<%BANNER%>

Effects of Interactive Online Media Type and Crisis Type on Public Trust during Organizational Crisis

Permanent Link: http://ufdc.ufl.edu/UFE0022582/00001

Material Information

Title: Effects of Interactive Online Media Type and Crisis Type on Public Trust during Organizational Crisis
Physical Description: 1 online resource (157 p.)
Language: english
Creator: Oyer, Seth
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: communication, corporate, crisis, effects, experiment, interactive, media, new, online, organization, organizational, oyer, public, relations, trust
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This 2 x 3 experimental design tests the effects of interactive online media type (specifically blog and streaming video) and crisis type (accident, intentional or victim) on public trust during organizational crisis. Coombs' (1997) Situational Crisis Communication Theory formed the foundation for the crisis type independent variable. Hon and Grunig's (1999) trust dimension measures from Relationship Theory were used as the basis for the dependent variables. Vonnetek Automotive, a fictional organization, was designed to test the variables. This is the first study of this kind to bridge the logical gaps between 'new media' effects, crisis communication and trust. Furthermore, the experimental methodology offers predictive qualities not offered in other existing research on the individual variables. In this study, the public trust of fictional organization Vonnetek Automotive (designed expressly for this study) was measured using reports of the faculty and staff at a major southern university and business members of a chamber of commerce of a southeastern city. The results indicated that the crisis type variable strongly affected perceived trust reports while interactive online media type did not. Interaction effects were found in only 1 of 11 dependent measures, indicating that in some cases there may be a more complex relationship between media type and trust, something that must be studied further.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Seth Oyer.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Mitrook, Michael A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022582:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022582/00001

Material Information

Title: Effects of Interactive Online Media Type and Crisis Type on Public Trust during Organizational Crisis
Physical Description: 1 online resource (157 p.)
Language: english
Creator: Oyer, Seth
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: communication, corporate, crisis, effects, experiment, interactive, media, new, online, organization, organizational, oyer, public, relations, trust
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This 2 x 3 experimental design tests the effects of interactive online media type (specifically blog and streaming video) and crisis type (accident, intentional or victim) on public trust during organizational crisis. Coombs' (1997) Situational Crisis Communication Theory formed the foundation for the crisis type independent variable. Hon and Grunig's (1999) trust dimension measures from Relationship Theory were used as the basis for the dependent variables. Vonnetek Automotive, a fictional organization, was designed to test the variables. This is the first study of this kind to bridge the logical gaps between 'new media' effects, crisis communication and trust. Furthermore, the experimental methodology offers predictive qualities not offered in other existing research on the individual variables. In this study, the public trust of fictional organization Vonnetek Automotive (designed expressly for this study) was measured using reports of the faculty and staff at a major southern university and business members of a chamber of commerce of a southeastern city. The results indicated that the crisis type variable strongly affected perceived trust reports while interactive online media type did not. Interaction effects were found in only 1 of 11 dependent measures, indicating that in some cases there may be a more complex relationship between media type and trust, something that must be studied further.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Seth Oyer.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Mitrook, Michael A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022582:00001


This item has the following downloads:


Full Text

PAGE 1

1 EFFECTS OF INTERACTIVE ONLINE MEDIA TYPE AND CRISIS TYPE ON PUBLIC TRUST DURI NG ORGANIZATIONAL CRISIS By SETH OYER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Seth Oyer

PAGE 3

3 To Breanne: without your love and suppor t this would have never been possible. To Emma: my sweetcakes, the day I met you wa s the best one I had ever known and each one since has been even better.

PAGE 4

4 ACKNOWLEDGMENTS I thank Richard Edelm an at Edelman Public Rela tions, Bill Properzio at the University of Florida and Travis Proctor at Artemis IT for s eeing and supporting the valu e of this research. Thanks to Michael Mitrook for his guidance, support and motivation.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .........................................................................................................................9ABSTRACT ...................................................................................................................... .............10 CHAP TER 1 INTRODUCTION .................................................................................................................. 112 LITERATURE REVIEW .......................................................................................................14General Public Relations Theory ............................................................................................14Measuring Trust in Public Relations ......................................................................................24Organizational Crisis Public Relations ...................................................................................28Public Relations in the New (Media) World .......................................................................... 35Hypotheses .................................................................................................................... ..........38Research Questions ............................................................................................................ .....383 METHODOLOGY ................................................................................................................. 40Pilot Study ..............................................................................................................................40Main Study ..............................................................................................................................41Participants .................................................................................................................. ....41Design ........................................................................................................................ ......41Operationalization of Independent Variables ..................................................................42Operationalization of Dependent Variable ......................................................................43Procedure ..................................................................................................................... ....43Data Analysis ...................................................................................................................444 RESULTS ....................................................................................................................... ........55Preparing the Data for Analysis .............................................................................................. 55Sample Demographics ............................................................................................................55Results of the Statistical Analysis ........................................................................................... 56Results Related to the Hypotheses .................................................................................. 92Results Related to the Research Questions .....................................................................94

PAGE 6

6 5 DISCUSSION .................................................................................................................... .....96Trust Dimensions ....................................................................................................................96Hypotheses .................................................................................................................... ..........99Research Questions ............................................................................................................ ...102Implications .................................................................................................................. ........105Implications For Interactiv e Online Media Research .................................................... 105Implications For Situational Crisis Communication Theory .........................................105Implications For Relationship Theory ........................................................................... 108Implications For Public Relations Practice ................................................................... 110Implications for Using Experimental Me thodology and Interactive Online Media ...... 112Implications Summary ................................................................................................... 112Limitations ................................................................................................................... ..115 APPENDIX A SURVEY MEASURES ........................................................................................................ 119B PILOT STUDY .....................................................................................................................120C SURVEY INSTRUMENT 1.................................................................................................127D SURVEY INSTRUMENT 2.................................................................................................131E INFORMED CONSENT ......................................................................................................137F MAIN EFFECTS ONE-WAY ANOVAS ......................................................................... 139LIST OF REFERENCES .............................................................................................................150BIOGRAPHICAL SKETCH .......................................................................................................157

PAGE 7

7 LIST OF TABLES Table page 3-1 Pilot test cr osstabs .............................................................................................................453-2 Pilot test ANOVA (DV: A TTRIBUTION, IV: CRISISTYPE) .............................................463-3 Pilot test post hoc Bonferroni (DV: ATTRIBUTION, IV: CRISISTYPE) ........................474-1 Two-way ANOVA descriptive st atistics (DV: TREATSFAIRLY) .......................................574-2 Two-way ANOVA descriptive sta tistics (DV: CONCERNEDABOUT) .............................584-3 Two-way ANOVA descriptive st atistics (DV: KEEPPROMISES) .....................................594-4 Two-way ANOVA descriptive sta tistics (DV: TAKESOPINIONSOF) ...............................604-5 Two-way ANOVA descriptive sta tistics (DV: CONFIDENTSKILLS) ...............................614-6 Two-way ANOVA descriptive sta tistics (DV: ABILITYACCOMPLISH) ...........................624-7 Two-way ANOVA descriptive sta tistics (DV: SOUNDPRINCIPLES) ...............................634-8 Two-way ANOVA descriptive sta tistics (DV: DOESNOTMISLEAD) ...............................644-9 Two-way ANOVA descriptive statis tics (DV: MAKEDECISIONSFOR) ...........................654-10 Two-way ANOVA descriptive sta tistics (DV: WATCHCLOSELY) ....................................664-11 Two-way ANOVA descriptive statis tics (DV: SUCCESSFULATTHINGS) .......................674-12 Tests of between-subjects effects (DV: TREATSFAIRLY) .................................................694-13 Tests of between-subjects effe cts (DV: CONCERNEDABOUT) ........................................704-14 Tests of between-subjects ef fects (DV: KEEPPROMISES) ...............................................714-15 Tests of between-subjects effe cts (DV: TAKESOPINIONSOF) .........................................724-16 Tests of between-subjects effe cts (DV: CONFIDENTSKILLS) ..........................................734-17 Tests of between-subjects effe cts (DV: ABILITYACCOMPLISH) .....................................744-18 Tests of between-subjects effe cts (DV: SOUNDPRINCIPLES) .........................................754-19 Tests of between-subjects ef fects (DV: DOESNOTMISLEAD) ..........................................764-20 Tests of between-subjects effe cts (DV: MAKEDECISIONSFOR) .....................................77

PAGE 8

8 4-21 Tests of between-subjects effects (DV: WATCHCLOSELY) ..............................................784-22 Tests of between-subjects effe cts (DV: SUCCESSFULATTHINGS) .................................794-23 Multiple comparisons post hoc Bonferroni (DV: TREATSFAIRLY) ..............................814-24 Multiple comparisons post hoc Bo nferroni (DV: CONCERNEDABOUT) .....................824-25 Multiple comparisons post hoc Bonferroni (DV: KEEPPROMISES) ............................834-26 Multiple comparisons post hoc Bo nferroni (DV: TAKESOPINIONSOF) ......................844-27 Multiple comparisons post hoc Bo nferroni (DV: CONFIDENTSKILLS) .......................854-28 Multiple comparisons post hoc Bo nferroni (DV: ABILITYACCOMPLISH) ..................864-29 Multiple comparisons post hoc Bo nferroni (DV: SOUNDPRINCIPLES) ......................874-30 Multiple comparisons post hoc Bonferroni (DV: DOESNOTMISLEAD) .......................884-31 Multiple comparisons post hoc Bo nferroni (DV: MAKEDECISIONSFOR) ..................894-32 Multiple comparisons post hoc Bonferroni (DV: WATCHCLOSELY) ...........................904-33 Multiple comparisons post hoc Bo nferroni (DV: SUCCESSFULATTHINGS) ..............91F-1 ANOVA (DV: TREATSFAIRLY, IV: CRISISTYPE) .........................................................139F-2 ANOVA (DV: CONCERNEDABOUT, IV: CRISISTYPE) ...............................................140F-3 ANOVA (DV: KEEPPROMISES, IV: CRISISTYPE) .......................................................141F-4 ANOVA (DV: TAKESOPINIONSOF, IV: CRISISTYPE) .................................................142F-5 ANOVA (DV: CONFIDENTSKILLS, IV: CRISISTYPE) .................................................143F-6 ANOVA (DV: ABILITYACCOMPLISH, IV: CRISISTYPE) .............................................144F-7 ANOVA (DV: SOUNDPRINCI PLES, IV: CRISISTYPE) .................................................145F-8 ANOVA (DV: DOESNOTMISLEAD, IV: CRISISTYPE) .................................................146F9 ANOVA (DV: MAKEDECISIONSFOR, IV: CRISISTYPE) .............................................147F-10 ANOVA (DV: WATCHCLO SELY, IV: CRISISTYPE) ......................................................148F-11 ANOVA (DV: SUCCESSFULATTHINGS, IV: CRISISTYPE) .........................................149

PAGE 9

9 LIST OF FIGURES Figure page 3-1 Vonnetek Automotive Home page both Pretest and Posttest ......................................... 483-2 Vonnetek Automotive About page both Pretest and Posttest ......................................... 493-3 Vonnetek Automotive Gallery page both Pretest and Posttest ....................................... 503-4 Vonnetek Automotive Events page both Pretest and Posttest ........................................513-5 Vonnetek Automotive News page Pretest treatment ...................................................... 523-6 Vonnetek Automotive News page Posttest Blog treatment ............................................ 533-7 Vonnetek Automotive News page Posttest Streaming Video treatment ........................ 544-1 Graph of interaction effects TAKESOPINIONSOF ....................................................... 95

PAGE 10

10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF INTERACTIVE ONLINE MEDIA TYPE AND CRISIS TYPE ON PUBLIC TRUST DURI NG ORGANIZATIONAL CRISIS By Seth Oyer December 2008 Chair: Michael Mitrook Major: Mass Communication This 2 x 3 experimental design tests the effects of interactive online media type (specifically blog and streaming vi deo) and crisis type (accident, intentional or victim) on public trust during organizational crisis. Coombs ( 1997) Situational Crisis Communication Theory formed the foundation for the crisis type independent variable. Hon and Grunigs (1999) trust dimension measures from Relationship Theory were used as the basis for the dependent variables. Vonnetek Automotive, a fictional orga nization, was designed to test the variables. This is the first study of this kind to bridge the logical gaps between new media effects, crisis communication and trust. Furthermore, the expe rimental methodology offers predictive qualities not offered in other existing rese arch on the individual variables. In this study, the public trus t of fictional organization Vonnetek Automotive (designed expressly for this study) was meas ured using reports of the faculty and staff at a major southern university and business members of a chamber of commerce of a southeastern city. The results indicated that the crisis type va riable strongly affected perceived trust reports while interactive online media type did not. Interaction effect s were found in only 1 of 11 dependent measures, indicating that in some cases there may be a mo re complex relationship between media type and trust, something that mu st be studied further.

PAGE 11

11 CHAPTER 1 INTRODUCTION Researchers across m ultiple disciplines have attempted to define and understand the concept of trust, often disagreeing on its definition and theoretical foundation (Watson, 2005). While trust research has been conducted in fi elds such as psychology, management, marketing, and organizational behavior, public relations scholars hold the con cept of trust in especially high regard because of its ro le within relationships a foundation of the public relations discipline (Broom, Casey, & Ritchey, 1997; Grunig, 1992; Grunig & Hon, 1999; Grunig & Huang, 1999; Ledingham & Bruning, 1998). But why is trust so im portant to these relationships? Kent and Taylor (2002) accurately summed it up when they stated, as is well known in public relations, once public trust has been lost it is difficult, sometimes impossible, to regain it. Understanding how trust is built and maintained is critical in organizational crisis public relations. One of the main goals of most organizations is to provide a re turn, or profit, for its shareholders. Trust is crucial to this goal as sh areholders must trust an organization to become a shareholder in the first place, and ongoing trust is needed to retain shareholders. Crises occur for a myriad of reasons but all of th em potentially threaten the trus t shared between an organization and its publics. Even when an organization has pr operly planned for an unexpected crisis, effects on public trust are still vague at best. Progre ss has been made, however, in distinguishing different types of crises. Coombs & Holladay (2002) identified three crisis attribution clusters helpful in explicating crisis types: accident clusters, intentional clusters, and victim clusters. Regardless of the attribution of the crisis, however, an organizations publics must be communicated with. In doing so, many organizations have integrated use of interactive online media tactics into its organiza tional communication strategies.

PAGE 12

12 New media has been a buzzword in communica tion since McLuhan (1964) first told us that the medium is the message. While the definition of new has changed with time and technological advances, it currently refers to the interactive use of the In ternet as the medium. Organizations, in general, and cris is public relations pract itioners, in particul ar, regularly use the Internet as a key communication medium henc e the use of interactive online media as an independent variable in this res earch. However, there is no resear ch to establish how interactive online media should be used during crisis and to what effect. Two type s of interactive online media increasingly used by organizations are blogs and streaming vi deo, but again there is little experimental research to test their effectiveness. This proposal details an experiment desi gned to better understand how public trust is affected by organizational use of interactive onli ne media during various t ypes of crises. Crisis public relations research has ofte n neglected experimental methods in favor of case studies for analysis. While such case studies are valuable, th ey lack the predictive quality of experimental research critical to further our understanding of crisis public relations. The focus of this experiment is to examine how organizational us e of blogs and streami ng video during different types of crises affect public trus t of the organization. This study a llows analysis of potential main effects of both crisis types a nd interactive online media types on trust, as well as possible interaction effects among vari ables. Furthermore, this study begins to fill the void of experimental research in the fi eld of crisis public relations. This research also furthers Situational Crisis Communication Theory (Coombs, 2007). While Situational Crisis Communication Theo ry has already proven valuable by adapting concepts from Attribution Theory, Situational Cr isis Communication Theory has not been tested experimentally. Furthering Situational Cr isis Communication Th eory through use of

PAGE 13

13 experimental method allows Situational Crisis Communication Theory to expand its theoretical application to the predictive a goal that both benef its scholars and prac titioners. Situational Crisis Communication theory also currently overl ooks media effects. This research begins to identify such media effects within Situationa l Crisis Communication Theory by examining blogs and streaming video.

PAGE 14

14 CHAPTER 2 LITERATURE REVIEW General Public Relations Theory In order to properly follow the developm ent of crisis and relationship theories, it is important to first examine the f oundations of public relations theory. Public relations as a form of communication has existed for centuries. The introductory public relations textbook Public Relations Strategies and Tactics (Wilcox and Cameron, 2007) give s the examples of Caesars Commentaries the Rosetta Stone, and even the earliest Olympic Games as examples of early public relations. And, like all sciences, public relations has st rived to identify a research paradigm within its disciple (Kuhn, 1996). One of the first, and certainly the largest, studies aimed at finding such a paradigm began in 1985 when the International Association of Business Communicators awarded a group of researchers $400,000 to fund a study focused on answ ering the following research questions: How, why, and to what extent does communicati on affect the achievement of organizational objectives? How does public relations make an organizatio n more effective, and how much is that contribution worth economically? What are the characteristics of a public relations function that are most likely to make an organization effective? The answers to these questions would form the basis for Excellence Theory (Grunig, Dozier, Ehling, Grunig, Repper, White, 1992). In 2006, Botan and Hazleton argued that there was not yet a paradigmatic theory in public relations, though Excellence Theory most closely approaches it. Grunig et al. id entified four models of public relations: press agentry, public information, two-way asymmetrical, and two-wa y symmetrical. The first, press agentry, involves one-way publicity for the organization at any means necessary. A classic example of this is P.T. Barnum and his publicity campaigns for his circus. While considered the lowest

PAGE 15

15 order of the four models, it is still used by some practitioners partic ularly with celebrity publicists. The next model, public information, is also a one-way comm unication model but the main difference is a focus on truthful information that is of importance to the public. This is a method often employed by government public affair s specialists. The third model is two-way asymmetrical communication. This mode l does acknowledge and include two-way communication but it is really more like two separate one-way models: organization to the public, and public to the organization. Feedback is used but usually for the benefit of the organization (for example, an organization might relay a message and see what type of reaction it gets, then revise the message and repeat in orde r to increase persuasion) Finally the highest order model is two-way symmet rical communication. In this model, both organization and public simultaneously act as sende r and receiver with constant feedback allowing for dynamic mutually beneficial communication. Through th is two-way communication, or mixed motive model, organization and public are constantly in states of negotia tion in order to create win-win scenarios. The Excellence study also identified four leve ls of public relations within organizations: Program level: Individual communication programs, often operating independently of organizational mission/goal. Functional level: Theoretical/benchmarking level Organizational level: Public relations must show contributi on to organizational effectiveness. (This was the original fo cus of the IABC proposal.) Societal level: Effective organizations must be socially responsible Grunigs team identified that, for excellence to be achieved, public rela tions must (a) be a part of the dominant coalition, (b) practice two-way sy mmetrical communication, (c) be educated in public relations, and (d) be public relations independently and not integrated with something else. In this way, public relati ons could be knowledgeab le and remain socially responsible while

PAGE 16

16 still have enough power to effectively work as a positive change agent wi thin the organization. The following section reviews the Excellence Study in more detail. The Excellence Study sampled four types of or ganizations in the United States, Canada, and the United Kingdom: corporations, not-for-profits, government agencies, and associations. The study was carried out in two phases. The first phase was a survey composed of mailed questionnaires to CEOs and senior managers of organizations, from which the authors built an Excellence Index to test the excellence of organizational public relations. Three years later, the second phase of the study involved c onducting case studies using long interviews CEOs and other dominant coalitio n members to test measures of organizations that were rated high in excellence as well as some that were measured low in excellence, seeking to measure economic value of public relations to its respective organizations. They then applied the Excellence Index to their qualitative second-phase findings to try and identify excellent public relations contribution to the organizational bottom line. This analysis provided the team with a number of variables to help answer their research questions. These included the value of commun ication, contribu tion to strategi c organizational functions, public relatio ns roles, models of public relatio ns, potential of the communications audit, activist pressure on the organization, employee variables, and the status of women. The value of communication. The team attempted to identify the value of communication within the organiza tions. This was achieved by CE O reports of perceived value of public relations within the organization. Contribution to strategic organizational functions. The team looked for contribution to four specific organizational functions: strategic planning, response to major social issues, major initiatives, and routine operations.

