KILLING THE MESSENGER: A SURVEY OF PUBLIC RELATIONS PRACTITIONERS AND ORGANIZATIONAL RESPONSE TO WHISTLEBLOWING AFTER SARBANES-OXLEY by CARY A. GREENWOOD A DISSERTATION Presented to the School of Journalism and Communication and the Graduate School of the University of Oregon in partial fulfillment of the requirements for the degree of Doctor of Philosophy September 2011 ii DISSERTATION APPROVAL PAGE Student: Cary A. Greenwood Title: Killing the Messenger: A Survey of Public Relations Practitioners and Organizational Response to Whistleblowing after Sarbanes-Oxley This dissertation has been accepted and approved in partial fulfillment of the requirements for the Doctor of Philosophy degree in the School of Journalism and Communication by: Dr. Patricia A. Curtin Co-Chairperson Dr. H. Leslie Steeves Co-Chairperson Dr. James K. Van Leuven Member Dr. Michael Russo Member Dr. Anne Parmigiani Outside Member and Kimberly Andrews Espy Vice President for Research and Innovation/Dean of the Graduate School Original approval signatures are on file with the University of Oregon Graduate School. Degree awarded September 2011 iii © 2011 Cary A. Greenwood iv DISSERTATION ABSTRACT Cary A. Greenwood Doctor of Philosophy School of Journalism and Communication September 2011 Title: Killing the Messenger: A Survey of Public Relations Practitioners and Organizational Response to Whistleblowing after Sarbanes-Oxley Approved: _______________________________________________ Dr. Patricia A. Curtin, Co-Chair Approved: _______________________________________________ Dr. H. Leslie Steeves, Co-Chair Whistleblowing has been a topic of media interest since the Vietnam War, and it continues to resonate strongly with the public. Several well-publicized whistleblowers have done much more than catch the attention of the world media. They arguably have changed the world. Whistleblowing refers to the reporting of illegal, wasteful, or unethical activities (i.e., wrongdoing) by current and former employees of an organization. Triggered by several highly publicized corporate financial failures, the Sarbanes-Oxley Act of 2002 requires publicly traded companies to provide an anonymous channel for employees to report financial wrongdoing and provides protection for those who do. Using resource dependence perspective and relationship management theory, this study uses e-mail to distribute an online survey to top-ranking public relations executives in the Fortune 1000 corporations to identify what role public relations executives have played in developing and publicizing anonymous whistleblowing channels, their knowledge of wrongdoing in their own organizations and elsewhere, their attitudes and v actions related to the wrongdoing, the consequences of their actions, and their relationships with their organizations. The study finds that only one-fifth of respondents helped develop the required anonymous communication channel, but two-thirds helped publicize it; almost one-half of respondents are aware of wrongdoing in their corporations or in other organizations, and two-thirds of those report such activities; those who report wrongdoing do so through internal channels within the corporation, with one exception; few who report wrongdoing suffer retaliation; and the vast majority enjoy positive relationships with their organizations. However, a small number of respondents experienced retaliation, and the research points to a broader exploration of this topic among public relations personnel within Fortune 1000 corporations to determine to what extent status, relationships, and benefits such as the “golden handcuffs” influence whistleblowing. Future research on whistleblowing and ethics in public relations is warranted. vi CURRICULUM VITAE NAME OF AUTHOR: Cary A. Greenwood GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon, Eugene Eastern Washington University, Cheney University of Oklahoma, Norman DEGREES AWARDED: Doctor of Philosophy, Communication and Society, 2011, University of Oregon Master of Fine Arts, Creative Writing, 1988, Eastern Washington University Bachelor of Arts, Journalism, 1974, University of Oklahoma AREAS OF SPECIAL INTEREST: Whistleblowing Evolutionary theory PROFESSIONAL EXPERIENCE: Operations and Policy Analyst, Oregon Department of Transportation, 2 ½ years Public Affairs Officer, Oregon Department of Human Services, 2 years Public Affairs Director, Oregon Department of Forestry, 7 years Communications Manager, Oregon Department of Revenue, 1 ½ years Public Affairs Officer, Oregon Department of Transportation, 2 ½ years Supervisor Public Relations/Corporate Communications, The Washington Water Power Company, 8 ½years Corporate/Financial Public Relations Coordinator, The Washington Water Power Company, 2 years Public Involvement Contractor, Oregon Department of Energy, 6 mos. Senior Legislative Assistant to Speaker of House, Oregon Legislative Assembly, 6 mos. Staff Writer/Speechwriter, Oregon Executive Department, 2 years vii PUBLICATIONS: Greenwood, C. A. (2010). Evolutionary theory: The missing link for conceptualizing public relations. Journal of Public Relations Research, 22, 456-476. Greenwood, C.A., & Kahle, L.R. (2007, June). Toward an evolutionary theory of marketing: Evolution and branding [Abstract]. In Joseph R. Priester, Deborah J MacInnis, & C. Whan Park, (Eds.), New Frontiers in branding: Attitudes, attachments, and relationships, 3-11. Santa Monica, CA: Society for Consumer Psychology. viii ACKNOWLEDGMENTS I wish to express my sincere appreciation to the members of my dissertation committee, who have guided and encouraged me throughout this process: Dr. Patricia A. Curtin, School of Journalism and Communication Chair in Public Relations; Dr. H. Leslie Steeves, Associate Dean, Graduate Affairs and Research, SOJC; Dr. James K. Van Leuven, former Professor and Endowed Chair in Public Relations, SOJC; Dr. Michael Russo, Lundquist Professor of Sustainable Management, Lundquist College of Business; and Dr. Anne Parmigiani, Assistant Professor of Management, Lundquist College of Business. They have given generously of their time and wisdom, and I sincerely thank each one of them. Dr. Curtin, in particular, has been with me every step of the way, and I owe her a huge debt of gratitude for helping me realize this dream. In addition, I would like to thank the many others in the academic community who have given me support. I cannot thank them all, but I especially would like to thank Dr. Tiffany Derville Gallicano, Assistant Professor, SOJC; Dr. John Orbell, Professor Emeritus, Political Science Department, and Dr. Lynn Kahle, Ehrman Giustina Professor of Marketing/Dept. Head, Lundquist College of Business, who shared their specialties with me. I would like to offer a special thanks to Dr. James E. Grunig, Professor Emeritus, and Dr. Larissa A. Grunig, Professor Emerita, University of Maryland, for their contributions to the field of public relations and for encouraging me to pursue this goal. Thanks go, too, to Petra Hagen, SOJC graduate programs, and Sue Varani, faculty services coordinator, for their tireless patience and support; to Kaiping Zhang, ix business/economics librarian, who helped with research; and to Spencer Smith, list master, who helped deliver the survey. I especially would like to thank my cohort, Anne Conaway, Dr. Mary Erickson, Dr. Janet Kwami, Yeong Doo Moon, Dr. Glenn Morris, and Dr. Kathleen Ryan for sharing their lives with me. And, finally, I would like to thank those public relations employees of Fortune 1000 corporations who responded to my survey. I sincerely appreciate your willingness to share your insights and perspectives. Without your help, this study would not have been possible. I hope we have another opportunity in the future to communicate about whistleblowing. x To my husband, Paul L. Nosbisch, whose love, dedication, and confidence never wavered and who truly made this possible; to my mother, Vonnett M. Greenwood, who continues to provide daily encouragement, and to my father, Orville Greenwood, who passed away October 2007, both of whom encouraged my pursuit of education throughout my life; to my children, Bridget Marie Greenwood-Geer and Todd Damien Greenwood-Geer, who have understood my need to learn and who have cheered me on, no matter what sacrifice; to my daughter-in-law, Dominic Maltais, who is embarking on an academic journey of her own; and to my brother, Richard P. Greenwood, and his family, who have kept the home front secure: thank you all for this gift of time, love and support. I love you. xi TABLE OF CONTENTS Chapter Page I. OVERVIEW OF WHISTLEBLOWING ................................................................. 1 Introduction............................................................................................................ 1 Significance of This Study..................................................................................... 4 U.S. History of Whistleblowing............................................................................. 6 Definitions of Whistleblowing......................................................................... 6 Whistleblowing during the 1960s and 1970s................................................... 7 Whistleblowing Protections in the Federal Government ................................. 9 Failure of Federal Whistleblower Protections ................................................. 10 Nixon-era Ties to Whistleblower Retaliation ............................................ 11 Retaliation against Whistleblowers............................................................ 12 Corporate Whistleblowers ............................................................................... 13 Enron—Sherron Watkins—2001............................................................... 13 WorldCom—Cynthia Cooper—2002 ........................................................ 15 The Sarbanes-Oxley Act of 2002........................................................................... 16 Expectations for Sarbanes-Oxley..................................................................... 16 Provisions of Sarbanes-Oxley.......................................................................... 17 Reality of Sarbanes-Oxley ............................................................................... 18 Whistleblowers as Defense against Fraud ....................................................... 20 II. LITERATURE REVIEW........................................................................................ 21 Federal Research on Whistleblowing .............................................................. 21 Whistleblowing in Management Literature ..................................................... 23 xii Chapter Page Resource Dependence Perspective ............................................................ 24 Research Methods...................................................................................... 25 Empirical Findings..................................................................................... 27 Retaliation .................................................................................................. 31 Whistleblowing in Communication Literature ................................................ 37 Upward Communication............................................................................ 37 Organizational Climate .............................................................................. 38 Interpersonal Closeness ............................................................................. 39 Communication Channels.......................................................................... 39 Whistleblowing in Public Relations Literature................................................ 40 Public Relations Roles ............................................................................... 42 Relationship Management ......................................................................... 44 Summary of Literature Review........................................................................ 47 Research Questions and Hypotheses ............................................................... 49 III. METHOD .............................................................................................................. 55 Introduction...................................................................................................... 55 Survey Research............................................................................................... 55 Survey Error..................................................................................................... 56 Coverage Error........................................................................................... 56 Construct Validity...................................................................................... 57 Measurement Error .................................................................................... 58 Nonresponse Error ..................................................................................... 58 xiii Chapter Page Survey Instrument............................................................................................ 59 Survey Components ................................................................................... 60 Section I ...................................................................................... 60 Section II ...................................................................................... 60 Section III ...................................................................................... 60 Section IV ...................................................................................... 60 Section V ...................................................................................... 63 The Measures ............................................................................................. 63 Incidence of Wrongdoing .................................................................... 63 Incidence of Whistleblowing ............................................................... 64 Incidence of Identified Whistleblowing............................................... 64 Incidence of Retaliation ....................................................................... 64 The Pilot Study ................................................................................................ 65 Changes Made to Instruments.................................................................... 65 Recommendations Related to Federal Studies........................................... 66 Recommendations Related to Relationship Scales .................................... 67 Population ........................................................................................................ 67 Developing the Mailing List: List 1................................................................. 68 Commercial Marketing Databases ............................................................. 68 Mechanical Turk ........................................................................................ 69 Mint Global................................................................................................ 69 Querying the Database......................................................................... 70 xiv Chapter Page Refining the Search Criteria................................................................. 71 Web Sites .................................................................................................. 71 Reducing Sampling Error ................................................................................ 73 Survey Administration ..................................................................................... 74 Survey Delivery: List 1.............................................................................. 75 Corrective Actions ..................................................................................... 77 Bounce Backs............................................................................................. 77 List Master ................................................................................................. 79 Aggregators, Newsletters, and Other Online Sources ............................... 80 Social Media .............................................................................................. 80 LinkedIn PR News Group.................................................................... 80 Public Relations Society of America Educators Academy E-Group... 81 Public Relations Society of America Membership List............................. 82 National Whistleblower Organizations...................................................... 82 Arthur W. Page Society ............................................................................. 83 PRSA Board of Directors........................................................................... 85 Arthur W. Page Society Revisited ............................................................. 85 Academic E-mail Marketing...................................................................... 86 Commercial Direct Marketing ................................................................... 86 Subject Lines ...................................................................................... 87 Subject Line Development................................................................... 88 Subject Line Testing Process ............................................................... 89 xv Chapter Page Subject Line Testing Implementation: Phase I .................................... 89 Subject Line Testing Implementation: Phase II................................... 92 Subject Line Testing Implementation: Phase III ................................. 92 Subject Line Test Results..................................................................... 93 Subject Line Test Bounce Backs ......................................................... 93 List 1 Distribution 2 (Reminder 1)............................................................. 94 List 1 Distribution 3 (Reminder 2)............................................................. 94 List 1 Results.............................................................................................. 95 Expanding the Population ................................................................................ 95 Qualtrics .................................................................................................. 96 The Loss of Mint Global............................................................................ 97 List 2 .................................................................................................. 97 List 3 .................................................................................................. 98 List 2 Distribution 1 ................................................................................... 99 List 2 Distribution 2 ................................................................................... 100 List 2 Distribution 3 ................................................................................... 100 List 3 Distribution 1 ................................................................................... 100 List 3 Distribution 2 ................................................................................... 101 List 3 Distribution 3 ................................................................................... 101 Summary .................................................................................................. 101 IV. RESULTS .............................................................................................................. 105 Response Rate.................................................................................................. 105 xvi Chapter Page Sample Size...................................................................................................... 107 Data Analysis ................................................................................................... 108 Demographics .................................................................................................. 109 Employment Characteristics ............................................................................ 110 Demographics and Characteristics of Top-ranking Public Relations Executives .................................................................................................. 113 Sarbanes-Oxley................................................................................................ 116 Whistleblowing Channels .......................................................................... 116 Knowledge of Illegal, Wasteful, or Unethical Activity: Wrongdoing ............................................................................................... 118 Third-party Awareness of Wrongdoing ..................................................... 118 Personal Awareness of Wrongdoing.......................................................... 122 How Prevalent Is the Wrongdoing?........................................................... 125 How Serious Is the Wrongdoing? .............................................................. 126 Was There Retaliation for Reporting Wrongdoing?.................................. 127 Elements of Power ........................................................................................... 127 Relationships: Outcomes and Types................................................................ 129 Relationship Quality Outcomes ................................................................. 131 Relationship Types..................................................................................... 135 Summary of Findings....................................................................................... 137 V. DISCUSSION ......................................................................................................... 138 Study Limitations............................................................................................. 138 Answering the Research Questions and Hypotheses ....................................... 141 xvii Chapter Page Developing the Anonymous Communication Channel ............................. 142 Publicizing Whistleblowing Channels....................................................... 143 Do Public Relations Executives Blow the Whistle? .................................. 144 Do Public Relations Executives Think Blowing the Whistle Is Part of Their Job?............................................................................................. 146 How Do Public Relations Executives Blow the Whistle? ......................... 148 Have Public Relations Whistleblowers Experienced Retaliation? ............ 150 Has Blowing the Whistle Affected Their Relationships with Their Employers? ......................................................................................... 152 Implications for Public Relations Theory........................................................ 155 Public Relations Roles or Organizational Turf .......................................... 155 Whistleblowing as Ethical Obligation ....................................................... 161 The Golden Handcuffs............................................................................... 163 Conclusion ....................................................................................................... 165 APPENDICES ............................................................................................................. 171 A. QUESTIONNAIRE FOR RANKING PUBLIC RELATIONS EXECUTIVE IN FORTUNE 1000 CORPORATION.............................. 171 B. RECRUITMENT CONSENT E-MAIL 2-24-10 .............................................. 179 C. PERMISSION/RELEASE FORM .................................................................... 181 D. FOLLOW-UP REMINDER.............................................................................. 182 REFERENCES CITED................................................................................................ 184 xviii LIST OF TABLES Table Page 1. Relationship Quality Outcome Measurement Items .............................................. 62 2. Relationship Types Measurement Items................................................................ 62 3. Research Questions, Hypotheses, Measurement, and Statistics ............................ 103 4. Research Questions, Measurement, and Statistics Revised ................................... 104 5. Respondent Demographics .................................................................................... 110 6. Employment Variables for All Respondents ......................................................... 111 7. Employment Variables for Top-Ranked Public Relations Executive Respondents ..................................................................................................... 114 8. Total Awareness of Wrongdoing ........................................................................... 126 9. Seriousness of Wrongdoing ................................................................................... 126 10. Comprehensiveness of Retaliation......................................................................... 127 11. Logistic Regression on Total Awareness of Wrongdoing ..................................... 129 12. Relationship Scale Cronbach’s Alpha ................................................................... 130 13. Exchange-Total Statistics....................................................................................... 131 14. Descriptive Statistics of Relationship Outcomes and Types ................................. 132 15. Descriptive Statistics of Relationship Quality Outcomes by Question ................. 133 16. Descriptive Statistics of Relationship Types by Question..................................... 136 1 CHAPTER I OVERVIEW OF WHISTLEBLOWING Introduction Whistleblowing has been a topic of media interest since the Vietnam War, and it has continued to resonate strongly with the public for almost 50 years. Several well- publicized whistleblowers have done much more than catch the attention of the world media. They arguably have changed the world. Whistleblowing emerged as a phenomenon in the 1960s and 1970s as a result of cynicism over corporate and governmental excesses (Glazer & Glazer, 1989) and “the civil rights, antiwar, consumer, and student protest movements” (Westin, 1981, p. 7). Although considered by some “a recent phenomenon in American history” (Glazer & Glazer, 1989, p. 4), the public’s awareness of whistleblowing may owe more to late 20th- century advances in technology and mass communication than to any sudden change in human nature. In other words, whistleblowing may be an ancient practice popularized by recent mass-communicated publicity. A brief review of the recent history of whistleblowing illustrates its importance to society as a mode of “ethical resistance” (Glazer & Glazer, 1989, p. 11), to organizations as a method of communication (Richardson, 2005), to business as a means of informing management of operational problems (Moberly, 2006), and to public relations as a means of dissent (Berger & Reber, 2006). But whistleblowing is not without a cost to the individual and to the organization. This study explores the extent of whistleblowing within the field of public relations. 2 The most common definition of whistleblower refers to a group member who goes against group norms and attempts to change improper group behavior by alerting an internal or external authority (Miceli & Near, 1984). The assumed results of such action are success, failure, and/or retaliation (Rehg, Miceli, Near, & Van Scotter, 2008). One of the most famous figures during the Watergate scandal of the early 1970s and one of the most enduring examples of whistleblowing in history is Deep Throat. Deep Throat provided confidential information that prompted The Washington Post’s investigation of the Watergate break-in (Von Drehle, 2005) and eventually led to Richard Nixon’s resignation as President. Deep Throat was viewed as a hero until he was unveiled in 2005 as W. Mark Felt, deputy director of the FBI at the time of Watergate. That information led to speculation he had been motivated not by altruism but by revenge for being passed over as successor to J. Edgar Hoover. In recent years, whistleblowers have alleged improper accounting methods by Enron and WorldCom and failure by the FBI to prevent 9/11 despite prior intelligence. Their actions led Time Magazine to label 2002 “The Year of the Whistle-Blowers” and label those involved—Sherron Watkins (Enron), Cynthia Cooper (WorldCom), and Coleen Rowley (FBI)—“Persons of the Year” (Lacayo & Ripley, 2002). The magazine elevated the whistleblower to hero, or in this case, heroine, and recognized the societal good produced by the whistleblower. The risks and punishments associated with whistleblowing were also clear. Sometimes those risks and punishments resulted from retaliation by the affected organization. Since the Civil War, the federal government has provided protection from those risks and punishments to private citizens who entered into actions against the 3 federal government. It has provided protections for federal employees since the 1950s. However, those protections would not be extended to whistleblowers in the private sector until the passage of the Public Company Accounting Reform and Investor Protection Act of 2002, also called the Sarbanes-Oxley Act. The U.S. Congress passed Sarbanes-Oxley to prevent financial fraud in corporations by increasing financial oversight and transparency, auditor independence, and corporate accountability (Brickey, 2003). The legislation not only changed corporate financial practices, but also, for the first time, mandated that publicly traded companies establish and maintain anonymous communication channels for reporting financial wrongdoing directly to external directors on the audit committee of the company’s board of directors. It also provided strong sanctions for retaliation against whistleblowers. Given this legislated mandate to establish communication channels and the newly strengthened sanctions for retaliation against whistleblowers, one would expect business and related fields, such as law and public relations, to take an increased interest in the phenomenon of whistleblowing. That has been the case for business and for law; recent journal articles on the subject are extensive in both fields. However, there is a dearth of publications examining whistleblowing in the major public relations journals. The connection between Sarbanes-Oxley’s internal communication requirements and public relations’ role in internal communications in many organizations does not appear to have been explored by scholars outside a series of studies on various influence strategies in public relations practice (Berger, 2005; Berger & Reber, 2006; Kang & Berger, 2009). Also unexplored in the public relations literature to date is the role relationships between an employee and employer play in the decision to blow the whistle. Studies of 4 federal employees in 1980, 1983, and 1992 established the role that relationships play both in the decision to report wrongdoing and also in the consequences to the whistleblower of reporting wrongdoing, yet these findings have not been applied to public relations to date. Significance of This Study While some research has addressed whistleblowing in public relations as part of a larger study of influence and dissent (Berger, 2005; Berger & Reber, 2006; Kang & Berger, 2009), it appears no study has focused exclusively on whistleblowing by public relations practitioners in the private sector in the United States. This study addresses that lack of research by surveying public relations executives of the Fortune 1000 corporations to identify whistleblowing these individuals have observed, whistleblowing these individuals have conducted, retaliation these individuals have experienced, and the relationships these individuals have with their organizations. This study combines resource dependence perspective (Pfeffer & Salancik, 2003) from management and relationship management theory from public relations to explore the predictors of whistleblowing and retaliation and their relation to relationship types and outcomes within the Fortune 1000 corporations. It informs the teaching of public relations in that students are told that public relations is a management function that requires them to advise management on strategic, tactical, and ethical issues. This study explores an area that has not yet received a great deal of attention in public relations research inside or outside the whistleblowing context: the potential repercussions for public relations executives for doing what they have been directed to do by public relations education and theory—gain a seat at the management table, provide 5 advice and counsel to senior management, act as boundary spanners, and, in the process, bring bad news about the organization to management. In whistleblowing literature, this is viewed as role-prescribed whistleblowing. Role-prescribed whistleblowing is the reporting of wrongdoing by those who are required to do so by their job descriptions, such as internal auditors (Miceli & Near, 1994). In this sense, public relations executives are role-prescribed internal whistleblowers. The admonition to management to listen to those bringing bad news forward and not to “shoot the messenger” has been articulated (Yamey, 2000, p. 70), but the extent of retaliation against those bringing bad news forward has not been explored. This study makes several original contributions to the existing research on whistleblowing. First, it extends the study of whistleblowing into the corporation at the firm level, as called for by Parmerlee, Near, and Jensen (1982). Second, it administers an adaptation of the original government whistleblowing instrument developed by the U.S. Merit Systems Protection Board (U.S. Merit Systems Protection Board [USMSPB], 1981) to individuals in the private sector to determine if public sector results are applicable to the private sector, as called for by Miceli and Near (1984). Third, it extends the research on whistleblowing into public relations by administering an adaptation of three longitudinal USMSPB survey instruments to public relations executives in private corporations. Fourth, it explores role-prescribed whistleblowing and its consequences in the private sector. Fifth, it tests relationship management theory in a whistleblowing context. Finally, this study is the first to use variables from the USMSPB whistleblowing survey design to explore impacts on public relations executives’ relationships with their organizations. 6 The dissertation is divided into five chapters. The remainder of Chapter I explores the history of whistleblowing and legal protections for whistleblowers in the United States. Chapter II reviews whistleblowing research from the federal government and from management literature, as well as relationship management and related studies from public relations literature. Chapter III discusses the design and administration of a Web- based survey of Fortune 1000 corporation public relations executives and the statistical tools used to analyze the data. Chapter IV describes the results of the analysis. Chapter V presents the implications of the analysis for public relations theory development, for the practice of public relations, for the teaching of public relations, and for future research. U.S. History of Whistleblowing Definitions of Whistleblowing The concept of whistleblowing has undergone subtle revisions over time as research has been conducted in the public and private sectors. The definition of who is a whistleblower, what type of activity is reported by a whistleblower, and to whom the activity is reported have changed over time. For example, some researchers consider reporting to an internal authority a political action and not whistleblowing (Near & Miceli, 1987). This study employs a broader definition of whistleblowing and whistleblowers. The first characteristic of a whistleblower is that he or she must be a current or former employee of the organization on which he or she is blowing the whistle. The second characteristic of a whistleblower is that he or she must report some type of negative activity. This is identified as “illegal or wasteful” in the first federal study 7 (USMSPB, 1981, p. 2) and as “illegal, immoral, or illegitimate” in management studies (Miceli & Near, 1984, p. 689). The third characteristic of a whistleblower is that he or she reports “to persons or organizations who may be able to affect action” (p. 689). This implies that the recipient of the report may be someone inside or outside the organization, and whistleblowers are characterized as either internal or external, depending on the location of the channels they use and to whom they report wrongdoing (Miceli & Near, 1984). The definition of whistleblowing used by most researchers (Miceli, Near, & Dworkin, 2008) is “the disclosure by organization members (former or current) of illegal, immoral, or illegitimate practices under the control of their employers, to persons or organizations that may be able to effect action” (Near & Miceli, 1985, p. 4). For the purpose of this study, whistleblowing is defined as disclosing a private corporation’s illegal, wasteful, or unethical practices to persons or organizations inside or outside the corporation that may be able to effect action. Although the definition of whistleblowing has always contained the components of internal or external reporting, whistleblowing became a household term because of the very public external reporting by certain individuals during and following the Vietnam War. Whistleblowing during the 1960s and 1970s “Ethical resisters—employees who publicly disclose unethical or illegal practices in the workplace—are a recent phenomenon in American history,” according to (Glazer & Glazer, 1989, p. 4). Although government had encouraged U.S. citizens to report on government fraud for a hundred years, whistleblowing emerged as a new and widely 8 publicized phenomenon in the 1960s and 1970s. This emergence occurred as a result of cynicism over corporate and governmental excesses (Glazer & Glazer, 1989) and “the civil rights, antiwar, consumer, and student protest movements” (Westin, 1981, p. 7). The actions of three former whistleblowers, Daniel Ellsberg, Deep Throat, and John Dean III, were directly related to U.S. involvement in the Vietnam War, while those of a fourth, Karen Silkwood, were directed at the nuclear industry. In 1971 defense contract employee Daniel Ellsberg leaked documents that exposed the unsavory antecedents of the Vietnam War (the Pentagon papers) to external sources: The New York Times, the Washington Post, and other newspapers. In retaliation, Nixon administration operatives (the White House plumbers) burglarized Ellsberg’s psychiatrist’s office in an attempt to find information that would discredit him. The plumbers subsequently burglarized and wiretapped the Democratic National Committee headquarters at the Watergate Hotel. Deep Throat’s leaks to Post reporters Woodward and Bernstein about the Watergate break-in prompted The Post’s investigation of the affair (Von Drehle, 2005). President Richard Nixon appointed White House Counsel John W. Dean III to investigate the break-in, but ongoing media coverage was limited, and Nixon was reelected in 1972. In 1973 Dean charged Nixon with participating in the Watergate cover-up in testimony before the Senate Watergate Committee. Nixon fired Dean for testifying, but the Senate’s investigations led to disclosure of the White House tapes, which confirmed Nixon’s role in the cover-up and eventually led to his resignation (The Washington Post, 2008). The suspicion of widespread corruption was not limited to politics and the federal government. Industry had its share of detractors, and the nuclear industry in particular 9 was under siege when Karen Silkwood reported alleged safety violations at Kerr- McGee’s plutonium fuels production plant in Oklahoma. Her subsequent death in 1974 fed speculation about murder and a cover-up. Silkwood’s autopsy showed elevated levels of plutonium; the plant closed a year after her death; and her estate eventually won a million-dollar settlement (Los Alamos National Laboratory, 1995). These highly publicized cases made whistleblowing a public issue, but whistleblowing had been a longstanding practice protected by the federal government since the Civil War. Whistleblowing Protections in the Federal Government Whistleblowing was first given statutory sanction in the 1863 federal False Claims Act as a means to assist the federal government. The act encouraged private citizens to report fraud in federal contracting by filing lawsuits on behalf of the federal government, as well as themselves. These lawsuits were known as qui tam suits, Latin for "he who brings an action for the king as well as for himself," or whistleblower lawsuits (The False Claims Act Legal Center, 2008). It was almost 100 years before the federal government again passed whistleblowing legislation. The 1958 federal Code of Ethics for Government encouraged federal employees “to place loyalty to the highest moral principles above loyalty to their departments” by publicly disclosing “inappropriate behavior” (USMSPB, 1993, p. 33). The next decade saw two major actions for whistleblowers in the legislative and judicial arenas. The 1966 Freedom of Information Act, which established government documents as public records, gave whistleblowers “a statutory justification for exposing misconduct” on the basis of “the public’s right to know” (USMSPB, 1993, p. 33). Two years later a 10 United States Supreme Court ruling, Pickering v. Board of Education, 391 U.S. 563 (1968), upheld the free speech protections in the First Amendment for federal employees who criticize the federal government (USMSPB, 1993). Retaliation by the Pentagon against a Defense Department employee, Ernest Fitzgerald, who disclosed cost overruns in defense contracting in 1968, played a role in the passage of the first whistleblower protection legislation (Glazer & Glazer, 1989; USMSPB, 1993). The Civil Service Reform Act of 1978 (CSRA) gave statutory protections to federal employees who disclosed “fraud, waste, and abuse” in the federal government and prohibited retaliation against them (USMSPB, 1993, p. 33). CSRA also created the United States Merit System Protection Board (USMSPB) to enforce those protections (USMSPB, 1993). Between 1978 and 1993, Congress enacted more than 25 laws to protect whistleblowers in various situations. The 1986 amendments to the False Claims Act gave protection to those who filed whistleblower lawsuits (The False Claims Act Legal Center, 2008). The Whistleblower Protection Act of 1989 (WPA) expanded on CSRA by giving federal employees appeal rights for personnel actions (USMSPB, 1993), and the WPA Amendments of 1994 strengthened protections for whistleblowers. Additional protections were provided by the Notification and Federal Employee Antidiscrimination and Retaliation (NO FEAR) Act of 2002 and the No FEAR Act of 2006. Failure of Federal Whistleblower Protections Not everyone agrees that protections for whistleblowers have been successful. The Government Accountability Project (GAP), which identifies itself as a nonprofit, government and corporate watchdog and a whistleblower advocate, believes the 11 Whistleblower Protection Act of 1989 and its amendments have weakened, rather than strengthened, protections for whistleblowers (Government Accountability Project, 2008 b). GAP places the blame for this on decisions by the U.S. Court of Appeals for the Federal Circuit, the court assigned to hear whistleblower complaints. GAP argues that