate reputation is a meaningful concept only as it applies to specific audien ces or publics, and t hat no across-the-board measure of reputation is or can be valid for all stakeholders (p. 249). In other words, reputation can vary by who is doing the measuring (e.g., customers, donors, or employees). In the Harris/Impulse study budget spending and co rporate reputation indicated a positive relationship; while in Fortunes Most Admired Company study, a relationship was found between company size and reputation. In regard to the types of communication, scholars have pointed out that proactive communication has a more positive relationship with reputation than reactive communication (Hu tton et al., 2001). Kiousis et al., (2007) defined corporat e reputation as a multidimensional construct with eight variables including financial performance. Adopted as a multidimensional item from the Reputati on Quotient by Harris Interactive and the 59

PAGE 60

Reputation Institute, six a ttributes and 20 sub-attribute items were suggested to measure stakeholders perceptio ns about corporate reputation (Kiousis et al., 2007): Vision and leadership: market opportuniti es, company has excellent leadership, and company/management has clear vision for the future; Social responsibility: company supports good causes, company is environmentally responsible, and company is responsible in the community; Emotional appeal: feel good about company, company inspires admiration and respect, and company inspires trust; Products and services: high quality products and/or services, innovative products and/or services, company provides good value for money, and company stands behind its products and/or services; Workplace environment: rewards employees fairly, good place to work, and good employees; and Financial performance: outperforms compet itors, company has record of bring profitable, company is a low risk investment with growth prospects (Kiousis et al., 2007, p. 155) Relationship between Message Expos ure and Corporate Reputation From a long-term relationship-building persp ective, corporations are involved in several social issues (global warming, health, or human rights) through socially responsible activities. Corporate reactions towa rd the social problems or issues affect the publics perception about t he corporation (Botan & Taylor 2004; Miller, 2010). The corporate social responsibility actions are one of the dimensions of corporate reputation measurements (Hutton et al., 2001). Bhattacharya et al. (2008) suggested the influence of corporate social responsibility activities on relationship quality outcomes. Issues have become one of the most prom inent concerns for corporations, and scholars have linked agenda-building and agenda-setting concepts to an issue management perspective (e.g., Heath & Cousino, 1990; Miller, 2010). Miller (2010) found that public relations practitioners can play a significant role as an issue advocacy agent who informs and 60

PAGE 61

persuades the public about certain social issues. Successful issue advocacy activities can benefit organizations with better relationship outcomes, such as approval rating toward the industry (Miller, 2010). From agenda-building and agenda-setting pers pectives, the role of media exposure is critical for corporate reput ation. Media exposure was defined as the aggregated news reports relating to a specif ic company within a prescribed period (Wartick, 1992, p. 34). Wart ick (1992) stated t hat media exposure of organizations activities is a critical factor in affecting corporate reputation. A case study showed that the magnitude of negative media ex posure of four companies re sulted in the decline of the corporations reputat ion (Weinberger & Romeo, 1989). Eyestone (1978) emphasized the role of mass media in devel oping social movements and issues and in shifting corporate reputation. Kiousis et al. (2007) expl ored the influence of public relations and the media on corporate reputat ion, as suggested as an outcome of public relations. However, empirical research between media exposure and corporate reputation is still limited. Fombrun and Shanley (1990) found that more media exposure had negative effects on corpor ate reputation no matter how the media portrayed the corporation (negatively or posit ively). Griffin, Babin, and A ttaway (1991) noted that the relationship between media exposure and cor porate reputation is mediated by other factors, such as source credibility, history, and other si tuational dynamics. On an international level, corporate r eputation was not thor oughly explored in an agenda-building and agenda-setting framework. However, some studies have been conducted in terms of nationa l reputation context (e.g., Kiousis & Wu, 2008; Zhang & Cameron, 2003). Analyzing U.S. media coverage before and after a Chinese 61

PAGE 62

governments public relations campaign, sch olars found significant changes in negative news coverage after the nation image camp aigning (Zhang & Cameron, 2003). Also, Kiousis and Wu (2008) found similar results s howing that the increase of public relations counsel can reduce the number of negative fore ign news coverage of a particular nation. The role of media exposure has been emphas ized since media portrayals of an object (nation-state or organizations) can affect how the public evaluates the object. Previous study results have also indicated that public relations efforts can influence the tone of media coverage (e.g., Kiousis & Wu, 2008; Miller, 2010; Zhang & Cameron, 2003). Hence, the role of public relations profession als through media relations is critical to making changes in corporate reputation or national reputation (e.g., Carroll, 2004; Carroll & McCombs, 2003; Kiousis & Wu, 2008; Zhang & Cameron, 2003). Hypotheses and Research Questions This study predicted the linkages among public relations messages, agendabuilding effects, and corporate reputation. Emotional arousal is tested as a mediator to link public relations and the salience of the agenda-building proc ess. The mediating effect refers to the effect of a third variable/construct intervening between two other related constructs (Hair, Black, Babin, Anderson, & Tatham, 2006, p. 844), and the addition of a third variable to this X -> Y relation, whereby X causes the mediator, M, and M causes Y, so X -> M -> Y (Lavigne & Bourbonnai s, 2010, p. 125). This study attempted to examine: 1) the relationship between the independent variables (issue ownership and issue attributes) and the mediator (emotional arousal), 2) the relationship between the mediat or (emotional arousal) and the dependent variables associated with agenda salience (issue and attribute salience), 3) the relationship between the independent variable s (issue ownership and issue attributes) 62

PAGE 63

and the dependent variables associated with agenda salience (issue and attribute salience), and 4) the relationship betw een the dependent variables associated with agenda salience (issue and attribute salience) and ones associated with public opinion (priming effect, trust, and co rporate reputation). Also, this study presented the direct effect between independent variables and agenda-building salience, while controlling for the indirect, mediated effect. In the relationship between the independent variables (issue ownership and issue attributes) and the mediator (emotional arousal), the follo wing hypotheses and research questions were proposed: H1 : When the issue is owned by the corporation in public relations messages, the participants will have higher emotional arousal. H2 : The negative tone of issue attributes in a public relations message will be positively related to higher levels of emoti onal arousal of partici pants than the positive tone of issue attributes. RQ1 : Does an interaction exist between the perceived issue ownership and the tone of issue attribute in a public relations message on the level of emotional arousal of participants? Also, in the relationship between the mediator (emotional arousal) and the dependent variables associated with agenda salience (issue and attribute salience), the following hypotheses and research questions were proposed: H3 : The higher the level of emotional arousal that participants have, the higher the level of issue salience tr ansfer of participants. H4 : The higher the level of emotional arousal that participants have, the higher the level of attribute salience transfer of participants. Then, in the relationship between the independent variables (issue ownership and issue attributes) and the dependent variables associated with agenda salience (issue 63

PAGE 64

and attribute salience), the following hy potheses and research questions were proposed: H5 : When the issue is owned by the corporation in public relations messages, the participants will have a higher issue salience transfer. H6 : When the issue is owned by the corporation in public relations messages, the participants will have a higher attribute salience transfer. H7 : The negative tone of issue attributes in a public relations message will be positively related to the higher levels of i ssue salience transfer of participants than the positive tone of issue attributes. H8 : The negative tone of issue attributes in a public relations message will be related more to the higher level of attribute salience transfer of participants than the positive tone of issue attributes. RQ2a: Does an interaction exist between the perceived issue ownership and the tone of issue attribute in a public relati ons message on the level of issue salience transfer of participants? RQ2b: Does an interaction exist between the perceived issue ownership and the tone of issue attribute in a public relations message on the level of attribute salience transfer of participants? The relationships from the previous set of hypotheses are also explored for the direct effect while controlling for the indire ct, mediated effect. With the first mediator, that is, the level of emoti onal arousal, the following four hypotheses are proposed: H9 : The level of emotional arousal of participants mediates the relationship between issue ownership perception and the level of issue salience transfer. H10 : The level of emotional arousal of participants mediates the relationship between issue ownership perception and the level of attribute salience transfer. H11 : The level of emotional arousal of participants mediates the relationship between the tone of issue attributes in a publ ic relations message and the level of issue salience transfer. H12 : The level of emotional arousal of participants mediates the positive relationship between the tone of issue attri butes in a public relations message and the level of attribute salience transfer. RQ3 : How do the mediating effects of em otion differ, dependi ng on the mode of emotions (anger, gratitude, guil t, pride, happiness, and sadness)? 64

PAGE 65

In the relationship between the dependent variables associated with agenda salience (issue and attribute salience) and ones associated with public opinion (priming effect, trust, and corporate reputation), the following si x hypotheses and research questions are proposed: H13 : The higher the level of the issue salie nce transfer of parti cipants, the higher the level of priming effect participants will have about the corporation in its public relations messages. H14 : The higher the level of the issue salie nce transfer of parti cipants, the higher the level of perceived trust participants will have about the corporat ion in its public relations messages. H15 : The higher the level of the issue salie nce transfer of participants, the more participants will have a favorable perceived cor porate reputation about the corporation from its public relations messages. H16 : The higher the level of the attribute salience transfer of participants, the higher the level of priming effect participant s will have about the corporation in its public relations messages. H17 : The higher the level of the attribute salience transfer of participants, the higher the level of perceived trust partici pants will have about the corporation in its public relations messages. H18 : The higher the level of the attribute salience transfer of participants, the more participants will have a favorable perceived cor porate reputation about the corporation from its public relations messages. RQ4a: Does an interaction effect exist bet ween the levels of issue and attribute salience transfer of participants on the priming effect of the corporation? RQ4b: Does an interaction exist between the levels of issue and attribute salience transfer of participants on the perce ived trust of the organization? RQ4c: Does an interaction exist between the levels of issue and attribute salience transfer of participants on the per ceived corporate reputation? Finally, the last two hypotheses propose the relationship between the priming effect, trust, and corporate reputation: H19 : The higher the level of priming effect that participants hav e, the higher the level will be of the publics perceived trust of the corporat ion in its public relations messages. 65

PAGE 66

H20 : The higher the level of the publics perceived trust on the corporation that participants have, the more favorable perce ived corporate reputat ion participants will have. This chapter has presented a review of literature about the agenda-building theory, corporate communication and issue ownersh ip, trust, and corporate reputation. Hypotheses and research questions were derived from theoretical frameworks discussed in this chapter. A graphical representation of the basic conceptual framework is presented in Figure 2-1, and a more det ailed conceptual mode l representing the relationships between each of the variables is shown in Figure 2-2. Furthermore, Figure 2-3 shows the rest of the theoretical relati onships among variables not tested in this study: 1) the direct relationships between p ublic relations (issue ownership and issue attributes) and public opinion (pri ming effect, trust, and reputation), 2) the relationships between emotion and public opinion (priming effect, trust, and reputation), 3) the relationship between issue salience and attribute salience, and 4) the relationship between dependent variables (pri ming effect and reputation). 66

PAGE 67

Figure 2-1. A theoretical framework: Explaining the relationship between public relations, emotional mediator, agenda sa lience, and public opinion. A: a relationship between the independent variables (issue ownership and issue attributes) and the mediator (emotion) B: a relationship between the mediator (emotion) and the dependent variables associated with agenda salience (issue and attribute salience). C: a relationship between the independent variables (issue ownership and issue attributes) and the dependent variables associated with agenda salience (issue and a ttribute salience). C: the direct effect while controlling for the indire ct, mediated effect. D: a relationship between the dependent variables associated with agenda salience (issue and attribute salience) and ones associat ed with public opinion (priming effect, trust, and corporate reputation). 67

PAGE 68

Figure 2-2. Proposed compr ehensive conceptual model 68

PAGE 69

69 Figure 2-3. Untested relationships among variables

PAGE 70

CHAPTER 3 METHODS This chapter describes the method employ ed in this research. It presents the choice of method, descriptions of t he participants, manipul ation, questionnaire construction, experimental procedure, pretest, and statistical analyses. Experimental Design Several methods were used to investigate the use, influence, and effects of public relations (Cornelissen, 2000). Particularly, Cornelissen (2000) explained the survey, discourse analysis, and experimental design methods. The survey (including interviews) was used to identify information that respondents received and to explore the factors used in information processing. The disc ourse analysis (also known as social framework analysis) was used to explore specific terms and vocabulary used by professionals and publics. Cornelissen (2000) also de fined experimental designs: Where use is assessed through the control of extraneous influences and the purposive manipulation of tr eatments to groups of respondents (p. 323). This study aimed to test the influence of issue ownership and attribute of public relations messages on public opinion regardi ng corporate trust and reputation through the agenda-building process. Th is study proposed relationships among eight variables (independent variables, a mediator, and dependent variables). Quantitative experimental research was conducted to establish whether a relationship was found among these eight variables and also to det ermine the direction of the effect. Corporate reputation is affected by media exposure (Griffin et al., 1991; Wartick, 1992; Weinberger & Romeo, 1989). Expanding the linkages in the social media context, this research attempted to establish that t he related variables in agenda-building affect 70

PAGE 71

corporate reputation as the consequences of the agenda-building proce ss. In particular, this study conducted a 2 x 2 between-subjects factorial design experiment, manipulating public relations messages in terms of whether issues were owned by a corporation, as well as how positively or negatively the i ssue was exposed by the media and the public. This research followed the quantitative and causal research design tradition. Traditionally, scholars have st ated that agenda-setting effects came from accessibility in memory (e.g., Iyengar & Kinder, 1987; Krosnick & Kinder, 1990). However, prior evidence was not enough to demonstrate a c ausal relationship in the agenda-setting process (Miller, 2007). Zikmund (2003) defined causal research as research to identify cause-andeffect relationships amongst vari ables (p. 56). In a causal nature, this study attempted to determine the relationships among attri butes of public relations messages, salience of agenda, and public opinion about the issue and corporate reputation. This study explored the effect of each i ndependent variable, as well as the interaction between the variables and mediating effect s with the emotional arousal va riable. The study included the interaction effect between two independent variables that have different levels, using a factorial causal design. A fact orial design allows for the simultaneous manipulation of two or more independent variables at various levels (Zikmund, 2003, p. 283). Since the study is a 2 x 2 factorial design, it needs four groups that have at least 24 participants in each group. The sample si ze was determined by the combination of an effect size and power when the significant alpha level was set to .05. Due to the exploratory nature of this study, the medium effect size and power were used. The 71

PAGE 72

effect size was set to .06, and the power was set to .60 (Erdfe lder, Faul, & Buchner, 1996, cited in Keppel & Wickens, 2004, p. 173). Each group was assigned to a particular combination of treat ment conditions, which allow ed this study to test the interaction effect. The combinations of each condition and each group design are shown in Tables 3-1 and 3-2, respectively. Participants Participants in this study consisted of adults (18 years old or olde r), and the unit of analysis was therefore the individual. The unit of analysis was defined as what or whom is being studied (Babbie, 2004, p. 94) In exploring the relationship between public relations effects and co rporate reputation, Wartick (1992) noted that researchers should be careful about who eval uates corporate reputation. Corporate reputation was defined as a perception by some individual or group on the effectiveness of the corporation (Winfr ey, 1989, p. 3), and Wart ick (1992) explained that different stakeholders would have va rious corporate reputation perceptions. According to Wartick (1992), corporate reputation will be predict ed on the single stakeholders view of company responses to their own expectations as well as the expectations of stakeholders other than itself (p. 35). This study recruited participants from a crowd-sourcing web service called Mechanical Turk. The subjects of the serv ice were considered representative of the U.S. population (Buhrmester, Kwang, & Gosli ng, 2011; Paolacci, Chandler, & Ipeirotis, 2010). The reward was usually less than $1 and the typical hourly wage was approximately $1.40 (Horton & Ch ilton, in press). The monetary reward for participants (15-20 minute experimental st udy) was $.20 per person. 72

PAGE 73

Procedure Public relations materials were desi gned to be associated with a fictitious international/global corpor ation involved in a global heal th issue. With the two independent variables (issue ownership and affective attribute of a public relations message), four copies of an online survey questionnaire were prepared. Participants were randomly assigned to one of the four conditions, and each participant read a corporate message (e.g., history and public relations campaign messages) along with news messages distributed by the corporati on. After participants read the manipulated materials, they viewed the list of questions of dependent and mediating variable items. Pretest and Manipulation Check This section describes the procedure of pretests and a description of the manipulation checks of the two independent variables, with discussion about how these variables were operationally defined. Pretests Procedure To observe that the two independent variables were manipulated successfully, a separate pretest was conducted. Participants were recruited and then assigned to four different cells. After confirming consent information, each participant reads the corporate public relations messages regardi ng an issue-related campaign along with redirected news stories about a global health issue. With r egard to issue ownership, the study included owned and non-owned conditions, while positive and negative conditions were included with regard to the issue tone. Participants were randomly assigned to one of the four conditions. After reading t he stimuli messages, participants answered a questionnaire about their perception about iss ue ownership status and the valence of the issue. 73