PAGE 17

17 Public relations roles. The team asked what role the t op communicator in an organization holds manager, senior advisor, technician, or media relations to identify public relations roles within excellent organizations. Models of public relations. The team analyzed which public relations models the organizations used: press agentry, public in formation, two-way asymmetrical, and two-way symmetrical. This information was used to iden tify organizational worl dview of in-house public relations. Potential of the communication unit. This variable measured if the communication unit in the organization actually had the ability to prac tice both management and technician functions. Even in organizations that strive for it, excel lence cannot be achieved if the communication unit does not have the potential ability to perform such public relations. Activist pressure on the organization. The team measured the extent to which an organization is exposed to activist pressure. The team contended that activist groups are, by definition, active and ready to be communicated with. Employee variables. Examining employees-as-unit-of-ana lysis to identify organizational culture, the team found tw o dominant factors: participative and authoritarian. Participative organizational cultures were much more like ly to value and utilize two-way symmetrical communication, a tenet of excellence. The status of women. The team posited that in order to effectively communicate with diverse publics, it must also value diversity within its organizational culture. Thus, the status of women was analyzed in order to help measure this The team argued that organizations that treat women well exhibit characteris tics critical to excellence.

PAGE 18

18 Based on this data, the Excellence Study iden tified seven characteristics of excellent communication departments, grouped into four categories: Empowerment of the pub lic relations function o The senior public relations executive is involved with the strategic management process of the organization, and comm unication programs are developed for strategic public identified as a part of this strategic management process. o The senior public relations executive is a member of th e dominant coalition of the organization or o The senior public relations executive has a direct repor ting relationship to senior managers who are part of the dominant coalition. o Diversity is embodied in a ll public relati ons roles. Communicator roles o The public relations unit is headed by a manager rather than a technician. o The senior public relations executive or others in the public relations unit must have the knowledge needed for the mana ger role, or the communication function will not have the potential to become a managerial function. o Both men and women must have equal opport unity to occupy the managerial role in an excellent department. Organization of the communicati on function, relationship to ot her functions, and use of consulting firms o Organizations must have an in tegrated communication function o Public relations should be a management function separate from other functions. Models of public relations o The public relations department and the dominant coalition share the worldview that the communication department shoul d reflect the two-way symmetrical, or mixed-motive, model of public relations. o Communication programs developed for specific publics are based on the twoway symmetrical, mixed-motive model. o The senior public relations executive or others in the public relations unit must have the knowledge needed for the two-way symmetrical model, or the

PAGE 19

19 communication function will not have the potential to practice that excellent model. In summary, the Excellence team identified fi ve keywords to describe excellent public relations: managerial, strategic, symmetrical, diverse, and ethical (Grunig, et al., 2002, p.306). Excellence Theory, however, made some assump tions in its findings. For instances, one assumption is that the idea of excellence is universal. This study, while large in population, was only studied in western cu ltures. Grunig, Grunig, Srirames h, Huang, and Lyra (1995) would later acknowledge this, identifying two additional models personal influence and cultural translator by modifying models of two-way communication. Berger (2005) suggested that Excellence Theory glosses over the power relationships regarding the dominant coalition, warning that two-way symmetrical communication may not be as eas y as it sounds in practice. Spicer (2007) suggested that a more critical defi nition of public relations practitioner may help alleviate some of Bergers concerns though research has not yet been conducted to test it. Stacks and Watson (2007) also proposed that to adequately measure and test symmetrical communication, a move away from regressi on-based, variable-ori ented methodology to a relational-based methodology from both metric and nonmetric data approaches to measurement and research is needed (p.67). Additionally, Molleda (2001) found that public rela tions practitioners act as change agents or as organizational conscience, focused on integr ating the organizations with the society in a communal relationship. This relationship is similar to Grunigs two-way symmetrical communication model in its regard for both organi zation and public, as well as inherent social responsibility. Not all scholars were satisfied with the Excellence Theory, however. One major alternative perspective is Conti ngency Theory. While not typically considered a critical theory,

PAGE 20

20 Cancel, Cameron, Sallot and Mitr ook introduced Contingency Theo ry in 1997. This theory is based on a continuum from total accommodation to total advocacy. Finding that many practitioners answered it depe nds when asked about best pract ices, the authors identified 86 variables to evaluate co mmunication responses based specific to publics. The authors originally contended that analysis of these variables for each key public could identify the communication strategy best used with each public (on the continuum). The variables introduced by the authors (Cancel et al., 1997 ) were distilled into external variables and internal variables, each with sec tions and subsections. External variables included threats, industry environment, general political/s ocial environment/external culture, the external public, and issue under question. Internal vari ables included organizational characteristics, public relations department character istics, characteristics of domina nt coalition, inte rnal threats, individual characteristics and relationship characteristics. A major difference between Contingency Theory and Excellence Theory is that, while Excellence Theory attempts to explain more generalizable or ganization-public relationships, Contingency Theory isolates the interactions of one organization and one public at a time (Cancel, Mitrook, & Cameron, 1999). Such interact ions, and not the results of them, are the focus of Contingency Theory. Contingency Theory is can be used to identify the many variables that can affect public relationships, though it may be overly complex. One assumption of this theory is that practitioners would be willing to use it (in order for it to be practically appli cable). In order to make the theory more parsimonious and prac titioner-friendly, Cancel, Mitrook, and Cameron (1999) interviewed organizational p ublic relations managers in orde r to refine the theory. This study reinforced the value of th e contingency continuum to pract itioners. However, the study

PAGE 21

21 also identified that the variab les most relevant to practit ioners could be grouped into predisposing and situational variables: predisposing variables refer to those variables which that their greatest influence on an organization by helping to shar e the organizations predisposi tion towards relations with external public situational va riables are the specific and often changing dynamics at work during particular situations involving an organization and an ex ternal public (Cancel et al., 1999). Based on their findings, the authors identified a multitude of variable groupings. The following quote is a synopsis of the authors findings regarding highl y supported variables: Of the predisposing variables cited by interviewees, the fo llowing variables received the most support in the overall data analysis: co rporation business exposure; public relations access to dominant coalition; dominant coal itions decision power and enlightenment; corporations size; and individual characteristics of involved persons. Of those situational variables cited by interviewees, those th at received the most support were the following: urgency of situation; ch aracteristics of extern al publics claims or requests; characteristics of external public; potential or obvious threats; and potential cost or benefit for a corporation from choosing va rious stances. Of thos e variables cited as being particularly pertinent to community relations decisions, the following variables were highly supported by the interviewee data: publics power to positively impact the corporation; support of the public by dominant coalition and employees; and availability of resources in the corporation. (Cancel et al., 1999, p. 189) Shin, Cameron, and Cropp (2006) furthered contingency theorys parsimony when they conducted a nationwide survey of public relations practitioners in order to find the most parsimonious explanation for con tingency theorys variables. Their results also reported statistical significance in the many variables. Ba sed on their findings, the authors used Cancel et al.s (1999) variables (within the previously established 11 categor ies) and their further findings to conduct a factor analysis. As with all f actor analyses, the fact ors were named based on commonalities among included variables. Based on the strong reliability in its factors (using Chronbach analysis), the authors identified the 26 external factors and 7 internal factors. Crisis and risk communication theory, while relatively quite new, have also become important areas of public relations. Stemming from something as simple as Murphys (1989)

PAGE 22

22 non-zero-sum game, organizations are realizing that mutually beneficial rela tionships are critical to their survival. Todays publics are much more business-savvy and also empowered like no other time in history (Friedman, 2005). These empowered publics often r ealize their leverage and regularly hold organizations socially responsible for their actions. But even those organizations that act socially responsibly undergo crises. In e ither case, the need for crisis public relations planning and management research is critical. Benoits (1995) image discourse theory, Coombs (1995) crisis responses, Threat Grid, and Crisis Type Matrix, and Coombs and Holladays (2002) Situational Cris is Communication Theory are a ll valuable contributions and being used regularly by scholars an d practitioners alike. Due to the newness of this type of theory and the fact that research done in th e area has not focused on one area but been very widely spread, it is too early in its evolution to generally evalua te crisis communication theories though specific theories are evaluated later in a separate secti on of this paper. As mentioned above, public relations methods have existed throughout history but it wasnt until the twentieth century that scholars began to search for theoretical underpinnings to the discipline. While studies like the Exce llence Theory and Contingency Theory have attempted to build a paradigmatic foundation for public relations, most re search leading up to this point was based on case study or practic al application (Ferguson, 1984). Ferguson recognized that public relations studies based on th eory-building were few. In arguably one of the most influential public relations works, Fer guson (1984) content analyz ed the state of public relations publications. Her rese arch examined published public relations journal articles and found only 4% of them to be theoretically-b ased. In 2003, Sallot, Lyon, Acosta-Alzuru, and Jones replicated this study and found 20% of articles to be th eoretically-based documenting a measureable increase in theoretical interest and research. Perhaps most important, however, was

PAGE 23

23 Fergusons observation that the field of public relations was in need of a logical unit-of-analysis. She concluded her 1984 paper by calling for the field to consider relationship as a unit-ofanalysis. Fergusons call was widely embraced by public relations scholars and has been increasingly adopted by practitioners. Since this initial call, relationship theory has developed into its own branch of public relations research and grown substantially since. In 1992, Ehling called this shift an important change in the pr imary mission of public relations. Broom, Casey and Ritchey (1997) furthered this perspectiv e by calling for an explication of the term organization-public relationship in order to further develop theory. Furthermore, they observed: The formation of relationships occurs when pa rties have perceptions and expectations of each other, when one or both parties need resources from the other, when one or both parties perceive mutual threats from an uncerta in environment, and when there is either a legal or voluntary necessity to associate (p. 95). This observation also translates well to cris is public relations pl anning and management. The mutual threats mentioned are an inherent component of crisis, and are discussed later in this paper. Ledingham (2001) identified four pivotal developments which spurred the emergence of the relational perspective as a framework for public relations study, teaching, and practice (p. 286): Recognition of the central role of relationshi ps in public relations. See Ferguson (1984) Reconceptualizing public relations as a manage ment function. See both Excellence Theory and Contingency Theory Identification of components a nd types of organization-public re lationships, their linkage to public attitudes, perceptions, knowledge a nd behavior, and relationship management strategies. Construction of organization-public relati onship models that accommodate relationship antecedents, process, and consequences.

PAGE 24

24 Two major models of the relationship theory have also been introduced. Broom, Casey and Ritchey (1997) developed a three-stage mode l based on antecedents, relationship state, and consequences of the relationship of interest. In 2000, Grunig and Huang developed a similar three-stage model based on situ ational antecedents, maintenance strategies, and relationship outcomes. Ledingham and Brunings (1998) relationship study identi fied five relationship dimensions: trust, openness, involvement, investment and commitment. Hon and Grunig (1999) advanced this line of research by further fles hing out relationship-management strategies and outcomes such as strategies, access, positivenes s, openness, assurance, networking, sharing of tasks, outcomes, control mutua lity, trust, satisfaction, commitmen t, communal relationships and exchange relationships. The dimension of trust, as defined in these st udies, is the dependent variable in this study and is discu ssed in-depth in th e next section. Finally, crisis public relations scholar Coombs (2000) also analyzed relationship theory from the crisis perspective. Hi s research found that relationships are affected (and in particular, can be damaged) by relational expectations and relationship history between an organization and its publics. Coombs crisis-focused research again highlighted the Broom, Casey, and Ritcheys 1997 findings. Measuring Trust in Public Relations Whenever philosophers, poets, statesmen, or theologians, has written about mans relationship to his fellow man, to nature, or to animals, the phenomena of trust and betrayal, faith and suspicion, responsibility and irresponsibil ity, have been discussed. -Morton Deutsch These words by Deutsch (1958), in his paper ent itled Trust and Suspicion, were written 50 years ago but the sentiment is no less powerfu l today when applied to organizational trust with its key publics. Morton util ized a type of two-person, non-zer o-sum game in order to test

PAGE 25

25 his hypotheses of trust. Based on game theory and today commonly referred to as Prisoners Dilemma, (Poundstone, 1993), he illustrated that if two parties could trust each other, they could also achieve mutually beneficial and desi rable circumstances. However, if either party violated the trust then both parties would be punished. While this mathematic model is too simplistic for direct application to most public relations research, its fo cus on the risks and rewards of mutu al trust do translate well. This model easily correlates to interpersonal relationshi ps but the tenets are th e essentially the same when analyzing relationships between an organiza tion and its publics. Simply put, shared trust between an organization and its key publics is necessary for the buildi ng and maintenance of mutually beneficial two-way comm unication. Public relations re search in the 1950s was nil but current research on trust within the discipline, like that reviewed in the previous section of this study, has acknowledged the critical importan ce of such organizat ion-public trust. As stated above, public relations scholars express such strong interest in the concept of trust because of its influence on relationships. P ublic relations theorists attempting to understand the implications of trust are often focused on rela tionship-based theories, as it is part and parcel of them. The unit-of-analysis fo r public relations has steadily evolved with the discipline, current research strongly supporting relationship as unit-of-analysis. Trust is often included as a component of such relationships. A review of this research follows. Relationship theories have grown in both number and relevance since Fergusons (1984) identification of relationship as a viable unit of analysis for pub lic relations scholars, as opposed to the (then) traditional use of individuals as unit of analysis (Ki & Hon, 2007). Grunig, Grunig, and Ehling (1992) furthered Fergusons explic ation of relationship-as-unit-of-analysis by including trust and credibility as additional dimensions of relationships. Broom, Casey, and

PAGE 26

26 Ritchey (1997) stated that Fergusons relationship orientatio n mix(es) characteristics of relationships with the perceptions of the parties in relationships, as well as constructs based on the reports of those in relationships. While th is could be problematic for researchers in the process of explicating a (then) new unit of analysis, it seems instrumental to truly understanding the implications of organizationalpublic trust relationships trust is based on just such a mix of characteristics. Organizational-public relatio nships, it seems, are largely based on varying degrees of trust or distrust. Many public relations researcher s agree with Kaspersons (19 86) reasoning that trust is multidimensional (eg., Hon & Grunig, 1999; Ki & Hon, 2007; Yang, 2007; Hall, 2006; Huang, 2001). Kasperson identified the following trust characteristics: competence, lack of bias, no hidden agenda or undue influence, and a sense th at the agency cares about those it serves and will provide adequate opportunities for people to air their concerns. While Kasperson is an environmental risk researcher, it is interesting that the dimensions identified are so closely aligned to contemporary public relations concep ts: no hidden agenda or undue influence and sense that the agency cares are very sim ilar to transparency a nd organizational social responsibility two terms regularly used in public relations resear ch and practice. Transparency is the full, accurate, and timely disclosure of in formation, a clear similarity with Kaspersons first dimension. Grunig (1992) defined social re sponsibility as excellent organizations manage with an eye of the effects of their decisions on society as well as on the organization (p.240). Publics that feel an or ganization is interested in the well-b eing of both itself and society will logically feel that the agency car es Kaspersons second dimension. As the evolution of the relationship manageme nt function of public relations progressed, so too has its research within the field. In fact, tw o influential public relations theories Excellence

PAGE 27

27 Theory and Contingency Theory are based h eavily on the relationshi p component of public relations. Excellence Theory (Grunig, Dozi er, Ehling, Grunig, Repper, & White, 1992) is structured on the concept of two-way symmetric al communication. Grunig et al. (1992) define two-way symmetrical comm unication as a model of public relati ons that is based on research and that uses communication to manage conf lict and improve understanding with strategic publics (p. 18). This improved understanding constitutes a relationshi p. Contingency Theory (Cancel, Cameron, Sallot, & Mitrook, 1997) is based on a continuum from total accommodation to total advocacy, which would be unnecessary unle ss it were applied to a relationship of some kind. Simply put, an organization must accommodate another party or advocate for itself in lieu of such accommodation also constituting a relationship. Watsons (2005) overview of trust research acr oss disciplines argued that each disciple must adapt its own measures of trust and conc luded the following: The question Do you trust them? must be qualified: trust them to do what, and how do you define trust? Hon and Grunigs (1999) Guide for Measuring Relationships in Public Relations provides an answer to that question by defining tr ust within public relations. Hon and Grunig (1999) identified six preci se components of relationships: control mutuality, trust, satisfaction, commitment, excha nge relationship, and communal relationship. In this context, Hon and Grunig defined trust as one partys level of confidence in and willingness to open oneself to the other party. Furthermore, they distilled three cl ear dimensions of trust: integrity, dependability, and competen ce. Integrity is defined as t he belief that an organization is fair and just, dependability as the belief th at an organization will do what it says it will do, and competence as the belief that an organization has the ability to do what it says it will do. Hon and Grunig (1999) tested these dime nsions with the following questions:

PAGE 28

28 1. This organization treats people like me fairly and justly. 2. Whenever this organization makes an important decision, I know it will be concerned about people like me. 3. This organization can be reli ed on to keep its promises. 4. I believe that this organization takes the opi nions of people like me into account when making decisions. 5. I feel very confident about this organizations skills. 6. This organization has the ability to accomplish what it says it will do. This conceptualization has been embraced by many public relations scholars (ie., Ki & Hon, 2007; Yang, 2007; Hall, 2006; Huang, 2001), a nd is used to conceptualize the trust variable in this study. Organizational Crisis Public Relations Crisis m anagement as an art has been around since the beginning of recorded time. Adam had to manage the first crisis after Eve pers uaded him to eat that apple. But crisis management as a science for American companies has been around just a few decades (Pines, 2000). Public relations itself is a relatively new discipline in respect to other communication disciplines, and younger still when compared to social sciences such as psychology and sociology. Furthermore, the sub-disc ipline of crisis public relations is in relative infancy. The first crisis public relations research was wr itten in 1982 (Ressler, 1982), not incidentally the same year as the well-known Tylenol cyanide cris is. Considered a paradigmatic case of how to effectively handle crisis, its case study is often analyzed to ascerta in effective methods of dealing with organizational crises (ie., Brassell-Ci cchini, 2003; Fearn-Banks 1994; LaPlant, 1999; Leeper, 1996; Marra, 1998). Organizations are always at risk for crisis, if for no othe r reason except that they cannot control all aspects of their environments. Ther efore it seems likely that organizational crises have existed since the formation of the first or ganizations. By exercising both their simple and

PAGE 29

29 complex operations, organizations accept the risks a ssociated with them. Furthermore, crises are a perceptual matter a crisis is determined by those affected by it, meaning that what might facilitate a crisis in one instan ce might not in another. For example, in the Intel Pentium chip crisis (Hearit, 1999), only chip users processing high-end calculat ions were noticeably affected by a design flaw in the chip. Most consumers would never even notice the flaw but once word of the flaw spread, the general publics perception of its effects created a full-fledged crisis for Intel soon all of Intels publics perc eived that they were affected. In public relations practice, a popular catchphrase is that perception is reality. In the case of Intel, it certainly seemed to be true; customers that were not actually functionall y affected by the flaw still perceived a crisis, constituting a very real one for Intel. Crises can have a number of facets, from mi squotes in the media to product recalls and from simple misunderstandings to accidents causi ng thousands of deaths. Even the most basic principles taught in introductory public relati ons courses such as accurate, honest and timely communication, audience analysis, transparenc y, social responsibility (Wilcox & Cameron, 2007) seem to be not only valid in a crisis, bu t may be even more sensitive than in normal everyday operations. Crises can highlight wh ere an organization has missed the mark or neglected certain areas of its public relations efforts. In order to understand crisis, it is helpful to look at how cris is is defined. The American Heritage Dictionary (1985) defines it as (a) a crucial or decisive poi nt or situation, turning point; (b) an unstable cond ition in political, in ternational, or economic a ffairs in which an abrupt or decisive change is impending. From pub lic relations practitioner Modzelweskis (1990) early definition as a situation getting out of c ontrol, the field of public relations has defined it in a number of ways. Crises have much to do with perceptions, especially those of external

PAGE 30

30 publics. Hayes (1985) suggested th at crisis occurs when there is a large, important difference between the expectations that organizational management has about the way its plans will interact with the environment a nd what actually happens. Pauc hant and Mitroff (1992) also offered a definition, stating crisis as a disruption that physically affects a system as a whole and threatens its basic assumptions, its subjectiv e sense of self, its existential core. Coombs (2002) substantiated th at a crisis can both carry th e potential to disrupt normal operations and damage an organizations reputa tion. Seeger, Sellnow, and Ulmer (1998) defined crisis as a specific, unexpected, and nonroutine even t or series of events that create high levels of uncertainty and threaten or ar e perceived to threaten an organizations high-prio rity goals. Seeger and Ulmer (2002) later refi ned their definition: a fundamental threat to the very stability of the system, a questioning of core assumptions and beliefs, and risk to high priority goals, including organizational image, legitimacy, pr ofitability, and ultimately survival. This definition will be used for the term crisis in this study. Some scholars have variations of their defin itions of crisis. Depending on what variables are being analyzed, scholars focus on different as pects of the situation. However it is clear that all research is concerned with potential risks th at can have a profound effect on the goals of an organization, whether they be normal operations, high-priority goals, or perceived responsibility, to name a few. It is important to note however that, regardless of the particular focus of the crisis analysis, much crisis research revolves around the concept of an organization either meeting or violating the publics expectations of integrity, depe ndability, and competence this studys conceptualization of trust. Examples mentioned above include Tylenols management of the cyanide crisis by meeting and exceeding public expectations and Intels mismanagement of the Pentium crisis by violating their customers expectations.