the court has issued rulings that have removed protections for many types of reporting of wrongdoing, including those made to fellow workers, management, and wrongdoers, as well as any that may be seen as part of one’s job duties (Government Accountability Project, 2008 a). The widely reported experiences of recent federal whistleblowers would tend to support that claim. In one example, Jessalyn Radack, an attorney with the Justice Department, was retaliated against for advising against interrogating the "American Taliban," John Walker Lindh (Leslie, 2003). In another, Justice Department attorney Thomas Tamm was retaliated against for blowing the whistle to the media about the Bush administration’s unauthorized government surveillance program against U.S. citizens (Isikoff, 2008). Nixon-era Ties to Whistleblower Retaliation The Nixon White House developed one of the most extensively documented processes of retaliation within the U.S. government in the Malek/May Manual, a training manual designed to instruct political appointees at the highest levels in the Nixon administration on ways to subvert the federal civil service system (U.S. Senate Committee on the Judiciary, 1986). Nixon believed federal agencies were top-heavy with Democrats put in place in the last days of the Johnson administration who would thwart his policies (U.S. Senate Committee on Governmental Affairs, 1978). The Malek/May 12 manual showed Nixon-appointees in federal agencies how to fire civil service employees “without violating civil service rules” and “served as a Watergate catalyst for civil service reform and creation of” the Office of Special Counsel in the U.S. Merit Systems Protection Board, (U.S. Senate Committee on the Judiciary, 1986, p. 123). The USMSPB was the agency created by CRSA to protect federal whistleblowers. Alex Kozinski, who served as Special Counsel in the USMSPB in the early 1980s, reportedly used the Malek/May manual to train upper-level federal managers on how to circumvent the civil service rules in order to retaliate against federal whistleblowers (U.S. Senate Committee on the Judiciary, 1986). Kozinski’s role in retaliating against the whistleblowers he was hired to protect came to light in his confirmation hearing for appointment to the Ninth Circuit Court of Appeals, to which he was confirmed (U.S. Senate Committee on the Judiciary, 1986). Retaliation against Whistleblowers Despite the extensive federal legislation enacted to prevent such actions, retaliation against whistleblowers continues to be seen by some as commonplace and institutionalized (Devine, 1997). Retaliation tactics, many of which were documented in the Malek/May manual (U.S. Senate Committee on Governmental Affairs, 1978), include diverting attention from the whistleblower’s message to the whistleblower, trying to discredit the whistleblower, starting a paper trail to document alleged whistleblower performance deficiencies, threatening the whistleblower, isolating the whistleblower, humiliating the whistleblower in front of peers, creating unreasonable work expectations that guarantee failure, prosecuting the whistleblower legally over manufactured charges, 13 assaulting the whistleblower physically, eliminating the whistleblower’s job, stalling the whistleblower’s career, and blacklisting the whistleblower (Devine, 1997). Federal employees, then, have in theory had protection from retaliation for whistleblowing, although in practice those protections may have been non-existent. Corporate employees, however, were without even those stated protections until widespread financial disasters among U.S. corporations prompted Congress to pass the Sarbanes-Oxley Act of 2002, the legislation designed to extend federal protections to corporate whistleblowers. Corporate Whistleblowers Two corporate cases, Watkins at Enron and Cooper at WorldCom, are presented in detail here because of their impact on the development of the federal Sarbanes-Oxley legislation and because of the significance of that legislation to the practice of public relations. Although these two cases deal with financial wrongdoing, they are germane to public relations because of the different channels of communication each whistleblower used, the results of their whistleblowing through those channels, and the response by Congress to mandate communication channels that would ensure that the concerns of employees were heard by those who would act on them. Enron—Sherron Watkins—2001 “It all began with Enron” (Brickey, 2003, p. 357): Sherron Watkins, vice president of corporate development at Enron and a certified public accountant, became an internal whistleblower in August 2001 when she outlined her concerns about “an elaborate accounting hoax” in an anonymous memo to Enron Chairman of the Board and 14 Chief Executive Officer Kenneth Lay (Anonymous Watkins Memorandum in Brickey, 2003, p. 361). Watkins sent her memo via a new communication channel, a company suggestion box created by Lay to elicit and allay employee concerns following the unexpected resignation of Chief Executive Officer and Chief Operating Officer Jeff Skilling (Brickey, 2003). Shortly thereafter, Watkins met with Lay and recommended he disclose the accounting irregularities, issue a restatement of earnings, and try to regain the trust of investors (Brickey, 2003). She warned Lay not to use Enron’s attorneys to investigate her allegations because they had authorized the existing processes, and she urged him to appoint an outside law firm instead. Lay did not follow her advice, and the subsequent report issued by Enron’s legal counsel found no evidence of the irregularities Watkins had exposed. Her supervisor, Chief Financial Officer Andrew Fastow, retaliated against her by removing her from her office, taking away her laptop, reassigning her duties, and making plans to terminate her, an action that was never carried out (Brickey, 2003). Within a few months, however, the irregularities Watkins reported caused the company to collapse: “On October 16, 2001, Enron stunned Wall Street by announcing that it had a $618 million net loss for the third quarter and would reduce shareholder equity by $1.2 billion” (Brickey, 2003, p. 357). Wall Street may have been stunned, but employees reportedly had joked about financial irregularities at company events (Moberly, 2006). The difference between those employees and Watkins was that she reported what she saw to a higher authority. However, her reporting was later called ineffective because she reported internally to a corporate executive (Lay) rather than to the full board of directors. By appointing the company’s 15 legal firm, which had been complicit in the actions Watkins had reported, to investigate her allegations, Lay was able to prevent information about the irregularities from reaching the full board or the public (Moberly, 2006). Enron collapsed quickly at the end of 2001. Five years later, Lay was convicted of six counts of fraud and conspiracy and four counts of bank fraud, and Skilling was convicted of 18 counts of fraud and conspiracy and one count of insider trading. Lay faced a lifetime in prison at the time of his death in July 2006, but a judge later vacated the ruling on the grounds that death prevents an appeal (Krishnamurthy, 2001). Skilling’s conviction was upheld on appeal, but his 24-year sentence was ordered revisited and the possibility was raised of a new trial based on evidence favorable to him that had not been presented to his defense attorneys (Norris, 2009). WorldCom—Cynthia Cooper—2002 At about the same time Enron’s financial situation was imploding, another major corporate financial disaster was becoming public at WorldCom, the world’s largest telecommunications company (Katz & Homer, 2008, February 1). In the course of her duties as vice president of internal audit, Cynthia Cooper uncovered accounting irregularities at WorldCom. The basis for WorldCom’s problems appeared to have been the heavy debt incurred by WorldCom’s acquisition of ailing telecom company MCI. Cooper encountered resistance from Chief Financial Officer Scott Sullivan, who asked her to delay the company’s capital-expenditure audit, and from Controller David Meyers, who argued against the need for capital expenditure audits. She first reported her concerns to an external party, the corporation’s auditing firm, Arthur Andersen. It was 16 only after Arthur Andersen ignored her that she reported her findings to the audit committee of the board of directors (Katz & Homer, 2008). Cooper, like Watkins, experienced retaliation. This included berating by Sullivan, accusations that her allegations were based on personal motives, legal maneuvers by the company, and the pressures of publicity (Schaal, 2008). In the end, the board fired Sullivan and publicized the fraud. The fact that Cooper reported to the board of directors, rather than to the CEO (as Watkins had), may have led to her success (Moberly, 2006). The Sarbanes-Oxley Act of 2002 Expectations for Sarbanes-Oxley As the Enron financial disaster began unfolding, Congress took action to reassure investors that corrective actions were in place to prevent a repeat of those events (Moberly, 2008). The legal ramification of Enron’s collapse was the swift drafting of the Sarbanes- Oxley Act of 2002. Named for its two authors, Senator Paul Sarbanes (D-MD) and Representative Michael G. Oxley (R-OH), Sarbanes-Oxley was introduced to prevent the kinds of financial fraud that had toppled Enron. As Congress was working on Sarbanes- Oxley, WorldCom’s situation was becoming public: “Just six months after Enron filed the largest bankruptcy case in U.S. history, WorldCom filed an even larger one” (Brickey, 2008, p. 626). As a result of Cooper’s whistleblowing experience at WorldCom, the U.S. Senate added Section 404, which increased the independence of public company auditors, to the Sarbanes-Oxley legislation that had been passed by the U.S. House of Representatives (Katz & Homer, 2008). Section 404 gave internal auditing a higher profile and increased the number of chief auditors reporting to the audit committee of the board of directors. 17 Following the actions of Watkins and Cooper, whistleblowers were seen as an important security mechanism to help prevent future corporate wrongdoing, and Sarbanes-Oxley (SOX) was seen as the mechanism to encourage and protect future whistleblowers. Among the whistleblowing provisions included in Sarbanes-Oxley were those to encourage employees who knew about financial wrongdoing in publicly traded corporations to report that wrongdoing to appropriate entities (i.e., outside members of the audit committees of their boards of directors), as well as provisions to protect whistleblowers from retaliation. Among Sarbanes-Oxley’s supporters were Professor Robert Vaughn, who saw SOX as “the most important whistleblower protection law in the world”; Tom Devine of the Government Accountability Project, who called it “the promised land”; and Taxpayers Against Fraud, who thought SOX offered the best possible protection against future disasters such as Enron (in Moberly, 2008, p. 68). Provisions of Sarbanes-Oxley The Sarbanes-Oxley Act of 2002 contains six main provisions designed to prevent financial fraud. It creates the Public Company Accounting Oversight Board, increases the independence of public company auditors, establishes regulation of corporate governance and responsibility, increases required financial disclosure, regulates securities analyst conflicts of interest; and adds new crimes and increases penalties for existing crimes under the securities laws. Within those six categories, SOX contains provisions that have major significance for corporate whistleblowers. Broadly, SOX extends to corporate whistleblowers the protections already in place for federal whistleblowers who report internally and those who report in the course of doing their jobs. More importantly for public relations, SOX specifically encourages internal 18 whistleblowing in corporations: it is the first legislation to require publicly traded companies to maintain a communication channel for employees to report financial irregularities (Moberly, 2006). SOX mandates that boards of directors of corporations establish a communication channel that allows anonymous reporting of corporate financial wrongdoing directly to outside directors on the audit committee of the board of directors. In addition, those employees are given protection from retaliation: “Employers may not ‘discharge, demote, suspend, threaten, harass, or in any other manner discriminate’ against whistleblowers” (Moberly, 2006, p. 1126). Retaliation that harms someone who gives information about a federal crime to law enforcement is a felony (Brickey, 2003). These provisions are designed to deal with two problems that have hindered whistleblowing in the past: “(1) the corporate norm of silence, and (2) the corporate tradition of blocking and filtering employee whistleblowing” (Moberly, 2006, p. 1109). The norm of silence is the tendency not to report negative information, and blocking and filtering are management techniques used by some to prevent whistleblowers’ reports of wrongdoing from reaching higher levels in the organization (Moberly, 2006). SOX also requires corporations to adopt a code of ethics for senior financial officers, to disclose whether or not they have adopted a code of ethics, and to notify the SEC of any change in or waiver of the code of ethics (Delikat, 2008). Reality of Sarbanes-Oxley Recent research into the application of Sarbanes-Oxley indicates that the expectations of its authors and supporters have not been realized. In the first few years since its passage, there has been a low rate of reporting misconduct by corporate 19 employees and a lower rate of successful claims for retaliation than other whistleblower laws have had (Moberly, 2006). The U.S. Department of Labor’s Occupational Safety and Health Administration (OSHA) is the agency charged with overseeing the administrative process for whistleblower retaliation claims brought under Sarbanes-Oxley and 13 other federal whistleblowing statutes. Under Sarbanes-Oxley, an employee of a corporation who feels he or she has been retaliated against for reporting financial fraud must file an administrative retaliation claim with OSHA within 90 days of the occurrence. If OSHA has not resolved the claim within 180 days, the employee may file a private lawsuit in federal district court. According to OSHA, through the third quarter of 2006 whistleblowers who have filed administrative claims have had a 2% success rate (claims that were judged valid), with another 13% gaining some satisfaction through settlement. The percentage of meritorious and settled cases for Sarbanes-Oxley is slightly lower than the percentage of successful claimants for other whistleblower statutes administered by OSHA, perhaps suggesting that the ‘stronger’ whistleblower protections of Sarbanes-Oxley do not result in more protections for whistleblowers. (Moberly, 2006, p. 1128) A subsequent study found 3.6% of whistleblowers prevailing in OSHA’s administrative process, with 6.5% of those who appealed the OSHA ruling to the Department of Labor winning on appeal (Moberly, 2008). Critics argue that the high hopes some had for Sarbanes-Oxley appear not to have been realized. Some attribute the low numbers of whistleblowers to frequent retaliation, including job loss, fear of ostracism, emotional strain, and blacklisting (Rapp, 2007). A 2005 survey attributes the decrease in whistleblowing, even with increased protections under Sarbanes-Oxley, to a culture that does not support whistleblowing, to concerns 20 about the value of whistleblowing in stopping wrongdoing, and to the retaliation experienced by the one-in-eight employees who report wrongdoing (Ethics Resource Center, 2007). Whistleblowers as Defense against Fraud Given the crisis in the U.S. housing and financial industries in late 2008 and the ongoing worldwide economic crisis this has created, the role of employees as the first line of defense in preventing financial and other types of fraud could be crucial. Both the federal government and Congress believe that to be the case and have encouraged whistleblowing within organizations as a form of communicating about wrongdoing before it triggers a crisis. The significance of Sarbanes-Oxley to public relations is that it mandates that publicly traded companies maintain a communication channel for whistleblowing that allows employees to report financial wrongdoing anonymously to an outside member of the audit committee of their company’s board of directors. With public relations having responsibility for overseeing communication in organizations, the role of public relations in the whistleblowing process deserves examination. The next chapter reviews the literature on whistleblowing in both management and public relations. It outlines the extensive prior use of resource dependence perspective in management to identify the independent variables associated with whistleblowing and retaliation for whistleblowing. It also examines recent work in relationship management theory and organization-public relationships that could demonstrate a connection between incidence of retaliation and quality of employer/employee relationships. 21 CHAPTER II LITERATURE REVIEW This section presents an overview of whistleblowing research in government, management, communication, and public relations literature, beginning with initial government descriptive studies conducted following enactment of federal whistleblower protections. Federal Research on Whistleblowing The majority of research conducted on whistleblowing appears in management literature, and much of the early management literature on whistleblowing was based on longitudinal, or trend, studies conducted on federal employees in the 1980s and 1990s by the U.S. Merit Systems Protection Board (USMSPB, 1981; 1984, 1993). The USMSPB was created by the Civil Service Reform Act of 1978 (CSRA) and was responsible for enforcing the whistleblower protections provided by CSRA. It was the first government agency to survey federal employees on the incidence of whistleblowing and retaliation (USMSPB, 1981). The USMSPB administered three surveys to a total of 39,000 federal employees over a period of 12 years to identify trends in the amount of illegal and wasteful activity observed by federal employees, their reporting (or lack of reporting) of those activities, and the level of retaliation they experienced. The USMSPB’s first survey in 1980 was based on a sample of 13,000 public employees from a population of 757,000 employees in 15 major federal organizations (USMSPB, 1981). About 45% of respondents said they had seen at least one instance of illegal or wasteful activity during the past year. Of those who saw wasteful or illegal activity, only 30% reported it. The vast majority (70%) of those who saw illegal or 22 wasteful activity did not report it, either because they felt nothing would be done (53%) or because they feared reprisals (19%). Of those who reported serious abuse, 24% were not identified as the source of the report (i.e., they reported anonymously). Of the 76% who were identified, 34% said they suffered negative consequences, or retaliation. Of those reporting abuse of any type, serious or not, 20% said they had suffered retaliation. Over a third of the identified reporters believe that reporting the incident resulted in some form of “negative experience” for them…Among a broader group of employees who reported any incident of misconduct—whether the most serious or not—within the last year and who were identified as the source of the report, approximately 20% claimed to have been the victim of reprisal or the threat of reprisal. (USMSPB, 1981, p. 33) The results of the 1980 survey distinguished between those who did not observe wrongdoing (nonobservers), those who observed wrongdoing but did not report it (inactive observers), and those who observed wrongdoing and reported it, either internally (internal whistleblowers) or externally (external whistleblowers). The data also showed that organizational climate played a significant role in whistleblowing. Knowing the organization would correct the wrongdoing was the most frequently cited (81%) condition that would encourage whistleblowing. The second USMSPB survey in 1983, which was sent to employees in all executive branches of the federal government, showed significant changes. The number of employees who reported seeing illegal or wasteful activity dropped by half, to 25%. As in 1980, only 30% of those who saw illegal or wasteful activity reported it, either because they felt nothing would be done (70%) or because they feared reprisals (37%). The 1983 report stated that, of those who reported wrongdoing, 23% said they had suffered retaliation, a slight increase over 1980. 23 Results of the 1992 survey, the third of three, showed that even fewer employees (18%) said they saw illegal or wasteful activities. However, 50% of those who saw such activity reported it. This was up from 30% in 1983 and in 1980. The most significant reported difference in the 1992 survey, however, was the increase in percentage of employees reporting such activity who said they had suffered retaliation. More than a third (37%) reported retaliation in 1992: “This is significantly higher than the 24 percent of employees in the Board’s 1983 survey who said they had experienced or had been threatened with reprisal after reporting an illegal or wasteful activity” (USMSPB, 1993, p. ii ). As noted previously, the 1983 percentage of 30% applied to all federal agencies, and it was lower than the number (34%) who reported retaliation in 1980. The most common forms of retaliation in the 1992 survey were “poor performance appraisals, shunning by coworkers or management, and verbal harassment or intimidation” (USMSPB, 1993, p. ii). Following the first USMSPB survey, management scholars used that data and the data from subsequent USMSPB surveys to provide more statistically sophisticated, theory-driven analyses of whistleblowing. Whistleblowing in Management Literature Most authors agree there is no overall theory guiding research on whistleblowing, and contributing theories from a variety of literatures, including management, sociology, and social psychology, abound. However, researchers frequently have turned to a theory from management—resource dependence perspective, or resource dependence theory—to explain at least some of the variables. 24 Resource Dependence Perspective Resource dependence perspective (RDP) has one simple premise: Organizations are dependent upon their environments for resources (Pfeffer & Salancik, 2003). Those resources upon which organizations are dependent include materials, financing, information, transportation, and workforce. Those resources are controlled by other actors who make demands on the organization. Management’s handling of those demands will determine the effectiveness and, therefore, the survival of the organization (Pfeffer & Salancik, 2003). The demands of internal actors, i.e., whistleblowers, and management’s handling of their demands is the focus of much of the whistleblowing research in management literature (Miceli, et al., 2008). RDP has three conceptual bases: social context is important, strategy can be effective in reducing obstacles, and power matters more than economic efficiency (Pfeffer & Salancik, 2003). The ways in which organizations are dependent upon their environments for resources occur within a social context. The dependencies between organizations and those who hold the resources they need “are often reciprocal and sometimes indirect” (Pfeffer & Salancik, 2003, p. xii), but they occur within social structures. When businesses stop thinking only about products and customers and begin to recognize the environment within which they operate, the strategies to overcome obstacles within the environment become more important (Pfeffer & Salancik, 2003, p. xii). Those strategies include developing relationships with those with whom one’s organization has ties (or linkages). Linkages have several uses. They provide information about other organizations, provide a “channel for communication” with those 25 organizations, help develop support for one’s own organization, and provide legitimacy for one’s own organization (Pfeffer & Salancik, 2003, p. 144). RDP focuses on the power relationships between organizations and other actors within their environments, including the interest groups from which organizations draw resources internally and externally: “The idea that power was important for understanding organization, as contrasted, for instance, with rationality or efficiency, was yet another way in which resource dependence ideas represented somewhat of a shift in focus for organization studies” (Pfeffer & Salancik, 2003, p. xiii ). Power is a function of an organization’s relationships, and external relationships affect internal relationships. Those individuals or groups within organizations that provide stability, control key relationships, and acquire resources are more powerful than those that cannot perform those functions because they play a key role in helping the organization remain viable (Pfeffer & Salancik, 2003). Resource dependence perspective is not the only theory used to explain whistleblowing, but it has been used extensively in management literature. The methods used to test resource dependence perspective and to examine the phenomenon of whistleblowing have included virtually all forms of research design. Research Methods Whistleblowing is a sensitive topic for individuals and for organizations, which makes research on whistleblowing difficult to conduct (Rehg, 1998). The methods used to study whistleblowing have included field experiments, quasi-experiments eliciting projected reactions to hypothetical situations, in-depth case studies, legal case studies, longitudinal nonexperimental field surveys, and meta-analysis (Miceli, et al., 2008). 26 However, the most prevalent form of research on whistleblowing has been surveys of employees about their actual experience with whistleblowing in organizations, and most of that survey research has been based on the three surveys conducted by the USMSPB. Those three USMSPB studies considered eight specific variables, plus demographic data (Near, Ryan, & Miceli, 1995). The variables were wrongdoing (the number who observed 10 types of wrongdoing); whistleblowing (the number who reported serious wrongdoing divided by the number who observed wrongdoing); identified whistleblowing (only whistleblowers who were identified could experience retaliation); “incidence of retaliation” (the number of identified whistleblowers who reported retaliation divided by the number of whistleblowers); “comprehensiveness of retaliation” (the number of incidents); lack of support from management or co-workers; seriousness of wrongdoing (dollar value and frequency); availability and knowledge of whistleblowing channels (internal and external); and demographic data: race, gender, education, organizational tenure, job tenure, pay grade, job classification, supervisory position, and most recent performance rating (Near, et al., 1995, p. 370). The broad categories of investigation into whistleblowing have included predictors of wrongdoing, predictors of whistleblowing, and predictors of organizational response to whistleblowing (Miceli, et al., 2008). Predictors of wrongdoing have been difficult to quantify: “The variance in employees, organizations, wrongdoing reported, and recipients of complaints makes it difficult to predict when wrongdoing will occur” (Rehg, 1998, p. 121). Predictors of whistleblowing have been identified as attributes of the individual, such as personality, demographics, and job situation; attributes of the wrongdoing, such 27 as seriousness, type, power or status of the wrongdoer, and support of supervisor; and attributes of the organization, such as organizational climate or supportiveness, and industry type (Miceli, et al., 2008). Predictors of organizational response to whistleblowing have been identified as “characteristics of the whistleblower and the job situation” (Miceli, et al., 2008, p. 105); and “situational characteristics related to the wrongdoing and whistle-blowing” (Miceli, et al., 2008, p. 109), such as support, attributes of the wrongdoer, attributes of the wrongdoing (seriousness and type), whistleblowing channels, interpersonal interactions, and characteristics of the social setting, such as organizational climate or culture and “societal, country, or cultural variables” (Miceli, et al., 2008, p. 122). Empirical Findings An early management study using the USMSPB (1980) data found that organizational dependence on wrongdoing as a resource, as measured by numerous actions of wrongdoing and threats of retaliation, results in organizational resistance to change and propensity toward retaliation. The threat of retaliation increases external reporting, leading the authors to conclude: Whistleblowers went ‘inside’ when the wrongdoing was at odds with other operations and the climate was less retaliatory; they went outside when the opposite conditions were present. Thus, resource dependence theory seems to be very useful in predicting both whether and to whom one would blow the whistle. (Miceli & Near, 1985, p. 540) The study also found that whistleblowing is more likely to occur when the observer has hard evidence of wrongdoing; it increases with the seriousness of the wrongdoing; and it increases if the whistleblower is affected by the wrongdoing. This study was the first to quantify the number of communication methods used to tell employees about the 28 communication channels available for whistleblowing and to quantify employees’ awareness of those channels, but neither the number of communication channels or awareness of the channels appeared to affect the decision to blow the whistle (Miceli & Near, 1985). In a further examination of whistleblowing using resource dependence theory, Perry (1992) found organizations respond with increasing hostility as the power of the whistleblower increases; hostility toward the whistleblower increases but resolution improves when the dominant coalition is involved in the wrongdoing (possibly because hostility forces the whistleblower to report externally); the greater the change required of an organization, the less likely it is to change; and change is more likely when there is an investigation of the whistleblower's claims by an outside group (Perry, 1992). Additional research on the 1980 USMSPB data had mixed results in finding differences between whistleblowers and two other groups: those who do not observe wrongdoing and those who observe wrongdoing but do not report it. Researchers found that whistleblowers believe more strongly than nonobservers and inactive observers that whistleblowing is a positive action, but they found no greater fear of retaliation among inactive observers than among whistleblowers (Miceli & Near, 1984), leaving open the question of why employees who observe wrongdoing do or do not report it. One answer might be found in a study of private sector first-level managers who were administered in the 1981 USMSPB survey. This research found that managers are more likely to blow the whistle internally when they are confident their bosses won’t retaliate (Keenan, 1988). That finding would seem to be borne out in the following study. 29 An analysis of the 1983 USMSPB data found that individuals will blow the whistle to benefit the larger group if it benefits themselves as well (Miceli & Near, 1988). In this case, whistleblowing is more likely to occur when observers are male professionals who have been in their positions a long time, who have positive reactions to their work, have been recognized recently for good performance, and are working in larger work groups in organizations known for responsiveness to complaints (Miceli & Near, 1988). The association of perceived organizational responsiveness with whistleblowing had also been found in prior research (Miceli & Near, 1985, in Miceli & Near, 1988). The researchers proposed that organizational responsiveness can be demonstrated by publicizing “through employee newsletters or posters, how problems were solved as a result of the reports” and by making ethical behavior a part of performance reviews and compensation plans (Miceli & Near, 1988, p. 278). In another of the few studies conducted of the private sector, and the only one to use a random sample (Miceli, et al., 2008), researchers administered a survey to directors of internal auditing, a group for whom some forms of internal whistleblowing are role- proscribed (Miceli, Near, & Schwenk, 1991). In that study, “about 90% of those who observed wrongdoing reported it” (Miceli, et al., 2008, p. 23). Researchers found that observers of wrongdoing who do not report it see themselves as “relatively poor performers, and they earn somewhat lower salaries than persons who report wrongdoing” (Miceli, et al., 1991, p. 126). They also do not report wrongdoing when reporting is not required morally, when reporting is not required by the job, when other observers witnessed the event, and when they see the organization as bureaucratic. Observers of wrongdoing report it when they view the wrongdoing as harmful to co-workers or the 30 public, when it involves theft, when the wrongdoers are low-level employees, and when there are few observers (Miceli, et al., 1991). A review of whistleblowing literature in the mid-1990s created a profile of whistleblowers. Although some results conflict, it appears that whistle-blowers are older or have more service, are better educated, and are more likely to be male than inactive observers (i.e., those observers of wrongdoing who do not report it). Whistle- blowers are also more likely than inactive observers to be highly paid, have high job performance, hold supervisory or professional status, and report that they have the role responsibility to report wrongdoing and the knowledge of channels for doing so. In some studies, whistle-blowers are more satisfied and committed than inactive observers and perceive the organization to be more just. (Near & Miceli, 1996, p. 511) A recent meta-analysis that explored much of the literature on whistleblowing distinguished between surveys of actual whistleblowers and studies that involved hypotheses about possible actions, or intent, regarding whistleblowing (Mesmer-Magnus & Viswesvaran, 2005). The analysis found a distinction between reported intent to blow the whistle and actual reported whistleblowing. For example, role prescription for whistleblowing, such as is found with internal auditors and, arguably, public relations executives, predicts intent to blow the whistle, but it does not predict actual whistleblowing. This finding is unexpected, as one would expect that the organization creates and staffs internal ‘watchdog’ positions to facilitate whistleblowing on unsanctioned practices (e.g., Miceli et al., 1991b; Near and Miceli, 1996). . . .Regardless, our data suggest that having a role-related responsibility to blow the whistle may not be sufficient to ensure whistleblowing action. (Mesmer-Magnus & Viswesvaran, 2005, p. 290) The analysis found that “a small, but positive correlation may exist between role responsibility and the use of an internal reporting channel” but not with an external channel (Mesmer-Magnus & Viswesvaran, 2005, p. 286). 31 Two additional analyses of the internal auditor data measured the effectiveness of whistleblowing, which is defined as success in getting the organization to stop wrongdoing. Both used resource dependence perspective to explain the situational variables that predicted success. These included reporting an activity that co-workers saw as wrong, such as theft; not going outside the organization; not reporting to the head of the organization or the manager over the employee doing wrong; holding a position of power within the organization; not having been retaliated against; and not reporting a wrongdoing on which the organization depends, as evidenced by how long the wrongdoing had been occurring (Near & Miceli, 1990; Perry, 1992). Retaliation Several aspects of whistleblowing have captured the attention of researchers, and retaliation is high on the list, even though some argue that retaliation has been reported infrequently (Near & Miceli, 1996). Retaliation is a complex phenomenon, but understanding why it occurs is necessary to preventing its occurrence, and preventing its occurrence may be one way to encourage whistleblowing: “Whistle-blowing represents an influence attempt in which organization member(s) try to persuade other members to cease wrongdoing; sometimes they fail; sometimes they succeed; sometimes they suffer reprisal” (Rehg, et al., 2008, p. 221). Researchers have suggested that an organization has several possible responses to whistleblowing, including cessation or continuation of the wrongdoing and either rewarding the whistleblower, treating the whistleblower as before, or retaliating against the whistleblower (Miceli & Near, 1992). Four progressive stages of organizational retaliation have been identified (O'Day, 1974) and used in management literature to analyze retaliation (Parmerlee, Near, & 32 Jensen, 1982). The first stage is nullification, in which the complaint is diminished. The second stage is isolation, in which the complainant is diminished. The third stage is defamation of character, in which the complaint is resolved through character assassination of the complainant. The fourth stage is expulsion, in which the complainant is ejected from the organization (O'Day, 1974). The management literature has identified several predictors of how an organization responds to whistleblowing (Miceli & Near, 1992). These include characteristics of the whistleblower, characteristics of the wrongdoing, characteristics of the organization, and the dependence of the organization on the wrongdoing, the wrongdoer, the whistleblower, and the person who receives the complaint. Whistleblower characteristics include credibility and loyalty. Wrongdoing characteristics include type and method. Organizational characteristics include types of retaliation, lack of support from management, and seriousness of the wrongdoing (Miceli & Near, 1992). In a study of the longitudinal data from the three federal studies, researchers predicted that enactment of the Civil Service Reform Act, the federal whistleblower protection law, would reduce wrongdoing, increase whistleblowing, increase identified whistleblowing, and reduce retaliation for identified whistleblowing (Near, et al., 1995). Using resource dependence perspective, they predicted that an organization that was dependent on wrongdoing, as evidenced by the seriousness of the wrongdoing, would be reluctant to change and would retaliate against whistleblowers. They also predicted that whistleblowers who had low power in the organization would experience retaliation. What they found was that the incidence of observed wrongdoing decreased by half in each subsequent study; the number of reports of wrongdoing increased by a third 33 in each study; the number of identified whistleblowers decreased slightly in each study; and the incidents of retaliation doubled from the first study to the last. Several variables predicted retaliation in one or two of the studies, but only two variables predicted retaliation in all three studies. Those variables predicting retaliation in all three studies were lack of supervisor support and lack of management support. Using external channels predicted retaliation in the 1980 and 1982 studies, and seriousness of wrongdoing predicted retaliation in the 1980 study (Near, et al., 1995). A later review of the government data predicted that CSRA whistleblower protections would result in “(1) less wrongdoing, (2) more whistle-blowing when wrongdoing occurred, (3) more identified (vs. anonymous) whistle-blowers, and (4) less retaliation (Miceli et al., 1999). Unfortunately, results were mixed” (Near & Miceli, 2008, p. 271). What the researchers found, as predicted, was a decrease in wrongdoing and an increase in whistleblowing, but, contrary to predictions, they found no reduction in fears of retaliation and an actual increase in retaliation (Near & Miceli, 2008, p. 271). Predictors of retaliation have varied over time with different studies, but lack of supervisor and management support and seriousness of wrongdoing have predicted retaliation in several studies (Mesmer-Magnus & Viswesvaran, 2005; Miceli, et al., 2008; Near, et al., 1995; Rehg, Miceli, Near, & Van Scotter, 2004; Rehg, et al., 2008; USMSPB, 1981, 1984, 1993). Using multiple regression analysis, Near et al., 1995, found that seriousness of wrongdoing (defined as dollar amount and frequency) predicted retaliation in 1980, but not in 1983 and 1992. However, hierarchical multiple regression of the same data in 1999 found that seriousness of wrongdoing predicted retaliation in 1983 only (Miceli, Rehg, Near, & Ryan, 1999; Near & Miceli, 2008). The meta-analysis 34 of whistleblowing research (Mesmer-Magnus & Viswesvaran, 2005) found seriousness of wrongdoing and frequency of wrongdoing predicted retaliation, although the authors did not define “seriousness of wrongdoing.” Rehg (1998) found seriousness of wrongdoing, defined as frequency and dollar amount, was only partially supported as a predictor of retaliation, but Rehg later found seriousness of wrongdoing, defined as “materiality, entrenchment, harm to multiple constituents,” a predictor of retaliation (Rehg, et al., 2004, p. E5). Seriousness of wrongdoing (measured as materiality and frequency) has also has been linked to the organization’s dependence on the wrongdoing (Miceli, et al., 2008). Seriousness of wrongdoing can be gauged by its entrenchment or long-term nature, its materiality, and its perceived harm to multiple constituencies. ‘Materiality,’ as defined by accountants, is wrongdoing that is ‘sufficiently important to influence decisions made by reasonable users of financial statements,’ measured by ‘both the dollar amount and the nature of the item; (Whittington et al., 1992, p. 47). Cost and frequency of wrongdoing, a similar measure, predicted retaliation in some studies (Lee et al. 2004, Parmerlee et al. 1982), but not others (Near and Jensen 1983). (Rehg, et al., 2008, p. 225) In the 1992 federal study, those who lacked co-worker support, those who were of a race other than white, and those who had poor performance ratings were more likely to have experienced retaliation (Near, et al., 1995). Coworker support and race did not predict retaliation in some studies, but did in others (Rehg, et al., 2008). Gender was not a variable in the 1980 USMSPB survey and not reported in the final survey in 1992, and, therefore, no comparisons were drawn. Subsequent studies also found “no consistent effect of gender on retaliation” until a purposive census of employees of a military base returned a large enough response to justify the use of 35 structural equation modeling (Rehg, et al., 2008, p. 223). That study found that women were much more likely than men to experience retaliation. Researchers continue to evaluate the impact of the interaction between organizational dependence on wrongdoing, the power of the wrongdoer, and the power of the whistleblower on retaliation. A recent study found that the more an organization depends on the wrongdoing, the greater the retaliation against the whistle-blower (Rehg, et al., 2004). However, that retaliation lessens with the whistleblower’s power, which was identified in the study as supervisory status and role prescription (Rehg, et al., 2004). The study also found that an organization's dependence on wrongdoing is positively associated with the whistle-blower's use of external channels to report the wrongdoing (Rehg, et al., 2004). A recent analysis placed more emphasis on the organization’s dependence on the wrongdoing or the wrongdoer. We believe that the whistle-blower’s power must be considered in the context of the relative power of the wrongdoer and the dependence of the organization on continuation of the wrongdoing. Where the organization depends heavily on the wrongdoer or the wrongdoing itself, even a whistle-blower with high status may not have sufficient relative power to escape retaliation. (Miceli, et al., 2008, p. 104) The relationship between the whistleblower and his or her direct supervisor and the relationship between the whistleblower and management are two other variables affecting both whistleblowing and retaliation. As noted previously, lack of supervisor support and lack of top management support were the only two variables that consistently predicted retaliation across all three federal studies. Supporting this view is a study in which lack of management support and age predicted retaliation among women who filed employment- related sex discrimination complaints (Parmerlee, et al., 1982). 