<%BANNER%>

Agenda-Building and Agenda-Setting in Corporate Proxy Contests

Permanent Link: http://ufdc.ufl.edu/UFE0041554/00001

Material Information

Title: Agenda-Building and Agenda-Setting in Corporate Proxy Contests Exploring Influence Among Public Relations Efforts, Financial Media Coverage and Investor Opinion
Physical Description: 1 online resource (429 p.)
Language: english
Creator: Ragas, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: activism, activist, agenda, building, business, candidate, communication, contest, contests, corporate, coverage, fight, financial, influence, investor, journalism, mass, media, opinion, proxy, public, relations, salience, setting, shareholder
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The rise of shareholder activism in recent years has sparked a resurgence in the number of contested corporate elections, known as corporate proxy contests or proxy fights, held each proxy season in the U.S. financial markets. This study investigated the role of mass mediated-influence during such contests from an agenda-building and agenda-building perspective. Specifically, this content analytic investigation assessed the transfer of issue and stakeholder salience among the information subsidies of the competing campaigns, financial news media coverage, and investor opinion in the 25 largest U.S. proxy fights for board representation or control that went to a shareholder vote over a five-year period (2005-2009). At a system-level, this study found that the sheer number of information subsidies disseminated during a contest did not appear to significantly affect the amount of raw media attention the contest received. Attention was linked with the financial size of the incumbent candidate in the contest. The larger the annual revenue, and, to a lesser extent, stock market capitalization of the incumbent, the more coverage the contest generally received. Shifting to analysis at an individual contest level, there was solid evidence of issue agenda-building linkages (with reciprocal influence) among candidate information subsidies and media coverage. News releases were more effective than shareholder letters in shaping issue salience. As for stakeholder salience, the campaigns seemed to largely take their salience cues from the media. The data provided mixed support for the classic agenda-setting hypothesis. A strong linkage was found between the media and investors (led by the media) for stakeholder salience, but only a weak to moderate linkage (with reciprocal influence) was found for issue salience. Candidate subsidies and investor opinion were significantly linked for issue salience (with reciprocal influence), but not for stakeholders. Releases were more effective than letters in shaping issue priorities among investors. Strong support was found for issue and stakeholder inter-candidate agenda-setting relationships among the competing campaigns. Finally, turning to agenda-setting consequences, this study revealed that the candidate with the stronger agenda-building and agenda-setting associations during a contest tended to win the election. The theoretical and practical implications of these findings are discussed.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Ragas.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Kiousis, Spiro K.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041554:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041554/00001

Material Information

Title: Agenda-Building and Agenda-Setting in Corporate Proxy Contests Exploring Influence Among Public Relations Efforts, Financial Media Coverage and Investor Opinion
Physical Description: 1 online resource (429 p.)
Language: english
Creator: Ragas, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: activism, activist, agenda, building, business, candidate, communication, contest, contests, corporate, coverage, fight, financial, influence, investor, journalism, mass, media, opinion, proxy, public, relations, salience, setting, shareholder
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The rise of shareholder activism in recent years has sparked a resurgence in the number of contested corporate elections, known as corporate proxy contests or proxy fights, held each proxy season in the U.S. financial markets. This study investigated the role of mass mediated-influence during such contests from an agenda-building and agenda-building perspective. Specifically, this content analytic investigation assessed the transfer of issue and stakeholder salience among the information subsidies of the competing campaigns, financial news media coverage, and investor opinion in the 25 largest U.S. proxy fights for board representation or control that went to a shareholder vote over a five-year period (2005-2009). At a system-level, this study found that the sheer number of information subsidies disseminated during a contest did not appear to significantly affect the amount of raw media attention the contest received. Attention was linked with the financial size of the incumbent candidate in the contest. The larger the annual revenue, and, to a lesser extent, stock market capitalization of the incumbent, the more coverage the contest generally received. Shifting to analysis at an individual contest level, there was solid evidence of issue agenda-building linkages (with reciprocal influence) among candidate information subsidies and media coverage. News releases were more effective than shareholder letters in shaping issue salience. As for stakeholder salience, the campaigns seemed to largely take their salience cues from the media. The data provided mixed support for the classic agenda-setting hypothesis. A strong linkage was found between the media and investors (led by the media) for stakeholder salience, but only a weak to moderate linkage (with reciprocal influence) was found for issue salience. Candidate subsidies and investor opinion were significantly linked for issue salience (with reciprocal influence), but not for stakeholders. Releases were more effective than letters in shaping issue priorities among investors. Strong support was found for issue and stakeholder inter-candidate agenda-setting relationships among the competing campaigns. Finally, turning to agenda-setting consequences, this study revealed that the candidate with the stronger agenda-building and agenda-setting associations during a contest tended to win the election. The theoretical and practical implications of these findings are discussed.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Ragas.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Kiousis, Spiro K.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041554:00001


This item has the following downloads:


Full Text

PAGE 1

1 AGENDA BUILDING AND AGENDA SETTING IN CORPORATE PROXY CONTESTS: EXPLORING INFLUENCE AMONG PUBLIC RELATIONS EFFORTS, FINANCIAL MEDIA COVERAGE AND INVESTOR OPINION By MATTHEW W. RAGAS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Matthew W. Ragas

PAGE 3

3 To my wife and better half, Traci: We did it. Without y our love and support, this would have never been possible. You make me a better person, and I love you.

PAGE 4

4 ACKNOWLEDGMENTS I have many individuals to thank for making this dissertation possible. I would first like to acknowledge my advisor and chair of this dissertation, Dr. Spiro Kiousis, for his guidance and support throughout this process. I could not ask for a better chair and mentor. I am also fortunate to have an excellent committee, and cannot thank them enough. Dr. Marilyn Roberts started me down this path and introduced me to agendabuilding and agendasetting theory. Dr. Mary Ann Ferguson and Dr. Renee MartinKratzer advanced my knowledge of research methods, and helped me gain confidence as an academic researcher. Professor Virginia Maurer challenged me to think about the intersection of the corporate sphere and mass communication. This dissertation would not have been possible without assistance from industry. I wish to thank John Laide and FactSet Research Systems SharkRepellent.net for providi ng essential inkind support. I am particularly grateful for the participation of the many financial public relations and investor relations professionals during the pretest interview phase of this project. Their expert insights into proxy fights were inv aluable. I am fortunate to have been surrounded by many supportive professors, staff, friends, and colleagues during my time at the University of Florida. In particular, I wish to thank Peter McKay in Library West for patiently demonstrating the Bloomberg Professional terminal. I also wish to acknowledge Dr. John Tedesco, Dr. Glenn Hansen, Luz Rivera, Sara Larsen, and Chris McCarthy and the Application Support Center team I send a special thank you to Jinsoo Kim and his RozelleCampbell baseline calculator Finally, I am very grateful for the support of my family throughout my doctoral program. My parents, Wade and Yvette Ragas, instilled in me a strong appreciation for education from an early age, and have always been there for me. I am grateful for

PAGE 5

5 hav ing such a supportive brother in Josh. My wife, Traci Ragas, uprooted her career to make this possible. Her love, patience, and support are immeasurable. God has truly blessed us. Finally, I thank my grandparents here and above for their support.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 9 LIST OF FIGURES ........................................................................................................ 19 ABSTRACT ................................................................................................................... 21 CHAPTER 1 INTRODUCTION .................................................................................................... 23 Need for Current Study ........................................................................................... 23 Theoretical Framework ........................................................................................... 25 Purpose and Overview of Study ............................................................................. 27 Theoretical and Practical Cont ributions .................................................................. 28 2 LITERATURE REVIEW .......................................................................................... 33 History of Shareholder Activism .............................................................................. 33 Gadflies, Corporate Governance and Social Responsibility ? ........................... 35 Institutional Investors and Entrepreneurial Activists ? ....................................... 38 Co rporate Proxy Contests ....................................................................................... 40 Structure of a Proxy Contest ............................................................................ 41 History of Proxy Contests ................................................................................. 43 The Proxy Access Rule .................................................................................... 45 Investor Relations and Financial Public Relations .................................................. 47 Empirical Research on the F ield ............................................................................. 48 Relationship with the Financial News Media ........................................................... 53 Evolution of AgendaSetting Theory ....................................................................... 57 Transfer of Object and Attribute Salience ......................................................... 59 Contingent Conditions of AgendaSetting ........................................................ 63 AgendaBuild ing and Public Relations ............................................................. 67 Candidateto Investor AgendaSetting ............................................................. 70 Inter Candidate AgendaSetting ....................................................................... 72 Consequences: Attitudes and Behavior ........................................................... 75 AgendaSetting in the Corporate Sphere ................................................................ 80 System -L evel AgendaSetting ................................................................................ 84 Hypotheses and Research Questions .................................................................... 86 System Level AgendaSetting .......................................................................... 87 System Level Contingent Conditions ............................................................... 87 Candidateto Media AgendaBuilding ............................................................... 87 Inter Candidate AgendaSe tting ....................................................................... 88 Mediato Investor (Traditional) AgendaSetting ................................................ 88

PAGE 7

7 Candidateto Investor AgendaSetting ............................................................. 89 Consequences of AgendaSetting .................................................................... 89 3 METHOD ................................................................................................................ 94 Study Design Overview ........................................................................................... 94 Pre Test .................................................................................................................. 97 Salient Issues ................................................................................................... 98 Salient Stakeholders ........................................................................................ 99 Information Subsidies ..................................................................................... 101 Computer Assisted Textual Analysis ..................................................................... 103 Selection of Sample .............................................................................................. 105 News Releases and Shareholder Letters ....................................................... 108 Investor Opinion ............................................................................................. 113 Financial Performance and Experience Indicators ......................................... 115 Coding Categories for Content ............................................................................. 117 Intercoder Reliability ............................................................................................. 122 Data Analysis Strategy .......................................................................................... 124 4 RESULTS ............................................................................................................. 136 System Level AgendaSetting .............................................................................. 136 System Level Contingent Conditions .................................................................... 138 Placement of Issues and Stakeholders on Agendas ............................................. 140 Candidateto Media AgendaBuilding ................................................................... 142 Inter Candidate AgendaSetting ........................................................................... 147 Mediato Investor AgendaSetting ........................................................................ 148 Candidateto Investor AgendaSetting .................................................................. 149 Consequences of AgendaSetting ........................................................................ 152 5 DISCUSSION ....................................................................................................... 240 Summary of Findings ............................................................................................ 240 Theoretical Contributions ...................................................................................... 248 Practical Implications ............................................................................................ 253 Implications for Communication Professionals ............................................... 253 Implications for Business Journalists .............................................................. 255 Implications for Financial Regulators .............................................................. 256 Limitations ............................................................................................................. 256 F uture Research ................................................................................................... 259 APPENDIX A ASSENT SCRIPT ................................................................................................. 268 B INTERVIEW GUIDE .............................................................................................. 269 C CODE SHEET ...................................................................................................... 271

PAGE 8

8 D CODE BOOK ........................................................................................................ 275 E INFORMATION SUBSIDIES CONTEST FREQUENCY TABLES ........................ 284 F MEDIA AND INVESTORS CONTEST FREQUENCY TABLES (ISSUE SALIENCE) ........................................................................................................... 309 G MEDIA AND INVESTORS CONTEST FREQUENCY TABLES (STAKEHOLDER SALIENCE) ........................................................................................................... 335 H CONTEST CORRELATION TABLES (ISSUE SALIENCE) .................................. 360 I CONTEST CORRELATION TABLES (STAKEHOLDER SALIENCE) ................... 386 LIST OF REFERENCES ............................................................................................. 411 BIOGRAPHICAL SKETCH .......................................................................................... 429

PAGE 9

9 LIST OF TABLES Table page 3-1 MergerMarket.com top ten financial PR M&A advisors for 2008 ...................... 128 3-2 Sample of 20052009 U.S. corporate proxy contests for study ........................ 129 3-3 Frequency of candidate news releases in proxy contests ................................ 130 3-4 Frequency of candidate shareholder letters in proxy contests .......................... 131 35. Financial media attention to proxy contests ......................................................... 132 3-6 Investor attention on message boards to proxy contests .................................. 133 3-7 Financial performance of incumbent candidates in proxy contests ................... 134 3-8 Activist campaign experience of challenger candidates in proxy contests ........ 135 4-1 System level descriptive statistics and correlations for 2009 proxy contests .... 155 4-2 System level descriptive statistics and correlations for 2008 proxy c ontests .... 158 4-3 System level descriptive statistics and correlations for 2007 proxy contests .... 161 4-4 System level descriptive st atistics and correlations for 2006 proxy contests .... 164 4-5 System level descriptive statistics and correlations for 2005 proxy contests .... 167 4-6 Aggregate media attention and investor attention to issues during proxy contests ............................................................................................................ 170 4-7 Media attention to issues during proxy contests across individual media outlets ............................................................................................................... 171 4-8 Candidate attention to issues in news releases during during proxy contests. 173 4-9 Candidate attention to issues in shareholder letters during proxy contests. ..... 174 410 Aggregate media attention and investor attention to stakeholders during proxy contests .................................................................................................. 175 411 Media attention to stakeholders during proxy contests across individual media outlets .................................................................................................... 176 412 Candidate attention to stakeholders in news releases during proxy contests. .. 178