PAGE 31

31 The State of Organizational Crisis Public Relations. A review of the existing research finds that most crisis public relations resear ch is based on case st udies and theory/model development. Case studies are designed to gain greater understandi ng of how crises have occurred and been managed. Some authors devo te entire books to them, as Kathleen FearnBanks (2002) did with Crisis Communication: A Casebook Approach a compendium of public relations crises and lessons learned. Some other organi zational case studies include (alphabetically, by author): Berger (1999) The Upjohn Company Bobo (1997) Hitachi Boje & Rosile (2003); Seeger and Ulmer (2003) Enron Brinson & Benoit (1999) Texaco Carney & Jordan (1993) Sears Christen (2005) AT&T Duhe & Zoch (1994); Tyler (1997) Exxon Gonzalez-Herrero & Pratt (1996 ) McDonalds and Intel Greenberg (1993) Pepsi Hearit & Brown (2004) Merrill Lynch Ihlen (2002) Mercedes Kauffman (1997, 1999, 2001, 2005) NASA LaPlant (1999) Tylenol Martinelli & Briggs (1998) Odwalla Stevens (1999) The Prudential Insurance Company Taylor (2000) Coca-Cola The volatile airline industry is also, not surprisingly, a popular field of research (alphabetically, by author): Cowden and Sellnow (2002) Northwest Airlines Downing (2004) American Airlines Englehardt, Sallot, & Springston (2004) ValuJet Greer and Moreland (200 3) United Airlines McKinney et. al. (2005) United Airlines The multitude of case studies available to cris is public relations schol ars and practitioners can be considered a strength of the field as a resource but do they stand on their own? Considering that it is critical fo r the discipline of public relations to understa nd how to best plan

PAGE 32

32 for and manage crises, it would seem that case studies alone will not serve our goals. Certainly we can learn from the mistakes and successe s of previous cases but how do scholars and practitioners best predict the effectiveness of future crisis management? The answer to predictive questions like this is experimental research research that is explicitly used for its predictive quality. Unfortunatel y, there is almost no experiment al research on crisis public relations. However some scholars have focused on furthering theory within the discipline. Scholars are still building the foundation on whic h to formulate comprehensive crisis public relations theory. Much of the research is intr oduced by adapting existing theory from other areas of public relations or associated social sciences or by creating relatively new theory altogether including some models for crisis public relations. A review of th eoretical crisis public relations research follows. Gonzalez-Herrero and Pratt (1996) introduced an integrated symmetrical model based on Grunigs situational theory and issues management They posit that a proactive, symmetrical crisis-management process characterizes the model s four main steps: (a) issues management, (b) planning-prevention, (c) crisis, and (d) postcrisis. They incl ude a matrix visualizing the theoretical framework of th e model as well as steps to be taken at each stage. It should also be noted the model is applicable in general crisis public relations, though focused on crises caused by organizational mismanagement. Crisis can also affect the image of the organization, another area of crisis public relations research. Benoit (1997) called for effective me ssages through the use of image restoration discourse theory. He contended that it allows for practitioners to critically analyze their messages both during and after use. The author suggested that image restoration discourse theory identified the following:

PAGE 33

33 Remember to use persuasive messages; Organi zations at fault shoul d admit responsibility immediately; Deny charges when innocent; Sh ift the blame when appropriate; Identify outside crisis catalyst, if po ssible; Report plans to correc t/prevent problems; Remember that minimization may not always work; and Employ multiple strategies when appropriate. (pp. 183-184) Grunig, Grunig, and Doziers (2002) Excellent Public Re lation and Effective Organizations furthered their advocacy of ongoing, two-way symmetrical communications. This type of public relations is designe d to allow the public relations practitioner to be tter measure the perceptions of key publics in order to more eff ectively communicate perceptions that define the perceived trustworthiness of an organization. Their research also called for spokesperson expertise and inclusion of public relations into the dominant coalition of decision-makers at the executive level. These strategies are important to general public relations but become critical in times of crisis. Cancel, et al. (1997) introduced the Conti ngency Theory of accommodation based on a continuum from pure accommodation to pure advocacy. While it argued that Grunigs Excellence Theory is too basic, it may be overly complex th e authors identify 87 different variables that can affect the st ance of organizational-public rela tionships. This stance strongly affects the relationship of the parties in general and the trust level of the parties specifically. How accommodating or advocating an organization is, and has been, can influence the trust of its publics. Furthermore, crises can change the dynamic of these relationship and its dominant variables. There has been theory specifically designed for public relations crises. Reynolds and Seeger (2005) presented the CERC, or crisis and emergency risk communication, model. It is a blend of risk and crisis communication founded on the evolution from risk to crisis. Based on the authors contention that stages of risk and crises will have some level predictability, the model attempts to reduce uncertainty and improve crisis management. While the authors do not

PAGE 34

34 contend that the model can predict all developm ents, it attempts to help managers anticipate potential risks and crisis. The au thors also provided a table of distinguishing features of risk and crisis communication and a working model of the CERC. Marshs (2006) research revisiting Ryans Rhet orical Stasis Theory contended that crisis managers can use the theory to select the most appropriate apologia in times of need. Marsh argued that Rhetorical Stasis Theory allows managers a hierarchical template for examining the range and depth of crisis response strategies. While not comprehensive, such theory application has begun to build a foundation on which to better study crisis public relations. Most recently, Coombs (2007) offered his theory of crisis response, Situational Crisis Communication Theory. It is based on Attrib ution Theory and Coombs (1995, 2002) earlier research on crisis response st rategies (nonexistence, ingratia tion, mortification, suffering), his Crisis Type Matrix (Intentional-Unintentional an d External-Internal quadrants), and the Threat Grid that was designed to asses the type and threat a crisis represented. As Coombs pointed out, an important consideration when evaluating organi zational crisis public relations research is the attribution of the crisis. While previous rese arch links Attribution Th eory and crises (ie., Bradford & Garrett, 1995; Coombs, 1995; Hrtel, McColl-Kennedy, & McDonald, 1998; Jorgenson, 1994, 1996; McDonald & Hrtel, 20 00, Stockmyer, 1996), Coombs and Holladay (2002) used Situational Crisis Communication Theory to iden tify three crisis attribution (responsibility) clusters that are he lpful in conceptualizing crisis t ype as an experimental variable (Coombs and Holladay, 2002): (1) Accident cluster minimal attributions of crisis re sponsibility (technical-error accident, technical-error product harm, and challeng e), event considered unintentional or uncontrollable by the organization (2) Intentional cluster very strong attributions of crisis responsibility (human-error accident, human-error product harm, and organiza tional misdeed), event considered to be purposeful

PAGE 35

35 (3) Victim cluster very weak attributions of crisis responsibility (natural disasters, workplace violence, product tampering, and ru mor), organization viewed as victim of event Coombs and Holladays (2002) crisis clus ters were used to conceptualize the crisis type variable in this study. This experi ment will identify what effect th e crisis attribution cluster type has on public trust of an organization. To do so will require an organization to have a relationship with its public, one based one communication and, more specifically, trust. While an organization communicates with its publics regularly, it is especially vital during organizational crises in order to preserve its established trust. As mentioned above, while the effects of traditional public relations methods (ie., news releases, news conferences, media interviews) have been previously explored, this experiment will identify how public trust of an organization is affected duri ng types of crisis through inte ractive online media channels. Public Relations in the New (Media) World Today we are beginning to notice that the new m edia are not just mechanical gimmicks for creating worlds of illusion, but new languages with new and unique powers of expression. -Marshall McLuhan (Carpenter and McLuhan, 1960) What is new media? This is a question that has been asked throughout media history and can be difficult to answer. When McLuhan made the statement, he was speaking not of a particular medium but of medi a innovation and its continuous e volution. Due to the dynamic nature of media and rapidly developing technology, todays new media is considered to be based on interactive Intern et technology and its uses. As Pavlik and Dozier (1996) stated, little is known with great certainty about th e future of the information supe rhighway except that the pace of technological change is likely to continue and accelerate. Over a decade later, scholars and practitioners do know more about the future of the information superhighway than Pavlik and Dozier did then but there is st ill uncertainty about the influen ce of interactive online media.

PAGE 36

36 Robert Logan (2004), a collaborato r with McLuhan in the 1970s, furthered McLuhans research by identifying fourteen differences between new media and traditional mass media: (1) two-way communication, (2) ease of access to and dissemination of information, (3) continuous learning, (4) alignment and integrati on, (5) community, (6) portabil ity and time flexibility, (7) convergence, (8) interoperability, (9) aggregation of content, (10) increased variety and choice, (11) reduced gap between producers and consumers, (12) soci al collectivity and cooperation, (13) remix culture, and (14) the tr ansition from products to services. There has been very little research conduc ted on how the Internet and interactive online media affect public relations, howev er a review of the existing research follows. Taylor and Perry (2005) applied Everett Rogers (2003) Diffusion of Innovations Theory to online organizational crisis communication. Approximate ly 50% of their participating organizations reported using the Internet when responding to crisis. Jo and Kim (2003) analyzed the relationship between web charac teristics and perceptions toward relational components. Capriotti and Moreno (2007) discussed the importance of organizations expressing corporate social responsibility (CSR) thr ough interactive websites. Finally, Gonzalez-Herrero and Ruiz de Valbuena (2006) found that large international or ganizations attribute high importance to webbased media centers, though there ar e still shortcomings in many of their designs and content. While these studies are valuable in their own right, a void still exists for research on how trust is affected by use of interactive online media. Two new mediums that are increasingly being used by organizations are weblogs, or blogs, and streaming video on their organizational webs ites. A blog is a website that typically combines text, images and links in a kind of personal journal (Goodman, 2006). Blog use has exponentially increased in recent years. According to Technorati, a company that presents data

PAGE 37

37 on the growth of blogs, it is currently track ing 106.1 million blogs over double the 50 million blogs tracked in 2006. Seltzer and Mitrook (2007 ) suggested that blogs could potentially be better suited for online relationship building that its traditional counterparts online relationships built, at least in part, on trust. Streaming video is being increasingly used on organizational websites as a method for communicating with key publics. One of the main advantages of using streaming video is that the organization can disseminate messages as th ey choose without th e filter of a media gatekeeper however there is currently no research testing its effects. This experiment will test how use of blogs and streaming vi deo affect public trust of an or ganization during different types of crisis. Using specific components of Situational Cr isis Communication Th eory and the trust concept from relationship theory, this experiment tests how trust is affected by use of interactive online media during different types of crises. This type of research is important for both scholarly and practical reasons. First, the theoretical foundation th at this experiment is based on is relatively young when compared with other p ublic relations theories. This experiment analyzes components of Situat ional Crisis Communication Theo ry (crisis clusters) and the trust concept of relationship theory. Second, this rese arch builds understanding of how crises can best be managed in particular situati ons. Finally, this research explor es how interactiv e online media, specifically blogs and streaming video, affect trust in times of crisis. This research clearly identifies the concept of trust as a key dimension to relationships between an organization and its publi cs. Organizations must rely on trust with its publics if they are to achieve their goals of survivability and prof it. Organizations are also to subject to crisis making it critical to understand how to best manage them in order to mainta in public trust. Two

PAGE 38

38 increasingly popular inte ractive online media types are blog s and streaming video, which are used to also test their effectiveness in maintaining public tr ust during crisis. Hypotheses The theoretical perspective this research is based on, Situationa l Crisis Communication Theory (Coom bs, 2007), supposes the crisis manage r examining the crisis situation in order to asses the level of the reputational threat of a crisis (p. 137). This inherent level of reputational threat is expected to damage trust regardless of th e crisis cluster, simply by its nature. In other words, trust is expected to be damaged (to some extent) simply because of the presence of the crisis. H1: All treatments will report lower trust scores after crisis than reports from pretest. However, Coombs and Holladay (2001) theorize d that strong crisis responsibility and predisposition to a negative reput ation evaluation should make it more difficult to protect the organizations reputation (p. 338). While reputation evaluati on is not included in this experiment, crisis respon sibility (type) should st ill have some effect. H2: Intentional cluster treatments will report lower trust scores than accident or victim, regardless of medium. Research Questions In addition to the hypotheses stated above, this experiment also hopes to answer two research questions: RQ1: Which interactive online medium, blog or streaming video, will report higher trust scores after organizational crisis treatments? RQ2: Will use of either blog or streaming vide o be better suited to maintain trust during particular types of crises?

PAGE 39

39 In summary, the logical connect between the variables in this experiment may be best summed up by borrowing from Harold Lasswell. Just as Lasswell (1948) famously described early models of communication as Who (says) What (to) Whom (in) What Channel (with) What Effect, this experiment tests Organization (says) Crisis Comm unication (to) Its Publics (in) Interactive Online Media (with) Wh at Effect on Trust? The remainder of this proposal outlines the methodology for testing the variables conceptualized above.

PAGE 40

40 CHAPTER 3 METHODOLOGY This experim ent used a 2 (interactive onlin e media: blog, streaming video) X 3 (crisis cluster: accident, intentional, victim) design to test perceptions of public trust during organizational crisis. A manipulation check for in teractive online media type was unnecessary; it was obvious if participants were exposed to either blog treatment or streaming video treatment. However, a pilot test was necessary to ensure th at participants perceive d the crisis type news stories in the way intended in this study. Before delving further into the main study, it will be helpful to review the pilot study methodology and its findings. The next section summarizes the results of the pilot study and its manipulation check testing. Pilot Study Participan ts ( n=206) were graduate and undergraduate students at a large southeastern university. Due to the nature of the pilot study questions and the fact th at participants would have had the same past experience (read: none ) as any other population, this population should not have had an effect on the results. Participan ts were exposed to one of three articles about Vonnetek Automotive to test crisis attribution based on message copy. Participants were also asked to read the article and answer a survey ra ting their feelings about its content (Appendix B). The articles were written to manipulation check the crisis clusters: The Article One crisis was attributed to accident by respondents, the Articl e Two crisis was attri buted to intentional by respondents, and the Article Three crisis was attributed to victim by respondents. Conclusions from the Pilot Study. All pilot tests resulted in statistically significant reports of successful manipulation checks (see Table 4-1, Table 4-2, Table 4-3).

PAGE 41

41 Main Study Participants Participan ts ( n=377) were faculty and staff from a large southeastern university and business members of a chamber of commerce in a southeastern city. Demographic data of participants including sex, age, job position, years worked, organization type, state in which organization conducts business, number of empl oyees and number supervised, tenure at the organization, and website use was taken for ev aluation purposes. For purposes of internal validity, all participants were assigned to th e same pretest and each participant was randomly assigned to one posttest treatment (blog/accide nt, blog/intentional, blog/victim, streaming video/accident, streaming vide o/intentional, streaming vide o/victim). This study used nonprobability consecutive sample as it was prov ided the best populati on representation of nonprobability methods. While using the universit y faculty and staff and chamber of commerce members were a convenience sample, each partic ipant was randomly assigned one of the six posttests. Design The 2 X 3 pretest/posttest design was used to assess the effects of interactive online media type and crisis cluster type on ratings of trust, incl uding dimensions of integrity, competence and dependability. Ratings of trus t, based on Hon and Grunigs (1999) measures, were taken during both preand posttest (Appendix A). 1. Vonnetek Automotive treats people like me fairly and justly. 2. Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me. 3. Vonnetek Automotive can be relied on to keep its promises. 4. I believe that Vonnetek Automotive takes the opinions of people like me into account when making decisions.

PAGE 42

42 5. I feel very confident about Vonnetek Automotives skills. 6. Vonnetek Automotive has the ability to accomplish what it says it will do. Hon and Grunigs (1999) measures of satisfaction and communal satisfaction were also taken during the data collection process to disguise the trust variable. Vonnetek Automotive. Vonnetek Automotive was a fictional custom automobile manufacturer based out of Toledo, Ohio. An organizational website was built for Vonnetek Automotive specifically for this study. It was po sted immediately before testing began in order to assure that no participants could have familiarity with it. This organizational website included links to the Homepage, the About page, a picture Gallery, an Ev ents page, and a News page. The Homepage (Figure 3-1) consis ted of custom automobile pict ures and general organizational information. The About page (Figure 3-2) incl uded a fictional history of Vonnetek Automotive as well as contact information (also fictional). The Gallery page (Figure 3-3) simply included pictures of custom automobiles, with the expecta tion that visitors would think they were created by Vonnetek Automotive. The Events page (Figure 3-4) consisted of a list of Car Shows. These events, locations and dates are all actually real car shows to add to the realism of the website. Finally, the News page included eith er A) a brief note about a charity Vonnetek Automotive was participating in (in the pretest, Fi gure 3-5) or B) both that note and either a blog or streaming video with one of six treatments (blog/accident, blog/intentional, blog/victim, streaming video/accident, streaming video/intent ional, streaming video/victim) in the posttest (Figure 3-6 and Figure 3-7). Operationalization of Independent Variables Interactive online med ia type. Operationalization of inte ractive online media type was very simple. In blog treatments, the posttest News page used a blog for the news story. In streaming video treatments, the posttest News page used streaming video for the news story. For

PAGE 43

43 each crisis type treatments (accident, intentional, victim), both interactive online media formats included the same copy (which was tested during Pilot Test). Crisis type. For each set of interactive online me dia treatments, one posttest treatment utilized a victim cluster (in which Vonnetek Automo tive was perceived as a victim in Pilot Test), one posttest treatment utilized an accident cluster (which was perceived as an accident for Vonnetek Automotive in Pilot Test), and one posttes t treatment utilized an intentional cluster (in which was perceived to be intentionally caused by Vonnetek Automotive). See Pilot test for more on crisis type operationalization. Operationalization of Dependent Variable The online s urvey used semantic differenti al scales (Appendix C) to measure trust, satisfaction, and communal relationship using Hon and Grunigs ( 1999) questions (Appendix A). A reliability analysis was conducted for the 11 de pendent measures. No measures were excluded (Chronbachs Alpha=.82). Procedure All treatments were web-based. Eight-t housand surveys were emailed and 377 were completed for a response rate of 5%. Participan ts were sent an emailing generally describing the experiment and asked to click a link to a website. Clicking the link first displayed the informed consent document and then directed participants to begin the treatment. Participants were asked to evaluate relationship characte ristics of Vonnetek Automotive. In the pretest, all particip ants were emailed a link to the same Vonnetek Automotive website and then they answered an online survey measuring th eir perceived trust of Vonnetek Automotive within the matrix of relationship questi ons. This pretest also established the control group for the experiment. One week later, partic ipants were exposed to (posttest) one of six treatments (blog/accident, blog/intentional, blog/victim, streaming video/accident, streaming

PAGE 44

44 video/intentional, streaming video/victim) and th en answered an online survey measuring their perceived trust (within the matrix of relations hip questions) of Vonnetek Automotive to measure effects of interactive online medi a type and crisis type on trust of the organization. This method tested for main and interaction effects of th e two independent variables, as also allowed replication of each of the six conditions. All tr eatments and measures were taken online. The participants were debriefed on the experiment and procedur e following completion of the dependent measures. Administrato rs at the university and the ch amber of commerce did not give permission at the same time. This resulted in the chamber of commerce being administered the survey two weeks before the university. In each case, all organizationa l participants were emailed instructions simultaneously, though simu ltaneous administration of treatments may not be ensured due to the nature of the online ad ministration. In each case, the posttest was administered 10 days after the pr etest to ensure consistency. Data Analysis Descriptive data was analyzed. Two-wa y ANOVAs were conducted to test main and interaction effects of independent variables interactive online media type and crisis type on dependent variable trust Bonferroni correction was used fo r each effect in order to ensure accuracy of statistical significance. Furthermore, one-ANOVAs were also conducted on each independent variable with each dependent variab le to reduce instances of chance statistical significance.