PAGE 74

Stimuli 1: Issue Ownership The operational definit ion of issue ownership received more attention from a medium associated with an issue in the context of corporate public relations messages (Meijer & Kleinnijenhuis, 2006). The issue ow nership hypothesis proposed that the more an organization received media attention, the higher the likelihood t hat the issue would became salient with an issue associated with an organization. Scholars have measured issue ownership by asking how well people think a political party (or an organization) handles a s pecific issue (e.g., Elmelund-Praestekaer, 2011; Green & Hobolt, 2008). Belanger and Meguid (2008) also asked which political party would be the best at dealing with cert ain issues. Elmelund-Praestekaer (2011) used a competence index ranging from -100 to +100. Cha et al. (2010) measured the degree of capability to solve certain probl ems or issues. Adopted from previous scholarship (e.g., Elmelund-Praestekaer, 2011; Green & Hobolt, 2008; McKnight et al., 2002a), the level of perceived issue ownersh ip was measured by three seven-point Likert scale items: How well do you think the organization handles global health issues? How well do you think the organization can solve problems related to global health issues? and I believe that the corporation is very knowledgeable about global health issues. Stimuli 2: Tone of Issue Attributes A seven-point semantic scale was used to measure the affectiv e attribute of the public relations messages. The tone of an issue in the public relations messages was measured by whether the tone was positiv e or negative and whether the tone was 74

PAGE 75

pleasant or unpleasant (e.g., Dunn & Schweitz er, 2005; Miller, 1997; Smith & Ellsworth, 1985). Pretest Results A pretest was conducted to ensure that the stimuli messages were manipulated accordingly. Using the Mechanical Turk service, 31 people participated in the pretest. The average duration of a complete survey was approximately 13 minutes ( M = 13.00, SD = 10.85), and answers completed in less than three minutes were excluded. One answer was excluded in this process. Among 30 participants, 17 people randomly viewed an issue owned message while 13 viewed a non-owned issue ownership message. Also, 15 of the total participant s randomly read a negative tone from the message while 15 read a positive tone from the message. Cronbachs alpha was calculated to examine the reliability of each index score of variables. The issue ownership index wa s the sum of thr ee items (Cronbachs = 88) and the tone of the issue index was the sum of two items (Cronbachs = .89). The independent sample t-test was run to che ck the manipulation conditions of the two variables. The t-test revealed a statisti cally significant di fference among the mean scores of owned ( M = 16.00, SD = 2.42) and non-owned ( M = 11.54, SD = 4.61) issue ownership conditions ( t (17.035) = 3.169, p < .05). Also, a statistically significant difference occurred between the mean scores of negative ( M = 5.60, SD = 2.44) and positive ( M = 11.93, SD = 1.58) tone of issue conditions ( t (28) = 8.430, p < .001). Questionnaire Construction and Measures The questionnaire for this study consisted of scales for dependent variables (issue salience, attribute salience, priming, trust, and corporate reputation), mediating factors (emotional arousal), demographic items, and open-ended questions. Demographic data 75

PAGE 76

included age, religion, education, gender, ra ce, country of origin, occupation, and income (Balmas & Sheafer, 2010; Kiousis & McDevitt, 2008). These socioeconomic items were used in previous agenda-setti ng and agenda-building studies. Every scale was revised by replacing the subject in the original statement wit h the corporation in the message. Individual items for six variabl es of the proposed model are presented in the following sections (Table 3-3), and the c opies of the consent form and questionnaire are included in Appendix C. Issue Salience A measure of issue salience transfer is the participants beliefs about the importance of the issue as presented in the message. Iyengar and Kinder (1987) measured the salience of global health issues: In your opinion, how important are global health issues today? How many news reports about global health issues do you regularly pay attention to? To what extent do you think global heal th issues are deserving of additional government action? and How often do you talk about global health issues in your everyday conversation? Adopted from Iyengar and Kinder (1987), an overall salience was measured to determine whether or not t he participants considered th e issue on the message as prominent, significant, impor tant, and well known. Cronbachs alpha for issue salience items was .71. According to Langdridge (2004), Cronbachs alpha coefficient is used to assess the internal reliability of items with scaled responses (e.g., strongly agree to strongly disagree) (p. 77), and .70 or higher is considered as an acceptable level. The scales were summed up to form an 76

PAGE 77

index score for issue import ance (first-level agenda-settin g effects). The response choices consisted of Likert scales ranging from 1 = Not at all to 7 = Extremely. Attribute Salience To measure attribute salienc e transfer, the participants beliefs about the valence of the issue were presented in the message. Two seven-point semantic Likert scale items were used to measure t he valence of issue attribute (positive and pleasant to negative and unpleasant). Cronbachs alpha fo r the two items was .86. The response choices consisted of Likert scales rangi ng from 1 = Negative/unpleasant to 7 = Positive/pleasant. Emotion The level of emotional arousal after reading a message was associated with how such messages provoked emotional appeal. To calibrate the corporations emotional appeal, a Likert scale item was used to ask pa rticipants to identify how much they would agree with the statement, I f eel excited. The response choices consisted of Likert scales ranging from 1 = Not at all to 7 = Extremely. Dunn and Schweitzer (2005) proposed several emotional conditions such as anger, pride, guilty, gratitude, happiness, and sadness. Anger has been characterized by high other-person control, sadness by high situati onal control, and guilt by high personal control (Dunn & Schweitzer, 2005, p. 8). Pride and guilt are more self-oriented, and happiness and sadness are more situationa l judgments (Dunn & Schweitzer, 2005). Emotions were also measured as follows : anger (angry, mad, irritated), gratitude (appreciative, grateful, thankfu l), guilt (guilty, remorseful sorry), pride (proud, selffulfilled), happiness (joyful, happy, elated) and sadness (gloomy, sad, upset). Cronbachs alpha for emotional items was .90 (anger: .93; gratitude: .95; guilt: .86; 77

PAGE 78

pride: .81; happiness: .92; and sadness: .93). The response choices consisted of Likert scales ranging from 1 = Not at all to 7 = Extremely. Priming Effect Priming effects were measured using re sponses from the following question: Do you support or oppose the efforts to solve gl obal health issues? (e.g., Lee, 2010). A seven-point Likert scale, anchored by ext remely oppose and extremely support, asked the subjects whether they personally supported or opposed the global issue shown in the message. Trust The perceived level of trust is a multid imensional concept (e.g., whether the organization is fair or whether it will do what it says) was measured based on previous research (e.g., Hon & J. Grunig, 1999; McKnight et al., 2002a). The following statements were used to measure perceived trust (Hon & J. Gr unig, 1999, pp. 28-29): This corporation would treat customers fairly; Whenever this corporation makes an im portant decision, I know it will be concerned about its customers; This corporation can be relied on to keep its promises; I believe that this corporation takes t he opinions of customers into account when making decisions; I feel very confident about this corporation; This corporation has the ability to accomplish what it says it will do; I believe sound principles gui de the corporations behavior; I am very willing to let this corporat ion make decisions for customers like me; and I believe the corporation does not mislead customers. 78

PAGE 79

The responses from the nine seven-point Likert scale items were then summed to measure perceived trust in the organization, where higher scores indicated higher trust. Cronbachs alpha for trust items was .93. T he response choices consisted of Likert scales ranging from 1 = Strongly di sagree to 7 = Strongly agree. Corporate Reputation To measure corporate reputation, Fortunes most admired corporations was widely used by many businesses and society scholars (Fombrun & Shanley, 1990; Wartick, 1992; Winfrey, 1989). Wartick (1992) defined the concept as the aggregation of a single stakeholders perceptions of how well organizational responses are meeting the demands and expectations of many or ganizational stakeholders (p. 37). Operationally, this study adopted a multidimensional item fr om the Reputation Quotient by the Harris Intera ctive and the Reputation Instit ute to measure stakeholders perceptions about corporate reputation using six attributes and 20 sub-attribute items (Kiousis et al., 2007, p. 155): Vision and leadership: market opportuniti es, company has excellent leadership, and company/management has a cl ear vision for the future; Social responsibility: company supports good causes, company is environmentally responsible, and company is responsible in the community; Emotional appeal: feel good about company, company inspires admiration and respect, and company inspires trust; Products and services: high quality products and/or services, innovative products and/or services, company provides good value for money, and company stands behind its products and/or services; Workplace environment: rewards employees fairly, good place to work, and good employees; and Financial performance: outperforms compet itors, company has record of being profitable, company is a low risk investment with growth prospects. 79

PAGE 80

Cronbachs alpha for corporate reputation item s was .95 (vision and leadership: .84; social responsibility: .85; emotional appeal: .91; products and services: .86; workplace environment: .85; and financial performance: .81), and the scales were summed up to form an index score for reputation. The res ponse choices consisted of Likert scales ranging from 1 = Strongly disagr ee to 7 = Strongly agree. Methodological Concerns As Golan and Wanta (2001) mentioned, exposure of media content is essential in the agenda-building and agenda-setting processes, with no guarantee that participants have actually seen and read t he manipulation materials. Al so, an unequal amount of message exposure existed among participants. Since this study used an online survey panel, less control occurred to force them to read stimuli messages and carefully answer the questionnaire. To reduce this conc ern, the consent form clearly indicated that each participant should read the public relations messages carefully before answering the questionnaires. Participants were exposed only to a few messages from the corporation, perhaps not enough to evaluate corporate reputation. Moreover, some in trinsic or prior interests in the issue or participants ordinary leve l of issue salience may have influenced the agenda-building process and effects. Statistical Analysis To test the 20 hypotheses and seven res earch questions, various data analyses were undertaken, as described below. A two-way analysis of variance (ANOVA) wa s used to test both interaction and the main effects of the two independent variables on each dependent variable. In the first part, issue and attribute salience were c onsidered as dependent variables; in the 80

PAGE 81

second part, priming, trust, and corporat e reputation were the dependent variables (Figure 2-2). To establish the relationship between these dependent variables (e.g., relationship between agenda salience and primi ng effect), a correlation and regression analysis was run. A path analysis was administrated using LISREL 8.80 statistical package to see whether or not the two m ediating factors mediated t he relationship between an independent and a dependent variable. According to Iacobucci (2008), mediation analysis is defined as follows: A set of statistical procedures used to investigate whether a particular data set exhibits a meditational structur e. A meditational structure posits a particular conceptualization of the mechanism through which an independent variable might affect a dependent variable-not directly, but rather through an intervening process, captured by the m ediator variable. (Iacobucci, 2008, p. 1) In the path of X -> M -> Y, X referred to an exogenous variable while M and Y referred to endogenous variables (Iacobucci, 2008). Structural equations modeling (SEM) examines the process by which an indepen dent variable X is thought to affect a dependent variable Y, directly, as X -> Y, or i ndirectly through a mediator, X -> M -> Y (Iacobucci, 2010, p. 93). 81

PAGE 82

Table 3-1. Treatment condition co mbinations of independent variables Independent variable: Issue ownership Independent variable: Issue attributes (tone) Positive Issue ownership Negative Positive Non-issue ownership Negative Table 3-2. Representation of each research cell Cell Independent variable: Issue ownership Independent variable: Issue attributes (tone) Minimum group size Cell 1 Issue ownership Positive 24 Cell 2 Issue ownership Negative 24 Cell 3 Non-issue ownership Positive 24 Cell 4 Non-issue ownership Negative 24 82

PAGE 83

Table 3-3. Measures Variables Description Issue ownership IO1 How well do you think that the company handles global health issues? IO2 How well do you think that the company can solve problems related to global health issues? IO3 I believe that the company is very knowledgeable about global health issues. Issue attribute IA1 Positively IA2 Pleasantly Issue salience IS1 In your opinion, how important is global health issues today? IS2 How many news reports about global health issues do you regularly pay attention to? IS3 To what extent do you think global health issues is deserving of additional government action? IS4 How often do you talk about global health issues in your everyday conversation? Attribute salience AS1 Please identify your general belie f about the issue: negative-positive AS2 Please identify your general be lief about the issue: unpleasantpleasant Emotion E0 I feel excited E1 I feel angry E2 I feel mad E3 I feel irritated E4 I feel appreciative E5 I feel grateful E6 I feel thankful E7 I feel guilty E8 I feel remorseful E9 I feel sorry E10 I feel proud E11 I feel self-fulfilled E12 I feel joyful E13 I feel happy E14 I feel elated E15 I feel gloomy E16 I feel sad 83

PAGE 84

84 Table 3-3. Continued Variables Description E17 I feel upset Priming effect P1 Do you support or oppose the ef forts to solve the issues? Trust T1 This company would treat customers fairly. T2 Whenever this company makes an im portant decision, I know it will be concerned about its customers. T3 This company can be relied on to keep its promises. T4 I believe that this company ta kes the opinions of customers into account when making decisions. T5 I feel very confident about this company's skills. T6 This company has the ability to accomplish what it says it will do. T7 I believe sound principles guide the companys behavior. T8 I am willing to let this company make decisions for customers like me. T9 I believe the company does not mislead customers. Corporate reputation CR1 The company has market opportunities CR2 The company has excellent leadership CR3 The company has clear vision for the future CR4 The company supports good causes CR5 The company is environmentally responsible CR6 The company is responsible in the community CR7 I feel good about company CR8 The company inspires admiration and respect CR9 The company inspires trust CR10 The company has high quality products and/or services CR11 The company has innovative products and/or services CR12 The company provides good value for money CR13 The company stands behind its products and/or services CR14 The company rewards employees fairly CR15 The company is a good place to work CR16 The company has good employees CR17 The company outperforms competitors CR18 The company has record of bring profitable CR19 The company is low risk investment CR20 The companys growth prospects

PAGE 85

CHAPTER 4 ANALYSES AND RESULTS This chapter describes data analysis procedures to test hypotheses and to answer research questions. Statistical analysis for descriptive statistics, analysis of variance, regression, and the structural equation model ar e explained. Description of Collected Data Two hundred twenty eight people completed the study during the three-week period in March, 2012. Among them, three responses were excluded due to their outstanding study duration longer than 200 minutes. When the three responses were excluded, the range of study dur ation dropped to 47 minutes ( M = 9.48, SD = 5.31) from 627 minutes ( M = 15.54, SD = 57.17). The average age of the three respondents was 26 years old and the majority were white (67%). There were no significant differences in demographic information between the remov ed responses and the responses analyzed. As a result, a total of 225 responses were used for the final data analysis. Among the 225 participants, about 43% of them were male and 57% of them were female, and the average age of the participants was 35 years old. The majority of the participants was white (75%), followed by Asi an (11%), Hispanic/Lati no (7%) or African American (5%), and Native Hawaiian or other Pacific islander (2%). Almost 90% of the participants originally came from the U.S. T he majority of participants held bachelors degree (38%), high school diplomas (31%), associates degrees (14%), or masters degrees (11%). About 44% of participants we re employed for wages, followed by selfemployed (16%), a student (14% ), and currently out of work (13%). The income range for the majority of participants was between $10,000 and $49,999 (51%). About 20% 85

PAGE 86

earned less than $10,000 and about 13% earned between $50,000 and $69,999 (Table 4-1). Descriptive Statistics Issue Ownership and Issue Attributes Manipulation Among 225 participants, 60 people random ly viewed an issue ownership and positive tone condition, 51 viewed an i ssue ownership and negative condition, 63 viewed a non-issue ownership and posit ive condition, and 51 viewed a non-issue ownership and negative condition (Table 4-2 and Table 4-3). Cons idering unequal cell sizes, Type III Sum of Squares were us ed in the factorial analysis of variance (Muenchen & Hilbe, 2010). The Shapiro-Wilk te st was used to check normality of data, and the normality assumption was satisfactory for all dependent variables (p > .05) (Shapiro & Wilk, 1965). Then, the Levenes te st results indicated that the data met homogeneity assumption (p > .05) (Keppel & Wickens, 2004). Cronbach's alpha was used to check the reliability of the index sco re of issue ownership and attribute tone of the condition. Reliability scores were satisf actory for both issue ownership (Cronbach's = .92) and tone of the issue (Cronbach's = .92). Independent sample t-test resu lts show that our two mani pulation conditions were manipulated accordingly. Mean differences between issue-ownership ( M = 5.49, SD = 1.06) and non-issue ownership ( M = 4.56, SD = 1.56) conditions were statistically significant ( t (223) = 5.254, p < .001). Mean differences between positive (M = 5.54, SD = 1.34) and negative ( M = 3.59, SD = 1.78) conditions were also statistically significant ( t (223) = 9.318, p < .001). 86

PAGE 87

Issue Salience An average score of four seven-point Like rt scale items was reported to indicate the level of issue salience ( M = 4.67, SD = 1.07) (Cronbach's = .71). Higher scores mean that participants would cons ider the issues more salie nt and they would pay more attention to the issues (Table 4-4). Attribute Salience Two seven-point Likert scale items were used to measure the attribute salience ( M = 3.53, SD = 1.51). Higher scores mean that par ticipants perceive the issues more pleasantly (Cronbach's = .86). Priming Effect Priming effect was measured by a seven-point Likert scale item, and its mean score was 5.73 ( SD = 1.24). The answer ranged from o ne (extremely opposed) to seven (extremely supported). Emotion Participants emotions were measured by how much they felt excited, ranging from one (not at all) to seven (extremely) ( M = 3.19, SD = 1.46). This study also measured different types of emotions: anger, gratit ude, guilt, pride, happiness, and sadness. Among the 17 items, nine items were used to measure the degree of negative feelings (anger, guilt, and sadness); while eight item s were used to measure the degree of positive feelings (gratitude, pride, and happiness). A princi pal axis factor analysis was run to see whether there are two components extracted for the type of emotion. Using the Oblimin rotation method (when variables ar e assumed to be related with each other), 17 items were loaded onto the two factors, and the two factors were named as emotional pain and emotional pl easure, respectably (Table 45). An emotional pleasure 87