PAGE 10

10 413 Candidate attention to stakeholders in shareholder letters during proxy contests. ........................................................................................................... 179 414 Summary of candidate news releases medi a issue agendabuilding relationships in proxy contests .......................................................................... 180 415 Summary of candidate shareholder letters media issue agendabuilding relationships in proxy contests .......................................................................... 181 416 Summary of candidate news releases media stakeholder agendabuilding relationships in proxy contests .......................................................................... 182 417 Summary of candidate shareholder le tters media stakeholder agendabuilding relationships in proxy contests ............................................................ 183 418 Summary of intercandidate issue and stakeholder agendasetting relationships in proxy contests .......................................................................... 184 419 Summary of mediainvestor issue and stakeholder agendasetting relationships in proxy contests .......................................................................... 185 420 Summary of candidate investor issue agenda setting relationships in proxy contests ............................................................................................................ 186 421 Summary of candidate investor stakeholder agendasetting relationships in proxy contests .................................................................................................. 187 5-1 Summary of findings for hypotheses and research questions ......................... 266 E-1 Candidate issue and stakeholder mentions during the 2005 Computer Horizons Crescendo Partners proxy contest. ................................................ 284 E-2 C andidate issue and stakeholder mentions during the 2005 BKF Capital Crescendo Steel Partners proxy contest. ......................................................... 285 E-3 C andidate issue and stakeholder mentions during the 2005 Six Flags Daniel Snyder proxy contest. ............................................................................ 286 E-4 C andidate issue and stakeholder mentions during the 2005 Exar Corporati on GWA Investments proxy contest. .................................................................. 287 E-5 C andidate issue and stakeholder mentions during the 2005 Blockbuster Inc. Carl Icahn proxy contest. ............................................................................... 288 E-6 C andidate issue and stakeholder mentions during the 2006 GenCorp Inc. Pirate Capital proxy contest. ............................................................................. 289 E-7 C andidate issue and stakeholder mentions during the 2006 Motient Corporation Highland Capital proxy contest. ................................................. 290

PAGE 11

11 E-8 C andidate issue and stakeholder mentions during the 2006 Massey Energy Third Point LLC proxy contest. ......................................................................... 291 E-9 C andidate issue and stakeholder mentions during the 2006 UbiquiTel Inc. Deephaven Capital proxy contest. .................................................................... 292 E10 C andidate issue and stakeholder mentions during the 2006 UbuiTel Inc. Deephaven Capital proxy contest. .................................................................... 293 E11 C andidate issue and stakeholder mentions during the 2007 OpenWave Systems Deephaven Capital proxy contest .................................................. 294 E12 C andidate issue and stakeholder mentions during the 2007 Motorola, Inc. Carl Icahn proxy contest. .................................................................................. 295 E13 C andidat e issue and stakeholder mentions during the 2007 H&R Block, Inc. Breeden Capital proxy contest. ...................................................................... 296 E14 C andidate issue and stakeholder mentions during the 2007 Atmel Corporation George P erlegos proxy contest. ................................................. 297 E15 C andidate issue and stakeholder mentions during the 2007 Arrow International, Inc. McNeil Trust proxy contest. ............................................... 298 E16 C andidate issue and stakeholder mentions during the 2008 Hexcel Corporation OSS Capital proxy contest. ........................................................ 299 E17 C andidate issue and stakeholder mentions during the 2008 International Rectifier Vishay Intertechnology, Inc. proxy contest. ..................................... 300 E18 C andidate issue and stakeholder mentions during the 2008 Micrel, Inc. Obrem Capital proxy contest. ........................................................................... 301 E19 C andidate issue and stakeholder mentions during the 2008 CSX Corporation TCI Fund Management proxy contest. .......................................................... 302 E20 C andida te issue and stakeholder mentions during the 2008 Biogen Idec, Inc. Carl Icahn proxy contest. ............................................................................... 303 E21 C andidate issue and stakeholder mentions during the 2009 Conseco, Inc. Keith Long proxy contest. ................................................................................. 304 E22 C andidate issue and stakeholder mentions during the 2009 PHH Corporation Pennant Capital proxy contest. ...................................................................... 305 E23 C andidate issue and stakeholder mentions during the 2009 Myers Industries, Inc. GAMCO Investors proxy contest. .............................................................. 306

PAGE 12

12 E24 Candidate issue and stakeholder mentions during the 2009 NRG Energy, Inc. Exelon Corporation proxy contest. .......................................................... 3 07 E25 Candidate issue and stakeholder mentions during the 2009 Target Corporation Pershing Square Capital proxy contest. ..................................... 308 F-1 Media and investor issue mentions during the 2009 Target Corporation Pershing Square Capital Management proxy contest ....................................... 310 F-2 Media and investor issue mentions during the 2009 NRG Energy, Inc. Exelon Corporation proxy contest ..................................................................... 311 F-3 Media and investor issue mentions during the 2009 Myers Industries, Inc. GAMCO Investors pro xy contest ...................................................................... 312 F-4 M edia and investor issue mentions during the 2009 PHH Corporation Pennant Capital Management proxy contest .................................................... 313 F-5 Media and investor issue mentions during the 2009 Conseco, Inc. Keith Long proxy contest .......................................................................................... 314 F-6 M edia and investor issue mentions during the 2008 Biogen Idec, Inc. Carl C. Icahn proxy contest ...................................................................................... 315 F-7 M edia and investor issue mentions during the 2008 CSX Corporation TCI Fund Management proxy contest ..................................................................... 316 F-8 M edia and investor issue mentions during the 2008 Micrel, Incorporated Obrem Capital Management proxy contest ...................................................... 317 F-9 Media and investor issue mentions during the 2008 International Rect ifier Vishay Intertechnology, Inc. proxy contest ....................................................... 318 F10 Media and investor issue mentions during the 2008 Hexcel Corporation OSS Capital Management proxy contest .......................................................... 319 F11 M edia and investor issue mentions during the 2007 Arrow International, Inc. McNeil Trust proxy contest ............................................................................... 320 F12 M edia and investor issue mentions during the 2007 Atmel Corporation George Perlegos proxy contest ........................................................................ 321 F13 M edia and investor issue mentions during the 2007 H&R Block, Inc. Breeden Capital Management proxy contest .................................................... 322 F14 Media and investor issue mentions during the 2007 Motorola, Inc. Carl C. Icahn proxy contest .......................................................................................... 323

PAGE 13

13 F15 Media and investor issue mentions during the 2007 Openwave Systems Harbinger Capital proxy contest ....................................................................... 324 F16 Media and investor issue mentions during the 2006 Career Education Corporation Steve Bostic proxy contest ......................................................... 325 F17 Media and investor issue mentions during the 2006 UbiquiTel Inc. Deephaven Capital Management proxy contest ............................................... 326 F18 Media and investor issue mentions during the 2006 Massey Energy Company Third Point LLC proxy contest ........................................................ 327 F19 Media and investor issue mentions during the 2006 Motient Corporation Highland Capital Management proxy contest ................................................... 328 F20 Media and investor issue mentions during the 2006 GenCorp Inc. Pirate Capital, LLC proxy contest ............................................................................... 329 F21 Media and investor issue mentions during the 2005 Blockbuster Inc. Carl C. Icahn proxy contest .......................................................................................... 330 F22 Media and investor issue mentions during the 2005 Exar Corporation GWA Investments proxy contest ................................................................................ 331 F23 Media and investor issue mentions during the 2005 Six Flags, Inc. Daniel Snyder proxy contest ........................................................................................ 332 F24 Media and investor issue mentions during the 2005 BKF Capital Group, Inc. Steel Partners II proxy contest .......................................................................... 333 F25 Media and investor issue mentions during the 2005 Computer Horizons Crescendo proxy contest .................................................................................. 334 G-1 Media and investor stakeholder mentions during the 2009 Target Corporation Pershing Square Capital Management proxy contest .................................... 335 G-2 Media and investor stakeholder mentions during the 2009 NRG Energy, Inc. Exelon Corporation proxy contest ..................................................................... 336 G-3 Media and investor stakeholder mentions during the 2009 Myers Industries, Inc. GAMCO Investors proxy contest ............................................................. 337 G-4 Media and investor stakeholder mentions during the 2009 PHH Corporation Pennant Capital Management proxy contest .................................................... 338 G-5 Media and investor issue mentions during the 2009 Conseco, Inc. Keith Long proxy contest .......................................................................................... 339

PAGE 14

14 G-6 Media and investor issue mentions during the 2008 Biogen Idec, Inc. Carl C. Icahn proxy contest ...................................................................................... 340 G-7 Media and investor issue mentions during the 2008 CSX Corporation TCI Fund Management proxy contest ..................................................................... 341 G-8 Media and investor issue mentions during the 2008 Micrel, Incorporated Obrem Capital Management proxy contest ...................................................... 342 G-9 Media and investor issue mentions during the 2008 International Rectifier Vishay Intertechnology, Inc. proxy contest ....................................................... 343 G10 Media and investor issue mentions during the 2008 Hexcel Corporation OSS Capital Management proxy contest .......................................................... 344 G11 Media and investor issue mentions during the 2007 Arrow International, Inc. McNeil Trust proxy contest ............................................................................... 345 G12 Media and investor issue mentions during the 2007 Atmel Corporation George Perlegos proxy contest ........................................................................ 346 G13 Media and investor issue mentions during the 2007 H&R Block, Inc. Breeden Capital Management proxy contest .................................................... 347 G14 Media and investor issue mentions during the 2007 Motorola, Inc. Carl C. Icahn proxy contest .......................................................................................... 348 G15 Media and investor issue mentions during the 2007 Openwave Systems Harbinger Capital proxy contest ....................................................................... 349 G16 Media and investor issue mentions during the 2006 Career Education Corporation Steve Bostic proxy contest ......................................................... 350 G17 Media and investor issue mentions during the 2006 UbiquiTel Inc. Deephaven Capital Management proxy contest ............................................... 351 G18 Media and investor issue mentions during the 2006 Massey Energy Company Third Point LLC proxy contest ........................................................ 352 G19 Media and investor issue mentions during the 2006 Motient Corporation Highland Capital Management proxy contest ................................................... 353 G20 Media and investor issue mentions during the 2006 GenCorp Inc. Pirate Capital, LLC proxy contest ............................................................................... 354 G21 Media and investor issue mentions during the 2005 Blockbuster Inc. Carl C. Icahn proxy contest .......................................................................................... 355

PAGE 15

15 G22 Media and investor issue mentions during the 2005 Exar Corporation GWA Investments proxy contest ................................................................................ 356 G23 Media and investor issue mentions during the 2005 Six Flags, Inc. Daniel Snyder proxy contest ........................................................................................ 357 G24 Media and investor issue mentions during the 2005 BKF Capital Group, Inc. Steel Partners II proxy contest .......................................................................... 358 G25 Media and investor issue mentions during the 2005 Computer Horizons Crescendo proxy contest .................................................................................. 359 H-1 Issue agendabuilding and agendasetting relationships during the 2009 Target Pershing Square proxy contest ........................................................... 361 H-2 Issue agendabuilding and agendasetting relationships during the 2009 NRG Energy Exelon proxy contest ................................................................. 362 H-3 Issue agendabuilding and agendasetting relationships during the 2009 Myers GAMCO proxy contest ......................................................................... 363 H-4 Issue agendabuilding and agendasetting relationships during the 2009 PHH Pennant Capital proxy contest ........................................................................ 364 H-5 Issue agendabuilding and agendasetting relationships during the 2009 Conseco Keith Long Capital proxy contest ..................................................... 365 H-6 Issue agendabuilding and agendasetting relationships during the 2008 Biogen Idec Carl Icahn proxy contest ............................................................. 366 H-7 Issue agendabuilding and agendasetting relationships during the 2008 CSX Corp. TCI proxy contest ................................................................................. 367 H-8 Issue agendabuilding and agendasetting relationships during the 2008 Micrel Obrem Capital proxy contest ............................................................... 368 H-9 Issue agendabuilding and agendasetting relationships during the 2008 International Rectifier Vishay proxy contest ................................................... 369 H10 Issue agendabuilding and agendasetting relationships during the 2008 Hexcel OSS Capital proxy contest ................................................................. 370 H11 Issue agendabuilding and agendasetting relationships during the 2007 Arrow McNeil Trust proxy contest .................................................................... 371 H12 Issue agendabuilding and agendasetting relationships during the 2007 Atmel George Perlegos proxy contest ............................................................. 372