36 Not surprisingly, researchers have found that retaliation against the whistleblower leads to negative feelings about the relationship with the direct supervisor (Rehg, et al., 2004). However, there is evidence to suggest that a positive relationship with the direct supervisor can lessen the amount of whistleblowing: "Specifically, supervisor support seems to increase the likelihood that an observer of wrongdoing will intend to blow the whistle, but to decrease actual whistleblowing behavior (e.g., Near and Miceli, 1986)" (Mesmer-Magnus & Viswesvaran, 2005, p. 291). The researchers hypothesized that a supportive relationship with one’s supervisor might increase the intent of an observer of wrongdoing to blow the whistle because of having confidence in organizational support but that actual whistleblowing might be deterred because of unwillingness to place the supportive supervisor in an untenable position (Mesmer-Magnus & Viswesvaran, 2005). Retaliation can be formal, something inflicted by the organization officially, or informal, something inflicted below the radar by co-workers, superiors, or subordinates without official organizational sanction; or, it can be both (Rehg, 1998). Dual-mode retaliation has been found to be the most frequent type of retaliation, and in that type of retaliation, “the supervisor’s lack of support has the strongest effect of any variable” (Rehg, 1998, p. 112). This finding has implications both for the management of relationships within organizations as well as for legislation relating to whistleblower protections: “Both formal and informal types of retaliation occur, and legislation against retaliation cannot prevent informal retaliation” (Rehg, 1998, pp. 121-122). What has been apparent in many of the studies identified thus far has been the role of relationships both in the decision to report wrongdoing and in the results to the whistleblower. The next section explores whistleblowing in communication and public 37 relations research, including the contribution of relationship management theory to the discussion on whistleblowing. Whistleblowing in Communication Literature A few scholars in both business and communication have identified whistleblowing as an act of communication (Keenan, 1988; King, 1997; Richardson, 2005; Richardson & McGlynn, 2007): “Although studies have examined whistleblowing primarily from an organizational behavior perspective, the process of revealing a wrongdoing is a communication phenomenon” (King, 1997, p. 419). In fact, communication has been called a fundamental aspect of whistleblowing (Richardson, 2005). However, communication scholars have not taken up the topic of whistleblowing to the degree business scholars have done. An early call by business scholars to “examine the role of organizational communications in affecting whistle-blowing decisions” (Miceli & Near, 1985, p. 359) has gone relatively unheeded, and little research has been conducted by communication scholars on the communication aspect of whistleblowing (Richardson & McGlynn, 2007). In the few studies in either field that have addressed whistleblowing as a function of communication, aspects of interest have included upward communication, organizational climate, interpersonal closeness, and communication channels. Upward Communication Upward communication is communication that originates within lower ranks and is transmitted to higher ranks within an organization. (Downward communication, or 38 communication of bad news from higher ranks to lower ranks, is considered discipline.) Whistleblowing, by its nature, is upward communication of bad news or “bad information” (King, 1997, p. 422), and that type of communication has been found to be less likely to be reported than good news (Jablin, 1978, in King, 1997). Upward communication also has been found to have a positive effect on performance. “Research on upward communication in organizations indicates that openness, trust, and receptivity tend to have beneficial effects in terms of organizational performance although one needs to be cautious in inferring direct relationships exist between the two (Eisenberg and Witten, 1987; Jablin, 1985; Glausser, 1982)” (Keenan, 1988, p. 247). Organizational Climate Organizational climate has been found to affect whistleblowing decisions (Miceli & Near, 1985). Organizational climate is defined as organizational dependence on wrongdoing, perceived potential for retaliation to the whistleblower, and encouragement of whistleblowing through internal and external communication. Although both whistleblowers and those who saw wrongdoing but did not report it were found to have encountered the same organizational climate, those who did not report said they were not aware of channels for reporting wrongdoing (Miceli & Near, 1985). The finding that whistleblowers are more likely to work in organizations known for their responsiveness to complaints (Miceli & Near, 1988) also was found in a study of organizational communication climate. Keenan (1988) administered the 1980 USMSPB survey to first-level managers and found them more likely to blow the whistle internally when they were confident their bosses wouldn’t retaliate (Keenan, 1988). Using Gibb’s (1961) typology of supportive or defensive organizational climate, Keenan (1988) found 39 “a high supportive and low defensive organization climate is a major factor in influencing the decision of first-level managers to blow the whistle or not” (p. 250). Interpersonal Closeness Interpersonal closeness between the wrongdoer and the observer of the wrongdoing has been found to have a negative correlation with whistleblowing. An employee is less likely to report wrongdoing by someone with whom he or she has a close interpersonal relationship than wrongdoing by someone with whom he or she does not have that relationship (King, 1997). It has also been found that a positive relationship between an employee and a supervisor does not lead to more whistleblowing. There is a difference between the intent to blow the whistle and the act of whistleblowing; that difference is related to support from the would-be whistleblower’s supervisor; and that support from the supervisor equates to a greater intent to blow the whistle but to less actual whistleblowing (Mesmer-Magnus & Viswesvaran, 2005). Communication Channels Communication channels have been shown to play an important role in whistleblowing, although the research has been conducted within a management framework and not a communication one. Observers of wrongdoing who did not report the wrongdoing cited a lack of awareness of communication channels for whistleblowing as a reason for not reporting (Miceli & Near, 1985). In addressing the lack of whistleblowing in certain instances by role-prescribed whistleblowers (internal auditors), researchers recommended that organizations communicate to employees those instances in which whistleblowing is appropriate and engage employees in “participation and two- 40 way communication” with the organization in order to encourage whistleblowing (Miceli, et al., 1991, p. 126). Whistleblowing in Public Relations Literature Only recently has whistleblowing been explored in the field of public relations, and then only as one of several approaches public relations practitioners use to gain power and exert influence in their organizations through dissent (Berger, 2005; Berger & Reber, 2006; Kang & Berger, 2009). Berger and Reber (2006) report on seven studies that identify three types of resistance strategies used by public relations practitioners to gain influence in their organizations: advocacy, dissent, and activism. Sanctioned influence tactics, such as sanctioned advocacy or “Alpha strategies,” include the traditional approaches of public relations: rational arguments, coalition and alliance- building, assertiveness and persistence, inspiration and personal appeals, among others. The authors note that “advocacy is accepted or sanctioned as a form of influence so long as it is seen to be in the interests of the organization and its objectives and is institutionalized within the office or position of the advocate” (Berger & Reber, 2006, p. 56). Unsanctioned influence tactics, which they call dissent tactics or “Omega approaches,” include leaking information, starting rumors, whistleblowing, and other activist tactics. Activist tactics could encompass sanctioned and unsanctioned forms of dissent, including “compliance hot lines and governance committees…leaking information, alerting stakeholders, and whistle-blowing” (Berger & Reber, 2006, p. 162). Dissent runs the gamut from relatively harmless complaining among colleagues “to whistle-blowing, one of the most extreme forms of dissent” (Berger & Reber, 2006, p. 41 58). Although the types of activities reported by public relations professionals varied among the studies based on differences in terminology and methodology, the findings showed the majority of public relations professionals opposed using unsanctioned influence tactics. In three different studies, “fewer than one in five professionals in each study reported using such approaches” (Berger & Reber, 2006, p. 165). In a survey of more than 700 public relations practitioners, Berger and Reber (2006, p. 37) asked, “What do you think it means to ‘do the right thing’ in public relations when management is making decisions that you believe are inappropriate?” Based on responses, they identified seven types of advocacy: loyalty, self protective, rational, stakeholder, ethical, activist, and exit. Berger and Reber (2006) argue that public relations executives who know about wrongdoing in an organization could choose to act as ethical advocates who bring bad news to management directly and make a case to management for proper behavior, or they could behave as activist advocates and use one or more dissent tactics. However, their findings show that dissent is not the choice of most practitioners. They found public relations practitioners used one of three advocacy strategies: rational advocacy, 24.61%; ethical advocacy, 23.76%; and exit advocacy, 19.94%. Rational advocacy was characterized as making persuasive arguments for one’s position, ethical advocacy as basing one’s arguments on ethical codes or philosophies, and exit strategy as leaving the organization. Only 8.35% reported they would use activist tactics, including 4% who said they would blow the whistle if the activity were illegal and only 1% who said they would use hot lines and audit committees (Berger & Reber, 2006). 42 Results for the topic of whistleblowing are consistent: “The vast majority of professionals in our studies said that whistle-blowing was not the right thing to do under virtually any circumstances” (Berger & Reber, 2006, p. 166). In interviews, 73.8% said they “would never employ activist tactics such as whistle-blowing” (p. 49). That finding was tempered, however, by a small but vocal group who said activist tactics were precisely what was needed in certain circumstances. Those findings led the authors to propose that “public relations professionals hold strong but polar feelings about the appropriateness of whistle-blowing” (Berger & Reber, 2006, p. 166). A recent study upheld these findings, with more than 75% of practitioners reporting they had never leaked information (Kang & Berger, 2009, p. 15). Public Relations Roles Good public relations has been called the corporate conscience—an indispensable attribute of modern and progressive business. (Hill, J. W., 1958, in Bowen, Heath, & Lee, 2006, p. 12) Despite the apparent unwillingness of public relations practitioners to engage in whistleblowing, public relations roles research provides the foundation for viewing whistleblowing as a role-prescribed function of public relations. In public relations roles research, two of the functions that distinguish the public relations manager from the public relations technician are involvement in management decision-making and informing management about issues (Dozier, 1992). One of the longstanding goals of public relations practitioners has been to gain a seat at the “decision-making table” in order to participate in management decision-making (Berger & Reber, 2006, p. 5). One of the ways in which public relations managers and practitioners accomplish that goal is by acting as boundary spanners who scan the environment (an area defined by the 43 organization as external to organizational boundaries), identify issues of concern to management, and raise those issues within the internal dominant coalition (Aldrich & Herker, 1977, in J. White & Dozier, 1992). As part of this boundary spanner function, public relations practitioners are expected to bring bad external news to management (Berger & Reber, 2006). However, nothing precludes the public relations manager from bringing bad internal news to management, as well. Jason Vines, former vice president of corporate communications for Chrysler, appears to be referencing bringing internal bad news to management (why else would the emperor have no clothes?) when he argues for having public relations report to the chief executive officer: You almost have to be the CEO’s alter ego…Once the relationship is developed, you can almost replace one another because you’re thinking along the same lines all the time…You have to have the guts to tell the CEO when he’s naked and to be able to say it without fear of retribution. (in Cobb, 2008, p. 6) Despite the fact that public relations practitioners choose overwhelmingly not to engage in external whistleblowing, almost a quarter report using ethical advocacy to argue for correct behavior on the part of the organization (Berger & Reber, 2006). The use of ethical advocacy implies that practitioners are aware of some real or potential issue or behavior that requires them to make an ethical argument, and the act of making that argument inside the organization may fall within the definition of whistleblowing. “Whistle-blowing represents an influence attempt in which organization member(s) try to persuade other members to cease wrongdoing” (Rehg, et al., 2008, p. 221). Public relations practitioners are in a unique position to critique their organization’s actions and motives, a role that some argue they should undertake more vigorously (Bowen, et al., 2006; L. A. Grunig, Grunig, & Dozier, 2002; J. W. Hill, 1958; 44 Holtzhausen, 2000; Holtzhausen & Voto, 2002). Although polarized about the role of ethical advisor to management (Bowen, 2008), many public relations practitioners consider this one of their most important roles (Bowen, et al., 2006). The task of keeping management informed opens the door to viewing whistleblowing as a role-prescribed function of the public relations manager. From the perspective of public relations as a management role, from the perspective of public relations as boundary spanning, and from the perspective of public relations as ethical counsel, the public relations function can be viewed as encompassing role-prescribed whistleblowing, just as internal auditing does. Relationship Management The management literature holds numerous references to the role of relationships in whistleblowing and retaliation, but management theory, including resource dependence perspective, is not well-equipped to deal with relationships. In fact, a recent review of whistleblowing research from its inception by the three leading whistleblowing scholars contained only one indexed reference to relationships, a reference to differing values placed on relationships by women and men (Miceli, et al., 2008). Public relations, on the other hand, deals directly with relationships. The concept of relationships has become a major focus of public relations theory and research over the past several decades (Sallot, Lyon, Acosta-Alzuru, & Jones, 2008), and it has increased significantly over the past 10 years (Yang, 2007). Because of this focus on relationships, public relations may offer the key to understanding the role of relationships in whistleblowing through the tool public relations has developed to study relationships and their management—relationship management theory. 45 Using relationship management theory may help researchers identify relationship variables that are predictive of whistleblowing and of retaliation. For example, one would expect that having a good relationship with one’s supervisor would lead to more internal and less external whistleblowing, as well as lead to less retaliation by that supervisor. Or, conversely, one might expect that a poor relationship with one’s supervisor would lead to more whistleblowing and retaliation. The insights gained from the study of relationship management in public relations could shed some light on the sometimes conflicting results about the predictors and outcomes of whistleblowing that Miceli et al. (2008) found in their recent review of the management literature. The study of relationship management in public relations has its roots in several theoretical traditions, including systems theory, resource dependency theory, exchange theory, marketing, organizational behavior, organizational communication, interpersonal communication, psychotherapy, and social psychology (Broom, Casey, & Ritchey, 1997; J. E. Grunig & Huang, 2000; Hon & Grunig, 1999; Toth, 2000). As a result, relationships between organizations and publics, or organization-public relationships, have been characterized in different ways (Ki & Shin, 2006). This study uses the Hon and Grunig (1999) model, which employs Huang’s (1997) four relationship quality outcomes (trust, control mutuality, commitment, and satisfaction) and two relationship types, communal and exchange (J. E. Grunig & Huang, 2000; Hon & Grunig, 1999). Relationship quality outcomes are defined as trust, which is the confidence in, and the willingness to be open to, another party; satisfaction, which is the result of a cost-benefit analysis that favors the relationship; commitment, which is a willingness to spend time on the relationship; and control mutuality, which is a sense that 46 power is shared appropriately in the relationship (J. E. Grunig & Huang, 2000; Hon & Grunig, 1999). Together, “trust, control mutuality, relational satisfaction, and relational commitment have been used extensively by scholars to evaluate the quality of an organization’s relationships with its publics” (Ni, 2007, p. 54). Relationship types include exchange relationships, in which there is an expected trade of benefits, and communal relationships, in which a benefit is extended but from which no reciprocal benefit is anticipated (Hon & Grunig, 1999). Exchange relationships have been identified as the norm for human interaction that has been hard-wired by evolution (Greenwood, 2010), while communal relationships have been identified as the higher order of relationships and the type to which public relations should aspire (Hon & Grunig, 1999). In studying relationships, scholars have found differing effects for relationship quality outcomes. For example, trust and commitment were “key predictors of attitude and behavioral intentions for members of a public who have strong relationships” (Ki, 2006, p. 163), while satisfaction had the strongest effect on attitude and behavioral intentions in a study of university students (Ki & Hon, 2007). Employees’ trust in their organization was increased by positive interpersonal communication by managers (Jo & Shim, 2005), and trust and satisfaction were negatively associated with asymmetrical communication from an organization (Kim, 2007). Trust also was found to be “key to understanding the relationship development process in crisis communication” (Huang, 2008, p. 319). In attempting to explain these divergent results, scholars have noted that Hon and Grunig (1999) appear to measure both interpersonal dimensions and organizational 47 dimensions, at least when dealing with relationships between employees and organizations (Gallicano, Curtin, & Matthews, 2010; Ni, 2007, 2009). For example, in a study of Chinese employees and public relations managers, Ni (2007) found that trust has an interpersonal dimension (trust in a supervisor) and an organization dimension (trust in the organization). In a study of Millennial Generation public relations practitioners (those born in 1982 and later), Gallicano et al. (2010) found that trust, control mutuality, and commitment all may have interpersonal and organizational dimensions. In the few instances in which whistleblowing has been addressed in the public relations literature, it has been presented as a rarely used dissent tactic (Berger, 2005; Berger & Reber, 2006; Kang & Berger, 2009). The researcher contends whistleblowing should also be viewed as a component of the management role of public relations, as an element of boundary spanning, as an element of the ethical advisor and counselor role, and as a variable influencing, and being influenced by, the relationship between an employee and his or her organization. Summary of Literature Review Whistleblowing has been a significant area of research by government, management, and law for the past 30 years. The federal research found that from 1980 to 1992 the incidence of wrongdoing observed by federal employees decreased, the incidence of whistleblowing by federal employees increased, and, unfortunately, the incidence of retaliation against federal employees also increased. Management research on the federal data set and other studies on whistleblowing undertaken by the academic community have identified factors that predict several aspects of whistleblowing, such as wrongdoing by an organization, whistleblowing by 48 current or former employees, and the response of the organization to whistleblowing. Factors that have been found to predict whistleblowing include attributes of the individual, such as personality, demographics, and job situation; attributes of the wrongdoing, such as seriousness, type, power or status of the wrongdoer, and support of the supervisor; and attributes of the organization, such as organizational climate or supportiveness. Factors that predict retaliation include attributes of the whistleblower and the whistleblower’s job, attributes of the wrongdoing, and attributes of the act of whistleblowing. However, despite these findings, management literature has not been able to reach a consensus on the relative role of these variables in predicting wrongdoing, whistleblowing, and organizational response to whistleblowing. This study argues that whistleblowing should be a significant area of research in public relations, as well as management and law, because of the nature of communication, the role of public relations in organizations, and the contribution public relations has made, and will continue to make, to the study of relationships. Because public relations frequently plays a role in internal communications within organizations, the role public relations has played in developing and publicizing the SOX communication channels is an obvious research question. However, no research has been conducted on the role of public relations in developing and/or publicizing those channels, or on the use of those channels by public relations practitioners. Despite the acknowledgement in management literature of the role of relationships in whistleblowing and organizational response to whistleblowing, little research has been conducted in management literature on relationship factors related to whistleblowing, and no research has been conducted in public relations on relationship 49 factors related to whistleblowing. In an effort to rectify this gap in the whistleblowing research and to gain a better understanding of whistleblowing in general and in public relations in particular, this study combines two theories and two literary traditions, one from management, resource dependence perspective, and one from public relations, relationship management theory, in an exploratory attempt to provide greater understanding of this topic. Research Questions and Hypotheses This study surveys high ranking public relations executive in each of the Fortune 1000 corporations. Because the public relations function could be expected to have an oversight role in the communication function in most organizations, these executives could be expected to have a role in the development of the anonymous communication channel for whistleblowing mandated by Sarbanes-Oxley. This leads to the following research question: RQ1: To what extent have public relations executives of Fortune 1000 corporations been involved in developing and publicizing Sarbanes-Oxley whistleblowing channels? Because of their duties as boundary spanners, public relations executives of Fortune 1000 corporations are more likely than most employees to know about wrongdoing within their organizations. It has been found that bad news, in general, is less likely to be reported upward than good news, and federal studies on whistleblowing have found that only one-quarter to one-half of those who see wrongdoing report it. In addition, the limited amount of research on whistleblowing in public relations has shown that public relations practitioners are divided on the use of dissent tactics such as 50 whistleblowing, with three-quarters saying they would never use such tactics. These findings lead to the following research question: RQ2: To what extent have public relations executives of Fortune 1000 corporations felt it necessary to blow the whistle internally or externally? The only study of role-prescribed whistleblowers thus far, a study of internal auditors, showed that they have a higher incidence of whistleblowing than other employees. In their role as advisors to senior management, public relations executives also are in position that might be viewed as role-prescribed in terms of bringing information about wrongdoing to upper management. However, research has shown that public relations practitioners are deeply divided on the role of public relations practitioners as ethical counsel to management. Some consider it their responsibility, and others do not. These divided findings regarding public relations’ ethical role within organizations lead to the following research question: RQ3: To what extent do public relations executives of Fortune 1000 corporations consider it their responsibility to blow the whistle internally or externally? Communication channels have been found to play a major role in several aspects of whistleblowing, including the decision to blow the whistle, the choice of channels for blowing the whistle, and the impact of the channel chosen on retaliation for whistleblowing. Research has shown that those who see wrongdoing are more likely to blow the whistle if the organizational climate supports reporting of wrongdoing, including making channels available to report wrongdoing. Those who see wrongdoing but do not blow the whistle say they were not aware of whistleblowing channels. Whistleblowers use external channels to blow the whistle when the organizational 51 climate causes them to fear retaliation, and most external whistleblowers have blown the whistle internally first. These findings lead to the following research question: RQ4: What channels do public relations executives of Fortune 1000 corporations use to blow the whistle either internally or externally? Internal auditors report that they favor internal channels slightly over external channels. However, only one-quarter of public relations practitioners say they would use dissent channels that include whistleblowing, and only 1% say they would use internal channels such as hot lines and audit committees. These findings lead to the following two hypotheses: H1: Public relations executives of Fortune 1000 corporations blow the whistle internally through the chain of command more often than they blow the whistle internally through SOX channels. H2: Public relations executives of Fortune 1000 corporations blow the whistle externally more often than they blow the whistle internally through SOX channels. The amount of retaliation experienced by whistleblowers is unclear. Reports of retaliation in the federal studies ranged from 20% to 37% of those who reported whistleblowing and were identified. However, public relations professionals report that they rarely blow the whistle. Therefore, no conclusions can be drawn about the instance of retaliation for public relations whistleblowers. That lack of information leads to the following research question: RQ5: How have public relations executives of Fortune 1000 corporations who have blown the whistle been treated by their Fortune 1000 corporations? 52 Power appears to confer some protection to whistleblowers. Whistleblowers with power, as measured by supervisory status, position, gender, race, education, tenure, and job performance, report less retaliation than those without power. This leads to the following hypothesis: H3: Public relations personnel with power (as measured by supervisory status, position, gender, race, education, tenure, and job performance) who blow the whistle will report less retaliation than public relations personnel who lack power. However, an organization’s dependence on wrongdoing may reduce the protection of power for the whistleblower. Dependence on wrongdoing has been measured by seriousness of wrongdoing, defined as dollar amount and frequency, and the duration of the wrongdoing. Seriousness of wrongdoing alone has been a predictor of retaliation in several studies, including the 1980 and 1983 federal studies. A meta- analysis of whistleblowing research found seriousness of wrongdoing (undefined) and frequency of wrongdoing predicted retaliation. These findings lead to the following hypothesis: H4: Public relations executives with power who blow the whistle on serious wrongdoing (as measured by dollar amount and frequency) will report more retaliation than those who blow the whistle on less serious wrongdoing. A common predictor of retaliation throughout the management studies has been the use of external reporting channels. This leads to the following hypothesis: H5: Public relations executives who blow the whistle externally will report more retaliation than those who blow the whistle internally. 53 Various studies have demonstrated that lack of supervisor support and lack of management support predict retaliation. This leads to the following hypothesis: H6: Public relations executives who report retaliation for blowing the whistle will report greater lack of supervisor support and lack of management support than those who do not report retaliation. This study uses the Hon and Grunig (1999) scale for relationship quality outcomes (trust, control mutuality, commitment, and satisfaction) to answer research questions and hypotheses about how relationships affect whistleblowing and retaliation and to measure the relationships between public relations executives who have blown the whistle and their organizations. Furthermore, it uses the short scale version of that instrument because the completion rate for that version is higher than the completion rate for the longer instrument (Hon & Grunig, 1999). In public relations studies using this scale, trust and commitment were the strongest predictors of attitude and behavioral intention on the part of members of an organization, while satisfaction and control mutuality were the strongest predictors of attitude and behavioral intention on the part of respondents. As noted previously, asymmetrical communication, a category which could include retaliation for whistleblowing, has been shown to reduce employee trust and satisfaction, while positive communication from the supervisor increases trust in the organization. These findings lead to the following research question: RQ6: In what ways has retaliation for whistleblowing by public relations executives affected their relationships with their Fortune 1000 corporation employers? 54 In addition, management research has found that whistleblowers who do not suffer retaliation have more positive relationships with their supervisors than those who do suffer retaliation. These findings lead to the following hypothesis: H7: Public relations executives who report more retaliation for whistleblowing will report more negative relationships with their organizations (in terms of trust, control mutuality, commitment, and satisfaction) than those who report less retaliation. One additional variable has been found to predict retaliation, and that is gender. However, the sample size and gender variation in those studies will not be replicated in this study, because most high ranking public relations executives in Fortune 1000 corporations are male. This leads to the following research question: RQ7. Do female public relations executives who blow the whistle experience more retaliation than male public relations executives who blow the whistle? Because whistleblowing is a new concept and area of study in public relations, this study first attempts to gather descriptive data about the role public relations has played in developing and/or publicizing the whistleblowing channels mandated by Sarbanes-Oxley. It uses an adaptation of the federal research questionnaire to gather descriptive data about how often whistleblowing occurs in public relations, how public relations executives view whistleblowing, the channels public relations executives have used to blow the whistle, and how those public relations executives who have blown the whistle have been treated. The study uses a relationship management scale with demonstrated validity and reliability to gather data that shows how the treatment of public relations practitioners has affected their relationships with their organizations. 55 CHAPTER III METHOD Introduction For this study, the researcher distributed an e-mail invitation to answer an Internet-based survey to the highest-ranking public relations executive at each of the Fortune 1000 corporations and to two additional lists of successively lower-ranked public relations practitioners at the same corporations. The survey used in these distributions was hosted on two Web sites, SurveyMonkey and Qualtrics, and the link to the survey was delivered via an encrypted link in an e-mail. The method of delivery is significant to the results of this research and deserves explanation. Wimmer and Dominick (2006) use the terms Internet survey and e-mail survey interchangeably. They refer to Internet surveys in discussing survey response rates, but they refer to e-mail surveys as either containing a questionnaire in the body of the e-mail, containing a survey as an attachment, or containing a link to a survey on a Web site. In this discussion, the researcher refers to her research as an e-mail survey because e-mail was the delivery mechanism for the link to the survey. Survey Research Surveys are the method of choice when quantitative observations are needed in order to understand how populations behave and to test hypotheses. Surveys are “a systematic method for gathering information from (a sample of) entities for the purposes of constructing quantitative descriptors of the attributes of the larger population of which the entities are members,” (Groves et al., 2004). Sometimes surveys attempt to measure everyone in a population (a census), and sometimes they attempt to measure a sample of 56 the population (Groves, et al., 2004). Self-administered surveys are the method of choice preferred by most survey administrators due to lower cost and the ability of the non- professional to administer them (Dillman, 2000). A common goal in survey research is to achieve a confidence interval of 5% at a confidence level of 95%. This equates to being 95% confident that the results of the survey are accurate within a margin of error of plus or minus 5%. Achieving this from a population of 1,000, such as the Fortune 1000 corporations, would require a total of 278 completed surveys, or a response rate of 27.8%. Opinions on response rates to e-mail or Internet surveys differ. A 27.8% response rate is near the upper range of anticipated response rates (1%-30%) for Internet survey delivery (Wimmer & Dominick, 2006). However, others have found a 70% response rate to Internet surveys, compared with a 61% response rate to mail surveys (Kiernan, Kiernan, Oyler, & Giles, 2005). The researcher determined that an e-mail survey was the method most likely to achieve the required response rate from her target population, Fortune 1000 public relations executives, at the least cost and in the most expeditious manner because of the sensitive nature of the survey topic and the busy schedules of the executives under study. Survey Error Surveys are subject to several types of errors, among them coverage error, sampling error, construct validity, measurement error, and nonresponse error (Groves, et al., 2004). Coverage Error Coverage error is the difference between the target population (an abstract concept) and the frame population or “sampling frame…a listing of all units (e.g., people, 57 employers) in the target population” (Groves, et al., 2004, p. 45). Coverage error results from not including all possible elements of the population in the grouping, or list, from which the sample is drawn: “An example would be the omission of people without telephones from a telephone survey” (Dillman, 2000, p. 9). The researcher intended to reduce coverage error by using a census of the ranking Fortune 1000 public relations executives, rather than a random sample. She also intended to use the list maintained by Money Magazine that was available through the University of Oregon library when she first designed her study. Unfortunately, that list was not available through library sources when she began her research. In response, she developed her own list of the ranking public relations executives in the Fortune 1000 to replace the Money Magazine list, which may have resulted in increased coverage error. The process of constructing this list is discussed in detail below. Construct Validity According to Groves et al., (2004, p. 41), “constructs are the elements of information that are sought by the researcher.” Construct validity is the extent to which a research instrument measures what it intends to measure: “A valid measure of an underlying construct is one that is perfectly correlated to the construct” (Groves, et al., 2004, p. 51). The researcher addressed construct validity by using and adapting survey instruments that had been tested repeatedly for validity. These instruments included the three surveys that the United States Merit System Protection Board (USMSPB) delivered to virtually all federal employees in 1980, 1983, and 1992. Instruments also included the Hon and Grunig (1999) relationship measurement scale, which had been validated 58 extensively in communication research (Ki, 2006). The researcher also validated the Hon and Grunig (1999) relationship scale herself through the use of Cronbach’s alpha, which measures the internal reliability, or consistency, of a scale (Pallant, 2005). Measurement Error Measurement error is the difference between the actual value for the item being measured and the measurement taken. Measurement error occurs as a result of bias or variance. Human factors, such as the unwillingness to report negative behaviors, introduce bias, while the vagaries of memory introduce variance (Groves, et al., 2004):“Common method bias is always a consideration when surveys are used,” (Miceli, et al., 1999, p. 147). The researcher considered the possibility that subjects in this study would not want to report what might appear as negative behavior (whistleblowing) or might not remember acts of wrongdoing or acts of whistleblowing. However, the researcher felt that the significance of the topic to individuals, organizations, and society made the effort to elicit this information worth undertaking. Nonresponse Error For a variety of reasons, some individuals will not complete a survey, which may result in the characteristics of those who do respond differing from the characteristics of the entire population. Nonresponse error occurs when those who respond to the survey differ from those who do not respond in a way that is important to the research (Dillman, 2000). Nonresponse error, as well as the resulting bias of the results, is a major concern of survey researchers; however, high response rates can reduce the risk of nonresponse error (Groves, et al., 2004). 59 The researcher attempted to address nonresponse error by increasing response rate. She followed the Dillman (2000) prescription of five contacts for optimal response as closely as the University of Oregon Institutional Review Board (IRB) guidelines allowed. These guidelines allowed the researcher to send three e-mail contacts, one initial invitation and two follow-up reminders, to the entire list of the highest-ranking public relations executives in the Fortune 1000 corporations and to the two subsequent Fortune 1000 lists that she developed. Survey Instrument The researcher developed the survey instrument by adapting questions from the three USMSPB whistleblowing surveys and by incorporating the shorter version of the Hon and Grunig (1999) relationship measurement scale. In consideration of the respondents’ time and in hopes of increasing the response rate, the researcher eliminated questions from the federal surveys that were not used in the analysis of retaliation and eliminated the 12-month limitation on awareness of wrongdoing. She also changed the phrasing of the questions to reflect a corporate, rather than a federal government, focus and to reflect the passage of the Sarbanes-Oxley Act of 2002 with its mandated anonymous whistleblowing communication channels. The researcher intended these modifications to allow her to conduct the kind of empirical examination of employees of private firms called for by management researchers (Near & Miceli, 1986), as well as allow comparisons with more recent analyses of all three federal studies. 