PAGE 88

index was created from the average of eight items me asuring gratitude, pride, and happiness (Cronbach's = .93, M = 3.33, SD = 1.39), and an emotional pain index was created from the average of nine items measuring anger, guilt, and sadness (Cronbach's = .94, M = 2.42, SD = 1.31). Trust Nine seven-point Likert scale items were used to measure the perceived trust toward the company. Higher scores mean that participants perceive the company more trustworthy (Cronbach's = .93, M = 5.01, SD = 1.01). Corporate Reputation Twenty seven-point Likert scale item s were used to measure the perceived corporate reputation. Higher scores mean that the participants consider the companys reputation more favorable. Overall, the av erage score of the corp orate reputation was 5.03 (Cronbach's = .95, SD = .85). Evidence for Research Questions and Hypotheses Hypotheses Testing and Answ ering Research Questions The first set of hypotheses and a res earch question were proposed in the relationship between the two independent variables (issue ownership and issue attributes) and emotional appeal. No intera ction effects were found between the two independent variables on emotions (Table 46). H1 predicted t hat participants who viewed a message in the issue ownership c ondition would have more emotional arousal than those in the non-issue owner ship condition. Two-way analysis of variance results showed that there was no main effect of issue ownership on the level of emotional arousal ( F = 2.127, df = 1, p > .05). There are no signifi cant differences between the issue-ownership ( M = 3.05, SD = 1.49) and non-issue ownership ( M = 3.33, SD = 1.43) 88

PAGE 89

conditions. The results are consistent for the two types of emotion: pain ( F = .243, df = 1, p > .05) and pleasure ( F = .351, df = 1, p > .05) (Table 4-7). H1 was not supported. H2 predicted that participants who view ed a message in the negative condition would have more emotional arousal than those in the positive condition. The two-way analysis of variance results s howed that there was no main effect of issue ownership on the level of emotional arousal ( F = 1.505, df = 1, p > .05) (Positive M = 3.30, SD = 1.46; Negative M = 3.06, SD = 1.46). H2 was not supported. However, the main effect was found when the two types of emotion were separately observed: pleasure ( F = 15.293, df = 1, p < .001) and pain (F = 3.584, df = 1, p > .05). As shown in Table 4-7, participants who viewed a message in the posit ive tone condition showed greater levels of emotional pleasure (Positive M = 3.65, SD = 1.32; Negative M = 2.94, SD = 1.39) than those who viewed a message in the negative condition. The first research question asked whether an interaction effect exists between issue ownership and issue tone on the level of emotional appeal. The two-way analysis of variance results showed no interaction e ffects between issue ow nership and the tone of the issue on emotional appeal ( F = .170, df = 1, p > .05), pleasure ( F = .121, df = 1, p > .05), and pain ( F = 1.253, df = 1, p > .05) (Table 4-6). The second set of hypotheses was proposed to examine the relationship between emotional arousal and issue and attribute salience. H3 pr edicted that participants who have higher levels of emotional arousal would have higher levels of issue salience. Table 4-8 shows a correlation matrix among the variables. Emotional appeal and issue salience had significant correlations ( r = .31, p < .001). To test H3, a regression analysis was run with the level of em otion as the independent variabl e and issue salience as the 89

PAGE 90

dependent variable. The results revealed that emotional appeal was a significant predictor of issue salience ( = .310, t (223) = 4.861, p < .001), and it accounted for about 9.6% of the total varianc e in issue salience. Hence, H3 was supported. Moreover, emotional pain was positively co rrelated with the issue salience ( r = .20, p < .001) while emotional pleasure was not significantly corr elated with issue salience. About 4% of the total variance in issue salience wa s explained by emotional pain ( = .198, t (223) = 3.003, p < .05). H4 proposed that participants who have hig her levels of emotional arousal would have higher levels of attribute salience. Overall, emotional appeal did not have a significant correlation with issue attribute salience ( = .110, t (223) = 1.642, p > .05). H4 was not supported. When the two types of emotion were separately observed, regression analysis results showed that emotional pleasure is a significant predictor of the level of attribute salience ( r = .28, p < .001) ( = .279, t (223) = 4.328, p < .001), and it accounted for about 7.8% of the total variance in attribute salience. The third set of hypotheses and research questions were proposed looking at the relationships between the two independent va riables (issue ownership and issue tone) and the dependent variables (issue and attri bute salience). H5 predicted that participants who viewed a message in the i ssue ownership condition would have a higher level of issue salience than those in the non-issue ownership condition. No interaction effects were found between the tw o independent variables on both issue and attribute salience (Table 4-9). The two-way analysis of variance results showed that there is no main effect of issue owner ship on the level of issue salience ( F = .852, df = 1, p > .05). There was no significant di fference between the issue-ownership ( M = 4.60, 90

PAGE 91

SD = 1.01) and non-issue ownership ( M = 4.73, SD = 1.12) conditions (Table 4-10). H5 was not supported. H6 predicted that participants who vi ewed a message in the issue ownership condition would have higher levels of iss ue attribute salience than those in the nonissue ownership condition. The two-way analysi s of variance results showed that there is no main effect of issue ownership on the level of issue attribute salience ( F = .067, df = 1, p > .05). There was no significant difference between issue-owned ( M = 3.55, SD = 1.56) and non-issue owned ( M = 3.50, SD = 1.46) conditions. H6 was not supported. H7 predicted that participants who view ed a message in the negative condition would have higher levels of issue salience th an those in the positive condition. The twoway analysis of variance results showed that t here is no main effect of issue tone on the level of issue salience ( F = .114, df = 1, p > .05). There was no significant difference between the positive ( M = 4.64, SD = 1.03) and negative ( M = 4.69, SD = 1.12) conditions. H7 was not supported. H8 predicted that participants who view ed a message in the negative condition would have a higher level of issue attribute sa lience than those in the positive condition. The two-way analysis of variance results showed that there is a main effect of issue tone on the level of issue attribute salience ( F = 6.664, df = 1, p < .05). There was a significant difference between the positive ( M = 3.76, SD = 1.52) and negative ( M = 3.25, SD = 1.45) conditions. Hence, H8 was not s upported. However, a main effect was found in an opposite direction. RQ2a asked whether an interaction effect exists between issue ownership and issue tone on the level of issue salience. The two-way analysis of variance results 91

PAGE 92

showed that there is no interaction effe ct between issue ownership and the tone of issue on the level of issue salience ( F = .511, df = 1, p > .05) Research Question 2b asked whether an interaction effect exists between issue ownership and issue tone on the level of iss ue attribute salience. The two-way analysis of variance results found no interaction effe ct between issue ownership and the tone of issue on the level of issue attribute salience ( F = 1.237, df = 1, p > .05) (Table 4-9). The fourth set of hypotheses and a re search question were proposed to explore the mediated effect of emotion in t he relationship between the two independent variables (issue ownership and issue tone) and the two dependent variables (issue salience and issue attributes salience). Path analysis was run using LISREL 8.80 to test Hypotheses 9, 10, 11, and 12. An estima ted model was drawn with three endogenous variables (emotional appeal, issue salience, and issue attributes salience) and two exogenous variables (issue ownership and is sue tone). The estimated model fits the data ( X2 (1) = 1.31, p = 0.25; CFI = .99, NNFI = .93, RMSEA = 0.04). Goodness of fit of a model was estimated with t he Comparative Fit Index (CFI), Non-Normed Fit Index (NNFI), and Root-Mean-Square Error of A pproximation (RMSEA). When the critical value of the CFI and NNFI is .90 or la rger, and the model which has CFI and NNFI value above .95, it is considered an exce llent model (Kline, 2005). The reasonable value range of RMSEA is between .06 and .08, and an RMSEA value less than .06 is excellent (i.e., Hu & Bentler, 1999; Za cchilli, Hendrick, & Hendrick, 2009). H9 examined the indirect effect of i ssue ownership on issue salience through emotion, and it was not supported (indirect effect coefficient = .03, st andard error = .02, p > .05) (Table 4-11). H10 predicted that there is an indirect effect of issue ownership on 92

PAGE 93

issue attribute salience through emotion. Tabl e 4-11 showed that the indirect effect was not supported (indirect effect coeffi cient = -.01, standard error = .01, p > .05). H11 and H12 were also not supported because the indire ct effect of issue tone on issue salience through emotional arousal (indirect effect coefficient = -.01, standard error = .01, p > .05) and the indirect effect of issue t one on issue attribute salience through emotion (indirect effect coefficient = .01, standard error = .01, p > .05) were not significant (Table 4-11). Moreover, RQ3 asked whether the indirect effects are different based on the types of emotions: emotional pleasure and emotional pain. The es timated model fits the data for both emotional pain ( X2 (1) = .015, p = 0.90; CFI = 1.00, NNFI = 1.42, RMSEA = 0.01) and emotional pleasure ( X2 (1) = .93, p = 0.76; CFI = 1.00, NNFI = 1.19, RMSEA = 0.01). Only emotional pleasur e mediated the indirect effe ct of issue tone on issue salience (indirect effect coeffici ent = -.05, standar d error = .02, p < .05) and issue attribute salience (indirect effect c oefficient = -.06, standard error = .02, p < .001) (Table 4-11). The next set of hypotheses and research questions were proposed to explore effects on public opinion in terms of primi ng, trust, and corporate reputation. H13 proposed that participants who hav e higher levels of issue sa lience would have higher levels of priming effect supporting the activi ties regarding the issues. As shown in Table 4-8, there is a significant correlation bet ween issue salience and priming effects ( r = .58, p < .001). Regression analysis results reveal ed that issue salience was a significant predictor of a priming effect ( = .581, t (223) = 10.657, p < .001), and this accounted for about 34% of variance in issue attribute salience. Hence, H13 was supported. 93

PAGE 94

H14 predicted that the parti cipants who have higher levels of issue salience would have higher levels of perceived trust in t he messages of the corp oration. Significant correlations were found between the two variables (r = .18, p < .001) (Table 4-8), and the regression analysis showed that issue salience was a significant predictor of perceived trust ( = .179, t (223) = 2.716, p < .05). Issue salience accounted for about 3% of the variance in the perceiv ed trust. H14 was supported. H15 proposed that participants who have hi gher levels of issue salience would have higher levels of perceived corporate reputation. Two variables were strongly correlated with each other (r = .32, p < .001) (Table 48). The regression analysis revealed that issue salience was a signifi cant predictor of perceived corporate reputation ( = .324, t (223) = 5.100, p < .001), and it accounted for about 11% of total variance. H15 was supported. H16 proposed that participant s who have higher levels of issue attribute salience would have higher levels of priming effects s upporting the activities regarding issues. As shown in Table 4-8, a significant corre lation was found between issue salience and priming effects ( r = -.23, p < .001). Regression anal ysis results revealed that issue attribute salience was a significant predictor of priming effect ( = -.232, t (223) = -3.566, p < .001). Hence, H1 6 was supported. H17 predicted that participants who have hig her levels of issue attribute salience would have a higher level of perceived trus t on the corporation in the messages. No significant correlation was found between the two variables (Table 4-8). Regression analysis also showed that issue attribute sa lience was not a significant predictor of perceived trust ( = -.057, t (223) = -.849, p > .05). Hence, H17 was not supported. 94

PAGE 95

H18 proposed that participant s who have higher levels of issue attribute salience would have higher levels of perceived corpor ate reputation. There was no significant correlation found between these two variables (Table 4-8). Regression analysis results also revealed that issue attr ibute salience was a not signifi cant predictor of perceived corporate reputation ( = -.119, t (223) = -1.784, p > .05). H18 was not supported. To answer the RQ4a, 4b, and 4c, issue and i ssue attribute salience variables were centered by subtracting the mean value from the original issue and issue attribute salience values. Then an interaction term wa s created as a product of the two centered variables (issue salience and issue attribute salience). Finally, r egression analysis was conducted (Aiken & West, 1991). Research Question 4a asked whether an interaction effect between issue and attribute salience exists on a priming effec t. The regression analysis revealed that there is a significant interaction effect on priming ( = .154, t (223) = 2.322, p < .05), and it accounted for about 2.4% of total variance. Research Question 4b asked whether an interaction effect between issue and attribute salience exists on the perceived trust. The regression analysis revealed that there wa s a significant inte raction effect ( = .194, t (223) = 2.940, p < .05) and it accounted for about 4% of total variance. Then, Research Question 4c asked whether an interaction ef fect between issue and attribute salience exists on perceived corporate reputation. A si gnificant effect was found as a result of regression analysis ( = .247, t (223) = 3.783, p < .001), accounti ng for about 6% of total variance. The last two hypotheses were proposed to explore the relationship between priming effect, trust, and co rporate reputation. H19 predict ed that participants who have 95

PAGE 96

higher levels of priming effect would have higher levels of perce ived trust in messages from the corporation. T he two variables were significantly correlated ( r = .31, p < .001), and regression analysis revealed that a priming effect was a significant predictor of perceived trust ( = .314, t (223) = 4.934, p < .001). This accounted for about 10% of the variance of perceived trus t. H19 was supported. Finally, H20 proposed that par ticipants who have higher levels of perceived trust would have higher levels of perceived corp orate reputation. There was a strong correlation between the two variables ( r = .86, p < .001) (T able 4-8). Regression analysis revealed that perceived trust was a si gnificant predictor of perceived corporate reputation ( = .855, t (223) = 24.579, p < .001). This accounted for about 73% of the variance of perceived trust. Hence, H20 was supported. Figure 4-2 summarized the results of proposed model. Additional Analysis: Path Analysis of Proposed Model In addition to ANOVA and regression analy sis, a structural equation model was estimated to analyze all the relationships simultaneously. The model includes both manipulated (Issue ownership and iss ue tone dummy variables) and measured variables (e.g., issue salience, emotion, priming effect, trust, and corporate reputation). The original path model did not fit the data ( X2 (30) = 135.59, p < 0.001). Based on the modification indices, the following paths were added: from the perceived issue ownership and perceived issue tone to trust, from the perceived issue ownership and perceived issue tone to corporat e reputation, from emotional pain to priming effect, and from emotional pleasure to trust and to corporate reputation. Even though these relationships were not included in the propos ed hypotheses, the direct paths from the 96

PAGE 97

two independent variables and public relation s outcomes (e.g., trust or corporate reputation) can be explained theoretically. The modified path model fits the data ( X2 (23) = 28.39, p = 0.20; CFI = .99, NNFI = .98, RMSEA = 0.03). Signifi cant path loadings were also found from emotional appeal to issue salience (coefficient = .28, p < .001), emotional pain to issue salience (coefficient = .14, p < .05), emotional plea sure to issue attributes salience (coefficient = .22, p < .001), issue salience to priming (coefficient = .62, p < .001), issue salience to corporate reputation (coefficient = .16, p < .001), attribute salience to priming (coefficient = -.20, p < .001), priming effects to trust (coefficient = .26, p < .001), and trust to corporate reputat ion (coefficient = .76, p < .001). These relationships were supported from the hypotheses testing in t he previous section, and the structural equation model added evidence to show that t hese relationships are robust (Figure 4-3). The analysis of variance (ANOVA) result s showed that there were no direct relationships from issue ownership to emot ional reactions (emotional appeal, emotional pleasure and emotional pain); however, a st ructural equation model reported significant coefficients from issue ownership to emotional appeal (coefficient = .17, p < .05). Also, a significant coefficient was found from attr ibute salience to corporate reputation (coefficient = .08, p < .05) as a result of the structural equation modeling. Moreover, the modified model included the following significant relationships: from perceived issue ownership to trust (coeffici ent = .39, p < .001), from perceived issue ownership to corporate reput ation (coefficient = .12, p < .001), from perceived issue tone to trust (coefficient = .17, p < .001), from emoti onal pain to trust (coefficient = -.19, p < .001), and from emotional pleasu re to trust (coefficient = .17, p < .001). These 97

PAGE 98

relationships were not tested in the hypothes es since the main goal of this current research was to test indirect influence of issue ownership and issue tone on the public opinion outcomes. However, the relationships can also be drawn theoretically (Figure 23 in Chapter 2). Hence, the path analysis re sults in Figure 4-3 suggest that there are some direct relationships between perceiv ed issue ownership and perceived issue tone on some public opinion outcomes (e.g., trus t and corporate reputation), in addition to the indirect influences through emoti ons and agenda salience as proposed in the current research. 98

PAGE 99

Table 4-1. Demographic profiles Variables Value Frequency Percentage (%) Gender Female Male 127 97 56.4 43.1 Age (years) 18-19 20-29 30-39 40-49 50-59 60-69 70-79 Unidentified 10 89 55 30 24 10 2 1 4.4 39.6 24.4 13.3 10.7 4.4 .9 .4 Race Hispanic or Latino Asian African American Native Hawaiian or other Pacific Islander White 16 24 12 4 168 7.1 10.7 5.3 1.8 74.7 Education No schooling completed Nursery school to 12th grade (no diploma) High school graduate Associates degree Bachelors degree Masters degree Professional degree Doctoral degree 5 1 70 29 77 21 4 2 1.0 .5 31.3 14.6 38.9 10.6 2.0 1.0 Occupation Employed for wages Self-employed Out of work and looking for work A homemaker A student Retired Unable to work 100 37 29 17 31 6 4 44.4 16.4 12.9 7.6 13.8 2.7 1.8 Income Less than $10,000 $10,000 to $29,999 $30,000 to $49,999 $50,000 to $69,999 $70,000 to $89,999 $90,000 to $109,999 $110,000 to $129,999 $130,000 or more 47 51 62 30 19 10 4 1 20.9 22.7 27.6 13.3 8.4 4.4 1.8 .4 99