PAGE 16

16 H13 Issue agendabuilding and agendasetting relationships during the 2007 H&R Block Breeden Capital proxy contest .............................................................. 373 H14 Issue agendabuilding and agendasetting relationships during the 2007 Motorola Carl Icahn proxy contest .................................................................. 374 H15 Issue agendabuilding and agendasetting relationships during the 2007 Openwave Harbinger Capital proxy contest ................................................... 375 H16 Issue agendabuilding and agendasetting relationships during the 2006 Career Education Steve Bostic proxy contest ................................................ 376 H17 Issue agendabuilding and agendasetting relationships during the 2006 UbiquiTel Deephaven Capital proxy contest .................................................. 377 H18 Issue agendabuilding and agendasetting relationships during the 2006 Massey Energy Third Point proxy contest ...................................................... 378 H19 Issue agendabuilding and agendasetting relationships during the 2006 Motient Highland Ca pital proxy contest .......................................................... 379 H20 Issue agendabuilding and agendasetting relationships during the 2006 GenCorp Pirate Capital proxy contest ............................................................ 380 H21 Issue agendabuilding and agendasetting relationships during the 2005 Blockbuster Carl Icahn proxy contest ............................................................. 381 H22 Issue agendabuilding and agendasetting relationships during the 2005 Exar GWA Investments proxy contest .................................................................... 382 H23 Issue agendabuilding and agendasetting relationships during the 2005 Six Flags Daniel Snyder proxy contest ................................................................. 383 H24 Issue agendabuilding and agendasetting relationships during the 2005 BKF Capital Steel Partners proxy contest .............................................................. 384 G25 Issue agendabuilding and agendasetting relationships during the 2005 Computer Horizons Crescendo Partners proxy contest ................................. 385 I-1 Stakeholder agendabuilding and agendasetting relationships during the 2 009 Target Pershing Square proxy contest .................................................. 386 I-2 Stakeholder agendabuilding and agendasetting relationships during the 2009 NRG Energy Exelon proxy contest ........................................................ 387 I-3 Stakeholder agendabuilding and agendasetting relationships during the 2009 Myers GAMCO proxy contest ................................................................ 388

PAGE 17

17 I-4 Stakeholder agendabuilding and agend asetting relationships during the 2009 PHH Pennant Capital proxy contest ...................................................... 389 I-5 Stakeholder agendabuilding and agendasetting relationships during the 2009 Conseco Keith Long Capital proxy contest ............................................ 390 I-6 Stakeholder agendabuilding and agendasetting relationships during the 2008 Biogen Idec Carl Icahn proxy contest .................................................... 391 I-7 Stakeholder agendabuilding and agendasetting relationships during the 2008 CSX Corp. TCI proxy contest ................................................................ 392 I-8 Stakeholder agendabuilding and agendasetting relationships during the 2008 Micrel Obrem Capital proxy contest ...................................................... 393 I-9 Stakeholder agendabuilding and agendasetting relationships during the 2008 International Rectifier Vishay proxy contest ........................................... 394 I10 Stakeholder agendabuilding and agendasetting relationships during the 2008 Hexcel OSS Capital proxy contest ........................................................ 395 I11 Stake holder agendabuilding and agendasetting relationships during the 2007 Arrow McNeil Trust proxy contest ........................................................... 396 I12 Stakeholder agendabuilding and agendasetting relationships during the 2007 Atmel George Perlegos proxy contest .................................................... 397 I13 Stakeholder agendabuilding and agendasetting relationships during the 2007 H&R Block Breeden Capital proxy contest ............................................ 398 I14 Stakeholder agendabuilding and agendasetting relationships during the 2007 Motorola Carl Icahn proxy contest ......................................................... 399 I15 Stakeholder agendabuild ing and agendasetting relationships during the 2007 Openwave Harbinger capital proxy contest ........................................... 400 I16 Stakeholder agendabuilding and agendasetting relationships during the 2006 Career Education Steve Bostic proxy contest ....................................... 401 I17 Stakeholder agendabuilding and agendasetting relationships during the 2006 UbiquiTel Deephaven Capital proxy contest .......................................... 402 I18 Stakeholder agendabuilding and agendasetting relationships during the 2006 Massey Energy Third Point proxy contest ............................................. 403 I19 Stakeholder agend abuilding and agendasetting relationships during the 2006 Motient Highland Capital proxy contest ................................................. 404

PAGE 18

18 I20 Stakeholder agendabuilding and agendasetting relationships during the 2006 GenCorp Pirate Capital proxy contest ................................................... 405 I21 Stakeholder agendabuilding and agendasetting relationships during the 2005 Blockbuster Carl Icahn proxy contest .................................................... 406 I22 Stakeholder agendabuilding and agendasetting relationships during the 2005 Exar GWA Investments proxy contest ................................................... 407 I23 Stakeholder agendabuilding a nd agendasetting relationships during the 2005 Six Flags Daniel Snyder proxy contest .................................................. 408 I24 Stakeholder agendabuilding and agendasetting relationships during the 2005 BKF Capital Stee l Partners proxy contest ............................................. 409 I25 Stakeholder agendabuilding and agendasetting relationships during the 2005 Computer Horizons Crescendo Partners proxy contest ........................ 410

PAGE 19

19 LIST OF FIGURES Figure page 1-1 Agendabuilding and agendasetting theoretical map during corporate proxy contests .............................................................................................................. 32 2-1 Timeline of shareholder activism ........................................................................ 90 2-2 Corporate proxy contests by year ....................................................................... 91 2-3 First and secondlevel media to pu blic agenda setting ..................................... 92 2-4 Agendabuilding and agendasetting theoretical map during corporate proxy contests (with hypotheses and research questions) ........................................... 93 4-1 Cross lag correlations for issue salience among candidate news releases and newspaper coverage. ................................................................................ 188 4-2 Cross lag correlations for issue salience among candidate news releases and news wire coverage. .................................................................................. 191 4-3 Cross lag correlations for issue salience among candidate shareholder letters and newspaper coverage. ...................................................................... 199 4-4 Cross lag correlations for issue salience among candidate shareholder letters and news wire coverage.. ...................................................................... 200 4-5 Cross lag correlations for stakeholder salience among candidate news releases and newspaper coverage. .................................................................. 205 4-6 Cross lag correlations for stakeholder salience among candidate news releases and news wire coverage.. .................................................................. 207 4-7 Cross lag correlations for issue salience among newspaper coverage and investor opinion. ............................................................................................... 213 4-8 Cross lag correlations for issue salience among news wire coverage and investor opinion. ............................................................................................... 215 4-9 Cross lag correlations for stakeholder salience among newspaper coverage and investor opinion. ........................................................................................ 219 410 Cross lag correlations for stakeholder salience among newspaper coverage and investor opinion.. ....................................................................................... 221 411 Cross lag correlations for issue salience among candidate news releases and investor opinion. ........................................................................................ 225

PAGE 20

20 413 Cross lag correlations for stakeholder salience among candidate news releases and investor opinion.. ......................................................................... 233 414 Cross lag correlations for stakeholder salience among candidate shareholder letters and investor opinion.. ............................................................................. 237 5-1 Agendabuilding and agendasetting theoretical map during corporate proxy contests (with hypotheses and research questions) ......................................... 267

PAGE 21

21 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy AGENDA BUILDING AND AGENDA SETTING IN CORPORATE PROXY CONTESTS: EXPLORING INFLUENCE AMONG PUBLIC RELATIONS EFFORTS, FINANCIAL MEDIA COVERAGE AND INVESTOR OPINION By Matthew W. Ragas May 2010 Chair: Spiro Kiousis Major: Mass Communication The rise of shareholder activism in recent years has sparked a resurgence in the number of contested corporate elections, known as corporate proxy contests or proxy fights, held each proxy season in the U.S. financial markets. This study investigated the role of mass mediatedinfluence during such contests from an agendabuilding and agendabuilding perspective. Specifically, this content analytic investigation assessed the transfer of issue and stakeholder salience among the information subsidies of the competing campaigns, financial news media coverage, and investor opinion in the 25 largest U.S. proxy fights for board representation or control that went to a shareholder vote over a fiveyear period (20052009). At a system level, this study found that the sheer number of information subsidies disseminated during a contest did not appear to significantly affect the amount of raw media attention the contest received. Attention was linked with the financial size of the incumbent candidate in the contest. The larger the annual revenue, and, to a lesser extent, stock market capitalization of the incumbent, the more coverage the contest generally received. Shifting to analysis at an individual contest level, there was solid

PAGE 22

22 evidence of issue agendabuilding linkages (with reciprocal influence) among candidate information subsidies and media coverage. News releases were more effective than shareholder letters in shaping issue salience. As for stakeholder salience, the campaigns seemed to largely take their salience cues from the media. The data provided mixed support for the classic agendasetting hypothesis. A strong linkage was found between the media and investors (led by the media) for stakeholder salience, but only a weak to moderate linkage (with reciprocal influence) was found for issue salience. Candidate subsidies and investor opinion were significantly linked for issue salience (with reciprocal influence), but not for stakeholders. Releases were more effective than letters in shaping issue priorities among investors. Strong support was found for issue and stakeholder inter candidate agendasetting relationships among the competing campaigns. Finally, turning to agendasetting consequences, this study revealed that the candidate with the stronger agendabuilding and agendasetting associations during a contest tended to win the election. The theoretical and practical implications of these findings are discussed.

PAGE 23

23 CHAPTER 1 INTRODUCTION Political elections in the public sphere are not the only setting in which opposing candidates attempt to influence the salience of issues on the media agenda, and in turn, on the public agenda. There is also intense competition for attention within the corporate sphere. Corporate proxy contests represent another venue in which opposing candidates in this case incumbent and challenger corporate actors attempt to build the media agenda and shape investor opinion to their advantage (Iyengar & Simon, 2000). When one or more shareholders are significant ly dissatisfied with a publicly held corporations board of directors and management, they may choose to try and unseat the incumbent board by nominating their own competing slate of challenger nominees as replacements (A. Miller, 1986). The resulting competition for shareholder votes leading up to the corporations annual or special meeting is known as a proxy contest or proxy battle (Holton, 2006). The term proxy comes from the legal right a shareholder has to designate an agent or proxy to vote their shares and act on their behalf (Holton, 2006). During a proxy contest, each side mails out its own proxy statements, proxy cards, and solicitation materials seeking shareholder support for its slate of director nominees. These highstakes contests between dueling corporate actors use many of the same public relations tools, such as news releases and shareholder letters, found in political campaigns. Need for Current Study To date, the mass communication aspect of proxy contests has received scant scholar ly research attention, even though shareholder activism is at near record levels (Laide & Mallea, 2009), and there is an acknowledged need for more scholarship

PAGE 24

24 specifically on investor relations and financial public relations (Botan & Hazelton, 2006). Acco rding to financial data provider FactSet SharkRepellent.net, more U.S. proxy fights were waged and went to a shareholder vote in 2008 than in any year since it began tracking contests in 2001 (Laide & Mallea, 2009). There were 123 proxy fights in the U.S. in 2008, compared with just 42 five years ago (Laide & Mallea, 2009). While this number is still relatively low compared with the hundreds of publicly traded U.S. firms, similar to a crisis situation, a proxy contest often serves as a transformational event that affects multiple stakeholder groups, and in which public relations plays an important role (Carney & Jorden, 1993). A companys reputation, the future of its management, employees and customers, as well as millions of dollars in stock market value may be gained or lost for its shareholders based on a contest outcome. Shareholder activism is an area that financial public relations, investor relations, and corporate communication professionals cannot afford to ignore. In recent years, large well known publicly held corporations such as Target, Wendys, Home Depot, Time Warner, H.J. Heinz, and Yahoo! have all been the subject of either full blown proxy contests or activist investor campaigns. A record $20 million dollars was collectively spent on campaign expenses, including financial public relations efforts, in the 2009 proxy contest between Target and Pershing Square Capital (Crosby, 2009). According to Marken (2005), an adviser to companies during proxy contests and takeovers, a solid PR strategy and program can be the strongest strategic and proactive weapon that can be used to advance managements case and point as well as control the issues (p. 22). So far, the growing expense and complexity of waging a proxy battle hasnt dissuaded large activist shareholders, such as hedge funds and