PAGE 45

45 Table 3-1 Pilot test crosstabs Response accidentResponse intentionalResponse victimTotal Accident treatment 60 6 1 67 Intentional treatment 6 63 0 69 Victim treatment 23 8 39 70 TOTAL 89 77 40 206

PAGE 46

46 Table 3-2 Pilot test ANOVA (DV: A TTRIBUTION, IV: CRISISTYPE) *p<.05 Source Sum of SquaresDf Mean SquareF Sig. Between Groups 64.98 2 32.49 145.35* .00* Within Groups 45.38 203.22 Total 110.35 205

PAGE 47

47 Table 3-3 Pilot test post hoc Bonferroni (DV: ATTRIBUTION, IV: CRISISTYPE) Dependent Variable Independent Variable Std. Error Sig. ATTRIBUTION CRISISTYPE Accident Intentional Victim .08 .08 .00* .00* IntentionalAccident Victim .08 .08 .00* .00* Victim Accident Intentional .08 .08 .00* .00* *p<.05

PAGE 48

48 Figure 3-1 Vonnetek Automotive Home page both Pretest and Posttest

PAGE 49

49 Figure 3-2 Vonnetek Automotive About page both Pretest and Posttest

PAGE 50

50 Figure 3-3 Vonnetek Automotive Gallery page both Pretest and Posttest

PAGE 51

51 Figure 3-4 Vonnetek Automotive Events page both Pretest and Posttest

PAGE 52

52 Figure 3-5 Vonnetek Automotive News page Pretest treatment

PAGE 53

53 Figure 3-6 Vonnetek Automotive News page Posttest Blog treatment

PAGE 54

54 Figure 3-7 Vonnetek Automotive News page Posttest Streaming Video treatment

PAGE 55

55 CHAPTER 4 RESULTS This chapter presents the results of th e experime nt described in Chapter 3. It begins with a review of data analysis conducted using the data collected during in the experiment. This review is followed by the results of hypothesi s testing and research questions. Preparing the Data for Analysis Online surv eys on www.surveymonkey.com were clos ed two weeks after posttest administration. Data was downloaded to an Excel spreadsheet and then tr ansferred into a SPSS data set. The surveys were designed so that each question was required to have an exact number of responses (for instance, one and only one answ er on Likert-type scales) and all questions were required to be answered in order to finish the surveys. This insured against missing values and that no impossible values were entered. Sample Demographics The average age of the respondents was 45.55. Fe males comprised 62.3% of respondents and 37.7% were male. The average respondent has been in their position for 8.99 years and supervises 5.25 employees. In terms of total ye ars at their organization, 35.3% reported working at their organization for More than 10 year s, 29.4% reported 1-5 years, 16.1% reported 6-10 years, and 6.3% reported Less than one year. For organization type, 75.8 % indicated Education, 9% indicated Corporation, 4.5% indicated Gove rnment, 4.1% indicated Not-for-profit, 1.2% indicated Consultant, and 5.3% indicated Other. These results were not surprising considering that respondents were faculty and staff at a large southeastern unive rsity and members of a chamber of commerce in a southeastern city. Th e vast majority of respondents indicated that their organization has a website (97.8%) and respondents reported spending an average of 9.9 hours per week conducting work-related activities on their website.

PAGE 56

56 Results of the Statistical Analysis The results of the statistical an alysis of this study are divide d into two s ections. The first section reviews how the analyses address the hypotheses introduced in Chapter 2. The second section evaluates the data analyses for answers to the research questions introduced in Chapter 2. Tables 4-1 through 4-11 illustrate descri ptives of two-way ANOVAs showing mean, standard deviation, and sample size (N). This is helpful to illustrate a general breakdown of the reported data. In order to test statistical si gnificance in these findi ngs, two-way ANOVAs were used to compare means across categories of one qualitative variable, contro lling for another.

PAGE 57

57 Table 4-1 Two-way ANOVA descriptive st atistics (DV: TREATSFAIRLY) Crisis type Media type MeanStd. DeviationN Pretest 4.02 .85 977 Accident Blog Streaming video Total 3.51 4.03 3.77 1.36 1.22 1.31 39 38 77 Intentional Blog Streaming video Total 2.73 2.61 2.67 1.30 1.43 1.37 45 51 96 Victim Blog Streaming video Total 4.66 4.55 4.60 1.19 1.42 1.31 35 42 77 Total Pretest Blog Streaming video Total 4.02 3.55 3.64 3.85 .845 1.50 1.60 1.20 377 119 131 627 Vonnetek Automotive treats people like me fairly and justly.

PAGE 58

58 Table 4-2 Two-way ANOVA descriptive st atistics (DV: CONCERNEDABOUT) Crisis type Media type MeanStd. DeviationN Pretest 3.58 1.19 377 Accident Blog Streaming video Total 3.54 3.84 3.69 1.32 1.33 1.32 39 38 77 Intentional Blog Streaming video Total 2.62 2.43 2.52 1.21 1.25 1.23 45 51 96 Victim Blog Streaming video Total 4.17 4.07 4.12 1.32 1.40 1.36 35 42 77 Total Pretest Blog Streaming video Total 3.58 3.38 3.37 3.49 1.19 1.42 1.52 1.31 377 119 131 627 Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me.

PAGE 59

59 Table 4-3 Two-way ANOVA descriptive statistics (DV: KEEPPROMISES) Crisis type Media type MeanStd. DeviationN Pretest 4.15 .846 377 Accident Blog Streaming video Total 3.82 4.05 3.94 1.34 1.11 1.23 39 38 77 Intentional Blog Streaming video Total 2.60 2.47 2.53 1.25 1.35 1.30 45 51 96 Victim Blog Streaming video Total 4.66 4.43 4.53 1.21 1.35 1.28 35 42 77 Total Pretest Blog Streaming video Total 4.15 3.61 3.56 3.93 .846 1.52 1.55 1.20 377 119 131 627 Vonnetek Automotive can be relied on to keep its promises.

PAGE 60

60 Table 4-4 Two-way ANOVA descriptive stat istics (DV: TAKESOPINIONSOF) Crisis type Media type MeanStd. DeviationN Pretest 3.73 1.35 377 Accident Blog Streaming video Total 3.51 3.84 3.68 1.23 1.29 1.26 39 38 77 Intentional Blog Streaming video Total 3.09 2.47 2.76 1.44 1.36 1.43 45 51 96 Victim Blog Streaming video Total 4.11 4.26 4.19 1.35 1.43 1.39 35 42 77 Total Pretest Blog Streaming video Total 3.73 3.53 3.44 3.63 1.35 1.40 1.57 1.41 377 119 131 627 I believe that Vonnetek Automotive takes the opi nions of people like me into account when making decisions.

PAGE 61

61 Table 4-5 Two-way ANOVA descriptive stat istics (DV: CONFIDENTSKILLS) Crisis type Media type MeanStd. DeviationN Pretest 4.40 1.14 377 Accident Blog Streaming video Total 3.79 3.45 3.62 1.61 1.29 1.46 39 38 77 Intentional Blog Streaming video Total 3.56 3.59 3.57 1.56 1.56 1.55 45 51 96 Victim Blog Streaming video Total 4.69 4.26 4.45 1.08 1.35 1.24 35 42 77 Total Pretest Blog Streaming video Total 4.40 3.97 3.76 4.19 1.14 1.52 1.45 1.32 377 119 131 627 I feel very confident about Vonnetek Automotives skills.

PAGE 62

62 Table 4-6 Two-way ANOVA descriptive stat istics (DV: ABILITYACCOMPLISH) Crisis type Media type MeanStd. DeviationN Pretest 4.40 .99 377 Accident Blog Streaming video Total 4.31 3.97 4.14 1.54 .85 1.25 39 38 77 Intentional Blog Streaming video Total 3.87 3.90 3.89 1.42 1.59 1.51 45 51 96 Victim Blog Streaming video Total 4.80 4.43 4.60 1.26 1.09 1.17 35 42 77 Total Pretest Blog Streaming video Total 4.40 4.29 4.09 4.32 .99 1.46 1.27 1.16 377 119 131 627 Vonnetek Automotive has the ability to accomplish what it says it will do.

PAGE 63

63 Table 4-7 Two-way ANOVA descriptive stat istics (DV: SOUNDPRINCIPLES) Crisis type Media type MeanStd. DeviationN Pretest 4.21 .91 377 Accident Blog Streaming video Total 3.64 4.11 3.87 1.31 1.43 1.38 39 38 77 Intentional Blog Streaming video Total 2.33 2.22 2.27 1.07 1.36 1.23 45 51 96 Victim Blog Streaming video Total 4.74 4.69 4.71 1.42 1.52 1.47 35 42 77 Total Pretest Blog Streaming video Total 4.21 3.47 3.56 3.93 .91 1.59 1.80 1.33 377 119 131 627 Sound principles seem to guide Vonnetek Automotives behavior.

PAGE 64

64 Table 4-8 Two-way ANOVA descriptive stat istics (DV: DOESNOTMISLEAD) Crisis type Media type MeanStd. DeviationN Pretest 4.00 .96 377 Accident Blog Streaming video Total 3.23 3.71 3.47 1.50 1.25 1.39 39 38 77 Intentional Blog Streaming video Total 2.20 2.24 2.22 1.08 1.41 1.26 45 51 96 Victim Blog Streaming video Total 4.34 4.31 4.32 1.28 1.37 1.32 35 42 77 Total Pretest Blog Streaming video Total 4.00 3.17 3.33 3.70 .96 1.55 1.62 1.30 377 119 131 627 Vonnetek Automotive does not mislead people like me.

PAGE 65

65 Table 4-9 Two-way ANOVA descriptive stat istics (DV: MAKEDECISIONSFOR) Crisis type Media type MeanStd. DeviationN Pretest 2.77 1.41 377 Accident Blog Streaming video Total 2.77 2.66 2.71 1.39 1.28 1.33 39 38 77 Intentional Blog Streaming video Total 1.98 1.82 1.90 1.03 1.11 1.07 45 51 96 Victim Blog Streaming video Total 3.14 2.90 3.01 1.61 1.45 1.52 35 42 77 Total Pretest Blog Streaming video Total 2.77 2.58 2.41 2.66 1.41 1.42 1.35 1.41 377 119 131 627 I am very willing to let Vonnetek Automotive make decisions for people like me.

PAGE 66

66 Table 4-10 Two-way ANOVA descriptive sta tistics (DV: WATCHCLOSELY) Crisis type Media type MeanStd. DeviationN Pretest 3.97 1.22 377 Accident Blog Streaming video Total 3.21 3.71 3.45 1.38 1.41 1.41 39 38 77 Intentional Blog Streaming video Total 2.53 2.78 2.67 1.08 1.72 1.46 45 51 96 Victim Blog Streaming video Total 4.11 3.93 4.01 1.41 1.31 1.35 35 42 77 Total Pretest Blog Streaming video Total 3.97 3.22 3.42 3.71 1.22 1.43 1.59 1.38 377 119 131 627 I think it is important to watch Vonnetek Automotiv e closely so that it does not take advantage of people like me.

PAGE 67

67 Table 4-11 Two-way ANOVA descriptive statis tics (DV: SUCCESSFULATTHINGS) Crisis type Media type MeanStd. DeviationN Pretest 4.19 .89 377 Accident Blog Streaming video Total 4.33 4.03 4.18 1.16 1.20 1.18 39 38 77 Intentional Blog Streaming video Total 4.07 3.88 3.97 .99 1.31 1.17 45 51 96 Victim Blog Streaming video Total 4.49 4.26 4.36 1.04 .99 1.01 35 45 77 Total Pretest Blog Streaming video Total 4.19 4.28 4.05 4.18 .89 1.07 1.18 .99 377 119 131 627 Vonnetek Automotive is known to be successful at the things it tries to do.

PAGE 68

68 Tables 4-12 through 4-22 illustrate these tests of between-subjects effects showing sum of squares, degrees of freedom, mean square, F-value, and p-value.

PAGE 69

69 Table 4-12 Tests of between-subjects effects (DV: TREATSFAIRLY) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 161.54 2 80.77 71.46* .00* MEDIATYPE .53 1 .53 .469 .49 CRISISTYPE*MEDIATYPE 5.31 2 2.66 2.35 .10 Error 700.75 6201.13 *p<.05 Vonnetek Automotive treats people like me fairly and justly.

PAGE 70

70 Table 4-13 Tests of between-subjects effe cts (DV: CONCERNEDABOUT) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 119.00 2 59.50 38.93* .00* MEDIATYPE .00 2 .00 .00 .98 CRISISTYPE*MEDIATYPE 2.83 2 1.42 .93 .40 Error 947.69 6201.53 *p<.05 Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me.

PAGE 71

71 Table 4-14 Tests of between-subjects ef fects (DV: KEEPPROMISES) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 184.95 2 92.47 86.04* .00* MEDIATYPE .11 1 .11 .10 .75 CRISISTYPE*MEDIATYPE 2.29 2 1.15 1.07 .35 Error 666.39 6201.08 *p<.05 Vonnetek Automotive can be relied on to keep its promises.

PAGE 72

72 Table 4-15 Tests of between-subjects effe cts (DV: TAKESOPINIONSOF) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 88.16 2 44.08 24.17* .00* MEDIATYPE .14 1 .14 .08 .78 CRISISTYPE*MEDIATYPE 11.13 2 5.57 3.05* .05* Error 1130.75 6201.82 *p<.05 I believe that Vonnetek Automotive takes the opi nions of people like me into account when making decisions.

PAGE 73

73 Table 4-16 Tests of between-subjects effe cts (DV: CONFIDENTSKILLS) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 41.10 2 20.55 12.82* .00* MEDIATYPE 3.73 1 3.73 2.33 .13 CRISISTYPE*MEDIATYPE 2.63 2 1.32 .82 .44 Error 993.60 6201.60 *p<.05 I feel very confident about Vonnetek Automotives skills.

PAGE 74

74 Table 4-17 Tests of between-subjects effe cts (DV: ABILITYACCOMPLISH) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 22.87 2 11.43 8.82* .00* MEDIATYPE 3.07 1 3.07 2.37 .12 CRISISTYPE*MEDIATYPE 2.23 2 1.12 .86 .42 Error 803.59 6201.30 *p<.05 Vonnetek Automotive has the ability to accomplish what it says it will do.

PAGE 75

75 Table 4-18 Tests of between-subjects effe cts (DV: SOUNDPRINCIPLES) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 267.29 2 133.65 109.42* .00* MEDIATYPE .53 1 .53 .49 .49 CRISISTYPE*MEDIATYPE 4.11 2 2.06 1.68 .19 Error 757.29 6201.22 *p<.05 Sound principles seem to guide Vonnetek Automotives behavior.

PAGE 76

76 Table 4-19 Tests of between-subjects ef fects (DV: DOESNOTMISLEAD) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 194.10 2 97.05 77.74* .00* MEDIATYPE 1.59 1 1.59 1.27 .26 CRISISTYPE*MEDIATYPE 3.05 2 1.52 1.22 .30 Error 773.98 6201.25 *p<.05 Vonnetek Automotive does not mislead people like me.

PAGE 77

77 Table 4-20 Tests of between-subjects effe cts (DV: MAKEDECISIONSFOR) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 58.81 2 29.41 15.63* .00* MEDIATYPE 1.74 1 1.74 .92 .34 CRISISTYPE*MEDIATYPE .16 2 .08 .04 .96 Error 1166.69 6201.88 *p<.05 I am very willing to let Vonnetek Automotive make decisions for people like me.

PAGE 78

78 Table 4-21 Tests of between-subjects ef fects (DV: WATCHCLOSELY) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 80.83 2 40.42 24.00* .00* MEDIATYPE 2.23 1 2.23 1.32 .25 CRISISTYPE*MEDIATYPE 4.70 2 2.35 1.39 .25 Error 1044.07 6201.68 *p<.05 I think it is important to watch Vonnetek Automotiv e closely so that it does not take advantage of people like me.

PAGE 79

79 Table 4-22 Tests of between-subjects effect s (DV: SUCCESSFULATTHINGS) Source Sum of Squaresdf Mean SquareF Sig. CRISISTYPE 6.81 2 3.41 3.50* .03* MEDIATYPE 3.50 1 3.50 3.60 .06 CRISISTYPE*MEDIATYPE .16 2 .08 .08 .92 Error 603.46 620.97 *p<.05 Vonnetek Automotive is known to be successful at the things it tries to do.

PAGE 80

80 Finally, Tables 4-23 through 433 illustrate multiple comparison post-hoc tests for each variable using Bonferroni correction. Bonferroni correcti on was used to reduce the chance of false statistical significance due to conducting multiple tests simultaneously on the data. For all variables exhibiting main effects, one-way ANO VAs were conducted to further reduce the odds of false significance based on chance (Appendix F). One-way ANOVAs were used to test equality of three or more means at one time by using variances. In all cases, the one-way ANOVAs confirmed statistical sign ificance of tested variables). Participant responses did not report statistical significance for any of the demographic data: sex, age, job position, years worked, organization type, state in which organization conducts business, number of employees and number supervised, tenure at the organizatio n, and website use was taken for evaluation purposes

PAGE 81

81 Table 4-23 Multiple comparisons post hoc Bonferroni (DV: TREATSFAIRLY) Dependent Variable Independent Variable Std. Error Sig. TREATSFAIRLY CRISISTYPE Pretest Accident Intentional Victim .13 .12 .13 .32 .00* .00* Accident Pretest Intentional Victim .13 .16 .17 .32 .00* .00* IntentionalPretest Accident Victim .12 .16 .16 .00* .00* .00* Victim Pretest Accident Intentional .13 .17 .16 .00* .00* .00* *p<.05 Vonnetek Automotive treats people like me fairly and justly.

PAGE 82

82 Table 4-24 Multiple comparisons post hoc Bo nferroni (DV: CONCERNEDABOUT) Dependent Variable Independent Variable Std. Error Sig. CONCERNEDABOUT CRISISTYPE Pretest Accident Intentional Victim .16 .14 .16 1.00 .00* .00* Accident Pretest Intentional Victim .16 .19 .20 1.00 .00* .19 IntentionalPretest Accident Victim .14 .19 .19 .00* .00* .00* Victim Pretest Accident Intentional .16 .20 .19 .00* .19 .00* *p<.05 Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me.

PAGE 83

83 Table 4-25 Multiple comparisons post hoc Bonferroni (DV: KEEPPROMISES) Dependent Variable Independent Variable Std. Error Sig. KEEPPROMISES CRISISTYPE Pretest Accident Intentional Victim .13 .12 .13 .55 .00* .02* Accident Pretest Intentional Victim .13 .16 .17 .55 .00* .00* IntentionalPretest Accident Victim .12 .16 .16 .00* .00* .00* Victim Pretest Accident Intentional .13 .17 .16 .02* .00* .00* *p<.05 Vonnetek Automotive can be relied on to keep its promises.