PAGE 100

Table 4-2. Participants in eac h manipulation condition Cell Independent variable: Issue ownership Independent variable: Issue attributes (tone) Total group size Cell 1 Issue ownership Positive 60 Cell 2 Issue ownership Negative 51 Cell 3 Non-issue ownership Positive 63 Cell 4 Non-issue ownership Negative 51 Table 4-3. Frequency for each condition Variables Frequency Percentage Issue ownership Issue-owned Non-issue owned 111 114 49.3 50.7 Issue tone Positive Negative 123 102 54.7 45.3 100

PAGE 101

Table 4-4. Means and standard deviations for measures Variables M SD Overall: Issue salience (Cronbach's = .75) In your opinion, how important is global health issues today? 5.85 1.17 How many news reports about global health issues do you regularly pay attention to? 4.34 1.51 To what extent do you think global health issues is deserving of additional government action? 5.27 1.49 How often do you talk about global health issues in your everyday conversation? 3.20 1.61 Overall: Issue attribute salience (Cronbach's = .86) Positive 3.72 1.69 Pleasant 3.33 1.53 Overall: Emotion I feel excited 3.19 1.46 Overall: Emotion (details) (Cronbach's = .86) I feel angry 2.04 1.39 I feel mad 2.03 1.41 I feel irritated 2.21 1.43 I feel appreciative 3.78 1.64 I feel grateful 4.05 1.65 I feel thankful 4.04 1.68 I feel guilty 2.50 1.70 I feel remorseful 2.32 1.66 I feel sorry 2.86 1.78 I feel proud 2.95 1.74 I feel self-fulfilled 2.92 1.68 I feel joyful 3.02 1.82 I feel happy 3.19 1.82 I feel elated 2.71 1.68 I feel gloomy 2.64 1.69 I feel sad 2.59 1.70 I feel upset 2.56 1.65 101

PAGE 102

Table 4-4. Continued Variables M SD Overall: Priming effect Do you support or oppose the effort s to solve the issues? 5.73 1.24 Overall: Trust (Cronbach's = .94) This company would treat customers fairly. 5.21 1.18 Whenever this company makes an im portant decision, I know it will be concerned about its customers. 5.08 1.21 This company can be relied on to keep its promises. 4.95 1.17 I believe that this company take s the opinions of customers into account when making decisions. 5.18 1.15 I feel very confident about this company's skills. 5.00 1.26 This company has the ability to accomplish what it says it will do. 5.11 1.29 I believe sound principles guide the companys behavior. 5.18 1.22 I am willing to let this company make decisions for customers like me. 4.47 1.56 I believe the company does not mislead customers. 4.86 1.24 Overall: Corporate reputation (Cronbach's = .95) The company has market opportunities 5.39 1.22 The company has excellent leadership 5.06 1.16 The company has clear vision for the future 5.42 1.24 The company supports good causes 5.70 1.16 The company is environmentally responsible 5.11 1.30 The company is responsible in the community 5.24 1.22 I feel good about company 5.07 1.31 The company inspires admiration and respect 5.01 1.34 The company inspires trust 4.98 1.27 The company has high quality pr oducts and/or services 5.17 1.18 The company has innovative products and/or services 5.31 1.13 The company provides good value for money 4.74 1.08 The company stands behind its products and/or services 5.27 1.09 The company rewards employees fairly 4.54 .98 The company is a good place to work 4.74 1.14 The company has good employees 4.80 1.12 The company outperforms competitors 4.76 1.10 The company has record of bring profitable 4.75 1.14 The company is low risk investment 4.42 1.18 The companys growth prospects 5.01 1.16 102

PAGE 103

Table 4-5. Principle Axis Factor loading of emotion Extracted factors* Measurement items Emotional pain Emotional pleasure I feel angry .802 I feel mad .818 I feel irritated .633 I feel appreciative .723 I feel grateful .747 I feel thankful .735 I feel guilty .760 I feel remorseful .797 I feel sorry .775 I feel proud .741 I feel self-fulfilled .798 I feel joyful .881 I feel happy .832 I feel elated .778 I feel gloomy .787 I feel sad .868 I feel upset .866 Eigenvalue 6.311 4.288 % of explained variance 37.124 25.222 *Rotation method: Oblimin with Kaiser Normalization 103

PAGE 104

Table 4-6. Two-way analysis of variance of issue ownership and issue tone on emotion Main and interaction effects Mean SD F df p eta2 power Emotional appeal Issue ownership 2.127 1 .15 .009 .31 Issue-ownership 3.05 1.49 Non-issue ownership 3.33 1.43 Issue tone 1.505 1 .22 .007 .23 Positive 3.30 1.46 Negative 3.06 1.46 Interaction .170 1 .68 .001 .07 Ownership and Positive 3.12 1.59 Ownership and Negative 2.96 1.39 None and Positive 3.48 1.32 None and Negative 3.16 1.54 Emotional pain Issue ownership 1.750 1 .41 .001 .08 Issue-ownership 2.38 1.37 Non-issue ownership 2.46 1.25 Issue tone 29.738 1 .12 .012 .33 Positive 2.27 1.30 Negative 2.60 1.29 Interaction .121 1 .73 .001 .06 Ownership and Positive 2.19 1.34 Ownership and Negative 2.59 1.39 None and Positive 2.34 1.28 None and Negative 2.61 1.20 Emotional appeal Issue ownership .379 1 .65 .002 .05 Issue-ownership 3.38 1.36 Non-issue ownership 3.29 1.43 Issue tone 12.113 1 .18 .064 .22 Positive 3.65 1.32 Negative 2.95 1.39 Interaction 1.253 1 .26 .005 .20 Ownership and Positive 3.61 1.27 Ownership and Negative 3.11 1.42 None and Positive 3.69 1.37 None and Negative 2.78 1.36 Note: Levenes test was run to conf irm the homogeneity of the sample ( p > .05) 104

PAGE 105

Table 4-7. Main effects of issue ownership and issue tone on emotion Main effects Mean SD F df p eta2 power Emotional appeal Issue ownership 2.127 1 .15 .009 .31 Issue-ownership 3.05 1.49 Non-issue ownership 3.33 1.43 Issue tone 1.505 1 .22 .007 .23 Positive 3.30 1.46 Negative 3.06 1.46 Emotional pain Issue ownership .243 1 .62 .001 .08 Issue-ownership 2.38 1.37 Non-issue ownership 2.46 1.25 Issue tone 3.584 1 .06 .016 .47 Positive 2.27 1.30 Negative 2.60 1.29 Emotional pleasure Issue ownership .351 1 .55 .001 .09 Issue-ownership 3.38 1.36 Non-issue ownership 3.29 1.43 Issue tone 15.293 1 .00* .065 .97 Positive 3.65 1.32 Negative 2.94 1.39 Note: Levenes test was run to conf irm the homogeneity of the sample ( p > .05) p < .001 105

PAGE 106

Table 4-8. Pearson product-moment correlation coefficients 1 2 3 4 5 5.1 5.2 6 7 8 9 10 1. Issue ownership 1 2. Issue tone n.s. 1 3. Perceived issue ownership n.s. n.s. 1 4. Perceived issue tone n.s. -.53** .18** 1 5. Emotional arousal (excited) n.s. n.s. .20** n.s. 1 5.1 Pain n.s. n.s. n.s. n.s. .23** 1 5.2 Pleasure n.s. -.25** .24** .26** .47** .17* 1 6. Issue salience n.s. n.s. n.s. n.s. .31** .20** n.s. 1 7. Issue attribute salience n.s. -.17* n.s. .34** n.s. n.s. .28** n.s. 1 8. Priming n.s. n.s. n.s. n.s. n.s. n. s. n.s. .58** -.23** 1 9. Trust -.14* n.s. .49** .25* .15* n.s. .28** .18** n.s. .31** 1 10. Corporate reputation -.15* n.s. .52** .17* .23** n.s. .29** .32** n.s. .41** .86** 1 Note: Issue ownership was coded as 0 (nonownership) and 1 (ownership); and issue tone was coded as 0 (positive) and 1 (negative). n.s. refers to not significant. p < .05, **p < .001 (two-tailed) 106

PAGE 107

Table 4-9. Two-way analysis of variance of issue ownership and issue tone on salience Main and interaction effects Mean SD F df p eta2 power Issue salience Issue ownership 1.419 1 .45 .003 .08 Issue-ownership 4.60 1.01 Non-issue ownership 4.73 1.12 Issue tone .234 1 .71 .001 .06 Positive 4.64 1.03 Negative 4.69 1.12 Interaction .511 1 .48 .002 .11 Ownership and Positive 4.53 .96 Ownership and Negative 4.68 1.07 None and Positive 4.75 1.09 None and Negative 4.70 1.17 Attribute salience Issue ownership .105 1 .80 .001 .05 Issue-ownership 3.55 1.56 Non-issue ownership 3.50 1.46 Issue tone 5.342 1 .26 .029 .14 Positive 3.76 1.52 Negative 3.25 1.45 Interaction 1.237 1 .27 .005 .20 Ownership and Positive 3.68 1.60 Ownership and Negative 3.39 1.50 None and Positive 3.83 1.44 None and Negative 3.10 1.40 Note: Levenes test was run to conf irm the homogeneity of the sample ( p > .05) 107

PAGE 108

Table 4-10. Main effects of issue ownership and issue tone on salience Main effects Mean SD F df p eta2 power Issue salience Issue ownership .852 1 .36 .004 .15 Issue-ownership 4.60 1.01 Non-issue ownership 4.73 1.12 Issue tone .114 1 .74 .001 .06 Positive 4.64 1.03 Negative 4.69 1.12 Attribute salience Issue ownership .067 1 .79 .001 .06 Issue-ownership 3.55 1.56 Non-issue ownership 3.50 1.46 Issue tone 6.664 1 .01* .030 .73 Positive 3.76 1.52 Negative 3.25 1.45 Note: Levenes test was run to conf irm the homogeneity of the sample ( p > .05) p < .05 108

PAGE 109

Table 4-11. Direct and indirect effects on endogenous variables Endogenous variables Paths Coefficients Standard errors Emotional appeal ( X2 (1) = 1.31, p = 0.25; CFI = .99, NNFI = .93, RMSEA = 0.04) Issue salience Issue ownership issue salience Issue ownership emotion issue salience Issue tone issue salience Issue tone emotion issue salience .03 .03 .25** -.03 .06 .02 .06 .02 Attribute salience Issue ownership issue attribute salience Issue ownership emotion issue attribute salience Issue tone issue salience Issue tone emotion issue attribute salience -.01 -.01 -.18** .01 .07 .01 .07 .01 Emotional pain ( X2 (1) = .015, p = 0.90; CFI = 1.00, NNF I = 1.42, RMSEA = 0.01) Issue salience Issue ownership issue salience Issue ownership emotion issue salience Issue tone issue salience Issue tone emotion issue salience .06 .01 .20** .02 .07 .01 .07 .01 Attribute salience Issue ownership issue attribute salience Issue ownership emotion issue attribute salience Issue tone issue salience Issue tone emotion issue attribute salience -.03 .01 -.19** .02 .07 .01 .07 .01 Emotional pleasure ( X2 (1) = 0.93, p = 0.76; CFI = 1.00, NNFI = 1.19, RMSEA = 0.01) Issue salience Issue ownership issue salience Issue ownership emotion issue salience Issue tone issue salience Issue tone emotion issue salience .07 -.01 .27** -.05* .06 .01 .07 .02 Attribute salience Issue ownership issue attribute salience Issue ownership emotion issue attribute salience Issue tone issue salience Issue tone emotion issue attribute salience -.01 -.01 -.11 -.06** .06 .02 .07 .02 Note: ** p < .01, *p < .05 (two-tailed) 109

PAGE 110

Table 4-12. Summary of hypotheses testing Relationship ANOVA or Regression SEM Coefficient H1 Issue ownership Emotional appeal Issue ownership Emotional pleasure Issue ownership Emotional pain Not significant Not significant Not significant .17** n.s n.s H2 Issue tone Emotional appeal Issue tone Emotional pleasure Issue tone Emotional pain Not significant Not significant Significant n.s n.s .16** H3 Emotional appeal Issue salience Emotional pleasure Issue salience Emotional pain Issue salience Significant Not significant Significant .28** n.s .14** H4 Emotional appeal Issue attribute salience Emotional pleasure Issue attribute salience Emotional pain Issue attribute salience Not significant Significant Not significant n.s .22** H5 Issue ownership Issue salience Not significant n.s H6 Issue ownership Issue attribute salience Not significant n.s H7 Issue tone Issue salience Not significant n.s H8 Issue tone Issue attribute sa lience Significant (opposite direction) n.s H9 Issue ownership Emotional appeal Issue salience Issue ownership Emotional pleasure Issue salience Issue ownership Emotional pain Issue salience n.s n.s n.s H10 Issue ownership Emotional appeal Issue attribute salience Issue ownership Emotional pleasure Issue attribute salience Issue ownership Emotional pain Issue attribute salience n.s n.s n.s H11 Issue tone Emotional appeal Issue salience Issue tone Emotional pleasure Issue salience Issue tone Emotional pain Issue salience n.s -.05* n.s H12 Issue tone Emotional appeal Issue attribute salience Issue tone Emotional pleasure Issue attribute salience Issue tone Emotional pain Issue attribute salience n.s n.s n.s 110

PAGE 111

Table 4-12. Continued Relationship ANOVA or Regression SEM Coefficient H13 Issue salience Priming Significant .62** H14 Issue salience Trust Significant n.s H15 Issue salience Corporate reputation Significant .16** H16 Issue attribute salience Priming Significant -.20** H17 Issue attribute salience Trust Not significant n.s H18 Issue attribute salience Corporate reputation Not significant .08* H19 Priming Trust Significant .26** H20 Trust Corporate reputation Significant .76** Note: ** p < .01, *p < .05 (two-tailed) 111

PAGE 112

Figure 4-2. Results of proposed model 112

PAGE 113

113 Figure 4-3. Path analysis of the proposed model ( X2 (23) = 28.39, p = 0.20; CFI = .99, NNFI = .98, RMSEA = 0.03) (errors were co rrelated among emotional appeal, emotional pleasure, and emotional pain)

PAGE 114

CHAPTER 5 DISCUSSION AND CONCLUSIONS This chapter presents a summary of results with theoret ical and practical implications. It also offers the studys lim itations, provides conclusions, and addresses future research direction. Theoretical Implications This study examined 1) the relationshi p between the independent variables (issue ownership and issue attributes) and the mediat or (emotion), 2) the relationship between the mediator (emotional arousal) and the dependent variables associated with agenda salience (issue and attribute salience), 3) the relationship bet ween the independent variables (issue ownership and issue attributes) and the dependent variables associated with agenda salience (issue and attribute salience), and 4) the relationship between the dependent variables associated with agenda salience (issue and attribute salience) and with public opinion (priming effe ct, trust, and corporate reputation). Twenty hypotheses were proposed to test the relationships: seven were supported and a relationship was found in the opposite direct ion from the prediction (Table 4-12 in Chapter 4). In the relationship between t he independent variables (issue ownership and issue attributes) and the mediator (emoti onal appeal), the hypotheses testing results showed that only issue tone was related to em otional appeal. This study predicted that a negative tone would lead to more emotional reactions; however, results showed that a positive tone would lead to more emotional appeal. A structural equation model also strengthened the positive relationship between issue tone and emotion. No direct effects were reported from issue ownership to emotions as a result of the analysis of variance; however, a structur al equation model showed a significant 114

PAGE 115

relationship between issue ownership and emot ional appeal. The mediator (emotional appeal) was then a significant predictor of issue salience. The results suggested that issue ownership indirectly influences issue salience via emotional appeal. Consequently, issue salience directly and indirectly influences public opinion (e.g., priming, trust, and corporate reputation). The direct effects of issue salience were found for priming and corporate reputat ion, while an indirect e ffect was found for perceived trust. Moreover, there was an indirect effect on public opini on through attribute salience. Issue tone was suggested as a significant predictor of issue attribute salience, and issue attribute salience was a predictor of pr iming and corporate repu tation effects. Also, interaction effects were found between Issue sa lience and attribute salience on priming, trust, and corporate reputation. Agenda-Building Theory One of the main theoretical contributions of this st udy was the exploration of agenda-building theory in detail in a corporat e communication setting. Particularly, this study suggested issue ownership and iss ue tone as antecedents of agenda-building effects, and public opinion (e.g., primi ng, trust, and corporate reputation) as consequences of agenda salience. Issue ownersh ip was operationalized as to whether or not the company, which distributed t he messages, handled the issues mentioned in messages successfully, and issue tone was about whether or not the issues were portrayed positively or negatively. No direct effect was found supporting the relationship between issue ownership and agenda salience. Ho wever, the results showed that the relationship was mediated by emotional appeals. Expanding from an agenda-setting pers pective (McCombs & Shaw, 1972), the agenda-building concept was explored by several public relations scholars (e.g., Kaid, 115

