|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 CREDIBILITY OF CORPORA TE BLOGS AND IMPACT ON ATTITUDE TOWA RD A COMPANY By JIUN-YI TSAI A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN MASS COMMUNICATION UNIVERSITY OF FLORIDA 2007
2 2007 Jiun-Yi Tsai
3 This work is dedicated to my closest family me mbers, especially my parents for their endless love and trust. Without their support, both em otionally and financially, I would not have accomplished the task.
4 ACKNOWLEDGMENTS This thesis could not have accomplished without the guidance of my supervisory committee. My supervisory committee chair, Dr. Spiro Kiousis, inspired me with an excellent research idea and provided c onstructive guidance to conduct the experimental design. Through benefiting from his research expertise, this study can integrate persuasion theories and the public relations scholarship in a quantif iable manner. Dr. Linda Hon gave insightful suggestions for me to carefully examine the contribu tion to research. Her inspirationa l lectures and discussions in Public Relations Theories cla ss also equipped me with a soli d theoretical understanding. Dr. Jennifer Robinson has been an energetic mentor who helped me develop an innovative research idea in terms of the emerging blogging issue. In addition, I would like to th ank Dr. Debbie Treise, Dr. Lynda Kaid and Dr. Hyojin Kim for giving all the intriguing classes that I enjoyed very much. I also want to express my gratitude to Dr. Jiunn-Jye Sheu for his warm assistance an d support through my Masters life. Furthermore, Jody Hedge helped me send out the questionnaire link. In particular, I give thanks to my colleague, Feng Shen, for his valuable advice in performing the statistical analysis and interpreting data. Furthermore, I like to thank my friends at UF who ever inspired, encouraged and comforted me during these stressful time of wr iting the thesis: Yi-Jong Tsai, Yi-Shan Hsu, ChunHsin Huang, Pei-Ying Chan, Marjorie Chen, Zyin Chen, Etta Chou, Louisa Chang, Johnson Chiang, Lance Fu, Vick Chiang, Daniel Tien and many more. Special thanks go to Peggy Tsou, Ivan Lin, a nd Nora Huang for their generous help in sending out my questionnaires and always cheer ing me up in these days. Without your great wholehearted companionship, I would not have become strong enough to combat with all the difficulties.
5 Finally, I highly appreciate my parents for nurturing me with endless love and strong courage to overcome all the cha llenges through my life. They are the most precious stones to me than anything else. Most of all, I would lik e to thank Randy Wang for ever-lasting emotional support, unconditional belief and sharing happiness and sadness along this journey.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......10 ABSTRACT ....................................................................................................................................11 CHAPTER 1 INTRODUCTION.................................................................................................................13 2 LITERATURE REVIEW......................................................................................................16 The Internet and Public Relations...........................................................................................16 Weblogs and Public Relations................................................................................................19 Similar, Yet Different from Web Pages..........................................................................20 Impact on Communication Industry................................................................................22 Definition and Element of Weblogs.......................................................................................23 Technological Features....................................................................................................24 Glossary of Blogging Terms...........................................................................................25 Genre of Weblogs in Public Relations....................................................................................26 Corporate Blogging............................................................................................................. ...29 Overview of Corporate Blogosphere...............................................................................29 Impact on Corporate Communication and Consumers...................................................31 Source Credibility Theory......................................................................................................33 Media Credibility Research.............................................................................................33 Blog Credibility...............................................................................................................34 Highlights of Source Credibility Research......................................................................35 Level of Involvement......................................................................................................40 Attitude and Brand Literature.................................................................................................43 Attitude....................................................................................................................... .....43 Attitude toward the Brand (Ab).......................................................................................44 Theory Development............................................................................................................. .46 Research Questions and Hypotheses......................................................................................46 3 METHODOLODY................................................................................................................48 Conducting an Experimental Design......................................................................................48 Stimuli........................................................................................................................ .....48 Sample and the Treatment...............................................................................................50 Pilot Study.................................................................................................................... ..........51 Main Study..................................................................................................................... .........52 Procedure...................................................................................................................... ...52
7 Operational Definitions: Independent Variable...............................................................52 Operational Definitions: Dependent Variables...............................................................53 4 RESULTS........................................................................................................................ ......57 Overview of Statistical Analysis............................................................................................57 Profile of Participants........................................................................................................ .....57 General Usage of Internet and Weblogs.................................................................................58 Manipulation Checks............................................................................................................ ..58 Reliability Check for Dependent Measures............................................................................59 Research Questions and Hypotheses Testing.........................................................................60 Test of Research Question 1............................................................................................60 Test of Research Question 2............................................................................................61 Test of Hypothesis 1........................................................................................................62 Test of Research Question 3............................................................................................63 5 DISCUSSION AND CONCLUSION...................................................................................77 Study Overview................................................................................................................. .....77 Overview of Research Qu estions and Hypothesis..................................................................79 Conclusion..................................................................................................................... .........85 Implications................................................................................................................... .........88 Theoretical Implications..................................................................................................88 Practical Implications......................................................................................................90 Limitations.................................................................................................................... ..........91 Suggestion for Future Research..............................................................................................92 Investigate Different Stakeholders..................................................................................93 Incorporate Other Variables in Persuasion Research......................................................94 Corporate Credibility and Endorser Credibility..............................................................95 Ethical Concern...............................................................................................................96 APPENDIX A BLOGGING GLOSSARY....................................................................................................99 B WEBLOG LAYOUT..........................................................................................................100 C INTRODUCTORY PARAGRAPH OF WEBLOG AUTHORS.......................................104 D BLOG POST ONE: NOW ON YOUTUBE.......................................................................105 E BLOG POST TWO: NOT TRA PPING USERS DATA=GOOD.....................................107 F BLOG POST THREE: SOME THOUGHTS.....................................................................109 G QUESTIONNAIRE FOR EXPERIMENT GROUP...........................................................111 H QUESTIONNAIRE FOR CONTROL GROUP.................................................................114
8 LIST OF REFERENCES.............................................................................................................116 BIOGRAPHICAL SKETCH.......................................................................................................125
9 LIST OF TABLES Table page 3-1. Conditions of 1x4 experi mental design with control group...............................................56 3-2. Likert-type measurement items of involvement level........................................................56 4-1. Demographic profile of the respondents............................................................................66 4-2. The number of participants in each cell..............................................................................66 4-3. General usage of Internet............................................................................................. .......67 4-4. General usage of Weblogs.............................................................................................. ....68 4-5. Cronbachs Alpha of Variables......................................................................................... .69 4-6. Perceived trustworthiness by source cue............................................................................69 4-7. F-test of perceived trustworthiness by sources cue............................................................69 4-8. Multiple comparison of mean difference of trust scale......................................................69 4-9. Perceived expertise by sources cue....................................................................................70 4-10. F-test of perceived expertise by sources cue......................................................................70 4-11. Multiple comparison of mean difference of expertise scale...............................................70 4-12. Attitude toward the company by sources cue.....................................................................70 4-13. F-test of attitude toward the company by sources cue........................................................70 4-14. Multiple comparison of attitude toward the company........................................................71 4-15. Tests of between-subject eff ects (dependent variable: attitudes).......................................71 4-16. Pearson correlation among perceived tr ustworthiness, expert ise and attitudes..................71 4-17. Hierarchical regression an alysis of the proposed model....................................................72 4-18. Regression analysis of the CEO source group....................................................................72 4-19. Regression analysis of the editor source group..................................................................72 4-20. Regression analysis of th e senior student source group.....................................................73 4-21. Regression analysis of the PR manager source group........................................................73
10 LIST OF FIGURES Figure page 2-1. Theoretical model of s ource credibility on the blogosphere.............................................47 3-1. Procedure fo r experimental groups....................................................................................56 4-1. Mean of trustworthiness among four source groups.........................................................74 4-2. Mean of expertise among four source groups...................................................................75 4-3. Mean of attitude to ward the company for 5 groups...........................................................76 5-1. Modified theoretical model in the present study...............................................................98 B-1. Weblog layout for the CEO source..................................................................................100 B-2. Weblog layout for the PR manager source......................................................................101 B-3. Weblog layout for the Busi nessWeek magazine editor source........................................102 B-4. Weblog layout for the senior student source....................................................................103
11 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Mast er of Arts in Mass Communication CREDIBILITY OF CORPORA TE BLOGS AND IMPACT ON ATTITUDE TOWA RD A COMPANY By Jiun-Yi Tsai August 2007 Chair: Spiro Kiousis Major: Mass Communication Started in 2004, Weblogs, which gain popularity as an effective altern ative public relations communication channel targeting Ne t citizens, allow corporate l eaders, managers, and industry watchers to speak directly to their desire d publics in a human voice tone. Many media-savvy corporations have ventured into the blogosphere in an effort to build positive relationships with employees and consumers. Little research has em pirically addressed the key issues of source credibility on the blogosphere a nd its effect on attitude toward a company. As Trammell points out, scholars should move from describing the contents of blogs to investigating the communication effects of this new medium. Therefore, our main objective was to explor e whether different sources of blog authors could affect peoples perception of credibility and attitude towa rd a company. We also tested whether level of involvement can m oderate the effect of perceived s ource credibility on attitudes. In addition, we verified the causal relations am ong attitudes, perceive d source credibility (trustworthiness and expertise) and level of involvement. We conducted a 1 by 4 (difference sources: CE O blog, journalist blog, p eer individual blog, public relations manager blog) factorial expe rimental design with a control group, which received no blog information. A to tal of 177 participants were recruited and randomly assigned
12 to one of the five conditions. While most experi mental designs were performed in classroom or laboratory settings, the innovative approach in this study that invited partic ipants to read online blogs could overcome the artificial ity and gain external validity. Results suggested the perceived source cred ibility (trustworthiness and expertise) and attitude toward a company can be affected afte r people read four popular sources of corporate blogging (CEO, industry watcher, PR manager, college student). Among the four types of authors, the CEO of Google Company was perc eived as a highly trus tworthy and expertise source. The finding was consistent with previous research which suggest ed that the CEOs are often considered as experts in their industry because of their extensive managerial experience and insights they possess. On the other hand, the senior student source, wh ich was defined as the peer individual source in this experiment, was perceived as the least tr ustworthy and expertise compared to the other three authors. The results did not indicate a pivotal role that the peer individual source has gained influence on the blogosphere. Participants who read the public relations manager source showed favorable ratings in perceived e xpertise but neutral opinions in perceived trustworthiness. In conclusion, the study confirmed the effects of source cues on perc eived credibility and attitude toward a company on the blogosphere. Thus higher perceived source credibility can lead to more favorable attitudes toward a company. Moreover, through running a regression analysis, the results found that perceived trustworthiness, expertise and level of involvement could be effective predictors for achieving positive attit udes toward the company in four source groups, respectively.
13 CHAPTER 1 INTRODUCTION During recent decades, the Internet has been growing at an unprecedented rate. According to the Pew Internet and American Life Report, survey results in 2006 in dicated that Internet penetration among adults in the U.S. has hit an all-time high. The percentage of Americans who say they use the Internet has continued to fluctu ate slightly. Seventy-three percent of respondents (about 147 million adults) are Internet users (Madden, 2006). A number of studies have sugge sted that the Internet is an influential medium in public relations. The advancement of new technologies such as the Internet and WWW (World Wide Web) can assist organizations in reaching their publics more effectively and building positive corporate images (Cooley, 1999). Furthermore, seve ral scholars explored th e interactive capacity of online communication a nd its impact on maintaining relati onships. They postulated that the enhancement of interactivity will enable corporations to realize two-way symmetric communication and maintain mutually beneficial relationships with their key stakeholders (Grunig, 2000; Johnson, 1997). In recent years, given the proliferation of Weblogs, th e blogsophere has provided a new channel for corporations to interact with a udiences in a more candid and responsive manner. Blogs emerged as a new public rela tions tool in maintaining orga nization-public relationships in that they allow more audience feedback and co mmunication in human voice. Corporate leaders, employees, journalists and individu als are using this new medium to directly share their thoughts and insights. In terms of consumer relations, the growth of the blogosphere is a remarkable trend that sees audiences and consumers gaining a vo ice and influence over institutions. For example, the negative complaint entries of Jeff Jarvis blog have become Dells nightmare as mainstream media, BusinessWeek Magazine, published the st ory. The fundamental shift from one-way or
14 even two-way communications to a form of in clusive communications requires a fundamental shift in managing organization-public relationships So far, few empirical studies have been conducted to explore the impacts of blog comm unication on attitudes toward a company. The advantage of the blogosphere for corpor ate communication is to boost credibility and get closer to customers. Genera lly, corporate executives recognize that a blog is an effective yet low-cost way to influence the public conversation about companies, enhance brand visibility and credibility and achieve custom er intimacy (Heires, 2005). Traditional persuasion research revealed a longstanding interest in so urce credibility. It has been well established that ch aracteristics of a message sources can affect attitudes. Source credibility has two major components: expertis e and trustworthiness (Hovland et al., 1953; McGuire, 1969). Ohanian (1990) studied the cr edibility of celebrity spokespersons in advertisements and emphasized two dimensions of credibility, expertise a nd trustworthiness, as the most applicable to corporate perceptions. Th us, the scale development in this study focused on measuring consumer perceptions of expertise and trustworthi ness to form the essential aspects of the perceived source credibility concept. Research began to explore the impacts of bl ogs regarding applicati ons in political and corporate communication settings (Kaid & Po stelnicu, 2006; Trammell & Metzgar, 2006). With exploratory and blending results from academic re search and survey reports that indicate blogs could be new potential communica tion channel to build corporat e credibility, the issue of different source cues of bloggers and impact of blogs on perceived credibility and attitudes still lack empirical support. Therefore, the theoretic al model of source cred ibility on the blogosphere (Figure 2-1) is proposed and test ed in this study employing a 1 by 4 factorial experimental design.
15 The blending of these two concepts, the pr oliferation of corpor ate cyber-chatter and increasing interest in exploring so urce credibility of Weblogs, gets at the heart of the research interest described here. Theref ore, the purpose of this study wa s to investigate the perceived source credibility among different bloggers and examine the relationships between credibility ratings and attitudes toward a company on the blogosphere.
16 CHAPTER 2 LITERATURE REVIEW The Internet and Public Relations In recent decades, the Internet has grown at an accelerating rate. According to the Pew Internet and American Life Repor t, surveys results in 2006 suggest ed that Internet penetration among adults in the U.S. has hit an all-time hi gh. The percentage of Americans who say they use the Internet has continued to fluctuate slightly. Seventy-th ree percent of people (about 147 million adults) are becoming Internet users (Madden, 2006). A number of studies suggeste d that the Internet is an influential medium in public relations. Research has found that the Internet has changed the trad itional ways of organizational communications in that it has allowed for more interactions between organizations and its publics. In addition, public relations professionals are steadily incor porating the attributes of the Internet as an integrated part of their communication strategi es with publics. Johnson (1997) explored public relations pract itioner perceptions of the Internet and WWW and described how the use of new communication tech nologies can influence public relations roles. He provided suggestions of how professionals can improve research, productivity, and two-way symmetric communication by utilizing new technologies. In a special issue of the Public Relations Review for Fall 1998, many scholars explored the utility of Web sites for public relations f unctions and provided seve ral examples of how organizations employ the Intern et to communicate with public s. The issues covered how corporations and activists groups increasingly uti lized the Internet to extend their reach. Heath (1998) took an issues management approach to new technologies and then explored the rich dialogues that can occur on the Web. Furthermore, Kent and Ta ylor (1998) suggested that building dialogic relationships via the Web was possible and proposed five principles to be
17 followed in building dialogic Web sites. The firs t principle is to crea te a dialogic loop. A dialogic loop allows publics to query organizations and, most impor tantly, it offers organizations the opportunity to respond to question, concerns and problems, they wrote (p.326). The second principle is to be sure that information on the We bsite is useful to all publ ics. The third principle is the generation of return visits by providing up dates of pertinent and valuable information. The fourth principle is the ease of interface which is easy to navigate. The last principle is to conserve visitors. The most distinguished feature of the Internet is the high level of interaction that it affords users. The Internet empowers receivers to intera ct with senders and fac ilitates dialogues between the two parties (Kelleher & Miller, 2006; Kent & Taylor, 1998; Seltzer 2005). Johnson (1997) stated that, through interactivity, the Web could facilitate media relations, employee communication, government relations, and customer relations. Cooley ( 1999) discussed four primary benefits of increasing interaction w ith the publics through corporate Web sites. The interactive characteristic s of the Web provide an opportunity to improve the corporate image, to collect and analyze public opinions more easily, to set corporate agendas to reflect the needs of the public and to increase cor porate accountability. She concluded that interactive Web sites serve as a channel to construct mutually beneficial dialogues and relationships. To further access the poten tial of Web-based communication, research started to investigate the impact of interactive features on maintaining positive relationships. OMalley and Irani (1998) suggested that highe r amounts of information and inte ractivity would increase the intention of revisiting a Web site. They concluded that the concept of interactivity in the Web is related to relationship building th rough attitudinal and behavioral change. Moreover, Jo and Kim (2003) conducted an experimental design to ex amine the impact of in teractivity on building
18 relationships. Overall, they sugge sted the enhancement of interac tivity in a Web site would bring better relationships with publics. In conslusi on, the intrinsic interac tivity of the Web can enhance the mutual relationshi p and collaboration between the organizations and publics. Bauer, Grether, and Leach (2002) looked at how the Internet can play a role in cultivating customer commitment, satisfaction, and trust. Results stated that C ustomers who trust a corporation feel more committed to it. In addition, the authors found that satisfaction had a greater effect on commitment than did trust. High availability via a Web site has a positive effect on trust, whereas efficient information transferri ng via a Web site and the opportunity to interact with an organization increase comm itment to that organization. Ki (2003) conducted a co ntent analysis of 286 Fortune 500 company Web sites to determine how organizations exhibit relations hip maintenance strategies- positivity, openness, sharing of tasks, networking, and access through their Web sites. Results indicated company Web sites be st performed openness, which was operationalizd by revealing company overview, releasing news, annual repor ts and stock price. Fortune 500 companies seemed to provide company related information to a much greater degree than information related to networks with extern al groups. This article implie d that relationship maintenance strategies on Web sites can play a pivotal role for orga nizations or corpora tions in maintaining relationships with their publics. Furthermore, Reber and Kim (2006) analyzed activist Web sites to determine how activist groups use online resources to maintain media relations. Evidence has found most activist Web sites built dialogues with gene ral publics through providing gene ral contact information and response mechanisms. However, there were fewer dialogic features such as contact information for specialists or experts and online press rooms that could serve to build trust and satisfaction
19 among journalists. They suggested activist organi zations could advocate for their organizations at the maximum capacity by taking advantage of th e characteristics of Internet communication. To sum up, public relations has generally been viewed as a dialogic profession that values meaningful interaction between an organizati on and its various publics or stakeholders. The Internet as a new communication channel generates more chances for organizations to realize excellent public relations communication in a two-way a nd symmetrical manner through maintaining appropriate re lationship strategies. Weblogs and Public Relations Currently, there has been much attention to emphasize the use of blogs as an emergent public relations tool. Blog-related research range d from facilitating the agenda setting process (Trammell & Kiousis, 2005) to an emphasis on how to employ blogs strategically. The top public relations firms have begun to offer blog ma rketing services for th eir clients and published white papers on blogs such as Edelman and Mak ovsky & Company. There is much evidence that public relations researchers, as we ll as practitioners, are increasingly interested in incorporating blogs into communication strategies. According to the Pew Internet and American Life Project, the report shows that blog readership grew by 58% in 2004 and more than 32 million people now read blogs. By the end of 2004, blogs had begun to establish themselves as a key part of online culture (Rainie, 2005, p. 1). Another telephone survey of a nationally-re presentative sample of bloggers in the United States has found 8 percent of Internet users, or about 12 million American adults, keep a blog. Thirty-nine percent of Internet users, or a bout 57 million American adults, read blogs a significant increase since the fa ll of 2005 (Pew Internet, 2006) Technorati, a technology company providing blog tracking service, current ly reports the blogosphe re is over 60 times bigger than it was only three years ago. Over 75,000 new Weblogs are created every day, which
20 means that on average, a new weblog is crea ted every second of every day and 19.4 million bloggers (55%) are still pos ting three months after their blogs are created. In addition, about 3.9 million bloggers update their blogs at least weekly. While previous research on the application of the Internet in public relations mainly focused on corporate Web site features, the Intern ets dialogic function and interactivity (Jo & Kim, 2003; Kent & Taylor, 1998), blogs are similar to Web sites in terms of their capacity and capability for high interactivity. Regarding comm unication on the blogosphere, public relations professions suggested Weblogs en able corporations to realize two-way, dialog ic communication. Marken (2005) suggested blogs provide manageme nt what they want to and need in todays world a fast, effective and economic means of carrying out two-way communication with the firms many audience. Blogs also represent a l ong-term commitment to reaching out to educate, inform, influence and hear directly from communities. With the popularity of the blogosphere, it unders cores an important shift in the structure and practice of communications. Blogs enable au diences to move away from the traditional pyramid of influence with its top-down, one-way information flow to a more fluid, horizontal peer-to-peer paradigm, in which brands and corporate reputations are built by engaging multiple stakeholders through continuous dialogue. Scholar s and professionals therefore predict that blogging will become a new commun ication tool for public relations practitioners to facilitate dialogues and manage relations hips with key publics (Law son-Borders & Kirk, 2005). Similar, Yet Different from Web Pages Although blogs share many common features with Web sites, such as multimedia presentation and ways of self-expression (Kat z & Rice, 2002), researcher s have distinguished several differences. According to Balnaves et al. (2004), what makes bl ogs different from Web sites is that they address unique topics that do not fit into the context of traditional media.