60 Survey Components Section I Section I gathered descriptive data on organizations’ compliance with the Sarbanes-Oxley mandate to develop an anonymous communication channel to the audit committee of the board of directors for reporting financial fraud, public relation’s role in developing and publicizing that channel, and the channels used to publicize it. Section II Section II was adapted from the federal surveys. It asked respondents to focus on whistleblowing activities by public relations executives other than themselves. The purpose of this section was to capture information about whistleblowing activities by public relations executives in a non-threatening manner. It allowed respondents to report on such activities without reporting on their own whistleblowing activities. The intent was to increase knowledge about whistleblowing in public relations. Section III Section III mirrored Section II but asked respondents to report on their own whistleblowing activities in their current organization. Section IV Section IV asked respondents to characterize their relationship with their current organization. This study followed Gallicano et al. (2010), Hall (2006), Ki and Hon (2007), Kim (2007), and Yang (2007) in using the Hon and Grunig (1999) scale to measure organization-public relationship outcomes. Furthermore, this study used the Hon and Grunig (1999) short scale of 26 questions. Other versions included the short plus one scale (30 questions), and the full scale (46 questions). However, the short scale was 61 “almost as reliable as the full scale” (Hon & Grunig, 1999, p. 27), and use of the short scale was “likely to increase the completion rate” (Hon & Grunig, 1999, p. 28). The short scale contained in this study is shown in Table 1 and Table 2. The scale contains 18 questions on relationship quality outcomes and eight questions on relationship type. Six items measure trust, and four items each measure control mutuality, commitment; satisfaction; exchange relationship; and communal relationship. Question 34, which measures control mutuality, and Questions 25, 37, and 40, which measure communal relationships, are reverse coded. The questions are presented in the random order used by Ki (2006). Although Hon and Grunig (1999) used a 9-point Likert scale, Yang’s (2007) pretest determined that his subjects had difficulty with the 9-point Likert scale. Therefore, this study follows Yang (2007) and Gallicano et al. (2010) in using a 5- point Likert scale. 62 Table 1. Relationship Quality Outcome Measurement Items Measurement Question Trust 24. This organization treats me fairly and justly. (Integrity) 28. Whenever this organization makes an important decision, I know it will be concerned about me. (Integrity) 35. This organization can be relied on to keep its promises. (Dependability) 38. I believe that this organization takes my opinions into account when making decisions. 44. I feel very confident about this organization’s skills. (Competence) 47. This organization has the ability to accomplish what it says it will do. (Competence) Control Mutuality 27. This organization believes my opinions are legitimate. 34. In dealing with me, this organization has a tendency to throw its weight around. (Reversed) 39. This organization really listens to what I have to say. 48. This organization and I are attentive to what each other say. Commitment 23. I feel that this organization is trying to maintain a long-term commitment to me. 26. There is a long-lasting bond between this organization and me. 33. I can see that this organization wants to maintain a relationship with me. 45. Compared to other organizations, I value my relationship with this organization more. Satisfaction 30. Both the organization and I benefit from the relationship. 36. Most people like me are happy in their interactions with this organization. 40. I am happy with this organization. 42. Generally speaking, I am pleased with the relationship this organization has established with me. Table 2. Relationship Types Measurement Items Measurement Question Communal 25. This organization does not especially enjoy giving others aid. (Reversed) 31. This organization is very concerned about my welfare. 37. I feel that this organization takes advantage of people who are vulnerable. (Reversed) 41. I think that this organization succeeds by stepping on other people. (Reversed) Exchange 29. Whenever this organization gives or offers something to me, it generally expects something in return. 32. Even though I have had a relationship with this organization for a long time, it still expects something in return whenever it offers me a favor. 43. This organization will compromise with me when it knows that it will gain something. 46. This organization takes care of people who are likely to reward the organization. 63 Section V Section V asked for demographic information: gender, race, professional job status, supervisory responsibility, organizational tenure, job tenure, age, education, pay, and job performance. Some of this information has been shown in management studies to act as a predictor of whistleblowing or retaliation. The Measures Two units of analysis were used for this study: the individual and the firm, or organization. The individual of interest in this study was the highest ranking public relations executive who could be reached in each corporation. Although management research frequently uses the firm as the unit of analysis, the researcher agreed with those who argued that the unit of analysis should be the manager because managers make decisions within corporations or firms (Scott & Lane, 2000). This study equated the public relations executive with Scott and Lane’s (2000) manager, a decision-making individual within the firm or organization. However, the practice of whistleblowing research in management literature traditionally has analyzed four variables at the level of the firm, or organization (Miceli, et al., 1999; Near, et al., 1995). This involves identifying the total number of responses in each category mentioned below and dividing that number by the number of respondents in each category to arrive at an average: Incidence of Wrongdoing This is the measure of the occurrence of wrongdoing in the environment; the number is calculated by dividing the number of respondents who knew of wrongdoing themselves, or through someone else, by the total number of respondents; 64 Incidence of Whistleblowing This is the measure of the occurrence of whistleblowing in the environment; the number is calculated by dividing the number of respondents who reported wrongdoing in any of the ways listed on the questionnaire (see Appendix A) by the number of respondents who indicated they knew of wrongdoing; Incidence of Identified Whistleblowing This is the measure of the occurrence of identified whistleblowing in the environment (only identified whistleblowers are considered subject to retaliation); the number is calculated by dividing the number of whistleblowers who indicate they were identified as the source of the information by the total number of whistleblowers; and Incidence of Retaliation This is the measure of the occurrence of retaliation in the environment; the number is calculated by dividing the number of whistleblowers who report retaliation by the total number of whistleblowers. Other variables were calculated at the individual level: comprehensiveness of retaliation was the total number of types of retaliation the person experience; role- prescribed whistleblowing was a measure of whether the person thought his or her job duties required reporting wrongdoing; lack of support measured lack of support from one’s coworkers, supervisor, or someone above one’s supervisor; seriousness of wrongdoing measured the dollar amount or the frequency of the wrongdoing; whistleblowing channel measured whether the whistleblower reported to someone inside his or her corporation or to someone in the external environment; and demographic data. 65 The Pilot Study The researcher asked seven public relations practitioners to review the survey instrument, recruitment consent e-mail, and permission/release form and to provide feedback. She selected participants for the pilot study based on her relationship with them and their experience in public relations but not on any known or perceived experience they might have had with her study topic of whistleblowing. Three public relations professionals and two Ph.D. students in the School of Journalism and Communication at the University of Oregon participated in the pilot study. Two of the public relations professionals were members of the Public Relations Society of America’s College of Fellows, an organization that requires 25 years of experience in public relations for membership, and one was the highest-ranking communications/public affairs officer in a state government agency. The two Ph.D. students had several years of public relations experience each, and both were focusing on public relations for their dissertations. The pilot study did not include public relations executives in Fortune 1000 corporations, and members of the pilot study were not eligible to receive the survey in an official capacity. No data were collected from the pilot study. Changes Made to Instruments The researcher used comments from the pilot study to refine the final survey instrument, recruitment/consent email, and permission release form. She evaluated all recommendations made by pilot study members and made several changes based on their 66 recommendations. These included punctuation and formatting changes as well as substantive recommendations. The researcher changed the range for the dollar value of possible wrongdoing in questions 6 and 15 based on a suggestion from a pilot study member. One member felt that the dollar range, “$1,000-$100,000,” was too broad based on her experience in the regional office of a major Fortune 1000 corporation. Her feeling was that items in the $10,000-$20,000 range would have been within the signatory authority of her department, but that anything more than $50,000 would have required multiple signatures and, therefore, would have been subject to numerous layers of oversight that would have minimized abuse. Although the answer option of “$1,000-$100,000,” had been used consistently throughout the USMSPB surveys, the researcher acknowledged the validity of the arguments raised and changed the response option in both questions to “$1,000- $49,999,” and “$50,000-$99,999” to address the issue raised by the member of the pilot study. Recommendations Related to Federal Studies In some cases, the researcher agreed with the suggestions made by members of the pilot study but chose not to make the changes because doing so would compromise her ability to make comparisons with other studies in the whistleblowing research tradition. For instance, members of the pilot study raised questions about language in the questionnaire that was taken from the federal studies. In all cases, the researcher chose to retain the federal language in order to provide a basis for comparison of data with the federal studies. 67 One pilot study member suggested adding a question that would have elicited non-monetary wrongdoing. Although subsequent open-ended responses from survey participants referenced non-monetary wrongdoing, the researcher chose not to add the question to the survey. In retrospect, the addition of “N/A” to questions 6 and 15, which asked the respondent to report on the dollar amount of the wrongdoing activity, would have resolved that concern without additional length and will be added to any future distribution of the survey instrument. Recommendations Related to Relationship Scale Several pilot study members also questioned wording in the Hon and Grunig (1999) relationship scale. Although professionals who had not been exposed to the instrument might have been expected to have the most concerns, in reality, both professionals and academics reported problems with the instrument. The researcher chose not to reword the relationship scale for this study because she felt it would compromise comparison with prior public relations research; however, she decided that in future research she will modify the statements that caused her pilot study members the most problem and measure what, if any, impact those changes have on the instrument’s internal validity through the application of Chronbach’s alpha. Population The target population for this project was the highest-ranking public relations executive in each of the companies on CCN Money Magazine’s 2009 Fortune 1000 list (CNNMoney, 2010). Early in her study, the researcher was told by a representative of the library that the names of the corporate communications contacts (i.e., the ranking public relations 68 executives) in the Fortune 1000 companies and at least some contact information were available in a database, ReferenceUSA, through the University of Oregon library. The only piece of information that was in question at the time the researcher first inquired was the availability of e-mail addresses. However, in the interim between her initial contact with the librarian and the initiation of her research, ReferenceUSA significantly changed the information it made available to the library. When she began her research, the researcher discovered that the names and contact information she needed were no longer available through that database. The only information available was a limited number of executive names, titles, addresses, and phone numbers, primarily those of chief executive officers but not public relations executives. According to a second librarian, who had contacted the database representative, even that information was not considered accurate by ReferenceUSA. The database representative indicated email addresses for the list were available through Money Magazine for $1,800, but the representative advised the librarian against purchasing them because of the poor quality of the list. These changes in the information available to the researcher through the library required her to reevaluate her approach and attempt to find contact information for her target population through other sources. Developing the Mailing List: List 1 Commercial Marketing Databases The researcher confirmed that the 2009 Fortune 1000 list of sanctioned executive contacts and contact information was available for purchase from Money Magazine for $1,800 (http://www.fortunedatastore.com/). However, that list did not include e-mail addresses. Those were available for an additional fee but only for mailings sent to the 69 individual by CNN Money Magazine. The use of the sanctioned Fortune 1000 list was not an option for the researcher because of the cost and because the distribution method approved by the University of Oregon’s Institutional Review Board (IRB) did not allow for distribution of the survey by a third party. The researcher rejected other commercial databases in part because of cost but also because she could not validate that the e-mail addresses provided for a fee were authentic e-mail addresses for the correct individuals in the target corporations. Mechanical Turk The researcher contacted an academic database systems researcher (A. Pavlo, personal communication, November 2009) who offered to write a computer program to extract information on the researcher’s target audience with augmentation by workers employed by Amazon Mechanical Turk (MTurk). As an option, he proposed she purchase a list from another commercial service. She decided not to pursue either because of concerns about the validity of MTurk’s work output and commercial lists. She believed that e-mail addresses supplied by unknown workers being paid pennies per address were not inherently more accurate than e-mail addresses purchased from commercial sources. In both cases, she would be relying on unknown third parties to identify the population for her survey, and she did not have confidence that the population would be identified correctly. Mint Global After trying to identify other sources for the list of ranking public relations executives in the Fortune 1000 corporations, the researcher concluded that the only way to reach her target population was to build the list herself. The only source of that 70 information available to the researcher through non-fee-based avenues was Mint Global, an international business database headquartered in Belgium, which provided widespread search capabilities and extensive contact information. The database was searchable using a variety of criteria, including revenue information (useful in screening the list to approximate the Fortune 1000), corporate identifier (useful in differentiating target corporations from subsidiaries and corporations with similar-sounding names), name and title, position information, personal information, and e-mail address. Furthermore, the information was available for export to her home computer for development of a customized database for purposes of e-mail research. In addition, by virtue of the library’s $15,000-a-year subscription, she had access to Mint Global corporate contacts both in the United States and in Belgium who could help her design database queries that would extract the maximum amount of data available. Querying the Database The researcher queried the Mint Global database multiple times to identify public relations, public affairs, corporate communications, marketing and advertising executives of publicly and privately traded companies incorporated in the United States with a minimum of $1.7 billion in annual revenue. Over time, based on the differing response these queries generated, she broadened those queries to include any job titles containing the words relations (public and investor), affairs (corporate and public), communications (corporate or stand-alone), marketing, and advertising. She designed her queries to extract files containing specific information in a standardized order: individual’s last name, company name, company identifier, job title, standardized position (position level within the corporation), department, full name, middle initial, first name, sex, age, 71 biographical information (provided for some but not all executives), latest operating revenue, compensation (rarely reported), and work e-mail addresses. By standardizing the query, she was able to incorporate information from multiple searches in Mint Global into an Excel spreadsheet on her personal computer to build the master list for her target population. Refining the Search Criteria Searching the Mint Global database was initially a matter of trial and error as the researcher attempted to elicit the correct job title and names from the hundreds of thousands of names in the database. Throughout this process, the researcher was required to use her professional knowledge of corporate structure and job titles to identify the most likely candidate. She refined the search criteria over time, reviewing and revising her results as appropriate. The initial criteria she used for selection of the target population incorporated the following: • First choice was an executive officer with stated public relations responsibilities. This information was sometimes identifiable in the biographical information; • Second choice was a vice president of public relations, public affairs, communication or corporate affairs; and • Third choice was a vice president of marketing or advertising. Web Sites When it was not possible to identify by title the person responsible for public relations among those listed in the database, or when the database contained no information on a particular corporation, the researcher accessed corporate Web sites and 72 cross-referenced names and titles against those in the database. In the course of the study, the researcher accessed more than 500 of the 1000 corporate Web sites representing her target population. In reviewing Web sites, she accessed hyperlinks on the Fortune 1000 list that connected her to the Fortune 1000 Web site, where she identified industry information and collected it on a spreadsheet. She then linked to the individual company Web site where she reviewed the home page and information contained under any of several headings, including “About Us,” “Company Information,” “Investor Relations,” “Media Relations,” and/or “Contact Us.” She typically found management information with biographies and job duties under “About Us,” “Company Information,” or “Investor Relations.” This information allowed her to validate the names contained in Mint Global and, when necessary, identify the decision-maker among the many names listed in the database. However, corporate Web sites rarely contained contact information for corporate officers or senior management. In the few cases in which corporate officers or senior management contacts were given, one or more executive officers (chief executive officer, chief financial officer, or investor relations officer) were named as contacts on news releases several layers down on the Web site. In those cases, the researcher used the contact information available for the investor relations officer. In most cases, contact information was available only for media representatives listed on the Web site under “Media Relations” or “Contact Us,” and those media contacts were rarely the individuals listed as executives on Mint Global. In those cases, the researcher used the contact information for the highest level position that appeared to have public relations 73 responsibilities. In a few cases, no contact information for any individual was listed on the Web site, and the public was invited to request information by responding to an anonymous e-mail address (“info@companyname.com”) or by filling out an online information request form. In a few cases, the researcher sent the survey to the anonymous e-mail address; she did not fill out any online information request forms because that was outside the parameters of her survey delivery process. Reducing Sampling Error Through the iterative, highly computerized, and Internet-enabled process described in “Building the Database,” the researcher developed a list that represented her best understanding of those individuals responsible for public relations in the May 2009 Fortune 1000 list of corporations. This list of names comprised, to the best of her understanding, the target population she had identified for her research: the ranking public relations executives in the Fortune 1000 corporations. Because the population was identified by the researcher, rather than by the corporations under study, the study is subject to increased sampling error. The population identified by the researcher may not have been the population intended. At the suggestion of a colleague in the business school, the researcher added two questions to the survey to help identify whether or not she had reached the target population. The survey already contained the question: “Are you the highest ranking public relations executive in your organization?” The researcher added: “What is the location of your position in relation to the Chief Executive Officer?” and “In what department do you work?” Answer options to the first question included the following: (1) I am the Chief 74 Executive Officer; (2) I report to the Chief Executive Officer; (3) I report to someone who reports to the Chief Executive Officer; (4) I report to someone who reports to someone who reports to the Chief Executive Officer; or (5) Other.” Answer options to the second question identified the location of the position based on the categories found in the Mint Global database. These were as follows: (1) Public Relations; (2) Corporate Communications; (3) Marketing; (4) Sales; (5) Finance & Administration; and (6) Other. The two additional questions allowed the researcher to reduce sampling error by identifying those among the respondents who were within her target population and correlating their role with their placement and location in the organization. Survey Administration The researcher’s original plan was to deliver a self-administered survey to her target population, the highest-ranking public relations executive at each of the Fortune 1000 corporations, via an e-mail containing an encrypted link to a secure, commercial, Web-based, survey-hosting service, SurveyMonkey. Following the protocol for the University of Oregon’s Office for Protection of Human Subjects, the researcher intended to make three e-mail contacts with the subjects. The researcher planned to send each e- mail contact to the entire list because she was ensuring confidentiality by collecting anonymous data, and, consequently, would not know who had responded and who had not. The first contact was a recruitment/consent e-mail containing an encrypted link to the survey on SurveyMonkey. This e-mail included an introduction to the researcher, an explanation of the purpose of the survey, a description of the benefit of the survey to the 75 academic and professional public relations communities, instructions for accessing the survey via an encrypted link to the Web-based survey, a commitment to protect the confidentiality of the data, and a recommendation to complete the survey away from the business environment if confidentiality were a concern. The first e-mail also included, as an attachment, a permission/release form that allowed subjects to agree to be contacted about possible future research. The second contact was to be a follow-up/reminder e-mail to be delivered to the entire subject list one week after the original invitation. It repeated the purpose of the study, the confidentiality disclaimer, and the encrypted link to the secure Web site. The third contact was to be an identical e-mail delivered to the entire subject list two weeks after the original. The researcher had intended to use SurveyMonkey’s Email Invitation Collector to distribute the e-mails because she believed the SurveyMonkey brand would increase the response rate. However, when she learned that SurveyMonkey’s distribution function, or collector, did not allow for collection of anonymous data, she developed an alternative survey delivery mechanism that involved assistance from the University of Oregon’s Listmaster and the use of the university’s list serve. Survey Delivery: List 1 The researcher sent the UO Listmaster a list of e-mail addresses for the public relations executives she had identified for the 999 companies on the May 2009 Fortune 1000 list. (Two of the original corporations had merged in 2009.) The Listmaster created a list, titled “fortune 1000,” for the researcher and populated it with those names. The researcher pasted the text of her recruitment/consent e-mail into the body of an e-mal 76 from her personal University of Oregon e-mail account and created a subject line. The researcher sent the invitational e-mail to the list serve. The list serve automatically distributed the e-mail to the target population as though it had come from the researcher’s own University of Oregon e-mail account. The return address on the e- mail was the researcher’s University of Oregon e-mail address. Returned or undeliverable e-mails were returned to the researcher’s university e-mail account and forwarded to her personal, commercial e-mail account. The subject line of the invitational e-mail was “Fortune 1000: Sarbanes-Oxley and Whistleblowing.” The invitational e-mail was 30 lines long. The last paragraph of the invitation read: As the ranking public relations executive in your corporation, your participation in this survey is of the utmost importance to research on this topic. You will be doing the profession of public relations and the teaching of public relations a great service by participating in this research. The researcher received 110 bounce backs, or e-mails identified as undeliverable, within two hours, for what she initially thought was an undeliverable rate of 11%. However, one of those e-mails was actually an aggregator e-mail from the university’s Mailer-Daemon that contained 131 bounce backs. The additional bounce backs increased the total number of bounce backs received within the first two hours to 241, or a percentage of 24%. Within twelve hours on the day of delivery, the researcher had received five responses to the survey. She also had received two e-mail messages from respondents who declined to participate on the basis of company policy against participating in surveys. One of those was from an international communications company and the other from an oil and gas company. Although the researcher asked both companies to 77 reconsider, the energy company declined and the communications company did not respond. The researcher identified both companies on her Master List as not available for further distribution. The researcher continued to receive notification of undeliverable e-mails for several days. The total number of bounce backs from this distribution was 278. The five responses the researcher received in the first twelve hours after distribution were the total number of response she received for that distribution. The timing of those responses validated the marketing concept that most marketing e-mails will be opened within 24 hours (Cangialosi, 2010). Corrective Actions Following the initial distribution of the survey, the researcher worked to increase the response rate on the next survey distribution. Her primary goal was to identify alternate contacts for the bounce backs. With this in mind, she ran new queries in Mint Global; reviewed corporate Web sites, blogs, newsletters, news releases, and annual reports; and accessed other online media sources, such as O'Dwyer's, PR Daily Newsfeed, the Arthur W. Page Society Page Turner Blog, and many others . She also implemented a variety of avenues to publicize the survey and redefine her target population. Bounce Backs The researcher developed a structured process to identify alternate contacts for, or clean, contacts on the bounce back list. Bounce back cleaning involved the following activities: • The list serve delivered e-mail notifications of “Undeliverable” e-mail invitations, or bounce backs, from her first distribution of List 1 to her 78 personal e-mail account; • The researcher compared the bounce backs with her Master List of Fortune 1000 public relations executives; • She downloaded new Mint Global lists for executive titles containing the words communications, relations, corporate, and affairs; • She checked names on the bounce back list against the new Mint Global lists; • If the previous name was on the new list, she attempted to confirm the e- mail address; • If the original contact was on the new list with the same e-mail address, she identified the original contact as “Confirmed, Treat as New,” and moved the original contact to a new “1st Contact List” with the assumption that the Permission/Release Form attachment could have caused the bounce back; • If the e-mail address for the name on the old list was not correct, she corrected the e-mail address, identified the original contact as “Corrected, Treat as New,” and moved the original contact to the new “1st Contact List;” • If she was unable to find the name on the new list, and there was no substitute name on the new list, she identified the original contact as “Retained, No Sub, Treat as New,” and moved the original contact to the new “1st Contact List;” 79 • If the original name was not on the new list, but a new, plausible name was on the new list, she substituted the new name and e-mail address and moved the contact to the new “1st Contact List;” and • She moved all contacts that had not bounced back onto a new “2nd Contact List,” which she later used as respondents for her second notice. List Master Concurrent with developing the process for bounce back cleaning, the researcher contacted the list manager for the University of Oregon list serve, who had helped her with her e-mail distribution, to inquire about possible reasons for the bounce backs. His response was that restrictions and filtering established by each company’s Internet Service Provider (ISP) could account for the bounce backs: “That’s not something we can modify. I would believe the rest were delivered in some fashion…that’s also not something we can verify one way or another” (S. A. Smith, personal communication, March 4, 2010). In response to suggestions for subject lines that would increase response rates, the list manager had a few suggestions. He noted the researcher’s subject line was intriguing and indicative of her topic, but he also noted the impact of spam e-mail on responses: The thing is, we are all of us sensitized to spam email, and if the subject line looks odd or we don't recognize the sender, we filter it out ourselves. I know that I do that; and I'd expect that 80% of the Internet does it by now. (S. A. Smith, personal communication, March 4, 2010) His primary suggestion was to find someone known to the recipient to introduce the researcher and her topic, but he acknowledged the difficulty this presented. He also suggested she consider an opt-in option: “Email response is a problem, but opt-in is a 80 solution. People care more deeply about things they choose (or, people choose to opt in to things they care deeply about)” (S. A. Smith, personal communication, April 23, 2010). Commercial e-mail marketers identify deliverability as one of the key areas of concern. Marketers have no way of knowing how any recipient’s e-mail system is configured: “…it could be just a straight pipe going in, or it could go through a third party to filter out questionable or unknown senders and may even go through a second internal system to remove any blacklisted senders” (Cangialosi, 2010, p. 19). Aggregators, Newsletters, and Other Online Sources To stay attuned to electronic media coverage of Fortune 1000 corporations that might contain information about corporate public relations executives, the researcher established NetVibes on her computer desktop to collect articles of interest from the Internet and frequently accessed electronic newspapers (The New York Times), newsletters (O’Dwyer), and blogs (Romenesko). While occasionally providing new insights into corporate management changes, the maintenance of this strategy proved too time-consuming, and she eventually discarded it as a primary approach. Social Media Acting on the list manager’s suggestion, the researcher reached out to her contacts in public relations in several ways: through LinkedIn’s PR Group, the Public Relations Society of America’s Educators Academy E-Group, and the Association for Education in Journalism and Mass Communication list serve. LinkedIn PR News Group Already a member of LinkedIn, the researcher joined the LinkedIn PR News group and posted a message requesting suggestions for reaching her target audience. The 81 one LinkedIn member who replied suggested she work with PRSA national and individual chapters. When she explained PRSA’s constraints on research, he suggested approaching the Arthur W. Page Society, or using LinkedIn’s commercial services to buy an e-mail list. The researcher reviewed LinkIn’s pricing structure, which included e-mails sent by LinkedIn to other members for 10 per month for $49.99 and 50 per month for $499. Her concern with this approach was that neither option would satisfy the IRB requirement for control over research, and the larger mailing size was not cost-effective. She chose not to pursue LinkedIn as a survey delivery mechanism. Public Relations Society of America Educators Academy E-Group During this time-frame, the researcher also posted a request for assistance with her research on the Public Relations Society of America’s Educators Academy E-Group Twenty-eight members of the Educators Academy viewed the posting, and three responded. Recommendations included using Dillman (2000), ensuring the survey instrument was free of errors, sending multiple reminders, soliciting a third-party endorsement or introduction, and consulting the American Association of Public Opinion Research (AAPOR) for information about expected return rates for e-mailed surveys. The researcher responded that she was using Dillman, that her university limited researchers to three contacts, and that she would consult AAPOR. She also requested and received information from one of the respondents about PRSA’s procedure for requesting access to their membership list for research purposes, which she was unable to find on the PRSA Web site. One of the respondents also suggested reducing the size of the study from the Fortune 1000 to the Fortune 100 and consulting the Arthur W. Page Society as an 82 alternate population for her study. Public Relations Society of America Membership List The researcher reviewed the application process for using PRSA’s membership list for research purposes, but she decided not to apply. She concluded that the PRSA research process was not compatible with her dissertation research because of conflicting requirements. The review process required approval by the PRSA Academic Research Task force and the PRSA board, and PRSA staff were responsible for pulling samples and sending e-mails. The PRSA application required inclusion of the IRB consent form, but the IRB process required prior delineation of sample selection processes and distribution before IRB approval was granted. She also had no way of know if the PRSA membership list would include significant members of the Fortune 1000 target population. National Whistleblower Organizations The researcher contacted the communications director for a national whistleblower organization and asked for his help in encouraging public relations executives in the Fortune 1000 to participate in her research project on whistleblowing. He responded that he disagreed with the premise that public relations executives were responsible for external and internal communication. He argued that human resources was the internal entity responsible for internal communication, and, therefore, the group to whom questions about Sarbanes-Oxley communication channels should be directed. After the researcher shared her perspective on whistleblowing as a communication function, he clarified his position to state that public relations should be involved in developing Sarbanes-Oxley-compliant channels to some degree, but that human 83 resources, legal, and executive management were the departments that shared responsibility for developing the channels. He offered no help with her current project but expressed interest in one of her proposed projects, interviews with self-identified government whistleblowers. The researcher also contacted an attorney affiliated with a second national whistleblower organization for assistance. He replied that he had little contact with corporate CEOs outside of legal actions against them and that a population of 999 was too small for him to influence. He also suggested contacting PRSA for assistance. The researcher contacted a member of the PRSA College of Fellows who was an educator and who had written a book on her target population for assistance. He, too, recommended requesting support from PRSA and the Arthur W. Page Society. He also suggested targeting the survey to the next level down in the organizational public relations hierarchy. Arthur W. Page Society For additional help in cross-checking titles, the researcher accessed specific public relations communications, including the Web site, annual reports, and blog, of the Arthur W. Page Society. The Arthur W. Page Society had been on her list of organizations to contact from the early stages of her research planning because its members comprised one-third of her target population, The Arthur W. Page Society is a professional association for senior public relations and corporate communications executives who seek to enrich and strengthen their profession. The membership consists primarily of chief communications officers of Fortune 500 corporations, the CEOs of the world's largest public relations agencies, and leading academics from the nation's top business and communications schools who have distinguished themselves teaching corporate communications…Today the Arthur W. Page Society has over 340 members. (Society, 2010) 84 The researcher made several attempts over a period of several months to contact individuals she knew who had ties to the Arthur W. Page Society. Her goal was to promote awareness of her project among its members and elicit support from the organization for participation by its members. Her first contact was to another fellow member of the PRSA College of Fellows. This individual invited a second member of the Society, whom the researcher also had met, to participate in the discussion. On behalf of both of them, the first individual requested a copy of the researcher’s questionnaire, background information on the study, and the researcher’s biographical information. The second contact suggested that she contact the Arthur Page staff and the Institute for Public Relations Research to request cooperation because participation would be difficult to achieve without endorsement by an organization or relevance to its members. After reading her materials, the first contact responded that the questionnaire was too long and some of the questions too legally sensitive. He asked if she would be willing to reduce the length of the questionnaire from 20 minutes (his assumption) to six and eliminate the sensitive questions. He felt those two changes were necessary to gain some response from the target population, either through the Arthur W. Page Society or the Institute for Public Relations Research. The researcher interpreted this response to mean her topic was too sensitive for corporate public relations officers and her efforts to pare down the extensive federal surveys had not achieved a level of simplicity and brevity required by corporate public relations officers. The researcher believed that complying with her contact’s suggestions would require changing the focus of her dissertation, starting over with her dissertation committee, and starting over with the IRB. 85 PRSA Board of Directors The researcher contacted members of the national board of the Public Relations Society of America for assistance. As a twenty-plus-year member of the organization, an Accredited member, and a member of the PRSA College of Fellows, the researcher thought she might be able to interest someone on the PRSA board of directors in her research project. She e-mailed four PRSA board members whom she knew personally to request their assistance in reaching her target population through their contacts with those individuals or their organizations. One of her contacts offered to contact his colleagues at the Arthur W. Page Society to enlist help. A second contact responded that he was forwarding the request to PRSA management to see what could be done through the organization, as well as in conjunction with the Arthur W. Page Society. The third contact responded to concur with the first two contacts’ suggestions. The fourth contact did not respond. The researcher heard nothing further from any of these contacts or from anyone at the Public Relations Society of America throughout the remainder of her research project. Arthur W. Page Society Revisited After deciding not to initiate the PRSA research process, the researcher again contacted her first Arthur W. Page Society contact and reissued her request for assistance. She again emphasized the groundbreaking nature of her research and her design of the questionnaire to allow comparisons with federal and management school studies. The second contact responded that conversation among board members and officers of the Arthur W. Page Society, the International Public Relations Association, and several professors involved with one or both organizations had led to the same result. The group 86 was in agreement that it would not be appropriate to ask members to participate in a survey that was not part of one of the organization’s own research projects and that they viewed as too long. Academic E-mail Marketing During the process of cleaning the list, the researcher contacted several members of the University of Oregon’s Charles H. Lundquist College of Business about online marketing and ways to improve e-mail research response rates. Suggestions included offering something of value, removing the IRB disclaimer from the e-mail invitation, removing the permission for future research form, not overstating the time it would take to complete the survey, and focusing on making the subject line and content enticing, employing a snowball sampling technique, using third-party endorsers, purchasing a paid panel, purchasing a commercial e-mail distribution service, and conducting interviews by phone. Although the suggestions were intriguing, the researcher felt constrained by the IRB’s three-contact rule and by the IRB’s requirement that the e-mails be delivered from her personal e-mail account. She rejected the suggestion of phone interviews on the basis of time it would take and the low probability that this group of people would be easier to reach by phone than they had been by e-mail. Commercial Direct Marketing Email is a direct descendant of the direct marketing field. For nonmarketing people, direct marketing (or DM for those in the know), is basically mailing stuff – catalogs, credit card apps, etc. It’s always been a profitable venture but often considered the uglier cousin of the glamorous world of advertising, especially TV. For a DM professional, email is like crack cocaine. In the past, the DM people would send something out and then wait 3 weeks for something to get mailed back. In email, your diehard fans (and unfortunately the unsubscribers) will start clicking almost immediately. (Cangialosi, 2010, p. 12) 87 The researcher reviewed e-mail marketing guides on ways to increase deliverability and response rate for her e-mail survey. She found several suggestions aimed at avoiding the appearance of spam, including being clear and honest in the To and From headers, using normal language in the subject line and body, not proclaiming the e- mail was not spam, not using unusual spellings and spacing, and not using words commonly used in spam, such as references to Rolex watches, drugs, debt, or sex (Markowitz, 2010). A second source advised e-mail marketers to “collect and count bounces,” “remove addresses that bounce repeatedly,” and not to “put removed addresses back on the list “just in case the bounces were a mistake” (Levine & Mason, 2010, p. 1). Another source recommended addressing each e-mail to an individual, rather than distributing to a list name. “Many tests have shown that using a person’s name in either the subject line or the body copy can significantly improve results” (Cangialosi, 2010, p. 8). In the first survey distribution, the researcher was not able to send to a named individual because the IRB required her survey to be anonymous because of the perceived sensitivity of her subject matter. Subject Lines E-mail marketing guides frequently referenced the importance of subject lines in the overall effectiveness of e-mail marketing campaigns. In fact, the importance of the subject line was highlighted by many commercial and academic sources. “An appropriate subject line in the invitation” was one of 13 response rate criteria recommended in an e- mail study of online communities (Andrews, Nonnecke, & Preece, 2003, p. 192). As one marketer advised, “use straightforward subject lines and message bodies. Misleading subject lines are a highly reliable way to identify viruses and spam” (Levine & Mason, 88 2010, p. 1). Another source emphasized the importance of making the subject line interesting. “Preparing the content—the subject line is all….Your reader is going to try to decide if it’s compelling enough to open on the subject line alone” (Cangialosi, 2010, p. 10). Reinforcing that advice, another source advised, “Without an accurate and compelling Subject Line, the chances of your email being viewed are very low” (Blue Sky Factory, 2010, p. 3). It’s a well-publicized best practice in email marketing to test and use different subject lines on your email campaigns. Load up a few subject lines, test with multivariate testing software...and then choose the subject line that performs best. Where most email marketers stumble is at the very start – how do you know what subject lines to test? (Penn, 2010) Blue Sky Factory, an online, e-mail marketing blog, offered advice on how to create compelling subject lines with the use of Google’s Wonder Wheel and Google Adwords Keywords (Penn, 2010). Wonder Wheel allowed the researcher to insert a phrase and see a graphic display of related phrases, and Adwords Keywords allowed her to rank those phrases based on how frequently those words are searched for on Google. Subject Line Development The researcher tested a number of keywords on Google Adwords Keywords related to her subject matter: academic; competitors; corporate; fortune; research; response; critical; dissertation; management; public relations; research; image; profit; reputation; corporate; Sarbanes-Oxley; ethics; firing; retaliation; whistleblowing; wrongdoing; HR; answers; data; fraud; legal; and relationships. The words reflecting the highest search volume on Google were corporate; research; management; image; data; fraud; and legal. 