PAGE 116

1976; Kiousis et al., 2006; Miller, 2010; Ohl et al., 1995). These public relations scholars emphasized the role of public relations effort s (e.g., information subsidies) in building the media and the public agenda (e.g., Kaid 1976; Miller, 2010; Ohl et al., 1995). A large portion of agenda-building research has focused on whether agenda salience transfer exists between information subs idies and media coverage or between information subsidies and public opinion. Th e study tried to expand the scope by focusing on what conditions agenda salienc e transfer would exist. As the results suggested, issue ownership and issue t one played a role of antecedence of agenda salience mediated by emotions. Two independent variables were proposed to explore the relationship between the first-level and second-level of agenda building. Scholars defined the salience of objects as the first-level of agenda building, and defined the salienc e of attributes as the second-level of agenda building (e.g., Ki ousis, 2005; McCombs & Shaw, 1972). For example, the first-level of agenda building focuses on what to talk about (e.g., issues, organizations, or candidates), while the se cond-level focuses on how to talk about the objects (e.g., frames or t one). Scholars have noted that first-level agenda setting is associated with the strength of an opinion, and that second-level of agenda setting is associated with the direction of the opini on (e.g., Kiousis, 2005; Lee, 2010; McCombs & Shaw, 1972; Sheafer, 2007). Also, scholars have explored the relationships between the first-level and second-level of agenda building. The compelling-arguments hypothesis was tested by demonstrating the influence of attr ibutes on object evaluations (Lee, 2010; Sheafer, 2007). 116

PAGE 117

Moreover, two dimensions exist on the second-level of agenda-setting and agenda-building: substantive and affective attributes (Kiousis, 2005; Lee, 2010; McCombs & Shaw, 1972; Sheafer, 2007). Substant ive attributes are related to cognitive thinking of objects, and affective attributes are related to the emotional (affective) evaluation of objects. In this study, iss ue ownership was suggested as an example of substantive attributes, and issue tone wa s suggested as an example of affective attributes in global issue-related messages. Based on the compelling arguments, hypotheses we tested to see whether or not the attributes of the messages influenced the overall evaluation of objects. In a po litical communication context, scholars have showed that voters would like to vote for a candidate who is competent regarding the issues salient to them (Belanger & Megui d, 2008; Cha et al., 2010; Green & Hobolt, 2008). Results of this study have also sugge sted that people woul d evaluate a company as being trustworthy or reputable when the co mpany showed issue ownership of certain issues. Moreover, through indirect relationships, this study explained that issue ownership would lead to higher emotional a ppeal that played as a predictor of issue salience, priming, tr ust, and reputation. In addition to issue ownership, the role of issue tone was explored and supported in the relationship. Participants, who viewed negative issue messages, reported a higher level of attribute salience for the issue, meaning that issues would have a stronger impact on how people perceived the va lence of the issues when they are negatively described. On the other hand, the e ffect of issue tone was different on issue salience. Participants in the positive issue condition reported a higher level of emotional appeal, which led to a higher level of issue sali ence. Still the effect of the valence of the 117

PAGE 118

issue is open for debate. While some studi es have highlighted the role of negative affective attributes in affecting overall i ssue importance (Semetko, 1992; Sheafer, 2007), some researchers have stated that positive attributes would also have an influence on the evaluation of the political candidate (Kim & McCombs, 2007). The results showed that information processing may be diffe rent based on valence of the issues. This study also suggested exploring the consequences of agenda building in a corporate communication cont ext. Priming, trust, and cor porate reputation were the three most important pub lic opinion consequences of agenda building. Scholars explored the agenda-building concept not only in the area of political communications but also in corporate communication settings (e.g., Cameron et al., 1997; Curtin, 1999; Gandy, 1982). Carroll and McCombs (2003) poi nted out that the key idea of agenda setting (or agenda building) is the transfer of agendas from one to another political party, and the theory can be easily appl ied to a corporate communica tion context. As with job approval ratings in a political communicati on context, trust, and corporate reputation were suggested as the most emphasized out comes of corporate communication efforts (Meijer & Kleinnijenhuis, 2006). Because Ferguson (1984) suggested a relations hip as the central unit of public relations, the organization-public relationshi p has been greatly studied and scholars have developed measurement item s for the relationship out comes. The most widely used dimensions of the measurement we re trust, openness, or commitment. In particular, trust has been suggested as one of the most essential outcomes of relationships (e.g., Swift, 2001; Yang & J. Grunig, 2005). In addition to trust, corporate reputation was suggested as a critical measurement of public opinion in a corporate 118

PAGE 119

communication context. Wartick (1992) defined corporate reputation as the publics perception about a corporati ons effectiveness, and emphasized the role of media portrayals of a company as a predictor of co rporate reputation. In the public relations literature, scholars have also suggested corporate reputat ion as an important outcome of public relations efforts influenced by information subsidies and media coverage (Kiousis et al., 2007). In spite of its importance, the effect on corporate reputation has not yet been fully explored. Scholars have noted that the role of media cover age on reputation or the role of mediating factors (e.g., s ource credibility, hist ory, or other situat ional factors) needs to be explored further (Fombr un & Shanley, 1990; Griffin et al., 1991). Results of this study added some evidence concerning the di rect and indirect effects on corporate reputation, exploring the two message fact ors (issue ownership and issue tone) and mediating influences (emotion). Also, the results suggested that participants perception of trust is a significant predictor of t he perception of corpor ate reputation. Finally, this study demonstrated a causal relationship in an agenda-building framework. Many previous agenda-se tting or agenda-building studies have used content analysis along with surveys (e.g., Iyengar & Kinder, 1987; Kiousis & McDevitt, 2008; Kiousis et al., 2006). Scholars have supported agenda-setting (or building) influences by showing correlation results between information subsidies and media coverage, and between media cove rage and public opinion. Ho wever, this study aimed to explore causal relationships in agenda buil ding by conducting an experimental study. The 2 x 2 factorial design experiment showed both direct and indire ct effects of the message attributes on agenda salience transfer and further public relations outcomes of 119

PAGE 120

the agenda-building effects. Hence, this st udy added to our theoretical understanding of the process of the agenda-building effect. Corporate Communication a nd Issues in Business The two independent variables of this study (issue ownership and issue tone) were proposed in consideration of the role of i ssues in corporate communication. Not only non-profit organizations but also for-profit organizations are now involved in various social issues, such as global warming, human rights, or public health. Scholars have suggested that corporate social responsibil ity has become one of the most prominent agendas for business corporations (Bhatta charya et al., 2008; Clark, 2000). Bhattacharya et al. (2008) stated that corporat e social responsibility activities can be a significant predictor of the relationship qualit y of the corporation wit h publics. Results of this study found supportive evidence of the role of issue-related activities of a company in affecting public perceptions of the com pany in terms of both trust and reputation. Emphasizing the role of media coverage, Carroll (2004) found that media description of issues or other attributes of a firm were associated with the publics perceptions about the firm. In a corpor ate communication context, Carroll and McCombs (2003) suggested several substantive attribute variables, such as familiarity, corporate citizenship, leadersh ip, or credibility. In addition the issue ownership concept was suggested as an example of substantive attributes in this study. Traditionally, the issue ownership concept was widely explored in a political communication context, and the agenda-building process has also been greatly studied in political communication settings, such as presidential election campaigns. Carroll and McCombs (2003) ex plained that the same th eoretical framework can be applied to other communica tion contexts such as corporate communication. 120

PAGE 121

However, empirical research evidence in a corporate communication context was limited compared to political communication ar eas. This study provided useful empirical evidence of the importance of an agenda-building process and the effects in a corporate communication context. Depending on how a company portrays issue-related information in its public relations messages, the public may perceive the issue more salient than other issues. The companys leve l of issue salience would then affect the publics overall evaluations about the company. Due to the multiple stakeholder interactions and reciprocal dialogue capability, issue-arena communications became an e ssential management strategy for an organization. Previous scholarship has s uggested several predictors of public opinion outcomes, such as the number of involved actors, the amount of media visibility, or the intensity of public interest in the iss ue (Luoma-aho & Vos, 2010). This study suggested two issue-related attributes as critical predictors of communica tion outcomes. Issue ownership and the tone of issues were explored together in the agenda-building process. In addition, this experimental st udy also measured the level of emotional appeal to examine the role of af fect in the relationship. Emphasizing the role of an evaluative tone of news media coverage on a political candidate, previous scholarship found the role of affective attributes in affecting public opinion about objects (e.g., Kepplinger et al., 1989; Kim & McCombs, 2007). For example, a candidates positive or negative portrayal in the media coverage was the predictor of public opinion about that candidate (Kim & McCombs, 2007). On the other hand, some previous studies highlighted the role of negative affective attributes in 121

PAGE 122

shaping overall issue importance, mentioning that a positive tone of media coverage would decrease the issue importanc e (Semetko, 1992; Sheafer, 2007). The results of this invest igation also suggested that the tone of an issue has a significant influence on issue salience. For example, supporting the indirect and direct effects of issue attributes on issue salienc e, the results suggested that a negative tone would have more of an impact on affective ev aluations of issues; on the other hand, a positive tone would also have an impact on the overall evaluations of the issues at the object level. Even though more empirical evidence should be added to compare the effects of the different types of affective attributes on issue salience, this study has attempted to expand the understanding of the ro le of affective attributes in the agendabuilding process. Mediating Factor in th e Agenda-Building Process Several studies have emphasized the impor tance of the effects of affective attributes in affecting cogni tive thinking (e.g., Arnold, 1985; Dillard & Wilson, 1993). Scholars have found that emotionally arousing messages can affect the public's cognitive information processing (Keller & Block, 1996), and the affective tone of messages can affect the overall evaluation of the objects (e.g., political leaders) (Sheafer, 2007). However, the role of affective attributes has not been explored as much, when compared to substantive attributes such as frames. From the agenda-setting or agenda-building liter ature, only the t one of objects or messages has been widely observed in terms of affective attributes. Defined as the feeling toward a stimulus that leads to relati ve preferences toward that stimulus out of a class of similar stimuli, affects or emoti ons are the dynamic conc ept (Batra, 1986, p. 54). In addition to the valence (positive or negative), scholars have also noted that 122

PAGE 123

another dimension of affect is related to t he level of arousal (e.g., Dunn & Schweitzer, 2005; Smith & Ellsworth, 1985). The effect s of emotionally arousing messages on cognitive information processing or overall evaluation toward an object have not yet been fully explored in the agenda-setting or agenda-building literature. In this study, emotion (emotional appeal, emotional pleasure, and emotional pain) was established as a successful mediator of issue ownership a nd issue attributes for the level of issue salience. Hence, the results of this study contributed to expand ing our understanding of the role of affective at tribute in agenda building. Dunn and Schweitzer (2005) suggested six different dimensions of emotions: anger, gratitude, guilt, pride, happiness, and sadness. Sev enteen items were used to measure the six dimensions of the emotion type. From the factor analysis results, 17 items were divided into two factors. Eight items were gr ouped to define the factor of emotional pleasure (gratitude, pride, and happiness items), wh ile nine items were used to define the factor of em otional pain (anger, guilt, and sadness items). These various emotional items showed differ ent mediating effects betw een issue ownership and issue salience dependent variables. Issue ownership has an indirect effect on issue salience via emotional appeal (single item measured by emotional excitement). The path analysis results also showed that issue t one has an indirect effect on trust through emotional pleasure. By demonstrating mediating relationships these results added empirical evidence to the argument, explaining that emotionally arousal messages influence further cognitive or rational thinking about certai n objects (Coleman & Wu, 2010; Keller & Block, 1996). The emotionally aroused feeling is a predictor of the prominence judgment of the 123

PAGE 124

objects (e.g., Miller, 2007). In addition to the general emotional appeal, this study conducted a factor analysis and suggested tw o emotional appeal fa ctors: emotional pleasure and emotional pain. Still, the direction and strengt h of the effects of emotions are underdeveloped. Wu and Coleman (2009) said that negative info rmation has more power to transfer the medias agenda of candidate attributes to th e public (p. 775). Sheafer (2007) further stated that negative affective attributes would have a stronger influence on the public's perception about issue importance. This cu rrent study added evidence that a negative message successfully draws participants' attention to an issue, consequently affecting the level of perceived corpor ate reputation. Operationally scholars defined the public agenda as one of the most impor tant problems facing the nation (Sheafer, 2007; Wu & Coleman, 2009), and the public would perceiv e an issue more prominently when the issue is described negativel y or problematically. On the other hand, this study also emphasiz ed the impact of positive emotions on the perceived trust toward a corporation, consequently affecting a corporation's perceived reputation. Research ers showed that the medias evaluative assessments of an object were closely related to public opinion about the object (Kepplinger et al., 1989). Kim and McCombs (2007) found that the medias evaluative assessments of a politician shifted public opinion about that politician. For example, positive news coverage of a nation would have a positive in fluence on the international community's perception of that nation in question (Wanta & Mikusova, 2010). The results of this study showed that a negative emotion has an effect on corporate reputation through perceived issue prominence whil e a positive emotion has an effect on the corporation's 124

PAGE 125

reputation through perceived trust. Even t hough it expanded our understanding of the effects by the type of emotions, the direction and strength of the effects should be tested further. Trust as a Relationship Quality Ou tcome and Corporate Reputation One of the primary goals of this study was to demonstrate the antecedents and consequences of the agenda-building process. Perceived trust and corporate reputation were suggested as the consequences of agenda building. These consequences were the most critical public relations outcomes in a corporate communication context (e.g., Swift, 2001; Yang & J. Grunig, 2005). From a relationship manag ement perspective, communication strategy plays an important role in cultivati ng the quality of relationships. J. Grunig and Huang (2000) explained that co mmunication strategies would link the antecedents and consequences of an organizationpublic relationship. The concept of relationships is understood as a byproduct of communication activities. This study explored the effect on trust as one of t he relationship's outcomes to evaluate the communication strategies in a corporate co mmunication context. Moreover, the public's overall evaluation of a company was meas ured in terms of corporate reputation. Previous research has suggested the role of communication efforts (e.g., information subsidies or media coverage) in affecting corporate reputation (e.g., Gotsi & Wilson, 2001; Wartick, 1992). As a result, issue ownership and iss ue tone were suggested as important predictors of trust and cor porate reputation, and the results showed that agenda building can provide a useful framework to evaluate communication strategies in a corporate communication setting. Based on the compelling argument of an agendabuilding perspective (e.g., Kiousis, 2005), th is study supported t he effects of the 125

PAGE 126

attribute-level of agenda on the object-leve l and other public opinion consequences. More importantly, the research expanded our understanding of the predictors or mediators (emotion) in t he agenda-building process. Practical Implications The issue management perspective has grown since the 1970s (Berger, 2001), and the role of public relations has also been emphasized to educate the public about certain issues, to manage communication about those issues, or to build relationships with the public's involvement in the iss ues (Curtin, 1999; Miller, 2010). From a corporate-stakeholder relationship perspective, corporate social responsibility has also become one of the most prominent concerns for corporations (Bhattacharya et al., 2008; Clark, 2000). Previous research has s uggested that corporat ions are concerned more about their commitment towards the well-being of a society or a community, and these corporate social responsibility activities affect the relationshi p quality or reputation outcomes of the corporation (Bhattacharya et al., 2008; Kotler & Lee, 2004). Moreover, corporate social responsibility activi ties have been suggested as one of the measurement items of the overall corporate reputati on (Hutton et al., 2001). Emphasizing the role of issues in a corporate communication context, this study explored the role of public relations in a ffecting the perceived corporate reputation through communication strategies. This study particularly explored communication strategies grounded in an agenda-building fram ework. Miller (2010) emphasized the role of public relations practitioners in in forming and persuading the public about certain social issues. When an issue is successfully advocated, the organization (or corporation) involved in the issue would receive favorable evaluations (e.g., higher approval rating) by the public (Miller, 2010) From a public relations perspective, 126

PAGE 127

communication strategies have played a signif icant role in educating and informing the public about the involvement of corporations regarding certain issues. Scholars have explored the role of media exposure in affe cting corporate reputat ion, suggesting that the news media are among the most important communication channels (Wartick, 1992). Previous research has found some evidence of the role of the mass media in shifting the public's perceived corporat e reputation (Eyestone, 1978; Fombrun & Shanley, 1990; Weinberger & Romeo, 1989). Weinberger and Romeo (1989) found that negative media exposure of a corporation w ould certainly not enhance its reputation (Weinberger & Romeo, 1989). On the other hand, Fombrun and Shanley (1990) found that the visibility of a corporation in the media affects corporate reputation negatively no matter how the corporation is portrayed in media coverage. Scholars have also pointed out that the relationship bet ween media exposure and corporate reputation may be mediated by several factors, su ch as history or credibility (G riffin et al., 1991). Hence, evidence of the relationship between media ex posure and corporate reputation is still underdeveloped. The results of this study have also added empirical evidence of the role of communication exposure in affecting publ ic relations outcomes. Even though the negative tone of an issue would affect attribut e salience more than a positive tone, the results showed that a positive tone of an issue would have more influence on issue salience (mediated by emotion) and the perceived level of trust. In the second-level agenda-building relationship, a negative tone of an issue had a stronger impact on the public's level of attribute salience, while in the compelling arguments relationship (between the second-level and fi rst-level agenda building), the role of a positive tone of 127