21 Moreover, the authoring partnership (p. 196) provides an added dimens ion of interactivity. Additionally, Trammell and Gasse ( 2004) summarized five structural differences between classic Web pages and blogs. The first key difference lies in the role of the hyperlink, which illustrates the differences in the intended impact of the cont ent on the page. For example, hyperlinks in blogs serve as a gateway to greater understanding and are often us ed to support comments made in a blog post (Blood, 2002). On the other hand, the common menta lity in linking strategy of Web sites is that Web masters do not want to send their readers aw ay by linking to an off-site resource. Chuck Olsen, the filmmaker behind Blogumentary.org, says, the beauty of (hyperlinking) is that you can let your readers decide and draw th eir own conclusions (Blogumentary, 2003). In addition, blogs can be understood as conve rsations punctuated by hyperlinks that tie different elements of a conversation together (W alker, 2005). Viewed from this angle, blogs do incorporate a comparatively higher level of interactivity. The comment feature makes conversations in the blogsphere interpers onal in natural which allows more audience participation. Third, blogs are designed to be read on a daily or weekly basis (Blood, 2002). The posts are presented in a reverse chronological order, which is different from traditional Web sites. In this way, readers can easil y perceive updated content. Moreover, the archive function of blogs main tains old entries in a chronological and systematic way which allows readers to find posts pertaining to certain month or topic. This is unique to blogs. Finally, blogs are distinct fr om Web pages in terms of the communities that surround them. The community as pect of Weblogs has long been emphasized by researchers (Halavais, 2003; Marlow, 2003). Blood (2002) noted that that ea rlier bloggers felt a strong identification with every other weblogger, even wh en their interests and ta stes differed in every
22 way (p. 101). Halavias (2003) constructed a metaphoric approa ch for investigating blogs by looking at the blogosphere as a city. Impact on Communication Industry Increasingly, professiona l publications such as Public Relations Tactics have discussed the blog phenomenon and implications for public relati ons practitioners. Trufelman (2005) described a blog as a form of consumer-generated media and suggested that blogs enable media consumers to have more media outlets than ever before. Key (2005) also poi nted out that Internet technology challenged old public re lation models and noted that there is a need to recognize communication trends in this field. On the bl ogosphere, the comment feedback mechanism, and candid disclosure all act as impetuses for the st art of dialogue (Burley, 2005). For these reasons, knowing the tone and content of major blogs and their impacts on organizations is important for practitioners (Albrycht, 2004). Given the buzz trend of blogging, scholars star ted to explore the impact of blogs in communication industry. Trammell (2005) analyzed the blogs posts, comments and trackbacks to examine different motivations of celebrity blogs readers. She found bloggers tended to post by self-expression while blog readers left comments for social interacti on, and those who create trackbacks desired to share information. Tramme ll and Kiousis (2005) studied the agenda-setting function between bloggers and readers. Their stu dy revealed that there was a positive correlation between Weblog interactivity and salient ag enda-setting. Porter, Trammell and Chung (2006) founded blog use by PR practitioners appeared to be on par with the general population. Most PR practitioners maintained their pe rsonal blogs and used blogs prof essionally at low level. Women lagged behind men in the strategic use of blogs. Moreover, Seltzer (2005) conducted a content analysis of 50 environmental Weblogs to identify their dialogue potential in relationship building, and the analysis indicated that Weblogs
23 could incorporate the existing dial ogic principles of traditional Web sites to build relationships online. Consequently, the pers onalized form of communication in blogs has been found to correlate positively with levels of trust and satisfaction in organization-public relationships (Kelleher & Miller, 2006). With the popularity and significant im pact of blog communication in mind, the following definition and characteri stics will garner a more comprehensive understanding of what blogs are. Definition and Element of Weblogs Weblogs, or blogs, began popping up on the Intern et as early as 1997, an d there have been several noted attempts at defining the technology and differentiating the medium from standard Web pages (Trammell, 2004). The current technical definition of a blog is a Web page with a series of dated entries arranged in reverse chronological order (B lood, 2002; Walker, 2005). However, the social definition better represents wh at blogs really are. Accord ing to Blood (2002), bloggers are opinionated people who use the ou tlet to communicate their th oughts, ideas, reflections, and politics. Walker (2005) confirms this by calling a blog a soapbox. The collection of all blogs in cyberspace is called the blogoshpere (Trammell, 2004). Wijnia (2004) further looked at Weblogs in a communicative way a nd suggested that a blog is a Web page on which the author publishes pi eces with the intention to start conversation. Conversation is achieved through a comment function to enable dialogue within the same blog. People are able to track convers ation through different Weblogs a nd use trackbacks every single entry. Finally, the content of a blog can be s yndicated for news aggregators via RSS (Really Simple Syndication) and easily linked to from other blogs. RSS feeds allow people to subscribe to blogs and thus be notified when bloggers has posted something new. They represent new ways
24 of tracking what people are writi ng about a particular organization or issue, both in traditional media outlet as well as in blogs. Technological Features Scholars agreed that Weblogs are more respon sive than Web sites (Seltzer, 2005) and have higher interactivity (T ramell & Kiousis, 2005). Wijnia (2006) argued Weblogs meet the two or more-sided communication. Gill (2004) indicated the primary characteristics of a blog: Reverse chronological journaling (format). Regular, date-stamped entries (timeliness). Links to related news articles, documents, and blog entries within each entry (attribution). Archived entries (old content remains accessible). Links to related blogs (blogrolling). RSS or XML feed (ease of syndication). Passion (voice). (p. 2) In addition to structural char acteristics, the presence of conversational human voice or authentic voices is particular ly unique to blog communication si nce profit-driven Web sites are criticized as less interesting and less relevant to online publics (Gill, 2004; Kelleher & Miller, 2006). The concept of human voice in blogs means a natural, direct and informal voice (Wack, 2005, para. 3). Kelleher and Miller (2006) operationaliz e human voice as communicating with a sense of humor, admitti ng mistakes, treating others as human and providing links to competitors. The interpersona l communication nature engages blog readers to the twoway flow of conversation. Also, the authoring partnership (Balnaves et al., 2004, p. 196) characteristic expands public dialogue. Smudde (2005) emphasized that the dialogue is the essence of blogs because people participate in the communication and exchange messages about various topics through blogs. Furthermore, the us es of hyperlinks and blogrolls widen the scope of available discussion topics. The shift in a pproach from targeting markets to markets as
25 conversations is consistent with public relations theory that calls for building collaborative relationships with constituencies rath er than targeting them (Grunig, 2000). Glossary of Blogging Terms The emergence of blogs has generated new term s for Internet users. The term blog can be used as a verb as well as a noun. The act of adding a new entry is known as blogging (Gill, 2004), and people who host blogs are referred to as bloggers. A post is the main body of the blog, and its contents are usually in te xt format. The blogroll refers to a list of links to other blogs (Trammell, 2004, p. 12). According to Trammell ( 2005), comments and trackbacks are feedback mechanisms. Comments allow reader s to post their reactions to orig inal authors po sts. Similarly, trackbacks are also feedback; however, they are utilized on readers own blogs when they are intrigued in certain issues discussed on others blog and they respond to the topics by posting remarks on their own blogs. While citing someone s blogs, trackbacks will automatically notify the original author that his/he r blogs are referenced by other bloggers. The trackback mechanism simplifies linking reference to other blogs. Throu gh trackbacks, readers can follow the flow of discussion on the blogsphere. In short, trackbacks are a list of other blogs to refer to and usually creates deeper discussions than comments. Permalinks, an abbreviation of permanent link, are permanent uniform resource locators (URLs) link to Weblogs archives. Once the blog cont ent is created, the URL link is stored in the database and remains unchanged regardless of wh ether the title of the post is removed or changed. Permalink is useful when blog archives grow to a considerab le number of pages. Permalinks enable readers to find specific inform ation within blogs, and links to blog posts are permanently valid. Category labels different posts photos and links and cla ssifies similar topics and discussions into the same category. Category he lps blog readers save time to find interesting topics. In addition, bloggers can c hoose whether they want their blogs to be tagged or not,
26 which is viewed as an increasing trend. Blogge rs can choose a most relevant subject according to their blog content and then tag their bl ogs. According to Sifry (2006), a tagging service enables bloggers to associate their posts with topics, and to make it easy for people to find interesting posts on a given subject (para. 9). In addition to basic communication functi on elements, content managing tools are developed to meet the needs of the growing numbers of blogs. To get content distributed actively and in a timely manner to readers, content feed tools such as RSS (Really Simple Syndication), Atom and XML (Extensible Markup Language) enable readers to receive the latest headlines and updated blog contents without browsing the bl ogs. Other blog glossary terms are compound vocabulary from existed media terms, such as podcasting (comes from iPod and broadcasting), moblog (comes from mobile and Weblog), and flog (fake blogs). See Appendix A for a more detailed glossary of blogging terms. Genre of Weblogs in Public Relations Herring et al. (2005) viewed Weblogs as a bridging genre because Weblogs form a de facto bridge between multimedia HTML document and text-based computer-mediated communication (p.143). They analyzed 203 randomly selected blogs and employed purpose as a key criterion for classifying blogs according to a modified version of Bloods (2002a) three blog types: filters, personal jo urnals and k-logs. Filter blogs refe r to contents primarily containing external observations and evaluations such as public events. Personal blogs means bloggers report their personal life and inte rnal reflections. The contents of k-logs focus on information around external topics, projects or products. Given the fluid nature of blog entries, we of ten see that blog content combines the above three kinds of information. Therefore, Smudde (2005) categorized blog types in a more comprehensive way to look at blog applications in public relations cont ext. Based on his four
27 types of blogs that are define d by discussion topics and blog au thors, this study divides public relations blogs into five types: Personal/ individual blogs. This type refers to blogs are created by anyone to express his or her personal convictions, observations, sugges tions, and other matters about selected topics that interest him or her. According to Smudde (2005), this type of blog is opinionated and hyperlinks to external Internet sources are abundant. In this re gard, personal blogging networks could be viewed as online micro communities that carry world of mouth marketing function to make or break a corporati ons reputation (Goodman, 2006). Political blogs. Political blogs are specialized diarie s in the sense that they document the authors or candidates political views rather th an his or her daily life (Pew Internet, 2006). The 2004 presidential election cycle feat ured the use of blogs with great intensity (Bichard, 2006). Both President Bush and John Kerry held their official blogs which ha d significant impacts on campaign communications. (Http:// GeorgeWBush.c om, Http:// JohnKerry. com) According to the Washington Post Election Day traffic on blogs rivaled that of traditional media. The most important campaign blog was Blog for American, which served as a nerve center for Governor Howard Deans insurgent presidential campaign. Kerbel and Bloom (2005) indicated his blog to be an example of how the Intern et is emerging as a vehicle fo r enhanced civic involvement. Based on the fact that more and more people sear ch online political information, candidate blogs are an inexpensive way to enhance citizen invol vement and participati on for citizens who are politically activ e (Gill, 2004). Publication-based blogs. According to Smuddes (2005) de finition, typical bloggers of this format are editors, journalist s, or freelancers that foster dialog about subjects of interest to the parent publications readers as outgrowths of established media outlets Bloggers as grassroot
28 reporters express opinions without editorial constraint and provide immediate news stories. This kind of blog is perceived as professional, be cause the bloggers knowle dge allows them to establish expertise in serving news-gathering purpose (p.36). Industry blogs. This type of blogs focuses on a specifi c topic history, development, trends and players in a certain industry. Industry blogs are characterized by blog gers with a credible track record of experience, knowledge, and su ccess that usually pres ent authoritative and influential opinions on the t opics discussed (Smudde, 2005). Re lying on accountability, bloggers impact and sometimes change readers initial thoughts. Corporate blogs. Corporate blogs, also known as busin ess blogs, are devoted to a single company and support organizational communicati on goals. According to Wack (2006a), the definition of a corporate blog is a blog publishe d by, or with the support of, an organization to reach that organizations goals (para. 8). Smudde (2005) regards the corporate blog as a hybrid of the personal blog, which are fairly new a nd feature the insights, assessments, commentary and other disclosure devoted to a single comp any (p. 3). In this way, blogs writers are organization members, from employees to top managers, including company presidents and Chief Executive Officers (CEOs). Famous corporat e blogging leaders are the vice chairman of General Motors, the president of Sun Microsyste ms, and the president of marketing of Boeing Commercial Airlines. Boards of directors could use the technol ogy to get their message out, and at the same time provide a forum for shareholders to offer informal input to their elected hoard representatives. On the other hand, Weil (2005), the author of The Corporate Blogging Book pointed out different views. She suggested co rporate blogs are written by some one hired especially for that job or by a small group within a company that pr oduces a document similar to an e-newsletter,
29 not really a corporate blog. With the attractiven ess of the corporate ex ecutive viewpoints/voice portion, good corporate blogs update information frequently and cons istently, often at least once weekly (Marken, 2005). However, in an ordinary business operation settin g, it seems impractical that corporate leaders would spend more time updating their blogs than on their job duties. Therefore, Marken (2005) suggested that an effective CEO blog post co nsists of executives writ[ing] out two or three key points which ar e viewed and messaged by a blog-savvy PR staff member( pp. 32-33). Corporate Blogging Overview of Corporate Blogosphere In 2004, corporations started to venture into the blogosphere to manage relationships mainly with their customers and employees. The cover story Blogs will change your business of Business Week in May 2, 2005, reported the increasing trend of blogs, discussed the feasibility of corporate blogging, and predicted that business blogging provi ded opportunities for corporations (Baker & Green, 2005). On the contrary, Forbes reported that attacks are rapidly expanding on the blogosphere and can destroy corporate brands and reputation (Lyons, 2005). The anticipated rise in blogging as a busine ss communication tool is gaining ground with corporations. Bill Gates, CEO of the Microsoft Corporation as well as an innovation evangelist, has discussed how blogging is going to be a fa d because it is very easy to communicate (Stevenson, May 21, 2004). Gates at his 2004 CEO summit said Web sites and email were outmoded means of business communication with blogs and RSS being the answer. Outside the technology field, General Motors was among the fi rst corporate to establish executives blog. Bob Lutz, Vice Chairman of Saturn, continued to post passionately about his companys product. For external communication, blogs could serve as a new medium, wh ich is significantly related to corporate image and performance. On the same note, major corporations also witnessed the
30 rapidly increasing numbers of employees blogs (Keller & Miller, 2006). Within corporations, blogs are used for internal communication and viewed as knowledge-sharing channels. HewlettPackard and International Business Machines Corpor ation are examples of companies that utilize blogs for maintaining employee relations (Jones, 2005). According to the Fortune 500 Blogging Wiki, as of October 5, 2006, 40 (8%) of the Fortune 500 companies reported active public bl ogs written by their employees about the company or its products. The number keeps increasing. Additionally, Backbone Media Inc. (2005) surveyed bloggers at hundred of companie s and conducted in-depth interviews with representatives from six corpora tions currently leading the way in blogging activities. Overall, the survey respondents agreed that they believe th ere is a broad array of benefits to starting a blog, including quick publishing, thought leadership, building community, sales and online public relations. Moreover, Maksovsky & Company Inc. (2006) car ried out a national telephone survey of 150 senior executives of Fortune 1,000 companie s asking their attitude toward corporate blogging. The findings suggested cor porate leaders were slow to re act to the growing credibility of corporate blogs as a communi cation medium. Most executives we re somewhat or not at all convinced of blogs credibility in as a commu tations medium, brand-building technique and a sales or lead generation tool. De spite this fact, 77% of them believed that their organizations should have corporate policies on blogging (Goodman, 2006). In br ief, corporations have begun to consider blogging as the ne xt step in corporate communi cation strategy to promote organizations reputation and visi bility. This trend encourages co rporations to develop effective strategies to engage on the blogoshpere as a way of reaching out and obtai ning direct feedbacks.
31 Impact on Corporate Comm unication and Consumers The growth of the blogosphere is a rema rkable trend that empowers audiences and consumers to gain a voice and influence over ins titutions (media, entertainment, brands) that were formerly under the tight control of author ity figures (Hirshberg, 2006). Blogs are worth watching because they provide the most complete example of consumers growing control of media and brands (Lewis, 2005). The advantage of the blogosphere for corporate communication is to boost credibility and get cl oser to customers. Generally, co rporate executives recognize that a blog is an effective yet low-cost way to in fluence public conversations about companies, enhance brand visibility, cred ibility and achieve customer intimacy (Heires, 2005). Many of the early adopters of blogs have b een technology companies that are eager to leverage blogs ability to pos ition a company executive for recogni tion as an expert in a given industry or on a specific topic. With his blog, J onathan Schwartz, the president and COO of Sun Microsystems, has established himself as a thou ght leader on issues pertinent to computer operating systems. Beside Sun Microsystems, companies that encourage blogging activity include General Motors, The Boeing Company, G oogle, Hewlett-Packard, IBM, Microsoft, Red Hat, Edelman, Stonyfield Farm and Yahoo! (Hei res, 2005). In addition, Sun Microsystems and Microsoft encourage employees to become bloggers for debate, free asso ciation, and collecting input about projects (Roush, 2005). The content is virtual word-o f-mouth communication that is typically opinioned and replete with links to other Internet sources. As for consumers, blogs can facilitate conve rsation between customers and corporations because they enable companies to receive direct feedbacks from their customers. Rank-and-file employees will blog about their companies while consumers will speak directly to people who share similar interests. These individuals have not been media trained. They are on the Web sharing ideas and collaborating. They are co-creating tomorrows products, brands and corporate
32 reputations continuously and spont aneously. At the core of the technology business, customers are writing and sharing their own open source applications with cu stom feature sets and cheaper price points therein challenging established proprietary software vendors. These characteristics make personal blogs perhaps the most potent type, because, depending on the bloggers agenda and attitude, subjects covered in the blog could be helpful or hur tful to the companys image, reputation and other aspects (Smudde, 2005). A recent report conducted by Edelman and Tech norati in 2005 revealed how companies interact with online community members and how interactions can be improved for the mutual benefit of companies. The report helped quan tify bloggers perspectiv es and illuminated the goals and objectives of individua l bloggers. In the Edelman/Technorat i survey, more than half of all respondents indicated that they write posts about comp anies, their products, or their employees at a rate of once per week or more. As many as nearly half of all bloggers (48%) reported never having contact with companies or their public rela tions representatives. This indicates that corporate communi cators are not venturing into th e blogosphere. Seventy percent respondents indicated that they would be interested in receivi ng products for evaluation and review while sixty-three percent respondents indi cated they trust others bloggers when seeking product information. This finding co uld suggest that peer-to-peer horizontal communications are more preferred in the online world where word-o f mouth is by far the most effective form of message delivery. Richard Edelman, CEO of Edel man Company, perceived the effect of the blogosphere as a network of cross influence. He suggested the network is a dynamic, real and virtual world of information exchange in which all participants demand a voice. Those who have been at the bottom of the pyramid of influence consumers and employees can use an array of new tools that enable them to sh are information and opinions online.
33 Source Credibility Theory Media Credibility Research A large body of media credibility research lies in comparing the perceived credibility among print, broadcasting (radio a nd television), and online news In the early 1960s, several Roper polls noted that televisi on was perceived as being more credible than newspapers. The findings were verified by several academic studi es suggesting television news has traditionally been viewed as most credible sources of information (e.g., Abel & Wirth, 1977; Gaziano & McGrath, 1986; Lemert, 1970). Because the evolution of new communicati on technology, especially the Internet, has changed media ecology, the impor tant shift to online news ha s generated a more complex scenario and yielded a valuable addition to media credibility sc holarship. Recently, the Internet has been the focus of studies comparing the credibility through traditional media outlets. Kiousis (2001) conducted a cross-sectional survey to ex plore the perceived credibility of television, newspaper and the Internet. Results revealed part icipants considered newspapers as the most credible news source, followed by online ne ws and television ne ws, although people are generally skeptical about news pr esented in three channels. The striking outcome that newspaper credibility ratings would exceed television news credibility ratings reversed the opposite trend since 1960s supporting television was found to lead the others in credibility by a large margin. Another surprising conclusion was the Internet ne ws garnered more credibility ratings than television news. Similarly, Brady (1996), Johnson and Kaye (1998) also concluded Internet news is perceived as credible as trad itional news, or even more credible. According the Pew Research Centers survey, it found that cr edibility of both print and broa dcasting news has declined in recent years. This evidence was consistent with Kiousiss (2001) statement that the introduction of online sources has changed public opinion of traditional media credibility (p. 395).