PAGE 25

25 private equity firms, from using the proxy mechanism to try and effect change. A pending rule by the U.S. Securities and Exchange Commission (S.E.C.), known as the proxy access rule, could significantly decrease th e cost and complexity of nominating alternative candidates to a companys board of directors (Younglai, 2009). The S.E.C. intends to have this proposed rule in place by the 2011 proxy season (Lynch, 2009) In short, this rule could have the effect of stimulating an additional increase in the number of contested director elections held annually, resulting in even greater interest among corporate managers, shareholder activists, and communication practitioners in better understanding proxy contest communication strategy and tactics. Theoretical Framework With 40 years of rich empirical history in political election contests to draw upon, the agendabuilding and agendasetting perspective provides a promising theoretical framework for analyzing the formation and transfer of salience among competing corporate actors, financial media, and investor opinion during proxy contests. Nearly 90 years ago, Walter Lippmann (1922), the intellectual forefather of agendasetting theory, described the mass media as painting the pictures in our heads and serving as a conduit to the world outside (p. 4). Lippmanns prognostications were first put to an empirical test by Max McCombs and Don Shaw in their study of a group of undecided voters in Chapel Hill, North Carolina during the 1968 U.S. presidential campaign (McCombs & Shaw, 1972). This initial agenda setting study, consisting of a content analysis of the news media and a survey of undecided voters, found that the media set the agenda of political issues for the public d uring the campaign. The more media attention an issue received, the greater the perceived importance the issue was accorded by the group of undecided voters

PAGE 26

26 Since the Chapel Hill study (McCombs & Shaw, 1972), several hundred subsequent empirical investigations have largely reaffirmed the initial agendasetting hypothesis the transfer of salience from the media agenda to the public agenda (Dearing & Rogers, 1996; McCombs, 2004). While agendasetting research into the traditional mediato public relationship remains active, the theoretical map has greatly expanded over the years to incorporate additional actors, conditions, and perspectives into this tradition. Over 25 years ago, Gandy (1982) helped set in motion a major new extension of agendasetting research by asking, who sets the media agenda? Research into the various sources, such as public relations professionals, which interact with journalists and try to build media agendas, have come to be known as agendabuilding. Finally, recent research by T edesco (2005a, 2005b) demonstrates that campaigns not only attempt to build media coverage, but also attempt to set the agendas of their opponents. This recent theoretical extension is known as inter candidate agendasetting (Tedesco, 2005a, 2005b). While the majority of agendasetting explorations continue to occur in the public sphere, namely political elections, the agenda setting framework has also been tested in new areas such as education (Rodriguez, 2000), professional sports (Fortunato, 2000; Seltzer & Dittmore, 2009), and organized religion (Buddenbaum, 20001). A particularly active area of investigation for agendasetting researchers in recent years has been the corporate sphere (e.g. Berger, 2001; Berger, Hertog, & Park, 2002; Carroll, 2004; Carroll & McCombs, 2003; Kiousis, Popescu, & Mitrook, 2007; Meijer & Kleinnijenhuis, 2006; Ohl, Pincus, Rimmer, & Harrison, 1995; Ragas, Kim, & Lim, 2009). Although most of this research has focused specifically on exploring linkages among

PAGE 27

27 corporate communication efforts, media coverage, and public perceptions of corporate reputation, at least two of these studies (Ohl et al., 1995; Ragas et al., 2009), have specifically focused on evaluating influence among competing corporate actors, the media and investors during hostile takeover attempts. Given that agendasetting theory has its roots in political election contests, the competing agenda environment of corporate election contests would appear to be a natural new setting for rigorously assessing the efficacy of this mass media effects theory in predicting and explaining the relationships among the competing campaigns, the financial news media, investor opinion, and the potential consequences of these interactions Purpose and Overview of Study With this in mind, the purpose of this investigation is to build theory that helps inform financial public relations and investor relations practice by testing for agendabuilding, agendasetting, and inter candidate agendasetting effects among competing corporate agendas (incumbent and challenger candidates), financial news media agendas, and investor opinion agendas in the new context of corporate proxy contests. More specifically, this investigation will test for the transfer of first level issue and stakeholder salience among the aforementioned agendas by content analyzing the candidatecontrolled communication materials (news releases and shareholder letters), financial news stories, and investor message board postings in 25 of the largest U.S. proxy battles held during the 20052009 proxy seasons that went to a vote. The purposive sample selected for this study consists of the five largest U.S. proxy battles held each year during this five year period that were for representation or control of a companys board of directors, and that went to a shareholder vote. The corporate proxy contest database of FactSet SharkRepellent.net, a provider of comprehensive

PAGE 28

28 data on shareholder activism campaigns and corporate governance, provided the sample frame of proxy contests by year. Given the size and scope of this study, it was necessary to use Diction 5.0, a computer assisted textual analysis (CATA) program, to accomplish the coding of the content encompassing the various agendas in each contest. To aid in the construction of the coding scheme for the content analyses, a pretest was conducted, consisting of indepth interviews with over 20 midto senior level financial public relations practitioners with prior proxy contest advisory experience. By selecting the five largest proxy contests held each year over a continuous fiveyear period for this studys sample, this multi case, multi year study design provides for rigorous testing of the individual hypotheses and research questions. The strength of this sampling approach and overall design is that it provides for repeated internal replications of many of the studys hypotheses and research questions, thereby raising confidence in the results and enhancing external validity. The use of multiple indicators for the incumbent and challenger candidate and media agendas in each contest enhances reliability (Chaffee, 1991). Theoretical and Practical Contributions This study holds the potential to make a variety of important contributions to the theory and practice of public relations mass communication, and corporate communication. Starting with theoretical contributions, this exploration is unique in that it simultaneously investigates multiple stages of agenda setting theory from an entire agenda perspective, thereby providing an opportunity for a comprehensive examination and tighter integration of the theorys components (McCombs, 2005; Tan & Weaver, 2007). See Table 11 for the agendabuilding and agendasetting theoretical map during corporate proxy contests, which will help gui de this investigation. Most research in this

PAGE 29

29 tradition is limited to separate examinations of agendabuilding or agendasetting, rather than looking at both of these processes in tandem along with the less studied aspects of the theory, such as the contingent conditions and potential consequences of agendasetting. In addition, this investigation builds and extends agendabuilding and agendasetting theory by introducing new actors (competing corporate candidates), a new type of object agenda (stakeholder agenda), a new type of information subsidy (shareholder letters, known as fight letters), and several underexplored or new business journalism sources (Dow Jones News, Bloomberg News, and the Financial Times ) into a major new setting (corporate proxy cont ests). This investigation not only builds and extends agendabuilding and agendasetting theory, but also contributes to the evaluation of the value of this theory in relation to other public relations, mass communication, and media effects theories. A recurring criterion in assessing the value of a theory within the social sciences is in determining its theoretical scope or generality (Chaffee & Berger, 1987; Littlejohn, 2002; Shoemaker, Tankard, & Lasorsa, 2004). If a theory is high in scope, it can be successfully applied to a number of different situations and settings. The usefulness of the agendabuilding and agendasetting perspective in explaining the relationships among source provided information subsidies, the news media, and the public during political election contests is well established. The setting for the current study contested corporate elections provides a compelling new situation for testing the theoretical scope of this theory. As previously noted, there is an acknowledged need for more scholarship specifically on financial public relations and investor relations, two subdisciplines of

PAGE 30

30 public relations (Botan & Hazelton, 2006). Much of the existing research on investor relations and financial public relations has been descriptive and atheoretical in nature, rather than explanatory and predictive. With this in mind, this study helps contribute to the overall body of knowledge within these two specializations, while introducing an established theoretical framework into these two understudied areas. By working within the agendasetting perspective, with its roots in the broader mass communication field, this study also answers calls by several public relation scholars (Botan & Hazelton, 2006; Broom, 2006; Gower, 2006; Grunig, 2006) for theory building that takes a multi disciplinary approach and incorporates external perspectives. Finally, the portion of this study focused on agendasetting consequences responds to appeals for more empirical research that measures the effectiveness of public relations efforts (Hon, 1998; Kim, 2001). This investigation also holds the potential to make important practical contributions. With the number of proxy contests and activist investor campaigns in the U.S. at near record levels (Laide & Mallea, 2009), this research directly addresses an area of growing importance to financial public relations, investor relations, and communication professionals. While communication professionals have been recognized as active participants in corporate proxy contests for at least 50 years (Aranow & Einhorn, 1956; Emerson & Latcham, 1953), the vast majority of scholarly research to date on proxy battles has been from a business and finance perspective. At least one prior study (A. Miller, 1986) into financial predictors of proxy contest winners suggested that future research incorporate non financial variables conceptualized around organizational conflict, power and influence, and/or intervention and change.

PAGE 31

31 Almost twenty five years later, still little is known about what effect public relations efforts, a decidedly nonfinancial variable, have during proxy battles on influencing media coverage and the agenda of the competing campaign, shaping investor opinion, and potentially even contributing to contest outcomes (i.e. voting behavior) This study provides an initial empirically grounded map for practitioners to use in navigating communication strategy and tactics during proxy fights and activist investor campaigns.

PAGE 32

32 Figure 1-1 Agenda building and agendasetting theoretical map during corporate proxy contests

PAGE 33

33 CHAPTER 2 LITERATURE REVIEW History of Shareholder Activism While shareholder activism has enjoyed a resurgence in recent years, shareholders have been trying to hold the board of directors and management of publ icly held U.S. corporations accountable for well over a century (Talner, 1983). For example, in 1885, dissatisfied shareholders of the Broadway and SeventhAvenue Railroad Company formed a committee of shareholders to demand more information from the company on its financial condition (Pound, 1992). As the owners of publicly held firms, shareholders have certain rights, including the right to elect a corporations board of directors. These elected directors, in turn, act as the agents of the shareholders. T he board is responsible for monitoring management, which is responsible for implementing corporate strategy and overseeing the day to day operations of the firm. According to Gillan and Starks (1998), the potential for shareholder activism arises when shareholders believe that the board of directors has failed in its duty, that is, they are dissatisfied with the performance of the board of directors (and presumably the firm) (p. 1). These scholars define an activist shareholder as an investor who tries to change the status quo through voice, without a change in control of the firm (Gillan & Starks, 1998, p. 3). Voice is most frequently expressed in shareholder activism by submitting shareholder proposals for inclusion on a corporations annual proxy statement, in private negotiations with management or board of directors over an issue, or by putting public pressure on a corporation by bringing the issue to the attention of the media (de Bakker & den Hond, 2008; Gillan & Starks, 1998, 2007; Marens, 2002;

PAGE 34

34 Vogel, 1983). The most aggressive and expensive form of shareholder activism is by waging a proxy contest, in which the dissatisfied activist attempts to gain enough shareholder support to replace one or more members of the corporations board of direc tors (Bagley & Savage, 2006; Dasgupta & Nanda, 1997; Ikenberry & Lakonishok, 1993; Manne, 1965). Activist shareholders can be divided into three categories: corporate governance activists, social responsibility activists, and entrepreneurial activists. Th ese first two groups are generally not interested in taking actions that will result in a new person or entity obtaining a fifty percent or greater ownership in the firm, also known as a change in control (Gillan & Starks, 1998). On the other hand, entrepreneurial activists may actively engage in activities, such as a proxy battle, that do result in a change of control (A. Miller, 1986). Corporate governance activists, a mixture of individual investors and pension funds, have historically been concerned with issues such as board composition, board and executive compensation, anti takeover provisions and shareholder rights (Marens, 2002). Social responsibility activists have historically been focused on issues such as labor, the environment, and international relations, as well as varied social and religious concerns (Vogel, 1983). While corporate governance activists generally link their corporate governance concerns with financial performance, social activists may buy stock in a corporation only to gain a platform for voicing their concerns (Marens, 2002). Entrepreneurial activists, such as hedge funds, private equity firms, and wealthy private investors, are generally concerned with improving corporate governance and may comment on some social issues, but they are most motivated by actions that will result in maximizing shareholder value (Gillan & Starks, 2007).