PAGE 128

an issue was more significant. Moreover, the results showed that the compelling arguments relationship was mediat ed by emotional appeal. In this experimental study, participants in the positive tone of an issue reported a higher emotional appeal than those in the negative tone of an issue. Also, this study defined the two types of emotional appeal, such as pleasure and pain, based on factor analysis. The role of the mediating factor of the emotional pleasure was found in affecting perceived trus t, while the role of emotional pain was found in affecting the issue salience and primi ng effect. The emotional pain (index of anger, guilt, and sadness) showed more impact in drawing attention to the issue. The effect of emotional pleasure (index of happiness, gratitude, and pride) is limited on issue attention; however, the results showed more effects on public opinion outcomes, such as trust. When designing communication strategies using emotional appeal, practitioners need to carefully consider their purpose of the communication. For example, they should think about whether or not their major goal of communication is to draw the public's attention to the issue it self. Even though the relationship should be explored further, this study suggested that practitioners should bear in mind how to deal with issues, not only what issues should be discussed. This study emphasized the role of affective attributes in public relations me ssages in affecting public relations outcomes, such as trust and reputation. Hence, the results of the study suggest t hat practitioners should monitor the issuerelated discussions in media coverage and am ong the public closely. Corporate social responsibility became one of the important criteria for overa ll corporate reputation. The results showed that the public's evaluations of corporate social actions are affected not 128

PAGE 129

only by the corporations one-time actions but also by its long-term effectiveness and the outcomes of various actions. To meet their goal, corporations need communication strategies to inform the public not only about what they are doing but also how the effects are portrayed in m edia coverage and public opin ion. Better public opinion outcomes follow when the pub lic understands the corporat ion's concern about the effectiveness and outcomes of its corporate acti ons. In other words, the public relations' goal can be reached when a corporati on becomes concerned about long-term communication outcomes instead of a super ficial symbolic corporate action. Communication strategies in this current study can be criticized if it means deceiving or manipulating the public about the r eality of the corporat ion. For example, corporate image, as one of t he domains of public relations research, has been criticized due to its negative connotati on. The perception is that attempts to develop a good image would mean deceiving the public about t he corporation (Grunig, 2003). Hence, the results of this study sugges ted that the role of public re lations practitioners would be to monitor how the media cover the issues, to listen to what the public says about the issues, and to communicate transparently, no matter whether the issues are portrayed negatively or positively. Also, this study suggested that the reciprocal communication efforts with the media and t he public should be expanded in future agenda-building research studies. In addition to affective attributes, this study also explored the role of issue ownership in corporate messages. In a po litical communication context, scholars have pointed out that issue-owned political parties would receive more attention from the public, and they would receive more electoral gains when the issue is primed well 129

PAGE 130

among the public (e.g., Kleinnijenhuis et al ., 2001; Petrocik, 1996; Sheafer & Weimann, 2005). Even though no direct effects were s upported, this study found that issue ownership has a significant indirect effe ct on the agenda-building process. Also, the path analysis results have suggested that perceived issue ownership would have an influence on the public's perceived trust and corporate reputation outcomes. When considering a corporation competent on cert ain social issues, the public would have a higher emotional appeal leading to str onger issue salience and priming effects. Consequently, the emotional appeal also affe cts further public relations outcomes, such as trust and corporate reputation. This resu lt would suggest that public relations practitioners can benefit their respective co rporations by educating the public about the corporations professionalism on certain social issues and by showing continued social responsibility. On a practical scale, this result would provide recommendations to practitioners about how to choose relevant soci al actions for their corporations. As the results suggested, corporate reputation is directly and in directly affected by the perceived issue ownership, and the effects woul d increase when corporations have a good fit with their supporting social causes. This current study tested the role of iss ue ownership when a corporation's socially responsible actions reflected the industry category of the corporations. For example, a technology-based corporation, created for an experimental study, had issue ownership when its technology was used for a social caus e. The results of this study showed that issue ownership could be viewed as an a ssociation between the industry category and social actions of a corporation, and issue ownership has a consequent effect on public 130

PAGE 131

opinion toward the corporation. Still, the c oncept of issue ownership is somewhat vague (Petrocik et al., 2003), and what issues a co rporation should become involved in and how the corporation should take action have not yet been fully studied (Meijer & Kleinnijenhuis, 2006). Hence, applying the concept of issue ownership, this study has expanded the understanding of the role of issues in an agenda-building process. Limitations and Suggested for Future Research This study showed that the agenda-buildi ng process and effects can be found not only in a political communication context but also in a corporate communication context. Emphasizing the role of issues in corporat e communication, the purpose of this study was to explore a compelling argument demonstrating the re lationship between attributelevel and object-level of agenda building. So me indirect relationships were found through emotional appeal based on the compelli ng arguments hypotheses. In spite of its theoretical and practical contributions to the field, the st udy is not free from certain limitations. First, the study suggested the priming effe ct, trust, and corporate reputation as public opinion outcomes of agenda building. Through an experimental design, the effects were measured after a one-time ex posure of a public relations message. However, some public relations outcomes may need a longer time period to be examined. Trust, for instance, has been suggested as one of the relationship quality outcomes, including satisfaction and commitmen t. The relationship perspective of public relations has been emphasized in longterm relationship-building between an organization and its publics. Hence, future research may need to conduct a longitudinal design study to measure long-term relationship quality outcomes. 131

PAGE 132

To explore the role of issues in a co rporate communication context, this study suggested issue ownership as an important predictor of agenda salience and public relations outcomes. Issue ownership is about a companys capability to deal with certain social issues, and this experimental study designed public relations messages with a fictitious firm. The firm wa s a technology-based business, and a global health issue was selected to manipulate messages. Emphasizing the importance of the corporate social responsibility activities, the role of issues has become significant in affecting corporate r eputation. However, an understanding of the direction and the strength of the re lationship is much more complicated. In this study, an interaction ef fect between issue ownership and issue tone has been explored, and the role of emot ional appeal has been examined. However, further evidence should be conduct ed to test the effects by t he type of industry, size of the company, or other si tuational conditions. To explore the causal relationships of the agenda-building proc ess, the main goal of the project was to demonstrate rela tionships among antecedent, agenda salience, and consequences of agenda building. Even though the direct effects were not predicted in the original hypotheses bet ween antecedents and consequences of agenda building, an additional path analysis tested both direct and indirect relationships among these three components (antecedents, sa lience, and consequences). Few previous studies have attempted to demonstrate caus al relationships in an agenda-building framework, and more empirical evidence shoul d be added. Future research studies may apply this framework to other communication settings. 132

PAGE 133

This study concentrated on exploring t he role of affect in the agenda-building process. Traditionally, the va lence of objects (positive or negative) has been widely used in terms of affective attribute of agenda, and affective attributes have been less explored compared to substantive attributes (e .g., frames). In addi tion to the valence, this study suggested emotion as another dimens ion of affects, and the results of the study expanded our understanding of the role of emotion in the agenda-building process. More empirical evidence needs to be added. Also, future research should be developed in terms of the diverse measurement of emotion. In addition to a single excitement measure, this study measured different ty pes of emotion, such as anger, happiness, sadness, or pride. Two emoti onal factors were then suggested. Hence, future research will be added to confirm the role of these emotional factors. Finally, exploring the mediating role of the emotional factor s, the study also contributed to our under standing of a contingent condition of agenda-setting and agenda-building processes. Scholars have expl ored in what conditions agenda-setting or agenda-building effects would be found (McCombs, 2004; Weaver, 1980; Wu & Coleman, 2009). Among the vari ous contingent conditions, the emotional factors have also been emphasized by prior scholars. Wu and Coleman (2009) found that negative information would have stronger agenda-setting e ffects. Including the emotional factors, future studies should test more conti ngent conditions of t he agenda-building process (e.g., cultures, prior issue propositions or expectations for corporate social responsibility). For example, McCombs (2004) suggested that the need for orientation is the most critical conditi on for the agenda-setting process. Weaver (1980) said that people who actively seek information woul d be more susceptible to agenda-setting 133

PAGE 134

134 effects. The future research would help to understand the agenda-building effects in an interactive media environment where the active role of the audience is emphasized. Conclusions To examine the role of issues in public re lations, this study focused on the role of public relations communication strategi es grounded in an agenda-building framework. Particularly, this study ex amined 1) the relationship bet ween the independent variables (issue ownership and issue attributes) and the mediator (emotion), 2) the relationship between the mediator (emotional arousal) and the dependent variables associated with agenda salience (issue and attribute salienc e), 3) the relationship between the independent variables (issue ownership and issue attributes) and the dependent variables associated with agenda salience (issue and attribute salience), and 4) the relationship between the dependent variables associated with agenda salience (issue and attribute salience) and with public opin ion (priming effect, trust, and corporate reputation). Twenty hypotheses were proposed to test the relationships and seven of them were supported. Also, a relationship was found in an opposite direction from the prediction (Table 4-12 in Chapter 4). As a result, major contributio ns of this study demonstrated causal relationships in an agenda-building framework to apply this framework into a corporate communication context, to find empirical evidence of the relationship between message exposure and co rporate reputation, and to expand our understanding of the role of attributes testing compelling arguments hypotheses. Furthermore, emphasizing the role of affect s in the agenda-building relationship, both valence and arousal level of affects have been explored. Finally, the mediating role of emotion was discussed.

PAGE 135

APPENDIX A STATEMENT OF INFORMED CONSENT Dear participants: I am currently conducting an online study to explore the linkages between corporate messages and public opinion about the corpor ation. In particular, you will make evaluations on the corporation after reading corporate messages. Please take your time to read the corporate site, so you r ead all the given messages carefully. This experimental study will take about 10 minutes. Your response is extremely important and valuable. Your answers will be kept for statistical purposes only and your identity will be kept c onfidential to the extent provided by law. You have the right to withdraw from the study at anytime without consequence. If you have any question about this survey, please contact the researcher at jy.kim@ufl.edu. If you have any questions about your rights as a study participant, please contact the University of Florida Inst itutional Review Boar d at (352) 392-0433. Thank you for your time and help in advance. Ji Young Kim Ph.D. Candidate College of Journalism and Communications University of Florida jy.kim@ufl.edu 135

PAGE 136

APPENDIX B EXPERIMENTAL STIMULI 136

PAGE 137

Issue ownership and positive issue attribute 137

PAGE 138

Issue ownership and negative issue attribute 138

PAGE 139

None issue ownership and positive issue attribute 139

PAGE 140

None issue ownership and negative issue attribute 140

PAGE 141

APPENDIX C QUESTIONNAIRE Section 1. Salience of issue, issue attribute, and priming 1. Now, you have read the companys mess ages. Id like ask your opinion again about global health issues. Please identify your answers with each statement ranging from 1 = Not at all to 7 = Extremely. Not at all Moderately Extremely In your opinion, how important is global health issues today? 1 2 3 4 5 6 7 How many news reports about global health issues do you regularly pay attention to? 1 2 3 4 5 6 7 To what extent do you think global health issues is deserving of additional government action? 1 2 3 4 5 6 7 How often do you talk about global health issues in your everyday conversation? 1 2 3 4 5 6 7 2. Also, please identify your general belief about the issue again: I feel that global health issues are. Strongly disagree neutral Strongly agree Negative 1 2 3 4 5 6 7 Positive Unpleasant 1 2 3 4 5 6 7 Pleasant 3. Please identify your answer with the following statement again ranging from 1 = Extremely Oppose to 7 = Extremely Support. Extremely oppose Moderately Extremely support Do you support or oppose the efforts to solve global health issues? 1 2 3 4 5 6 7 141

PAGE 142

Section 2. Emotion 1. In this section, Id like to ask your feeling after readi ng the messages. Please identify your answer with the follo wing statement ranging from 1 = Not at all to 7 = Extremely. Not at all Moderately Extremely I feel excited 1 2 3 4 5 6 7 2. The following question asks more details on how you feel. Id like to ask your feeling after reading the public relations messages Please identify your answers with each statement ranging from 1 = Not at all to 7 = Extremely. Not at all Moderately Extremely I feel angry 1 2 3 4 5 6 7 I feel mad 1 2 3 4 5 6 7 I feel irritated 1 2 3 4 5 6 7 I feel appreciative 1 2 3 4 5 6 7 I feel grateful 1 2 3 4 5 6 7 I feel thankful 1 2 3 4 5 6 7 I feel guilty 1 2 3 4 5 6 7 I feel remorseful 1 2 3 4 5 6 7 I feel sorry 1 2 3 4 5 6 7 I feel proud 1 2 3 4 5 6 7 I feel self-fulfilled 1 2 3 4 5 6 7 I feel joyful 1 2 3 4 5 6 7 I feel happy 1 2 3 4 5 6 7 I feel elated 1 2 3 4 5 6 7 I feel gloomy 1 2 3 4 5 6 7 I feel sad 1 2 3 4 5 6 7 I feel upset 1 2 3 4 5 6 7 142

PAGE 143

Section 3. Trust In this section, I ask how you feel about the companys relationship with their stakeholders. Nine statements are listed below. Please identify your level of agreement or disagreement with each statement. Strongly disagree neutral Strongly agree This company would treat customers fairly. 1 2 3 4 5 6 7 Whenever this company makes an important decision, I know it will be concerned about its customers. 1 2 3 4 5 6 7 This company can be relied on to keep its promises. 1 2 3 4 5 6 7 I believe that this company takes the opinions of customers into account when making decisions. 1 2 3 4 5 6 7 I feel very confident about this company's skills. 1 2 3 4 5 6 7 This company has the ability to accomplish what it says it will do. 1 2 3 4 5 6 7 I believe sound principles guide the companys behavior. 1 2 3 4 5 6 7 I am willing to let this company make decisions for customers like me. 1 2 3 4 5 6 7 I believe the company does not mislead customers. 1 2 3 4 5 6 7 143

PAGE 144

Section4. Corporate reputation In this section, I ask how you feel about the companys r eputation with their stakeholders. 20 statements are listed below. Please identify y our level of agreement or disagreement with each statement. Strongly disagree neutral Strongly agree The company has market opportunities 1 2 3 4 5 6 7 The company has excellent leadership 1 2 3 4 5 6 7 The company has clear vision for the future 1 2 3 4 5 6 7 The company supports good causes 1 2 3 4 5 6 7 The company is environmentally responsible 1 2 3 4 5 6 7 The company is responsible in the community 1 2 3 4 5 6 7 I feel good about company 1 2 3 4 5 6 7 The company inspires admiration and respect 1 2 3 4 5 6 7 The company inspires trust 1 2 3 4 5 6 7 The company has high quality products and/or services 1 2 3 4 5 6 7 The company has innovative products and/or services 1 2 3 4 5 6 7 The company provides good value for money 1 2 3 4 5 6 7 The company stands behind its products and/or services 1 2 3 4 5 6 7 The company rewards employees fairly 1 2 3 4 5 6 7 144

PAGE 145

Strongly disagree neutral Strongly agree The company is a good place to work 1 2 3 4 5 6 7 The company has good employees 1 2 3 4 5 6 7 The company outperforms competitors 1 2 3 4 5 6 7 The company has record of bring profitable 1 2 3 4 5 6 7 The company is low risk investment 1 2 3 4 5 6 7 The companys growth prospects 1 2 3 4 5 6 7 Section 5. Mani pulation items 1. Regarding the company's prior commitment activities you read (on the left column of the previous page), please indicate your le vel of agreement or di sagreement with each statement about the company described on the messages. Extremely little neutral Extremely well How well do you think that the company handles global health issues? 1 2 3 4 5 6 7 How well do you think that the company can solve problems related to global health issues? 1 2 3 4 5 6 7 I believe that the company is very knowledgeable about global health issues. 1 2 3 4 5 6 7 2. The message described what news media and people say about global health issues. How was the issues described on the messages? Negatively 1 2 3 4 5 6 7 Positively Unpleasantly 1 2 3 4 5 6 7 Pleasantly 145

PAGE 146

Section 6. Demographic information Now, the following questions ask demographic information. 1. In what year were you born? ________________ 2. What is your gender? Male Female 3. How would you describe yourself? Hispanic or Latino Asian African American Native Hawaiian or other Pacific Islander White 4. What is your religious affiliation? Protestant Christian Roman Catholic Evangelical Christian Jewish Muslim Hindu Buddhist Other: _________________ 5. Where is your country of origin? ________________ ____________ 6. What is your highest leve l of school you have completed? If currently enrolled, mark the previous grade or hi ghest degree received. No schooling completed Nursery school to 12th grade (no diploma) High school graduate Associates degree (e.g., AA) Bachelors degree (e.g., BA or BS) Masters degree (e.g., MA, MS, or MBA) Professional degree (e.g., MD or JD) Doctoral degree (e.g., PhD or EdD) 7. What is your occupation? Employed for wages Self-employed Out of work A homemaker A student Retired Unable to work 146

PAGE 147

8. How would you describe your employer? For-profit company Not-for-profit organization Local government State government Federal government Self-employed Others (Specif y:___________ ____________) 9. What is your average income level? Less than $10,000 $10,000 to $29,999 $30,000 to $49,999 $50,000 to $69,999 $70,000 to $89,999 $90,000 to $109,999 $110,000 to $129,999 $130,000 or more 10. What role do you believe issue-related information (i.e., global health) plays in your attitudes, beliefs, and perc eption about the company? ________________________ _______________________ _______________ ________________________ _____________________ _________________ ________________________ _____________________ _________________ ________________________ _____________________ _________________ ________________________ _______________________ _______________ 11. If you have additional co mments or thoughts on this study please list them below. ________________________ _____________________ _________________ ________________________ _____________________ _________________ ________________________ _______________________ _______________ ________________________ _____________________ _________________ ________________________ _____________________ _________________ Thank you so much for your participation! 147