89 Based on the responses to her Adwords Keywords results, the researcher crafted three new subject lines that she tested for response: • Subject Line B: Fraud research gives PR managers scientific data, • Subject Line C: Scientific research gives PR managers fraud data, and • Subject Line D: Fraud reporting data gives management insight. Subject Line Testing Process The researcher completed cleaning the bounce back list using the process referenced previously. The bounce back list included the original 278 bounce backs plus an additional 28 names that she identified as either incorrect contacts or incorrect e-mail addresses. She moved the bounce back list to a separate workbook in her master spreadsheet and divided it alphabetically into three equal segments of 102 names each for the subject line test (Phase I, Phase II, and Phase III). Lacking access to the multivariate testing software recommended by marketers (Cangialosi, 2010, pp. 22-23), and having no way to distinguish respondents because of anonymity restrictions, the researcher tested her subject lines consecutively over several days. Because response to the original distribution had come within 24 hours, as predicted by marketers, the researcher determined it was not necessary to wait longer than 24 hours between distributions in order to identify which subject line had resulted in the greater response. Subject Line Testing Implementation: Phase I The researcher sent Phase I of the subject line test to 102 e-mail addresses that contained the revised subject line, “Fraud Reporting Gives PR Managers Scientific 90 Data,” rather than the original subject line, “Fortune 1000: Sarbanes-Oxley and Whistleblowing.” In addition to the revised subject line, the researcher changed several elements of the invitation. She recognized that changing more than one variable in the distribution invalidated the subject line change as the sole predictor of response rates. However, without access to an e-mail distribution service or multivariate testing software, she felt she needed to make as many of those changes as might increase response rates as possible. The additional changes included the following: • She shortened her introductory paragraph by eliminating her PRSA credentials in the first sentence; • She substituted the words “fraud reporting” for “whistleblowing” in her revised introductory sentence: “I’m writing to ask you to participate in my dissertation research about how Sarbanes-Oxley and fraud reporting have, or have not, affected your roles as counselor to C-Level Management”; • She mentioned the incentive she was offering, a copy of the survey’s Executive Summary, in the second paragraph, rather than in the fifth; • She removed the e-mail attachment that contained the Permission/Release Form regarding permission to be contacted for future research; • She provided an embedded link in the e-mail to the researcher’s University of Oregon webpage, where the form was posted; • She reduced the estimated completion time from 20 minutes to 10-15 minutes (based on the researcher’s best estimate of how much time the survey would take to complete); • She changed the final sentence from, “No one has done this research 91 before, and you will be helping me break new ground,” to “No one has done this research before, and you will be helping to break new ground.” She hoped the change in emphasis from the researcher to the research would encourage response. The researcher retained all safeguards and cautions directed at respondent security and confidentiality in the revised e-mail invitation. Phase I resulted in 55 bounce backs (54%), all of which were received in the first hour after delivery, and zero responses. The researcher sent an aggregated list of 39 bounce backs that had been delivered to her by the university list serve to the list manager for his review. His assessment was that most of the bounce backs were the result of bad e-mail addresses, and very few were the result of spam filters. Because of the lack of response to her first subject line test, the researcher took additional actions: • She changed the title of the survey on SurveyMonkey from “Fortune 1000 Public Relations: Sarbanes-Oxley and Whistleblowing,” to “Fortune 1000 Public Relations” because she was concerned that the original title was alarming potential respondents; • She verified with the list manager that the list serve would not allow her to send e-mails to individual names; • She added her PRSA honorifics (“APR” and “Fellow PRSA”) to her name on her university e-mail account; and • She reinstated her credentials in the first sentence of the e-mail invitation for Phase II of the subject line test because, despite adding them to her 92 account, she was not confident that the honorifics would be displayed in the “Sender” line of the e-mail distributed by the list serve. Subject Line Testing Implementation: Phase II The researcher sent Phase II of the subject line test to 102 e-mail addresses the contained the revised subject line, “Scientific Research Gives PR Managers Fraud Data,” rather than the Phase I subject line, “Fraud Reporting Gives PR Managers Scientific Data.” Phase II of the subject line test incorporated the changes listed above and represented the second one-third of the alphabetized list of bounce backs from the original distribution list. She again used the university’s list serve to deliver the e-mails. At the time of distribution, the total response to her survey stood at five. Phase II resulted in 66 bounce backs (61%) and generated two additional survey responses. It also resulted in an accolade from one of the recipients. He complimented the researcher on the quality of writing in the invitation and for addressing potential concerns about anonymity and confidentiality. Subject Line Testing Implementation: Phase III The researcher sent Phase III of the subject line test to 102 e-mail addresses identical to the Phase II distribution with the exception of a revised subject line, “Fraud Reporting Data Gives Management Insight,” rather than the Phase II subject line, “Scientific Research Gives PR Managers Fraud Data.” Phase III resulted in another personal note of encouragement from a fellow APR who wished her well and indicated her willingness to take the survey. The Phase III distribution resulted in two additional survey responses and 43 (44%) total bounce backs. At the conclusion of Phase III of the subject line test, the researcher had received a total 93 of nine responses to the survey since the first distribution on February 24, 2010, including four from the last two subject line tests. Subject Line Test Results The subject line test was inconclusive. Two of the subject lines resulted in two responses each, while the third yielded no response. The researcher had intended to send the first reminder to the cleaned list with the subject line that had garnered the most responses. However, based on the results, she was not convinced that she had found the subject line that would generate maximum response. She concluded that she should continue to experiment with subject line wording in subsequent distributions. Subject Line Test Bounce Backs The subject line test resulted in 164 (54%) bounce backs. She followed the following procedure to address the bounce backs: • She chose a recent, comprehensive list she had compiled from Mint Global; • If the name from the bounce back was on the Mint Global list with the same e-mail address and no alternative contact, she left the original name on the list; • If the same name was on the Mint Global list with a different e-mail address, she retained the name and inserted a line on her database with the new e-mail address; • If the same name was on the Mint Global list, but there was also an additional contact with a higher title, she inserted and substituted the higher name and title; and 94 • If the name was not on the Mint Global list, she inserted a line on her database with the next highest name and title. List 1 Distribution 2 (Reminder 1) The researcher sent the Reminder 1 e-mail to the 978 names that represented the original distribution list and the cleaned bounce back list. This distribution carried the revised subject line, “Fortune 1000 PR Survey Needs Your Response,” rather than the Phase III subject line, “Fraud Reporting Data Gives Management Insight.” It also incorporated a colleague’s suggestion to personalize the text and reference the positive response the original had generated. Although the response to the original survey was not positive, the researcher felt justified in adding the following introductory paragraph to the invitational text distributed in Phase II and Phase III: If you are among the Fortune 1000 public relations executives who have responded to my survey, please accept my sincere thanks and feel free to delete this message. If not, please take the survey and contribute to an understanding of public relations in the Fortune 1000. This distribution resulted in nine responses to the survey. It also resulted in one negative and one positive personal e-mail responses to the invitation. The negative response was a request to be removed from the distribution list, which the researcher did. The positive response was a note of encouragement and agreement to respond to the survey. List 1 Distribution 3 (Reminder 2) Following the distribution of Reminder 1, the researcher cleaned the bounce back list. She then sent the Reminder 2 e-mail to the 890 names that represented the original distribution list and the cleaned bounce back list from Reminder 1. This distribution 95 carried the revised subject line, “PR Student Needs Your Help with Dissertation Research on Fortune 1000,” rather than the Reminder 1 subject line, “Fortune 1000 PR Survey Needs Your Response.” This distribution resulted in 10 more responses to the survey. It also resulted in one personal e-mail from a home products company that declined to participate because of corporate policy. The researcher removed the respondent from the list. Based on past experience, the researcher did not try to encourage a response to the survey. In an effort to extract the maximum possible response, the researcher distributed a reminder identical to List 1 Distribution 3 (Reminder 2) to the bounce backs from List 1 (List 1 BB Contact 1) with the last subject line to 55 names from the bounce back list from List 1 Distribution 3. Subsequently, she distributed a reminder (List 1 BB Contact 2) to the same list of 55 with the revised subject line, “PR Dissertation on Fortune 1000 Needs Your Response.” Because of time constraints, she did not clean the bounce back list prior to either distribution; she did not receive any additional response to the survey from these two distributions, and she did not distribute the invitation to the bounce back list for the third time. List 1 Results The researcher received a total of 32 responses from the distribution of List 1 via SurveyMonkey. Expanding the Population Based on the poor response to the first list, the researcher began using Mint Global to develop a second list that was directed at what she perceived to be the second tier of public relations executives in each of the Fortune 1000 corporations. This list 96 focused more directly on executives with the words “public relations” in their titles. The rationale for the second list was that (a) the first list may have captured executives above the level of the person who identified him or herself as the highest ranking public relations executive in the organization, and (b) those with “public relations” in their titles might be more receptive to the researcher and the goals of the survey, regardless of their level in the corporate hierarchy. For the second list, the researcher identified the executive with the highest title containing “public relations,” “communications,” or “public affairs” in each of the corporations, wherever possible. Qualtrics In August 2010, the University of Oregon announced it had purchased access to Qualtrics for the entire university and encouraged all students to begin using it. Based on the low response rate from SurveyMonkey, the researcher initiated the process of moving her survey instrument from SurveyMonkey to Qualtrics. This was not an easy task because the Qualtrics process differed in many respects from SurveyMonkey. The key difference the researcher noted in functionality between SurveyMonkey and Qualtrics was the ability of Qualtrics to distribute e-mails with embedded links to the survey while collecting data anonymously. SurveyMonkey did not have the capacity to deliver e-mails without collecting data from respondents. The researcher was hopeful that using the Qualtrics distribution function would increase the response rate because of Qualtric’s stated efforts to reduce spam filtering of its e-mail distribution and because of its reputation as an academic survey research organization. 97 The Loss of Mint Global Two weeks after beginning compilation of the second list, and while she was initiating the move to Qualtrics, the researcher learned that the library had canceled its subscription to Mint Global because of cost and limited usage. In an attempt to capture what she could of the necessary data before the subscription expired, the researcher downloaded a file containing the names and contact information of executives with all of the possible titles she had identified. Due to the size of the file and the record limitations of Microsoft Excel, the researcher moved the data to Microsoft Access in order to manage it more effectively. The new download contained new contact information that she incorporated into List 2 using the process previously described. List 2 For the first distribution of her survey using Qualtrics, the researcher made several changes to the invitational e-mail (Letter 4): • She revised the subject line to “Please help PR student study Fortune 1000”; • She revised the second sentence in the first paragraph from, “…I am writing to ask you to participate in my dissertation research about how Sarbanes-Oxley and fraud reporting may have affected your role as counselor to C-Level management,” to “…I respectfully request your help with my dissertation research on PR’s role in developing Sarbanes-Oxley reporting channels and the possibility that reporting bad news to management may have affected your relationship with your Fortune 1000 corporation”; 98 • She revised the second paragraph from, “I am studying public relations in the Fortune 1000, and I have invited you to participate because you are the highest ranking public relations executive in your Fortune 1000 corporation. As such, your response is critical to my research. This research is cutting edge, and I will send you a copy of the executive summary before I publish it anywhere else,” to, “As a public relations executive in a Fortune 1000 corporation, you are critical to my research, and I will send you a copy of the executive summary.” • She removed the reference and the embedded link to the permission/release form concerning future research on the topic; • She revised the last sentence from, “No one has done this research before, and you will be helping to break new ground,” to, “Your response is crucial to understanding this topic, and by completing the survey you will be helping to break new ground.” List 3 At the same time the researcher was developing List 2, she was also developing List 3. At the time, she had no indication that she would use the third list, but given the current state of her research, she felt it wise to try to maximize her options. List 3 contained all available names with public relations in the title or titles potentially having public relations responsibilities. In building the two lists, the researcher exhausted the supply of names available from the static Mint Global database. As a result, she decided not to attempt to clean bounce backs from List 2 because she no longer had access to a source of possible replacement names. 99 List 2 Distribution 1 The researcher used Qualtrics to send the first distribution of List 2 to the 863 public relations executives in the Fortune 1000 list of companies whom she had identified as potentially functioning at the next level down in the corporate hierarchy from the individuals on List 1. Within 90 minutes the researcher had received 12 responses, and within 24 hours she had received a total of 14 responses. One large computer company declined to participate on the basis of policy. The researcher removed the respondent from the list. Through this distribution, the researcher learned that Qualtrics reports as a response anyone who opens the survey, regardless of whether or not that person answers a question. In addition to the increased, reported response rate, the researcher noted that Qualtrics did not deliver bounce backs to her inbox. In fact, Qualtrics reported it had delivered all e-mails. The researcher doubted this but could not disprove it. Despite the positive response, the researcher experienced a significant setback with Qualtrics: one of her questions in the Sarbanes-Oxley section would not allow for multiple responses. She attributed the problem to three issues: her unfamiliarity with the Qualtrics, the process required to transfer her survey from SurveyMonkey to Qualtrics, and the process involved in testing the system. She tested the survey after she transferred it, and it appeared to operate as it had on SurveyMonkey. However, she subsequently received two e-mails notifying her that Question 4, which asked the respondent to identify all channels used to publicize the anonymous communication channel, would not allow more than one response. This was a major setback because of the lack of responses overall. She had anticipated contributing that 100 new information to the discussion of the role of public relations in Sarbanes-Oxley implementation, but the error in the survey administration prevented that from happening. As an alternative, she decided to calculate the responses to that question from the 32 respondents to the SurveyMonkey questionnaire, in which the survey did work correctly, and draw conclusions from those responses. List 2 Distribution 2 The researcher sent the second distribution of List 2 to 863 public relations executives in the Fortune 1000 list of companies. This distribution contained the revised subject line, “Please help PR student study Fortune 1000.” The distribution received nine responses. In addition, one recipient declined to participate on the basis of not having been involved in the development of the anonymous communication channel, and another recipient reported passing the survey to someone else who might be better suited to answer it: List 2 Distribution 3 The researcher sent the third distribution of the survey with the same subject line to List 2, 863 public relations executives in the Fortune 1000 list of companies. This distribution received eight responses. List 3 Distribution 1 On November 17, 2010, the researcher sent the first distribution of the survey with the same subject line to List 3, 558 public relations executives and practitioners in the Fortune 1000 list of companies. This distribution received nine responses. 101 List 3 Distribution 2 On November 23, 2010, the researcher sent the second distribution of the survey with the same subject line to List 3, 558 public relations practitioners in the Fortune 1000 list of companies. This distribution received five responses. She received one personal e- mail request asking to be removed from the list on the basis of not being a public relations professional. An out-of-office response was received from someone who had left the organization and who suggested an alternate contact. The researcher noted the alternative contact in the Master list but did not send a survey because no title was supplied and because this was one of two individuals contacted at that company. List 3 Distribution 3 On November 30, 2010, the researcher sent the third distribution of the survey with the same subject line to List 3, 558 public relations executives and practitioners in the Fortune 1000 list of companies. The distribution received five responses. In addition, one recipient representing a utility declined to participate based on being over-surveyed; one declined to participate based on not being a public relations professional and asked to be removed from the mailing list, and one declined to participate because he did not know the answer to the first question concerning Sarbanes-Oxley compliance and the question did not include a “Don’t know” answer category. Summary In all, the researcher distributed the survey to a total of 2,423 public relations executives in the Fortune 1000 corporations through a series of three e-mail lists. As noted previously, the researcher identified those with public relations responsibilities based on corporate Websites, news releases, annual reports and other published materials 102 that contained indicators of rank. Not everyone the researcher identified as having public relations responsibilities held a title containing the words, “public relations.” The researcher distributed List 1 via SurveyMonkey to 999 executives in the Fortune 1000 companies, of whom 157 held public relations titles, and received 32 responses to the survey. The researcher distributed List 2 via Qualtrics to 863 executives in the Fortune 1000 companies, of whom 205 held public relations titles, and received 34 responses to the survey. The researcher distributed List 3 via Qualtrics to 558, of whom 195 held public relations titles, and received 19 responses to the survey, for a total of 80 responses. As previously noted, not everyone identified as a respondent by Qualtrics answered a question; and of those who answered one or more questions, not all completed the survey. Table 3 shows how the survey was intended to answer the research questions and hypotheses. It lists the research questions, the hypotheses, the constructs being measured, the survey question or questions that address the research questions and hypotheses, and the statistical analysis technique the researcher intended to use. 103 Table 3. Research Questions, Hypotheses, Measurement, and Statistics RQ H Measurement Question # Statistic RQ1 SOX 1-4 Descriptive statistics sRQ2 Wrongdoing 5, 14 " Whistleblowing 8, 11, 17, 20 RQ3 Responsibility 9, 18 " RQ4 " H1 Channel 10, 19 " H2 Channel 10, 19 " RQ5 H3 Power Whistleblowing Retaliation 50-59 8, 11, 17, 20 12, 21 Multiple Regression H4 Power Whistleblowing Seriousness Retaliation 55-64 8, 11, 17, 20 6, 7, 15, 16 12, 21 " H5 Whistleblowing Channel Retaliation 8, 11, 17, 20 10, 19 12, 21 " H6 Whistleblowing Retaliation Lack support 8, 11, 17, 20 12, 21 12 " RQ6 H7 Whistleblowing Retaliation Relationships 8, 11, 17, 20 12, 21 23-48 " RQ7 Whistleblowing Retaliation Gender 8, 11, 17, 20 12, 21 57 " 104 Table 4 shows the statistics the researcher actually employed based on the level of response she received. Table 4. Research Questions, Measurement, and Statistics Revised RQ H Measurement Question # Statistic RQ1 SOX 1-4 Descriptive statistics sRQ2 Wrongdoing 5, 14 " Whistleblowing 8, 11, 17, 20 RQ3 Responsibility 9, 18 " RQ4 " H1 Channel 10, 19 " H2 Channel 10, 19 " RQ5 H3 Power Whistleblowing Retaliation 50-59 8, 11, 17, 20 12, 21 Correlations, Chi- squares, t-tests H4 Power Whistleblowing Seriousness Retaliation 55-64 8, 11, 17, 20 6, 7, 15, 16 12, 21 Correlations, Chi- squares, t-tests H5 Whistleblowing Channel Retaliation 8, 11, 17, 20 10, 19 12, 21 Correlations, Chi- squares, t-tests H6 Whistleblowing Retaliation Lack support 8, 11, 17, 20 12, 21 12 Correlations, Chi- squares, t-tests RQ6 H7 Whistleblowing Retaliation Relationships 8, 11, 17, 20 12, 21 23-48 Correlations, Chi- squares, t-tests, Cronbach’s alpha RQ7 Whistleblowing Retaliation Gender 8, 11, 17, 20 12, 21 57 Correlations, Chi- squares, t-tests Total Awareness of Wrongdoing Logistic Regression 105 CHAPTER IV RESULTS This study was designed as a census of the top-ranking public relations executive in each of the Fortune 1000 companies named on the May 2009 Fortune 1000 List. However, the issues recounted in Chapter 3 explains why that target population was abandoned in favor of gathering data from at least three levels of public relations executives within those corporations. Those levels were identified as (1) those in charge of public relations; (2) those who are second-in-command of public relations; and (3) those working in public relations. Response Rate Internet surveys sent to specific individuals are similar to mail surveys for purposes of calculating response rates (AAPOR, 2008). Four categories of responses are considered when evaluating mail surveys and Internet surveys: completed and partial questionnaires; eligible, no returned questionnaire (non-response); unknown eligibility, no questionnaire returned; and not eligible (2008). Because the researcher developed the e-mail list herself, she was not able to identify eligible or not eligible recipients. However, she identified those who reported to her via e-mail that they would not take the survey for one reason or another as eligible, no returned questionnaire (non-response). She counted unresolved bounce backs as unknown eligibility. The researcher counted questionnaires with all questions answered (completed questionnaires) and questionnaires with any questions answered (partial questionnaires) as responses in calculating response rate. She divided the number of responses (complete 106 and partial) by the number of questionnaires sent, minus the number of no responses bounce backs, to determine response rate. The researcher sent a total of 2,423 surveys to the three separate lists of Fortune 1000 public relations executives. This included 999 names on List 1, 865 names on List 2, and 559 names on List 3. Only List 1, which used the University of Oregon listserve for distribution, contained bounce backs. Qualtrics identified all surveys in both List 2 and List 3 as delivered. After bounce backs were taken into account, the total number of surveys presumed delivered was 2,299. Of those, eight refused to participate. This resulted in a total presumed distribution of 2,291. The total number of responses, complete and partial, was 80, which resulted in the response rate of 3.5%. This response rate is within the anticipated response rate for Internet surveys (1%-30%) identified by Wimmer and Dominck (2006) and the response rate for Internet surveys (1-8%) identified by the researcher’s marketing colleagues (I. Parkman, personal communication, April 15, 2010; J. Chen, personal communication, April 16, 2010). It is lower than the response rate for Internet surveys (70.1%) identified in a study of the effectiveness of Web surveys versus mail surveys (Kiernan, et al., 2005) and for the mean response rate (36.83%) for e-mail surveys identified in 31 studies over the period 1986 to 2000 (Sheehan, 2001). Researchers recently have noted that response rates for all surveys are declining, and response rates for Web surveys appear to be declining more than others (Couper & Miller, 2008). The response rate to this study and the results from those who answered the survey suggest that those in the target population—public relations executives in the 107 Fortune 1000 corporations—may be hesitant to talk about whistleblowing, which is discussed in more detail later. It also may reflect the lower response rate for surveys of corporate executives identified by management researchers (Baruch & Holtom, 2008, p. 1154). Sample Size Despite the tolerance for lower response rates from an executive population (Baruch & Holtom, 2008), the researcher has been careful not to generalize to the larger population of top-ranking public relations executives in the Fortune 1000. As was demonstrated in the last chapter, she delivered her survey to increasingly lower levels in the corporate hierarchy in hopes of garnering additional responses. Statisticians and other researchers differ on the numbers needed for various parametric analyses. Some statisticians have argued that small samples can be representative of the larger population if they reach a certain threshold (Churchill & Peter, 1984; Hair, Anderson, Tatham, & Black, 1998; Hanke & Reitsch, 1991; Keppel & Wickens, 2004). Sample sizes ranging from 12 (Keppel & Wickens, 2004) to 30 have been deemed sufficient “to ensure the normal shape of the sampling distribution” (Hanke & Reitsch, 1991, p. 206). In multiple regression, a small sample is one with “fewer than 20 observations” (Hair, et al., 1998, p. 164), and sample size has been correlated negatively with reliability (Churchill & Peter, 1984). These scholars seem to support the validity of parametric analyses with small samples and suggest that one can generalize to a larger population using small samples. However, the researcher has been cautious about doing so. Instead, she has chosen to use the results of her sample to suggest trends and suggest areas of 108 interest for more research, but she has been careful not to draw cause and effect relationships based on those samples. Data Analysis The researcher used SPSS 19 to analyze responses. The researcher had planned to calculate four new constructs around the Sarbanes-Oxley questions at the firm level. However, firm-level analysis was not possible because more than one individual from each firm could have responded to the three survey distributions. For that reason, the researcher merely reported descriptive statistics for these items. The researcher intended to analyze the data further using multiple regression following Miceli et al. (1999). Multiple regression is widely used because of “its ability to predict and explain metric variables” (Hair, et al., 1998, p. 240). The researcher had intended to perform a regression to determine if certain independent variables (comprehensiveness of retaliation, lack of support, seriousness of wrongdoing, whistleblowing channels, and demographic data) predicted retaliation as some of them had in prior management research. Additionally, she planned to follow Huang (2008) in using trust, control mutuality, commitment, and satisfaction as dependent variables in a hierarchical regression to determine the extent to which retaliation impacts those relationship quality outcomes. However, the response for those dependent variables precluded regression analysis. Even though a multiple regression can be run on a sample of 20 (Hair, et al., 1998), the researcher’s dependent variable of interest, retaliation, had only 7 responses. Instead, the researcher chose to examine a different variable, the amount of wrongdoing 109 in the environment, or total awareness of wrongdoing, using a logistic regression. The results of the analyses the researcher was able to conduct are presented below. Demographics The demographic characteristics of the sample suggest it is representative of the target population, which other scholars have thought to be older, better-educated, and higher-ranking white males (Berger, 2005; L. A. Grunig, et al., 2002; Inroads, 2010; Near & Miceli, 1996; Public Relations Coalition, 2005). Comparing the demographic characteristics of the sample and the previously identified characteristics of the target population is an accepted method of addressing nonresponse error, which occurs when respondents and nonrespondents differ (Dooley & Lindner, 2003). As shown in Table 5, respondents generally fit this profile. Respondents were male (60.7%), older (M = 48.40, SD = 8.676), white (91.1%) college graduates (100%).Racial diversity was concentrated in the younger age group. Of the four non-white respondents, three were in the 31-45 age group. 110 Table 5. Respondent Demographics Variables Category Frequencies Percent Gender Male 34 60.7 Female 22 39.3 Age Group 31-45 16 32.0 46-64 34 68.0 Education No college 0 0.0 Some college 0 0.0 A four-year college degree 20 35.7 Some graduate or professional school 12 21.0 Graduate or professional degree 24 42.9 Race American Indian/Alaskan Native 0 0.0 Asian/Asian American 1 1.3 Black or African American 2 3.6 White, not Hispanic 51 91.1 Hispanic/Latino 1 1.3 Other 0 0.0 White, not Hispanic and Hispanic/Latino 1 1.3 Employment Characteristics On average, respondents were relatively new to their companies (M=10.48, SD 9.741) and their positions (M =5.16, SD =4.48), although they may have had long careers in other organizations (Table 6). A post-hoc analysis showed a moderate correlation between length of time with the company and length of time in position (r = .453, n = 52, p = .001). Their annual salaries averaged slightly more than $200,000 (M = 2.17, SD = .778). More than half (n = 32; 57.1%) worked in corporate communications departments, and more than three-quarters (n = 44; 78.6%) wrote performance appraisals for others. Respondents were highly successful in their organizations, as indicated by their most recent performance ratings (M = 4.05, SD = .803), and almost three-quarters (n = 40; 71.5%) reported either to the chief executive officer or to someone who reported to the chief executive officer. 111 Table 6. Employment Variables for All Respondents Variables Category Frequencies Percent How many years have you worked for this employer? 1-10 33 63.5 11-49 19 36.5 How many years have you been in your current position? 1-5 39 70.9 6-25 16 29.1 What is your current salary range? <$100,000 11 20.8 $100,000-$199,999 23 43.4 $200,000-$500,000 18 34.0 >$500,000 1 1.9 Are you the highest-ranking public relations executive in your organization? Yes 36 64.3 No 20 35.7 What is your gender? Male 34 60.7 Female 22 39.3 What is the location of your position in relation to the Chief Executive Officer? Report to CEO 10 17.9 Report to someone who reports to CEO 30 53.6 Report to someone who reports to someone who reports to CEO 11 19.6 Other 5 8.9 In what department do you work? Public relations 12 21.4 Corporate communications 32 57.1 Marketing 6 10.7 Finance and Administration 2 3.6 Other 4 7.1 Do you now write performance appraisals for other employees? Yes 44 78.6 No 12 21.4 112 Table 6. (Continued) Variables Category Frequencies Percent Which of the following most closely describes the performance rating you received at your last appraisal? Unacceptable 0 0.0 Minimally successful 1 1.8 Fully successful 13 23.6 Exceeds fully successful 23 41.8 Outstanding 18 32.7 113 Demographics and Characteristics of Top-ranking Public Relations Executives Almost two-thirds (n = 36; 64.3%) of respondents were the top-ranking public relations executives in their Fortune 1000 corporations, which was the original target population of this study, and more than half (n = 21; 58.3%) of those were males (Table 7). 114 Table 7. Employment Variables for Top-Ranked Public Relations Executive Respondents Variables Category Frequencies Percent How many years have you worked for this employer? 1-10 20 58.8 11-49 14 41.2 How many years have you been in your current position? 1-5 23 65.7 6-25 12 34.3 What is your current salary range? <$100,000 6 17.6 $100,000-$199,999 10 29.4 $200,000-$500,000 17 50 >$500,000 1 2.9 What is your gender? Male 21 58.3 Female 15 41.7 What is the location of your position in relation to the Chief Executive Officer? Report to CEO 9 25 Report to someone who reports to CEO 20 55.6 Report to someone who reports to someone who reports to CEO 4 11.1 Other 3 8.3 In what department do you work? Public relations 4 11.1 Corporate communications 21 58.3 Marketing 5 13.9 Finance and Administration 2 5.6 Other 4 11.1 Do you now write performance appraisals for other employees? Yes 30 83.3 No 6 16.7 Which of the following most closely describes the performance rating you received at your last appraisal? Unacceptable 0 0.0 115 Table 7. (Continued) Variables Category Frequencies Percent Minimally successful 1 2.9 Fully successful 8 22.9 Exceeds fully successful 13 37.1 Outstanding 13 37.1 116 Sarbanes-Oxley One goal of this study was to gather information about Sarbanes-Oxley whistleblowing channels. As stated previously, Sarbanes-Oxley, also known as the Public Company Accounting Reform and Investor Protection Act, or Public Law 107-204, mandated that boards of directors of publicly traded corporations establish a communication channel to allow employees to report corporate financial wrongdoing anonymously to outside directors who sit on the audit committees of those corporations’ boards of directors. Whistleblowing Channels Questions 1 through 4 (Section I; see Appendix A) were designed to elicit information about the extent have public relations executives of Fortune 1000 corporations been involved in developing and publicizing Sarbanes-Oxley whistleblowing channels? Question 1 asked if the respondent’s organization had an anonymous communication channel to reach the audit committee of the board of directors. Of the 80 who responded to this question, 53 (66.3%) reported their organization had such a communication channel. The researcher could not draw conclusions about Fortune 1000 compliance with this provision of Sarbanes-Oxley, however, because not all companies on the Fortune 1000 list are publicly traded (CNNMoney, 2010). In the second question (Question 2), the survey asked if the respondent or someone who worked for him or her in the public relations department had played a role in developing the anonymous communication channel. The results suggest public relations executives and their staffs do not play a major role in developing this 117 anonymous communication channel for their corporations. Of the 70 (87.5%) who responded to the question, 55 (68.8%) reported neither they nor someone who worked for them had played such a role, while only 15 (21.4%) reported they or someone who worked for them had. Question 3 asked if the respondent or someone who worked for him or her within the public relations department had played a role in publicizing the anonymous communication channel. In a reversal of the response to the last question, almost two- thirds (63.1%) of respondents reported that they or someone who worked for them within the public relations department had done so. Of the 65 (81.3%) who responded to this question, 41 (63.1%) reported they or someone who worked for them had played such a role. The fourth question asked about the types of communication channels (electronic, print, verbal, and other) used by the organization to publicize its anonymous communications channel. Due to a malfunction of the answer categories for this question on the second and third survey distributions, only the results from the first distribution are discussed here. In the first survey distribution, 88.8% (N = 32) of respondents stated they used electronic communication alone or in conjunction with other communication channels. One (3.7%) reported using another communication channel, which leaves one to question what that might have been, and 2 (7.4%) reported that their anonymous communication channel had not been publicized. 