34 Blog Credibility Researchers continue to raise the concern about the issue of blog credibility as the blogoshpere grows at an ever fast pace. As Tr ammell (2005) point out, scholars can move from describing and defining blogs to wo rking on future understanding the uses and effects of blogs. Johnson and Kaye (2004) started to examine whet her blog users perceived blog information as credible or not compared to other media. Th ey found blog readers beli eve blogs are the most credible source of information because blogs provide more depth and more thoughtful analysis than is available in other media. The 2006 AEJMC annual conference witnessed a proliferation in research papers on blogging issue. Several articles discussed bl og credibility in the journalism, political communication and public relations fields. Armstrong and McAdams (2006) examined how gender and occupational cues of bloggers influen ced weblog credibility through an experimental design. They found that while male authors were d eemed more credible than female authors in a main effect, that difference disappeared among bl og users. The relationship between gender cues and credibility was moderated by blog usage. Ka id and Postelnicu (2006) were concerned with whether young citizens find the political informa tion posted on blogs credible enough to use it for decision making. They tested how young citizens ev aluate expert sources (a career politician), trustworthy sources (an independe nt group) and socially attractiv e sources (a celebrity). Their findings lack of statistical significance may be related to media consumption. Because the Internet is the top source information of college students, they tend to trust that source and accept its validity without ch allenging the sources. Furthermore, using a national telephone su rvey, Banning and Trammell (2006) revisited the issue to explore the relations hip between credibility, third-pers on effect, and likelihood to act. In addition, Trammell et al. (2006) investigated the use of bl og in the communication industry
35 using an online survey of professionals working in journalism and public relations. It appeared that practitioners are using blogs as extensions of their traditional work and active users are more likely to think blogs more cred ible. Also, Trammell and Metzgar (2006) explored how two types of blogs, independent personal bl ogs and organizational blog, to explore the impact of source credibility on relationship outcomes during a cris is situation by an experimental method. Their findings indicated that those w ho read both blogs perceived lowe r levels of crisis for an organization than those who were not exposed to blogs. A reverse relationship between perceptions of crisis and blog credibility was identified. That is, if audiences believe the credibility of a blog is low, they are more likel y to perceive high crisis. These promising clues may suggest blogs have the pot ential as an organizational crisis management tool. Highlights of Source C redibility Research Traditional persuasion research has revealed a long-standing interest in source credibility. It has long established that char acteristics of a message source can affect attitudes. One of the first researchers to explore source credibility wa s Hovland. In fact, he and his colleagues devoted an entire chapter of their book Communication and Persuasion to the topic of communicator credibility. In this chapter, they emphasized the important ro le the source has on persuasion. They claimed that source credibility influen ces persuasion and opinion change: High (vs. low) credibility source genera ted better evaluations of the presen tation and opinion change in the direction advocated by the comm unicator. On the contrary, a co mmunication by a low credibility source was perceived as less fair and less justified than the same information was given by a high credibility source (Hovland et al., 1951). Subsequent studies have yiel ded similar findings. Johnson et al. (1968) observed people yielded to message more when it was delivered by a high credible source than by a low credible source. McGinnies and Ward (1974, 1980) reported th at a source identified as both expert and
36 trustworthy produced more attitude change in the direction advocat ed than a source lacking these two attributes. Maddux and Rogers (1980) obs erved that subjects agreement with a communication was higher when the source was an expert in the topic (vs. a non-expert) regardless of the presence or absence of supporting arguments. Credibility has two major components: expertis e and trustworthiness (Hovland et al., 1953; McGuire, 1969). The two concepts affected acce ptance of the message and changes in opinions. Expertise is how well informed and intelligent th e audience believes the source is (Hovland et al., 1953, p 21-23) and his/her ability to have a correct stand on the issue (McGuire, 1969). Trustworthiness is the degree of confidence the recipient has in th e sources intent or motivation to accurately communicate the information (Hovla nd et al., 1953, p 21-23). Since these original studies, many variables have been examined to assess source credibility. (Adoni, Cohen, & Mane, 1984; Gaziano & McGrath, 1986; Me yer, 1988; Slater & Rouner, 1996). Perceived expertise, bias, fairness, truthfulness, accuracy, amount of use, depth or completeness of message, prior knowledge and message quality have all emer ged as components of credibility. Although scholars disagree about the exac t number of dimensions measuring source credibility, trustworthiness and expertise are the two most widely used dimensions of source credibility (Dutta-Bergman, 2004). Ohanian (1990) studied th e credibility of celebrity spokespersons in advertisements and emphasized two dimensions of credibility, expertise a nd trustworthiness, as the most applicable to corporate perceptions. Th us, the scale development in this study focused on measuring consumer perceptions of expertise and trustworthi ness to form the essential aspects of the perceived source credibility concept. Credibility is a characteris tics defined by readers judgments. Many scholars defined credibility not as an objectiv e property of the source, but as a receiver perception (Gunter,
37 1992, p.148). Fogg et al. (2002a) held the same view. Cr edibility, in general, can be seen as an entire set of percepti ons that receivers hold toward a s ource (Bettinghaus, 1969). Therefore, in this study, the term credible refers to a perception of credibility rather than a direct measure of actual quality. Source credibility has been proposed as an im portant antecedent to attitudes toward an advertisements and ultimately, advertising e ffectiveness (Lutz et al., 1983). Many studies supported the generalization that perceived source (i.e., spokesp erson) credibility influence attitudes and behavior inten tions. MacKenzie and Lutz (1989) analyzed the relationships between advertiser credibility, attitude toward the ad, attitude toward the brand, and other important ad-related constructs. Their findings su ggested how important the credibility of an advertiser is in influencing consumers attitu des toward the advertisement. Specifically, MacKenzie and Lutz (1989) found significant rela tionships between advertiser credibility and attitude toward the advertiser and between advertiser credibility and ad credibility. In addition, their results reveal a significant link between attitude toward the advertiser and attitude toward the ad, which in turn significantly affects at titude toward the brand. Goldberg and Hartwick (1990, p. 173) also state the importance of cor porate reputation or cr edibility: As with spokesperson credibility, companies with positiv e reputations would seem to be in a better position to get consumers to believe their adve rtising claims. LaBarbera (1982, p. 223) agrees: A firm without a reputation has a problem with credibility; that is the firm is lacking in trustworthiness and expertise. In turn, this cr edibility deficit is likely to influence persons receiving the message to respond unfavorably. Perloff (2003) suggested authority, credibilit y, and social attractiveness as the main attributes of a trustworthy source. Authorit y persuades because it makes people compliant by
38 offering rewards or punishment in exchange for atti tude change; credibility consists of expertise, trustworthiness, and goodwill; and attractiveness play s into peoples desire to model or associate with socially or physically appealing persons (Perloff, 2003). Bloggers are communicators whose potential to influence is derived from either their credibili ty (i.e., expertise on a subject, or perception of being trustworthy, unbiased, and independent) or thei r social attractiveness (are famous, well regarded individuals). Although opi nions are split about which type of message source is more persuasive as the effectiveness of the source is influenced by many variables such as the audience characteristics, previous studies have indicated that expert and trustworthy sources tend to be rated slightly higher than socially attractiv e sources. To extend this concept to analyze corporate blogging, CEOs are often considered as experts in their industry because of the extensive managerial experience and insights they possess. In the advertising literature, using CEOs of companies to endorse their products an d services becomes more popular because some evidence suggests that CEOs are perceived to be more expert and likely to use the products/services endorsed (Poinde xter, 1983). Journalist bloggers often serves as industry insiders and are viewed as unbiased sources, which allows them to become online opinion leaders. Among these popular bloggers, the effects of source credibility seem to be intertwined. According to the Edelmans Trust Barometer re port (2006), blogs and bloggers could act as potential corporate credibility indexes as th e Internet remains steadily in the third place position (19 %) among media providi ng trustworthy information/news. When people consider the credibility of information provided about a company and the trustworthiness of the source, regular employees of a company are rated at 45 percent credibility; the CEO of company garners a 28 percent rati ng, and Public Relations executives ar e rated at 17 percent. It is noted that Blogger (13 %) has become a new index of credible informati on sources regarding a
39 company in the 2006 survey. Notably, according to 27 percent of people in the United States, employee or CEO blogs are considered to be one of the corporate attributes that build trust. In addition, the practice of a corpor ate CEO acting as a spokesperson to convey the information about a company has grown in credibility in the pa st three years in the United States (from 11 % in 2003 to 27 % in 2006). This empirical report provided some implications for using corporate blogs as a means to bolster trust in publics. When people are asked about the credibility of information sources about a company, they tend to choose a person like your self or your peer (meaning an average person) as the most cred ible person to communicat e information about a company, reflected in the 68 percent rating in the United States. Also, the majority of people (72 %) tend to believe multiple sources of informati on about a company. These survey results, which incorporate characteristics of blogs, are applicab le to the setting of corporate blogs and are indicative of public opinion. Resu lts suggested that providing a t op executives viewpoint as a lens with which to look at corporate polic y and products makes co rporate blogging a more credible format than corporate adver tisements and official newsletters. The image of public relations and its practitioners were sk ewed to negative (Callison, 2001). According to the National Cr edibility Index (1999) released by Public Relations Society of America, results revealed that the low ranki ng attained by public relations specialists. The study, involved more than 2,500 interviews, measur ed the extent to which the American public perceived its leaders and public fi gures to be believable sources of information and guidance on major issues. The study investigated the credibility leve l of 25 to 44 categor ies of leaders from Supreme Court justices, teachers and politicians to public relations specialists. With the average credibility index score at 61.5, lo cal news reporters received a 65.8, Wall Street executives by comparison received 57.9 and PR specialists received 47.6, which ranked near the bottom of the
40 list. The results suggested the public is generally cynical about the credibility of information provided by public relations specialis ts. Therefore, the study raised major issues for corporations to determine who the most credible figur es are to use as information sources. In summary, source credibility can be seen as the perceived expertis e and trustworthiness or truthfulness of firm. If corpor ate credibility is lacking, then any promotional message the firm presents to the public may elicit an unfavorable response (LaBarbera, 1982) Overall, credibility plays an important role in forming perceptions of corporate brand. The credibility of a firm contributes to building brand equi ty for its products, both tangible goods and intangible services (Newell & Goldsmith, 2001). Finally, the reputation or credibility of a firm also contributes to the success of its new products (Cooper, 1994). With exploratory and blending results from academic research and survey reports that indicate blogs could be new potenti al managerial tools to build corporate credibility, the issues of different source cues of bloggers and impact of bl ogs on perceived credibility still lack empirical supports. Credibility of corporate blogging remain s a burgeoning and unsolved issue. Therefore, this study aims to investigate the perceived credibility among differe nt corporate blogging sources and peer individuals. Research question 1: What is the difference in pe rceived source credibility among different sources of bloggers (CEO, journa list, peer individual, and public relations manager)? Level of Involvement Beyond establishing the impact of source cred ibility on persuasion, research has also explored the robustness of such effects across potential moderating conditions. Several studies have examined influence of source credibility under different levels of ability and motivation process by varying: ego involvement, the importa nce of the issue, exposure rate, direct or indirect experience with the pr oduct, the involvement with the issue (Eagly & Chaiken, 1993;
41 Petty & Cacioppo, 1986). Most studies found that when people were motivated and/ or able to process the message, source credibil ity did not affect persuasion; in other words, people were not persuaded more by a highly credible source than by a low credible source. However, they found that when people lacked the motivation and/ or the ability to process the messages, a high credibility sources was mo re persuasive than a low credibility source. Why should source effects depend on a person s motivation and ability to process the message? Both the elaboration likelihood model (ELM) proposed by Petty and Cacioppo (1986) and the heuristic-systematic model (HSM) de veloped by Eagly and Chaiken (1993) suggested two information-processing mechanisms. Accord ing to ELM theory, Pe tty and Cacioppo (1986) postulated that people will not devote the same amount of cognitive effort to every aspect of the message all the time. Instead, people elaborat e more or less along the continuum anchored by low elaboration (called the peri pheral route to persuasion) an d high elaboration (called the central route to persuasion). When people lack the ability and motivat ion to undertake extensive elaboration, persuasion via the peripheral route occurs. Rece ivers taking the peripher al route processing are persuaded by affective or heuristic cues. Thes e peripheral cues are s timuli devoid of issue relevant information that induce attitude ch ange without necessitati ng a careful and thorough scrutiny of the information. Typical peripheral cues used in persuasion re search are number of arguments, attractiveness of source (e.g., a beautifu l model in an advertisement) and the number of sources endorsing a position (Petty & Cacioppo, 1986). In the central route, the audience will devot e a higher amount of cognitive effort to the message and will scrutinize the true merits of th e arguments presented in the message (Petty & Cacioppo, 1986). People are more likely to take th e central route processi ng when they are (1)
42 believe an argument has a personal relevance, (2) are knowledge about a given topic, (3) are motivated, (4) are able to process the informa tion in a message (Petty & Cacioppo, 1986). In other words, level of involvement and familiarity of the issue will moderate how readers process the information and evaluation. In the parlance of the ELM, many studies have suggested the view that the credibility of the communication source acts as a peripheral cue in the formati on of attitudes under conditions of low elaboration (McGinnies, 1973; McNeil and Stoltenberg, 1988; Moore et al., 1986; Ratneshwar and Chaiken, 1991; Wu and Shaffer, 1987). When elaboration is low, the presence of a high credible source may enhance persuasi on regardless of the arguments used in the communication. These findings that manipulated involvement supported source credibility affected persuasion at low elaboration levels. Although source credibility is gene rally perceived to serve as a peripheral cue, it may also affect persuasion in the presence of high elaboration by serving as an argument germane to the product. A sources endorsement may be viewed by consumers as a highly relevant piece of information when evaluating products. When the recommendation comes from someone possessing high credibility (e.g., a doctor), persua sion should be enhanced so long as consumers engage in sufficient high levels of elaboration. People will vi ew source credibility as an additional argument in the message to affect pe rsuasion. In other words, if consumers are asked to list the salient reasons behind buying the products, they will incl ude the endorser as one of the reasons that persuade them. Gotlieb and Swan (1990) have purported to show that source credibility can serve as an argument in hi gh elaboration conditions. Based on previous discussions, the following hypothesis was formed: Hypothesis 1: The respondents level of involvement of the given issue will moderate the effect of source cues on att itude toward a company.
43 Undoubtedly, corporate blogs are giving establ ished companies and obscure brands alike the ability to connect with their audiences on a more personal level, build trust, collect valuable feedback and foster strengthene d business relationships. The ne xt section discusses how the consumers attitudes towards a br and are evolved and measured. Attitude and Brand Literature Attitude Attitude has been defined as a construct that combines belief, affect, and conation intervening between stimulus and response. Allpor t (1935) considered it as one of the most unique and essential concepts in modern soci al psychology. Conceptually, an attitude is a psychological tendency that is e xpressed by evaluating a particular entity with some degree of favor or disfavor (Eagly & Chaiken, 1993). Eagly and Chaiken (1993) defined psychological tendency as a state that is internal to the pe rson and evaluating as all classes of evaluative responding, whether overt or covert, c ognitive, affective, or behavioral. According to Fishbein and Ajzen (1975), attitude is a function of his/her salient beliefs at a given point in time (p. 222). Mitchell and Olso n (1981) defined attitude as an individuals internal evaluation of an object such as a branded product (p 318). Beliefs are the subjective associations between any two diffe rentiable concepts and salient be liefs are those activated from memory and considered by the person in a gi ven situation (Fishbein & Ajzen, 1975). Attitudes play a key role in predicting purchase behavior for particular brands. For this reason, much study has concentrated on the cognitive and affective de terminants of attitudes in hopes of predicting the conative factor. The a ffective link has become the main play er in todays marketplace (Batra, Myers, & Aaker, 1996). Assessing individuals opinions, attitudes, a nd preferences becomes extremely important when seeking to evaluate the overall impact or effectiveness of a particular public relations
44 program or campaign. Lindenmann (2002) asserted th at attitude research measures not only what people say about something, but also what they know and think (their mental or cognitive predispositions), what they feel (their emoti ons), and how they are in clined to act (their motivational or drive tendencies). Given the impor tance of attitude research in measuring public relations outcomes, the relationship between pe rceived credibility and consumers attitudes toward brands needs to be further investigated. Attitude toward the Brand (Ab) Brand attitudes are considered as important phenomena in consumer behavior, marketing, and advertising. Brand attitudes are defined in terms of consum ers overall evaluations of a brand (Kim, 2003). While substantial studies have been created to measuring attitude toward the brand and purchase intention as the effect of attitude toward advertising, no empirical research has looked at the effect of source credibility on attitudes on the blogoshpere. Attitude toward the brand is defined by Ajzen and Fishbein (1980) as a predisposition to respond in a consistently favorable or unfavorable manner to a particul ar brand. Mitchell a nd Olson (1981) defined attitude toward the brand as consumers overa ll evaluation of good or bad. Semantic differential scales measuring brand attitude are frequently used in the mark eting and advertis ing literature. Bruner and Hensel (1996) summarized 66 publis hed studies which measured brand attitude, typically as the dependent variab le in research on product line exte nsion or advertising effects. Brand attitude serves a key stra tegy to differentiate from mark et competitors. According to Olins (2000), Brands are the device we use to differentiate between otherwise almost indistinguishable competitors. Without clear brandi ng, in some fields, we literally could not tell one product or service from another (P. 61). Olins (2000) also sugge sted that people can have a relationship with a brand: the y have an immense emotional c ontent and inspire loyalty beyond reason (p. 63). Olinss (2000) di scussion suggests that brands c ould consist of the following
45 three factors: the behavior of an organization often defined as a component of organizational identity, communications/messages to define diffe rentiating attributes of an organization or product, or relationships with an organization as people c onceptualize that organization. In public relations scholarship, Ledingha m and Bruning (1998) found a link between relationships and public loyalty toward an organization. Ledingham and Brunings (1998) research showed that consumers who ranked the or ganization high with rega rd to the dimensions of trust, openness, involvement, commitment, and investment said they would stay with that company in the face of competition. In other words, the research showed that building effective relationships can center evaluation of public rela tions activities on attitu dinal and behavioral outcomes (Bruning, 2002). In addition, Kim and Chan-Olmsted (2005) conducted an empirical study to link the effects of orga nization-public relationships on attitude toward the brand and purchase intention via brand attitude; and the e ffects of product-related attributes on attitude toward the brand and purchase intention via the brand attitude. They found that organizationpublic relationships were significant ly related to attitude toward the brand. Also, attitude toward the brand was a strong mediator between inde pendent variables (OPRs and product-related attribute beliefs) and purchase intention. Recently, studies have begun to investigate the potential of relationship maintenance strategies pertaining to blog co mmunication, which enables corpor ations to build positive brand attitudes. Kelleher and Miller (2006) found that organizational blogs can improve relationship outcomes through perceived relational strategies which facilitates communication process beyond what regular web pages can do. Tramme ll and Metzgar (2006) further confirmed organizational blogs can achieve more desirable relationshi p through conversational human voice and responsiveness/customer service factor. Although the stre ngth of corporate blogging is
46 ubiquitous, little relationship betw een perceived credibil ity and attitude toward a company on the blogosphere has been explored. This stimulates to form the following research questions: Research question 2: Will different source cues affect respondents attitude toward a given company? Research question 3: How is the perceived credibility of different source cues related to attitude toward a given company? Theory Development Draw on the literature review in the previous section, th e conceptual model of the relationships among different sources of blogging, perceived credibility an d attitude toward a company were proposed a nd tested (Figure 2-1). Research Questions and Hypotheses Based on the preceding literature review, the study explored the following research questions and hypotheses: Research question 1: What is the difference in pe rceived source credibility among different sources of bloggers (CEO, journa list, public relations manager and peer individual)? Research question 2: Will different source cues affect respondents attitude toward a given company? Hypothesis 1: The respondents level of involvement of the given issue will moderate the effect of source cues on at titude toward a company. Research question 3: How is the perceived credibility of different source cues related to attitude toward a given company?