PAGE 148

LIST OF REFERENCES Abelson, R. P. (1995). Attitude extremity. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 25-42). Mahwah, NJ: Lawrence Erlbaum. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interaction. Thousand Oaks, CA: Sage. Andsager, J., & Smiley, L. (1998). Eval uating the public information: Shaping news coverage of the silic one implant controversy. Public Relations Review, 24 (2), 183-201. Arnold, V. D. (1985). T he importance of pathos in persuasive appeals. The Bulletin 2627. Aula, P., & Mantere, S. (2008). Strategic reputation managem ent towards a company of good. London: Routledge. Babbie, E. R. (2004). The practice of social research Belmont, CA: Wadsworth. Balmas, M., & Sheafer, T. (2010). Candidate image in election campaigns: Attribute agenda setting, affective priming, and voting intentions. International Journal of Public Opinion Research, 22 (2), 204-229. Batra, R. (1986). Affective advertising: Role, process, and measurement. In R. A. Peterson, W. D. Hoyer, & W. R. Wilson (Eds.). The role of affect in consumer behavior: Emerging theories and applications (pp. 53-85). Lexington, MA: Lexington Books. Behr, R. L., & Iyengar, S. (1985) Television news, real-world cues, and changes in the public agenda. Public Opinion Quarterly, 49 (1), 38-57. Belanger, E., & Meguid, B. M. (2008). Issue salience, issue ownership, and issue-based vote choice. Electoral Studies, 27 477-491. Berger, B. K. (2001). Privat e issues and public policy: Locating the corporate agenda in agenda-setting theory. Journal of Public Re lations Research, 13 (2), 91-126. Berger, B. K., Hertog, J. K. & Park, D. J. (2002). The po litical role and influence of business organizations: A communication perspective. In W. B. Gudykunst (Ed.), Communication yearbook (Vol 26, pp. 160-200). Mahwah, NJ: Lawrence Erlbaum. Bernstein, D. (1984). Company image and reality. A critique of corporate communications New York: Holt, Rinehart and Winston. 148

PAGE 149

Best, J. (1995). Images of issues: Typifying contemporary social problems New York: Aldine de Gruyter. Bhattacharya, C. B., Korschun, D., & Sen, S. (2008). Str engthening stakeholdercompany relationships through mutually be neficial corporate so cial responsibility initiatives. Journal of Business Ethics, 85 257-272. Blumer, H. (1966). The mass, the public, and public opinion. In B. Berelson & M. Janowitz (Eds.), Reader in public opinion and communication (pp. 43-50). New York: Free Press. Botan, C. H., & Taylor, M. (2004). P ublic relations: State of the field. Journal of Communication, 54 (4), 645661. Bromley, D. B. (1993). Reputation, image, and impression management West Sussex, England: Wiley. Budge, I., & Farlie, D. J. (1983). Explaining and predicting elec tions: Issue effects and party strategies in twenty-three democracies London: George Allen and Unwin. Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazons Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6 (1), 3-5. Cameron, G. T., Sallot, L. M., & Curtin, P. A. (1997). Public relations and the production of news: A critical review and theoretical framework. Communication Yearbook, 20, 111-155. Carroll, C. (2004). How the mass media influ ence perceptions of corporate reputation: Exploring agenda-setting e ffects within business news coverage. Ph.D. dissertation. The University of Texas at Austin, Austin, TX. Carroll, C. E., & McCombs, M. E. (2003). Agenda-setting effects of business news on the publics images and opinions about major corporations. Corporate Reputation Review, 6 (1), 36-46. Cha, H., Song, Y., & Kim, J. R. (2010). Effects of issue ownership and issue obtrusiveness on corporate reputati on at two Korean corporations. Public Relations Review, 36 289-291. Clark, C. E. (2000). Differences between public relations and corporate social responsibility: An analysis. Public Relations Review, 26 (3), 363-380. Clore, G. L., Schwarz, N., & Conway, M. (1994). Affective causes and consequences of social information processing. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition: Basic processes (pp. 323-419). Hillsdale, NJ: Lawrence Erlbaum Associates. 149

PAGE 150

Cobb, R. A., & Elder, C. D. (1972). Participation in American politics: The dynamics of agenda building Boston: Allyn & Bacon. Cohen, B. C. (1963). The press, the public and foreign policy Princeton, NJ: Princeton University Press. Coleman, L. S. (1982). The asymmetric society New York: Syracuse University Press. Coleman, R., & Wu, H. D. (2010). Proposing emotion as a dimension of affective agenda setting: Separating affect into two components and comparing their second-level effects. Journalism and Mass Communication Quarterly, 87 (2), 315327. Cornelissen, J. P. (2000). Toward an under standing of the use of academic theories in public relations practice. Public Relations Review, 26 (3), 315-326. Curtin, P. A. (1999). Reevaluating public rela tions information subsidies: Market driven journalism and agenda-buildi ng theory and practice. Journal of Public Relations Research, 11 (1), 53-90. Cutlip, S. M., Center, A. H., & Broom, G. M. (1994). Effective public relations Englewood Cliffs, NJ: Prentice Hall. Dearing, J. W., & Rogers, E. M. (1996). Agenda setting London: Sage. Deephouse, D. L. (2000). Medi a reputation as a strategic re source: An integration of mass communication and resource-based theories. Journal of Management, 26(6), 1091-1112. Dewey, J. (1927). The public and its problems Oxford, England: Holt. Dillard, J. P., & Wilson, B. J. (1993). Co mmunication and affect: Thoughts, feelings, and issues for the future. Communication Research, 20, 637-646. Domhoff, G. W. (1970). The higher circles New York: Random House. Domhoff, G. W. (1979). The power that be New York: Random House. Dozier, D. M., Grunig, L. A., & Grunig, J. E. (1995). Managers guide to excellence in public relations and communication management Mahwah, NJ: Lawrence Erlbaum Associates. Dunn, J. R., & Schweitzer, M. E. (2005). Feel ing and believing: The influence of emotion on trust. Journal of Personality and Social Psychology, 88 (5), 736-748. Elmelund-Praestekaer, C. (2011). Issue ow nership as a determinant of negative campaigning. International Political Science Review, 32 (2), 209-221. 150

PAGE 151

Entman, R. (1993). Framing: Toward cl arification of a fr actured paradigm. Journal of Communication, 43 (4), 51-58. Erdfelder, E., Faul, F., & Buchner, A. (1996). GPOWER: A general power analysis program. Behavior Research Methods, In struments, and Computers, 28 1-11. Eyestone, R. (1979). From social issues to public policy New York: John Wiley & Sons. Ferguson, M. A. (1984). Building theory in public rela tions: Interorganizational relationships. Paper presented to the Association for Education in Journalism and Mass Communication, Gainesville, FL. Fombrun, C. J. (1996). Reputation: Realizing value from the corporate image Boston: Harvard Business School Press. Fombrun, C. J. (1998). Indice s of corporate reput ation: An analysis of media rankings and social monitors ratings. Corporate Reputation Review, 1 327-340. Fombrun, C. J., Gardberg, N. A., & Server, J. M. (2000) The reputation quotient: A multi-stake-holder measure of corporate reputation. Journal of Brand Management, 7 (4), 241-255. Fombrun, C., & Shanley, M (1990). What s in a name? Reputation building and corporate strategy. Academy of Manage ment Journal, 33 233-256. Forgas, J. P. (1995). Mood and judgment: The affect infusion model. Psychological Bulletin, 117 (1), 39-66. Freeman, R. E. (1984). Strategic management: A stakeholder approach Boston: Pitman. Gandy, O. H. (1982). Beyond agenda setting: Information subsidies and public policy Norwood, NJ: Ablex. Gans, H. J. (2003). Democracy and the news New York: Oxford University Press, Inc. Ghanem, S. (1996). Media coverage of crime and public opinion: An exploration of the second level of agenda setting. Ph.D. dissert ation. The University of Texas at Austin, Austin, TX. Ghanem, S. (1997). Filling in the tapestry: The second level of agenda setting. In M. McCombs, D. Shaw & D. Weaver (Ed.), Communication and democracy (pp. 314). Mahwah, NJ: Lawrence Erlbaum. Goffman, E. (1959). The presentation of self in everyday life New York: Overlook. 151

PAGE 152

Golan, G., & Wanta, W. (2001). Second-level agenda setti ng in the New Hampshire Primary: A comparison of coverage in three newspapers and public perceptions of candidates. Journalism and Mass Communications 78 (2), 247-259. Gorn, G., Pham, M. T., & Sin, L. Y. ( 2001). When arousal influences ad evaluation and valence does not (and vice versa). Journal of Consumer Psychology, 11 (1), 4355. Gotsi, M., & Wilson, A. M. (2001). Cor porate reputation: Seeking a definition. Corporate Communications: An international Journal, 6 (1), 24-30. Gray, E. R., & Balmer, J. M. T. (1998). Managing image and corporate reputation. Long Range Planning, 31 (5), 685-692. Green, J., & Hobolt, S. B. ( 2008). Owning the issue agenda: Pa rty strategies and vote choices in British election. Electoral Studies, 27 460-476. Griffin, M. B., Babin, J., & At taway, J. S. (1991). An empiri cal investigation of the impact of negative public publicity on consumer attitudes and intentions. Advances in Consumer Research, 18 334-341. Grunig, J. E. (1979). A new measure of public opinions on corporate social responsibility. Academy of Management Journal, 22 (4), 738-764. Grunig, J. E. (2003). Image and substance: From symbolic to behavioral relationships. Public Relations Review, 19 (2), 121-139. Grunig, J. E. (2006). Furnis hing the edifice: Ongoing resear ch on public relations as a strategic management function. Journal of Public Relations Research, 18 (2), 151-176. Grunig, J. E., & Huang, Y. H. (2000). From organizational effectiveness to relationship indicators: Antecedents of relationshi ps, public relations strategies, and relationship outcomes. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to public relations (pp. 23-53). Mahwah, NJ: Lawrenc e Erlbaum Associates, Inc. Grunig, J. E., & Repper, F. (1992). Strategic management, publics, and issues. In J. E. Grunig (Ed.), Excellence in public relations and communication management: Contributions to effective organizations (pp. 117-158). Hills dale, NJ: Lawrence Erlbaum Associates. Grunig, J. E., & White, J. ( 1992). The effect of worldviews on public relations theory and practice. In J. E. Grunig, D. M. Dozier, W. P. Ehling, L. A. Grunig, F. C. Repper, & J. White (Eds.). Excellence in public relations and communication management (pp. 31-64). Hillsdale, NJ: Lawrence Erlbaum Associates. 152

PAGE 153

Grunig, J. E., Grunig, L. A., & Ehling, W. P. (1992). What is an effective organization. In J. E. Grunig (Ed.), Excellence in public rela tions and communication management (pp. 65-90). Hillsdale, NJ: Lawrence Erlbaum Associates. Grunig, L., Grunig, J. & Dozier, D (2002). Excellence in public relations and effective organizations: A study of communica tion management in three countries Mahwah, NJ: Lawrence Erlbaum Associates. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis (6th Ed.). Upper Saddle River, NJ: Pearson. Hallahan, K. (1999). Seven models of fram ing. Implications for public relations. Journal of Public Relations Research, 11 205-242. Hallahan, K. (2001). The dynamics of iss ue activation and response: An issues processes model. Journal of Public Relations Research, 13 27-59. Harrison, J. S., & Freeman, R. E. (1999) Stakeholders, social responsibility, and performance: Empirical evidence and theoretical perspectives. Academy of Management Journal, 42 (5), 479-485. Heath, R. (2006). Onward into more f og: thoughts on public relations research directions. Journal of Public Relations Research, 18 93-114. Heath, R., & Cousino, K. R. (1990). Issue management: E nd of first decade progress report. Public Relations Review, 16 (1), 6-18. Hilgartner, S., & Bosk, C. L. (1988). The rise and fall of social problems: A public arenas model. American Journal of Sociology, 94 53-78. Hirt, E. R., Levine, G. M., McD onald, H. E., Melton, R. J., & Martin, L. L. (1997). The role of mood in quantitative and qualitativ e aspects of performance: Single or multiple mechanisms? Journal of Experimental Social Psychology, 33 (6), 602629. Hon, L. C. (1997). What have you done for me lately? Exploring effectiveness in public relations. Journal of Public Relations Research, 9 (1), 1-30. Hon, L. C., & Gruni g, J. E. (1999). Guidelines for measuring relationships in public relations Gainesville, FL: The Institute for Public Relations. Horton, J., & Chilton, L. (i n press). The labor economics of paid crowdsourcing. Proceedings of the 11th ACM Conference on Electronic Commerce, Cambridge, MA. Hu, L., & Bentler, P. M. (1999). Cutoff criter ia for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidi sciplinary Journal, 6 (1), 1-55. 153

PAGE 154

Huang, Y. H. (1998). Public relations strategies and organization-public relationships Paper presented at the annual conference of the Association for Education in Journalism and Mass Communica tion, Baltimore. Huang, Y. H. (1999). The effects of public relations st rategies on conflict management. Paper presented at the 49th annual conference of the International Communication Association, San Francisco. Hutton, J. G., Goodman, M. B., Alexander, J. B, & Genest, C. M. (2001). Reputation management: The new face of corporate public relations? Public Relations Review, 27, 247-261. Iacobucci, D. (2008). Mediation analysis Thousand Oaks, CA: Sage. Iacobucci, D. (2010). Structural equations modeling: Fit indices, sample size, and advanced topics. Journal of Consumer Psychology, 20 90-98. Isaacs, H. R. (1972). Images of Asia: American view of China and India New York: Harper Torchbooks. Iyengar, S., & Kinder, D. R. (1987). News that matters: Television and American opinion Chicago: University of Chicago Press. Iyengar, S., & Simon, A. (1993). News coverage of the Gulf cr isis and public opinion: A study of agenda-setting, priming, and framing. Communication Research, 20 365-383. Iyengar, S., Peters, M. D., & Kinder, D. R. (1982). Experiment al demonstrations of the "not-so-minimal" consequences of television news programs. The American Political Science Review 76(4), 848-858. Izard, C. E. (1993). Four systems for emotion activation: Cognitive and noncognitive processes. Psychological Review, 100 68-90. Judd, C. M., & Johnson, J. T. (1981). Attitu de, polarization, and diagnosticity: Exploring the effects of affect. Journal of Personality and Social Psychology, 41 25-36. Kaid, L. L. (1976). Newspaper treatment of a candidates news releases. Journalism Quarterly, 53 135-137. Keller, P. A., & Block, L. G. (1996). Increasing the persuas iveness of fear appeals: The effect of arousal and elaboration. Journal of Consumer Research, 22 448-459. Keppel, G., & Wickens, T. D. (2004). Design and analysis of a researchers handbook (4th Ed.). Upper Saddle River, NJ: Pearson Inc. 154

PAGE 155

Kepplinger, H., Donsbach, W., Brosius, H ., & Staab, J. (1989). Media tone and public opinion: A longitudinal study of media co verage and public opinion on Chancellor Helmut Kohl. International Journal of Public Opinion Research, 1 326-342. Ki, E. J., & Hon, L. (20 09). A measure of relationshi p cultivation strategies. Journal of Public Relations Research, 21 (1), 1-24. Kim, K., & McCombs, M. ( 2007). News story descriptions and the public opinions of political candidates. Journalism and Mass Communication Quarterly, 84 299-314. Kim, S. H., Scheufele, D. A., & Shanahan, J. (2002). Agenda se tting, priming, framing, and second-levels in local politics. Journalism & Mass Communication Quarterly, 79(1), 7-25. King-Casas, B., Tomlin, D., Anen, C., Camerer, C., Quartz S. R., & Montague, P. R. (2005). Getting to know you: Reputati on and trust in a two-person economic exchange. Science, 308 (5718), 78-83. Kiousis, S. (2003). Job approval and favorability: The impact of media attention to the Monica Lewinsky Scandal on public op inion of President Bill Clinton. Mass Communication & Society, 6 (4), 435-451. Kiousis, S. (2005). Compelling argument and attitude strength: Exploring the impact of second-level agenda-setting on public opinion of presidential candidate images. Harvard International Journal of Press/Politics, 10, 3-27. Kiousis, S., & McDevitt, M. (2008). Agenda se tting in civic development: Effects of curricula and issue import ance on youth voter turnout. Communication Research, 35(4), 481-502. Kiousis, S., & Strmbck, J. (2010). The Wh ite House and public relations: Examining the linkages between presidential communications and public opinion. Public Relations Review, 36, 7-14. Kiousis, S., & Wu, X. (2008). Inter national agenda-buildi ng and agenda-setting. International Communication Gazette, 70 (1), 58-75. Kiousis, S., Bantimaroudis, P., & Ban, H. (1999). Candidate image attributes: Experiments on the subst antive dimension of second level agenda setting. Communication Research 26(4), 414-428. Kiousis, S., Mitrook, M., Wu, X., & Seltzer, T. (2006). Firstand second-level agendabuilding and agenda-setting effects: Exploring the linkages among candidate news releases, media coverage, and publ ic opinion during the 2002 Florida gubernatorial election. Journal of Public Relations Research 18, 265-285. Kiousis, S., Popescu, C., & Mitrook, M. ( 2007). Understanding influence on corporate reputation: An examination of public re lations efforts, media coverage, public 155