PAGE 35

35 Shareholder activism has been credited with helping to shakeup a number of major U.S. corporations over the years, such as I.B.M., Kodak, American Express, Westinghouse and General Motors (Thompson & Davis, 1997). Taking a step back, though, and looking at the bigger picture reveals that shareholder activists still face an uphill battle in effecting change at corporations. Approximately 8,600 shareholder proposals were put on corporate proxies during 19732004 with less than 10% of these proposals winning majority support from shareholders (Gillan & Starks, 2007). However, more than half of the proposals receiving majority support occurred from 2000-2 004, which suggests that the tide could slowly be turning in favor of the activists (Gillan & Starks, 2007). Corporate governance related proposals typically receive more shareholder support than social responsibility related proposals. For example, governance proposals received 24% shareholder support on average during the 1997 proxy season, while social proposals received less than 7% support on average for this same year (Campbell, Gillan, & Niden, 1998). Turning to the most aggressive form of shareholder activism, during 2008, a total of 479 activist investor campaigns against U.S. publicly held firms were announced, resulting in activists ultimately securing board seats at 84 companies (Laide & Mallea, 2009). Gadflies, Corporate Governance and Social Responsibility ? Up until the mid1980s shareholder activism was largely the domain of individual investors rather than large institutional investors, such as pension funds or hedge funds. (See Figure 21 for a timeline of shareholder activism in the U.S.) The dawn of modern shareholder activism traces its roots to the 1932 annual shareholders meeting of New York Citys Consolidated Gas Co. At this meeting, a small shareholder named Lewis Gilbert watched as the CEO of Consolidated Gas Co. ignored calling on sh areholders

PAGE 36

36 who had raised their hands for questions (Gillan & Starks, 1998; Marens, 2002; Talner, 1983). The meeting was adjourned by inviting shareholders to a buffet (Marens, 2002). This disheartening event helped set in motion Lewis Gilbert and his brother John spending the next 60 years aggressively challenging corporate directors and management in an attempt to level the playing field (Minow, 2002). Small activist shareholders like the Gilbert brothers, who persistently made their case for corporate go vernance improvements in person at scores of corporate annual meetings, came to be called corporate gadflies (Marens, 2002). Another notable gadfly of this era was Wilma Soss, whom founded the Federation of Women Shareholders in response to being rebuffed at the 1947 U.S. Steel shareholders meeting after asking about appointing a female director to the board (Marens, 2002). Gadflies such as the Gilbert brothers, Wilma Soss, and later Evelyn Y. Davis, actively courted media attention and attempted to use i t to their advantage (Marens, 2002). The adoption in 1943 of Rule 14a8 by the U.S. Securities and Exchange Commission (S.E.C.) served as a breakthrough event for the shareholder activism movement (Vogel, 1983). This rule, which governs the submission of shareholder proposals, significantly strengthened shareholder access to the corporate proxy mechanism (Vogel, 1983). With the adoption of Rule 14a8, shareholder activists gained a way to submit corporate governance proposals for inclusion in the proxy statement that is mailed to shareholders, rather than by simply making their case to shareholders on the floor of a companys annual meeting (Gillan & Stacks, 1998). Under Rule 14a8 companies are allowed to exclude proposals that relate to the daily operatio ns of the firm, such as board of director nominations and solicitations in opposition to

PAGE 37

37 management proposals, such as approval for a merger or acquisition (Bagley & Savage, 2006; Gillan & Starks, 1998). In addition, most shareholder submitted proposals ar e advisory and non binding, meaning that even if shareholders vote in favor of a proposal, management doesnt legally have to act upon it (Thompson & Davis, 1997). That being said, such a vote sends a strong signal to a corporations board of directors, and corporations risk facing more aggressive activism, such as a proxy contest, if they choose to ignore such advisory votes. Even with its share of limitations, Rule 14a8 provided corporate governance activists with more of a fighting chance in making their case to shareholders. Until the early 1970s, gadflies such as the Gilbert Brothers accounted for half or more of all shareholder proposals submitted to proxies each year (Marens, 2002). Social responsibility activism became much more prevalent following a 1970 ruling by the U.S. Court of Appeals in which it ordered the S.E.C. to review its support for Dow Chemicals decision to omit a shareholder proposal from its proxy regarding restricting the companys sale of napalm (Gillan & Stacks, 1998, 2007). In turn, the S.E.C. expanded the interpretation of Rule 14a8 so that social responsibility related issues could also be submitted by shareholders for inclusion on corporate proxy statements. This same year, an organization called the Project on Corporate Res ponsibility held a press conference in which it announced that it was submitting nine social responsibility related proposals to be voted on at the annual meeting of General Motors (Vogel, 1983). The S.E.C. ruled that General Motors was required to include two of these Campaign GM social proposals on its proxy statement (Vogel, 1983). While each proposal received less than 3% shareholder support, this campaign helped usher in a

PAGE 38

38 new era of proxy proposals with a social responsibility bent (Vogel, 1983). Major social issues submitted for inclusion on proxies during the first half of the 1970s including stopping investment in South Africa, ceasing military production, and consumer and environmental protection (Vogel, 1983). 1975 marked the first year that a s ocial proposal submitted by shareholders received more than 10% support (Vogel, 1983). In 1977, the University of Minnesota became the first educational institution to sponsor a shareholder resolution related to South Africa (Vogel, 1983). Institutional I n vestors and Entrepreneurial Activists ? As institutional investors, such as pension funds, became larger players in the stock market, their involvement and interest in shareholder activism not surprisingly increased as well. The formation of the Council of Institutional Investors by the California state treasurer in 1985 signaled that public pension funds were no longer content with being passive shareholders (Gillan & Starks, 1998; Thompson & Davis, 1997). Large public pension funds, such as the California Public Employees Retirement System (CalPERS) began submitting shareholder proposals in 1986 (Gillan & Starks, 1998). These proposals were primarily focused on corporate governance, such as changing rules in voting, increasing board independence, and repealing anti takeover amendments (Gillan & Starks, 1998). Harkening back to the era of the gadflies, in the early 1990s, public pension funds began using the media to bring attention to issues at companies in which the funds were seeking changes (Gillan & Starks, 1998, 2007). In the mid1990s union pension funds started to become more active with their stock holdings as well (Gillan & Starks, 2007). With over 140 public, labor and corporate pension funds as members, controlling in excess of $3 trillion in capit al, the Council of

PAGE 39

39 Institutional Investors is today arguably the most influential force in corporate governance related shareholder activism (Gillan & Starks, 2007). The past decade has also witnessed the rise of entrepreneurial activists, such as hedge f unds, private equity funds, and wealthy private investors. Klein and Zur (2009) define an entrepreneurial activist as an investor who buys a large stake in a publicly held corporation with the intention to bring about change and thereby realize a profit on the investment (p. 187). Submitting shareholders proposals is but one way an entrepreneurial activist attempts to implement change and unlock shareholder value. According to Pound (1992), entrepreneurial activists generally use two broad strategies to g enerate financial returns: 1) direct acquisition of a controlling interest in the target company (p. 8) and 2) use of a relatively small amount of voting power to bargain, usually through public suasion (p. 8). While the first strategy is epitomized by acquiring the target and the second by holding a proxy contest, these are but two of many possible moves an entrepreneurial activist can make. Pound (1992) elucidates: Viewed on a broader canvas, there are virtually endless ways entrepreneurs can combine share ownership with suasion so as to exert influence on management. Tactics may be friendly or hostile, and can range from subtle negotiation to brute exertions of financial force. The entrepreneur, seeking maximum profits, chooses those tactics that combine acquisition and suasion so as to exert the critical degree of influence over management at minimum cost. (p. 8) The reasons entrepreneurial activists state for initiating an activist shareholder campaign include changing the composition of the board of directors, opposing a merger, selling or merging a company, changing management, improving the company dividend, or implementing a share repurchase plan (Gillan & Starks, 2007; Klein & Zur, 2009). The use of the proxy contest as a form of shareholder activism has increased with the rise of entrepreneurial activists. The number of proxy fights rose from just 30 in

PAGE 40

40 2004 to 91 in 2006 (Gillan & Starks, 2007). While the decline in the U.S. stock market during 2008 curtailed the overall level of entrepreneurial activism (479 campaigns in 2008 versus 506 in 2007), a new record was set in terms of the number of proxy contests (123 proxy fights in 2008 versus 108 in 2007). Large, well known corporations that have been the subjects of entrepreneurial shareholder activism in recent years have included Home Depot, Wendys, Time Warner and Applebees (Gillan & Starks, 2007). At least during the 20032005 timeframe, entrepreneurial activists were successful in getting corporations to acquiesce to their initial demands over 60% of the time (Klein & Zur, 2009). Corporate Proxy Contests At the vast majority of the annual meetings of publicly held corporations held each year, the company provided nominees for the board of directors run unopposed. In advance of the annual meeting, shareholders receive a proxy statement and proxy card in the mail that provides them with essentially two choices either vote for the nominees recommended by the corporation or abstain by withholding their vote (Monks & Minow, 2008). At many corporations, this abstention carries only symbolic weight, though, as it only takes one yes vote on a proxy from a single shareholder to get a slate of unopposed director nominees elected (Bagley & Savage, 2006). The term proxy comes from the legal right a shareholder has to designate an agent or proxy to vote their shares and act on their behalf (Holton, 2006). A proxy has historically allowed shareholders to exercise their rights as owners without having to show up in person at the annual meeting to cast their vote (Holton, 2006). When one or more activist shareholders are significantly dissatisfied with the performance of a corporations board of directors and management, they may choose to

PAGE 41

41 try and unseat all or part of the board by nominating their own competing slate of nominees as replacements (A. Miller, 1986). The resulting competition for shareholder votes is known as a proxy contest or proxy battle, since the competing group also mails out its own proxy statements and proxy cards seeking the support of shareholders for its alternative slate (Holton, 2006). There are two types of proxy contests: representation contests and control contests (Miller, 1986). A representation contest occurs when an activist shareholder group attempts to win less th an 50% of the directorships on a firms board (Miller, 1986). A recent example of a representation contest was the 2005 battle between investor Carl Icahn and Blockbuster in which he won a minority number of seats on the companys board (Bagley & Savage, 2006). A control contest, on the other hand, occurs when an activist shareholder group attempts to win more than 50% of the directorships on a firms board, resulting in a change in control of ownership (Miller, 1986). Icahns unsuccessful attempt in 2008 to gain majority control of Yahoo! Inc.s board is an example of a control contest (Ragas et al., 2009). While proxy battles generally center around shareholders choosing between two competing director slates, a contest may also occur when an activist opposes a major transaction supported by the incumbent board (Bagley & Savage, 2006). The 2001 merger between Hewlett Packard and Compaq, which was unsuccessfully opposed by the Hewlett family, is an example of this type of contest (Cone, Fischel, Pelnar, & Ross, 2004). Structure of a Proxy Contest As previously discussed, a proxy contest is generally the most aggressive, costly, and procedurally complex expression of shareholder activism. The dissatisfied shareholder group, generally known as either the challenger or dissident, must pay for

PAGE 42

42 campaign expenses (proxy mailings and solicitations, legal and financial advisor fees, communication consultants, etc.) out of its own pockets, while the corporation, known as the incumbent, has the advantage of being able to tap the corporate treasury for its campaign (Aranow & Einhorn, 1956; Ikenberry & Lakonishok, 1993; Manne, 1965; A. Miller, 1986; Wilcox, 2004). Generally, the challenger is only reimbursed if its slate is elected (Bagley & Savage, 2006). Campaign expenses during a contest can run into the millions of dollars, as illustrated by the highprofile 2009 Target Corporation Pershing Square Capital contest, in which the two sides collectively spent $20 million (Crosby, 2009). In many ways, the communication aspect of proxy contests resemble political election campaigns with the public relations efforts of the two competing corporate actors each attempting to build the media agenda and garner shareholder votes (Sauerhaft, 1997; Wilcox, 2004). According to Marken (2005), an adviser to firms during proxy battles and takeovers, a solid PR strategy and program can be the strongest strategic and proactive weapon that can be used to advance managements case and point as well as control the issues (p. 22). Just as nati onal political campaigns include a mix of internal campaign staffers and outside consultants, the modern proxy contest also includes a diverse group of specialists. The internal players on the incumbent side of a contest generally include members of top ma nagement and the investor relations and corporate communication teams (Wilcox, 2004). External advisors that are typically retained by corporations during a proxy battle include an investment banker, corporate legal counsel, public relations counsel, and a proxy solicitor (Wilcox, 2004). Many, if not all, of these same types of advisors will be retained by the challenger, particularly if the activist