47 Independent Variable Dependent Variables Figure 2-1. Theoretical model of s ource credibility on the blogosphere. Source Cues CEO as a blogger Journalist as a blogger Peer individual as a blogger Public Relations Manager as a blogger Perceived Source Credibility Trustworthiness Expertise Attitude toward a Company
48 CHAPTER 3 METHODOLODY Conducting an Experimental Design This study employed a 1 by 4 (difference sources: CEO blog, journalist blog, peer individual blog, public relations manager blog) factorial experimental design with a control group, which received no blog information. The purpose of this study was to measure the credibility ratings among diffe rent sources of online corpor ate messages and explore the relationship between and perceived credibility an d attitude toward a company. In other words, this study examined the interaction effect be tween source cues, perc eived credibility, and participants attitudes towards a company. Respondents level of involvement was treated as a moderator variable that was possibly predictive of the outcome under this study. Given the purpose of this study, an experiment al design was considered as an appropriate approach to answer research questions and hypotheses. The e xperimental method gains major advantages in that it allows researchers to es tablish causality between two or more variables (Wimmer & Dominick, 2003). In terms of source credibility research, most studies were completed by using experimental designs. In addi tion, Parasuraman et al. (1988) suggested that in order to evaluate the quality of an organiza tions service, a good approach is to measure the publics perception of it. Although experiments ge nerally lack external validity, the copy test situation is more real than nor mal, in that it incorporates real news event for an existing brand, yet allows for control of the variables in question. Stimuli Four blogs were created to serve as stimuli for this experiment. The discussion topic used in this study was Google, the leading searching engine on the Internet, announced on October 9, 2006, that it bought popular online video site Y ouTube for $1.65 billion in stock. YouTube,
49 founded in February 2005, has quic kly become the most well-known of several online video sites. More than 100 million videos, many of which are short videos created by the sites users, are downloaded a day on the site. Acco rding to the Internet research firm Hitwise, YouTube has about a 46 percent share of the online video ma rket. The combination of Google and YouTube aroused plenty of discus sions on the blogosphere. The rationale for choosing the event was that technology companies ar e usually known as early adopters of innovative te chnology. Corporate blogging is a common strategy for Internetsavvy technology companies. In fact, the Google Company holds an official blog space in an effort to provide insights into product, technology news and their unique company culture. The Google blog features that ha ving different company employees voice their experiences helps users learn about the company and its people. In addition, given the significant attribute of this issue and high visibility of Google Company, college students will be familiar with ongoing events and may be interested in reading blog posts for more in-depth information. After reading blogs written by different sources on this topic, I composed three blog posts, which I presented on four different blogs. In each case, the blog contents and layouts were identical, but the source of the blogs was identified as different. The first blog was attributed to the CEO of Google Company, Dr. Eric Schmidt (http://ceoofgoogle.blogspot.com/index.html ). He led that companys strategic planning, mana gement and technology development (Figure B-1). The second blog was held by public relatio ns manager of Google Company, Elliot Schrage. He is a lawyer and business a dvisor with 20 years of experien ce at the intersection of global business strategy and public policy (http://publicrelaions2.blogspot.com/ ) (Figure B-2). The third blog was established by a senior male editor of BusinessWeek Magazine who specialized in technology and the Internet issue for fifteen years (http://bweditor.blogspot.com/ ) (Figure B-3).
50 The fourth blog was maintained by a college ma le senior student majoring in computer engineering who was considered to be a peer individual so urce among college participants (http://cbradshow.blogspot.c om/) (Figure B-4). Each blog contained one post discussing the advantage of combina tion that brought to online users and arguments on disputing copyrig ht issues of YouTube (Appendix D). Another post, titled Not trapping users data= G ood, presented the philos ophy of Googles service (Appendix E). The other post kept some thoughts on responding to an allegation from other bloggers (Appendix F). All four blog s were identical in design: par ticipants visibl y identified the owner through an introductory paragraph of the given author (Appendix C). Blog posts were in reverse chronological order and writ ten in a personal voice and informational tone. The length of each post was approximately 500 words because it was estimated to be substantial but not burdensome for an experiment with college students (Armstrong & McAdams, 2006). The gender cue of blog authors was controlled since f our bloggers in the stimuli were all males. Sample and the Treatment A convenient sample was used in this study. It was considered appropriate to use a convenient sample due to the theo retical testing nature of the e xperiment (Calder, Phillips, & Tybout, 1981). Calder, Phillips, and Tybout (1981) argued that a homogeneous sample for an experiment permits more exact theoretical pr edictions than from a he terogeneous group. In addition, a homogenous sample also decreases th e chance of making a false conclusion about whether there is a covariati on between the variables under study (Cook & Campbell, 1979). Participants were recruited from college students at a large southeastern university. College students are heavy users of comput ers and Internet. They are more likely to have the Internet access and tend to be more familiar with blog is sues, have personal blogs or involved in the discussions on the blogosphere. Ac cording to a Perseus study, the bulk of bloggers are within the
51 under-30 demographic. Over 90 per cent of blogs are authored by people between the ages of 13 and 29, with 51.5 percent between the ages of 13 and 19. Another survey authored by Viegas (2004) for the MIT Media Lab, estimated 46.3 percen t of blogs are starte d by people between the ages of 21 and 30. Therefore, college stude nts were relevant for the present study. A convenience and volunteer sample was use d. A convenience sample is a collection of readily accessible subjects for st udy while a volunteer sampling m eans that subjects are willing to participate in the research project voluntar ily. The advantage of the sampling strategy was convenience, high response rate and cost-effective. Also, mass media researchers frequently use nonprobability sampling, particularly in the form of available sample s and volunteer samples (Wimmer & Dominick, 2003). Although convenience sampling was constantly employed in the past research, one major limitation of this sa mpling strategy is gene ralization. Although a university experimental design is likely to generate a relatively di verse sample of students, the results of the present stu dy cannot be generalized be yond this specific population. Four blogs were established to serve as stim uli in this study. Participants were randomly assigned to one of five conditi ons: (1) CEO source, (2) public relations manager source, (3) BusinessWeek editor source (4) college senior st udent, (5) control group, which received no blog information (Table 3-1). In each cell, about 30 pa rticipants are recommen ded. Therefore, a total of 150 respondents are required for this experiment. Pilot Study In order to make sure the manipulation check s of different sources work successfully, a pilot test with 30 participants were executed before the main st udy. All students were recruited on the campus of the University of Florida and as signed to one of the four treatment conditions randomly.
52 From the pilot test, a successful manipulati on of different source cues was obtained. 93.3% (28 participants) can clearly det ect the author of blog, which sugge sted that the manipulation of source cues on the blogoshpere was successful. Main Study Participants for the experiment were recruite d from University of Florida through sending invitation emails to listserv. The advantage of this method was that it allowed students to complete the experiment at their convenience. Students read an introduction paragraph in which explained the research topic a nd purpose of this study and volunt eered to participate in the research project. Procedure The procedure followed two steps and took approximately 20 minutes to complete. Participants were randomly assigne d to the stimuli, and filled out posttest questions. Four Web site links of blogs were randomly attached in the invitation email. Thus students were randomly assigned to one of the four blogs and browse the posts online. After reading the blog entries, participants proceeded to a Web-based surv ey asking their evaluation on perceived source credibility, attitude toward th e Google Company, their general Internet and blog use, level of involvement and basic demographic information (Figure 3-1). The survey was comprised of scales adapted from previous studies meas uring blog usage (Armstrong and McAdams, 2006), source credibility (Ohanian, 1991; Pornpitakpa n, 2003) and brand attitude (Woo, 2001). Operational Definitions: Independent Variable Different source cues Source cue was an independent variable and ma nipulated into four di fferent authors: CEO of Google Company source, editor of BusinessW eek Magazine source as a senior industry watcher, college senior student as a peer in dividual and public rela tions manager of Google
53 Company source. After participants exposed to the stimuli, one question was proposed to check the manipulation asking them to identify the blog authors. A moderator variable: Level of involvement Involvement was a readers characteristic that moderates how individuals process information. In this study, involvement was define d in terms of issue involvement which can be explained by the extent to which the attitudi nal issue under considerat ion is of personal importance (Petty & Cacioppo, 1979, p.1915). Persona l relevance occurs when people expect the issue to have significant consequences for their own lives (A psler & Sears, 1968). Therefore, it is clear that partic ipants with high level of involveme nt would be more motivated to engage in thoughtful consideration of the issu e and gauge the arguments presented in blogs because they would be affected personally wher eas in low involvement conditions, they would not. Involvement level was measured using four questions which closely paralleled by the study of Petty, Capcioppo and Schumann (1993 ) and Lin (2005) (Table 3-2). It asked participants to evaluate their in terests on a 5 point Likert scale: (1) How important are the issues about the development of technology companies to you personally?; (2) How important are the issues about the development of technology prod ucts and services to you personally?; (3) How much are you concerned about the issues on th e development of technology companies?; (4) How much are you concerned about the issues on the development of technology products and services? Operational Definitions: Dependent Variables Perceived source credibility Despite the variety of dimensi ons on operationalizing source credibility, source credibility has been generally considered as consisting of tw o major aspects: trustworthiness and expertise.
54 This study included these two aspe cts to evaluate perceived sour ce credibility. The questionnaire contained 10 seven-point se mantic differential items taken from Ohanian (1990) and Pornpitakpan (2003) to measure celebrity endorse rs credibility. The scale has shown to have high reliability and validity with a reliability coefficient of 0.82 or higher (Ohanian, 1991). Pornpitakpan (2003) further validated the same s cale within Asia countries context. The results showed that the original scales factor st ructure fitted the Sing aporean data well and demonstrated satisfactory reliability. The words dependable/undependable, honest/dis honest, reliable/unr eliable, sincere/ insincere and trustworthy/untrust worthy anchored those measuri ng trustworthiness. The words expert/not an expert, experienced/i nexperienced, knowledgeable/unknowledgeable, qualified/unqualified, and skilled /unskilled anchored the items measuring expertise. Attitude toward a company To measure attitude toward the company (the dependent variable), the most frequently used multi-dimensional scale (unfavorable-favorable, bad-good, dislike-like, and negativepositive) in the 1990s Journal of Advertisi ng was adopted (Woo, 2001). Studies in the 1990s using that multi-dimensional scale reported hi gh reliability coefficients of items, ranging from .84 to .97. Brand attitude has been one of the most widely examined constructs in consumer behavior (Berger & Mitchell, 1989). The present study a dopted Wilkies (1990) conceptualization and defines attitude toward the brand as consumers overall evaluations of a brand. In the present study, attitude toward a company was measured us ing a seven-point semantic differential scale with regards to four questions. The scales ( unfavorable-favorable, ba d-good, dislike-like, and negative-positive) were borrowed from prior research by Holbrook and Batra (1987). Their
55 reliability coefficient alpha for these items wa s .98, suggesting the brand attitude measure had extremely high internal consistency. In addition, th irteen advertising studies that used the same multi-item scales reported Cronbachs alpha s ranging from .84 to .97 (Woo, 2001). The respondents were asked, What is your attitude toward the Google Company after reading the blog posts? They answ ered on four semantic differenti al items using a 7-point scale regarding the following items: unfavorable /favorable, bad/good, dislike/like, and negative/positive. The survey ended with a section of demographic questions such as gender, age, and education.
56 Table 3-1. Conditions of 1x4 experi mental design with control group CEO source Editor of BusinessWeek blog Senior student blog Public Relations manager source Receive no blog information Group (1) Group (2) Group (3) Group (4) Control group Table3-2. Likert-type measurement items of involvement level Figure 3-1. Procedure fo r experimental groups. 1) How important are the i ssues about the development of technology companies to you personally? 2) How important are the issues about the deve lopment of technology products and services to you personally? 3) How much are you concerned about the i ssues on development of technology companies? 4) How much are you concerned about the i ssues on development of technology products and services? Randomly Assigned Expose to the stimuli Posttest Questions 1. Identify blog source 2. Evaluations of perceived credibility 3. Attitude toward Goolge 4. Level of involvement 5. General Internet and blog use 6. Demographics
57 CHAPTER 4 RESULTS Overview of Statistical Analysis To analyze the data collect ed for the present study, the SPSS program 14.0 version was used. The data set contained a total of 177 cases. Th e analyses of this study involve two sections. The first section provides descriptive statistics of all respondents involved in the experiment. The second part is devoted to addres sing the research questions and hypotheses. To answer the first and the second research questi on, a MANOVA analysis was used to investigate whether there are differences in respondents perceived source credibility and attitudes after they expose to different source cues. Correlation and regression analysis was run to answer the third research question exploring relationship between perceive d source credibility and attitude toward the company. To test the hypothesi s, a MANOVA analysis was perf ormed to determine whether level of involvement moderates th e effect of perceived source credibility on attitude toward a company. Profile of Participants All respondents used for analysis in this e xperiment were universi ty students, including undergraduate and graduate students at the University of Florida. A total number of 177 students participated in this study. Of the total 177 part icipants, 82 (46.3%) were males, 95 (53.7%) were females (Table 4-1). The mean age was 24.58 years old. Most of them were graduate students (61 %), and the remainder were undergraduate students. In terms of academic classificati on, 6 respondents (3.4%) were freshmen, 8 (4.5%) were sophomores, 20 (11.3%) were juniors, 30 (16.9%) were seniors and 108 (61%) were graduate students. Respondents age ranged from 18 to 36. However, 67.8% of respondents were in the 18-25 age group. One-third of part icipants were between 26-35 years old.
58 One hundred and seventy-seven participants we re randomly assigned to one of the five groups. Thirty-one participants were assigne d in the control group, and the remaining 146 participants were assigned in th e following treatment groups: 1) Thir ty-six in the blog attributed to the CEO of Google Company group, 2) Thirty-sev en in the blog attributed to the editor of BusinessWeek Magazine group, 3) Thirty-seven in the blog held by a senior student majoring in computer engineering group, 4) Thirty-six in th e blog written by the public relations manager of Google Company group (Table 4-2). Chi-square te st showed that there were no significant differences in age, gender and education among these five groups. General Usage of Internet and Weblogs In terms of the general usage of Internet, 99.4 % (n = 176) of respondents use a computer and 175 (98.9%) use the Internet at least occasionally. About si xty-seven percent (n = 118) of them have been Internet users for six years or more. Ninety-one percent (n = 161) of respondents go online for getting information several times a day. About sixty-seven percent (n = 118) of them spend 2-5 hours getting news or information from the Internet everyday. When asking about general usage of Weblogs, 62.1% (n = 110) of respondents do have a personal blog, 58.2% of them read 2-5 blog posts and 68.4% read 5 blogs or less in the past week. Forty-five percent of par ticipants spend one hour reading and posting bl ogs per week. Half (58.2%) of them never write posts about companie s and their products (n = 103). One-third write posts about companies and their products less than once a week. Manipulation Checks Consistent with the results from the pilot test, the manipulation check of source cues worked successfully. There are 151 students w ho were randomly exposed to four treatment groups and turned in survey results. Approximate ly ninety-seven percent (n=146) of respondents could clearly recognize the auth or through the introductory para graph on the blogs. Only five
59 participants provided other answers such as M SN space, dont know, in terms of replying to blog author question and they did not complete all the posttest ques tions. Thus, the five participants were eliminated from analysis. Reliability Check for Dependent Measures The present study averaged th e value of all items for the dependent variable. As a prerequisite for averaging, the scale items with in each variable should have a high internal reliability. In order to establish th e reliability of dependent measures used in this study, reliability analyses were conducted for each of construct, including perceived source credibility, attitude toward the company and level of involvement. Consequently, Cronbachs alpha was computed to evaluate if the items within each index have high internal reliability. Alpha is a coeffici ent that indicates how well the items measuring the same characteristic correlate with one another (Hon & J. E. Grunig, 1999). Generally, reliability coefficients over .90 are considered excellent, over .80 are very good, and values over .70 are perceived as ad equate (Kline, 1998). The index of trustworthiness and expertise cons ists of five questions respectivel y, using a 7-point semantic differential scale. Participants attitudes toward the comp any were evaluated by a 7-point semantic differential scale with four items Level of involvement was measured though 5-point likert scale addressing f our questions. Reliability tests confirmed that the indexes of overall trustworthiness (Cronbachs =.91), expertise (Cronbachs =.94) and attitude toward the company (Cronbachs =.92) applied in this research were appropriate. Therefore, the statistical analysis indicated th at the dependent measures used in this experiment had a high internal reliability. Table 4-5 il lustrates that Cronbachs alpha of each scale exceeded .80, which means that all scales can be us ed statistically in this study.
60 Research Questions and Hypotheses Testing Test of Research Question 1 Research Question 1: What is the difference in pe rceived source credibility among different sources of bloggers (CEO, journali st editor, public relations manager and peer individual)? To answer research question 1, a multivar iate analysis of variance (MANOVA) was performed with different sources as the independent variable a nd measures of trustworthiness and expertise as the dependent variables. MANOVA is a technique which determines the effects of independent categorical vari ables on multiple continuous dependent variables. It is usually used to compare several groups with respec t to multiple continuous variables. MANOVA is useful in experimental situa tions where at least some of the independent variables are manipulated. It has several advantages over ANOVA. First, by measur ing several dependent variables in a single experiment, there is a be tter chance of discovering which factor is truly important. Second, it can protect against Type I errors that might occur if multiple ANOVAs were conducted independently. Additionally, it can reveal differences not discovered by ANOVA tests (Huck, 2000). The test yielded a significant source cue effect s on perceived trustwor thiness, F (3, 143) = 14.44, p<.001. The results in Table 4-6 and Table47 indicated that there were significant mean differences among the four treatment groups. The m ean score of the trustw orthiness scales for the CEO source group was 5.39, the highest one among four groups. The mean score of the trustworthiness scales for the senior student source group was 4.14, the lowest one among four groups (Figure 4-1). In an effort to understand the true pattern of the population means, a Scheffe post hoc comparison test was undertaken (Huck, 2000). Ta ble 4-8 revealed the mean scores of trustworthiness for the CEO and the editor sour ce group was significantly higher than those of
61 the senior student and the PR manager group. In addition, the mean score of trustworthiness for the PR manager group was significan tly higher than that of the senior student group. In other words, participants in the CEO and editor so urce group rated the highe st score on perceived trustworthiness while the senior student source group was rated the lowest. In addition, participants in the PR manger source group had a lower score on perceived trustworthiness than the CEO and the editor group. Consistent with results of perceived trustworthiness, the MANOVA test indicated a significant source cue effect on perceived expertise, F (3, 14 3) = 30.91, p<. 001. The results in Table 4-9 and Table4-10 summari zed that there were significant mean differences of perceived expertise among the four treatment groups. The m ean score of the expertise scale for the CEO source group was 6.06, the highest one among four groups, and PR manager was ranked the second regarding expertise (M = 5.00). The mean score of the expertise scale for the senior student source group was 3.91, the lowest one am ong four groups (Figure 4-2). The Scheffe post-hoc test was run to compare means among fo ur groups. Table 4-11 revealed the mean score of expertise for the CEO source group was signi ficantly higher than thos e of the other three groups. In addition, the mean scores of expertis e for the PR manager and the editor group was significantly higher than that of the senior student group. Test of Research Question 2 Research Question 2: Will different source cues affect respondents attitudes toward a given company? The test indicated a significant source cue e ffect on attitude toward a company, F (4, 173) = 20.90, p<.001. The results in Table 4-12 and Ta ble4-13 summarized that there were significant mean differences of attitudes between the four treatment groups and the control group. The mean score of attitudes for the CEO source group was 5.91, the highest one among four
62 groups, and the attitude of edito r group was ranked the third (M = 5.01). Similarly, the mean score of attitudes for the senior student s ource group was 4.20, which was perceived as the lowest one among five groups (Figure 4-3). The attitude of the PR manager source group was slightly higher than the se nior student group (M = 4.67). It was interesting that the attitudes of contro l group were higher than that of the editor, senior student and PR manager group. This c ould be explained by the strong brand equity and positive attitudes toward the Google Company am ong general publics. According to Forbes.Com, Google made its debut in the number one position on Fortunes 10th annual 100 Best Companies to Work For list in 2007, which suggested the brand equity of Google was becoming stronger and stronger. To shed light on comparing the means between five groups, a Scheffe post-hoc test was run. Table 4-14 revealed that the mean scor e of attitudes for the CEO source group were significantly higher than those of the other three groups. The mean score of attitude for the editor group was significantly higher than that of the senior student group. The attitudes of control group were significantly higher th an that of the PR manager source and senior student group. Test of Hypothesis 1 Hypothesis 1: The respondents level of involvement of the given issue will moderate the effect of source cues on at titude toward a company. In order to answer the hypothesis, the le vel of involvement of participants was dichotomized into two groups. The level of i nvolvement was measured through 5-point likert scale. The mean score of 146 participants in four treatment groups was 3.76 (SD = 0.818). To better understand the distribution of the data, the mode and th e median were 4. Thirty-four people out of 146 answered 4 (somewhat important ) when asking their level of involvement.