PAGE 84

84 Table 4-26 Multiple comparisons post hoc Bo nferroni (DV: TAKESOPINIONSOF) Dependent Variable Independent Variable Std. Error Sig. TAKESOPINIONSOF CRISISTYPE Pretest Accident Intentional Victim .17 .15 .17 1.00 .00* .04* Accident Pretest Intentional Victim .17 .21 .22 1.00 .00* .10 Intentional Pretest Accident Victim .15 .21 .21 .00* .00* .00* Victim Pretest Accident Intentional .17 .22 .21 .04* .10 .00* MEDIATYPE Pretest Blog Streaming Video .14 .14 .46 .11 Blog Pretest Streaming Video .14 .17 .46 1.00 Streaming Video Pretest Blog .14 .17 .11 1.00 *p<.05 I believe that Vonnetek Automotive takes the opi nions of people like me into account when making decisions.

PAGE 85

85 Table 4-27 Multiple comparisons post hoc Bo nferroni (DV: CONFIDENTSKILLS) Dependent Variable Independent Variable Std. Error Sig. CONFIDENTSKILLS CRISISTYPE Pretest Accident Intentional Victim .16 .15 .16 .00* .00* 1.00 Accident Pretest Intentional Victim .16 .19 .20 .00* 1.00 .00* IntentionalPretest Accident Victim .15 .19 .19 .00* 1.00 .00* Victim Pretest Accident Intentional .16 .20 .19 1.00 .00* .00* *p<.05 I feel very confident about Vonnetek Automotives skills.

PAGE 86

86 Table 4-28 Multiple comparisons post hoc Bo nferroni (DV: ABILITYACCOMPLISH) Dependent Variable Independent Variable Std. ErrorSig. ABILITYACCOMPLISH CRISISTYPE Pretest Accident Intentional Victim .14 .13 .14 .41 .00* 1.00 Accident Pretest Intentional Victim .14 .17 .18 .41 .84 .08 IntentionalPretest Accident Victim .13 .17 .17 .00* .84 .00* Victim Pretest Accident Intentional .14 .18 .14 1.00 .08 .00* *p<.05 Vonnetek Automotive has the ability to accomplish what it says it will do.

PAGE 87

87 Table 4-29 Multiple comparisons post hoc Bo nferroni (DV: SOUNDPRINCIPLES) Dependent Variable Independent Variable Std. Error Sig. SOUNDPRINCIPLES CRISISTYPE Pretest Accident Intentional Victim .14 .13 .14 .09 .00* .00* Accident Pretest Intentional Victim .14 .17 .18 .09 .00* .00* IntentionalPretest Accident Victim .13 .17 .17 .00* .00* .00* Victim Pretest Accident Intentional .14 .18 .17 .00* .00* .00* *p<.05 Sound principles seem to guide Vonnetek Automotives behavior.

PAGE 88

88 Table 4-30 Multiple comparisons post hoc Bonferroni (DV: DOESNOTMISLEAD) Dependent Variable Independent Variable Std. Error Sig. DOESNOTMISLEAD CRISISTYPE Pretest Accident Intentional Victim .14 .13 .14 .00* .00* .12 Accident Pretest Intentional Victim .14 .17 .18 .00* .00* .00* IntentionalPretest Accident Victim .13 .17 .17 .00* .00* .00* Victim Pretest Accident Intentional .14 .18 .17 .12 .00* .00* *p<.05 Vonnetek Automotive does not mislead people like me.

PAGE 89

89 Table 4-31 Multiple comparisons post hoc Bo nferroni (DV: MAKEDECISIONSFOR) Dependent Variable Independent Variable Std. Error Sig. MAKEDECISIONSFOR CRISISTYPE Pretest Accident Intentional Victim .17 .16 .17 1.00 .00* .94 Accident Pretest Intentional Victim .17 .21 .22 1.00 .00* 1.00 IntentionalPretest Accident Victim .16 .21 .21 .00* .00* .00* Victim Pretest Accident Intentional .17 .22 .21 .94 1.00 .00* *p<.05 I am very willing to let Vonnetek Automotive make decisions for people like me.

PAGE 90

90 Table 4-32 Multiple comparisons post hoc Bonferroni (DV: WATCHCLOSELY) Dependent Variable Independent Variable Std. Error Sig. WATCHCLOSELY CRISISTYPE Pretest Accident Intentional Victim .16 .15 .16 .01* .00* 1.00 Accident Pretest Intentional Victim .16 .20 .21 .01* .00* .05* IntentionalPretest Accident Victim .15 .20 .20 .00* .00* .00* Victim Pretest Accident Intentional .16 .21 .20 1.00 .05* .00* *p<.05 I think it is important to watch Vonnetek Automotiv e closely so that it does not take advantage of people like me.

PAGE 91

91 Table 4-33 Multiple comparisons post hoc Bonf erroni (DV: SUCCESSFULATTHINGS) Dependent Variable Independent Variable Std. ErrorSig. SUCCESSFULATTHIN GS CRISISTYPE Pretest Accident Intentional Victim .12 .11 .12 1.00 .28 1.00 Accident Pretest Intentional Victim .12 .15 .16 1.00 .95 1.00 IntentionalPretest Accident Victim .11 .15 .15 .28 .95 .06 Victim Pretest Accident Intentional .12 .16 .15 1.00 1.00 .06 *p<.05 Vonnetek Automotive is known to be successful at the things it tries to do.

PAGE 92

92 Results Related to the Hypotheses H1: All treatments will report lower trust scores after crisis than reports from pretest. To test Hypothesis 1, descriptive analysis was first performed to see if posttest trust scores were lower than pretest treatments and then tw o-way ANOVAs were conducted to find statistical significance. Hypothesis 1 was strongly supported as pretests reported st atistically significant higher means than Intentional tr eatments for 9 of 11 dependent variables: Vonnetek Automotive treats people like me fairly, (4.02 vs. 2.67), F (2, 620)=71.46, p=.00) Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me (3.58 vs. 2.52), F (2, 620)=38.93, p=.00, Vonnetek Automotive can be relied on to keep its promises (4.15 vs. 2.53), F (2, 620)=86.04, p=.00, I feel very confiden t about Vonnetek Automotive s skills (4.40 vs. 3.57), F (2, 620)=12.82, p=.00, Vonnetek Automotiv e has the ability to accomplish what it says it will do (4.40 vs. 3.89), F (2, 620)=8.82, p=.00, Sound principles seem to guide Vonnetek Automo tives behavior (4.21 vs. 2.27), F (2, 620)=109.42, p=.00, Vonnetek Automotive does not mislead people like me (4.00 vs. 2.22), F (2, 620)=77.74, p=.00, I am very willing to let Vonnetek Auto motive make decisions for people like me (2.77 vs. 1.90), F (2, 620)=15.63, p=.00, and I think it is importa nt to watch Vonnetek Automotive closely so that it does not take advantage (3.97 vs 2.67), F (2, 620)=24.00, p=.00. Pretests also reported statistically significant higher means than Accide nt treatments in 3 of 11 dependent variables: I feel very confident about Vonnetek Automotives skills (4.40 vs. 3.62), F (2, 620)=12.82, p=.00, Vonnetek Automotive does not mislead people like me (4.00 vs. 3.47), F (2, 620)=77.74, p=.00, and I think it is importa nt to watch Vonnetek Automotive closely so that it does not take advantage (3.97 vs. 3.45), F (2, 620)=24.00, p=.00.

PAGE 93

93 However, Pretests also reported significantly lower means than Victim treatments in 4 of 11 dependent variables: Vonnetek Automotive treats people like me fairly (4.02 vs. 4.60), F (2, 620)=71.46, p=.00, Whenever Vonnetek Automotive ma kes an important decision, I know it will be concerned about people like me (3.58 vs. 4.12), F (2, 620)=38.93, p=.00, Vonnetek Automotive can be relied on to keep its promises (4.15 vs. 4.53), F (2, 620)=86.04, p=.00, and Sound principles seem to guide Vonnete k Automotives behavior (4.21 vs. 4.71), F (2, 620)=109.42, p=.00. H2: Intentional cluster treatments will repor t lower trust scores than accident or victim, regardless of medium. To test Hypothesis 2, descriptive analysis was first performed to see if the Intentional treatment mean was lower than Accident a nd Victim treatments and then two-way ANOVAs were conducted to find statistical significance. Hypothesis 2 wa s strongly supported as all 11 of 11 Intentional treatments reported lower trust sc ore means than Accident and Victim and the difference was statistically significant in 9 of 11 dependent variables: Vonnetek Automotive treats people like me fairly, F (2, 620)= 55.67, p=.00, Whenever Vonnetek Automotive makes and important decision, I know it will be concerned about people like me, F (2, 620)= 27.57, p=.00, Vonnetek Automotive can be re lied on to keep its promises, F (2, 620)= 72.86, p=.00, I believe that Vonnetek Automo tive takes the opinions of people like me into account when making decisions, F (2, 620)= 18.40, p=.00, Sound principles seem to guide Vonnetek Automotives behavior, F (2, 620)= 93.05, p=.00, Vonnetek Automotive does not mislead people like me, F (2, 620)= 74.37, p =.00, I am very willing to let Vonnetek Automotive make decisions for people like me, F (2, 620)= 12.51, p =.00, and I think it is important to watch Vonnetek Automotive closely so th at it does not take advantage, F (2, 620)= 28.20, p=.00.

PAGE 94

94 Results Related to the Research Questions RQ1: Which interactive online medium, blog or streaming video, will report higher trust scores after organizational crisis treatments? To answer Research Question 1, two-way ANOVAs were conducted to test for main effects of independent variable Media Type (Blo g, Streaming Video). None of the 11 dependent variables reported statistically significant main effects. RQ2: Will use of either blog or streaming vi deo be better suited to maintain trust during particular types of crises? To answer Research Question 2, further tes ting from Research Question 1 was required. Two-way ANOVAs were conducted to test for interaction effects of independent variables Crisis Type (Accident, Intentional, Vi ctim) and Media Type (Blog, Streaming Video). One dependent variable reported statistically significant interaction e ffects: I believe th at Vonnetek Automotive takes the opinions of people like me into acc ount when making decisi ons (Table 4-4b), a dependability dimension measure. For I believe that Vonnetek Automotive takes the opinions of people like me into account when making decisions, F (1, 620)=3.05, p=.05, Streaming Video in Accident treatments (3.84) reported significantly higher means than Blog in Accident treatments (3.51) and Streaming Video in Victim treatments (4.26) reported signi ficantly higher means than Blog in Victim treatments (4.11). However, Blogs in Intentiona l treatments (3.09) reported significantly higher means than Streaming Video in Inten tional treatments (2.47) (Figure 4-1).

PAGE 95

95 I believe that Vonnetek Automotive takes the opi nions of people like me into account when making decisions. Figure 4-1 Graph of interac tion effects TAKESOPINIONSOF victim intentional accident pretest Crisis type accident, intentional, or victim 4.5 4 3.5 3 2.5 Mean I believe that Vonn etek Automotive takes the opinions of people like me into account when making decisions. streaming video blog pretest Media type blog or streaming video

PAGE 96

96 CHAPTER 5 DISCUSSION This study used an experim ental design to test potential effects of interactive online media type and crisis type on public trust of an organization during an organizational crisis. The goal of this research was two-fold. The first goa l was to further the body of knowledge of public relations research in order to better understand an d improve the field at large. Secondly, this research attempted to identify effective predictive tactics for use by public relations practitioners. The study was designed such that both goals could be met. Trust Dimensions In order to test for independent variable effects, two-way ANOVAs were conducted. The independent variables tested in this study were Interactiv e Online Media Type (blog and stream ing video) and Crisis Type (using Coombs cr isis clusters: accident, intentional, victim). For the dependent variable, Hon and Grunigs (1999) definition of trust was used. Composed of several dimensions integrity, dependability, and competence their definitions of these terms (p. 19) are below followed by the questions used in the study to measure respondent reactions. The original questions were written using thi s organization instead of Vonnetek Automotive. Integrity: the belief that an organization is fair and just o Vonnetek Automotive treats people like me fairly and justly o Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me o Sound principles seem to guide Vonnetek Automotives behavior o Vonnetek Automotive does not mislead people like me Dependability: the belief that an orga nization will do what it says it will do o Vonnetek Automotive can be relie d on to keep its promises o I am very willing to let Vonnetek Au tomotive make decisions for people like me o I think it is important to watch Vonnetek Automotive closely so that it does not take advantage of people like me Competence: the belief that an organization has to ability to do what it says it will o I feel very confiden t about this Vonnetek Automotives skills o Vonnetek Automotive has the ability to accomplish what it says it will do

PAGE 97

97 It is important to discuss the roles that each vari able plays in predicting reports of trust. In other words, how can scholars and practitioners u tilize this information for the practice of public relations? Hon and Grunigs dimensions of trus t clearly identify how the dependent measures used in this research translate into public pe rceptions of organizational trustworthiness during crisis. Based on the consistency with which measur es of each trust dimension were reported either positively or negatively, it seems unlik ely that they contai n hidden meanings to respondents. In short, th is research works to sup ports the future use of such measures as reliable instruments of trust. The two-way ANOVAS indicated main effects fo r Crisis Type in several variables: Main effects were found for the following m easures of Integrity: Vonnetek Automotive treats people like me fairly and justly, Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me, Sound principles seem to guide Vonnetek Automotives behavior, and Vonnetek Automotive does not mislead people like me. Main effects for Dependability included Vonnetek Automotive can be relied on to keep its promises, I am very willing to let Vonne tek Automotive make decisions for people like me, and I think it is important to watch V onnetek Automotive closely so that it does not take advantage of people like me. Finally main effects were found for the following measures of Competence: I feel very confident about th is Vonnetek Automotives skills, and Vonnetek Automotive has the ability to acco mplish what it says it will do. The main effects represent respondents feeling significantly di fferent answering these questions based on the type of crisis the organization had undergone (acci dent, intentional, victim).

PAGE 98

98 In moving forward, a moment should be taken to evaluate the other variables. While no specific study or report has been conducted on the percentages of organiza tions using interactive online media, their organizational use seems to be very prevalent. Organizational blogs and streaming videos are often menti oned or featured in outlets that are not technology focused such as news broadcasts, news magazine articles and newspapers. This representation to the general public suggests that interactive online medi a use by organizations has become common knowledge. While it seems that blogs and streaming vide o are indeed widely used by organizations, their effects are unknown. Therefor e this experiment includes both blogs and streaming video as media to begin studying their effects during organi zational crisis. Does the method in which the crisis message is deliv ered matter and, if so, how does it matter? Is one be tter than another all of the time? In certain scenarios? In order to find answers to some of thes e questions, two-way ANOVAs were conducted to test for main effects of Media Type. No main effects were found based on this research. This finding would begin to answer one of this studys re search questions, analyzed in a later section. Again, Is one [media type] better than another [media type] all of the time? In certain scenarios? While testing for main effects already handled the former question, testing for interaction effects needed to be done for the la tter. Interaction effect s suggest that it is situational; similar to other situational theo ries like Contingency Theory, It Depends. Interaction effects indicate when variables affect each in one d irection in one case and another direction in another. Indeed interaction effects of the independent variables were found for one of the dependent measures. Interaction eff ects for Crisis Type and Media Type were found for I believe that Vonnetek Automotive takes th e opinions of people like me into account when

PAGE 99

99 making decisions (F=3.05, p=.05). This m easure was one of Hon and Grunigs (1999) dependability dimension questions, that [Vonnetek Automotive] will do what it says it will do. Hypotheses Now that the data analyses have been covered, a review of the hypotheses and research questions is in order. Data an alysis led to the following results: H1: All treatments will report lower trust scores after crisis than reports from pretest. While it was hypothesized that a crisis of a ny type would damage public trust of the organization, this hypothesis was not supported. It was expected that an organization undergoing any crisis, regardless of who was responsible fo r it, would report lower test scores than an organization not in crisis (as in the Pretest). In fact, Victim treatments reported higher mean scores than Pretes t treatments in all dependent measures with statistic al significance in 5 of 11 dependent variables: Three of these measures tested the Integrity dimension of the trust variable: Vonnetek Automotive treats people like me fairly and justly, Whenev er Vonnetek Automotive makes an important decision, I know it will be concerned about people like me, andSound principles seem to guide Vonnetek Automotives behavior. Two sta tistically significant va riables measured the Dependability dimension of the tr ust variable: Vonnetek Automotive can be relied on to keep its promises and I believe that Vonnetek Automotiv e takes the opinions of people like me into account when making decisions. In terestingly, none of the Competency measures were affected by the Crisis Type. This may indicate that Compet ency needs to have a historical basis, whereas Integrity and Dependability do not. For example, perhaps Vonnetek Automotives actions mentioned in, and including, the communication ga ve the respondents a feeling for if Vonnetek

PAGE 100

100 was acting with integrity and in a dependable way. It is possible that competence is based on past actions of the organization; in this case Vonnetek does not have any. In summary, this data indicates that a organi zation who has been a victim of a crisis is likely to have its publics find it more trustwort hy on scales of treating its publics fairly and justly, being concerned about people like the resp ondent, being reliable for keep its promises, considering opinions of its publics when making decisions, and being guided by sound principles, than before it ever being involved in a crisis at all. This is an intriguing finding, perhaps due to respondent sympathy for the organizati on as a victim of the crisis. In fact, that seems to be a very likely possibility. In this particular study, it is hard to imagine another reason for this result. However, this data also suggest s that competency is not rated higher in Victim treatments than Pretest. It is possible that that a respondent c ould potentially feel sympathy for the organization but not report higher organizational competence scores. While crisis attribution was mani pulated in this experiment, othe r types of crisis could also potentially influence the results. For example, w ould the results be consis tent using a crisis in which people were hurt or killed? Due to the drama tically different consequences of this type of crisis (compared with odometer tampering used in this study), it seems possi ble that participants could respond even more strongly For purposes of comparison, this study should be replicated using a re al organization. It would be especially interesting to test with multiple organizations with pretest scores ranging from very low to very high. It may be that an organization that is rated very low on trust because of its past may report same or lower ratings after a crisis because the publi cs thought they had it coming. Or perhaps an organization with extr emely high pretest ratings gets lower scores because the crisis highlights a previously unknown flaw in the organization. While it may be

PAGE 101

101 that all organizations will report higher trust scores when a victim than in pretest (as resulted in this study), that will remain unknown without furt her research. Furthermore, future research should consider concluding the data analysis by measuring questions such as On the following scale, please rate the organizations level of trust/integrity/dependability/competence. This would provide either data to s upport the reliability of the depende nt measures or potential flaws in design if the reports were not cons istent with the dependent measures. H2: Intentional cluster treatments will repor t lower trust scores than accident or victim, regardless of medium. It was hypothesized that, no matter what the medium, intentional treatments would report lower trust scores after crisis. This was based on the thought that once respondents knew that the organization had undergone crisis due to its ow n intentional actions, th ey would essentially punish the organization with lowe r trust scores. In other words, an organization that would be in crisis due to an intentional act would surely not be trusted. This hypothesis was fully supported. All depe ndent measures reported significantly lower mean scores in Intentional treatments than Accident treatments or Victim treatments. Therefore, the data suggest that the publics of a organization in an Intenti onal scenario would report the organization to not treat them fair ly or justly, not be concerned about people like them, not be reliable with its promises, not consider opinions of its publics when making decisions, not able to accomplish what it says it will, not be guided by sound principles, be misleading, need to be watched closely, not be successful at the thi ngs it tries to do, and th e publics will not be confident in the organizations sk ills or willing to let it make decisions for them, compared to that organization in either an Accident or Victim scenario.