PAGE 43

43 shareholder is a hedge fund or private equity firm. Financial public relations firms have been recognized as a standard component of proxy contests for over fifty years (see Aranow & Einhorn, 1956; Emerson & Latcham, 1953). These external financial public relations advisors assist with developing the communication strategy and issue positions, as well as preparing the communication materials throughout the campaign. The proxy solicitor contributes to the strategy and manages the solicitation, projection, and tabulation of votes. MacKenzie Partners (2009), a large proxy solicitor, describes its work as follows: Pr oxy contests are decided by the quality of communications to shareholders, by how well we devise and disseminate arguments for changing or defending against a change in control. Whether we advise a dissident shareholder or incumbent management, one of our key strategic roles is to frame the issues and shape the message to be delivered to a companys shareholders. The goal is to convince shareholders to vote their proxies in favor of our client and against the opponent. (paras. 1011) History of Proxy Conte sts The corporate proxy contest and the threat of waging a contest are not new additions to the activist shareholder toolbox. Early notable proxy contests included W.C. Durants successful battle for control of General Motors in 1915 and John D. Rockefelle rs successful campaign in 1929 to oust the chairman of his own board of directors at Standard Oil (Pound, 1992). While wealthy financiers were the most adept at using the proxy contest to enact their corporate agenda, small investors also eventually found success through this route. In 1950, for the first time, a group of small investors successfully banded together and won enough shareholder support to defeat an incumbent slate of directors (Emerson & Latcham, 1953). Owning just 1% of the total stock outstanding, investors in Sparks Withington Company, a New York Stock Exchange listed company, were able to secure enough additional shareholder support

PAGE 44

44 to seat five new directors and get two of its three proposals enacted (Emerson & Latcham, 1953). The use of proxy contests began to decline in the late 1950s with the emergence of the cash tender offer, which was deemed to be an easier, more cost effective tool for gaining corporate control (Pound, 1992). Unlike a proxy contest with its high upfront costs and uncertain outcome, in making a tender offer the financier simply announces the price at which they are willing to pay for a stock and then waits to see if investors tender to her a sufficient number of shares (Pound, 1992). The proxy contest enjoyed a renaissance in the late 1970s and into the mid1980s after it was rediscovered by the corporate raiders aggressive financiers who took ownership stakes in companies and then threatened to take over these companies unless their demands were met (Bruck, 1989; Pound, 1992). The goal of many proxy fights during this time period was to gain enough board seats to force the sale of a company at a premium to its stock price, thereby generating a handsome profit for the raider (Bruck, 1989). Other times, the announcement of a proxy fight would attract enough attention that the target company would soon thereafter be acquired, negating the need to follow through on trying to replace the board (Bruck, 1989). A decline in the fortunes of the raiders, combined with a rise in the costs to wage proxy battles, eventually curtailed the use of this tactic once again (Pound, 1993; Gillan & Starks, 2007). The latest wave of shareholder activism utilizing proxy contests began in the mid1990s with the rise of activist entrepreneurs, namely profit minded, deeppocketed hedge funds and private equity firms (Pound, 1993; Klein & Zur, 2009). The activist entrepreneurs of today are generally more long term oriented and relationshipfocused

PAGE 45

45 than the corporate raiders of the early -1 980s (Gillan & Starks, 2007). With this in mind, activist entrepreneurs generally are loath to threaten or launch a proxy battle for board of director seats and other changes right away. Instead, these activists first try to employ less confrontational tactics to advance their corporate agendas, such as submitting shareholder proposals and requesting meetings with management (Gillan & Starks, 2007). Only after being rebuffed do activist entrepreneurs generally take their campaign to the public arena and the financial press. A proxy contest often serves as more of a last resort (Wilcox, 2004). This being said, there has been considerable growth in the number of proxy contests waged in recent years. Even with the downturn in the capital markets in 2008, the number of contests in the U.S. remained at high levels (Laide & Mallea, 2009). As previously referenced, there were 123 contests in the U.S. in 2008, nearly triple the number of contests in 2004 (Laide & Mallea, 2009). Figure 22 tracks the growth in the number of proxy contests by year for 20042008. Through the latter part of March 2009, 71 proxy fights for board seats had been announced in 2009, compared with 69 at the same point last year (Sorkin, 2009). The Proxy Access Rule While entrepreneurial activi sts have the resources needed to wage a proxy contest, for many shareholder activists, a proxy contest is too expensive and too complex to be a viable option for changing the composition of a corporations board of directors (Westbrook, 2009). A new proxy access proposal by the S.E.C. could significantly reduce the expense of proxy fights by providing certain long term shareholders under certain conditions with direct access to the corporate proxy materials as a mechanism for nominating directors (Monks & Minow, 2008). Under this proposed rule, an activist shareholder wishing to nominate its own slate of director

PAGE 46

46 candidates would no longer need to prepare and mail its own proxy statements and cards seeking support for its alternative slate (Bagley & Savage, 2006). Instead, corporations would be required to permit direct shareholder nominations of directors, with these alternative nominees appearing within the corporate proxy statements which are mailed out to shareholders (Bagley & Savage, 2006). Under such a scenario, shareholders would still get to choose between two competing sets of director candidates (one proposed by the incumbent and one proposed by the challenger), but the cost of waging the challenger campaign would likely be greatly reduced (Youngl ai, 2009). According to S.E.C. chairwoman Mary Schapiro, the agency believes that enabling investors to participate meaningfully in the nomination of directors through proxy access will foster a sense of enhanced legitimacy to the director nomination proc ess (Younglai, 2009, para. 13). Under the proposed proxy access rule, a shareholder must meet certain ownership thresholds before they are allowed to nominate directors directly through the corporate proxy. A shareholder would be required to have owned for at least one year at least one percent of the stock outstanding in a large company (defined as a market capitalization of greater than $700 million) before they would be able to list their director candidates on the proxy (Westbrook, 2009). This ownership threshold for direct nominations would rise to three percent ownership for medium sized companies (market capitalization of $75 million to $700 million) and five percent for small companies (market capitalization of less than $75 million) (Westbrook, 2009). The rule would also limit the number of directors that a shareholder could nominate. If a board had three members, the shareholder could only nominate one director (Westbrook,

PAGE 47

47 2009). If the board had more than three members, the shareholder could nominate directors representing no more than 25% of the total number of board seats (Westbrook, 2009). While opposed by various business groups, the proxy access rule could be in place for the 2011 proxy season, potentially resulting in a new wave of chall enger campaigns by smaller activist shareholders (Lynch, 2009) In short, the current investigation into communication during corporate proxy contests is already a timely topic for practitioners, as evidenced by the rise in the number of proxy battles in r ecent years; it could become even more so with the implementation of the S.E.C.s proxy access rule. Investor Relations and Financial Public Relations In response to the increasing complexity of the capital markets and growing shareholder activism, investor relations and financial public relations have emerged over the past two decades as important communication functions for publicly traded corporations (Dolphin & Fan, 2000; Dolphin, 2003; Marston, 1996; Rao & Sivakumar, 1999; Useem, 1996). The National Investor Relations Institute (NIRI), a large U.S. based trade association for investor relations professionals, defines investor relations as a strategic management responsibility that integrates finance, communication, marketing and securities law compliance to enable the most effective twoway communication between a company, the financial community, and other constituencies (National Investor Relations Institute, 2003, para. 1). According to NIRI (2003), investor relations ultimately contributes to a companys securities achieving fair valuation (para. 1). K. Clarke (2004) notes that the terms investor relations and financial public relations are often used synonymously, but argues that these are separate functions. According to K. Clarke (2004), while investor relations focuses on managing relationships with

PAGE 48

48 large institutional investors, financial public relations is the more general management of communication between a publicly traded company and its financial audiences. As submitted by K. Clarke (2004), financial PR is concerned with more general communication with investors as well as raising awareness and building understanding among opinion formers, such as the media and analysts who influence both small and large investors and potential investor s (p. 205). While investor relations and financial public relations have arguably grown into two of the more influential areas within the corporate communication field (J. Davis, 1995; Regester, 1990), scholars in disciplines affiliated with these two specialty areas have paid them relatively little attention (Farragher, Kleinman, & Bazaz, 1994). For example, while investor relations is acknowledged as a specialty discipline within the broader public relations field, as evidenced by its discussion in major public relations text books (e.g. Cutlip, Center, & Broom, 1999; Newsom, Scott, & Turk, 1993; Wilcox, Ault, & Agee, 1995), only a handful of empirical investigations on investor relations have appeared in public relations academic journals (e.g. Cameron, 1996; Laskin, 2006, 2009; Niedziolka, 2007; Petersen & Martin, 1996). Given this inactivity, public relations scholars Botan and Hazelton (2006) lament that if we continue to ignore this area it will become the exclusive domain of academic fields that do pay attention to it (p. 16). Botan and Hazelton (2006) cite economics, finance, and management as other fields that could potentially take a leadership role in conducting academic research on investor relations. Empirical Research on the Field Through books and book chapters (e.g. K. Clarke, 2004; Cole, 2004; Collis, 2004; Miller, 1991, Sauerhaft, 1997; Useem, 1996) practitioners have provided broad brush

PAGE 49

49 stroke overviews of investor relations and financial public relations. Most of the empirical research to date on investor relations and financial public relations has focused on exploring these two functions in the context of large corporations based in either the U.S. or the U.K. A series of studies conducted by Dolphin and a colleague (e.g. Dolphin, 2003, 2004; Dolphin & Fan, 2000) tracked the growth of investor relations within large U.K. companies. Dolphin (2003) found that seven (39%) of the eighteen corporations in his study had set up a separate investor relations department. Thirteen (72%) of th ese companies identified shareholders as a key audience (Dolphin, 2003). Dolphin (2004) also found that a strong majority of the communication executives he interviewed thought that investor relations had increased in importance over the past decade. Research by Laskin (2006, 2009) has explored investor relations within Fortune 500 companies in the United States. A survey by Laskin (2006) found that approximately 65% of Fortune 500 companies have a separate investor relations department, while 27% of these firms assign investor relations to the finance/treasury department. Only 7% of respondents reported managing the investor relations function from within a PR department (Laskin, 2006). While the Dolphin and Laskin studies demonstrate that investor relati ons is receiving growing attention within at least large U.S. and U.K. corporations, perceptions of this disciplines importance seem somewhat mixed outside of corporate communications. A survey of chief executive officers of Floridabased publicly traded companies, many of which were relatively small in size, found that these CEOs did not perceive investor relations as a strategic management activity, but rather more of a technician role that carries out duties set by others (Petersen & Martin, 1996). Thes e

PAGE 50

50 Florida CEOs also favored having financial affairs executives and departments, rather than public relations executives, managing the investor relations function (Petersen & Martin, 1996). A survey of Public Relations Society of America (PRSA) members rev ealed that public relations practitioners perceived investors relations as falling within the realm of public relations (Hong & Ki, 2006). However, the view among survey respondents in which investor relations wasnt practiced within their organizations was that they didnt think investor relations was an important function of public relations (Hong & Ki, 2006). In short, there remains at least some skepticism over the value and importance of investor relations. Survey research by Laskin (2009) has provided arguably the most detailed data on the importance investor relations professionals within large U.S. corporations accord various investor publics, how influential they perceive these publics in effecting the companys share price, and how much time they spend on various investor relations activities. The two activities that investor relations professionals say they participate in most frequently are responding to requests from shareholders, analyst or stockbrokers, and roadshows, presentations and conferences (Laskin, 2009). Respondents also reported spending the most time communicating with institutional investors (almost 40% of time), followed by professional stock analysts (almost 33% of time) (Laskin, 2009). Institutional investors were rated the most important public for investor relations professionals, followed by stock analysts, and internal publics (Laskin, 2009). Finally, the three publics perceived by respondents as being the most important in influencing the stock price were institutional investors, stock analysts, and mass media (Laskin, 2009).