63 In that regard, it seemed impossi ble for the researcher to divide the sample into two groups that had the same number of people. Thus, the data was divided into two groups by overall mean score. Participants of mean scores between1 to 3.76 were categorized into the low involvement group. Scores of 3.76 to 5 were labeled as the hi gh involvement group. Sixty-six were in the low involvement group and eighty were in the high involvement gr oup. Table 4-15 revealed that there was no significant interacti on effect of source cue and in volvement on attitude toward a company. Therefore, Hypothesis 1 was rejected. Ho wever, the level of involvement still had a significant effect on the attitude toward a company ( p<.001 ). Test of Research Question 3 Research Question 3: How is the perceived credibility of different source cues related to attitude toward a given company? In order to determine any si gnificant relationships between perceived source credibility (trustworthiness and expertise) and attitudes toward the compa ny, a Pearson correlation analysis was undertaken. A Pearson correlation coefficient is a measure of association which caries from -1 to +1, while 0 indicating no linear relationship, -1 indi cating a perfect negative linear relationship and +1 indicating a perfec t linear relationship (Garson, 2005). In Table 4-16, a Pearson correlation coefficient showed that there was a positive correlation between perceived trustworthiness, perceived expertise and attitude toward the company. This finding suggested that perceived trustworthiness and perceived expertise were highly associated with attitudes toward the compa ny. It could be interpreted that when a person perceives the blog author as trustworthy and expert it is likely that he or she would have more positive attitude toward the company, which the blog author writes posts about or vice versa. However, whether a particular variable has a casual impact on a diffe rent variable cannot be determined by measuring the two variables simultaneously and then correlating the two sets
64 of data (Huck, 2000). Regression equations can be used to obtain predicted or fitted values of the dependent variable for given values of the independent variable. Th erefore, to gain more insight on relationships between perceived trustworthiness, perceived expe rtise and attitude toward the company, a multiple regression was utilized to pred ict and explain scores on attitudes toward the company based upon the information regarding pe rceived trustworthiness and expertise. First, the data were examined with a hier archical regression an alysis (Table 4-17). Attitude toward a company was the dependent va riable for this hierarch ical regression model. Two blocks of variables were en tered into the hierarchical regression analysis. Participants demographics consisted of age, gender, a nd education were dumm y-coded (Model 1). The variables of perceived trustworth iness, perceived expe rtise and level of involvement were also entered (Model 2). In this hierarchical regression analysis, th ree participants demographic variables were entered first. No significant influence was found from entering this block of variables. Since these demographics were considered as contro l variables in this e xperiment, it could be interpreted that individual differences that may hurt the results of this experiment were controlled. The second block of the analysis (Model 2) observe d a significant impact of perceived trustworthiness ( = .29, p <. 001), perceived expertise ( = .41, p <. 001) and level of involvement ( = .28, p <. 001) on the dependent variable. In addition, a multiple regression was perfor med, using perceived trustworthiness, expertise and level of involvement as the indepe ndent variables and attitude toward the company as the dependent variable. Table 4-18 to Table 4-21 gave the results of these three variables ability to predict attitude towa rd the company in four treatmen t groups respectively. In the CEO source group, the regression was a good fit ( R2 adj = 70.4%), and the overall relationship was
65 significant ( F 3, 33 = 28.779, p < 0.001). Only the positive effect of perceived expertise was significant ( =0.733, p < 0.001). In the editor source group, these variables accounted for 23.9% of the variance and the overa ll relationship was significant ( F 3, 34 = 4.776, p < 0.01). In the senior student group, these variables accounted for 75.1% of the variance and the overall relationship was significant ( F 3, 34 = 37.160, p < 0.001). The results indicated a significant and positive effect of perceived trustworthiness, expertise on attitude toward the company. In the PR manager group, these variables accounted for 22.8 % of the variance and the overall relationship was significant ( F 3, 33 = 4.443, p < 0.05). Perceived trustworth iness had a significant and positive effect on attitudes. To sum up, the results showed that perceive d trustworthiness, expertise and level of involvement could be effective predictors for ac hieving positive attitudes toward the company. A positive relationship between percei ved trustworthiness, expertise and level of involvement and attitudes toward the company wa s corroborated. Perceived expertis e could be inte rpreted as a reliable and significant contributor to positive attitudes in the CEO source group ( =0.733, p < 0.001). For the PR manager source group, perceived trustworthiness was a si gnificant contributor to achieve positive attitude toward the company ( =0.425, p < 0.05).
66 Table 4-1. Demographic profile of the respondents Frequency Percent Gender Male 82 46.3 Female 95 53.7 Total 177 100 Age 18-25 120 67.8 26-35 53 30.0 Over 35 2 1.1 Not Reported 2 1.1 Total 177 100 Education Level Freshman 6 3.4 Sophomore 8 4.5 Junior 20 11.3 Senior 30 16.9 Graduate Students 108 61.0 Not Reported 5 2.8 Total 177 100 Table 4-2. The number of participants in each cell Source Cues CEO Blog Editor of BusinessWeek blog Senior student blog PR manager blog Control Total Number/ Percent 36 (20.3%) 37 (20.9%) 37 (20.9%) 36 (20.3%) 31 (17.5%) 177 (100%)
67 Table 4-3. General usage of Internet Frequency Percent Years for being an Internet user Six months or less 4 2.3 A year ago 1 0.6 Two or three years ago 11 6.2 Four years ago 24 13.6 Five years ago 19 10.7 Six years or more 118 66.7 Total 177 100 How often for going online for getting information Several times a day 161 91.0 About once a day 13 7.3 3-5 days a week 2 1.1 Every few weeks 1 0.6 Total 177 100 Average hours on getting information from the Internet everyday 1 35 19.8 2-5 118 66.7 6-10 19 10.7 More than 10 5 2.8 Total 177 100
68 Table 4-4. General usage of Weblogs Frequency Percent Have a personal blog Yes 110 62.1 No 67 37.9 Total 177 100 Blog posts that you read in the past week 0 13 7.3 2-5 103 58.2 6-10 31 17.5 11-15 30 16.9 Total 177 100 Blogs that you read in the past week 0 9 5.1 1 40 22.6 2-5 72 40.7 6-10 33 18.6 More than 10 23 13.0 Total 177 100 Average hours on reading and posting blogs per week 0 13 7.3 1 80 45.2 2-5 63 35.6 6-10 19 10.7 More than 10 2 1.1 Total 177 100 How often for writing posts about companies and their products Never 103 58.2 Less than once a week 64 36.2 About once a week 4 2.3 More than once a week 4 2.3 Daily or almost daily 2 1.1 Blog devoted to product or company 0 0 Total 177 100
69 Table 4-5. Cronbachs Alpha of Variables Variable Cronbachs Alpha Perceived Source Credibility Trustworthiness .91 Expertise .94 Attitude toward the Company .92 Level of Involvement .89 Table 4-6. Perceived trustworthiness by source cue N Mean of Trust Scale SD CEO source 36 5.39 0.84 Editor source 37 4.89 0.80 Senior student source 37 4.14 0.72 PR manager source 36 4.33 1.19 Table 4-7. F-test of perceived trustworthiness by sources cue Source Sum of Squares df Mean squares F Between groups 35.27 3 11.76 Within groups 115.61 143 0.81 14.44 *** Note. *** p < .001 Table 4-8. Multiple comparison of mean difference of trust scale Mean Difference of trust scale (I J) CEO source (J) Editor source Senior student source PR manager source CEO source (I) 0.50 1.25* 1.06* Editor source 0.75* 0.56* Senior student source -1.872* PR manager source Note. 1. Scheffe post-hoc test was performed. 2. The mean difference is si gnificant at the .05 level.
70 Table 4-9. Perceived expertise by sources cue N Mean of Expertise Scale SD CEO source 36 6.06 1.06 Editor source 37 4.93 0.80 Senior student source 37 3.91 0.85 PR manager source 36 5.00 1.08 Table 4-10. F-test of percei ved expertise by sources cue Source Sum of Squares df Mean squares F Between groups 84.24 3 28.08 Within groups 128.99 143 0.91 30.91 *** Note. *** p < .001 Table 4-11. Multiple comparison of mean difference of expertise scale Mean Difference of trust scale (I J) CEO source (J) Editor source Senior student source PR manager source CEO source (I) 1.13* 2.15* 1.06* Editor source 1.02* -0.70 Senior student source -1.09* PR manager source Note. 1. Scheffe post-hoc test was performed. 2. The mean difference is si gnificant at the .05 level. Table 4-12. Attitude toward the company by sources cue N Mean of Expertise Scale SD CEO source 36 5.91 0.80 Editor source 37 5.01 0.93 Senior student source 37 4.20 0.77 PR manager source 36 4.67 0.84 Control Group 31 5.48 1.03 Table 4-13. F-test of attitude to ward the company by sources cue Source Sum of Squares df Mean squares F Between groups 64.14 4 16.03 Within groups 131.96 173 0.77 20.90 *** Note. *** p < .001
71 Table 4-14. Multiple comparison of attitude toward the company Mean Difference of attitude scale (I J) CEO source (J) Editor source Senior student source PR manager source Control group CEO source (I) 0.90* 1.71* 1.24* 0.43 Editor source 0.81* 0.34 -0.47 Senior student source -0.47 -1.28* PR manager source -0.81* Control group Note. 1. Scheffe post-hoc test was performed. 2. The mean difference is si gnificant at the .05 level. Table 4-15. Tests of between-subject effect s (dependent variable: attitudes) MS df F Intercept 2303.54 1 3867.17*** Source of author 2.80 3 4.71** Involvement 16.81 1 28.22*** Source of author*Involvement 1.39 3 2.34 Error 0.631 138 Note. 1. ** p < .01, *** p < .001 Table 4-16. Pearson correlation am ong perceived trustworthine ss, expertise and attitudes Perceived Trustworthiness Perceived expertise Attitudes Perceived Trustworthiness 1 0.687** 0.630** Perceived expertise 0.687** 1 0.741** Attitudes 0.630** 0.741** 1 Note. 1. ** p < .01 (2-tailed).
72 Table 4-17. Hierarchical regression analysis of the proposed model Attitude toward a company as a dependent variable Model 1 Model 2 Standard Error Standard Error Block 1 Demographics Age -.41 .06 .02 .05 Gender -.69 .06 -.09 .05 Education .01 .07 .01 .05 Block 2 Variables Trust .29*** .07 Expertise .41*** .07 Involvement .28*** .07 R2 .07 .63 Adjusted R2 .06 .62 F 11.06 59.66*** Note. *** p < .001 Table 4-18. Regression analys is of the CEO source group Dependent variable: Attitude toward the company R R2 Adjusted R2 F Sig. Beta t Sig. Perceived trustworthiness 0.078 0.614 0.544 Perceived expertise 0.733 4.714 0.000 *** Level of involvement 0.854 0.730 0.704 28.779 0.000 *** 0.089 0.690 0.495 Note. 1. p < .05, ** p < .01, *** p < .001 Table 4-19. Regression analysis of the editor source group Dependent variable: Attitude toward the company R R2 Adjusted R2 F Sig. Beta t Sig. Perceived trustworthiness 0.166 0.972 0.338 Perceived expertise 0.247 1.417 0.166 Level of involvement 0.550 0.303 0.239 4.776 0.007 ** 0.352 2.367 0.024 Note. 1. p < .05, ** p < .01, *** p < .001
73 Table 4-20. Regression analysis of the senior student source group Dependent variable: Attitude toward the company R R2 Adjusted R2 F Sig. Beta t Sig. Perceived trustworthiness 0.386 2.669 0.012 Perceived expertise 0.443 3.046 0.005 ** Level of involvement 0.878 0.772 0.751 37.160 0.000 *** 0.247 2.892 0.007 ** Note. 1. p < .05, ** p < .01, *** p < .001 Table 4-21. Regression analysis of the PR manager source group Dependent variable: Attitude toward the company R R2 Adjusted R2 F Sig. Beta t Sig. Perceived trustworthiness 0.425 2.081 0.046 Perceived expertise 0.233 1.304 0.202 Level of involvement 0.542 0.294 0.228 4.443 0.010 0.316 1.803 0.081 Note. 1. p < .05, ** p < .01, *** p < .001
74 Figure 4-1. Mean of trustworth iness among four source groups.
75 Figure 4-2. Mean of expertise among four source groups.
76 Figure 4-3. Mean of attitude to ward the company for 5 groups.
77 CHAPTER 5 DISCUSSION AND CONCLUSION The chapter begins with an overview of the present study, followed by a detailed analysis of conclusions related to the research questions and hypothese s. Next, a discussion of the implications for theoretical and practical perspe ctives in public relations is provided, followed by limitations of this research. Finally, the chapter concludes with recommendations for future research. Study Overview The present study represents an empirical examination of th e effects of source cues on perceived source credibility and attitude toward the company on the blogosphere. The level of participants involvement is also inve stigated as a moderator variable. As public relations scholars have emphasize d, the interactive features of Internet communication enable organizations to maintain a potential two-way dial ogic relationship with their publics. Started in 2004, Weblogs are gainin g popularity as an effect ive alternative public relations communication channel targeting Net citizens. Many me dia savvy corporations have started to venture on the blogosphere in an effort to manage re lationships with employees and consumers. Scholars have also confirmed that Weblogs have the potentia l to facilitate open dialogues (Seltzer, 2005; Trammell, 2004) and build trust and satisfaction in organization-public relationships (Kelleher & Miller 2006). Moreover, blogs present in a conversational form of communication, and the human quality of voice makes readers feel that the blog author is talking to them directly. The blending features of pe rsonal diary and Web pages enable Weblogs to become an effective channel which engages audiences.
78 Research has examined the applications of Weblogs in different theoretical frameworks, including use and gratifi cation theory, users profile of blogs and comparing credibility of blogs over traditional media channels. Little research has empirically addressed the key issues of source credibility and its effect on attitude toward a company. As Trammell (2005) pointed out, scholars should move from describing the conten ts of blogs to investigating the communication effects of this new medium. As such, the main objective of the present study was to explore whether different sources of blog authors could affect peoples percepti on of credibility and attitude toward a company. Also, the present st udy tested whether level of involvement can moderate the effect of perceived source credib ility on attitudes. In addition, the present study verified the causal relationships among attitude toward a compa ny, perceived source credibility (trustworthiness and expertise) and level of involvement. By performing an experimental study (N =177), th is inquiry sheds light on research of blog communication and persuasion. The results sugg ested the perceived source credibility (trustworthiness and expertise) a nd attitude toward a company can be affected after exposing to four popular sources of corporate blogging (CEO, industry watcher, PR manager, senior student). Attitude toward a company can be predicted by perceived trustworthiness, perceived expertise and level of involvement. From the public relations perspectives, this study contributed to e xplore the effectiveness of blog communication as a proa ctive way of managing relati onships. The results provide preliminary evidences for future research rega rding this developing fo rm of blog communication used in the context of cor porate communication settings. In addition, this study allows for greater understanding of the source credibil ity in corporate blogs. This study could inform the work of scholars interested in examini ng source credibility of Weblogs as well as aid in understanding
79 the strategic use of Weblogs in maintaining pos itive attitudes. As the diffusion of Weblogs continues to grow more prevalent (Tsai, 2006), the strategic communicati on of corporate blogs will garner more importance. Overview of Research Questions and Hypothesis In the following section, each of the thr ee research questions and one hypothesis is discussed in detail based on the results of this study. Research Question 1: What is the difference in pe rceived source credibility among different sources of bloggers (CEO, journa list, public relations manager and peer individual)? The first research question investigated the e ffect of different source sues on participants perceived source credibility, in cluding two dimensions of pe rceived trustworthiness and perceived expertise. To answer the question, a MANOVA analysis was performed. The results yielded a significant source cue effect on perc eived trustworthiness a nd perceived expertise (p<.001). With regard to trustw orthiness ratings, the CEO source was perceived as the highest one (M = 5.39), followed by the editor source (M = 4.89). The public relations manager was ranked the third (M = 4.33) and th e senior student source was perc eived as the least trustworthy (M = 4.14). In terms of perceived expertise, th e CEO source was perceived as the highest one (M = 6.06), followed by the public relations mana ger source (M = 5.00). The editor was ranked the third (M = 4.93) and the senior student source was perceived as the least expertise (M = 3.91). In order to better understand the mean differe nces among four treatment groups, a Scheffe post hoc comparison test was undertaken. The resu lts indicated participants in the CEO and editor source group rated the highe st score on perceived trustworth iness while the senior student source group was rated the lowest. Additionally, pa rticipants in the PR manger source group had a lower score on perceived trustworthiness than that of the CEO and the editor group. Moreover, the mean score of expertise for the CEO source group was significantly higher than those of the
80 other three groups. In addition, the mean score of expertise for the PR manager and the editor group was significantly higher than th at of the senior student group. To sum up, the CEO source was perceived as high ly trustworthy and e xpertise. The results were consistent with previous research findings. Corporations seek to employ credible endorsers who can add credibility to their communicati on messages and win trust among stakeholders (Kim, 2006). Bloggers are communicators whose potenti al to influence is derived from either their credibility (i.e., expertis e on a subject, or perception of being trustworthy, unbiased, and independent). In terms of corporate blogging, Ts ai (2006) found about half of corporate blog authors are the organizations CEO. The major motivations for corporate leaders to keep a personal blog are to provide useful informati on to blog readers and to share professional experiences regarding the industr y in which author works. As Edelman Trust Barometer report pointed out, CEOs are often considered as expe rts in their industry because of the extensive managerial experience and insights they possess. Al so, 27 percent of people in the United States think employee or CEO blogs are cons idered to be one of the corporat e attributes that build trust. In the advertising literature, using CEOs of co mpanies to endorse their products and services becomes more popular because some evidence sugge sts that CEOs are perceived to be more expert and likely to use the products/s ervices endorsed (Poindexter, 1983). The senior student source, which was define d as the peer individual source in this experiment, was perceived as the least trustwor thy and expert compared to the other three authors. The results did not indi cate a pivotal role that the pe er individual source has gained influence on the blogosphere. According to prev ious Edelman surveys, it revealed that 68 percent of people in the United States tend to choose a person like yourself or your peer (meaning an average person) as the most cred ible person to communicat e information about a
81 company when people are asked about the credib ility of information sources about a company. One explanation for the lack of support of this phenomenon is that most of participants were graduate students (61%) and they tend to cons ider college students as untrustworthily and inexperienced because age and level of knowledge may be higher than undergraduate students and thus skew the results. Therefore, to test the effect of source cues on perceived source credibility, future research shoul d consider including prior knowledge toward certain topics as another independent variable. Furthermore, participants who read the public relations manager source showed favorable ratings in perceived expertise but neutral opinions in perceived tr ustworthiness. In other words, participants believed the public relations manger was well informed and intelligent to reveal argument on the issue. However, peoples percei ved trustworthiness of public relations manger remained questionable. The results can be e xplained by existing resear ch on public relations, which reported that perceptions of public relations and its practitioners were skewed toward the negative (Newsom et al., 1993; Callison, 2001). Ca llison (2001) argued that these perceptions are based on peoples association of public relations with reporting bias (Eag ly et al., 1978) in favor of clients and their companies. Research Question 2: Will different source cues affect respondents attitude toward a given company? The second research question addressed the im pact of different source cues on attitude toward a company. The results showed a significan t difference in attitude change between four treatment groups and the control group. Consiste nt with previous finding, the mean score of attitudes for the CEO source group was 5.91, th e highest one among four groups, and the attitudes of editor group was ranked the third (M = 5.01). Sim ilarly, the mean sc ore of attitudes for the senior student source group was 4.20, which was perceived as the lowest one among five
82 groups. The attitude of the PR manager source was slightly higher than th e senior student group (M = 4.67). The results were predicted by some of the existing research on source trustworthiness. Priester and Petty (2003) reported that there was a main effect of endorser trustworthiness from their series of experiments. They reported that participants in their experiments had more favorab le attitudes toward a produc t which was endorsed by a high trustworthiness source than that of a low trus tworthiness source. They concluded subjects who were disclosed to a high trustwor thiness source condition had a mo re favorable attitude toward the company than subjects disclosed to a low trustworthiness source condition. The present study confirmed the effects of source cues on perceive d credibility and attit ude toward a company on the blogosphere. Specifically, the CEO and seni or industry watcher wo uld be perceived as bloggers who possess higher level of trustworthin ess and expertise. Thus, the higher perceived source credibility can lead to more fa vorable attitudes toward a company. One the other hand, it was intriguing that th e attitudes of the co ntrol group were higher than that of the editor, senior student and PR manager group. This could be explained by the strong brand equity and positive attitudes towa rd the Google Company among general publics. Google has been the number one search engine in the online market for years. In addition, thanks to large amount of positive coverage in the main stream media and free services they provide, the Netcitizens are highly acquainted with its in terface and all kinds of innovative services. Therefore, it was reasonable that the attitudes of control group which did not read any blog information reflected a fairly favorable attitude. Hypothesis 1: The respondents level of involvement of the given issue will moderate the effect of source cues on at titude toward a company. Based on previous persuasion research, bot h the elaboration likelihood model (ELM) proposed by Petty and Cacioppo (1986) and the heuristic-systematic mo del (HSM) developed by
83 Eagly and Chaiken (1993) suggested the source effects of attitude changes will depend on a persons motivation and ability to process th e message. The hypothesis suggested that the respondents level of involvem ent of the issues regarding the development of technology products and companies in this experiment will m oderate the effect of source cues on attitude toward a company. To test the hypothesis, a M ANOVA analysis was conducted. Th e level of involvement of participants was dichotomized into two groups low and high involvement. However, the results indicated no significant interaction effect of source cue and level of involvement on attitude toward the company. Thus, Hypothesis 1 was rej ected. Although the intera ction effect does not reach statistically significant, th e participants level of involveme nt still had direct effect of predicting attitude toward a company. One explanati on of the results is that the sample size in some of subgroups was too limited to generate si gnificant interaction eff ect. For example, there were only three participants who were exposed to the CEO source and they had low level of involvement. Only two participants who were e xposed to the editor sour ce had low level of involvement. Another explanation for the lack of interaction effect may be that more than half of participants answered four on the 5-piont liker scale and thus diminish the effect. Another disadvantage may argue that self-report of involveme nt level measured in this experiment did not reflect respondents motivation to process the messages presente d in the stimuli blogs. Given the fact that the level of involvem ent was not manipulated into anot her independent variable, it is appropriate to assume that the am ounts of effort participants inve sted in reading the blog entries and answering the questionnaire may differ from i ndividuals. Put it differe ntly, participants who were categorized as the high i nvolvement group were not necessarily more motivated to process
84 the stimuli even though their answers on the 5-poin t likert scale regarding level of involvement were higher than that of the medium and low involvement group. Consequently, future research should consider manipulating level of involvement as another indepe ndent variable to control the sample size or conceiving other approaches to enhance participants mo tivation in responding to experiment materials. Research Question 3: How is the perceived credibility of different source cues related to attitude toward a given company? The third research question examined the relati onship of two dependent variables. In order to determine any significant relationship between perceived source credib ility (trustworthiness and expertise) and attitude to ward the company, a Pearson corre lation analysis was performed. The Pearson correlation coefficient showed that there was a positive correlation among perceived trustworthiness, perceived expe rtise and attitude toward the company. This finding suggested that perceived trustworthiness and perceived ex pertise were highly associated with attitude toward the company. It could be interpreted th at when a person perceives the blog author as trustworthy and expert, it was likely that he or she would have more positive attitude toward the company which the blog author wr ites posts about or vice versa. Furthermore, a multiple regression was utilized to predict and explain scores on attitude toward the company based upon the informati on regarding perceived trustworthiness and expertise in an effort to gain more insight on relationship between variables. The results found that perceived trustworthiness, expertise and leve l of involvement could be effective predictors for achieving positive attitudes toward the company in four source groups, respectively. The four regression models all reached statistical signif icance. A positive relatio nship between perceived trustworthiness, expertise and level of involvement and atti tude toward the company was corroborated. Perceived expertise co uld be interpreted as a reliable and significant contributor to
85 positive attitude in the CEO source group ( =0.733, p < 0.001). For the PR manager source group, perceived trustworthiness was a significant c ontributor to achieve positive attitude toward the company ( =0.425, p < 0.05). As outlined earlier, previous research has claimed that source credibility influences persuasion and opinion change: High (vs. low) credibility source generated better evaluations of the presentation and opinion change in the dir ection advocated by the communicator (Hovland et al., 1951; McGinnies and Ward, 1974, 1980). Trustwor thiness and expertise are two major and widely-accepted components of perceived source credibility. The regression analysis in the present study confirmed the persua sion effects of perceived source credibility of attitude toward the company, which suggested the higher perceived source credibility can lead to more favorable attitude toward a company. Level of involvement could act as an effectiv e predictor for gaining positive attitude toward the company. Based on the summarized results, the origin al proposed model can be established and slightly modified. Figure 5-1 illustrated the rela tionships between variables found in this study. Conclusion The ways in which online communication contri butes to organizational goals have been a crucial topic in public relations research. Recently examinations of the relationship that exist between an organization and its key publics has emerged as a significant paradigm for public relations scholarship and pract ice (Bruning & Ledingham, 1999) Thank to the interactive capacity of Internet technol ogy, organizations are provided with a potential communication channel to realize dialog ic relationships targeting an increas ing numbers of online population. In the Internet era, the Internet has is gaining more popularity as an alternative mass media as well as a public relations tool.