PAGE 102

102 In short, for an organization to be trusted by its publics, one thing that it must prevent is intentionally causing a crisis. It should be noted, however that this study us ed an organization undergoing an intentional crisis due to acting in a very socially irresponsible manner. It is possible that an organization co uld intentional create a crisis but do so with admirable and socially responsible intentions, only to have them go wrong. Fo r example, an organization that decides not to chop down a forest for environmen tal issues but then has a crisis when its employees revolt because of lack of work. Th is concept leads back to Situational Crisis Communication Theory, which looks at both crisis attribution and relationship history. Would an organization with a long history of acting socially responsible be perceived similarly during the treatments in this experiment as an organiza tion with a tradition of ac ting irresponsibly? It seems unlikely but future research must be conducted to te st that hypothesis. Research Questions RQ1: Which interactive online medium, blog or streaming video, will report higher trust scores after organizational crisis treatments? This research question was trying to find out if either blog or streaming video were simply better in all cases. If so, practitioners could support focusing more of their efforts on the particular medium. Based on the two-way ANOVAS above, Media Type did not show any main effects: measures of dependent variables were no t significantly affected by use of Blog instead of Streaming Video, or vice versa. In other words, neither blogs nor streami ng video rose to the top as the consistently higher scor er in terms of trust scores. These results do not indicate that practitioners should stop using either blogs or streaming video. In fact, they simply s uggest that there may not be a di fference in public perception of organizational trust during crises.

PAGE 103

103 Future research pilot tests should incl ude questions asking respondents to rate trustworthiness of interactive online media types. This woul d help establish a benchmark upon which to measure the posttests. In fact, in clusion of many types of both traditional and interactive online media could be a valuable reference for many future studies. RQ2: Will use of either blog or streaming vi deo be better suited to maintain trust during particular types of crises? This research question again harkens back to the earlier questi on, Is one better than another all of the time? In certain scenarios ? Specifically, the second question is answered by RQ2. In order to find out, independent variable s Crisis Type and Media Type were tested for interaction effects. Once again, interaction effects indicate when variables affect each in one direction in one case and anot her direction in another. Interaction effects were found for 1 of 11 de pendent measures: I believe that Vonnetek Automotive takes the opinions of people like me into account when making decisions (p=.05). Streaming Video treatments report ed a significantly higher mean than Blog treatments in Accident (3.84 vs. 3.51, respectively) and Victim crisis types (4.26 vs. 4.11, respectively). However, Blog treatments (3.09) reported a significantly higher mean than Streaming Video treatments (2.47) in In tentional treatments. In summary, when a organization is in an A ccident crisis scenario, its use of Streaming video would likely report higher values for I believe that Vonnetek Automotive takes the opinions of people like me into account when ma king decisions, a Dependability dimension of trust. This may be due to the fact that the streaming video treatments were very similar in look to a newscast (they were, in fact, filmed in a ne ws studio). Respondents may be more used to

PAGE 104

104 watching newscast that they consider trustworthy report accidents in newscas ts. If so, there may be some projection of trust by the respondent from one source (news) to a nother (organization). If this experiment were conducted again, it s hould add another variable: source. Everything could be exactly replicated excep t to use two versions of each st reaming video treatment, one in which the speaker identifies himself/herself as a newscaster and another in which they identify themselves as a organizational spokesperson. It is possible that the newscaster treatment receives significantly higher trust scores than the exact same treatment introduced as an organizational spokesperson. The results alternatively indicate that if an organization is in the midst of an Intentional crisis scenario, based on this re search its use of a Blog would lik ely report higher values for I believe that Vonnetek Automotive takes the op inions of people like me into account when making decisions. This research strongly supports that an organiza tion in an Accident or Victim crisis should use Streaming Video, but use a Blog in Intentional crisis situations. In this case, it may be that the respondents are looking for some answers or justification from the organization and blogs may seem to be more intimate. Future research should again test blogs but against streaming video of a news conference, for exampl e, in which the same questions are answered. It could be that this too could fulfill the respondents potential desire for answers. This study was conducted using the faculty and st aff of a large southeastern university and members of a chamber of commerce fo r a city in the southeastern United States. In order to test whether this made a difference, testing was conduc ted to test for statisti cal significance between the university faculty and staff responses comp ared with the chamber of commerce members responses. The responses were not significan tly different.

PAGE 105

105 Implications Implications For Interactive Onlin e Media Research This research highlights a large void in new media effects research. While many, including scholars and practitioners are interested in using in using both mediums, the effects are still unknown. As stated above, this study did not find any main effects for blogs or streaming video treatments. Beyond blogs and streaming video, there are many other interactive online mediums that need to be studied both in terms of general understa nding and in terms of effects-specific research. Some of these in clude instant messaging, online social networking, virtual communities, as well as some that will sure ly be introduced in the near future. Further research is definitely needed to flesh out this area of study. As interactive online media are introduced to both organizations a nd their publics, it will be important to continue studying their effects. For example, interactive online media have reputations for embodying certain a ttributes: blogs as intimate, in stant messaging as quick, social networking sites as communities establishing thei r own distinct cultures,etc. While these descriptions may indeed be accurate, future research is necessary to prove such claims. Implications For Situational Crisis Communication Theory This study m akes several contributions to th e body of academic knowledge of Situational Crisis Communication Theory. Mo st important of these contribut ions is expanding the scope of the theory. Previous Situational Crisis Co mmunication Theory research has not explored application within the scope of interactive on line media the same me dia increasingly being used as focus for research and standard pr actice for many professiona ls. While furthering experimental research in the area of crisis public relati ons, it also tested variables that had not been experimentally measured previously. As mentioned above, Situational Crisis Communication Theory (Coombs, 2007), supposes that crisis manager examining the crisis

PAGE 106

106 situation in order to asses the level of the reputational threat of a crisis (p. 137), with this inherent level of reputational thr eat is expected to damage trust regardless of the crisis cluster, simply by its nature. In other words, trust is expected to be damaged (to some extent) simply because of the presence of the crisis. The findings of this research, specifically Victim treatments reporting higher values in 11 of 11 dependent variables suggest th at this is surely not the case. This research is the first to test aspects of Situation Crisis Communication Theory and media effects. Due to the fact that most organizational crises ar e going to have a media component, this is an area where much more expl oratory research must be conducted. It must be noted that the media suppliers ar e also steadily changing. It is not enough as an organization in crisis to simply understand that the traditional media now has interactive online components or that the organization can comm unicate via its own website. In creasingly, opinion leaders are emerging out of the Web 2.0. No longer does a jour nalist need to be hired by a large national newspaper to have his/her stories read, with the proliferation of We b 2.0 anyone with some knowledge of any subject, access to the Internet, and the time it takes to type a blog can become an opinion leader know worldwide and many are! Understanding all of these different types of media is critical during organizational crisis, though it is also becoming necessary to simply operate from day-to-day successfully. Situatio nal Crisis Communication Theory has provided a sufficient foundation on which to build these bran ches of crisis public relations research. Based on the results of the manipulation checks pilot test and this study itself, Coombs and Holladays (2002) Situational Crisis Communication Theory crisis clusters seem to be very strong in their design. Respondent s not only identified cr isis types distinctly but also reported statistically significan t reactions to them.

PAGE 107

107 This experiment also acknowledges that Si tuational Crisis Communication Theorys organizational history component may be very important for understanding how an organization can best maintain trust during a cr isis. It seems likely, as Situational Crisis Communication Theory suggest, that an organization with a re putation for acting responsibly would receive higher trust scores than an irrespon sible organization. Furthermore, this research begins to make a case to consider integrati ng the trust dimension into Situational Crisis Communication Theory as well as acknowledge media effects with in its construct. While Situational Crisis Communication Theory is alr eady built in a way that can accommodate these variables, they are important enough to the scope of the theory itself that they should be implemented as fully functioning components. While crisis public relations theory is still too young to be generally evaluated, the findings of this research build upon the fields understanding of each of the variables. This study supports researchers moving forward using Coom bs and Holladays clusters with increased confidence. Furthermore, simply by isolating th e clusters outside of the Situational Crisis Communication Theory framework, confirming thei r own strength independently, and refolding them back into Situational Crisis Communicati on Theory, it strongly contributes to building upon Situational Crisis Communication Theory and by extension the sub-discipline of crisis public relations as a theoretical whole. Fina lly, the power of the clusters retrospectively strengthens past research using them as well. Finally, these effects on Situational Crisis Communication Theory should have a domino effect for Crisis Response Theory. This resear ch illustrates that additional variables may be needed to effectively manage crises, whethe r using either theory. Situational Crisis Communication Theory attempts to predict resu lts of crisis communication. Crisis Response

PAGE 108

108 Theory focuses on the responses themselves. When combined, they are complementary to one another by first identifying how best to communi cate in crisis, followed by confirming that the chosen response should have its desired effect. Implications For Relationship Theory The results of this research further Relatio nship Theory by experi mentally applying the trust dimension measures to crisis scenarios. Clea rly trust is a critically important factor before, during and after organizational cris is and this research furthers our understanding of how trust is affected throughout this process. Organizations exist, at least in large part, to achieve success. This success if often measured by longevity and abili ty to make money. Trust is critical to both of these goals. While money can be made in the short term, even without trust, an organization performing in such fashion would not achieve long evity. Therefore, in order for an organization to be successful, it must be ab le to cultivate and maintain trus t with its publics publics that it strives to create and keep rela tionships with. It follows that without such relationships, an organization would not make money nor be able to survive for any dece nt length of time. Hon and Grunig (1999) introduced the PR Relationship Measurement Scale which included guidelines for measuring relationships, an obvious component to Relationship Theory. They identified trust, control mutuality, sa tisfaction, commitment, exchange relationship, communal relationship. In short, symmetrical relationships are those with the highest likelihood of maintaining the longevity mentioned above an d these six elements seem to be present: The most productive relationships in the long ru n are those that benef it both parties in the relationship rather than those designed to be nefit the organization on ly. Public relations theorists have termed these types of re lationships symmetrical and asymmetrical, respectively. (p. 19) An effective way to test components of Relati onship Theory is to take them out of their framework and test them independently of the ov erall theory. While this has been done with

PAGE 109

109 trust variables in the past, this is the first instan ce in which the trust dimensions were tested using crisis scenarios. Just as a crisis can bring abou t the best and worst in people when it arises, so too can crisis testing bring out the best and worst in a theory. It can disprove a theory, at least disprove it as being applicable du ring a crisis scenari o. Or on the other hand, it can still remain a reliable theoretical found ation even when tested in crisis s cenarios. In this case, the trust variables were tested in just such a way. Trust is clearly important in any relationship but it can sometimes be the only thing holding an organizat ional -public relationshi p together when the chips fall. For each of the dependent measures all of them trust dimensions from Hon and Grunigs research, trust measures seemed to be str ong. This was due in large part to the fact that there were no instances in which trust scores were way out of sync with the majority. Again, by testing the trust component of Rela tionship Theory outside of its normal framework, finding it effective, and folding it ba ck into Relationship Theory should strengthen the overarching theory in general. As the sayi ng goes, a chain is only as strong as its weakest link. Therefore, each of the components should also be tested more inde pendently. In fact, a replication of this study using a re al organization would allow for a ll elements above to be tested and evaluated for strengths and weaknesses. This is especially true for trust, which could then be compared with the findings in this study. Due to the explorative nature of this rese arch (based on the unique combination of variables), it is difficult to make solid predicti ons for future studies using the trust scale. However, this research does succeed at servi ng as a roadmap for future research by identifying new areas for such study. The literature reviewed in this study identi fied Excellence Theory as a precursor to Relationship Theory. The experiment conducted also seems to support the Excellence teams

PAGE 110

110 two-way models in terms of its critical importa nce to cultivating and maintaining trust. However, future research should consider as king ask respondents which of the four public relations models press agen try, public information, two-wa y asymmetrical, or two-way symmetrical they actual feel each media type is (Grunig, et al., 1992). It could be, for example, that blog would be reported as two-way symmet rical and streaming video as public information but we cannot assume anything in this regard without testing this in futu re research. For the same reason, future research using media type as a variable should ask if respondents feel they are traditional or interactive to further our understanding of interactive online media perceptions. This is especially true with populations that are heavily personally invested in technology. Technology changes ex tremely quickly. With it, so too do the mediums used and the perceptions of them by their users. For example, there are likely publics of the same medium that may think a simple website is interactive and others who insist that it is just another version of traditional media. Understanding this on a more de tailed level in future research could strongly enhance any potential findings. Finally, this research give s further support to the Relationship-Theory-as-eventualparadigm camp. The variables simply fit too well into public relations research to dismiss this possibility. In fact, it would be very difficult to think of a public relations research study in which it could not be logically applied. Implications For Public Relations Practice This rese arch highlights important implications for public relations practitioners. The first is that practitioners should consider exactly what they really know about interactive online media effects before spending potentially vast amounts of money on them. In some cases, the answer is probably not that much. It is well known that many agency practitioners and in-house organizational practitioners (as well as consultants and the like) believe that it is very important

PAGE 111

111 to have an interactive online presence, but to wh at effect? Budgets are not unlimited, especially in todays economy. Public rela tions practitioners have long dea lt with how to show their worth within an organization (Grunig, et al., 1992). Spending money with no real understanding of what the return-on-investment is w ill not help that cause. This is certainly not to say that public relations practitioners should not pur sue such media, it must just be a priority to practitioners to be able to evaluate the value of each endeavor. For example, this research suggests that an organization in a Victim situation should make sure that it is communicati ng that it is a victim to its publics. If the results of this research do indeed predict practic al application, it could essentially build trust instead of losing it. That is a measurable objectiv e for any practitioner to take to his or her management. The second contribution to public relations practice is furt her building the knowledge of how to practically manage public relations cris es. This is a critical function within any organization and C-level executives are increasing ly calling for crisis planning and management from their staff. Studies like this one can immediately translate into resources for such practitioner/ Beyond simply applying these findings to publ ic relations planning and implementation, organizations like Edelman Public Relations, the university involved and the chamber of commerce involved have already ex pressed that they find this st udy to be extremely valuable. Each of them recognizes that this research will help to focus and streamline their crisis planning and management. Furthermore, this knowledge ca n be applied to any organization, or for that matter many organizations. And practical unders tanding of how an organizational practitioner should communicate during a crisis does not need to be asymmetrical communication. By

PAGE 112

112 maintaining trust throughout such crises, both organizations and th eir publics can benefit through openness, honesty, and clear lines of communication. Implications for Using Experimental Methodology and Interactive Online Media The m ethodology used in this study illustrated some strengths and weaknesses of attempting to use experimental method to meas ure interactive online me dia. First, while methodology should always be handled with care, special consideration should be taken when using interactive online media to test interactive online media. Future research should consider using other methods in tandem with interactive online media to ensure that this issue does not confound the variables. Second, using Survey Monkey as the collection tool also provided some strengths and weaknesses. While the service does an adequate job of collecting the act ual data and exporting it for use in statistical software, it is also not expressly designed for experimental research. Because of this, Survey Monkey had to be linked to from within the Vonnetek website. This was effective but it did not allow for visual consiste ncy. The only way to achieve that would have been to design Vonnetek.com to match the Surv ey Monkey website, a design that may not suit many studies. A better data collection website design would implement th e actual design of all online components in order to create such consistency. Finally, likely the strongest aspe ct of Survey Monkey is its user-friendliness and ability to store massive amounts of data in an easily manageab le format. It also allows for exporting data into a variety of formats, which can save extra steps when preparing to analyze data. Implications Summary The m ost important implication, and overall contribution to the body of knowledge, that this study makes is to begin to connect the logical variables of interactive online media, crisis type and trust. Perhaps most important in this regard is that it promotes the value of theory to

PAGE 113

113 practitioners. This research is the first to ex plore and attempt to create a roadmap for future studies analyzing how these three variables work toge ther. This area of research is one that both scholars and practitioners agree is important and thus should help bridge the often present rift between academia and practice. The fact that th is theoretical study was fully funded by Edelman Public Relations further supports this claim. This study applied aspects of both Excellen ce Theory and Contingency Theory in its design. By the organization communicating its cr isis scenario honestly, even in the case of Intentional treatments, Excellence Theorys con cept of acting in a socially responsible manner was (at least partially) applied. It must be noted that this resp onsibility refers only to the aftereffects in Intentional treatments due to the fact that an intentional cris is should not occur when acting responsibly. Contingency Theory was also partially applied in this study because Vonnetek Automotive used the inte ractive online media types to co mmunicate the crisis to its publics. This essentially bega n a negotiation between the orga nization and its publics for the destination of its relationships on the accommodation-advocacy continuum. The data from this research additionally ra ises questions about Seltzer and Mitrooks (2007) suggestion that blogs coul d be better suited for online relationship building than its traditional counterparts. As mentioned above future research must measure respondent perceptions of the media type. If, indeed, streaming video is perceived as a non-traditional medium then perhaps blogs are not better suited. Finally, this research brings up the question, Can the trust dimension stand on its own? Based on the results of this study, the answer seems to be yes. The only way to find out for sure, however, will be to replicate th is study with additional measures perhaps other Relationship Theory dimensions and test if the same results are reported.

PAGE 114

114 This study enhances the valu e of the theories that it builds upon. Shoemaker (2003) established guidelines to test the value of theory. While this study does not create theory nor is it focused on one specific theory, Shoemakers guide lines are still helpful to gauge this studys contribution to the theories that it uses: Testability This is achieved when a theory is able to be tested to veri fy its claims. This study could easily be replicated to prove its testability. This re search also introduced new ways of testing the theories involved. By utilizing experimental design, th is study begins to push crisis public relations research beyond the traditional case st udies offered previously. It also is the first study of its kind to combine testing of online interactive media, cris is, and trust variables in order to understand their connections. Falsifiability. A good theory must be able to be falsified. Based on the design of this study, it would be fairly simple to achieve falsifiability in future research, a strength of theory per Shoemakers guidelines. This could be achieved by creating new scenarios in which the current findings would be violated. Parsimony. This does not refer to being simple, pe r se. Instead, parsimony is achieved while only being as complex as necessary in order to achieve the desired results. While including 11 dependent variables, they were closely related an d only two independent variables were tested making this study only as complex as needed. The connections this research makes between variables is also quite simple to unde rstand, thereby further s upporting its parsimonious nature. Explanatory power. In essence, the more that a theo ry can explain the better. The findings from this study enhance th e explanatory power of the theori es that it uses. It achieves

PAGE 115

115 this first by testing them in new ways and s econdly by proving that the theories do indeed provide some explanation in those scenarios. Predictive power. Can the theory or theories being test provide prediction for what might happen in a similar scenario in the future? This is of special importance to practitioners. One of the biggest strengths of using expe rimental method is its predictive nature. Furthermore, the use of real world participants instead of student s, for example increase s the accuracy of this predictive power by enhancing th e studys external validity. Scope. How applicable is the theory to a large range of use? In some instances, scope can come at the expense of predictive power. The scope of this research was narrowed by use of only two interactive online media types and use of an organization. Th is design was used to increase the predictive power for those selections. However, it is likely th at other organizations can apply the findings of this resear ch to their communication efforts. Heuristic value. Does the theory stimulate further in terest in knowledge? This study does increase the heuristic value of the theories whic h it uses by applying its va riables to applications of interest to both scholars and practitioners. Furthermore, it offers several suggestions for future research which also enhance the heuristics. Limitations One of the most interesting findings within this research is the fact that Victim treatment means were significantly higher than the Pretests. The reasons for this could be two-fold. First, it may be that the respondents felt sympathy for Vonnetek Automotiv e as a victim. However, it may be a result of Vonnetek Automotive being a fictional organization. Ki and Hon (2007) found that stakeholder publics with stronger evaluations of the re lationship with an organization reported more positive attitudes toward it. Wit hout a previous relationshi p with the organization, the respondents may have felt challenged in answer ing some Pretest questions resulting in lower

PAGE 116

116 Pretest scores or Pretest scores near the medi an. While the advantages of using a fictional organization included having full control over the website and its content and ease of manipulation checks, the limitations above must be considered. Another limitation was the method of admini stration. While the emails containing the surveys were simultaneously sent for each orga nization, both organizations should have been administered the surveys simultaneously for all pa rticipants. The respons es from the university participants were not significantl y different than those of the ch amber members but if they were it would have been difficult to te ll whether it was due to their or ganizational membership and/or when they were administered the survey. Also while organizational members were emailed the surveys simultaneously, there was no practical way of achieving simultaneous administration of the actual instruments. This also likely had an effect on the low 5% response rate. While it was expected that the response rate would be lower due to the su rvey administration during the summer break at the participating university (many faculty and staff on vacation), using the email method of administration also typically has a lower response rate than some other methods. Due to the time and budgetary constraints, however, it was deemed the most appropriate method for this study. Additionally, future research should consider inclusion of traditional media types in order to act as control. Th is would both create opportunitie s for comparison as well as potentially further knowledge of thos e traditional media. The lack of a true control group in this design makes it difficult to confirm that the effects are due specifi cally to the interactive online media types and not simply media. There may be a limitation within the explana tion of the interacti on effects found for I believe that Vonnetek Automotive takes the opin ions of people like me into account when

PAGE 117

117 making decisions. In other words, does an or ganization with high scor es for this variable correlate to it being trustworthy? For example, some people feel that tobacco companies take opinions of people into account when making deci sions but then use them in an untrustworthy fashion. While considered part of the trust matrix, further testing should be conducted to correlated higher scores with that measure and hi gher levels of trust. Again, could respondents feel that an organization take s their opinions into account bu t they still dont trust the organization? Could an organi zation take a respondents opin ion into account only to use the information in an untrustworthy manner? Furthe r testing must be conduc ted to explicate this information. It should, at minimum, be manipulation checked in future research. It seems unlikely that this study is the one to totally disprove McLuhans (1964) The medium is the message, though the results show very little in the way of significance between the media types used in this experiment. More preferable would be further test these findings by replication. This would be simple enough by repli cating the study with more types of interactive online media. As mentioned above, many people seem to believe that blogs are more intimate than other forms on interactive online media. This research introduced the question of whether respondents might have felt that the Blog treatm ents were more intimate (and less well-known) and Streaming Video treatments were widely known because of its similarity to broadcast television. In order to test, this study should be replicated using all streaming video treatments in which one treatment has the speaker state that th ey are an independent reporter and another in which the speaker states they ar e an organizational spokesperson. This study could be replicated surveying the faculty and staff of another university to further the external validity of its findings. It would also be interesting to have the workforce of

PAGE 118

118 a large organization(s) as the sample for comparison. While no significa nt differences were found between the participants working at the university a nd the business members of the chamber of commerce, future research could focus more specifically on how more detailed demographics might affect the data. Finally, it would be very intere sting to create an experiment from a combination of this design and Coombs (2000) research findings that relationships are damaged during crisis by relational expectations and history between an organization and its publics. It would likely require an existing and known organi zation in order for the publics to have a relationship history with it. Would such a relationship still be damaged using the Victim treatments from this study?