PAGE 51

51 As a potential launching point for theory building within the field, Palter, Rehm, and Shih (2008) introduced a model for segmenting professional investors by stock ownership behavior. This simple model divides professional investors into three groups: intrinsic investors, mechanical investors, and traders (Palter et al., 2008). Intrinsic investors invest in a stock only after rigorous due diligence and generally hold shares for several years. These investors generally take large ownership positions, are actively engaged in understanding the strategy of the business, and tend to support management through short term stock price volatility (Palter et al., 2008). Mechanical investors invest in a large number of stocks, make investment decisions based on mathematical models and are not interested in understanding company strategy (Palter et al., 2008). Traders seek short term financial gain by moving quickly in and out of stocks. These investors seek to gain short term informational advantages about a company, but arent interested in understanding the company at a deep level (Palter et al, 2008). Based on this model, Palter et al. (2008) assert that corporate executives should focus their communication efforts on building relationships with intrinsic investors, rather than the other two investor segments. According to Palter et al. (2008), the result of this approach should be a better alignment between a companys intrinsic value and its market value (para. 4). Outside of the research by Palter et al. (2008) on shareholder segmentation, this review of the investor relations and financial public relations literature reveals that research in this area has been largely exploratory or descriptive and atheoretica l in nature. Explanatory or predictive research is rare. The state of scholarship in this emerging, still relatively young field doesnt seem that different than the state of

PAGE 52

52 scholarship in the broader public relations field in the early 1980s. A now classic content analysis by Ferguson (1984) of public relations journal articles published from 19751984 found that only 4% of these articles engaged in theory building. A subsequent analysis of 25 years worth of PR journal articles (through the year 2000) found that nearly 20% of these articles had contributed to theory development, suggesting that scholars are doing a better job at predicting and explaining the field (Sallot et al., 2003). By applying the agendabuilding and agendasetting framework to the study of PR efforts, media coverage, and investor opinion during proxy contests, the current investigation hopes to contribute to the emerging theory development within investor relations and financial public relations. While theory building remains at a nascent stage, investor relations and financial public relations scholarship has begun to investigate more specialized areas of the field. For example, research has analyzed perceptions of the annual statement (G. Clarke & Murray, 2000), the management (Bollen, Hassink, de Lange, & Buijl, 2008) and communication aspects (Hong & Kiousis, 2007) of investor relations Web sites, issues with communicating corporate social responsibility to investors (Hockerts & Moir, 2004), and the perceived credibility of investor relations versus public relations spokespersons (Hong & Cho, 2008). So far, though, most investor relations and financial public relations scholarship has yet to empirically evaluate communication strategy and tactics during company transforming events, such as initial public offerings (IPOs), bankruptcy reorganizations, hostile or friendly mergers and acquisitions (M&A), shareholder activism, and corporate proxy contests. The lack of attention to this latter area is ironic

PAGE 53

53 as research (e.g. Rao & Sivakumar, 1999) has identified shareholder activism as a force behind the growth of modern investor relations departments within corporations. Relationship with the Financial News Media Particularly relevant to the current study is research that has probed the r elationship between investor relations or financial public relations efforts and the financial press. While practitioners such as Silver (2004) claim that management is often perceived through the lens of the financial news media (p. 61) and the media d rives what much of Wall Street thinks about public companies (p. 73) these claims are often made absent of empirical evidence. Aeron Davis, a media sociologist, has conducted perhaps the most detailed empirical investigations into the role of the mass media in investor relations and financial public relations (e.g. A. Davis, 2002, 2006). In addition to reviewing various secondary data, A. Davis conducted over 100 semi structured interviews with financial PR practitioners, business journalists, and investment managers in the U.K. Based on his analysis, A. Davis (2006) concluded that on the one hand, the financial news media appear to be of decreasing importance to the professional communicators of financial messages (p. 8). On the other hand, he found that the media still appears to be central to much market communications (p. 8) and, at times, can still have a powerful impact on investment patterns (A. Davis, 2006, p. 7). His interviews found that shaping coverage was most important for practitioners w hen their firm or client was facing bad news situations and crisis moments (A. Davis, 2002, p. 10). According to A. Davis (2002), the importance of effective news management is also vital during takeover activity (p. 10). A. Davis has also probed indepth the broader relationship between corporate communication efforts and business journalism in the U.K. (A. Davis, 2002). In his view,

PAGE 54

54 corporate influence plays an increasing role in shaping business news coverage because corporations have become so paramount to the success of business media organizations (A. Davis, 2002). Corporations serve as the primary advertisers, news sources, and consumers of business news (A. Davis, 2002). He sums up his reasoning as follows: In this news sector, businesses pay f or all advertising, dominate as sources, have a monopoly in the supply of information subsidies and are the main consumers of business and financial news. It therefore stands to reason that, even if companies are in conflict, business agendas and business norms and values will shape financial and business news. (A. Davis, 2002, p. 60) Davis is not alone in finding that corporations and the financial press are intertwined and that corporations often win at what Gans (1979) classically called the source journ alist tug of war (p. 117). While a variety of studies have documented the influence of sourceprovided information, such as public relations materials, on general media content (e.g. Cameron, Sallot, & Curtin, 1997; Cutlip, 1962; Kaid, 1976; Sigal, 1973) at least two studies have looked specifically at the influence of corporate actors on business media content. Blyskal and Blyskal (1985) estimate that nearly 50 percent of the business news stories in The Wall Street Journal are influenced by public relations. Similarly, over 60 percent of the content in the companies and market section of the Financial Times is believed to be influenced by public relations efforts (Press lets its prejudice, 1994). Just as there have been few empirical investigations into the relationship between investor relations or financial public relations and the financial news media, the scholarly literature on communication efforts specifically during corporate proxy contests is similarly scarce. While practitioners have written several book chapters (e.g. K. Clarke,

PAGE 55

55 2004; Collis, 2004; Miller, 1991; Sauerhaft, 1997; Wilcox, 2004) that discuss communication strategy and tactics during mergers and acquisitions, takeovers, and proxy contests, the most detailed empirical investigations of the role of financial public relations during these transformational events have been conducted by A. Davis (2002) and Newman (1983). The only scholarly research to date that has explicitly tried to link theory and practice in this area has been Ragas et al. (2009), which assessed the dueling PR efforts, media coverage, and investor opinion during the 2008 Yahoo! Icahn proxy contest from an agendabuilding and agendasetting perspective. Taking a case study approach, A. Davis (2002) analyzed the com peting communication campaigns during Granada Groups attempted takeover of Forte plc in the midnineties. A. Davis (2002) found that these two large U.K. based companies exerted significant resources in trying to influence media coverage. Over the course of the two month battle, the dueling campaigns collectively generated 200 news releases (A. Davis, 2002). While both sides devoted significant resources to financial public relations, case study interviewees provided mixed responses on how significantly fi nancial public relations contributed to the ultimate outcome of the contest (A. Davis, 2002). Granada ultimately successfully persuaded Forte shareholders to support its buyout offer, but only after significantly increasing its offer price. While media cov erage may have influenced shareholder opinion at the margin, by and large, interviewees believed that the value of the offer was the key factor in deciding the outcome of this takeover (A. Davis, 2002). An analysis by Newman (1983) of twelve corporate take over battles in the U.K. from 19581982 yielded a similar conclusion. She concluded that

PAGE 56

56 public relations and media coverage played a role in these battles, but that, ultimately, the price or value of the offer decided the outcome (Newman, 1983). While financial public relations efforts may have limited influence in directly deciding the outcome of a contested corporate transaction, shareholders are but one stakeholder that is affected during such an event. Effective financial public relations and investor relations during hostile takeovers and proxy contests has been credited with maintaining the morale of employees, especially when a company is under attack from a hostile acquirer (L. Davis, 1989). In addition, financial public relations efforts, both dir ectly and indirectly through the financial press, during takeovers and proxy contests is credited with maintaining the confidence of key external stakeholders, such as customers and suppliers (Sauerhaft, 1997). Finally, in some contested transactions, fina ncial public relations is tasked with appealing for government agency or legislative support for blocking or approving a transaction (Sauerhaft, 1997). Jack Levy, chief of mergers and acquisitions for Merrill Lynch, a large investment banking firm, described the value of financial public relations and the financial press during a transaction as follows: Theyre important today because so many of the strategic transactions are story deals. You need skilled public relations professionals to make sure the story is conveyed clearly, consistently and frequently, so the message is understood by the various constituencies: investors, employees, customers, what have you. (Bagli, 1997, para. 21) In short, while corporate executives and financial professionals, such as investment bankers and corporate lawyers, largely seem skeptical that a corporation will win or lose a corporate takeover battle or proxy contest based solely on who controls the financial media and investor agendas, they view the financial news media as a playing field that they cant afford to ignore either (Bagli, 1997; A. Davis, 2002; L. Davis, 1989; Newman,

PAGE 57

57 1983). In short, the successful prioritization and framing of a corporations issue positions in the financial press may not only have an effect on influencing the views of shareholders and investment analysts, but it also has the potential to shape the perceptions of other key stakeholders of the firm, such as employees, customers, suppliers, government agencies, and legislators (Wilcox, 2004). Evolution of Agenda Setting Theory The notion of the press influencing what the public thinks about and potentially even what it thinks stretches back at least several decades before the start of modern mass communication research in the 1940s and 1950s. In his classic Public Opinion (1922), Walter Lippmann described the role of the press as being like the beam of a searchlight that moves restlessly about, bringing one episode and then another out of the darkness into vision (p. 364). Lippmanns timeless analogy recognizes that there are many topics and issues swirling about at any one time within the public sphere, many of which the public learns about mainly secondhand through sources such as the press, rather than through direct experience. Lippmann (1922) posited that those topics and issues which the media chooses to shine its searchlight upon often become the topics and issues that the public perceive as most important. In short, the media helps paint the pictures in our heads and serves as an important conduit to the world outside (p. 3). Decades later, Lippmanns conjectures on the world outside and the pictures in our heads, found new life in the now classic observation by Bernard Cohen (1963) that the media: may not be successful much of the time in telling people what to think but it is stunningly successful in telling its readers what to think about (p. 13). Cohens assertion was first put to an empirical test by Max McCombs and Don Shaw in their

PAGE 58

58 study of a group of undecided voters in Chapel Hill, North Carolina during the 1968 U.S. presidential campaign (McCombs & Shaw, 1972). Combining a content analysis of media coverage with a survey of a small group of undecided voters, this initial test of the agendasetting hypothesis found a significant positive correlation between the media and public agendas regarding political issue salience. The issues receiving the most media attention those issues that were the most salient were perceived as being the most important by these voters. In the parlance of agendasetting, the mass media set the agenda of issues for the pubic during the campaign. This finding helped shake up the limited effects paradigm of the time (e.g. Klapper, 1960), providing important new empirical weight behind the notion that the media may not significantly influence what the public thinks (attitudes and opinions), but it can greatly influence what the public thinks about (cognitions). Since the publication of the seminal Chapel Hill study (McCombs & Shaw, 1972), the original agendasetting proposition the transfer of salience from the media agenda to the public agenda has been largely reaffirmed in several hundred subsequent empirical inquiries (Dearing & Rogers, 1996; McCombs, 2006). Key to the early th eoretical development of agendasetting was a series of longitudinal studies (e.g. Funkhauser, 1973; Shaw & McCombs, 1977; Weaver, Graber, McCombs, & Eyal, 1981; Winter & Eyal, 1981) which detected a cause andeffect relationship between the media agenda and the public agenda. In other words, these studies found that the media generally influence s the publics priorities, rather than merely reflect ing the publics priorities. After this initial wave of longitudinal studies, a series of controlled laboratory experiments conducted by Iyengar and Kinder (1987) provided further causal evidence

PAGE 59

59 that the media agenda predominantly influences the public agenda, rather than vice versa. Agendasetting theory has expanded in both depth and breadth over the past four decades. While the majority of agendasetting investigations continue to occur in the public sphere, namely political elections, the transfer of salience from one agenda to another has also been explored in areas such as education (Rodriguez, 2000), profes sional sports (Fortunato, 2000; Seltzer & Dittmore, 2009), and organized religion (Buddenbaum, 2001). In particular, a stream of agendasetting research has emerged in the corporate sphere (e.g. Berger, 2001; Berger et al., 2002; Carroll, 2004; Carroll & M cCombs, 2003; Kieffer, 1983; Kiousis et al., 2007; Meijer & Kleinnijenhuis, 2006; Ohl et al., 1995; Ragas et al., 2009), which is the setting for the current study. While the relationship between the mass media and the public remains a focal point of agend asetting research, the theoretical map has been significantly expanded since the seminal Chapel Hill study (McCombs & Shaw, 1972). McCombs (2005) describes the chronological development of agendasetting as having progressed along five distinct stages, all of which remain active areas of inquiry: I. Basic agendasetting effects II. Attribute agendasetting III. Psychology of agendasetting effects IV. Sources of the media agenda V. Consequences of agendasetting effects The following sections will review the key concepts that form the foundation of agendasetting theory and the specific aspects of the theory which encompass each stage. Transfer of Object and Attribute S alience As agendasetting theory has consistently expanded into new domains over the past forty years, the core theoretical proposition has remained consistent the transfer

PAGE 60

60 of salience from one agenda to another. In the vast majority of agendasetting investigations, the unit of analysis on each agenda has been an object a p ublic issue (McCombs & Shaw, 1993, McCombs, 2005). The term object is used by agendasetting scholars similar to how social psychologists refer to an attitude object (McCombs, 2005, p. 546). For example, the original Chapel Hill study (McCombs & Shaw, 1972) found a positive, near perfect correlation between the agenda of public issues, or objects, presented by the news media and the perceived importance accorded this same set of public issues, or objects, by a group of undecided voters during the 1968 presidential election. This transfer of object salience from one agenda to another has come to be known as first level agendasetting (McCombs, Lopez Escobar, & Llamas, 2000). To revisit Cohens (1963) classic assertion, first level agendasetting evaluates the salience of what to think about. While agendasetting is best known for studying issues in the public sphere during political elections, McCombs and his colleagues have noted on several occasions (e.g. Lopez Escobar, Llamas, McCombs, & Lennon, 1998; McCombs, 2005, 2006; McCombs et al., 2000; McCombs, Llamas, Lopez Escobar, & Rey, 1997) that public issues need not be the only objects which can be analyzed from the agendasetting perspective. The agendasetting metaphor of a set of objects competing for attention is applicable in many settings (McCombs & Shaw, 1993). According to McCombs et al. (1997), when viewed in the abstract, the objects could be a set of political candidates, competing brands of consumer goods, or whatever (p. 441442). The hier archy of importance conferred upon objects by news media coverage and the communications of other actors extends well beyond just political issues (Perloff, 1998).