86 Weblogs, emerging as a new communication cha nnel, allow corporate leaders, managers and industry watchers to speak di rectly to their desired publics in a human voice tone (Kelleher and Miller, 2006). Marken (2005) suggested blogs provide manageme nt what they want to and need in todays world a fast, effective and economic means of carrying out two-way communication with the firms many audience. With the popularity of the blogosphere, it underscores an important shift in the structure and practice of communications. Blogs enable audiences to move away from the traditional pyr amid of influence with its top-down, one-way information flow to a more fluid, horizontal peer-to-peer paradigm, in which brands and corporate reputations are built by engaging multip le stakeholders through continuous dialogue. Consequently, scholars and professionals began to define the nature of Weblogs and its applications in social inter action, political communication and public relations. While most study analyzed the blog posts using content analysis, us e and gratification theo ry of blog authors and readers through survey method, little research has em pirically addressed the key issues of source credibility and its effect on at titude toward a company. To br idge such a gap, this study integrated persuasion literature and public relations scholarship in an attempt to empirically examine the effects of different source cues on pe rceived source credibility and attitude toward a company. In addition, the present study tested wh ether level of involvement can moderate the effect of perceived source cred ibility on attitude toward the co mpany. Furthermore, the present study verified causal relations am ong attitudes, perceived source cr edibility (trustworthiness and expertise) and level of involvement. By conducting an experimental study (N =177), this contributes to the research of blog communication and persuasion. The results sugg ested the perceived source credibility (trustworthiness and expertise) and attitude to ward a company can be affected after people
87 exposed to four popular sources of corporate blogging (CEO, industry watcher, PR manager, college student). Among four popular authors, th e CEO of Google Company received the highest scores regarding perceived trus tworthy and expertise source. Th e finding was consistent with previous research, which suggest ed the CEOs are often viewed as experts in their industry because they hold extensive managerial experi ence and insightful thoughts. On the other hand, the senior student source, which was defined as the peer individual source in this experiment, was perceived as the least trustworthy and expertise compared to the other three authors. The results did not reflect a pivotal role that the peer individual so urce has gained influence on the blogosphere as previous research suggested. Intriguingly, participants who read the public relations manager source showed favorable ratings in perceived expertise but neutral opinions in perceived tr ustworthiness. In other words, participants believed the public relations manger was well informed and intelligent to reveal argument on the issue. However, people remained somewhat cynical about the blogging messages of the public relations manager source regarding trustworthy issue. As such, future research should investigate the key issue of trustworthiness of public relations professionals and make recommendations to improve the image in terms of blog communication. In conclusion, the present study confirmed the effects of source cues on perceived credibility and attitude toward a company on the blogosphere. Thus the higher perceived source credibility can result in more favorable attit udes toward a company. Moreover, through running a regression analysis, the results found that perceived trustworthin ess, expertise and level of involvement could be effective predictors for ac hieving positive attitudes toward the company in four source groups, respectively.
88 Implications Findings of this study could pr ovide several theoretical and practical implications for public relations researchers as well as professional practitioners. Theoretical Implications How Web-based communication offers a poten tial way to facilitat e a two-way dialogic communication has been a crucial topic in public relations res earch. Recently, an increasing emphasis has been placed on examining how We blogs could evolve as a new communication channel to maintain beneficial relationships with target publics though demonstrating a high level of conversation-style communication (Ke lleher & Miller, 2006). As Kelleher and Miller (2006) pointed out, the conversati onal speaking tone used in blogs plays an important role in building and maintaining computer-mediated rela tionships. However, the impact of blogging has not been empirically investigated. The present study integrated persuasion research and online public relations literature aiming to understand whether source cue of bloggers could affect peoples perceived source credibility and attitude toward the company. The results yielded inspiring conclusions that blogs could serv e as an effective communication resource to foster trust and bui ld positive attitudes between corporations and consumers depending on the perceived trustworth iness and expertise of a given source. For public relations scholarship, the re sults of this study may contribu te or be related to existing public relations theories such as excellence theory (Grunig & Hunt, 1984) and relationship management paradigm (Dozier, L. A. Grunig, & J. E. Grunig, 1995; Huang, 2001) in developing or maintaining more favorable rela tionships with target publics. Excellence theory provides an ethical framewor k for organizations to achieve effectiveness by emphasizing the two-way symmetrical mode l. The symmetrical model provides more equitable dialogues with publics, and the organi zations communication is likely to be more
89 effective when influence flows two ways. Th e advent of the Internet and Web-based communication has provided a pot ential way to facilitate the practice of effective two-way communication. This study would add a new pers pective on how employing different sources of corporate blogging would affect the credibility of a communica tion message, and how perceived source credibility works as a predictor of bu ilding positive attitudes. As such, a dialogic relationship could be enhanced by choosing the right person to share th eir insights through the interactive mechanisms on the blogosphere. According to Ehling (1992), the relationship management perspective shifts public relations practice away from the manipulation of public opinion toward building, nurturing, and maintaining organization-public rela tionships. The perspective of re lationship management is the conception of public relations as the management of relationships betw een an organization and its key publics (Ledingham & Bruning, 2000). H on and J. E. Grunig (1999) looked at the psychology literature in order to id entify characteristics of interpersonal rela tionships. Hon and J. E. Grunig (1999) concluded that control mutual ity, trust, satisfaction, commitment, exchange relationship, and communal relationship are good indicators of successful interpersonal relationships. In addition, Kim (2003) conducted an empirical study to vali date the relationship between OPRs and attitude toward the brand usin g relationship indicators established by J. E. Grunig and Huang (2000). She found four dimensi ons of OPRs, the perception of satisfaction with the company had a significant imp act on attitude toward the brand. The study would generate insights on this pa radigm by presenting how trustworthiness and expertise of a source can play a pivotal role in maintaining positive relationships on the blogoshpere. Furthermore, Weblogs offer the onlin e users an opportunity to establish a virtual community that shares specific interests, whic h makes the issue of ma naging relationship on the
90 blogosphere more complex. Future research should consider who starts th e dialogic loop, what relationship outcomes will be generated and how to incorporate other variables to optimize communication effects and achieve an organizations goal. Practical Implications For public relations practitioners, the study c onfirmed the belief that Weblogs could be a potential communication cha nnel to foster trust and gain favora ble attitudes in terms of corporate and marketing communication. As outlined earlier, blogs provide opportunities for greater interactivity, transparent di scussion, and participation in the communication process. The distinguishing features between co rporate blogs and other public relations tools such as news releases and newsletters lie in the naturalnes s, objectivity and persona l intimacy (Kim, 2006). In addition, Weblogs are suitable for setting corporate agendas in that they are not subject to the gate keeping process, which is constrained by limited time and space in the mass media. Thus, corporate blogs are gaining the popularity as an alternative channel to open a dialogue and achieve effective communication. From a strategic planning point of vi ew, a company should carefully take source credibility into account if they plan to venture in corporat e blogging. For example, the CEOs and senior industry watchers would be appropriate persons to start to blog for strong brand awareness corporations like Google Company since both of them obtained higher scores in perceived trustworthiness and perceived e xpertise as well as attitude toward the company. The messages from public relations mangers source were consid ered to be expert. Neve rtheless, the issue of trustworthiness remains questionable, which is consistent with the long standing negative image of public relations professionals as researcher s mentioned the reporting bias associated with public relations (Callison, 2001). In that regard, public relations practitioners should increase
91 trustworthiness to overc ome the negative image through talki ng in an honest, transparent and candid manner, as well as having a long-term commitment to interact ing with desired audiences. Last but no least, communication practitione rs should keep monito ring and reacting to online opinions. Even though the st udy did not find a significant e ffect of peer communication, the issue of the peer individual sources effect in affecting attitude toward a company needs to be further tested. It should be noted that the dissemination speed of negative news over the Internet is far faster than expected. Therefore, it become s more and more essential for organizations to keep an eye on what is happening on the blogoshp ere to conduct environm ent scanning and crisis audit. Limitations No research is without limitations. Unlike most experiments which were conducted in classroom settings, this study ove rcame the artificiality and generated external validity through presenting four authentic online blogs to respondents (Wimmer & Dominick, 2003). People browsed the online blogs and were in vited to take an Internet surv ey. This enabled them to have a more realistic experience (Kim, 2006). Although this experiment took place in a natural online browsing environment, it was subject to problems with control, such as the time difference and cognitive efforts that each participant spent on reading each blogs (Wimmer & Dominick, 2003). A second limitation was associated with sa mpling method used to collect data. All participants were university stude nt at a large southeastern pub lic school in the United States. Respondents were recruited using convenien ce sampling. The mean age of them was 24.58 years old. Although previous survey indicated co llege students are heavy users of Internet technology, they do not represent the whole population. Therefore, the results of present study cannot be generalized beyond this specific population.
92 In addition, the situation used in the experiment was Google Company. It was assumed that university students would be familiar with the companys services and development. The effect of familiarly with the company was not cont rolled and tested. It is reasonable to assume that perceived source credibility and attitude toward the company might be influenced by familiarity with the Google Company. Moreover, because of the strong brand awaren ess and equity of Google Company reported in major business media outlets, the attitude s of the control gro up which received no blog information were higher than that of the edito r source, public relations manager source and senior source. The higher mean score of attitude s in control group may reflect general favorable attitudes toward Google Company and thus skewed the results. As such, future research should employ different stimuli with other organizations to researchers to retest the effect of blog communication more sophisticatedly. A comp arison between companies based on brand awareness with different organizations w ould make the results much stronger. Finally, the study focused on cons umers perceived source cred ibility and attitude toward the company. As scholars point out, the goal of p ublic relations professi onals is to maintain mutually beneficial relationshi ps among various stakeholders. In that regard, other company public such as investors, employees, community members and so on could be analyzed to examine whether the model holds true in terms of blog communication. Suggestion for Future Research The study contributes to exam ining the impact of blog comm unication in the context of building corporate images. The expe riment should be replicated w ith other organizations in an attempt to generalize the above conclusions a nd garner different insights. To enhance the applicability of the proposed model, future re search could investigate diverse companies, industries and issues. Accordi ng to the Fortune 500 Blogging Wi ki, as of October 5, 2006, 40
93 (8%) of the Fortune 500 companies reported active public blogs written by their employees about the company or its products. As the diffusion of blog adoption continue s to grow, it appears essential to explore source credibility issues w ithin different types of organizations, including nonprofit, small etc. In addition, the study take s a quantitative appro ach to analyze the relationships between given variables. It shoul d be noted that future research could employ qualitative methods, such as focus group or in -depth interviews to better understand why consumers perceive different levels of trustworthiness and expertise among four popular blogging sources. The advantage of qualitative anal ysis is to gain res earch depth (Wimmer& Dominick, 2003). Investigate Different Stakeholders The present study examined the impact of bl og communication with regard to one of the external publics consumers. As discussed in previous literature, ma jor corporations also witness the rapidly increasing numbers of em ployees blogs (Keller & Miller, 2006). Within corporations, blogs are used for internal co mmunication and viewed as knowledge-sharing channels. Hewlett-Packard and International Bu siness Machines Corporation are examples of companies that utilize blogs for maintaining em ployee relations (Jones, 2005). In addition, Sun Microsystems and Microsoft encourage employ ees to become bloggers for debate, free association, and collecting input about projects (Roush, 2005). The content is virtual word-ofmouth communication that is typically opinioned and replete with links to ot her Internet sources. Therefore, future efforts could be ventured into examining whether corporate leaders and employees blogs can increase the intimacy and atti tude toward the company in the context of internal communication. Besides, the opinions of other stakeholders such as community members, investors and so on s hould be further investigated.
94 Incorporate Other Variables in Persuasion Research Apparently, there are other variables that ma y affect the results of blogging messages. The study proposed a basic model for an alyzing the impacts of corporate blogs. Future research could incorporate other variables in th eir experimental designs. For exam ple, other variables such as argument quality, message sidedness, prior knowledge and order effect of message sidedness that have been frequently tested in the persuasion li terature would allow fo r intriguing results for future research. In this experiment, argument quality was not taken into account. However, Priester and Petty (2003) argued that argument quality affects the effectivene ss of source trustworthiness. They suggested that a strong argument from a low trustworthiness source is more effective than a weak argument from a high trustw orthiness source, or almost or equally effective as strong argument from a high trustworthiness source, especially because an endorser of low trustworthiness can lead to grea ter scrutiny of product related information than an endorser of high trustworthiness (p. 417). Blogs, which ar e not constrained by limited time and space like other mass media, enable authors to make th e arguments in depth. Therefore, it would be essential to examine the argument qua lity presented on the blogoshpere. In terms of message sidedness, two approach es of a one-sided message and a two-sided message are often contrasted and measured by res earchers to examine various effects on attitudes toward an issue or a brand (OKeefe, 1999). Ot her researchers also looked at the different impact of refutational two-sided message a nd the nonrefutational two-sided message and suggested that the message sidedness effect ma y vary significantly depending upon whether the opposing arguments are refuted (Allen, 1991, 1994; Cr owley & Hoyer, 1994). Future research could consider adding messages sidedness as an other independent variable to explore the interaction effect of source cred ibility and message sidedness.
95 Previous studies have indicat ed that prior knowledge about the issue influences the effectiveness of each type of messages. More sp ecifically, one-sided messages are more effective when the audience is uninformed about the issue than when the audience is knowledgeable. On the contrary, two-sided messages tend to pers uade well-informed recipients more than individuals who are unfamiliar with the issue (C hu, 1967). This experiment asked participants how much they are concerned about technology issue and produc t to measure their level of involvement. It should also be noted that individuals motiva tion to engage in processing messages influences the effectiveness of the t ypes of sided messages. Future research should create a more sophisticated manipulation of cognitive involvement and incorporate questions regarding prior knowledge. Corporate Credibility a nd Endorser Credibility The reputation of a corporation is often cite d as an important ingredient in a firms success. Fombrun (1996) defines corporate reputa tion as a perceptual representation of a companys past actions and future prospects th at are an aggregate of many personal judgments about the company. Fombrun (1996, p. 72) explicitly incorporates corporate credibility as one important aspect of corporate re putation. In this context, cred ibility is the degree to which consumers, investors and other constituents believe in the company's trustworthiness and expertise. Keller (1998, p. 426) defines corporate cr edibility as the exte nt to which consumers believe that a firm can design and deliver products and services that satisfy customer needs and wants. In essence, corporate credibility is the perceived expertise and trustworthiness of a firm. In the present study, individual blogs and corpor ate leaders blogs are used as the stimulus materials. Many companies hold official blogs, wh ich are often associated with their companys culture, products and services. The study did not distinguish company blogs from individual
96 blogs. Thus, it would be interesting to examine the interaction effect of corporate credibility and endorser credibility on the attitude changes. Ethical Concern The case of Wal-Mart enlisting bloggers in a pu blic relations campaign stimulated a debate whether using a weblog as a communication tool is ethical or not (Barbaro, 2006). Wal-Mart, long criticized for low wages and its health benef its, began working with bloggers in late 2005 to promote a positive corporate image. In 2006, a usual couple who drove their RV around the United States, parking in Wal-Mart parking lots as they went la unched an interesting blogs called Wal-Mart Across America (http://walmartingacrossamerica.com/ ). In the blog, they kept defending the policies of Wal-Mart. A news story in the New York Times revealed the blog was actually a carefully scripted and funded campaign from the esteemed public relations firm, Edelman. Industry watchers criticized the entire Walmarting Across America campaign was built upon a lie, a duplicitous back-story th at while not unusual in marketi ng, was in direct violation of the Word of Mouth Marketing Associations C ode of Ethics, which Edelman helped craft. Obviously, the approach violated honesty and transparency in marketing and public relations practice. After the report, the CEO of Edelman, Ri chard Edelman, apologized for his firms error for failing to be transparent. The case aroused pr actitioners and researcher s to address the ethical concern on the blogosphere since Weblogs are c onsidered as a potential channel to build intimacy and foster trust with stakeholders. Futu re research should further address the issue of ethical practice in blogging disclosure. This study empirically contribu tes to persuasion research and public relations theories as well as practitioners. This study would add a new perspective on how employing different sources of corporate blogging would affect the credibility of a communication message, and how
97 perceived source credibility works as an antecedent for managing organization-public relationship. In that regard, a two-way dial ogic communication loop could be enhanced by choosing the right person to share their insights on the blogosphere. For public relations practitioners, the study conf irmed the belief that Weblogs could be an effective channel to cultivate trust and gain favorable attitudes in terms of corporate and marketing communication. A company should carefully take source credibili ty into account if they plan to venture in corpor ate blogging. Also, public relations practitioners should increase trustworthiness to overc ome the negative image through talki ng in an honest, transparent and candid manner, as well as having a long-term commitment to intera cting with desired audiences.