PAGE 119

119 APPENDIX A SURVEY MEASURES Trust: 1. Vonnetek Automotive treats people like me fairly and justly. 2. Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me. 3. Vonnetek Automotive can be relied on to keep its promises. 4. I believe that Vonnetek Automotive take s the opinions of people like me into account when making decisions. 5. I feel very confident about Vonnetek Automotives skills. 6. Vonnetek Automotive has the ability to accomplish what it says it will do. 7. Sound principles seem to guide Vonnetek Automotives behavior. 8. Vonnetek Automotive does not mislead people like me. 9. I am very willing to let Vonnetek Automotive make decisions for people like me. 10. I think it is important to watch Vonnete k Automotive closely so that it does not take advantage of people like me. 11. Vonnetek Automotive is known to be succe ssful at the things it tries to do.

PAGE 120

120 APPENDIX B PILOT STUDY Welcom e, Thank you for participating in this study. This research is designed to test your perception of a company and your analysis of a short articl e. Your answers will be used to help further research on public perception. Today you will be asked to complete a surve y. Please carefully follow the directions for each section. All responses are anonym ous and kept strictly confidential. Once again, thank you for your cooperation. Please continue to the next page.

PAGE 121

121 SECTION Ia. Please carefully read the following article. Once finish ed, you will be asked several questions. Please select the answer that best matches how you feel about each statement. ARTICLE ONE (Accident) In May 2006, odometers in cars manufactured by Vonnetek Automotive Corporation were reported to be faulty. By late 2007, the Vonne tek Automotive Corporatio n had discovered that some odometers had malfunctioned due to a ra re electrical issue. Vonnetek Automotive executives were unaware of the malfunction whil e the cars were on the assembly line because the electrical impulse causing the problem did not affect any other components. Some of the affected cars had been taken from the production line to be tested, a common practice in the automobile industry. Because it is illegal to disconnect the odometers on the test cars, they arrive at the dealerships with test miles on th em. The test drive is an important check for quality control. These tests invo lve driving the cars many hard miles. The cars are driven on a test track located at the manu facturing facility. The people who bought the test cars from Vonnetek Automotive did not know the cars had b een test driven. In fact, the customers believed the cars had virtually no miles on them. Vonnetek Automotive ex ecutives investigated and diagnosed the technical failure The facts of the case were revealed at a news conference held by Vonnetek Automotive Corporation following its investigation. Affected customers have been contacted by Vonnetek Automotive in order to correct the situati on. Vonnetek Automotive is headquartered in Toledo, Oh io and specializes in manufact uring of exotic and replica automobiles. Please continue to the next page.

PAGE 122

122 SECTION IIa. The following items measure your feelings toward Vonnetek Automotive. For each item, you will be asked whether you agree or disagree with the statement. Please select the choice that best matches how you feel about the statement. Indicate agreement or disagr eement with each statement by selec ting a number between 1 and 7, where 1 means I strongly disagree and 7 means I strongly agree. Please look at the Vonn etek Automotive website. 1. Which of the following are victims in this situation? (Check all that apply) Vonnetek Dealerships Customers Automotive industry 2. Who do you feel is at fault in this crisis? 3. Do you think that Vonnetek Automotiv e is responsible for this crisis? Yes No

PAGE 123

123 SECTION Ib. Please carefully read the following article. Once finish ed, you will be asked several questions. Please select the answer that best matches how you feel about each statement. ARTICLE TWO (Intentional) In May 2006, odometers in cars manufactured by Vonnetek Automotive Corporation were reported to be faulty. By late 2007, the Vonnetek Automotive Corporation was found guilty of odometer tampering. Two Vonnetek Automotive executives had ordered the odometers to be disconnected on the cars tested from the assembly line. It is common prac tice in the automobile industry to take cars from the produ ction line and to test them. Ho wever, it is illegal to tamper with the odometers on the test cars. The test dr ive is an important check for quality control. These tests involve driving the cars many hard miles. The cars are driven on a test track located at the manufacturing facility. The people who bought th e test cars from Vonnetek Automotive did not know the cars had been test driven. In fact, th e customers believed the cars had virtually no miles on them. Vonnetek Au tomotive executives knowingly committed fraud. The facts of the case were revealed at a news conference held by Vonnetek Automotive following the guilty verdict. Affected custom ers have been contacted by Vonnetek Automotive in order to correct the situation. Vonnetek Automotive is headquartered in Toledo, Ohio and specializes in manufacturing of exotic and replica automobiles. Please continue to the next page.

PAGE 124

124 SECTION IIb. The following items measure your feelings toward Vonnetek Automotive. For each item, you will be asked whether you agree or disagree with the statement. Please select the choice that best matches how you feel about the statement. Indicate agreement or disagr eement with each statement by selec ting a number between 1 and 7, where 1 means I strongly disagree and 7 means I strongly agree. Please look at the Vonn etek Automotive website. 1. Which of the following are victims in this situation? (Check all that apply) Vonnetek Dealerships Customers Automotive industry 2. Who do you feel is at fault in this crisis? 3. Do you think that Vonnetek Automotiv e is responsible for this crisis? Yes No

PAGE 125

125 SECTION Ic. Please carefully read the following article. Once finish ed, you will be asked several questions. Please select the answer that best matches how you feel about each statement. ARTICLE THREE (Victim) In May 2006, odometers in cars manufactured by Vonnetek Automotive Corporation were reported to be faulty. By late 2007, an i ndependent investigation found that Vonnetek Automotive Corporation had been a victim of odometer tampering. Vonnetek Automotive executives discovered that some dealerships sell ing their cars had tampered with the odometers in order to charge higher prices. The dealerships invol ved are not owned by Vonnetek Automotive. It is common pract ice in the automobile industry to take cars from the production line and to test them. However, it is illegal to tamp er with the odometers on the test cars. The test drive is an important check for quality control. These tests invol ve driving the cars many hard miles. The cars are driven on a test track loca ted at the manufacturing facility. Because of the testing, Vonnetek Automotive execu tives insist that all dealersh ips sell them at a discount. The people who bought the test cars from the dealership s did not know the cars had been test driven. In fact, the customers believed the cars had virtually no miles on them. Vonnetek Automotive executives were unaware that the dealerships had committed fraud. The facts of the case were revealed at a news conference held by Vonnetek Automotive Corporation following the investigation. Affected customers have been contacted by Vonnetek Automotive in order to correct the situation. Vo nnetek Automotive is headquartered in Toledo, Ohio and specializes in manufacturing of exotic and replica automobiles. Please continue to the next page.

PAGE 126

126 SECTION IIc. The following items measure your feelings toward Vonnetek Automotive. For each item, you will be asked whether you agree or disagree with the statement. Please select the choice that best matches how you feel about the statement. Indicate agreement or disagr eement with each statement by selec ting a number between 1 and 7, where 1 means I strongly disagree and 7 means I strongly agree. Please look at the Vonn etek Automotive website. 1. Which of the following are victims in this situation? (Check all that apply) Vonnetek Dealerships Customers Automotive industry 2. Who do you feel is at fault in this crisis? 3. Do you think that Vonnetek Automotiv e is responsible for this crisis? Yes No

PAGE 127

127 APPENDIX C SURVEY INSTRUMENT 1 Attention UF Faculty and Staff In its ongoing efforts to provide a safe environment for students, faculty and staff, the university is working to im prove its response to di sasters. In addition to our internal planning efforts, we are attempting to implement best prac tices from other institutions and support studies that may provide us new insights to disaster mana gement. One such effort is a doctoral research study by Seth Oyer, a student in UFs College of Journalism & Communications. A brief description of Seths project a nd the need for volunteers to view a web site and provide input is outlined. There are two parts to Seths survey. Seth has piloted his surveys and indicates that it will take 7 to 8 minutes (total for both) to view and respond to his information. The second part of this survey will be sent out in a few days. I encourage you to consider participating in this study. William Properzio, Ph.D. Director, UF Environmental Health and Safety Project Title: Effects of Media on Trust

PAGE 128

128 Thank you for considering participation in this study. This research is designed to test media effects during certain scenarios. Your answers will be used to help researchers better understand particular media effects. This survey is two-part. First you will be as ked to look at a website and complete a short survey (Part One). Within one week, you will be contacted via email (like you were today) to complete Part Two of the survey. Please carefully review the organizational we bsite you will be direct ed to with the link below. Explore each tab. Once you are finished please click the News tab, then the Survey link and please complete the short survey. All responses are anonymous and kept strictly confidential. Once again, thank you for your cooperation. Please click on this link to begin: www.vonnetek.com/1/ Sincerely, Seth Oyer soyer@ufl.edu Prim ary Researcher & Doctoral Candidate University of Florida College of Journalism & Mass Communications Department of Public Relations Please continue to the next page

PAGE 129

129 SECTION I. The following items measure your feelings toward Vonnetek Automotive. For each item, you will be asked whether you agree or disagree with the statement. Please select the choice that best matches how you feel about the statement. Indicate agreement or disagreement with each statement by selecting a number between 1 and 7, where 1 means I strongly disagree and 7 means I strongly agree. STRONGLY DISAGREE DISAGREE SOMEWHAT DISAGREE NEITHER AGREE NOR DISAGREE SOMEWHAT AGREE AGREE STRONGLY AGREE Vonnetek Automotive treats people like me fairly and justly. 1 2 3 4 5 6 7 Whenever Vonnetek Automo tive makes an important decision, I know it will be concerned about people like me. 1 2 3 4 5 6 7 Vonnetek Automotive can be relied on to keep its promises. 1 2 3 4 5 6 7 I believe that Vonnetek Automotive takes the opinions of people like me into account when making decisions. 1 2 3 4 5 6 7 I feel very confident about this Vonnetek Automotives skills. 1 2 3 4 5 6 7 Vonnetek Automotive has the ability to accomplish what it says it will do. 1 2 3 4 5 6 7 Sound principles seem to guide Vonnetek Automotives behavior. 1 2 3 4 5 6 7 Vonnetek Automotive does not mislead people like me. 1 2 3 4 5 6 7 I am very willing to let Vonnetek Automotive make decisions for people like me. 1 2 3 4 5 6 7 I think it is important to watch Vonnetek Automotive closely so that it does not take advantage of people like me. 1 2 3 4 5 6 7 Vonnetek Automotive is known to be successful at the things it tries to do. 1 2 3 4 5 6 7 I am happy with Vonnetek Automotive. 1 2 3 4 5 6 7 Both Vonnetek Automotive and people like me benefit from the relationship. 1 2 3 4 5 6 7 Most people like me are happy in their interactions with Vonnetek Automotive. 1 2 3 4 5 6 7 Generally speaking, I am pleased with the relationship Vonnetek Automotive has established with people like me. 1 2 3 4 5 6 7 Most people enjoy dealing with Vonnetek Automotive. 1 2 3 4 5 6 7 Vonnetek Automotive fails to satisfy the needs of people like me. 1 2 3 4 5 6 7 I feel people like me are important to Vonnetek Automotive. 1 2 3 4 5 6 7

PAGE 130

130 This concludes the survey. Thank you for your time and your participation. If you have any questions regarding this st udy, please feel free to contact the primary researcher. Seth Oyer Doctoral Candidate University of Florida College of Journalism & Communication Department of Public Relations soyer@ufl.edu In general, I believe that nothing of value has been accomplished between V onnetek Automotive and people like me. 1 2 3 4 5 6 7 Vonnetek Automotive does not especially enjoy giving others aid. 1 2 3 4 5 6 7 Vonnetek Automotive is very concerned about the welfare of people like me. 1 2 3 4 5 6 7 I feel that Vonnetek Automotive takes advantage of people who are vulnerable. 1 2 3 4 5 6 7 I think that Vonnetek Automotive succeeds by stepping on other people. 1 2 3 4 5 6 7 Vonnetek Automotive helps people like me without expecting anything in return. 1 2 3 4 5 6 7 I dont consider Vonnetek Automotive to be particularly helpful. 1 2 3 4 5 6 7 I feel that Vonnetek Automotive tries to get the upper hand. 1 2 3 4 5 6 7

PAGE 131

131 APPENDIX D SURVEY INSTRUMENT 2 Attention UF Faculty and Staff In its ongoing efforts to provide a safe environment for students, faculty and staff, the university is working to im prove its response to di sasters. In addition to our internal planning efforts, we are attempting to implement best prac tices from other institutions and support studies that may provide us new insights to disaster mana gement. One such effort is a doctoral research study by Seth Oyer, a student in UFs College of Journalism & Communications. A brief description of Seths project a nd the need for volunteers to view a web site and provide input is outlined. There are two parts to Seths survey. Seth has piloted his surveys and indicates that it will take 7 to 8 minutes (total for both) to view and respond to his information. The second part of this survey will be sent out in a few days. I encourage you to consider participating in this study. William Properzio, Ph.D. Director, UF Environmental Health and Safety Project Title: Effects of Media on Trust

PAGE 132

132 Thank you once again for participating in this study. This research is designed to test media effects during certain scenarios. Your answers will be used to help researchers better understand particular media effects. This is the second and final part of the survey you began last week. Today you will again be asked to look at a website and news st ory followed by a short survey (Part Two). Please carefully navigate to the News tab of the organizational website you will be directed to with the link below. Pleas e look at the news story on the webpage. Once you are finished please click the Survey link below the news stor y and please complete th e short survey. All responses are anonymous and ke pt strictly confidential. Once again, thank you for your cooperation. Please click on this link to begin: www.vonnetek.com Sincerely, Seth Oyer soyer@ufl.edu Prim ary Researcher & Doctoral Student University of Florida College of Journalism & Mass Communications Department of Public Relations

PAGE 133

133 SECTION I. The following items measure your feelings toward Vonnetek Automotive. For each item, you will be asked whether you agree or disagree with the statement. Please select the choice that best matches how you feel about the statement. Indicate agreement or disagreement with each statement by selecting a number between 1 and 7, where 1 means I strongly disagree and 7 means I strongly agree. STRONGLY DISAGREE DISAGREE SOMEWHAT DISAGREE NEITHER AGREE NOR DISAGREE SOMEWHAT AGREE AGREE STRONGLY AGREE Vonnetek Automotive treats people like me fairly and justly. 1 2 3 4 5 6 7 Whenever Vonnetek Automo tive makes an important decision, I know it will be concerned about people like me. 1 2 3 4 5 6 7 Vonnetek Automotive can be relied on to keep its promises. 1 2 3 4 5 6 7 I believe that Vonnetek Automotive takes the opinions of people like me into account when making decisions. 1 2 3 4 5 6 7 I feel very confident about this Vonnetek Automotives skills. 1 2 3 4 5 6 7 Vonnetek Automotive has the ability to accomplish what it says it will do. 1 2 3 4 5 6 7 Sound principles seem to guide Vonnetek Automotives behavior. 1 2 3 4 5 6 7 Vonnetek Automotive does not mislead people like me. 1 2 3 4 5 6 7 I am very willing to let Vonnetek Automotive make decisions for people like me. 1 2 3 4 5 6 7 I think it is important to watch Vonnetek Automotive closely so that it does not take advantage of people like me. 1 2 3 4 5 6 7 Vonnetek Automotive is known to be successful at the things it tries to do. 1 2 3 4 5 6 7 I am happy with Vonnetek Automotive. 1 2 3 4 5 6 7 Both Vonnetek Automotive and people like me benefit from the relationship. 1 2 3 4 5 6 7 Most people like me are happy in their interactions with Vonnetek Automotive. 1 2 3 4 5 6 7 Generally speaking, I am pleased with the relationship Vonnetek Automotive has established with people like me. 1 2 3 4 5 6 7 Most people enjoy dealing with Vonnetek Automotive. 1 2 3 4 5 6 7 Vonnetek Automotive fails to satisfy the needs of people like me. 1 2 3 4 5 6 7 I feel people like me are important to Vonnetek Automotive. 1 2 3 4 5 6 7

PAGE 134

134 Please continue to the next page. In general, I believe that nothing of value has been accomplished between V onnetek Automotive and people like me. 1 2 3 4 5 6 7 Vonnetek Automotive does not especially enjoy giving others aid. 1 2 3 4 5 6 7 Vonnetek Automotive is very concerned about the welfare of people like me. 1 2 3 4 5 6 7 I feel that Vonnetek Automotive takes advantage of people who are vulnerable. 1 2 3 4 5 6 7 I think that Vonnetek Automotive succeeds by stepping on other people. 1 2 3 4 5 6 7 Vonnetek Automotive helps people like me without expecting anything in return. 1 2 3 4 5 6 7 I dont consider Vonnetek Automotive to be particularly helpful. 1 2 3 4 5 6 7 I feel that Vonnetek Automotive tries to get the upper hand. 1 2 3 4 5 6 7

PAGE 135

135 SECTION II. The following questions ask for some genera l demographic information. Please select the appropriate response or fill in the blank as needed. Sex ( ) Female ( ) Male Age __________ What is your position at your organization? __________ How many years have you work ed in that position? __________ Which of the following best descri bes your organization (select one)? ( ) Corporation ( ) Not-for-profit ( ) Government ( ) Education ( ) Consultant ( ) Other In which state do YOU conduct most of your business __________ How many employees currently wo rk for your organization? __________ How many years in total have you worked at your organization (please choose one)? ( ) Less than one year ( ) 1-5 years ( ) 6-10 years ( ) More than 10 years Does your company have a website? ( ) Yes ( ) No Please estimate the number of hour s you spend in a typical work week involved in business interactions via your website (please round to the nearest hour): __________ Please continue to the next page.