PAGE 156

opinion, and financial performance fr om an agenda-building and agenda-setting perspective. Journal of Public Relations Research, 19 (2), 147-165. Kleinnijenhuis, J., Maurer, M., Kepplinger, H. M., Oegem a, E. (2001). Issues and personalities in German and Dutch televi sion news: Patterns and effects. European Journal of Communication, 16 337-359. Kline, R. B. (2005). Principles and practices of structural equation modeling New York: Guilford. Kotler, P., & Lee, N. (2004). Corporate social responsibil ity: Doing the most good for your company and your cause Hoboken, NJ: Wiley & Sons, Inc. Krosnick, J. A., & Kinder, D. R. (1990). Al tering the foundations of support for the president through priming. American Political Science Review, 84, 497-512. Kunczik, M. (1997). Images of nations and inte rnational public relations Mahwah, NJ: Erlbaum. Kunczik, M. (2003). States, international organizations, and the news media: Problems of image cultivation. In P. J. Maareck & G. Wolfsfeld (Eds.), Political communication in a new era: A cross-national perspective (pp. 117-138). London: Routledge. Lang, A., Dhillon, K., & Dong, Q. (1995). T he effects of emotional arousal and valence on television viewers cogni tive capacity and memory. Journal of Broadcasting & Electronic Media, 39 (3), 313-327. Lang, G. E., & Lang, K. (1966). The mass medi a and voting. In B. Berelson & M. Janowitz (Eds.), Reader in public communication (pp. 455-472). New York: Free Press. Lang, G. E., & Lang, K. (1981). Watergate: An exploration of the agenda building process. In G. C. Wilhoit & H. de Bock (Eds.), Mass communication review yearbook (Vol. 2, pp. 447-468). Beve rly Hills, CA: Sage. Langdridge, D. (2004). Introduction to research methods and data analysis in psychology Harlow, England: Pearson Prentice Hall. Lavigne, ., & Bourbonnais, R. (2010). Psyc hosocial work environment, interpersonal violence at work and psychotropic drug use among correctional officers. International Journal of Law and Psychiatry, 33 122-129. Ledingham, J. A. (2001). Govern ment-community relationships: extending the relational theory of public relations. Public Relations Review, 27 285-295. 156

PAGE 157

Ledingham, J. A., & Bruning, S. D. ( 1998). Relationship management and public relations: Dimensions of an organization-public relationship. Public Relations Review, 24 55-65. Ledingham, J. A., & Bruning, S. D. (2000a). A longitudinal st udy of organization-public relationship dimensions: Defining the role of communication in the practice of relationship management. In J. A. Ledi ngham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to public relations (pp. 55-70). Mahwah, NJ: Lawr ence Erlbaum Associates. Ledingham, J. A., & Bruning, S. D. (2000b). Background and current trends in the study of relationship management. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A re lational approach to public relations (pp. xi-xvii). Mahwah, NJ: Lawr ence Erlbaum Associates, Inc. Lee, G. (2010). Who let priming out? Analysis of firstand second-level agenda setting effects on priming. International Communication Gazette, 72(8), 759-776. Lee, R. M., & Yoo, H. C. ( 2004). Structure and measurement of ethnic identity for Asian Americans. Journal of Counseling Psychology, 51 263. Lopez-Escobar, E., Llamas, J. P., & McCo mbs, M. E. (1998a). Agenda setting and community consensus: First and sec ond level effects. International Journal of Public Opinion Research, 10( 4), 335-348. Lopez-Escobar, E., Llamas, J. P., McCombs, M., & Lennon, F. R. ( 1998b). Two levels of agenda-setting among advertising and new s in the 1995 Spanish elections. Political Communication, 15 (2), 225-238. Luoma-aho, V., & Vos, M. (2010). Toward s a more dynamic stakeholder model: Acknowledging multiple issue arenas. Corporate Communications: An International Journal, 15 (3), 315-331. Marcus, G., Neuman, W. R., MacKuen, & M. (2000). Affective intelligence and political judgment Chicago: University of Chicago Press. McCombs, M. (2004). Setting the agenda: The mass me dia and public opinion Malden, MA: Polity. McCombs, M., & Estrada, G. (1997): T he news media and the pictures in our heads, In S. Iyengar & R. Reeves (Eds.). Do the media govern? (pp. 237247). London: Age. McCombs, M., & Ghanem, S. (2001). The convergence of agenda setting and framing. In S. D. Reese, O. H. Gandy Jr., & A. E. Grant (Eds.), Framing public life: Perspectives on media and our understanding of the world (pp. 67-82). Mahwah, NJ: Lawrence Erlbaum. 157

PAGE 158

McCombs, M., Llamas, J., Lopez-Escobar E., and Rey, F. (1997). Candidate images in Spanish elections: Se cond-level agenda-setting effects. Journalism and Mass Communication Quarterly, 74 703-717. McCombs, M., Lopez-Escobar, E., & Llam as, J. P. (2000). Setting the agenda of attributes in the 1996 S panish general election. Journal of Communication, 50(2), 77-92. McCombs, M., & Reynolds, A. (2002). Ne ws influence on our pictures of the world. In J. Bryant & D. Zillman (Eds.), Media effects: Advances in theory and research (pp. 1-18). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. McCombs, M., & Shaw, D. L. (1972). The agenda-setting function of the mass media. Public Opinion Quarterly 36, 176-187. McCombs, M., & Shaw, D. L. (1993). The evolution of agenda-setting research: Twenty-five years in the marketplace of ideas, Journal of Communication, 43(2), 58-67. McKinnon, L. M., Tedesco, J. C., & Lauder, T. (2001). Polit ical power through public relations. In R. L. Heath (Ed.), Handbook of public relations (pp. 557-563). Thousand Oaks: Sage. McKnight, D. H., & Choudhury, V. (2001). Trus t and distrust definitions: One bite at a time, in R. Falcone, M. Singh, & Tan, Y. H. (Eds.), Trust in cyber-societies: Integrating the human and artificial perspectives (pp. 27-54). Berlin: Springer. McKnight, D. H., Choudhury, V., & Kacmar, C. (2002a). Developing and validating trust measures for e-Commerce: An integrative typology. Information Systems Research, 13 (3), 334-359. McKnight, D. H., Choudhury, V., & Kacmar, C. (2002b). The impact of initial consumer trust on intentions to transact with a Web site: A trust building model. Journal of Strategic Information Systems, 11 (3-4), 297-323. Meijer, M-M, & Kleinnijenhuis, J. (2006). Issue news and cor porate reputation: Applying the theories of agenda setting and issue owner ship in the field of corporate communication. Journal of Communication, 56 543-559. Miel, P., & Faris, R. (2008) News and information as digital media come of age, overview 2008. The Berkman Center for Internet & Society at Harvard University. Retrieved on October 2, 2011, from http://cyber.law.harvard.edu/sites/cyber.law.harvard.edu/files/Overview_MR.pdf Milgrom, P. R. (1981). Good news and bad news: Representation theorems and applications. Bell Journal of Economics, 12 (2), 380-391. 158

PAGE 159

Miller, B. M. (2010). Community st akeholders and marketplace advocacy: A model of advocacy, agenda buil ding, and industry approval. Journal of Public Relations Research 22(1), 85-112. Miller, J. M. (2007). Examining the medi ators of agenda setting: A new experimental paradigm reveals the role of emotions. Political Psychology, 28 (6), 689-717. Morris, J. P., Squires, N. K ., Taber, C. S., & Lodge, M. (200 3). Activation of political attitudes: A psychophysiological examinat ion of the hot cognition hypothesis. Political Psychology, 24 (4), 727-745. Muenchen, R. A., & Hil be, J. M. (2010). R for stata users New York: Springer. Ohl, C. M., Pincus, J. D ., Rimmer, T., & Harrison, D. (1995). Agenda building role of news releases in corporate takeovers. Public Relations Review, 21 89101. Olasky, M. N. (1987). Corporate public relations and Amer ican private enterprise: A new historical perspective Hillsdale, NJ: Lawrence Erlbaum. Olins, W. (1978). The corporate personality: An inqu iry into the nature of corporate identity London: Design Council. Paolacci, G., Chandler, J., & Ipeirotis, P. G. (2010). Ru nning experiments on Amazon Mechanical Turk. Judgment and Decision Making, 5 (5), 411-419. Petrocik, J. R. (1996). Iss ue ownership in presidentia l elections, with a 1980 case study. American Journal of Political Science, 40 825-850. Petrocik, J. R., Benoit, W. L., & H ansen, G. J. (2003). Issue ownership and presidential cam paigning, 1952-2000. Political Science Quarterly, 118 (4), 599-626. Phillips, D., & Young, P. (2009). Online public relations: A practical guide to developing an online strategy in the world of social media London: Kogan Page. Reber, B., & Cameron, G. T. (2003). Measur ing contingencies: Using scales to measure public relations practitioner limits to accommodation. Journalism and Mass Communication Quarterly, 80(2), 432-446. Renfro, W. L. (1993). Issues management in strategic planning Westport, CT: Quorum Books. Salwen, M. B., & Matera F. (1989). Public salience of foreign nations. Journalism Quarterly, 69 (3), 623-632. Schachter, S., & Singer, J. E. (1962). C ognitive, social, and physiological determinants of emotional state. Psychological Review, 69 379-399. 159

PAGE 160

Schattschneider, E. E. (1960). The semi-sovereign people: A realists view of democracy in America. New York: Holt, Rinehart, and Winston. Schoenbach, K., & Semetko, H. (1992). Agenda-setting, agenda-reinforcing or agendadeflating: A study of the 1990 German national election. Journalism Quarterly, 69 837-846. Seltzer, T., & Zhang, W. ( 2011). Toward a model of polit ical organization-public relationships: Antecedent and cultivation strategy influence on citizens relationships with political parties. Journal of Public Relations Research, 23 (1), 24-45. Semetko, H. A., Brzinski, J. B., Weaver, D., & Willnat, L. (1992). TV news and U.S. public opinion about foreign countries: T he impact of exposure and attention. International Journal of Public Opinion Research, 4 (1), 18-36. Sethi, S. P. (1979). Inst itutional/image advertising and idea/issue advertising as marketing tools: Some public policy issues. Journal of Marketing, 43 68-78. Severin, W. J., & T ankard, J. W. (2001). Communication theories: Origins, methods, and uses in the mass media (5th Ed.). San Francisco: Longman. Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52 (3/4), 591-611. Shaw, D. L., & Mc Combs, M. (1977). The emergence of American political issues: The agenda setting function of the press St. Paul, MN: West. Sheafer, T. (2007). How to evaluate it: T he role of story-evaluative tone in agenda setting and priming. Journal of Communication 57(1), 21-39. Sheafer, T., & Weimann, G. (2005). Agenda bu ilding, agenda setting, priming, individual voting intentions, and the aggregate results: An analysis of four Israeli elections. Journal of Communication, 55 (2), 347-365. Shrauger, S. (1967). Cognitive differentiation and the impression-formation process. Journal of Personality, 35 402-414. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and social psychology, 48 813-838. Solis, B. (2009). The Facebook era, tapping social net works to build better products, reach news audiences, and sell more stuff Boston: Prentice-Hall, Pearson Education. Swift, T. (2001). Trust, reputation and co rporate accountability to stakeholders: Business ethics. A European Review, 10(1), 16-26. 160

PAGE 161

Tankard, J., Hendrickson, L., Silberm an, J., Bliss, K., & Ghanem, S. (1991). Media frames: Approaches to concept ualization and measurement. Paper presented at the annual meeting of the Association for Education in Journalism and Mass Communication, Boston. Tesser, A. (1978). Selfgenerated attitude change. In L. Berkowitz (Ed.), Advances in experimental social psychology (vol.11, pp. 289-338). San Diego, CA: Academic Press. Turk, J. V. (1986). Information subsidies and media content: A study of public relations influences on the news. Journalism Monographs, 100 1-29. Walters, T. N., Walters, L. M., & Gray, R. (1996). Agendabuilding in the 1992 presidential campaign. Public Relations Review, 22 9-24. Walton, R. E. (1969). Interpersonal peacemaking: C onfrontations and third party consultation Reading, MA: Addison-Wesley. Wang, T. L. (2000). Agenda-se tting online: An experiment testing the effects of hyperlinks in online newspapers. Southwestern Mass Communication Journal, 15(2), 59-70. Wanta, W. (1997). The public and the national agenda Mahwah, NJ: Lawrence Erlbaum. Wanta, W., & Golan, G., & Lee, C. (2004). Agenda setti ng and international news: Media influence on public perc eptions of foreign nations. Journalism & Mass Communication Quarterly, 81(2), 364-377. Wanta, W., & Hu, Y-W. ( 1993). The agenda-setting effects of international news coverage: An examination of differing news frames. International Journal of Public Opinion Research, 5 (3), 250-264. Wanta, W., & Mikusova, S. (2010). The agenda-setting proce ss in international news. Central European Journal of Communication, 2 221-236. Wartick, S. L. (1992). The re lationship between intense media exposure and change in corporate reputation. Business & Society, 31 33-49. Weaver, D. H. (1977). Political issues and vote r need for orientation. In D. L. Shaw & M. E. McCombs (Eds.), The emergence of American political issues: The agendasetting function of the press (pp. 107-119). St. Paul, MN: West. Weaver, D. H. (1980). A udience need for orientati on and media effects. Communication Research, 3 361-376. Weaver, D. H., Graber, D. A., McCombs, M. E., & Eyal, C. (1981). Media agenda setting in a presidential election: Issues, images, and interest New York: Praeger. 161

PAGE 162

Weinberger, M. G., & Romeo, J. B. (1989). The impac t of negative product news. Business Horizons, 32 (1), 44-50. Wilcox, D., Ault, P ., & Agee, W. (1989). Public relations: Strategies and tactics Philadelphia: Harper and Row. Wilson, L. J. (2000). Buil ding employee and community relationships through volunteerism: A case study. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to public relations (pp. 137-144). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Winfrey, F. L. (1989). Firm reputation and CSR: A tale of two constructs. Paper presented at the 49th annual meeting of the Academ y of Management, Social Issues in Management Division, Washington, DC. Winter, J., & Eyal, C. (1981). Agenda setting for the civil rights issue. Public Opinion Quarterly, 45 376-383. Wu, H. D., & Coleman, R. (2009). Adv ancing agenda-setting theory: The comparative strength and new contingent conditions of the two levels of agenda-setting effects. Journalism & Mass Communi cation Quarterly, 86 (4), 775-789. Yagade, A., & Dozier, D. M. (1990). The media agenda-setting ef fect of concrete versus abstract issues. Journalism Quarterly, 67 3-10. Yang, S-U, Shin, H., Lee, J-H, & Wrigley, B. ( 2008). Country reputation in multidimensions: Predictors, effe cts, and communication channels. Journal of Public Relations Research, 20 (4), 421-440. Yang, S-U., & Grunig, J. E. (2005). Decomposing organization al reputation: The effects of organization-public relationship outco mes on cognitive representations of organizations and evaluations of organizational performance. Journal of Communication Management, 9(4), 305-325. Yioutas, J., & Segvic, I. (2003). Revi siting the Clinton/Lewinsky scandal: The convergence of agenda setting and framing. Journalism & Mass Communication Quarterly, 80 (3), 567-82. Zacchilli, T. L., Hendrick, C., & Hendrick, S. (2009). The rom antic partner conflict scale: A new scale to measure relationship conflict. Journal of Social and Personal Relationships, 26 (8), 1073-1096. Zaller, J. R. (1992). The nature and orig ins of mass opinion. Cambridge, U.K: Cambridge University Press. Zhang, J., & Cameron, G. T. (2003). Ch inas agenda building and image polishing in the US: Assessing an international public relations campaign. Public Relations Review, 29 13-28. 162

PAGE 163

Zikmund, W. G. (2003). Business Research Methods (7th Ed.) Mason, OH: ThomsonSouth West. Zoch, L. M., & Molleda, J. C. (2006). Buildi ng a theoretical model of media relations using framing, information subsidies and agenda building. In C. H. Botan & V. Hazleton (Eds.), Public relations theory II (pp. 279-309). Mahw ah, NJ: Lawrence Erlbaum Associates, Inc. Zucker, H. (1978). The variable nature of news media influence. In B. D. Ruben (Ed.), Communication Yearbook (Vol. 2, pp. 225-240). New Brunswick, NJ: Transaction Books. 163

PAGE 164

164 BIOGRAPHICAL SKETCH Ji Young Kim received her Ph.D. from the University of Florida majoring in mass communication in the summer of 2012. Her spec ialty is public relations and her outside concentration is international business. As an alumni fellow, Ji Young has taught several courses in the Public Relations Department, including Public Relations Research, Public Relations Campaigns, and International Public Relations. Ji Youngs primary research interests fall under public relations and strategic communications in both international and nat ional contexts. To date, Ji Young has presented several research papers at majo r national and inter national research conferences in the field, and some of t he work won best research awards at the Association for Journalism and Mass Communi cation (AEJMC, 2009), the International Communication Association (ICA, 2010 & 2011), and the International Public Relations Research Conference (IPRRC, 2011). She also received an outstanding international student award from the University of Flori da in 2010. Six of her collaborative articles have been published in j ournals including, Public Relations Review and Public Relations Journal, and she has an upcoming manuscr ipt which will be published in Journalism and Mass Communication Quarterly Ji Young also has professional experience as a graduate intern at the International Organization for Migration in Switzerland, a daily newspaper, and a communication consultancy firm in Korea. Prior to attending the University of Florida, Ji Young earned a masters degree in New Media at Syracuse University in 2006, and she was enrolled in the Public Diplomacy program at Syracuse University between 2007 and 2008. She received her bachelors degrees in Business Administration and Mass Communications from Sogang University in Korea in 2005.