98 Independent Variable Dependent Variables Figure 5-1. Modified theoretica l model in the present study. Source Cues CEO as a blogger Journalist as a blogger Peer individual as a blogger Public Relations Manager as a blogger Perceived Source Credibility Trustworthiness Expertise Attitude toward a Company Level of Involvement
99 APPENDIX A BLOGGING GLOSSARY Glossary term Explanation Archive Archives are links in the sidebar to older entries. Blogs Weblogs (Blogs) A regularly updated online publication Blogger People who run blogs are bloggers Blogging The act of adding an entry is blogging. Flog Fake blogs created by co rporations to promote themselves and their products. K-log Knowledge-log is a blog that is designed for knowledge sharing. Moblog A word derived from mobile and Weblog. Category Category refers to subjec ts that the entry discusses. Feedback Mechanism Comment The comments section a llows readers to post their thoughts on original authors blogs. Trackback Trackback is a list of other blogs to refer to which usually generate more extensive discussions. Links Blogroll A list of links to other Weblogs. Hyperlink A reference in a hypertex t document to another document or other resource. Permanlink Permanent uniform reso urce locators (URLs) or links to Weblogs archives. Syndication Atom An XML-based language designed for blogs and syndicated news delivery. RSS Really Simple Syndication (R SS) or a RSS feed allows people to display content for Web di stribution or syndication. XML Extensible Markup Langua ge (XML) is a programming language that allows application developers to facilitate common data sharing. Tage A service for bloggers to associ ate their posts with topics, and to make it easy for people to find interesting posts on a given subject. Timestamp Date and time the post was published. (Gill, 2004; Public Relations Tactics 2005; Trammell, 2004; Trammell 2005)
100 APPENDIX B WEBLOG LAYOUT Figure B-1. Weblog layout for the CEO source. http://ceoofgoogle.blogspot.com/index.html
101 Figure B-2. Weblog layout for the PR manager source. http://publicrelaions2.blogspot.com/
102 Figure B-3. Weblog layout for the Busi nessWeek magazine editor source. http://bweditor.blogspot.com/
103 Figure B-4. Weblog layout for th e senior student source. http://cbradshow.blogspot.com
104 APPENDIX C INTRODUCTORY PARAGRAPH OF WEBLOG AUTHORS About Me: CEO sources Dr. Eric Schmidt Mountain View, CA, US I am the chairman and CEO of Google. I led that companys strategic planning, management and technology development Since coming to Google, I have focused on building the corporate infrastructure needed to maintain Google's rapid growth as a company and on ensuring that quality rema ins high while product development cycle times are kept to a minimum. I have a B. S. degree in electrical engineering from Princeton University, and a mast er's and Ph.D. in computer science from the University of California-Berkeley. About Me: Public Relations Manager sources Elliot Schrage Mountain View, CA, US At Google, I am the public relations mana ger and serves as a spokesperson. I am responsible for corporate communications a nd public affairs, which encompass media relations, stakeholder outreach and policy stra tegy. I am a lawyer and business advisor with 20 years of experience at the intersec tion of global business strategy and public policy. About Me: BusinessWeek Editor sources Tim Weber New York City, NY, US I am a senior editor of BusinessWeek Mag azine. I specialized in technology and the Internet issue for 15 years. I l ove reading, movie and traveling. About Me: College Senior Student sources Chris Bradshow Gainesville, FL, US I am a senior student majoring in computer engineering at UF. I am a huge technology fan. Hope I can get a job in the major bi g companies after graduating from college.
105 APPENDIX D BLOG POST ONE: NOW ON YOUTUBE Sunday, March 25, 2007 Now on Youtube: Good move or Copyright nightmare? When talking about Google s takeover of Youtube (website news ), some people may think it is a nifty business move. For the company's legal t eam, however, it may soon turn into a long and nasty nightmare. I found an interesting vote on MSNBC website. Is acquiring YouTube a good move for Googl e? (there are total 3023 responses.) Yes, this makes them even more dominant : 55 % No, YouTube is this year's version of Napster : 22 % No, YouTube is just th e flavor of the month : 18 % What is this YouTube thing you speak of? 5.5 % Showing videos on the internet is nothing new. Their clever idea was to create a model that makes it easy not just to watch the films, but also to share them. Want to show a film on YouTube? You don't have to mess about with video standards. Just upload your film and the website does all the hea vy lifting. Just make sure you have labelled the clip correctly, so that the re st of the world can find it. Watching is just as easy. No worries about having the ri ght video player. Plus you can ra te films, recommend them to friends, comment on them and even integrat e them into your own website, without any technical knowledge. Little wonder that YouTube has been a huge success. In August 2005 the site had a measly 2.8 millio n users a month. One year later YouTubes audience had grown to 72 million people. This has created its own dynamic. People will po st their films on YouTube because that's where the audience is, and the audien ce will grow ever larger because of the extra content. It's social networking in overdrive.
106 Until now most copyright holders had little incentive to sue YouTube. The company was young and rapidly burning through its ve nture capital. However, now th at YouTube is part of the Google, with a market capitalization of $129 billion, there is a serious incen tive to let the lawyers off the leash. We obviously don't see ourselves as content pirates. I argue that we act fully within the law, ba sed on general "fair use" standards, and more importantly the safe harbor provisions in sec tion 512 of the US Digita l Millennium Copyright Act (DMCA) of 1998. The act was designed to ensure copyright protection works in the digital age although its authors clearly did not anticipate toda y's dynamic and on-demand digital world. Section 512 helps "service providers to avoid liability for acts of copyright infringement committed by third parties; it gives them a safe harbor. It's a complicated piece of legislation, but here is one example of how it is supposed to work: A service provider (YouTube) stores material (a pirated movie clip ) on its system at the direction of a user (YouTube member). Meeting these cond itions may help it qualify for the safe harbour provision at least as long as YouTube makes it easy for copyright holders (a film studio) to complain about the infringement, and quickly rem oves pirated material that has been brought to its attention. So far the US courts have failed to rule what most of the DMCA's statutes actually mean. Already, though, a couple of "512" defences have gone disastrously wrong for file-sharing services Napster and Grokster. Some copyright experts wonder how it will play in the courts if Google has made advertising dollars on the back of pirated material on YouT ube. They also predicted YouTube will continue to get sued. No doubt we will have to work hard to steer YouTube into safe waters. Solid content identification, video watermarking, royalty reporti ng and clearer upload guidelines for YouTube members are a must.
107 APPENDIX E BLOG POST TWO: NOT TRAP PING USERS DATA=GOOD Wednesday, March 7, 2007 Not Trapping users data=GOOD When users get what they want from you quickly and easily, theyre more likely to come back next time. (Shh. Dont tell anyone else this vital secr et.) Part of that is fe eling that they arent trappedthat they can leav e you behind if they want. We build a very good targeting engine and a lot of business success has come from that. We run the company around the usersso as long as we ar e respecting the rights of end users and make sure we dont do anything agains t their interest, we are fine. We would never trap user data I was asked if users could get al l of their search history and e xport it to Yahoo. We would like to do that, as long as it is authenti cated.If users can switch it keeps us honest. It echoes the send your users away happy and theyll come back ph ilosophy. It also gives guidance to teams at Google. So I started making a list of the ways that Google lets you access your data: Gmail This ones easy. Google provides free POP access so that anyone can fetch their email out of Gmail. Search. If you sign in with your Google account to search, Google can offer not only personalized search but also let you retrieve your search history Mihai Parparita did some digging a while ago for example. The ability to securely access your se arch history as an RSS feed is documented in our help pages now. For example, the url https://www.google.com/sear chhistory/?output=rss works very well if youre logged into your Google account. I believe you can add things like &num=250 so that you dont have to access 10 items at a time either. This feature is secu red by password-protection (you ha ve to be logged in), but it provides a nice way to access your own search es. Oh, and dont forget to try out your personal search trends If youre logged in, the url is http://www.google.com/psearch/trends and youll get all sorts of neat data like your most frequent searches, clicks, and when you tend to search. Okay, enough about search. Lets look at some othe r products that let you ge t to your data easily. Google Docs and Spreadsheets let you export your stuff in more formats than I know: Word, Rich Text Format, CSV, HTML, XLS (Excel), and PDF. Even one I did not know: .ods? Ah, OpenOffice. Nice. Google Calendar. Also easy. From its launch, Google calend ar has allowed iCal (.ics) and RSS export of calendar data.
108 Google Talk uses the open XMPP protocol. The VoIP pa rt of Google Talk is done with Jingle, another open protocol that Google helped with. I like that our IM chat is open to other clients so you can talk from iChat and GAIM to Trillian Pro and Blackberries. Google Reader easily exports your list of feeds in OP ML format, and can import OPML files as well. Blogger. Blogger can export data via FTP or SFTP and backup your blog Google AdWords. I dont use AdWords myself, but Googl e provides a free application called the AdWords Editor and its features include a snapshot export feature: Save a delimited file with your AdWords account informa tion and show it to a colleague or keep it for reference. So Im assuming its not too hard to suck down your AdWords info. Yup, a couple minutes of searching found references to impor ting your Google ad campaigns into Microsoft and Yahoo Google Groups. I was dreading checking on this one. B ack in August, someone wrote to me and said I run a Google Group with 7,500+ subscrib ers and I need to download the subscriber list, but I dont see an op tion for that. It turns out that we di d not offer that as a feature back in August. We were able to help the fellow, but it did not sound like an often -requested feature, so I did not think the Groups team had gotten a chance to do this. But I check ed and it looks like the Groups team got a chance to add this. Yay! Fo r a group I owned, I clicked Manage and then Browse membership list. At the bottom right will be a button Export member list and clicking that will download a comma-separated value (CSV) file. Lets see, where else can you store data at Google? Ah, a Custom Search Engine Theres even a bookmarklet to let you add sites to your custom search engine as you surf the web. Can you get your entire list of sites expor ted from your Custom Search Engine? Yup. Go to your search engines control panel and c lick on the Advanced tab. Youll get options to download your sites in XML or tab-separate d value (TSV) file format. Lots of products like Google Analytics and th e Google Webmaster Console also give options to export data in various formats. Okay, so looking down this list, it looks like Goog le does pretty well in offering open access to your data, at least for all the im portant services that I checked If you know of some way that Google doesnt let you download your data please feel free to mention it.
109 APPENDIX F BLOG POST THREE: SOME THOUGHTS Sunday, February 25, 2007 Some thoughts Recently, I saw Elinor Mills writing about an interesting allegation Ill include the whole content of the allegation: [Just to clarify, this is an allegation th at Elinor is passing on from a newsletter, not a claim that Elinor is making herself directly.] In the past, when you launched a website, or Googl e wasnt picking up your stuff, you could call the friendly people over there and theyd look at your website to see if you were legit, look at their search results, and adjust th eir code appropriately. It used to be this all occurred in the same day. Then it was 24 hours. So, imagine our dismay when www.wesrch.com wasnt even being picked up two weeks after we launched. We had cal led Google two days into the launch and they apologized, saying their search e ngines were backlogged with so many sites to monitor. We called after a week and then called again and ag ain, with no better answer. We even tried posting ads with Google and they couldnt find us. Clearl y, we had tried their patience, as in the end they threatened to BLACKLIST our Web sites so no one would ever find us again. Now is that power or what? Funny thing is, Yahoo found us fa ster and more reliably. So, Google is no longer my home page. More importantly, they are s howing all the signs of a monopolist trying to forcibly extract revenues for nothing. Whenever th is happens, its a sign th at revenue growth has peaked and they are trying to force it in order to maintain high stock valu ations. So watch out if you are an investor. When Elinor asked for a comment about this, several of our employees read the original complaint, and I have to admit that we we re perplexed. Google doesnt provide phone support for webmasters; as Vanessa Fox recently noted, over 1 million webmasters have signed up for our webmaster console alone, so offering phone support for every site owner in the world wouldnt really scale that well. They talk about buying ads later in the paragraph; we wondered maybe they were talking to phone support fo r AdWords? But I cant imagine anyone at Google on the ads side or anywhere else saying our search engines were backlogged with too many sites to monitor. The Google index is designe d to scale to billions of webpages, and it does that job pretty well. Its even harder for me to imagine anyone at Google saying on the phone that they would BLACKLIST our Web sites so no one would ever find us again, because again, we dont provide webmaster support over th e phone, and I believe AdWords phone support would know better than to claim our index was b acklogged or to threaten to remove anyones site from our index. Maybe a call to AdWord s support reached such a fever pitch that a representative declined to run an ad? At any rate, Im sorry for any negative interactions that wesrch.com had with Google. The current description of the issue doesnt give en ough concrete details to check out, but if anyone from that domain wanted to clarify or to provide emails or dates/times/names of phone calls (did they call AdWords? Randomly try to hop into the Google phone tree? Talk to a receptionist?), our staff would be happy to try to look into it more.
110 In the absence of more details about their interaction, I tried to dig more into the crawling of wesrch.com. I did not see any negative issues (n o spam penalties or anything like that) for the domain. I saw attempts to crawl the site as far back as October 2006, but th at earliest attempt got an authentication crawl error (tha t would have been a 401 or a 407 HTTP status code). I believe that this allegation went out Feb. 2nd, and I believe we had at least one page from that site at that point. I did notice that visiting th e root page of the domain give s a 302 (temporary) redirect to the HTTPS version of the domain. Thats kinda unusua l, but we should still be able to crawl that. The other thing to look at is curre nt coverage. Heres what I saw: Search Engine Number of pages Google over 450+ pages Yahoo 1 page Live about 176 pages Ask 0 pages (Note that if you just do [site:wesrch.com] on MS N/Live, you might get results estimates as high as 500+ results, but the way to veri fy results estimates is to go to the final page of results, and MSN/Live stops after 176 results.) It looks like Google crawls wesrch.com at least as deeply as any other major search engine. Im still confounded who the folks at wesrch.com coul d have talked to at Google, but Ill leave open the offer to dig into it more if they want to provi de more details. And Ill wish them well for their new domain in the future.
111 APPENDIX G QUESTIONNAIRE FOR EXPERIMENT GROUP INTRODUCTION Thank you for taking the time to answer the que stions in this survey. I am studying the credibility of corporate blogs. Please circle th e number that best de scribes your thoughts or feelings. Your answers will be used only for st atistical purposes and will remain strictly confidential to the extent provided by law. Please read the in structions and questions carefully. SURVEY QUESTIONS 1. Please identify the author of the blog you read. CEO of Google [ ] Editor of BusinessWeek [ ] College Senor majoring in Computer Engineering [ ] Public Relations Manager of Google Company [ ] Section 1. Evaluation of Perceived Credibility Please evaluate the blog pos ts you just read about the following statements. (Please circle one number between 1 and 7 for EACH of the ten answers. If you feel that you have no reaction, please circ le the number 4 to indi cate your neutrality.) Trustworthiness 1) undependable 1 2 3 4 5 6 7 dependable 2) dishonest 1 2 3 4 5 6 7 honest 3) unreliable 1 2 3 4 5 67 reliable 4) insincere 1 2 3 4 5 67 sincere 5) untrustworthy 1 2 3 4 5 67 trustworthy Expertise 1) not an expert 1 2 3 4 5 67 expert 2) inexperienced 1 2 3 4 5 67 experienced/ 3) unknowledgeable 1 2 3 4 5 67 knowledgeable 4) unqualified 1 2 3 4 5 67 qualified 5) unskilled 1 2 3 4 5 67 skilled Section 2. Attitude toward the company After reading the blog post, please evaluate how you feel about Google Company by circling a number on each of the scales below. If you feel that you have no reaction, please circle the number 4 to indicate your neutrality. 1) Unfavorable 1 2 3 4 5 6 7 Favorable 2) Bad 1 2 3 4 5 6 7 Good 3) Dislike 1 2 3 4 5 6 7 Like 4) Negative 1 2 3 4 5 6 7 Positive
112 Section 3. Level of involvement Section 4. General Internet and Blog Use 1. Do you use a computer at your workplace, at sc hool, at home, or anywhere else on at least an occasional basis? Yes [ ] No [ ] 2. Do you use the internet or World Wide Web, at least occasionally? Yes [ ] No [ ] 3. About how many years have you been an internet user? Six month or less [ ] A year ago [ ] Two or three years ago [ ] Four years ago [ ] Five years ago [ ] Six years or more [ ] 4. About how often do you go online for getting information? Several times a day [ ] about once a day [ ] 3-5 days a week [ ] 1-2 days a week [ ] every few weeks [ ] every few months [ ] less often [ ] Never [ ] 5. What is the average number of hours you spend on getting news or information from the Internet everyday? 0 [ ] 1 [ ] 2-5 [ ] 6-10 [ ] more than 10 [ ] 6. Do you have a personal blog? Yes [ ] No [ ] 7. How many blog posts did you read in the past week? 0 [ ] 1-5 [ ] 6-10 [ ] 11-15 [ ] 8. How many online blogs did you read in the past seven days? 0 [ ] 1 [ ] 2-5 [ ] 6-10 [ ] more than 10 [ ] 9. What is the average number of hours you spend reading and/ or posting to blogs per week? 0 [ ] 1 [ ] 2-5 [ ] 6-10 [ ] more than 10 [ ] 10. How often do you write posts about companies and their products? Never [ ] Less than once a week [ ] About once a week [ ] More than once a week [ ] Daily or almost daily [ ] Blog devoted to product or company [ ] Please evaluate your interests in the following issues. Not important -----------Very Important At all 1) How important are the issues about the development of technology companies to you personally? 1 2 3 4 5 2) How important are the issues about the development of technology products and services to you personally? 1 2 3 4 5 3) How much are you concerned about the issues on development of technology companies? 1 2 3 4 5 4) How much are you concerned about the issues on development of technology products and services ? 1 2 3 4 5
113 11. Please evaluate the use of following s ources when you seek corporate and product information. Section 5. Demographics 1. Gender Male [ ] Female [ ] 2. Age _______ 3. What is your current level of education? [ ] Freshman [ ] Sophomore [ ] Junior [ ] Senior or post-baccalaureate [ ] Graduate Student THANK YOU VERY MUCH FO R YOUR PARTICIPATION! Never Most Often CEO blog 1 2 3 4 5 Industry watcher blog 1 2 3 4 5 Individual peer blog 1 2 3 4 5 Public Relations Specialist blog 1 2 3 4 5
114 APPENDIX H QUESTIONNAIRE FOR CONTROL GROUP INTRODUCTION Thank you for taking the time to answer the questi ons in this survey. I am studying the general usage of corporate blogs. Please circle the numbe r that best describes your thoughts or feelings. Your answers will be used only for statistical pur poses and will remain strictly confidential to the extent provided by law. Please read the instructions and questions carefully. SURVEY QUESTIONS Section 1. Attitude toward the company Please evaluate how you feel about Google Comp any by circling a number on each of the scales below. If you feel that you have no reaction, pl ease circle the number 4 to indicate your neutrality. 1)Unfavorable 1 2 3 4 5 6 7 Favorable 2) Bad 1 2 3 4 5 6 7 Good 3) Dislike 1 2 3 4 5 6 7 Like 4) Negative 1 2 3 4 5 6 7 Positive Section 2. Level of involvement Section 3. General Internet and Blog Use 1. Do you use a computer at your workplace, at sc hool, at home, or anywhere else on at least an occasional basis? Yes [ ] No [ ] 2. Do you use the internet or World Wide Web, at least occasionally? Yes [ ] No [ ] 3. About how many years have you been an internet user? Six month or less [ ] A year ago [ ] Two or three years ago [ ] Please evaluate your interests in the following issues. Not important -----------Very Important At all 1) How important are the issues about the development of technology companies to you personally? 1 2 3 4 5 2) How important are the issues about the development of technology products and services to you personally? 1 2 3 4 5 3) How much are you concerned about the issues on development of technology companies? 1 2 3 4 5 4) How much are you concerned about the issues on development of technology products and services ? 1 2 3 4 5
115 Four years ago [ ] Five years ago [ ] Six years or more [ ] 4. About how often do you go online for getting information? Several times a day [ ] about once a day [ ] 3-5 days a week [ ] 1-2 days a week [ ] every few weeks [ ] every few months [ ] less often [ ] Never [ ] 5. What is the average number of hours you spend on getting news or information from the Internet everyday? 0 [ ] 1 [ ] 2-5 [ ] 6-10 [ ] more than 10 [ ] 6. Do you have a personal blog? Yes [ ] No [ ] 7. How many blog posts did you read in the past week? 0 [ ] 1-5 [ ] 6-10 [ ] 11-15 [ ] 8. How many online blogs did you read in the past seven days? 0 [ ] 1 [ ] 2-5 [ ] 6-10 [ ] more than 10 [ ] 9. What is the average number of hours you spend reading and/ or posting to blogs per week? 0 [ ] 1 [ ] 2-5 [ ] 6-10 [ ] more than 10 [ ] 10. How often do you write posts about companies and their products? Never [ ] Less than once a week [ ] About once a week [ ] More than once a week [ ] Daily or almost daily [ ] Blog devoted to product or company [ ] 11. Please evaluate the use of following s ources when you seek corporate and product information. Section 4. Demographics 1. Gender Male [ ] Female [ ] 2. Age _______ 3. What is your current level of education? [ ] Freshman [ ] Sophomore [ ] Junior [ ] Senior or post-baccalaureate [ ] Graduate Student THANK YOU VERY MUCH FOR YOUR PARTICIPATION! Never Most Often CEO blog 1 2 3 4 5 Industry watcher blog 1 2 3 4 5 Individual peer blog 1 2 3 4 5 Public Relations Specialist blog 1 2 3 4 5
116 LIST OF REFERENCES Ajzen, I., & Fishbein, M. (1980). Understanding Attitudes and predicting social behavio r. Englewood Cliffs, NJ: Prentice Hall. Albrycht, E. (2004). Turning blogs into useful communication tools. Public Relations Tactics, 11 (3), 14-15. Allen, M. (1991). Meta-analysis comparing the persuasiveness of one-sided and two-sided messages. Western Journal of Speech Communication, 55 (4), 390-404. Allport, G. W. (1935). Attitudes, in C.A. Murchison (E d.) A Handbook of social psychology (Vol. 2, pp. 810-815). New York: Russell & Ressell. Apsler, R., & Sears, D. O. (1968). Warning, personal involvement, a nd attitude change. Journal of Personality and Social Psychology, 9 162-166. Armstrong, C., & McAdams, M. (2006). Believing blogs? Examini ng the influence of gender cues on credibility. Paper presented at the Associati on for Education in Journalism and Mass Communication annual conference, August 4, San Francisco, CA, 1-36. Backbone Media Inc. (2005) Blogging success st udy. Boston, MA, 1-40. Retrieved October 10, 2006 from: http://www.scoutblogging.c om/success_study/index.html Baker, S., & Green, H. (2005). Blogs will change your business. Business Week, 3931 56-57. Balnaves, M., Mayrhofer, D., & Shoesmith, B. (2004). Media Prof essions and the New Humanism. Journal of Media & Cultural Studies, 18 (2), 191-203. Banning, S., & Trammell, K. D. (2006). Revisiting the issue of blog credibility: A national survey. Paper presented at the Association for Education in Journalism and Mass Communication annual conference, A ugust 4, San Francisco, CA, 1-30. Barbaro, M. (2006). Wal-Mart en lists bloggers in P.R. camp aign. New York City, NY, 1-3 Retrieved March 13, 2007, from: http://www.nytimes.com/2006/03/07/technology/07blog.html?ex=z12993876 00zzzzz&en=ae7585374bf280b9&ei=5088 Batra, R., Myers, J. G., & Aaker, D. A. (1996). Advertising management (pp. 316-328). Upper Saddle River, NJ: Prentice Hall. Bauer, H. H., Grether, M., & Leach, M. (2002). Building customer relations over the Internet. Industrial Marketing Management, 31 (2), 155-163.