118 Knowledge of Illegal, Wasteful, or Unethical Activity: Wrongdoing Section II and Section III were modified versions of the surveys administered to federal employees by the United States Merit System Protection Board in 1980, 1983, and 1992. Section II (Questions 5 through 13; see Appendix A) asked respondents to report on wrongdoing known to others within the field of public relations without implicating themselves. Third-party Awareness of Wrongdoing Question 5 asked if, during the course of his or her entire career, the respondent had known another public relations executive who was aware of illegal, wasteful, or unethical activity in his or her organization. Of the 73 (91.3% of respondents) who responded to this question, 25 (34.2%) knew another person in the field of public relations who was aware of illegal, wasteful, or unethical activity in his or her own organization. The next question asked the dollar amount of the illegal, wasteful, or unethical activity of which the other person was aware. When the dollar amount was known, it was sometimes substantial. Of the 23 (28.8%) who responded to Question 6, 3 (13%) reported the dollar amount was between $50,000 and $99,999; 2 (8.7%) reported it was between $100,000 and $500,000 and 1 (4.3%) reported it was greater than $500,000. Almost half of those who reported the frequency of the activity said it was a rare occurrence. However, one quarter said it was not rare. Of the 23 (28.8%) who responded to Question 7, 11 (47.8%) reported it happened rarely; 6 (26%) said it happened occasionally or frequently; and 6 (26.2%) did not know. 119 Question 8 asked if the person who had observed the wrongdoing had reported the activity to an individual or group. Of the 21 (26.3%) who responded, more than half (11; 52.4%), said the other person had reported the activity; 3 (14.3%) said the other person had not reported the activity; and 7 (33.3%) did not know. Twenty respondents (25%) answered Question 9, which asked if the person who had reported the activity had done so because he or she thought reporting was a part of his or her job. Of the 20, 9 (45%) said the other person did not consider it a part of his or her job; 2 (10%) said the other person considered it a part of his or her job; and 9 (45%) did not know if the other person considered it a part of his or her job. Question 10 asked how the person who had reported the activity had done so. Answer choices were internally through the chain of command; internally through an anonymous whistleblowing channel; externally, through channels outside the organization; or through a combination of the three. Respondents could also answer that they did not know how the reporting had occurred. Half of the respondents said the reporting was internal and the other half did not know how the other person had reported the activity. No one knew of anyone who had reported externally. Of the 18 (22.5%) who responded to this question, 6 (33.3%) said the other person had reported internally through the chain of command; 2 (11.1%) said the other person had reported internally through an anonymous channel; and 1 (5.6%) said the other person had reported both internally through chain of command and internally through an anonymous channel. Of the 18 (22.5%) who responded to Question 11, which asked whether the other person who had reported the wrongdoing had been identified as the source of the report, 6 (33.3%) said the other person was not identified; 1 (5.6%) said the other person was 120 identified; and 11 (61.1%) did not know if the other person was identified as the source of the report. Question 12 asked what had happened to the person who had been identified as the source of the report of wrongdoing. Oddly enough, although only one person in the responses to the previous question indicated the other person had been identified as the source of the report of wrongdoing, 4 respondents to this question reported lack of management support for the other person. Of the 16 (20%) who responded to this question, 6 (37.5%) said nothing happened to the other person; 2 (12.5%) said the person’s supervisor was unhappy with him or her; 2 (12.5%) said someone above the person’s supervisor was unhappy with him or her; and 6 (37.5%) did not know what happened. The last question was open-ended. It asked respondents to share any additional information about incidents in which they knew of another person who had been aware of corporate wrongdoing. Thirteen respondents provided comments, which are included throughout the results as appropriate to add context and clarification. In discussing why she had answered “no” when asked if s/he had known of another person who was aware of illegal, wasteful, or unethical activities, a 50-year-old, white female responded: “I answered "no" because the individuals were not aware of the activity at the time it was going on, but rather later.” Based on this response, the researcher determined that this person had known another person who had been aware of wrongdoing. In order to reflect that knowledge, the researcher recoded the response to the previous question (Question 5) as “yes.” 121 A 45-year-old white female who also had responded “no” when asked if she had been aware of another person who knew about illegal, wasteful, or unethical activities, responded: “I saw fraud while at a public relations agency--saw two senior executives charge time and expenses that were not legitimate.” She reported the fraud was greater than $500,000 and happened frequently. Because her response was not about third party knowledge but about her own knowledge, the researcher did not change the coding on this question. A white male of unknown age reported he had known someone who was aware of rarely occurring wrongdoing in the less-than-$50,000 range, who had reported the activity internally through the chain-of-command, and whose co-workers had been unhappy with him: “Episode I am thinking of is not SOX related but manager-level ethical issues involving money. Whistle-blowing was done during exit interviews and ASAIK [as far as I know] no action was taken. I left that company at my earliest opportunity.” In this case, the person who is responding to the survey both knew someone else who reported wrongdoing while employing exit strategy and left rather than report wrongdoing. The second response is an example of exiting the organization as a result of retaliation. A 60-year-old white male commented on a person he knew who had reported occasional wrongdoing of an unknown amount internally through the chain-of-command and who had experienced the unhappiness of someone above his supervisor: “Person cited in previous questions subsequently retired.” Five respondents took advantage of the open-ended question to offer comments about the ethical values of their companies. One, a 52-year-old white male, commented 122 on his own situation, rather than that of a third party: “By and large, ethical behavior rules. I have not witnessed any unethical actions.” A 47-year-old white female commented: “We have signs posted in our buildings to make sure that team members understand how importanat [sic] of an issue ethics is.” A 56-year-old white male commented: “Ethics and ethical behavior is a core value of our company. All employees are educated to report ethical violations immediately, either to their supervisors or to the compliance officer for their business unit.” A 64-year-old, white male commented: “I've worked in PR for 41 years for two Fortune 500 companies. Both had very ethical management.” A 34-year-old white female commented: “We have annual Code of Conduct training - all employees must participate.” One other comment worth noting is that of a 52-year-old white male, who objected to the previous questions about awareness of another person with knowledge of wrongdoing because, he said: “involvement of PR people was BECAUSE of their role, not as the person observing or reporting. Therefore virtually all questions in this section do not apply.” His comment appears to assert that the role of the public relations person does not include observing or reporting wrongdoing. Personal Awareness of Wrongdoing Question 14 through Question 21 mirrored Questions 6 through 13 but addressed personal awareness of wrongdoing in the corporation in which he or she currently was employed. Fewer respondents reported personal awareness of illegal, wasteful, or unethical activity their own organizations than reported second-hand knowledge of the activity. Of the 65 (81.8%) who responded to Question 14, 14 (21.5%) said they had been 123 aware of illegal, wasteful, or unethical activity in their own company, and 51 (78.5%) said they had not. More than 50% of respondents to Question 15 were aware of wrongdoing valued at more than $100,000 in their own companies. Of the 14 (17.5%) who reported the dollar amount involved in the wrongdoing activity, 2 (14.3%) said the dollar amount was between $1,000 and $49,999; 2 (14.3%) said it was between $50,000 and $99,999; 5 (50%) said it was between $100,000 and 500,000; and 1 (4.3%) said it was more than $500,000. Seven (50%) of the 14 (17.5%) respondents to Question 16, which asked about the frequency of the wrongdoing in the respondent’s own company, said the wrongdoing happened occasionally or frequently; 5 (35.7%) said the wrongdoing happened rarely; and 2 (14.3%) did not know. Almost three-quarters of the respondents who personally knew about wrongdoing reported it to an individual or group. Of the 14 (17.5%) who responded to Question 17, 10 (71.4%) said they had reported the wrongdoing activity, and 4 (28.6%) said they had not. The majority of respondents who reported the activity did not do so because they thought reporting was a part of their job. Of the 13 (16.3%) who responded to Question 18, 11 (84.6%) said reporting the activity was not a part of his or her job, and 2 (15.4%) said it was. Respondents who reported wrongdoing did so almost exclusively through internal channels, and the majority used the chain of command. Of the 11 (13.8%) respondents to Question 19, 7 (63.6%) said they had reported internally through chain-of-command; 2 124 (18.2%) said they had reported internally through an anonymous channel; 1 (9.1%) said he or she had reported internally through chain-of-command and internally through an anonymous channel; and 1 (8.9%) said he or she had reported externally. Eight (10%) of the 11 (13.7%) respondents to Question 20 said they were identified as the source of the report of wrongdoing, and 3 (27.3%) said they were not. Although only 8 (10%) said they were identified as the source of the report of wrongdoing, 10 (12.5%) told what happened to them after being identified as the source of the report (Question 21). Of those, 7 (70%) said nothing happened; 1 (10%) said his or her supervisor was unhappy; and 2 (20%) experienced retaliation. The last question in this series was open-ended to allow respondents to share additional information. Of the 5 who responded to Question 22, three commented on whistleblowing; one commented on ethics training; and one commented on public relations roles. The 52-year-old white female who previously reported seeing fraud at a public relations agency reported that she had been retaliated against and had left the organization as a result: “One person who committed the fraud was fired. The other was not, and she retaliated against me. I eventually left the firm.” A white male who previously reported he had known another person who was aware of a non-Sarbanes-Oxley-related issue involving managerial misuse of funds expanded on his previous statement: Need to clarify: We have an anonymous Ethics hotline. When there was an alleged breach of ethics against an employee it was reported to the client, a public-sector entity in an open-records state. Upon learning of the breach, our attorneys alerted PR and told the client, knowing that this would result in our termination. When the media asked for details we worked with the client to provide accurate factual information. I have no knowledge of any wrongdoing by the company itself on a SOX related matter or anything else. 125 This report appears to concern an ethics breach in this respondent’s company, a public relations agency hired by a government agency. The public relations agency had an ethics hotline; it was an employee of the public relations agency who was guilty of the ethics breach; and it was the public relations agency that experienced repercussions from the wrongdoing. A review of this respondent’s other answers reveals he reported an incident of wrongdoing externally. However, it is not possible to know if he was the person who reported the ethics breach through the hotline and characterized that as external reporting, or if he characterized the “accurate factual information” he provided to the media as external reporting. The issue of public relations roles was again raised by the 52-year-old white male who objected to the first line of questioning about third-party knowledge of wrongdoing on the basis that it was his job to deal with situations involving wrongdoing, but it was not his job to report them: Again, in my position, I am involved in responding to and addressing most significant incidents in this regard. So most of these questions really do not apply at ALL. And I am becoming concerned in responding to this questionnaire, as I fear answers may not be understood. remember, in this position, we are involved in addressing and/or managing almost every incident of this nature. How Prevalent Is the Wrongdoing? Total awareness of wrongdoing was 44.4% (Table 8). The researcher computed total awareness of wrongdoing as a surrogate for the amount of illegal, unethical, or wasteful activity in the environment. In this study, total awareness of wrongdoing was the number of respondents who knew of someone who knew of wrongdoing, or who knew of wrongdoing themselves, divided by the number of respondents. This procedure followed the practice in management literature. 126 Table 8. Total Awareness of Wrongdoing Frequency Percent Valid Percent Cumulative Percent Not Aware of Wrongdoing 40 50.0 55.6 55.6 Aware of Wrongdoing 32 40.0 44.4 100.0 Valid Total 72 90.0 100.0 Missing System 8 10.0 Total 80 100.0 How Serious Is the Wrongdoing? Although more than one-half (n = 18; 56.3%) of the incidents of wrongdoing were identified as least serious, one-quarter (n = 8; 25% were identified as most serious (Table 9). That means they involved $100,000 or more or they happened frequently. This study uses the same technique used in management studies to calculate the seriousness of the wrongdoing by looking at the dollar amount and frequency of the activity reported. Incidents involving less than $1,000 were labeled least serious; incidents involving $1,000 to $99,999 were labeled somewhat serious; and incidents involving $100,000 or more were labeled most serious. If no dollar amount was given, the frequency level was considered. Rarely was labeled least serious; occasionally was labeled somewhat serious; and frequently was labeled most serious. Table 9. Seriousness of Wrongdoing Frequency Percent Valid Percent Cumulative Percent Least serious 18 22.5 56.3 56.3 Somewhat serious 6 7.5 18.8 75.0 Most serious 8 10.0 25.0 100.0 Valid Total 32 40.0 100.0 Missing System 48 60.0 Total 80 100.0 127 Was There Retaliation for Reporting Wrongdoing? Almost one-third (n = 7; 31.8%) of those who reported wrongdoing experienced retaliation (Table 10). Comprehensiveness of retaliation was calculated on the basis of existence or absence of retaliation, which included lack of management and co-worker support, as well as incidents the respondent saw as retaliation. If the respondent knew of someone who had experienced one of those situations after reporting wrongdoing, or if the respondent had experienced one of those situations after reporting wrongdoing, retaliation was deemed to have occurred. Table 10. Comprehensiveness of Retaliation Frequency Percent Valid Percent Cumulative Percent No Retaliation 14 17.5 66.7 66.7 Retaliation 7 8.8 33.3 100.0 Valid Total 21 26.3 100.0 Missing System 59 73.8 Total 80 100.0 Elements of Power In prior management studies of whistleblowing, the key constructs of the variable of power were age, education, race, professional position, and salary. Gender was significant only in studies in which the population was large enough (N = 3,288) to allow for advanced multivariate statistical techniques, such as structural equation modeling. The results of this study suggest that age has its privileges; older public relations executives hold more power than younger executives. Post-hoc Chi-square analyses revealed significant differences for age group and salary, supervision, and tenure with the company. The proportion of executives in the older age group who earned $200,000- 128 $500,000 was significantly higher than those in the younger age group (χ2(3, N = 49) = 14.108, p < .05); the proportion of executives in the older age group who supervised others was significantly higher than those in the younger age group (χ2(1, N = 50)= 4.756, p < .05); and the proportion of executives in the older age group who had been with the company 11-49 years was significantly higher than those in the younger age group (χ2(1, N = 46) = 3.969, p < .05). The researcher conducted a direct logistic regression analysis on total awareness of wrongdoing (a surrogate for wrongdoing), using the power indicators of gender, supervision, position, age, reporting relationship, performance rating, and education. A test of the full model with all seven predictors against a constant-only model was statistically reliable (χ2(7, N = 49) = 14.986, p < .05), indicating that the predictors distinguished between those who were aware of wrongdoing and those who were not aware of it. Table 11 shows the regression coefficients, Wald statistics, odds ratios, and 95% confidence intervals for odds ratios for each of the seven predictors. According to the Wald statistic, both gender (being male; z = 4.186, p < .05) and rank (being the top- ranking public relations executive; z = 2.089, p < .05) were reliable predictors of total awareness of wrongdoing. In other words, male, top-ranking public relations executives were more likely than other public relations executives to be aware of wrongdoing. 129 Table 11. Logistic Regression on Total Awareness of Wrongdoing 95% C.I.for EXP(B) B S.E. Wald df Sig. Exp(B) Lower Upper male(1) 1.685 .824 4.186 1 .041 5.395 1.073 27.114 supervise(1) .303 .989 .094 1 .759 1.355 .195 9.402 rank2(1) 2.089 .829 6.356 1 .012 8.077 1.592 40.978 age .058 .052 1.248 1 .264 1.060 .957 1.175 CEO .113 .496 .052 1 .820 1.120 .423 2.963 rating -.002 .450 .000 1 .997 .998 .413 2.413 education -.772 .441 3.070 1 .080 .462 .195 1.096 Step 1a Constant -.895 4.482 .040 1 .842 .408 a. Variable(s) entered on step 1: male, supervise, rank2, age, CEO, rating, education. A Chi-square analysis revealed no significant findings for gender with age group (χ2(1, N = 50) = 2.285, p = .131), education (χ2(2, N = 56) =.764, p = .683), salary (χ2(3, N = 53) = 1.980, p = .577), race or ethnicity (χ2(4, N = 56) = 3.922, p = .417), and position as the top-ranking public relations executive (χ2(1, N = 56) = .042, p = .838). In other words, there were no differences between men and women in any of those categories. The next section of the survey, Section IV, examined the relationship the respondents had with their employers. Relationships: Outcomes and Types The researcher used the Hon and Grunig (1999) short scale to measure relationship quality outcomes in terms of trust, satisfaction, control mutuality, and commitment and to measure two relationship types, exchange and communal. In the short scale, four items each measured satisfaction, control mutuality, and commitment; six items measured trust; and four items each measured exchange and communal 130 relationships. All measures employed five-point Likert scales. The researcher used Cronbach’s alpha to measure the internal reliability (consistency) of the scale. An alpha of .7 or above is usually considered acceptable (Hon & Grunig, 1999). As shown in Table 12, the commitment, trust, control mutuality, satisfaction, and communal relationship scales demonstrated internal reliability above acceptable levels. However, the exchange relationship scale was not reliable. The last statement in the exchange relationship scale has a low item-correlation (Table 13). For that reason, the researcher removed it from the exchange scale, which increased scale reliability to acceptable levels (α=.771). Table 12. Relationship Scale Cronbach’s Alpha N Scale Mean Standard Deviation Cronbach's Alpha Trust 57 24.32 4.067 .850 Control Mutuality 55 16.49 2.693 .799 Commitment 55 16.42 3.201 .856 Satisfaction 58 16.90 2.580 .758 Communal Relationship 58 17.38 2.628 .706 Exchange Relationship 56 13.20 2.799 .642 131 Table 13. Exchange-Total Statistics Scale Mean if Item Deleted Scale Variance if Item Deleted Corrected Item-Total Correlation Squared Multiple Correlation Cronbach's Alpha if Item Deleted Whenever this organization gives or offers something to me, it generally expects something in return. 9.86 4.270 .518 .450 .499 Even though I have had a relationship with this organization for a long time, it still expects something in return whenever it offers me a favor. 10.25 4.009 .645 .487 .395 This organization will compromise with me when it knows that it will gain something. 10.55 4.179 .541 .302 .479 This organization takes care of people who are likely to reward the organization. 8.93 7.268 .010 .053 .771 Relationship Quality Outcomes As shown in Table 14, respondents rated satisfaction (M = 4.22, SD = .645) most highly, followed by control mutuality (M = 4.12, SD = .673), commitment (M = 4.10, SD = .800), and trust (M = 4.05, SD = .678). The communal relationship scale rated higher (M = 4.34, SD = .657) than the 3-item exchange relationship scale (M = 2.98, SD = .899). 132 Table 14. Descriptive Statistics of Relationship Outcomes and Types N Range Mean Std. Deviation Cronbach’s Alpha Satisfaction 58 2.50 4.22 .645 .758 Control Mutuality 55 2.50 4.12 .673 .799 Commitment 55 3.00 4.10 .800 .856 Trust 57 2.67 4.05 .678 .850 Communal Relationship 58 2.50 4.34 .657 .706 Exchange 3-item 56 4.00 2.98 .899 .771 Valid N (listwise) 53 The means of responses to the relationship factors (M = 4.1, SD = 2.48) indicate public relations executives have a favorable view of their relationships with their corporations (Table 15). 133 Table 15. Descriptive Statistics of Relationship Quality Outcomes by Question Variables N Range Mean SD Satisfaction Both the organization and I benefit from the relationship. 59 4 4.42 .792 Most people like me are happy in their interactions with this organization. 58 4 4.03 .878 I am happy with this organization. 58 3 4.22 .839 Generally speaking, I am pleased with the relationship this organization has established with me. 58 4 4.21 .874 Control Mutuality This organization believes my opinions are legitimate. 61 4 4.33 .870 In dealing with me, this organization has a tendency to throw its weight around. [R] 58 4 3.83 1.028 This organization really listens to what I have to say. 58 3 4.14 .736 This organization and I are attentive to what each other say. 56 2 4.14 .672 Commitment I feel that this organization is trying to maintain a long-term commitment to me. 63 4 4.10 .946 There is a long-lasting bond between this organization and me. 61 4 3.90 1.121 I can see that this organization wants to maintain a relationship with me. 58 4 4.19 .926 Compared to other organizations, I value my relationship with this organization more. 56 4 4.23 .894 Trust This organization treats me fairly and justly. 63 4 4.33 .880 Whenever this organization makes an important decision, I know it will be concerned about me. 60 4 3.30 1.030 This organization can be relied on to keep its promises. 58 4 4.12 .900 I believe that this organization takes my opinions into account when making decisions. 58 3 4.07 .792 I feel very confident about this organization’s skills. 57 3 4.12 .927 This organization has the ability to accomplish what it says it will do. 57 3 4.37 .747 3.54 4.1 2.48 Note: N = sample size after pairwise deletion for missing variables. [R] = the statement was reverse coded. 134 Almost all respondents (93.2%) viewed their relationship with their organization as mutually beneficial (M = 4.42, SD =.792), a concept that measures satisfaction. Well more than three-quarters (84.5%) of respondents rated their own happiness with the organization as high (M = 4.22, SD = .839), and 82.8% were pleased with the relationship the organization had established with them (M = 4.21, SD = .874). Respondents had a slightly lower rating for the happiness they thought others who were like them felt for the organization (M = 4.03, SD = .878), but it was still a highly favorable rating. Respondents rated control mutuality slightly lower than satisfaction. Among the measures of control mutuality, most (90.2%) respondents rated highest their view that the organization believed their opinions were legitimate (M = 4.33, SD = .870). Slightly fewer (84%) respondents (M = 4.14, SD = .672) believed the organization and respondent were attentive to each other, and almost as many (82.8%) respondents (M = 4.14, SD = .736) thought the organization listened to the respondent. Respondents gave the lowest rating (M = 3.83, SD = 1.028) to the concept that the organization throws its weight around (this question was reversed-coded). Respondents rated commitment slightly lower than control mutuality. They rated their relationship with their organization highest (M = 4.23, SD = .894) and their belief that the organization wanted to maintain a relationship with them slightly lower (M = 4.19, SD = .926). More than three-quarters (78.8%) of respondents agreed that the organization was trying to maintain a long-term commitment to them (M = 4.10, SD = .946), but slightly fewer (75.4%) agreed there was a long-lasting bond between them and their organizations (M = 3.90, SD = 1.121). 135 Trust rated lowest of the four relationship outcomes. A belief in the organization’s ability to accomplish what it says it will do rated highest (M = 4. 37, SD = .747), followed closely by a belief that the organization treated the respondent fairly and justly (M = 4.33, SD = .880). Respondents’ belief that the organization keeps its promises (M = 4.12, SD = .900) tied with their confidence in the organization’s ability to do what it says it will do (M = 4.12, SD = .747). However, respondents were less confident that the organization took their opinions into account when making decisions (M = 4.07, SD = .792), and respondents had even less confidence that the organization was concerned about them when it made decisions (M = 3.30, SD = 1.030). Relationship Types As shown in Table 16, most respondents strongly disagreed with the concept that their organizations get ahead by stepping on others (M = 4. 57, SD = .840). Overall, respondents had good relationships with their employers, as signified by higher means for the communal relationship statements than for the exchange relationship statements. That interpretation was further strengthened by significant positive correlations between the communal relationship scale and relationship outcomes, and negative correlations between the 3-item exchange relationship scale and relationship outcomes. 136 Table16. Descriptive Statistics of Relationship Types by Question Variables N Range Mean SD Communal This organization does not especially enjoy giving others aid. [R] 62 3 4.42 .933 This organization is very concerned about my welfare. 58 3 3.98 .868 I feel that this organization takes advantage of people who are vulnerable. [R] 58 3 4.40 .954 I think that this organization succeeds by stepping on other people. [R] 58 3 4.57 .840 4.34 .899 3-item Exchange Whenever this organization gives or offers something to me, it generally expects something in return. 59 4 3.31 1.087 Even though I have had a relationship with this organization for a long time, it still expects something in return whenever it offers me a favor. 57 4 2.93 1.050 This organization will compromise with me when it knows that it will gain something. 57 4 2.65 1.094 2.96 1.077 Note: N = sample size after pairwise deletion for missing variables. [R] = the statement was reverse coded. The researcher conducted independent-samples t-tests at the 95%, 90%, and 85% levels on scores for the four relationship quality outcomes (trust, control mutuality, commitment, and satisfaction) and the communal relationship type scale and the 3-item exchange relationship scale with gender, age group, and ranking. She found no significant differences. There were no detectable differences between males and females, those between 31 and 46 and those between 46 and 64, and those who were the top- ranking public relations executives and those who were not in terms of their perceived trust, control mutuality, commitment and satisfaction with their organization, or their view of the types of relationships they have with their organizations. 137 Summary of Findings Respondents were primarily middle-aged, well-educated, well-paid white males who were new to their companies, worked in corporate communications departments, and held highly favorable views of their relationships with their employers. The amount of wrongdoing in large corporations eight years after the enactment of Sarbanes-Oxley fraud sanctions and whistleblowing protections was comparable to that found in the federal government in 1980. About one-quarter of the wrongdoing was considered serious either in dollar amount or frequency of occurrence. Top-ranking male public relations executives were more likely than others to be aware of wrongdoing but were no more likely to report it (i.e., blow the whistle). Public relations executives did not consider reporting wrongdoing part of their jobs. However, those who did report wrongdoing almost always reported internally rather than externally. Those who reported wrongdoing generally did not report retaliation, and those with power reported less retaliation than others. However, when retaliation did occur, informal retaliation in the form of lack of management support most frequently occurred. Respondents generally experienced good relationships with their employers. 138 CHAPTER V DISCUSSION This study made five original contributions to the existing research on whistleblowing. It administered an adaptation of the original federal whistleblowing survey developed by the U.S. Merit Systems Protection Board (U.S. Merit Systems Protection Board, 1981) to individuals in the private sector. It extended the management research on whistleblowing into public relations. It gathered data on the role of public relations departments in the implementation of Sarbanes-Oxley whistleblowing provisions. It explored whistleblowing as a role-prescribed function of public relations practice. And it tested relationship management theory in the context of whistleblowing. The study also provided a set of data on whistleblowing in the Fortune 1000 and on public relations executives’ relationships with their corporations that has implications for relationship management theory, public relations roles theory, the excellence study, and resource dependence perspective. Study Limitations The limitations the researcher encountered resulted, in part, from the fact that a sampling frame for her target population, public relations executives in the Fortune 1000, does not exist. As a result, it became necessary for the researcher to deduce the names of public relations executives, or those responsible for public relations functions, from business databases, corporate Web sites, and other corporate communications vehicles. The lack of an independent profile for the target population made demonstrating the validity of the sample difficult. The researcher also adapted a mail survey designed three decades ago for the federal government to an e-mail delivered, Web-based survey for a 139 corporate audience. As noted previously, response rates for Web surveys are down (Couper, 2000), and lower responses rates have been found in surveys directed at corporate executives (Baruch & Holtom, 2008). The overall length of the survey, which reflected an earlier era of survey research when people were more willing to spend time on them, and the electronic protections in place in most corporations to filter e-mail messages from unknown senders may have added to the difficulty. Also of note is a recent U.S. Supreme Court ruling that allows public employers to read employees’ text messages, which may have a negative effect on response rates to online surveys. As noted in a whistleblower blog: “The bottom line for whistleblowers is that they should not conduct whistleblowing or other personal business on computers, phones or other devices provided by their employers” (Renner, 2010, p. 1). The researcher attempted to allay concerns by suggesting that employees forward the survey to personal e-mail accounts. However, doing so could have triggered the type of employer oversight described in the article, “10 things your boss won’t tell you” under the heading, “Yes, we are reading your emails… and your IMs,” which reported that 20% of all large U.S. corporations had fired an employee in the past year for violations of email policies (C. Hill, 2011, p. 1). As the author noted: “They use an email monitoring software to flag certain types of messages and keywords within messages….Every day, they end up reading %5 to 10% of the messages employees send…Most email investigations pertain to issues of employees leaking sensitive, confidential or embarrassing information, or theft” (C. Hill, 2011, p. 1). Further support for the concept of employer oversight of electronic communication came from a public relations industry source who alleged public relations 140 practitioners had been deployed by employers to restrict employee speech and blow the whistle on them: Desk-bound PR pros now are the Gestapo of their employers, monitoring all communications of employees including e-mails and blogs. Anyone remotely connected to the organization is also tracked. PR pros immediately snitch on anyone wavering from the corporate line. They also serve as blockers for their CEOs and keep outgoing information to a minimum. Their names are rarely seen on company websites. They’ve almost become the “spooks” of their organizations, operating in the shadows. (in O'Dwyer, 2010, p. 2) Surveying this population, then, may be increasingly problematic as electronic surveillance of employees becomes more common. Other scholars have noted the difficulty of gaining access to higher levels of corporate personnel to conduct research (B. H. Reber, personal communication, May 24, 2008; B. K. Berger, personal communication, August 9, 2008), and the sensitive nature of the topic of whistleblowing was almost certainly a factor in this study. As one scholar said: "Whistleblowing deals with very sensitive issues from both the individual and the organizational perspective. Such sensitivity limits easy access to information and prevents the use of traditional methodological approaches" (Keenan, 2000, p. 200). Another scholar commented on the difficulty of conducting research on retaliation because of the apprehensions of corporate managers about allowing access and the fears of corporate employees about admitting to whistleblowing: The sensitivity of the topic of retaliation makes it a difficult phenomenon to research regardless of the methodology employed. Corporate leaders are generally not enthusiastic about sponsoring controversial research on wrongdoing in their organizations – research which may itself lead to greater employee unrest. Where access to an organization is gained, the danger perceived by whistle-blowers if they are identified decrease their willingness to participate, despite reassurances of confidentiality. (Rehg, 1998, p. 4) 141 Despite the difficulty of gaining access to this population for research, the research effort remains worthwhile because of the implications for society’s understanding of wrongdoing. The collapse of Enron and WorldCom because of corporate wrongdoing had multi-billion-dollar impacts on stockholders (Brickey, 2008; Moberly, 2008). Information gained about the role played by public relations executives who report to, or who act within, the power structure (i.e., dominant coalition) of large corporations with regard to corporate wrongdoing might help avoid such catastrophic failures in the future, might improve the practice of public relations, and might advance public relations theory and resource dependence perspective. For those reasons, whistleblowing research remains of import to both public relations and management scholarship, despite the inherent difficulties. Much can be gleaned from the data, however. It provides a window into a subset of public relations practice: public relations executives and other personnel who work for the large corporations that make up the Fortune 1000 group of companies. Although this group has been studied by other public relations scholars (e.g., Berger, 2005; Berger & Reber, 2006; Dozier & Grunig, 1992; J. E. Grunig, 1992; L. A. Grunig, et al., 2002), this study represents a different research track by examining whistleblowing among this group and the effect of retaliation on relationships with employers. Answering the Research Questions and Hypotheses Scholars support the idea that whistleblowing is a communication function (Keenan, 1988; Keenan & McLain, 1992; King, 1997; Richardson, 2005; Richardson & McGlynn, 2007), which led to the first research question. 142 RQ1: To what extent have public relations executives of Fortune 1000 corporations been involved in developing and publicizing Sarbanes-Oxley whistleblowing channels? Public relations professionals played a much greater role in publicizing the Sarbanes- Oxley whistleblowing channels (63.1%) than in developing them (21.4%), although a post-hoc analysis found a moderate correlation between the two (r = .396, n = 64, p = .001), Developing the Anonymous Communication Channel In post-hoc analysis, the researcher found a moderate correlation between one’s role in developing an anonymous communication channel and years employed with the company (r = .332, n = 46, p < .05). The findings may indicate that longer-term public relations executives are more likely to be entrusted with the task of developing the anonymous communication channel, or working in conjunction with other departments to develop it, than are shorter-term employees. In one prior study, longer-term employees were found to identify with the organization more than shorter-term employees (Cheney & Christensen, 2001). Additional post-hoc analyses found that working for a company that had established an anonymous communication channel correlated moderately with being aware of another person who knew about wrongdoing (r = .321, n = 73, p = .006) and correlated slightly with total awareness of wrongdoing (r = .253, n = 72, p < .05). Working for a company that had established an anonymous communication channel, however, did not correlate with whistleblowing (r = .055, n = 14, p = .852). These findings are contrary to earlier research that showed knowledge of whistleblowing channels and encouragement of whistleblowing by employers were 143 factors that supported whistleblowing in corporations (Keenan, 2000). However, the differences in findings between this study and Keenan’s may be explained by population surveyed. Keenan surveyed mid- and upper-level managers in all organizational departments in U.S. corporations. In contrast, this study surveyed only public relations executives and practitioners, who may be less willing to report wrongdoing than managers from other organizational departments. Publicizing Whistleblowing Channels Electronic communication, alone or in conjunction with other communication channels, was by far the most widely used (88.8%) method for publicizing the Sarbanes- Oxley whistleblowing channels. This finding is not surprising given the acceptance of e- mail communication in the market place. A recent survey found that 96% of respondents thought the use of e-mail communication would stay the same or increase in the workplace, and 53% thought it was the most effective form of communication (Warren, 2011). However, the question remains whether electronic communication is the best approach to communicating with employees about a topic as sensitive as fraud reporting and anonymous whistleblowing channels. The widespread use of electronic means to communicate with employees has drawbacks, as recent academic researchers have noted: “Information is not the same as communication. The study found that although e-mail is efficient for information exchange, the preference for communication among all groups of employees is still face-to-face interaction” (C. White, Vanc, & Stafford, 2010). The fact that electronic communication, including e-mail and the Internet, is so widely used to publicize whistleblowing channels to employees, then, does not address 144 whether electronic communication is an adequate communication channel for this topic. Keenan (2000) noted that organizations need to communicate their whistleblowing policies and procedures to employees, but recent research indicates organizations may not be communicating these policies and procedures to the extent necessary to make the whistleblowing channels effective. In its annual global study on fraud, the Global Economic Crime Survey, PriceWaterhouseCoopers found that whistleblowing channels were not widely effective in identifying fraud, and it attributed part of that ineffectiveness to inadequate publicity and lack of management support: Globally, 7% of frauds were detected through formal whistle-blowing procedures. This may suggest either the ineffectiveness or the absence of such procedures, which could be due to a lack of support within organizations, insufficient publicity and/or leadership not being seen to take whistle-blowing seriously. (PriceWaterhouseCoopers, 2009, p. 9) Taken together, these results suggest that using electronic communication channels to communicate to employees about whistleblowing may not be sufficient to encourage employees to use those channels. The communications director for the whistleblower protection advocacy group quoted previously also questioned how much publicity organizations provide about whistleblowing out of fear of calling into question how ethical their practices might be. He encouraged organizations to make whistleblowing information available on their Web sites, however (D. Blaylock, personal communication, March 10-11, 2010). Do Public Relations Executives Blow the Whistle? The excellence study has linked the communication manager role as part of, or at least reporting to, the dominant coalition (J. E. Grunig, 1992; L. A. Grunig, et al., 2002), 145 and whistleblowing research at times has linked job level to whistleblowing (Mesmer- Magnus & Viswesvaran, 2005), which led to research question 2. RQ2: To what extent have public relations executives of Fortune 1000 corporations felt it necessary to blow the whistle internally or externally? To report wrongdoing, respondents must first be aware of it. Almost half (n = 32, 44.4%) of those responding reported they or someone they knew was aware of wrongdoing, and two-thirds (n = 21, 65.6%) of those who knew about wrongdoing reported it. This level of awareness of wrongdoing was comparable to the 45% found in the 1980 federal study, almost twice as much as found in the 1983 study (25%), two and one-half times as much as found in the 1992 study (USMSPB, 1993), and higher than the 30% found in a recent worldwide report of corporate fraud (PriceWaterhouseCoopers, 2009). The level of reporting wrongdoing was higher than that found in the 1980 and 1983 (30%) studies and the 1992 (50%) study of federal employees after implementation of whistleblower protections (USMSPB, 1993). In the federal studies, observed wrongdoing decreased and reporting of wrongdoing increased after whistleblower protections were put in place (Near, et al., 1995). Researchers interpreted this to mean that whistleblower protections led to less wrongdoing and more employees willing to report wrongdoing in the federal government (Near & Miceli, 2008; Near, et al., 1995). There is no baseline level for wrongdoing or reporting of wrongdoing among the population for this study. However, the finding that the level of wrongdoing in the corporate sector 8 years after implementation of Sarbanes-Oxley is as high as that found in the federal government 30 years ago raises questions about the effectiveness of Sarbanes-Oxley whistleblowers provisions. 