PAGE 136

136 This concludes the survey. Thank you for your time and your participation. If you have any questions regarding this st udy, please feel free to contact the primary researcher: Seth Oyer Doctoral Candidate University of Florida College of Journalism & Communication Department of Public Relations soyer@ufl.edu

PAGE 137

137 APPENDIX E INFORMED CONSENT Protocol T itle: Measurement of Media Effects Please read this consent document carefully before you decide to participate in this study. Please feel free to print this do cument for your research. Purpose of the research study: To measure the effects of me dia in certain scenarios. If you choose to participate in the study, you will be asked to: 1) Look at an organizational website and answer a short online survey about the organization. 2) Within ten days, you will be asked to an swer another short online survey about the organization. Time required: 5-7 minutes (total) Risks and benefits: There are no risks or benefits associated with participating in this st udy. You will be able to contribute to the development of social science and public relations research. Compensation: You will not receive compensation for participation in this study.

PAGE 138

138 Confidentiality: Your identity will be kept confidential to the extent provided by law. Your name will not be attached in any way to the results of the st udy. No one will be able to identify you as a respondent. Voluntary participation: Your participation in this study is completely voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to withdraw from th e study at any time w ithout consequence. _____________________________ ___________ Participant Signature Date _Seth Oyer___________________ _5/9/08____ P.I. Signature Date Principal Investigator Seth Oyer, 352-256-0910, soyer@uufl.edu Supervisor Dr. Michael Mitr ook, 352-392-8730, mmitrook@jou.ufl.edu For questions about your rights as a research participant, contact th e IRB at 352-392-0433 or irb2@ufl.edu.

PAGE 139

139 APPENDIX F MAIN EFFECTS ONE-WAY ANOVAS Table F-1 ANOVA (DV: TREATSFAI RLY, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 189.36 3 63.12 55.67* .00* Within Groups 706.43 6231.13 Total 895.79 626 *p<.05 Vonnetek Automotive treats people like me fairly and justly.

PAGE 140

140 Table F-2 ANOVA (DV: CONCERNED ABOUT, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 126.21 3 42.07 27.57* .00* Within Groups 950.52 6231.53 Total 1076.73 626 *p<.05 Whenever Vonnetek Automotive makes an important decision, I know it will be concerned about people like me.

PAGE 141

141 Table F-3 ANOVA (DV: KEEPPRO MISES, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 234.65 3 78.23 72.86* .00* Within Groups 668.83 6231.07 Total 903.48 626 *p<.05 Vonnetek Automotive can be relied on to keep its promises.

PAGE 142

142 Table F-4 ANOVA (DV: TAKESOPINIONSOF, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 101.24 3 33.75 18.40* .00* Within Groups 1142.39 623 Total 1243.63 626 *p<.05 I believe that Vonnetek Automotive takes the opi nions of people like me into account when making decisions.

PAGE 143

143 Table F-5 ANOVA (DV: CONFIDENTSKILLS, IV: CRISISTYPE) *p<.05 I feel very confident about Vonnetek Automotives skills. Source Sum of SquaresDf Mean SquareF Sig. Between Groups 83.79 3 27.93 12.41* .00* Within Groups 999.38 6231.60 Total 1083.17 626

PAGE 144

144 Table F-6 ANOVA (DV: ABILITYACCOM PLISH, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 29.07 3 9.69 7.47* .00* Within Groups 808.40 6231.30 Total 837.47 626 *p<.05 Vonnetek Automotive has the ability to accomplish what it says it will do.

PAGE 145

145 Table F-7 ANOVA (DV: SOUNDPRINCIPLES, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 341.37 3 113.79 93.05* .00 Within Groups 761.82 6231.25 Total 1057.23 626 *p<.05 Sound principles seem to guide Vonnetek Automotives behavior.

PAGE 146

146 Table F-8 ANOVA (DV: DOESNOTMISLEAD, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 278.77 3 92.92 74.37* .00* Within Groups 778.46 6231.25 Total 1057.23 626 *p<.05 Vonnetek Automotive does not mislead people like me.

PAGE 147

147 Table F9 ANOVA (DV: MAKEDECISIONSFOR, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 70.38 3 23.46 12.51* .00* Within Groups 1168.58 6231.88 Total 1238.96 626 *p<.05 I am very willing to let Vonnetek Automotive make decisions for people like me.

PAGE 148

148 Table F-10 ANOVA (DV: WATCHCLO SELY, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 142.75 3 47.58 28.20* .00* Within Groups 1051.15 6231.69 Total 1193.90 626 *p<.05 I think it is important to watch Vonnetek Automotiv e closely so that it does not take advantage of people like me.

PAGE 149

149 Table F-11 ANOVA (DV: SUCCESSFULATTHINGS, IV: CRISISTYPE) Source Sum of SquaresDf Mean SquareF Sig. Between Groups 6.95 3 2.32 2.38 .07 Within Groups 607.04 623.97 Total 613.99 626 *p<.05 Vonnetek Automotive is known to be successful at the things it tries to do.

PAGE 150

150 LIST OF REFERENCES The American Heritage Dictionary (2nd college edition) (1985). Boston: HoughtonMifflin. Benoit, W.L. (1997). Image repair di scourse and crisis communications. Public Relations Review, 23 (2), 177-186. Berger, B.K. (1999). The Halcion affair: Pub lic relations and the construction of the ideological world view. Journal of Public Relations Research, 11(3), 185-203. Bobo, C. (1997). Hitachi faces crisis with textbook response. Public Relations Quarterly, 42(2), 18-21. Boje, D.M., & Rosile, G.A. (2003). Life imitates art: Enrons epic and tragic narration. Management Communication Quarterly, 17 85-125. Botan, C.H. & Hazelton, V. (2006). Public relations theory II Lawrence Erlbaum Associates. Bradford, J.L., & Garrett, D.E. (1995). The effectiveness of cor porate communicative responses to accusations of unethical behavior. Journal of Business Ethics, 14 875-892. Brassell-Cicchini, L.A. (2003). The sh areholder value of crisis handling. Risk Management, 50(5), 48. Brinson, S.L., & Benoit, W.L. (1999). The tarn ished star: Restoring Texacos damaged public image. Management Communication Quarterly, 12 483-510. Broom, G.M., Casey, S., & Ritchey, J. ( 1997). Toward a concep t and theory of organization-public relationships. Journal of Public Relations Research, 9 (2), 83-98. Cancel, A.E., Cameron, G.T., Sallot, L.M ., & Mitrook, M.A. (1997). It depends: A contingency theory of accommodation in public relations. Journal of Public Relations Research, 9(1) 31-63. Cancel, A.E., Mitrook, M.A., & Cameron, G.T. ( 1999). Testing the contingency theory of accommodation in public relations. Journal of Public Relations Review, 25 (2), 171-197. Capriotti, P., & Moreno, A. (2007). Corporat e citizenship and pub lic relations: The importance and interactivity of social res ponsibility issues on corporate websites. Public Relations Review, 22 84-91. Carney, A., Jordan, A. (1993). Prepare for business-related crises. Public Relations Journal, 49(8), 34-35.

PAGE 151

151 Carpenter, E., & McLuhan, M. (1960). Explorations in Comm unication: An anthology Beacon Press. Christen, C.T. (2005). The restructuring and reengineering of AT&T : Analysis of a public relations crisis us ing organizational theory. Public Relations Review, 31 239-251. Coombs, W.T. (1995). Choosing the right words: The development of guidelines for the selection of the appropriate crisis response strategies. Management Communication Quarterly, 8 447-476. Coombs, W.T. (2000). Crisis management: Adva ntages of a relational perspectives. In J.A. Ledingham & S.D. Bruning (Eds.) Public relations as relationship management: A relational approach to public relations 73-93. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Coombs, W.T. (2002). Deep and surface threats: Conceptual and practical implications for crisis vs. problem. Public Relations Review, 28(4), 339-345. Coombs, W.T. (2004). West Phar maceuticals explosion: Stru cturing crisis discourse knowledge. Public Relations Review, 20 467.473. Coombs, W.T. (2007). Attribut ion theory as a guide for post-crisis communication research. Public Relations Review, 33, 135-139. Coombs, W.T., & Holladay, S.J. (2002). Helpi ng crisis managers protect reputational assets. Management Communication Quarterly, 16(2), 165-186. Deutsch, M. (1958). Trust and suspicion. The Journal of Conflict Resolution, 2 (4), 265279. Duhe, S.F., & Zoch, L.M. (1994). Framing the medias agenda during a crisis. Public Relations Quarterly, 39(4), 42-45. Ehling, W.P. (1992). Estimating the value of public relations and co mmunication to an organization. In J.E. Grunig, D.M. Dozier, W.P. Ehling, L.A. Grunig, F.C. Repper, & J. White (Eds.), Excellence in public relations and communication management 616-638. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Fearn-Banks, K. (2002). Crisis communications: A casebook approach (2nd edition). Mahwah, NJ: Lawrence Erlbaum. Ferguson, M.A. (1984, August). Building theory in public relations: Interorganizational relationships as a public relations paradigm Paper presented at the meeting of the Association for Education in Journalism and Mass Co mmunication Annual Convention, Gainesville, Florida.

PAGE 152

152 Gonzalez-Herrero, A., & Pratt, C.B. (1996). An integrated symmetri cal model for crisiscommunications management. Journal of Public Relations Research, 8 (2), 79-105. Goodman, R. (2006). Corporate America behind the curve: Makovsky 2006 state of corporate blogging survey. The Institute for Public Relations Gainesville, FL. Greenberg, K.E. (1993). Pepsis big scare. Public Relations Journal, 49 (8), 6-7. Grunig, J.E., Dozier, D.M., Ehling, W.P., Gruni g, L.A., Repper, F.C., & White, J. (1992). Excellence in public relations and communications management Hillsdale, NJ: Lawrence Erlbaum. Grunig, J., Grunig, L., Sriramesh, K., Huang, Y ., & Lyra, A. (1995). Models of public relations in an in ternational setting. Journal of Public Relations Research, 7 (3), 163-186. Grunig, J.E., & Huang, Y.H. (2000). From orga nizational effectivene ss to relationship indicators: Antecedents of relationships, public relati ons strategies, and relationship outcomes. In J.A. Ledingham & S.D. Bruning, (Eds.), Public relations as relationship management: A relational approach to the st udy and practice of public relations Mahwah, NJ: Lawrence Erlbaum. Hall, M.R. (2006). Corporate philanthropy and corporate community relations: Measuring relationship-building results. Journal of Public Relations Research, 18 (1), 121. Hrtel, C., McColl-Kennedy, J.R., & McDonal d, L. (1998). Incorporating attribution theory and the theory of reas oned action within an affective events theory framework to produce a contingency predictive model of consumer reactions to organizational mishaps. Advances in Consumer Research, 25 428-432. Hayes. (1985). Corporate crisis management as adaptive control. In S.J. Andriole (Ed.), Corporate Crisis Management Princeton: Petrocelli Books. Hearit, K.M. (1999). Newsgroups, activist publics, and corporat e apologia: The case of Intel and its Pentium chip. Public Relations Review, 25 (3), 291-308. Hearit, K.M., & Brown, J. (2004). Merrill Lync h: Corporate apologia and business fraud. Public Relations Review, 30 459-466. Heath, R.L. (2001). Shifting foundations: Public relations as relati onship building. In R.L. Heath (Ed.), Handbook of Public Relations Thousand Oaks, CA: Sage. Hon, L.C., & Grunig, J.E. (1999). Guidelines for measuring relati onships in public relations. A report for the Co mmission on Public Relations Measurement and Evaluation, the Institute for Public Relations Gainesville, Florida.

PAGE 153

153 Huang, Y. (2001). OPRA: A cross-cultural, multiple-item scale for measuring organization-public relationships. Journal of Public Relations Research, 13 61-91. Huang, Y.-H. (2001). Values of public re lations: Effects on organization-public relationships mediati ng conflict resolution. Journal of Public Relations Research, 13 (4), 265-301. Ihlen, O. (2002). Defending the Mercedes A-Class: Combining and changing crisisresponse strategies. Journal of Public Relations Research, 14 (3), 185-206. Jo, S., & Kim, Y. (2003). The effect of web characteristics on relationship building. Journal of Public Relations Research, 15(3), 199-223. Jorgenson, B.K. (1994). Consumer reaction to co mpany-related disasters: The effect of multiple versus single explanations. Advances in Consumer Research, 21 348-352. Jorgenson, B.K. (1996). Components of consumer reaction to company-related mishaps: A structural equati on model approach. Advances in Consumer Research, 23 346-351. Kauffman, J. (1997). NASA in crisis: The space agencys public relations efforts regarding the Hubble space telescope. Public Relations Review, 23 (1), 1-10. Kauffman, J. (1999). Adding fuel to the fire : NASAs crisis communications regarding Apollo 1. Public Relations Review, 25 (4), 421-432. Kauffman, J. (2001). A successful failure: NASAs crisis communications regarding Apollo 13. Public Relations Review, 27 437-448. Kauffman, J. (2005). Lost in space: A critique of NASAs cr isis communications in the Columbia disaster. Public Relations Review, 31 263-275. Kasperson, R.E. (1986). Six propositions on public participation and their relevance for risk communication. Risk Analysis, 6 275-281. Kennedy, J.F. (1959). Speech in Indianapolis to Convocation of the United Negro College Fund, April 12. Kent, M.L., & Taylor, M. (2002). Toward a dialogic theory of public relations. Public Relations Review, 28 21-37. Ki., E.-J., & Hon, L.C. (2007). Testing th e linkages among the organization-public relationship and attitude a nd behavioral intentions. Journal of Public Relations Research, 19(1), 1-23. Kuhn, T.S. (1996). The structure of scientific revolutions (third edition) The University of Chicago Press.

PAGE 154

154 LaPlant, K. (1999). The Dow Corning crisis: A benchmark. Public Relations Quarterly, 44(2), 32-33. Lasswell, H.D. (1948). The structure and f unction of communicati on in society. In Lyman Bryson (Ed.) The Communication of Ideas Harper and Row. Ledingham, J., & Bruning, S. (1998). Relatio nship management in public relations: Dimensions of an organization-public relationship. Public Relations Review, 24 (1), 5565. Ledingham, J.A., & Bruning, S.D. (2000). Intr oduction: Background and current trends in the study of relationship management. In J.A. Ledingham & S.D. Bruning (Eds.), Public relations as relationship management: A relational approach to the study and practice of public relations Mahwah, NJ: Lawrence Erlbaum. Ledingham, J.A. (2001). Government and citizen ry: Extending the relational perspective of public relations. Public Relations Review, 27 285-295. Ledingham, J.A. (2003). Explicating relationship management as a general theory of public relations Journal of Public Relations Research, 15 (2), 181-198. Leeper, K.A. (1996). Public relations ethics and communitarianism: A preliminary investigation. Public Relations Review, 22 (2), 163-179. Logan, R.K. (2004). The sixth language: Learning a liv ing in the internet age, 2nd edition The Blackburn Press. Marra, F.J. (1998). Crisis communi cation plans: Poor predictors of excellent crisis public relations. Public Relations Review, 24(4), 461-474. Marsh, C. (2006). The syllogism of apologia: Rhetorical statis th eory and crisis communication. Public Relations Review, 32, 41-46. Martinelli, K.A., & Briggs, W. (1998). Integr ating public relations and legal responses during crisis: The case of Odwalla, Inc. Public Relations Review, 24 (4), 443-460. McDonald, L., & Hrtel, C.E.J. (2000). A pplying the involvement construct to organisational crises. In Proceedings of the Australian and New Zealand Marketing Academy Conference (pp. 799-803). Gold Coast, Aust ralia: Griffith University, Department of Marketing. McLuhan, M., McLuhan, E., & Zingrone, F. (1996). Essential McLuhan Basic Books. Modzelewski, J. (1990). What I would do: CEOs consider corporate crises. Public Relations Quarterly, 35(1), 12-14. Molleda, J.C. (2001). International paradigm s: The Latin American school of public relations. Journalism Studies, 2 (4), 513-530.

PAGE 155

155 Pavlik, J.V., & Dozier, D.M. (1996). Managi ng the information superhighway: A report on the issues facing communications professionals. The Institute for Public Relations Gainesville, FL. Pauchant, T.C., & Mitroff, I.I. (1992). Transforming the crisis prone organization. San Francisco, CA:Jossey-Bass Publishers. Pines, W.L. (2000). Myths of crisis management. Public Relations Quarterly, 45 (3), 1517. Poundstone, W. (1993). Prisoners dilemma Anchor Books. Ressler, J.A. (1982). Crisis communications. Public Relations Quarterly, 27 (3), 8. Reynolds, B., & Seeger, M.W. (2005). Crisis and emergency risk communication as an integrative model. Journal of Heath Communication, 10 (1), 43-55. Rogers, E. (2003). Diffusion of innovations (5th edition) Free Press. Sallot, L.M., Lyon, L.J., Acosta-Alzuru, C., & Jones, K.O. (2003). From aardvark to zebra: A new millennium analysis of theory development in public relations academic journals. Journal of Public Relations Research, 15 (1), 27-90. Seeger, M.W., Sellnow, T.L., & Ulmer R.R. (1998). Communication, organization and crisis. Communication Yearbook, 21, 231-275. Seeger, M.W., & Ulmer, R.R. (2001). Virtuous responses to organizational crisis: Aaron Feuerstein and Milt Colt. Journal of Business Ethics, 31 (4), 369-376. Seeger, M.W., & Ulmer, R.R. (2002). A post-crisis discourse of renewal: The cases of Malden Mills and Cole Hardwoods. Journal of Applied Communication Research, 30 (2), 126-142. Seeger, M.W., & Ulmer, R.R. (2003). Explai ning Enron: Communication and responsible leadership. Management Communication Quarterly, 17 58-84. Seltzer, T., & Mitrook, M.A. (2007) The dialogic potential of weblogs in relationship building. Public Relations Review, 33 227-229. Shin, J-H., Cameron, G.T., & Cropp, F. (2006) Occams Razor in the contingency theory: A national survey on 86 contingent variables. Public Relations Review, 32 282286. Shoemaker, P. (2003). How to build social science theories. Sage Publications, Inc.

PAGE 156

156 Stevens, B. (1999). Persuasion, probity, and paltering: The Prudential crisis. Journal of Business Communication, 36 (4), 319-334. Stockmyer, J. (1996). Brands in crisis: Consumer help for deserving victims. Advances in Consumer Research, 23 429-435. Taylor, M. (2000). Cultural va riance as a challenge to gl obal public relations: A case study of the Coca-Cola scare in Europe. Public Relations Review, 26 (3), 277-293. Tyler, L. (1997). Liability means never bei ng able to say youre so rry: Corporate guilt, legal constraints, and defensiv eness in corporate communication. Management Communication Quarterly, 11 51-73. Watson, M.L. (2005). Can there be just one trus t? A cross-disciplinary identification of trust definitions and measurement. Submitted to the Institute for Public Relations for the 2004 Ketchum Excellence in Public Relations Research Award. Wilcox, D.L., Cameron, G.T. (2007). Public relations strategies and tactics, study edition 8/e. Boston: Allyn and Bacon. Yang, S.-U. (2007). An integrated model for organization-public relational outcomes, organizational reputation, and their antecedents. Journal of Public Relations Research, 19(2), 91-121.

PAGE 157

BIOGRAPHICAL SKETCH Seth Oyer is an accom plished public relations professional with over a decade of executive and management experience in the technology indus try. A crisis communicat ions specialist, he has acted as a public relations cons ultant to corporate, not-for-pro fit, and political organizations. An award-winning instructor, Oyer has experience teaching smalland large-lecture and interactive classes, resulting in excellent studen t evaluations (with normal grading curves). He has also provided service during his doctoral studies on faculty search committees and as president of Graduate Students in Mass Communication Association. Oyer earned his BA as a double-major in E nglish and communicati ons from the State University of New York College at Fredonia. He earned his MA in mass communication from the University of Central Florida. He comple ted his PhD at the University of Florida and accepted an assistant professor at Bowling Green State University, starting in fall 2008. Oyer has been published and has several conf erence papers/publications in progress.