117 Berger, I. E., & Mtichell, A. A. (1989). The effect of advertising on attit ude accessibility, attitude confidence, and the attit ude-behavior relationship. Journal of Consumer Research, 16 269-279. Bichard, S. L. (2006). Building blogs: A multi-dimens ional analysis of the distribution of frames on 2004 presidential candidate web sites. Journalism & Mass Communication Quarterly, 83 (2), 329-345. Blogumentary.org (2006). Blogumentary: Open s ource me. Twin Cities, MN, 1-3. Retrieved September 12, 2006, from http://blogumentary.org/video/blogumentary.mov Blood, R. (2002). The Weblog handbook: Practical advice on creating and maintaining your blog Cambridge, MA: Perseus Publishing. Bruner, G. C., & Hensel, P. J. (1996). Marketing Scales Handbook American Marketing Association, Chicago, IL. Bruning, S. D., & Ledingham, J. A. (1989). Perc eptions of relationshi p and evaluations of satisfaction: An exploration of interaction. Public Relations Review, 26 (1), 85-95. Burley, D. P. (2005). Get the conversation started with blogs. Public Relations Tactics, 12 (8), 10. Calder, B. J., Phillips, L. W., & Tybout, A. M. (1981). Design research for application. Journal of Consumer Research, 8 (2), 197-201. Callison, C. (2001). Do PR practitioners have a PR problem? The effect of associating a source with public relations and client-negative news on audience perception of credibility. Journal of Public Relations Research, 13 (3), 219-234. Cook, T. D., & Campbell, D. D. (1979). Quasi-experimentation: De sign & analysis issues for field settings Boston, MA: Houghton Mifflin Co. Cooley, W. T. (1999). Inte ractive communicatio n: Public relations on the Web. Public Relations Quarterly, 44 41-42. Cooper, R. G. (1994). New products: The factors that drive success. International Marketing Review, 11 (1), 60-76. Crowley, A. E., & Hoyer, W. D. (1994). An in tegrative framework for understanding two-sided persuasion. Journal of Consumer Research, 20 (4), 561-574. Eagly, A., & Chaiken, S. (1993). Psychology of Attitudes New York: Harcout, Brace Jovanovich. Edelman. (2006). 2006 Edelman trust barometer: Summary of findings New York: Edelman, 132. Retrieved September 10, 2006 from: http://www.edelman.co.uk/insights/trus t/Edelman%20Trust%20Barometer%202006.pdf
118 Ehling, W. P. (1992). Estimating the value of public relations and communication to an organization. In J. E. Grunig, D. M. Dozier W. P. Ehling, L. A. Grunig, F. C. Repper, and J. White (Eds.), Excellence in Public Relatio ns and Communication Management Hillsdale, NJ: Lawrence Erlbaum Associates, 617-638. Fishbein, M., & Ajzen, I. (1975). Belief, attitude intention and behavior: An introduction to theory and research. Readi ng, Mass.: Addison-Wesley. Gill, E. K. (2004). How can we measure the in fluence of the blogosphere? Seattle, WA, 1-10. Retrieved September 10, 2006 from: http://faculty.washington.edu/kegi ll/pub/www2004_blogosphere_gill.pdf Goldberg, M. E., & Hartwick, J. (1990). The effects of advertis er reputation and extremity of advertising claim on adve rtising effectiveness. Journal of Consumer Research, 17 172179. Goodman, R. (2006). Corporate America behind the curve: Makovsky 2006 state of corporate blogging survey Gainesville, FL: The Institute of Public Relations. Gotlieb, J. B., & Swan, J. E. (1990). An app lication of the Elaboration Likelihood Model. Journal of Academy of Marketing Science, 18 (3), 221-228. Grunig, J. E. (2000). Collectivism, collaboration, a nd societal corporation as core professional values in public relations. Journal of Public Relations Research, 12 (1), 23-48. Grunig, J. E., & Hunt, T. (1984). Managing public relations New York: Holt Rineheart and Winston. Heath, R. L. (1998). New communi cation technologies: An issue management point of view. Public Relations Review, 24 (3), 273-288. Heires, K. (2005). The blogosphere b eckons: Should your company jump in? Harvard Management Communication Letter, 2 (4), 3-5. Hirshberg, P., & Edelman, R. (2006). Public Relationship: Communica tion in the age the age of the personal media New York: Edelman, 1-24. Retrieved September 10, 2006 from: http://www.edelman.com/image/insights/c ontent/BloggerSurvey_FINAL_fromprinterRV SD.pdf Holbrook, M. B., & Batra, R. (1987). Assessing the role of emotions as mediators of consumer responses to advertising. Journal of Consumer Research, 14 404-420. Hon, L. C., & Grunig, J. E. (1999). Guideline for measuring rela tionships in public relations Gainesville, FL: The Institute of Public Relations.
119 Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and Persuasion New Haven, CT: Yale University Press. Huang, Y. (2001). Values of pub lic relations: Effects on organi zation-public relationships mediating conflict resolution. Journal of Public Relations Research, 12 (4), 265-301. Huck, S. W. (2000). Reading Statistics and Research New York: Longman. Jo, S., & Kim, Y. (2003). The effect of web characteristics on relationship building. Journal of Public Relations Research, 15 (3), 199-223. Johnson, J. T., & Kaye, K. B. ( 1998). Cruising is believing? Compar ing Internet and traditional sources on media credibility measures. Journalism and Mass Communication Quarterly, 75 (2), 325-340. Johnson, J. T., & Kaye, K. B. (2004). Wag the blog: How reliance on traditional media and the internet influence credibility per ceptions of Weblogs among blog users. Journalism and Mass Communication Quarterly, 81 (3), 622-642. Johnson, M. A. (1997). Public relations a nd technology: Practiti oner perspectives. Journal of Public Relations Research, 9 213-236. Jones, D. (2005 May 10). CEOs refuse to get tangled up in messy blogs. USA Today, p. 1B. Kaid, L. L., & Postelnicu, M. (2006-in press). Cr edibility of political me ssages on the Internet: A comparison of blog sources. In Blogging, citizenship and the future of media New York: Routledge. Kelleher, T. M., & Miller, B. (2006). Organi zational blogs and the human voice: Relational strategies and relational outcomes. Journal of Computer-Mediated Communication, 11 (2), 1-30. Kent, M. L., & Taylor, M. ( 1998). Building dialogic relationshi p through the World Wide Web. Public Relations Review, 24 (3), 321-334. Kerbel, M. R., & Bloom, D. J. (2005). Bl og for America and civic involvement. Harvard International Journal of Press/Politics, 10 (4), 3-27. Key, R. J. (2005). How the profession can flour ish in this new dig ital age: Why you must challenge old PR model. Public Relations Tactics, 12 (11), 18-19. Ki, E. (2003). Relationship mainte nance strategies on web site. Ma sters thesis, University of Florida, Gainesville, FL. Kim, J. (2006). An experimental test of public relations messages: Si dedness, and corporate goodwill and trustworthiness. Ph.D. Dissertation, University of Florida, Gainesville, FL.
120 Kim, J., & Chan-Olmsted, S. M. (2005). Co mparative effects of Organization-Public Relationships and product-related attributes on brand Attitude. Journal of Marketing Communication, 11 (3), 145-170. Kiousis, S. (2001). Public trust or mistrust? Per ception of media credibility in the information age. Mass Communication and Society, 4 (4), 381-403. Kline, R. B. (1998). Principles and practices of structural equation modeling NY: The Guilford Press. LaBarbera, P. A. (1982). Overcoming a no-re putation liability through documentation and advertising regulation. Journal of Marketing Research, 19 (2), 223-228. Lawson-Borders, G., & Kirk, R. (2005) Blogging in campaign communication. American Behavioral Scientist, 49 (4), 548-559. Lin, S. (2005). A public relations campaign of cor porate social responsibili ty: A test of cognitive processing model of a CSR message Masters thesis, University of Florida, Gainesville, FL. Lindenmann, W. K. (2002). Guidelines for measuring the effectiveness of PR programs and activities Gainesville, FL: The Instit ute for Public Relations. Lutz, R. J., MacKinzie S. B., & Belch, G. E. ( 1983). Attitude toward the ad as a mediator of advertising effectiveness: Determinants and consequences. In R. P. Bagozzi & A. M. Tybout (Eds.), Advances in consumer research, Vol. 10 (pp. 532). Ann Arbor, MI: Association for Consumer Research. Lyons, D. (2005, November 14). Attack of the blogs. Forbes, 176, 128-138. Madden, M. (2006). Internet Penetration and Impact Washington, DC: Pew Internet & American Life Project, 1-5. Retrieved September 2, 2006, from: http://www.pewinternet.org/ pdfs/PIP_Internet_Impact.pdf MacKenzie, S. B., & Lutz, R. J. (1989). An empi rical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. Journal of Marketing, 53 48-65. Maduxx, J. E., & Rodgers, R. W. (1980). Effects of source expertise, physical attractiveness and supporting arguments on persuasion: A case of brains over beauty. Journal of Personality and Social Psychology, 39 (2), 235-244. Marken, G. A. (2005). To blog or no t to blog, That is the Question. Public Relations Quarterly, 50 (3), 31-33.
121 Marlow, C. (2003). Classifying emergent communities through diffusion Association of Internet Researchers annual conferen ce, Toronto, Ontario. McGinnies, E., & Ward, C. D. (1974) Persuasibility as a function of source credibility and locus of control: Five cross-cultural experiments. Journal of Personality, 42 360-371. McGinnies, E., & Ward, C. D. (1980). Better liked than right: Trustworthiness and expertise as factors in credibility. Personality and Social Psychology Bulletin 6 (3), 467-472. McGuire, W. J. (1969). Handbook of social psychology New York: Random House. McGuire, W. J. (1969). The nature of attitude s and attitude change. In G. Lindzey, & E. Aronson (Ed.), The handbook of social psychology (2nd ed., Vol. 3, pp. 136-314). Reading, MA: Addison-Wesley. McNeil, B. W., & Stoltenberg, C. D. (1988). A test of the Elaboration Likelihood Model for therapy. Cognitive Therapy and Research, 12 69-80. Mitchell, A. A. & Olson, J. C. (1981). Are pr oduct attribute beliefs th e only mediator of advertising effects on brand attitude? Journal of Marketing Research, 18 318-332. Moore, D. L., Hausknecht, D., & Thamodaran, K. (1986). Time compression, response opportunity, and persuasion. Journal of Consumer Research, 13 85-99. Newell, S. J., & Goldsmith, R. E. (2001). The development of a scale to measure perceived corporate credibility. Journal of Business Research, 52 (3), 235-247. Newsom, D. A., Ramsey, S. A., & Carrell, B. J. (1993). Chameleon chasing II: Areplication. Public Relations Review, 19 33-47. Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers perceived expertise, trustwor thiness, and attractiveness. Journal of Advertising, 19 (3), 3952. Ohanian, R. (1991). The impact of celebrity spokespersons perceived image on consumers intention to purchase. Journal of Advertising Research 46-54. O Keefe, D. J. (1999). How to handle opposing arguments in persuasive messages: A metaanalytic review of the effect s of one-sided and two-sided messages. In M. E. Roloff (Ed.), Communication Yearbook 22 (pp. 209-249): Sage Publication Inc. Olins, W. (2000). How brands are taking over the corporation. In M. Schultz, M. J. Hatch, & M. H. Larsen (Eds.), The expressive organization: Liki ng identity, reputation, and the corporate brand (pp. 51-65). Oxford, England: Oxford University Press.
122 O Malley, M., & Irani, T. (1998). Public relations and the web: Measuring the effect of interactivity, information, and the access to information in Web site. Paper presented at the Association of Education in Journalis m and Mass Communication annual conference, August 5, Baltimore, MD, 1-23. Parasuraman, A., Zeithaml, V., & Berry, L. (1988). SERVQUAL: A multiple-item scale for measuring consumer perceptions of service quality. Journal of Retailing, 64 12-40. Perloff, R. M. (2003). The dynamics of persuasion. Communi cation and attitudes in the 21st century (2nd ed.). Mahwah, NJ: Lawrence Erlbaum. Petty, R. E., & Cacioppo, J. T. (1979). Issue-invol vement can increase or decrease persuasion by enhancing message-relevant cognitive responses. Journal of Personality and Social Psychology, 37 1915-1926. Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Mode of persuasion New York: Academic Press. Pew Internet and American Life Project. (2006). Buzz, blogs, and beyond: The Internet and the national discourse in the fall of 2004 Washington, DC: Pew Internet & American Life Project, 1-15. Retrieve d on October 10, 2006 from http://www.pewinternet.org/ Pew Internet and American Life Project. (2006). Bloggers Washington, DC: Pew Internet & American Life Project, 1-23. Retrieved September 13, 2006 from: http://www.pewinternet.org/ PPF/r/186/report_display.asp Poindexter, J. (1983). Voices of authority. Psychology Today, 26 (3), 28-31. Pornpitakpan, C. (2003). The effect of celebri ty endorsers perceive d credibility on product purchase intention: The case of Singaporeans. Journal of International Consumer Marketing, 16 (2), 55-74. Porter, L., Trammell, K. D., & Chung, D. (2006). Roles and blogs in public relations. Paper presented at the Association for Educa tion in Journalism and Mass Communication annual conference, August 4, San Francisco, CA, 1-29. Priester, J., & Petty, R. E. ( 2003). The influence of spokesper son trustworthiness on message elaboration, attitude strength, a nd advertising effectiveness. Journal of Consumer Psychology, 13 408-421. Public Relations Society of America. (1999). Na tional credibility index. New York City: NY, 12. Retrieved September 20, 2006 from: http://www.stonypointpr.com/credibility_index.htm
123 Rainie, L. (2005). The state of blogging Washington, DC: Pew Internet & American Life Project, 1-4. Retrieved September 3, 2006 from: http://www.pewinternet.or g/pdfs/PIP_blogging_data.pdf Ratneshwar, S., & Chaiken, S. (1991). Comprehens ions role in persuasion: The case of its moderating effect on the persua sive impact of source cues. Journal of Consumer Research, 18 52-62. Reber, B. H., & Kim, J. (2006). How activist grou ps use Web sites in media relations: Evaluating online press rooms. Journal of Public Re lations Research, 18 (4), 313-333. Seltzer, T. (2005). The dialogic potential Weblog s in relationship building. Paper presented at the Association for Education in Journalis m and Mass Communication annual conference, August 3, San Antonio, Texas, 1-26. Sirfy, D. (2006). State of the blogoshere. New York City, NY, 1-36. Retrieved September 3, 2006 from: http://technorati.com/weblog/2006/02/81.html Smudde, P. M. (2005). Blogging, ethics and public re lations: A proactive and dialogic approach. Public Relations Quarterly, 50 (3), 34-38. Socialtext Company. (2005). Fortune 500 bloggi ng wiki. Palo Alt o, CA, 1-4. Retrieved September 10, 2006 from http://www.so cialtext.net/bizblogs/index.cgi Stevenson, R. (2004). Microsofts ga tes touts blogging as business tool. Routers News Trammell, K. D. (2004). Celebrity blog: Investig ation in the persuasive nature of two-way communication. Ph.D. Dissertation, Universi ty of Florida, Gainesville, FL. Trammell, K. D. (2005). Looking at the pieces to understand the whole: An analysis of blog posts, comments, and trackbacks. Paper presented at the International Communication Association annual meeting, Ma y 26, New York, NY, 1-25. Trammell, K. D., & Gasser, U. (2004). Deconstructing Weblogs: An analytical framework for analyzing online journals. Paper presented at the International Communication Association, May 27, New Orleans, LA, 1-27. Trammell, K. D., & Kiousis, S. (2005). Agenda-setting and the blogs : Issue and attributes salience influence on celebrity web site. Paper presented at the Association for Education in Journalism and Mass Communicati on, August 5, San Antonio, TX, 1-26. Trammell, K. D., & Metzgar, E. (2006). Impact of blogs on relati onship management during a crisis. Paper presented at the Association for Education in Journalism and Mass Communication annual conference, A ugust 4, San Francisco, CA, 1-30.
124 Trammell, K. D., Porter, L ., Chung, D., & Kim, E. (2006). Credibility and the uses of blogs among professionals in the communication Industry Paper presented at the Association for Education in Journalism and Mass Co mmunication annual conference, August 4, San Francisco, CA, 1-26. Trufelman, L. P. (2005). Consumer-generated media: Challenges and opportunities for public relations. Public Relations Tactics, 12 (5), 17. Tsai, Y. (2006). Exploratory study on corporate blogs: Potential of corporate blogging to be a new public relations tools. Masters thesis, University of Florida, Gainesville, FL. Viegas, F. B. (2005). Bloggers expectation of privacy and accountability: An initial survey. Journal of Computer-Mediated Communication, 10 (3), 1-10. Wack, F. (2005). Whats a blog? Malmo, Sw eden, 1-2. Retrieved September 10, 2006 from: http://www.corporateblogging.i nfo/basics/what/default.asp#2 Wack, F. (2006a). Business blogging basics Malmo, Sweden, 1-2. Retrieved September 10, 2006 from: http://www.corporateblogging.i nfo/basics/what/default.asp Walker, J. (2005). Weblog (definition). In D. Herman, M. Jahn, & M.-L. Ryan (Eds.). In Routledge Enyclopedia of Narrative Theory London: Routledge. Weil, D. (2005). The corporate blogging book: Absolutely e verything you need to know to get it right NY: Portfolio Hardcover. Wijinia, E. (2004). Understanding Weblogs: A co mmunicative perspective. Masters Thesis, Twente University, The Netherlands. Wilkie, W. L. (1990). Consumer behavior, 2nd ed. New York: Johnson Wiley & Sons. Wimmer, R. D., & Dominick, J. R. (2003). Mass media research: An Introduction Belmont, CA: Wadsworth. Woo, C. (2001). Optimal Aad-Ab-PI measures in the journal of advertising: 1991 through 2000. Masters thesis, University of Florida, Gainesville, FL. Wu, C., & Shaffer, D. (1987). Susceptibility to persuasive appeals as a function of source credibility and prior experi ence with attitude object. Journal of Personality and Social Psychology, 52 677-688.
125 BIOGRAPHICAL SKETCH Jiun-Yi Tsai was born in Taichung, Taiwan. In 2004, she obtained her B.A. degree in radio and television from National Chengchi Universit y, one of the most prestigious colleges of mass communication in Taiwan. With extensive interest in mark eting communication, she also fulfilled the requirement of their advertising planning program. She realized the dream of pursing advance study in the United States in 2005 and completed her Master of Arts in Mass Communi cation specializing in publ ic relations from the University of Florida in 2007. During her graduate studies, her research interests involved two different areas. First, she focused on how exce llent public relations can be enhanced by online communication. Second, she examined crisis /risk communication and the media framing process. As a masters student, she presented a re search paper at the Association for Education in Journalism and Mass Communication annual conf erence and won the topsecond student paper in the International Communication Division. After graduation, she hopes to co ntinue her public relations career in academic field in the United States.