146 Post-hoc Chi-square analyses comparing knowledge of wrongdoing between top- ranking public relations executives and lower-level public relations personnel (χ2(1, N = 56) = 3.700, p = .054) and comparing amount of reporting wrongdoing between the same groups (χ2(3, N = 11) = 7.639, p = .054) were not significant. Top-ranking public relations executives were more likely than other public relations executives to be aware of wrongdoing, z = 2.089, p < .05. That finding lends some support to prior whistleblowing research that demonstrated an association between upper-level management and whistleblowing: Upper-level managers “feel less pressured to conform and thus feel freer to blow the whistle on wrongdoing…Findings of the study indicate that upper-level managers have greater knowledge and information about where to blow the whistle and perceive greater degrees of organizational encouragement of whistleblowing. This relates to their position within the hierarchy of their organization, length of tenure, and greater degree of involvement with policy matters (Katz & Kahn, 1978).” (in Keenan, 2002, p. 29) The findings in this study show that wrongdoing of some degree of seriousness exists in large U.S. corporations despite Sarbanes-Oxley provisions designed to prevent financial fraud and encourage whistleblowing. The findings also suggest that public relations personnel in Fortune 1000 corporations may report wrongdoing at a higher rate than did government employees in the prior federal studies. Do Public Relations Executives Think Blowing the Whistle Is Part of Their Job? Reporting negative news to management is part of the job of an internal auditor, making whistleblowing a role-prescribed function of the job. Internal auditors also have more frequently blown the whistle than other employees (Miceli, et al., 1991). Given the role proscribed for public relations practitioners as boundary-spanners (J. White & Dozier, 1992) and their role in decision-making and informing management about issues 147 (Dozier, 1992; L. A. Grunig, et al., 2002) by excellence theory, public relations executives also could be viewed as role-prescribed whistleblowers. This literature led to research question 3. RQ3: To what extent do public relations executives of Fortune 1000 corporations consider it their responsibility to blow the whistle internally or externally? Public relations executives clearly do not think whistleblowing is part of their job. Of the 21 respondents who said they or another person had reported wrongdoing, only four (19%) said they reported it because reporting was part of their job. A post-hoc analysis found a strong, negative correlation between thinking reporting is part of one’s job and supervision of others (r = -.640, n = 13, p < .05). This finding may mean that supervisors in the corporate structure do not think reporting wrongdoing is part of their jobs. These data suggest that public relations executives do not see themselves as advisers and counselors to management, contrary to what was found in the excellence study (L. A. Grunig, et al., 2002). Public relations executives may view whistleblowing as an external activity and not something that occurs internally through the chain of command or through anonymous whistleblowing channels. Conversely, management may direct internal auditors to report wrongdoing they find in the course of their audits, whereas management directs public relations executives to perform communication functions that do not include reporting wrongdoing as part of their job duties. As researchers have pointed out: “Typical employees, who do not see whistle-blowing as part of their jobs, are not inclined to act when they see wrongdoing” (Miceli, et al., 2008, p. 23). As a respondent in another study reported when asked about reporting wrongdoing 148 through internal channels: “I can’t visualize PR people doing that…I can’t imagine we would report our own executives. Ratting people out just isn’t very acceptable in most contexts” (Berger, 2005, p. 21). How Do Public Relations Executives Blow the Whistle? Previous studies have shown employees are more likely to blow the whistle when the organization encourages whistle blowing and communicates with employees about internal and external whistleblowing channels (Keenan, 2000; Miceli & Near, 1985), which led to research question 4. RQ4: What channels do public relations executives of Fortune 1000 corporations use to blow the whistle internally or externally? Public relations executives who blew the whistle did so almost exclusively through internal channels. Of the 21 who said they or someone else had reported wrongdoing, 20 (95%) said they had reported internally. This lack of external reporting is consistent with earlier studies that found that public relations executives and personnel are strongly opposed to leaking information and blowing the whistle externally (Berger, 2005; Berger & Reber, 2006; Kang & Berger, 2009). This finding is also consistent with findings by other researchers that public relations practitioners prefer to express disagreement with management internally rather than externally: “Practitioners tend to prefer to keep their dissent behaviors aboveboard and inside the organization” (Berger & Reber, 2006, p. 195). This study suggests that public relations practitioners either do not see their role as reporting wrongdoing, have not seen wrongdoing sufficient to require external reporting, or believe that internal reporting mechanisms are adequate to correct any wrongdoing they have reported. 149 Given theory that suggests public relations practitioners serve as advisers to, if not members of, the dominant coalition of decision-makers (Berger, 2005; Berger & Reber, 2006; Bowen, 2008; J. E. Grunig, 1992; L. A. Grunig, et al., 2002), the researcher proposed hypothesis 1. H1: Public relations executives of Fortune 1000 corporations blow the whistle internally through the chain of command more often than they blow the whistle internally through SOX channels. Hypothesis 1 was supported. Of the 21 who told how they or another person reported wrongdoing, 13 (61.9%) reported they had used internal chain of command channels rather than anonymous whistleblowing channels. Although the sample is small, this finding may lend some support to previous work that found public relations executives do not use formal whistleblowing channels: All of the executives interviewed indicated their companies had ethical codes and mechanism for reporting violators, but few could recall actual use or enforcement of such codes. None of the executives interviewed had used this form of resistance, and they felt it was little used by other public relations practitioners. (Berger, 2005, p. 21) The literature suggesting that public relations executives would not use anonymous communication channels led to hypothesis 2. H2: Public relations executives of Fortune 1000 corporations blow the whistle externally more often than they blow the whistle internally through SOX channels. H2 was not supported. Of the 21 respondents who told how they reported wrongdoing, only one (4.7%) reported externally. These findings lend some credence to a prior study’s findings that 85.7% of public relations professionals preferred internal channels over external channels for reporting wrongdoing (Berger & Reber, 2006). As the researchers noted: “The most acceptable response to questionable management decisions was to 150 assertively confront management,” (Berger & Reber, 2006, pp. 175-176). Their respondents said they would rather leave the organization than report wrongdoing externally, a strategy the researchers refer to as exit strategy (Berger & Reber, 2006). Have Public Relations Whistleblowers Experienced Retaliation? The amount of retaliation experienced by whistleblowers to date has been unclear (Miceli, et al., 2008). This lack of information led to research question 5. RQ5: How have public relations executives of Fortune 1000 corporations who have blown the whistle been treated by their Fortune 1000 corporations? Of the 21 individuals who reported how they or others reported wrongdoing, one-third (7, 33.3%) suffered retaliation as defined in this study. Of those, 3 reported that their supervisor was unhappy with them, and two reported that someone above their supervisor was unhappy with them. These five accounts reflect Rehg’s category of informal retaliation: “Lack of support from supervisors and management was the strongest and most consistent predictor of comprehensiveness of retaliation” (Rehg, 1998, p. 120). A post-hoc analysis found that retaliation had a strong, negative correlation with being identified as the source of the report of wrongdoing (r = -.713, n = 9, p < .05). This finding appears to lend support to Rehg’s (1998) finding that retaliation against whistleblowers is rare. Although the absolute numbers are small, they indicate that some level of retaliation exists in Fortune 1000 corporations. In previous whistleblowing research, the power of the whistleblower has appeared at times to confer protection from retaliation (Near, et al., 1995; Rehg, et al., 2004, 2008), which led to hypothesis 3. H3: Public relations executives with power (as measured by supervisory status, position, gender, race, education, tenure, and job performance) who 151 blow the whistle will report less retaliation than other public relations employees. H3 was not supported. Of the 7 respondents who experienced retaliation, 3 (42.9%) were the top-ranking public relations executives in their respective organizations. The finding may suggest that public relations executives do not occupy a position in the dominant coalition as evidenced by their experiencing retaliation from the dominant coalition. As researchers have noted, predicting retaliation is difficult because the whistleblower’s power, the power of the wrongdoer, and the dependence of the organization on whistleblowing all play a role in retaliation (Miceli, et al., 2008; Rehg, 1998). In keeping with resource dependence perspective, an organization’s dependence on wrongdoing, quantified as seriousness of wrongdoing, may predict retaliation (Mesmer-Magnus & Viswesvaran, 2005; USMSPB, 1981, 1984), or it may reduce the protection of power for the whistleblower (Rehg, et al., 2004). These findings led to hypothesis 4. H4: Public relations executives with power who blow the whistle on serious wrongdoing (as measured by dollar amount and frequency) will report more retaliation than those who blow the whistle on less serious wrongdoing. H4 was not supported. Of the three top-ranking public relations executives who reported retaliation, 1 reported the most serious level; 1 reported the least serious level; and 1 gave no information about level of seriousness. The small sample size does not provide enough data to discuss the implications for the larger population of public relations executives who have experienced retaliation. The use of external reporting channels has been a predictor of retaliation in some management studies (Near, et al., 1995; USMSPB, 1981, 1984), which led to hypothesis 5. 152 H5: Public relations executives who blow the whistle externally will report more retaliation than those who blow the whistle internally. Hypothesis 5 was not supported. Only one respondent (4.7%) reported external whistleblowing, and that respondent did not report retaliation. The finding is consistent with public relations research in which most public relations personnel said they would not report wrongdoing externally (Berger, 2005; Berger & Reber, 2006). Lack of supervisor support and lack of management support have predicted retaliation through various federal and management studies (Miceli & Near, 1992; Near, et al., 1995), which led to hypothesis 6. H6: Public relations executives who report retaliation for blowing the whistle will report greater lack of supervisor support and lack of management support than those who do not report retaliation. Hypothesis 6 was not supported. However, of the 7 respondents who reported retaliation in the broader sense, 5 (71.4%) reported some form of management lack of support. Although the numbers are small, they appear to point to a possible relationship between retaliation and lack of support that might be explored more fully with a larger set of data. In addition, one respondent indicated his supervisor was unhappy with him for reporting wrongdoing through an anonymous channel. The contradictions between reporting anonymously, not being identified, and yet having an unhappy supervisor are puzzling. They may indicate that the survey instrument needs further refinement, or they could indicate that the anonymous communication channel the respondent used was less than anonymous. Has Blowing the Whistle Affected Their Relationships with Their Employers? This study used the Hon and Grunig (1999) relationship scales from the public relations literature to assess relationships between public relations executives who have 153 blown the whistle and their organizations. The literature that has shown a relationship between superior-subordinate relationships and retaliation led the researcher to develop research question 6 and hypothesis 7. RQ6: In what ways has retaliation for whistleblowing by public relations executives affected their relationships with their Fortune 1000 corporation employers? The public relations executives who responded to this section of the survey had an extremely favorable view of their relationships with their corporations. However, one respondent objected to the relationship management line of questioning: “I am often perplexed by individuals who feel that their organization 'owes' them anything. This line of questioning seems to reinforce that mindset.” This respondent appeared to be arguing for what one researcher characterized as a market relationship: Maybe the employment relationship amounts to nothing more than the bare legal structure of a principal and an agent under a contract of employment at will with no enduring moral obligations over time. According to this market view, managers are obligated to perform a series of legally defined and ethically constrained obligations. (Soule, 1998) The market relationship described by Soule (1998) is comparable to what Hon and Grunig (Hon & Grunig, 1999, p. 20) identified as an exchange relationship: “In an exchange relationship, one party gives benefits to the other only because the other had provided benefits in the past or is expected to do so in the future.” This study found that the Hon and Grunig (1999) short (4 question) exchange relationship scale was not reliable. This could reflect the difference between the population used to develop the scale, a general population sample, and the population to which it was administered in this study. Ki and Hon (2007) found a comparable reliability issue with both relationship type scales and a population of university students. 154 However, when the researcher removed the last of the 4 statements in the exchange scale, the 3-item scale was reliable (α=.771). This indicates that the 4-item exchange relationship scale should be evaluated and possibly refined to ensure it is measuring what it is intended to measure. H7: Public relations executives who report more retaliation for whistleblowing will report more negative relationships with their organizations (in terms of trust, control mutuality, commitment, and satisfaction) than those who report less retaliation. Hypothesis 7 was not supported. Respondents had very positive relationships with their employers and responses to the relationship outcome scales and the communal relationship scale were quite high. In comparison to a study of young public relations professionals (Gallicano, et al., 2010) born 1982 and later (the Millennial Generation), respondents to this study had higher scores for all relationship outcome measures. Respondents to this study showed greater satisfaction (M = 4.22) with their relationships with their employers than Millennials (M = 3.82), greater control mutuality (M = 4.12) than Millennials (M = 3.77), greater commitment (M = 4.10) than Millennials (M = 3.78), and greater trust (M = 4.05) than Millennials (M= 3.67). Interestingly, both groups gave almost identical low scores to the concept of the organization’s taking their interests into account when making decisions (M = 3.30). Gender has been a concern in whistleblowing research since its inception, but the role of gender in whistleblowing has not yet been undetermined. Gender is a concern in this study because of the increasingly feminized role of public relations (Tsetsura, 2011) That concern led to research question 7. RQ7. Do female public relations executives who blow the whistle experience more retaliation than male public relations executives who blow the whistle? 155 A post-hoc Chi-square analysis found gender was not significant for retaliation (χ2(2, N = 10)= .816, p = .665), but the numbers who reported retaliation (2 males and 1 female) were small. As shown in prior research, gender differences have been found only with large samples (Rehg, et al., 2008). Implications for Public Relations Theory Public relations role theory and excellence theory, which formed the basis for many of the research questions in this study, do not appear to explain many of the study results. Contrary to what would be expected from professionals who work as managerial employees and are part of the dominant coalition, respondents did not consider reporting wrongdoing as part of their jobs. They also reported very strong relationships with their employers. Other theoretical approaches appear to be necessary, then, to account for these results. Public Relations Roles or Organizational Turf This study provides a different view of the traditional communication manager and technician roles, in which communication managers are primarily responsible for communication program results, communication policy, and counseling senior management; and communication technicians are primarily responsible for writing and producing communication materials (Dozier, 1992; L. A. Grunig, et al., 2002; J. White & Dozier, 1992). In this study, public relations professionals reported operating more as technicians than as communication managers on Sarbanes-Oxley implementation. Within the context of roles theory, helping to develop an organization’s Sarbanes- Oxley whistleblowing channels would be an example of operating in the communication manager role, while developing publicity to inform employees about the whistleblowing 156 channel would be an example of operating in the technician role. This finding suggests that the distinction between the communication manager role and the technician role may be less pronounced in practice than was characterized by CEOs and communication managers in prior research (L. A. Grunig, et al., 2002), at least with regard to sensitive topics. This finding also has implications for the role of communication managers as members of the dominant coalition. Excellence theory suggests that organizations that rely predominantly on negotiation and collaboration (i.e., two-way symmetrical communication), rather than persuasion and manipulation (i.e., two-way asymmetrical communication), in their practice of public relations will entrust higher-level communication functions to their public relations leaders and will be more likely to include those managers as members of the dominant coalition of decision-makers (L. A. Grunig, et al., 2002). As one scholar noted: “The CEOs seem to believe that public relations has its greatest value when it fulfills the managerial role specified in the Excellence theory” (L. A. Grunig, et al., 2002, p. 77). The finding that public relations professionals operate more as technicians than as communication managers on the topic of Sarbanes-Oxley implementation could mean that public relations executives are not part of the dominant coalition to the degree predicted by role theory and excellence theory. Conversely, the finding may support the concept that Sarbanes-Oxley implementation is perceived by organizations as a human resources or legal department function, rather than as a public relations function. For example, the public relations director of a national whistleblower support organization objected to the premise that 157 public relations should be involved in the development of the Sarbanes-Oxley whistleblowing channels: But not all communication activities, especially with staff, fall to public relations officials…These things are handled by HR. And I think that whistleblowing processes are for the most part as well. It is the HR, and organizational head’s, responsibility to comply with the law. This is a legal requirement. It is usually not the job of the PR official at an organization to make sure that actions comply with the law…the primary responsibility (by far) falls to the HR position, legal...and Executive Management. And I think that this, in actuality, is where it primarily is handled. (D. Blaylock, personal communication, March 10-11, 2010, emphasis in original). This perspective was supported by at least one member of the target population who declined to participate in the survey: “I was not involved in developing our Sarbanes- 0xley reporting strategy. In our organization this was all handled by our legal department.” The lack of public relations involvement in the development of Sarbanes-Oxley whistleblowing channels lends support to prior research on the encroachment of legal departments into public relations decision-making (Fitzpatrick & Rubin, 1995; J. Lee, Jares, & Heath, 1999; Reber, Cropp, & Cameron, 2001). The relationship between public relations practitioners and lawyers has been historically troublesome. Simon (1969) noted research that showed public relations practitioners saw lawyers as deterrents to their work. Research found public relations people agreed with statements like “Lawyers do not understand the importance of public attitudes” and “Lawyers are too rigid” (pp.8-9). Simon dubbed legal and public relations professionals “the oil and water team” (p.7). More than 30 years later, many believe this rift between the professions remains. (Reber, et al., 2001, p. 187) As one public relations practitioner who is also an attorney phrased it: “Over the years, lawyers have seated themselves importantly at management’s decision-making table while we have waited in an ante room for the news—so we could publicize it after the fact” (Corbin, 1997, p. 17). The ambivalent position occupied by corporate public 158 relations personnel was echoed by a corporate vice president in another study: “Sometimes we’re partners in decision making in the eyes of others: always we are technicians and wordsmiths, we prepare and deliver messages” (in Berger, 2005, p. 14). The way in which public relations executives view themselves, and how they are viewed by the CEO, in terms of dominant coalition membership may not always coincide (Bowen, 2009). Scholars also cite human resources as another function that has encroached on public relations, and encroachment by the human resources and legal functions is thought to occur when the public relations manager lacks a vision of public relations as having a powerful role in the organization (Lauzen & Dozier, 1992). Although, the excellence study did not find many instances of the public relations function reporting to the human resources function (L. A. Grunig, et al., 2002), the views expressed by Blaylock (2010) concerning the responsibility of human resources for employee communication raise questions about the relationship between human resources and public relations in the area of whistleblowing communication. That relationship has not been explored to the degree to which the relationship between public relations and legal departments has been explored, but such research could prove instructive. In order to examine and extend role theory and excellence theory, further research is needed among public relations professionals at all levels to determine how they view the role of public relations, how they define whistleblowing, how they currently receive information about their organization’s whistleblowing policies and channels, how they would like to receive such information, and whether current channels are effective; what they view as their obligations to their employers, their profession, and to society in terms 159 of whistleblowing; under what circumstances they have been, or would be, willing to blow the whistle; what channels they have used or would use and why; and what level of retaliation they have experienced for past acts of whistleblowing. This research could be conducted through surveys delivered to a broad spectrum of public relations professionals at all organizational levels using membership lists of public relations organizations or commercial databases. This study demonstrates that survey research on this topic is difficult to conduct, and employee privacy concerns may increase the difficulty for online survey research. It should also be noted that survey research has other limitations, including offering only a snapshot in time, lacking depth, and offering little context for the data derived. Future research should incorporate qualitative methods in addition to surveys in order to alleviate these concerns and provide a more complete understanding of whistleblowing. Given the problems the researcher encountered with this study, gathering this additional data through surveys could prove difficult, and personal interviews with whistleblowers who have been identified in the media and are willing to discuss their experiences may offer the only avenue to gain additional data. Qualitative research could prove helpful in determining if public relations executives were active in counseling senior management on Sarbanes-Oxley implementation even though they may not have identified themselves as participating in its development. Qualitative research could also more fully explore organizational culture and climate in terms of how supportive the organization is to whistleblowers and to what degree it encourages whistleblowing. Researchers have found that a supportive climate encourages the decision to blow the whistle internally (Keenan, 1988), that upper-level managers “perceive greater degrees of 160 organizational encouragement of whistleblowing” (Keenan, 2002, p. 29), and that organizations can encourage whistleblowing through policies and procedures that promote whistleblowing and sanctions against retaliation for whistleblowing (Miceli, Near, & Dworkin, 2009). All of these topics would lend themselves to qualitative methods of inquiry and provide more complete data on organizational approaches. Future research needs to examine the channels used in whistleblowing because reporting through chain-of-command channels could be a way for public relations executives to gain or retain access in the dominant coalition. Research is needed to determine under what circumstances public relations executives choose to use chain-of- command channels, anonymous whistleblowing channels, and external channels, and what those decisions mean for their relationships with their employers. This research could be conducted using interviews with public relations executives who may not be identified as whistleblowers but who may have been in a position to report wrongdoing, or with those who have been identified as whistleblowers in media reports. In order to further the understanding of public relations roles, additional research is needed to identify which departments handle Sarbanes-Oxley compliance issues, including developing and publicizing the anonymous communication channel for whistleblowers. Phone or mail surveys directed at human resources departments could identify the overall organizational structure within which Sarbanes-Oxley compliance is housed. The survey approach to human resources departments may avoid the emotion- laden issues of whistleblowing and could provide an increased response rate. In addition, Web sites, corporate annual reports, and Securities and Exchange Commission 10-K reports could be reviewed to determine which companies discuss their anonymous 161 whistleblowing channels, how those channels are constructed, and the ways in which those companies promote the channels. In both cases, however, the information gathered would be descriptive only. Whistleblowing as Ethical Obligation A recent study of public relations executives indicated they placed a high value on ethical behavior (Lee & Cheng, 2011), but the current study finds that 81% of public relations executives do not believe reporting wrongdoing is part of their job duties. The finding suggests that public relations executives may be willing to report wrongdoing, but they do not believe that reporting wrongdoing is required of them (i.e., not a role- prescribed function of public relations). Some might argue that a job duty is, by definition, prescribed by management, and that, in this case, management has not made reporting wrongdoing part of the public relations function at their organizations. Reporting wrongdoing, however, also has an ethical component. Some literature has suggested public relations practitioners should provide ethical counsel to management (Bowen, 2008; Bowen, et al., 2006; L. A. Grunig, et al., 2002; Holtzhausen, 2000; Holtzhausen & Voto, 2002). Yet one study found that many public relations practitioners oppose the ethical counselor role that scholars have advocated: “Some practitioners voiced vehement opposition to the idea of public relations acting as ethical or corporate conscience” (Bowen, 2008, p. 285). Others in the same study supported the ethical counselor role: The concurrent emergence of both pro- and antiethical conscience viewpoints in relation to the same set of basic questions reflects the divergence and polarization of views of the role of ethics and ethical responsibility among practitioners. In focus group settings, both perspectives were hotly debated and aroused deeply- held notions about the very nature of the public relations function. (Bowen, 2008, p. 285) 162 This study casts doubt on the acceptance by public relations executives of the role of ethical adviser to management. It suggests that public relations executives seek a seat at the management table, as evidenced by their close working relationships with the dominant coalition, but are unwilling to take an ethical stance about reporting wrongdoing. This dichotomy could reflect the conflict between normative theory and what is possible in practice (Curtin & Boynton, 2001), or it could reflect the division in public relations practice between those who feel strongly that ethical advising is not the role of public relations and those who feel strongly that it is. Further research is needed to understand the views of public relations practitioners on the topic of ethics. For example, public relations executives’ views about reporting wrongdoing could reflect their interpretation of professional codes of conduct, one of which emphasizes loyalty: “We are faithful to those we represent, while honoring our obligation to serve the public interest” (PRSA, 2008, p. 11). Future research should examine how public relations practitioners view the twin admonitions of organizational loyalty and public interest through surveys and personal interviews. The researcher also proposes further research on organization-public relationships to determine if the current view of relationships between organizations and individuals or groups is valid and if the relationship outcomes and relationship type scales measure what they are intended to measure. One approach could be further survey testing of the individual items in each of the relationship scales that might parse out the multiple dimensions of the current measures. 163 The Golden Handcuffs Another notable finding from this study is that these public relations executives with Fortune 1000 companies have highly favorable relationships with their employers, much more so than those recorded for younger agency employees (Gallicano, et al., 2010). Although respondents were highly positive about their organizations and viewed their relationships with their organizations as communal, however, their somewhat lower level of trust in their organizations and their belief that their organizations do not have their best interests at heart suggest that more is going on than might be readily explained by relationship management theory. The notion of the “golden handcuffs,” rewards that bind an employee to an organization, may provide part of the explanation for those apparently conflicting findings. In public relations literature, golden handcuffs has specifically referred to high salaries, extensive benefits, and the power that goes with higher-level positions. Berger (2005) believes the golden handcuffs also have ethical implications for public relations managers: “The pressures of organizational compliance and corresponding material and social benefits that accrue to public relations managers in the dominant coalition may render doing the right thing even more difficult” (Berger, 2005, p. 14). In management literature, the golden handcuffs has referred to a variety of employee benefits, including salary, insurance, fringe benefits, and pension plans (Capelli, 2000; Goldfarb, 2009; Morse, Hall, & Lake, 1997). As one author noted: “Golden handcuff agreements indeed tie an executive to the company, but the reward can be substantial and may even include a share of ownership” (Goldfarb, 2009, p. 1). For 164 example, employee stock ownership plans often reward longevity with the organization (Sengupta, Whitfield, & McNabb, 2007). In this study, the high average salaries; management status, which implies compensation beyond salary; and positive performance reviews reported by respondents suggest that the golden handcuffs are very much present in this population and may offer a more robust explanation for the good relationships public relations executives have with their organizations. Further research is needed to parse out exactly how practitioners might feel beholden to, and invested in, their organizations, and how this affects their relationships with their employers. One possibility is to look at relationships through evolutionary theory (Greenwood, 2010), a fundamental concept of which is reciprocity or the trading of favors (Byrne & Whiten, 1997). Exchange relationships, which demonstrate reciprocity, are clearly evident in nonhuman primate behavior. The evidence for communal relationships, which demonstrate reciprocal altruism, however, is limited almost exclusively to humans. In that respect, perhaps, the normative aspect of symmetrical relationships is accurately portrayed. (Greenwood, 2010, p. 467) When exchange theory is applied to organizations, it is known as resource dependence perspective (Prior-Miller, 1989). As one management researcher noted, resource dependence perspective has been viewed as a theory about dependence on external resources: “The theory’s central proposition is that organizational survival hinges on the ability to procure critical resources from the external environment” (Casciaro & Piskorski, 2005, p. 167). However, Pfeffer and Salancik (2003) acknowledged that the resources on which the organization depends can also be internal, and their position has been demonstrated empirically in a 165 study of the banking industry (Morris, 2007). Much more research needs to be undertaken to understand how internal dependencies affect an organization’s participation in wrongdoing and its response to whistleblowing, but evolutionary theory, the notion of the golden handcuffs, and resource dependence perspective seem best suited to explain the results of this study and to offer an avenue for future research. Conclusion Whistleblowing is a complex issue, much like “international trade, bribery and corruption, emergency preparedness, and global outsourcing,” that could benefit from a cross-disciplinary approach using multiple levels of theory, measurement, and analysis (Hitt, Beamish, Jackson, & Mathieu, 2007, p. 1395). With its emphasis on mandated communication channels, culturally supported whistleblowing, and codes of ethics, Sarbanes-Oxley has built a bridge between public relations and corporate finance, for example, which should not be ignored. The potential for future collaboration on investigations into whistleblowing includes partnerships with management, marketing, psychology, sociology, political science, and law, in addition to public relations, communication, and organizational communication. Challenges to cross-disciplinary studies are inherent in the ways in which different disciplines define terms, the units of analysis they employ, and the theories they use to understand data and phenomena. Public relations literature tends to promote normative theories, such as relationship management theory and excellence theory, that describe how individuals and organizations should behave based on professional practice and ethical considerations. Management literature, on the other hand, tends to rely on theories, such as the theory of the firm and resource dependence perspective, that 166 describe the nature of reality. These two perspectives can be difficult to reconcile unless researchers are aware of the differences, delineate the scope of their research, define terms, and explain their theories in terms of origins and connectivity to the other discipline. For example, the situational theory of publics, a public relations theory that predicts group behavior based on interest in an issue, has its roots in management theory (Verčič & Grunig, 2000). The linkages between theories would prove valuable to the disciplines involved if they were acknowledged and incorporated routinely into cross- disciplinary work. A further suggestion to public relations scholars is to extend theory-building in the area of mid-range theories. The reason for this approach is that neither the descriptive management theories nor the normative public relations theories appear to account for the findings of this study. This discrepancy would argue for a third course, which is that of an instrumental, or predictive, approach to theory-building that might explain the data. Such an approach has been found useful in developing theory in the management field (Berman, Wicks, Kotha & Jones, 1999), and it might prove useful in this arena, as well. One such path to theory-building could begin with evolutionary theory as the “umbrella under which all mid-range public relations theories can shelter” (Greenwood, 2010, p. 469). Within that concept, the notion of the golden handcuffs could be developed to predict behavior within corporations as a function of the current and future rewards offered to executives. Another area of concern in cross-disciplinary research is the unit of analysis that each study employs. Relationship management theory purports to measure the relationship between the individual and the organization. However, the Hon and Grunig 167 (1999) instrument contains questions taken from several disciplines, including interpersonal relationship, marketing, and management (J. E. Grunig & Huang, 2000), and appears to measures relationships at the individual level and the organizational level. Resource dependence perspective is measured at two levels, as well: the organization and the group. However, the unit of analysis for resource dependence perspective is generally considered to be the firm, or the organization (Casciaro & Piskorski, 2005). One way to resolve the issue of varying units of analysis may be to look at how research has been handled in a discipline with connections to public relations and communication. Hitt et al. (2007, p. 1386) describe three areas of management studies that have been conducted with little attempt “to integrate theory or conduct research that crossed these levels of inquiry or analysis.” The disciplinary heritages of scholars working in these areas reinforced their differences. The micro approach was rooted in psychology and focused on understanding the thoughts, feelings, and actions of individuals. The macro approach was rooted in sociology and economics: it focused on understanding organizations and markets. The middle ground was rooted in social psychology and closely related fields such as communications. These authors recommend that organizational studies that address individual-, group-, and firm-level issues should identify the level at which the researcher wants to generalize (the level of theory or focal unit), the level from which the data comes (the level of measurement), and the level at which the data are analyzed (the level of analysis). Confusing these levels results in “fallacies of the wrong level” (Hitt, et al., 2007, p. 1389). Given that resource management theory appears to measure at least some aspects of organization public relationships at the interpersonal level, researchers need to determine at what level or levels the Hon and Grunig (1999) scale is actually measuring 168 relationships. It appears more work is needed to segregate those aspects of the relationship management scale that measure organization-level constructs from those that measure individual- or interpersonal-level constructs. Future studies could characterize findings as related to one or the other construct. Another alternative may be to incorporate aspects of other scales, such the Leader-Member-Exchange (LMX), which has been used in management studies for decades to measure relationships between a manager and an individual employee (Graen & Uhl-Bien, 1995). One question that could be answered at the organizational level is the extent and type of golden handcuffs provided to corporate public relations executives. This data would be important because it would contribute to building a mid-range theory from what currently is only the notion of the golden handcuffs. A corollary question that could be answered at the interpersonal level is the degree to which golden handcuffs influence public relations executives’ ethical decision-making. A recent study of public relations agency employees found a high level of moral reasoning (Coleman & Wilkins, 2009). The discrepancy between those findings and the finding in this study that corporate executives do not view reporting wrongdoing as one of their job duties raises questions about the impact of the monetary rewards given to corporate employees on ethical decisions. Further research is needed to test whether the promise of monetary gain compromises moral reasoning. Additionally, Coleman and Wilkins’ (2009) finding that professionals believed their managers have higher ethical standards then they have suggests that employees may become socialized to corrupt behaviors, resulting in the internalization of those behaviors and the normalization of corruption (Ashforth & Anand, 2003). These questions could be addressed through interviews with the corporate 169 population on the impact of the golden handcuffs on their views of ethical behavior within the corporation. However, the larger issue raised by this study may involve the type of reasoning that occurs when those with professional training address issues of ethics. Coleman and Wilkins (2009) reported that increased levels of education have been found to lead to higher levels of moral reasoning. However, it may be possible that increased levels of education in professional practices, such as public relations, may also lead to a greater ability to see all sides of an issue, to make finer distinctions among options, and, possibly, to convince oneself that one’s course of action, because it is consistent with the organizational direction, is ethical. Despite the inherent difficulties in studying whistleblowing, the effort needs to be made because of the import of the issue to society. What this study has demonstrated is that Fortune 1000 companies do not have a culture that actively promotes and sanctions whistleblowing. Instead, the inclination is still to shoot the messenger. In looking ahead, the prognosis for the whistleblower is not promising. While legally federal civil servants may have technical protections to rely upon, they will likely be subjected to a continuing culture that brands them as disloyal if they choose to report questionable activities. Until the idea of disclosure is seen as an ethical benefit for the entire organization, rather than a violation of trust perpetrated by the whistleblower, civil servants will remain reluctant to come forward. Ultimately, lingering perceptions of whistleblowing is all about shooting the messenger, rather than heeding the content of the message. The fact that the whistleblower is doing nothing more than handing off information as they know it to an authority illustrates the ultimate irony. The whistleblower is just the messenger wondering why they end up shot. (Saunders & Thibault, 2008, p. 22) Thus wrongdoing continues to occur, despite legislation such as Sarbanes-Oxley that is designed to help prevent it. Researchers into whistleblowing should be both comforted and concerned to know that the human condition will provide opportunities for research 170 into organizational wrongdoing, individual whistleblowing, and organizational and individual retaliation for the foreseeable future. 171 APPENDIX A QUESTIONNAIRE FOR RANKING PUBLIC RELATIONS EXECUTIVE IN FORTUNE 1000 CORPORATION SECTION I 1. Does your corporation have an anonymous communication channel to the audit committee of the board of directors? (Please select ONE box.) 1. No 2. Yes 2. If your organization has such a channel, did you or someone who worked for you in public relations help develop it? (Please select ONE box.) 1. No 2. Yes 3. If your organization has such a channel, did you or someone who worked for you in public relations help publicize it? (Please select ONE box.) 1. No 2. Yes 4. If your organization has such a channel, which of the following have been used to publicize it? (Please select ALL that apply.) 1. Electronic communication (e-mail/Internet/Intranet) 2. Print communication 3. Verbal communication 4. Other (Please specify) 5. Not publicized SECTION II 5. During your entire career, have you ever been aware of a public relations executive other than you who was aware of an illegal, wasteful, or unethical activity involving the organization for which he or she worked? (Please select ONE box.) 1. No 2. Yes 172 6. What was the dollar amount of this activity? (Please select ONE box.) 1. <$1,000 2. $1,000-$100,000 3. >$100,000 4. Don’t know 7. How frequently did this activity occur? (Please select ONE box.) 1. Rarely 2. Occasionally 3. Frequently 4. Don’t know 8. Did this person report this activity to any individual or group? (Please select ONE box.) 1. No 2. Yes 3. Don’t know 9. If this person reported this activity, was it because reporting such activity was a routine part of his or her job? (Please select ONE box.) 1. No 2. Yes 3. Don’t know 10. If this person reported this activity how did he or she report it? (Please select ALL that apply.) 1. Internally through chain of command 2. Internally through anonymous channel 3. Externally 4. Don’t know 11. If this person reported this activity, was this person identified as the source of the report? (Please select ONE box.) 1. No 2. Yes 3. Don’t know 173 12. If this person was identified as the source of the report, what happened to this person? (Please select ALL that apply.) 1. Person was rewarded. 2. Nothing happened to person. 3. Person’s co-workers were unhappy with him/her. 4. Person’s supervisor was unhappy with him/her. 5. Someone above person’s supervisor was unhappy with him/her. 6. Person experienced retaliation. 7. Don’t know 13. What else would you like to share about this topic? (Please do not include any identifying information.) ________________________________________________________________ SECTION III 14. During your career with your current employer, did you ever personally observe or obtain direct evidence of an illegal, wasteful, or unethical activity involving your organization? (Please select ONE box.) 1. No 2. Yes 15. What was the dollar amount of this activity? (Please select ONE box.) 1. <$1,000 2. $1,000-$100,000 3. >$100,000 4. Don’t know 16. How frequently did this activity occur? (Please select ONE box.) 1. Rarely 2. Occasionally 3. Frequently 4. Don’t know 17. Did you report this activity to any individual or group? (Please select ONE box.) 1. No 2. Yes 174 18. If you reported this activity, was it because reporting such activity is a routine part of your job? (Please select ONE box.) 1. No 2. Yes 19. If you reported this activity, how did you report it? (Please select ALL that apply.) 1. Internally through chain of command 2. Internally through anonymous channel 3. Externally 20. If you reported this activity, were you identified as the source of the report? 1. No 2. Yes 21. If you were identified as the source of the report, what happened to you? (Please select ALL that apply.) 1. I was rewarded. 2. Nothing happened to me. 3. My co-workers were unhappy with me. 4. My supervisor was unhappy with me. 5. Someone above my supervisor was unhappy with me. 6. I experienced retaliation. 22. What else would you like to share about this topic? (Please do not include any identifying information.) ________________________________________________________________ 175 SECTION IV Please answer the following questions as they pertain to your relationship with your CURRENT employer. Responses range from Strongly Disagree to Strongly Agree, with 3 being Neutral. For example, if you strongly agree with a statement, select “5.” If you feel neutral about a statement, select “3.” Strongly Disagree Somewhat Disagree Neither Agree nor Disagree Somewhat Agree Strongly Agree 1 2 3 4 5 ___ 23. I feel that this organization is trying to maintain a long-term commitment to me. ___ 24. This organization treats me fairly and justly. ___ 25. This organization does not especially enjoy giving others aid. ___ 26. There is a long-lasting bond between this organization and me. ___ 27. This organization believes my opinions are legitimate. ___ 28. Whenever this organization makes an important decision, I know it will be concerned about me. ___ 29. Whenever this organization gives or offers something to me, it generally expects something in return. ___ 30. Both the organization and I benefit from the relationship. ___ 31. This organization is very concerned about my welfare. ___ 32. Even though I have had a relationship with this organization for a long time, it still expects something in return whenever it offers me a favor. ___ 33. I can see that this organization wants to maintain a relationship with me. ___ 34. In dealing with me, this organization has a tendency to throw its weight around. ___ 35. This organization can be relied on to keep its promises. ___ 36. Most people like me are happy in their interactions with this organization. ___ 37. I feel that this organization takes advantage of people who are vulnerable. 176 ___ 38. This organization really listens to what I have to say. ___ 39. I am happy with this organization. ___ 40. I think that this organization succeeds by stepping on other people. ___ 41. I believe that this organization takes my opinions into account when making decisions. ___ 42. Generally speaking, I am pleased with the relationship this organization has established. ___ 43. This organization will compromise with me when it knows that it will gain something. ___ 44. I feel very confident about this organization’s skills. ___ 45. Compared to other organizations, I value my relationship with this organization more. ___ 46. This organization takes care of people who are likely to reward the organization. ___ 47.This organization has the ability to accomplish what it says it will do. ___ 48. This organization and I are attentive to what each other say. 49. What else would you like to share about this topic? (Please do not include any identifying information.) ________________________________________________________________ SECTION V 50. How many years have you worked for this employer? ___ 51. How many years have you been in your current position? ___ 177 52. Are you: (Please select ONE box.) ___ Male ___ Female 53. What is your age? ___ 54. What is your highest education level? (Please select ONE box.) ___ No college ___ Some college ___ A 4-year college degree ___ Some graduate or professional school ___ Graduate or professional degree 55. What is your current salary range? (Please select ONE box.) ___Under $100,000 ___ $100,000 to $199,999 ___ $200,000-to $500,000 ___ More than $500,000 56. Are you: (Please select ALL that apply.) ___ American Indian/Alaskan Native ___ Asian/ Asian American ___ Black or African American ___ White, not Hispanic ___ Hispanic/Latino ___ Other 57. Are you the highest-ranking public relations official in your organization? (Please select ONE box.) ___ Yes ___ No 58. What is the location of your position in relation to the Chief Executive Officer? ___I am the Chief Executive Officer ___I report to the Chief Executive Officer ___I Report to someone who reports to the Chief Executive Officer ___I report to someone who reports to someone who reports to the Chief Executive Officer ___Other. 178 59. In what department do you work? ___Public Relations ___Corporate Communications ___Marketing ___Sales ___Finance & Administration ___Other. 60. Do you now write performance appraisals for other employees? ___ Yes ___ No (Note: The answer to the following question has been found to be significant in some studies of whistleblowing.) 61. Which of the following most closely describes the performance rating you received at your last appraisal? ___ Unacceptable ___ Minimally successful ___ Fully successful ___ Exceeds fully successful ___ Outstanding Thank you for completing the survey! 179 APPENDIX B RECRUITMENT CONSENT E-MAIL 2-24-10 Dear Public Relations Colleague: As an Accredited member of the Public Relations Society of America, a member of the PRSA College of Fellows, and a doctoral candidate in the School of Journalism and Mass Communication at the University of Oregon, I would like to invite you to participate in my dissertation research. My study involves a survey of public relations executives and organizational response to whistleblowing, including response to the Sarbanes-Oxley Act of 2002. You are eligible to participate in this study because you are the highest ranking public relations executive in your Fortune 1000 corporation. The confidentiality of your information is of prime importance to this study. If you decide to participate, the encrypted link in this email, which is being sent from my University of Oregon email account, will take you to a survey on SurveyMonkey that should take no more than 20 minutes to complete. https://www.surveymonkey.com/s/Fortune_1000_Public_Relations Your responses will be completely anonymous, and your identifying information will not appear anywhere on the survey. Please do not include any identifying information on the survey, including in the open-ended questions. You should take the survey on your home computer if you feel uncomfortable about possible work-related issues. Remember, this is completely voluntary. You may choose to be in the study or not. Completing the questionnaire constitutes your consent to participate. Please complete the survey within one week. Everyone contacted for this study will receive a copy of the Executive Summary of the survey results and analysis after the study is completed. If you are willing to be contacted about future research on this topic, please mail the signed Permission/Release Form to me at my private mailbox. These forms will be stored in a secure location under lock and key. There will be no way for anyone to link your information on the Permission/Release Form to the anonymous survey data. If you have any questions about the study, please email me at cgreenwo@uoregon.edu or call me at 541-968-6180. You may also contact my advisor, Dr. Patricia Curtin, at pcurtin@uoregon.edu or 541-346-3752. If you have any questions regarding your rights as a research subject, please contact the Office for Protection of Human Subjects at the University of Oregon, (541) 346-2510. This Office oversees the review of the research to protect your rights and is not involved with this study. As the ranking public relations executive in your corporation, your participation in this survey is of the utmost importance to research on this topic. You will be doing the 180 profession of public relations and the teaching of public relations a great service by participating in this research. Thank you for your help. Sincerely, Cary A. Greenwood, APR, Fellow PRSA Doctoral Candidate, School of Journalism and Communication University of Oregon 1275 University of Oregon Eugene, OR 97403-1275 541-968-6180 (PH) cgreenwo@uoregon.edu 181 APPENDIX C PERMISSION/RELEASE FORM Thank you for allowing me to contact you about my current research project. I may have future opportunities for you to participate in additional research studies on this topic. If you would be interested in hearing about these opportunities, I would be happy to contact you by phone or email when they become available. Should you change your mind and decide that you do not want to be contacted, let me know at any time and I will destroy this form. If you wish to be contacted, please provide the following information: Name: ________________________________________________________________ Phone number: __________________________________________________________ Email address: ___________________________________________________________ What is the best way to reach you? ___Phone _____Email Is it okay to leave a voicemail about research studies? ____Yes ____No Are there any restrictions on contacting you (e.g., don’t call before 10 a.m.; don’t contact after October 2010, etc.)? Please mail this form to my private mailbox at the following address: Cary Greenwood, 597 Country Club Rd. #17, Eugene, OR 97401. Thank you for your consideration. Sincerely, Cary A. Greenwood, APR, Fellow PRSA Doctoral Candidate School of Journalism and Communication University of Oregon 1275 University of Oregon Eugene, OR 97403-1275 541-968-6180 (PH) cgreenwo@uoregon.edu 182 APPENDIX D FOLLOW-UP REMINDER Dear Public Relations Colleague: Last week you were sent information about an Internet survey on Sarbanes-Oxley and whistleblowing in the Fortune 1000. If you have already completed the survey, thank you for your time! Please feel free to delete this message as it is simply a reminder that your participation in this research is appreciated. If you have not completed the survey, please let me repeat the previous information. As an Accredited member of the Public Relations Society of America, a member of the PRSA College of Fellows, and a doctoral candidate in the School of Journalism and Mass Communication at the University of Oregon, I would like to invite you to participate in my dissertation research. My study involves a survey of public relations executives and organizational response to whistleblowing, including response to the Sarbanes-Oxley Act of 2002. You are eligible to participate in this study because you are the highest ranking public relations executive in your Fortune 1000 corporation. The confidentiality of your information is of prime importance to this study. If you decide to participate, the encrypted link in this email, which is being sent from my University of Oregon email account, will take you to a survey on SurveyMonkey that should take no more than 20 minutes to complete. https://www.surveymonkey.com/s/Fortune_1000_Public_Relations Your responses will be completely anonymous, and your identifying information will not appear anywhere on the survey. Please do not include any identifying information on the survey, including in the open-ended questions. You should take the survey on your home computer if you feel uncomfortable about possible work-related issues. Remember, this is completely voluntary. You may choose to be in the study or not. Completing the questionnaire constitutes your consent to participate. Please complete the survey within one week. Everyone contacted for this study will receive a copy of the Executive Summary of the survey results and analysis after the study is completed. If you are willing to be contacted about future research on this topic, please mail the signed Permission/Release Form to me at my private mailbox. These forms will be stored in a secure location under lock and key. There will be no way for anyone to link your information on the Permission/Release Form to the anonymous survey data. If you have any questions about the study, please email me at cgreenwo@uoregon.edu or call me at 541-968-6180. You may also contact my advisor, Dr. Patricia Curtin, at pcurtin@uoregon.edu or 541-346-3752. If you have any questions regarding your rights 183 as a research subject, please contact the Office for Protection of Human Subjects at the University of Oregon, (541) 346-2510. This Office oversees the review of the research to protect your rights and is not involved with this study. As the ranking public relations executive in your corporation, your participation in this survey is of the utmost importance to research on this topic. You will be doing the profession of public relations and the teaching of public relations a great service by participating in this research. Thank you for your help. Sincerely, Cary A. Greenwood, APR, Fellow PRSA Doctoral Candidate, School of Journalism and Communication University of Oregon 1275 University of Oregon Eugene, OR 97403-1275 541-968-6180 (PH) cgreenwo@uoregon.edu 184 REFERENCES CITED AAPOR. (2008). Standard definitions: Final dispositions of case codes and outcome rates for surveys 5. Retrieved from http://www.aapor.org/AM/ Template.cfm?Section=Standard_Definitions&Template=/CM/ContentDisplay.cf m&ContentID=1273 Andrews, D., Nonnecke, B., & Preece, J. (2003). Electronic survey methodology: A case study in reaching hard-to-involve Internet users. International Journal of Human- Computer Interaction, 16(2), 185-210. Ashforth, B. E., & Anand, V. (2003). The normalization of corruption in organizations. In R. M. Kramer & B. M. Staw (Eds.), Research in organizational behavior: An annual series of analytical essays and critical reviews (Vol. 25, pp. 1-52). Amsterdam: Elsevier. Baruch, Y., & Holtom, B. C. (2008). Survey response rate levels and trends in organizational research. Human Relations, 61(8), 1139-1160. Berger, B. K. (2005). Power over, power with, and power to relations: Critical reflections on public relations, the dominant coalition, and activism. Journal of Public Relations Research, 17(1), 5-28. Berger, B. K., & Reber, B. H. (2006). Gaining influence in public relations: The role of resistance in practice. Mahwah, NJ: Lawrence Erlbaum Associates. Berman, S. L., Wicks, A. C., Kotha, S., & Jones, T. M. (1999). Does stakeholder orientation matter? The relationship between stakeholder management models and firm financial performance. Academy of Management Journal, 42(5), 488-506. Blue Sky Factory. (2010). Email design guidelines. Retrieved February 25, 2010, from http://www.blueskyfactory.com/?s=2010+email+design+guidelines Bowen, S. A. (2008). A state of neglect: Public relations as "corporate conscience" or ethics counsel. Journal of Public Relations Research, 20, 271-296. Bowen, S. A. (2009). What communication professionals tell us regarding dominant coalition access and gaining membership. Journal of Applied Communication Research, 37(4), 418-443. 185 Bowen, S. A., Heath, R. L., & Lee, J. (2006, June). An international study of ethical roles and counsel in the public relations function. Paper presented at the International Communication Association Annual Conference, Dresden, Germany. Brickey, K. F. (2003). From Enron to Worldcom and beyond: Life and crime after Sarbanes-Oxley. Washington University Law Quarterly, 81, 357-402. Brickey, K. F. (2008). From boardroom to courtroom to newsroom: The media and the corporate governance scandals. Journal of Corporation Law, 33(3), 625-663. Broom, G. M., Casey, S., & Ritchey, J. (1997). Toward a concept and theory of organization-public relationships. Journal of Public Relations Research, 9(2), 83- 98. Byrne, R. W., & Whiten, A. (1997). Machiavellian intelligence. In A. Whiten & R. W. Byrne (Eds.), Machiavellian intelligence II: Extensions and evaluations (pp. 1- 23). Cambridge, England: Cambridge University Press. Cangialosi, G. (2010, February 25). The ultimate guide to email. Retrieved April 11, 2010, from http://blog.blueskyfacotry.com Casciaro, T., & Piskorski, M. J. (2005). Power imbalance, mutual dependence, and constraint absorption: A closer look at resource dependence theory. Administrative Science Quarterly, 50, 167-199. Cheney, G., & Christensen, L. T. (2001). Public relations as contested terrain: A critical response. In R. L. Heath (Ed.), The handbook of public relations (pp. 167-182). Thousand Oaks, CA: Sage. Churchill, G. A., & Peter, J. P. (1984). Research design effects on the reliability of rating scales in marketing. Journal of Marketing Research, 21, 360-375. CNNMoney. (2010). Fortune 500 FAQ definitions and explanations. Retrieved March 28, 2011, from http://money.cnn.com/magazines/fortune/fortune500/210/faq/ Cobb, C. (2008). Driving public relations: Chrysler moves PR under the HR umbrella, spurs debate about where PR reports. Strategist, 14(3), 6-11. 186 Coleman, R., & Wilkins, L. (2009). The moral development of public relations practitioners: A comparison with other professions and influences on higher quality ethical reasoning. Journal of Public Relations Research, 21(3), 318-340. Corbin, J. (1997). Lawyers and us—A synergistic relationship. Public Relations Quarterly, 42(4), 15-17. Couper, M. P. (2000). Web surveys. Public Opinion Quarterly, 64(4), 464-494. Couper, M. P., & Miller, P. V. (2008). Web survey methods. Public Opinion Quarterly, 72(5), 831-835. Curtin, P. A., & Boynton, L. A. (2001). Ethics in public relations: Theory and practice. In R. L. Heath (Ed.), The handbook of public relations (pp. 411-421). Thousand Oaks, CA: Sage. Delikat, M. (2008). Corporate whistleblowing in the Sarbanes-Oxley era (3rd ed.). New York: Practicing Law Institute. Devine, T. (1997). The whistleblower's survival guide: Courage without martyrdom. Washington, DC: Fund for Constitutional Government. Dillman, D. A. (2000). Mail and Internet surveys: The tailored design method (2nd ed.). New York: Wiley. Dooley, L. M., & Lindner, J. R. (2003). The handling of nonresponse error. Human Resource Development Quarterly, 14(1), 99-110. Dozier, D. A. (1992). The organizational roles of communications and public relations practitioners. In J. E. Grunig, D. A. Dozier, W. P. Ehling, L. A. Grunig, F. C. Repper & J. White (Eds.), Excellence in public relations and communication management (pp. 327-355). Hillsdale, NJ: Lawrence Erlbaum Associates. Dozier, D. A., & Grunig, L. A. (1992). The organization of the public relations function. In J. E. Grunig (Ed.), Excellence in public relations and communication management (pp. 395-417). Hillsdale, NJ: Lawrence Erlbaum Associates. 187 Ethics Resource Center. (2007). The 2007 national business ethics survey: An inside view of private sector ethics. Arlington, VA: Ethics Resource Center. The False Claims Act Legal Center. (2008). What does “qui tam” mean? Retrieved May 30, 2008, from http://www.taf.org/faq.htm#q2 Gallicano, T. D., Curtin, P. A., & Matthews, K. (In press). I love what I do, but...a relationship management survey of Millennial Generation public relations agency employees. Journal of Public Relations Research. Gibb, J. R. (1961). Defensive communication. Journal of Communication, XI(3), 141- 148. Glazer, M. P., & Glazer, P. M. (1989). The whistleblowers: Exposing corruption in government and industry. New York: Basic Books, Inc. Goldfarb, A. (2009, March 22). “Golden handcuffs” can hold key to locking up top executives. Fenton Report, Retrieved April 29, 2011, from http://www.fentonreport.com/2009/03/22/entrepreneurs/%E2%80%9Cgolden- handcuffs%E2%80%9D-can-hold-key-to-locking-up-top-executives/684 Government Accountability Project. (2008 a). Legislation. Retrieved June 8, 2008, from http://www.whistleblower.org/template/page.cfm?page_id=15 Government Accountability Project. (2008 b). S. 494 - H.R. 1317: Whistleblower Protection Act Amendments. Retrieved June 8, 2008, from http://web.archive.org/ web/20080604151524/http://www.whistleblower.org/template/page.cfm?page_id =146 Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6(2), 219-247. Greenwood, C. A. (2010). Evolutionary theory: The missing link for conceptualizing public relations. Journal of Public Relations Research, 22(4), 456-476. Groves, R. M., Fowler, F. J., Couper, M. P., Lepkowski, J. M., Singer, E., & Tourangeau, R. (2004). Survey methodology. Hoboken, N. J.: Wiley. 188 Grunig, J. E. (Ed.). (1992). Excellence in public relations and communication management. Hillsdale, NJ: Lawrence Erlbaum Associates. Grunig, J. E., & Huang, Y.-H. (2000). From organizational effectiveness to relationship indicators: Antecedents of relationships, public relations strategies, and relationship outcomes. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to the study and practice of public relations (pp. 23-54). Mahwah, NJ: Lawrence Erlbaum Associates. Grunig, L. A., Grunig, J. E., & Dozier, D. M. (2002). Excellent public relations and effective organizations: A study of communication management in three countries. Mahwah, NJ: Lawrence Erlbaum Associates. Hair, J. F., Jr., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5th ed.). Upper Saddle River, N. J.: Prentice Hall. Hall, M. R. (2006). Corporate philanthropy and corporate community relations: Measuring relationship-building results. Journal of Public Relations Research, 18(1), 1-21. Hanke, J. E., & Reitsch, A. G. (1991). Understanding business statistics. Boston: Richard D. Irwin, Inc. Hill, C. (2011, January 3). 10 things your boss won't tell you. SmartMoney. Retrieved from http://www.smartmoney.com/personal-finance/employment/10-things-your- boss-wont-tell-you-1293757446378/#ixzz1AK6TMy3E Hill, J. W. (1958). Corporate public relations: Arm of modern management. New York: Harper & Brothers. Hitt, M. A., Beamish, P. W., Jackson, S. E., & Mathieu, J. E. (2007). Building theoretical and empirical bridges across levels: Multilevel research in management. Academy of Management Journal, 50(6), 1385-1399. Holtzhausen, D. R. (2000). Postmodern values in public relations. Journal of Public Relations Research, 12(1), 93-114. 189 Holtzhausen, D. R., & Voto, R. (2002). Resistance from the margins: The postmodern public relations practitioner as organizational activist. Journal of Public Relations Research, 14(1), 57-84. Hon, L. C., & Grunig, J. E. (1999). Guidelines for measuring relationships in public relations. Gainesville, FL: The Institute for Public Relations. Huang, Y.-H. (2008). Trust and relational commitment in corporate crises: The effects of crisis communicative strategy and form of crisis response. Journal of Public Relations Research, 20, 297-327. Inroads. (2010). Leadership! campaign: Inroads and the promise of corporate diversity. Retrieved February 25, 2011, from http://www.inroads.org/sites/default/files/files/INROADS_Case.pdf Isikoff, M. (2008, December 13). The fed who blew the whistle. Newsweek. Retrieved April 24, 2009, from http://www.newsweek.com/2008/12/12/the-fed-who-blew- the-whistle.print.html Jo, S., & Shim, S. W. (2005). Paradigm shift of employee communication: The effect of management communication on trusting relationships. Public Relations Review, 31(2), 277-280. Kang, J.-A., & Berger, B. K. (2009, August ). The influence of organizational conditions on public relations practitioners' dissent. Paper presented at the Association for Education in Journalism and Mass Communication Conference, Boston. Katz, D. M., & Homer, J. (2008, February 1). WorldCom whistle-blower Cynthia Cooper. CFO Magazine, 38-40. Retrieved November 14, 2008, from http://www.cfo.com/article.cfm/10590507?f=search Keenan, J. P. (1988, August ). Communication climate, whistle-blowing, and the first- level manager: A preliminary study. Paper presented at the Annual Meeting of the Academy of Management, Anaheim, CA. Keenan, J. P. (2000). Blowing the whistle on less serious forms of fraud: A study of executives and managers. Employee Responsibilities and Rights Journal, 12(4), 199-217. 190 Keenan, J. P. (2002). Whistleblowing: A study of managerial differences. Employee Responsibilities and Rights Journal, 14(1), 17-32. Keenan, J. P., & McLain, D. L. (1992, August). Whistleblowing: A conceptualization and model. Paper presented at the Annual Meeting of the Academy of Management Las Vegas, NV. Keppel, G., & Wickens, T. D. (2004). Design and analysis: A researcher's handbook (4th ed.). Upper Saddle River, NJ: Pearson Prentice Hall. Ki, E.-J. (2006). Linkages among relationship maintenance strategies, relationship quality outcomes, attitude, and behavioral intentions. Unpublished doctoral dissertation, University of Florida, Gainsville. Ki, E.-J., & Hon, L. C. (2007). Testing the linkages among the organization-public relationship and attitude and behavioral intentions. Journal of Public Relations Research, 19(1), 1-23. Ki, E.-J., & Shin, J.-H. (2006). Status of organization-public relationship research from an analysis of published articles, 1985-2004. Public Relations Review, 32(2), 194- 195. Kiernan, N. E., Kiernan, M., Oyler, M. A., & Giles, C. (2005). Is a Web survey as effective as a mail survey? American Journal of Evaluation, 26(2), 245-252. Kim, H.-S. (2007). A multilevel study of antecedents and a mediator of employee- organization relationships. Journal of Public Relations Research, 19(2), 167-197. King, G., III (1997). The effects of interpersonal closeness and issue seriousness on blowing the whistle. The Journal of Business Communication, 34(4), 419-436. Krishnamurthy, S. (2001). A comprehensive analysis of permission marketing. Journal of Computer-mediated Communication, 6(2), 0-0. Lacayo, R., & Ripley, A. (2002, December 30). Persons of the year. Time, 160(27), pp. 30-33. Retrieved January 18, 2008, from http://www.time.com/time/magazine/ article/0,9171,1003998,00.html 191 Lauzen, M. M., & Dozier, D. M. (1992). The Missing Link: The Public Relations Manager Role as Mediator of Organizational Environments and Power Consequences for the Function. Journal of Public Relations Research, 4(4), 205- 220. Lee, S. T., & Cheng, I.-H. (2011). Characteristics and dimensions of ethical leadership in public relations. Journal of Public Relations Research, 23(1), 46-74. Leslie, G. (2003). Justice and the whistleblower. News Media & the Law, 27(3), 4-6. Levine, J., & Mason, J. (2010). Bulk e-mail HOWTO. Retrieved from http://spam.abuse.net/marketerhel/bulk-howto.shtml Los Alamos National Laboratory. (1995, November 23). The Karen Silkwood story. Los Alamos Science, 23. Retrieved May 30, 2008, from http://www.pbs.org/wgbh/pages/frontline/shows/reaction/interact/ silkwood.html Markowitz, S. (2010, February 23). Some tips for legitimate senders to avoid false positives. Retrieved from http://wiki.apache.org/spamassassin/AvoidingFpsForSenders?action=print Mesmer-Magnus, J. R., & Viswesvaran, C. (2005). Whistleblowing in organizations: An examination of correlates of whistleblowing intentions, actions, and retaliation. Journal of Business Ethics, 62, 277-297. Miceli, M. P., & Near, J. P. (1984). The relationships among beliefs, organizational position, and whistle-blowing status: A discriminant analysis. Academy of Management Journal, 27(4), 687-705. Miceli, M. P., & Near, J. P. (1985). Characteristics of organizational climate and perceived wrongdoing associated with whistle-blowing decisions. Personnel Psychology, 38, 525-544. Miceli, M. P., & Near, J. P. (1988). Individual and situational correlates of whistle- blowing. Personnel Psychology, 41, 267-281. Miceli, M. P., & Near, J. P. (1992). Blowing the whistle: The organizational and legal implications for companies and employees. New York: Lexington Books. 192 Miceli, M. P., & Near, J. P. (1994). Relationships among value congruence, perceived victimization, and retaliation against whistle-blowers. Journal of Management, 20(4), 773-794. Miceli, M. P., Near, J. P., & Dworkin, T. M. (2008). Whistle-blowing in organizations. New York: Routledge. Miceli, M. P., Near, J. P., & Dworkin, T. M. (2009). A word to the wise: How managers and policy-makers can encourage employees to report wrongdoing. Journal of Business Ethics, 86, 379-396. Miceli, M. P., Near, J. P., & Schwenk, C. R. (1991). Who blows the whistle and why? Industrial and Labor Relations Review, 45(1), 113-130. Miceli, M. P., Rehg, M., Near, J. P., & Ryan, K. C. (1999). Can laws protect whistle- blowers?: Results of a naturally occurring field experiment. Work and Occupations, 26(1), 129-151. Moberly, R. E. (2006). Sarbanes-Oxley's structural model to encourage corporate whistleblowers. Brigham Young Law Review, 2006(5), 1107-1180. Moberly, R. E. (2008). Unfulfilled expectations: An empirical analysis of why Sarbanes- Oxley whistleblowers rarely win. William and Mary Law Review, 49(1), 65-156. Morris, T. (2007). Internal and external sources of organizational change: Corporate form and the banking industry. The Sociological Quarterly, 48, 119-140. Near, J. P., & Miceli, M. P. (1985). Organizational dissidence: The case of whistle- blowing. Journal of Business Ethics, 4, 1-16. Near, J. P., & Miceli, M. P. (1986). Retaliation against whistle blowers: Predictors and effects. Journal of Applied Psychology, 71(1), 137-145. Near, J. P., & Miceli, M. P. (1987). Whistle-blowers in organizations: Dissidents or reformers? Research in Organizational Behavior, 9, 321-368. 193 Near, J. P., & Miceli, M. P. (1990, August). When whistleblowing succeeds: Predictors of effective whistleblowing. Paper presented at the Annual Meeting of the Academy of Management, San Francisco. Near, J. P., & Miceli, M. P. (1996). Whistle-Blowing: Myth and reality. Journal of Management, 22(3), 507. Near, J. P., & Miceli, M. P. (2008). Wrongdoing, whistle-blowing, and retaliation in the U.S. Government: What have researchers learned from the Merit Systems Protection Board (MSPB) survey results? Review of Public Personnel Administration, 28(3), 263-281. Near, J. P., Ryan, K. C., & Miceli, M. P. (1995, August). Results of a human resource management "experiment": Whistle-blowing in the federal bureaucracy, 1980- 1992. Paper presented at the Annual Meeting of the Academy of Management, Vancouver, B.C. Ni, L. (2007). Refined understanding of perspectives on employee-organization relationships. Journal of Communication Management, 11(1), 53-70. Ni, L. (2009). Strategic role of relationship building: Perceived links between employee- organization relationships and globalization strategies. Journal of Public Relations Research, 21(1), 100-120. Norris, F. (2009, January 6). Could Skilling get a new trial? Retrieved July 10, 2008, from http://norris.blogs.nytimes.com/2009/01/06/could-skilling-get-a-new- trial/?pagemode=print O'Day, R. (1974). Intimidation rituals: Reactions to reform. The Journal of Applied Behavioral Science, 10(3), 373-386. O'Dwyer, J. (2010, June 21). Left and right mix it up at PR seminar. O'Dwyer's Blog. Retrieved from http://www.odwyerpr.com/blog/index.php?/archives/991-Left- and-Right-Mix-It-up-at-PR Pallant, J. (2005). SPSS Survival Manual (2nd ed.). Chicago: Open University Press. 194 Parmerlee, M. A., Near, J. P., & Jensen, T. C. (1982). Correlates of whistle-blowers' perceptions of organizational retaliation. Administrative Science Quarterly, 27, 17-34. Penn, C. (2010, February 8). How to choose an email subject line for maximum deliverability. Blue Sky Factory Blog. Retrieved February 25, 2010, from http://blog.blueskyfactory.com/deliverability/how-to-choose-an-email-subject- line-for-maximum-deliverability/ Perry, J. L. (1992, August). The consequences of speaking out: Processes of hostility and issue resolution involving federal whistleblowers. Paper presented at the Annual Meeting of the Academy of Management, Las Vegas, NV. Pfeffer, J. P., & Salancik, G. R. (2003). The external control of organizations: A resource dependence perspective Stanford, CA: Stanford University Press. PriceWaterhouseCoopers. (2009). The global economic crime survey: Economic crime in a downturn. from http://www.pwc.com/en_GX/gx/economic-crime- survey/pdf/global-economic-crime-survey-2009.pdf Prior-Miller, M. (1989). Four major social scientific theories and their value to the public relations researcher. In C. H. Botan & J. Hazleton, V. (Eds.), Public relations theory (pp. 67-81). Hillsdale, NJ: Lawrence Erlbaum Associates. PRSA. (2008). Public Relations Society of America Member Code of Ethics 2000. Retrieved 31 May 2008, 2008, from http://www.prsa.org/aboutUs/ethics/ preamble_en.html Public Relations Coalition. (2005). Public relations coalition diversity tracking survey. Retrieved from http://www.instituteforpr.org/topics/diversity-tracking-survey- 2005/ Rapp, G. C. (2007). Beyond protection: Invigorating incentives for Sarbanes-Oxley corporate and securities fraud whistleblowers Boston Law Review, 87, 91-156. Reber, B. H., Cropp, F., & Cameron, G. T. (2001). Mythic Battles: Examining the Lawyer-Public Relations Counselor Dynamic. Journal of Public Relations Research, 13(3), 187-218. 195 Rehg, M. T. (1998). An examination of the retaliation process against whistleblowers: A study of federal government employees. Unpublished doctoral dissertation. Indiana University, Bloomington, IN. Rehg, M. T., Miceli, M. P., Near, J. P., & Van Scotter, J. R. (2004, August). Predicting retaliation against whistle-blowers: Outcomes of power relationships within organizations. Paper presented at the Academy of Management, New Orleans, LA. Rehg, M. T., Miceli, M. P., Near, J. P., & Van Scotter, J. R. (2008). Antecedents and outcomes of retaliation against whistleblowers: Gender differences and power relationships. Organization Science, 19(2), 221-240. Renner, R. (2010, June 17). Supreme Court allows government to read employee text messages. Whistleblower Protection Blog. Retrieved June 19, 2010, from http://www.whistleblowersblog.org/2010/06/articles/whistleblowers-government- empl/ Richardson, B. K. (2005, May). Expanding whistle-blowing scholarship: How stakeholder theory, organizational structure, and social influence processes can inform whistle-blowing research. Paper presented at the International Communication Association Annual Conference, New York. Richardson, B. K., & McGlynn, J. (2007, May). Gendered retaliation, irrationality, and structured isolation: Whistle-blowing in the collegiate sports industry as gendered process. Paper presented at the International Communications Association Annual Conference, San Francisco. Romenesko, J. Romenesko. Retrieved from http://www.poynter.org/column.asp?id=45 Sallot, L. M., Lyon, L. J., Acosta-Alzuru, C., & Jones, K. O. (2008). From aardvark to zebra redux: An analysis of theory development in public relations academic journals into the 21st century. In T. L. Hansen-Horn & B. Neff, Dostal (Eds.), Public relations: From theory to practice (pp. 343-387). Boston: Allyn & Bacon. Schaal, D. (2008). Days in the life of a whistleblower. The CRO Blog. Retrieved from http://www.thecro.com/node/648 Scott, S. G., & Lane, V. R. (2000). A stakeholder approach to organizational identity. Academy of Management Review, 25(1), 43-62. 196 Sheehan, K. B. (2001). E-mail Survey Response Rates: A Review. Journal of Computer- mediated Communication, 6(2), 0-0. Society, A. W. P. (2010). About the Society. from http://www.awpagesociety.com/site/about/ Soule, E. (1998). Trust and managerial responsibility. Business Ethics Quarterly, 8(2), 249-272. Toth, E. L. (2000). From personal influence to interpersonal influence: A model for relationship management. In J. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to the study of public relations (pp. 205-219). Mahwah, NJ: Lawrence Erlbaum Associates. Tsetsura, K. (2011). Is public relations a real job? How female practitioners construct the profession. Journal of Public Relations Research, 23(1), 1-23. U.S. Merit Systems Protection Board. (1981). Whistle-blowing and the federal employee: Blowing the whistle on fraud, waste, and mismanagement—who does it and what happens. Washington, DC: U.S. Government Printing Office. U.S. Merit Systems Protection Board. (1984). Blowing the whistle in the federal government: A comparative analysis of 1980 and 1983 survey finds. Washington, DC: U.S. Government Printing Office. U.S. Merit Systems Protection Board. (1993). Whistleblowing in the federal government: An update (No. MS1.2:W57/2/Update). Washington, DC: U.S. Government Printing Office. U.S. Senate Committee on Governmental Affairs. (1978). The whistleblowers: A report on federal employees who disclose acts of governmental waste, abuse, and corruption. Washington, DC: U.S. Government Printing Office. U.S. Senate Committee on the Judiciary. (1986). Confirmation hearings on appointments to the federal judiciary and the Department of Justice. Washington, DC: U.S. Government Printing Office. 197 Verčič, D., & Grunig, J. E. (2000). The origins of public relations theory in economics and strategic management. In D. Moss, D. Vercic & G. Warnaby (Eds.), Perspectives on public relations research (pp. 9-58). London: Routledge. Von Drehle, D. (2005, June 1). FBI's no. 2 was 'Deep Throat': Mark Felt ends 30-year mystery of The Post's Watergate source. The Washington Post. Retrieved May 30, 2008, from http://www.washingtonpost.com/wp- dyn/content/article/2005/05/31/AR2005053100655.html Warren, T. (2011, May 5). Microsoft: Email still a key means of communication. Retrieved May 30, 2011, from http://www.winrumors.com/microsoft-email-still- a-key-means-of-communication/ The Washington Post. (2008). The Watergate Story. Retrieved May 30, 2008, from http://www.washingtonpost.com/wp-srv/politics/special/watergate/part2.html Westin, A. F. (Ed.). (1981). Whistle blowing! Loyalty and dissent in the corporation. New York: McGraw-Hill. White, C., Vanc, A., & Stafford, G. (2010). Internal communication, information satisfaction, and sense of community: The effect of personal influence. Journal of Public Relations Research, 22(1), 65-84. White, J., & Dozier, D. A. (1992). Public relations and management decision making. In J. E. Grunig (Ed.), Excellence in public relations and communication management. Hillsdale, NJ: Lawrence Erlbaum Associates. Wimmer, R. D., & Dominick, J. R. (2006). Mass media research: An introduction (8th ed.). Belmont, CA: Thomson Wadsworth. Yamey, G. (2000). Protecting whistleblowers: Employers should respond to the message, not shoot the messenger. British Medical Journal, 320(7227), 70-71. 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.