PAGE 61

61 Beyond probing for traditional political issuebased agendasetting effects, Tedesco has also probed on multiple occasions for what he calls strategy agendasetting (Tedesco, 2001, 2005b). According to Tedesco (2001), a significant aspect of political campaign strategy involves not just a focus on making certain issues salient, but also a frami ng of the campaign discourse around certain audiences and processes. With this in mind, the strategy agenda conceptualized by Tedesco consists of two sets of frames: 1) political campaign processes, such as ads, polls, debates, speeches, and endorsements, and 2) appeals to audience segments, such as conservatives, liberals, minorities, women, and the elderly. Tedesco (2001) found convincing evidence of both issue and strategy agendasetting during the 2000 presidential primaries. A similar investigation of the hot phase of the 2004 presidential election also found evidence of issue agendasetting, while signs of strategy agendasetting were more muted (Tedesco, 2005b). Following the lead of Tedesco in examining multiple types of object agendas, an analysis of the 2005 Virginia gubernatorial election found evidence of both issue and audience agendasetting between the candidates and the media, as well as between the candidates (Dunn, 2006, 2009). In an effort to break new theoretical ground in evaluating traditional agendasetting effects, the current study will also analyze two different sets of objects the traditional issue agenda and the stakeholder agenda (similar to audience segments) used in the communication efforts of candidates, news media coverage, and the discussions of investors during corporate proxy contests. While the new construct of a stakeholder agenda will be applied in a corporate context in the current study, this construct could also have applicability to other settings, such as politi cal, crisis, non profit, and public

PAGE 62

62 information/social action communications. More specifically, the current investigation will test for agendasetting relationships among media coverage and investor opinion regarding both issue salience and stakeholder salience in the new context of corporate proxy contests. In addition to probing for the presence of agendasetting relationships the current study will assess the direction of influence between the priorities in news media content (as measured by newspaper coverage and news wire coverage) and investor opinion regarding these two different types of object agendas (issues and stakeholders). The mass media does more than just increase the salience of an object or a set of objects competing for attention on the public agenda. Each object has numerous attributes the characteristics and traits which help describe and fill out the picture of an object, be it a political candidate, elected official, a public issue, brand or an organization (McCombs et al., 1997) When the mass media focuses its attention on an object, it chooses to emphasize some attributes of the object more than others. Many attributes are ignored entirely. Other attributes only receive a passing mention. Just as objects vary in salience on the media agenda so do the attributes of these objects (McCombs et al., 1997). For example, during a political election contest, a significant amount of news media coverage discussing the qualifications of a particular candidate could increase the publics perceived salience of this attribute when thinking about that candidate. As outlined by McCombs and colleagues (2000), both the selection by journalists of objects for attention and the selection of attributes for detailing the pictures of these objects are powerful agendasetting roles (p. 78). The transfer of attribute

PAGE 63

63 salience from one agenda to another has come to known as secondlevel agendasetting (McCombs et al., 2000). To once again revisit Cohens (1963) classic assertion, secondlevel agenda set ting asserts that the media may not only tell the public what to think about (object salience), but may also influence how to think about (attribute salience) those objects. Said another way, first level agendasetting signifies attention to media cont ent, while secondlevel agendasetting signifies comprehension of media content. Figure 23 provides a graphical representation of the transfer of object (first level) and attribute (secondlevel) salience among the media and public. Two different types o f attributes that aid in comprehension of media content have been identified in the literature: substantive attributes and affective attributes (McCombs & Evatt, 1995; McCombs et al., 2000). Substantive attributes are the cognitive characteristics that describe objects in media content, while affective attributes refer to the valence dimension (i.e. positive, negative, or neutral) of attribute salience in media content (Kiousis et al., 2007; Sheafer, 2007). A particular set of substantive attributes that have received attention in the literature are called candidate image attributes (Kiousis, Mitrook, Wu, & Seltzer, 2006; McCombs et al., 1997; McCombs et al., 2000). An example of these attributes is ideology issue positions, biographical information, perce ived qualifications, personality, and integrity (Kiousis et al., 2006). Contingent Conditions of Agenda -S etting An important, but often overlooked, area of agendasetting theory is the research into the psychology of agendasetting effects. While agenda -se tting theory has demonstrated that the sheer frequency of objects appearing on the media agenda often leads to increased concern for these objects on the public agenda, this perspective

PAGE 64

64 does not mark a return to the bullet or hypodermic needle theories (Severin & Tankard, 2001). The public are not passive automatons waiting patiently to be programmed by the priorities in media content (McCombs, 2006). Rather, the media has a stronger influence on public opinion under certain conditions than under others (Perloff, 1998). As explained by Rogers and Dearing (1988), agendasetting does not operate everywhere, on everyone, and always (p. 569). This sentiment is echoed by Takeshita (1997), whom observed that, agenda setting is a very important but hardly univer sal effect (p. 27). Research into the moderators, or conditions, that enhance or limit agendasetting effects has come to be known in the literature as the contingent conditions of agenda setting (McCombs, 2005, 2006; McCombs & Shaw, 1993; Perloff, 1988; Rogers & Dearing, 1988, 2000). The most well known contingent condition of agendasetting is the concept of need for orientation developed by Weaver and McCombs (McCombs & Weaver, 1973; Weaver, 1977, 1980). Need for orientation refers to the tendency of an individual to seek out information about an issue or object in the news media. Research into need for orientation has demonstrated that those who engage in high informationseeking on a topic or object are more susceptible to agendasetting effects, whe reas these effects are minimized for those who are low in need for orientation on a topic or object (e.g. Matthes, 2005; Wanta, 1997; Weaver, 1977, 1980). Need for orientation has traditionally been comprised of two lower order concepts: relevance and uncertainty Need for orientation is at its highest when a topic is deemed highly relevant by an individual and this person is also uncertain of their position on this topic. Conversely, when a topic is low in both relevance and uncertainty to the individual, there is a low

PAGE 65

65 need for orientation. A situation in which relevance is high, but uncertainty is low, suggests a moderate need for orientation, and a moderate agendasetting effect. In retrospect, while need for orientation had not yet been explicitly conc eptualized, the original Chapel Hill study (McCombs & Shaw, 1972), which surveyed a small group of registered, undecided voters during the 1968 U.S. presidential election, was an optimal test of this psychological explanation (McCombs, 2005, 2006). As prev iously discussed, this study found a near perfect positive correlation between the media agenda and the public agenda. The fact that the individuals comprising the public agenda were registered to vote indicates that the relevance of news about the campaig n was high. In addition, the fact that these voters were still undecided at the time of the study indicates that they were also high in uncertainty With this in mind, these voters would appear to have been high in need for orientation and thereby highly susceptible to agendasetting effects. In short, the original Chapel Hill study consisted of a stratified sample of voters that were predisposed to attending to the patterns of news coverage. The vast majority of research into the contingent conditions of agendasetting can be sorted into two groups audience characteristics and media characteristics (Winter, 1981) Besides need for orientation, other audience characteristics identified in the literature as moderators of agendasetting effects include: is sue or object obtrusiveness/unobtrusiveness (Demers, Craff, Choi, & Pession, 1989; Manheim, 1986; Zucker, 1978), interpersonal communication (Atwater, Salwen, & Anderson, 1985; Lasora & Wanta, 1990; Wanta & Wu, 1992), and political involvement (McLeod, Bec ker, & Byrnes, 1974). Mediacentered contingent conditions of agendasetting effects

PAGE 66

66 include: credibility of the news sources (Rogers & Dearing, 1988; Wanta & Hu, 1994), news story placement (Behr & Iyengar, 1985; Iyengar & Kinder, 1987; McCombs, 2006), th e type of news media outlet (McCombs, 1977; McClure & Patterson, 1976; Palmgreen & Clarke, 1977; Wanta, 1997; Winter, 1981) and the amount of audience exposure to the mass media (Chiang, 1995; Ragas & Roberts, 2009; Shaw and Martin, 1992). Most of the research on contingent conditions has focused on examining the aforementioned moderators of the mediato public agendasetting relationship, rather than on identifying potential moderators of the source to media agendabuilding or sourceto public agenda -sett ing relationships. A notable exception is research by Wanta (1991) which identified presidential approval ratings as a moderator of the relationship among the communication efforts of the president and public opinion. A content analysis of the presidents weekly summary of activities/documents and Gallup Poll data from 1970 to 1988 found that popular presidents influence the perceived salience of issues on the public agenda more than unpopular presidents (Wanta, 1991). Less evident was the role of presidential approval as a moderator of the relationship among presidential communication efforts and news media coverage. The current study aims to add to the research on contingent conditions involving news media sources by looking at the role of financial performance indicators (market value, revenue, assets, and net income) as a potential moderator of the relationship among the communication efforts of incumbent candidates and media coverage during proxy contests. Several company performance indicators (revenue, assets, and net income) have been examined in at least one prior agendasetting study (Kiousis et al., 2007), but not in this context. Financial performance could moderate the relationship

PAGE 67

67 among t he public relations efforts of the firm (incumbent) and the amount of media attention it receives. Similarly, this study will also test whether the experience of the challenger candidate in a proxy fight, as measured by the number of activist campaigns a challenger has waged previously, moderates the relationship among the public relations efforts of the challenger and the amount of media attention the challenger receives. Agenda Building and Public R elations For over 30 years, research has indicated that public relations efforts play a significant role in shaping media coverage (Cutlip, 1962; Kaid, 1976; Sigal, 1973). In Beyond Agenda Setting, Gandy (1982) suggested that scholars look beyond the traditional news media to public agendasetting relationship and ask the question, who sets the media agenda? Gandys question helped open a new phase of agendasetting scholarship in which the media agenda moved from being the independent variable, the key causal factor in the shaping of public opinion, to the dependent variable, the outcome that is to be explained (McCombs, 2006). Key to exploring the news source to news media relationship in agendasetting is the concept of information subsidies, such as source provided news releases, advertising, speeches, letters, and other public relations materials, which attempt to intentionally shape the news agenda by reducing journalists costs of gathering information (Berkowitz & Adams, 1990, p. 723). Agendasetting research that focuses on answering the question of who sets the media agenda and explores the role of sour ce provided information subsidies influencing media content has come to be known as agendabuilding (Berkowitz & Adams, 1990; Turk, 1985; Turk & Franklin, 1987). It is important to note that the agendabuilding perspective is premised on a dynamic, multi -d irectional exchange

PAGE 68

68 between journalists and news sources, such as government officials, public relations professionals, lobbyists, and advertisers. News media content is the result of this symbiotic relationship between sources and journalists, not simply the result of the placement of messages by sources (Kiousis, 2004). As elaborated by Kiousis et al. (2007): The broader concept of agendabuilding views the process of salience formation as one involving reciprocal influence among multiple groups in addi tion to media and public opinion, such as policy makers, interest groups, and corporations. Through the use of information subsidies, such as news conferences, news releases, and interviews, public relations programs and campaigns can have a profound impact on shaping news content. (pp. 149150) While Gandy (1982) provided a diverse range of information subsidies as examples in Beyond Agenda Setting, most agendabuilding scholarship to date has focused on evaluating the effects of just two kinds of information subsidies (news releases and advertisements), often in the context of political campaigns. Empirical support for agendabuilding effects generated by candidatecontrolled news releases or advertising have been found in senatorial (Ghorpade, 1986), gubernatorial (Evatt & Bell, 1995; Lancendorfer & Lee, 2003; Kiousis et al., 2006), and presidential elections (Boyle, 2001; Miller, Andsager, & Reichert, 1998; Tedesco, 2001, 2005b; Roberts & McCombs, 1994), as well as in international election settings (Lopez Escobar et al., 1998; McCombs et al., 2000). Moving beyond studying news releases and ads, Wanta and colleagues identified agendabuilding effects generated by the presidents State of the Union speeches (Johnson, Wanta, & Byrd, 1995; Wanta, Stephenson, Turk, & McCombs, 1989) and the Weekly Compilation of Presidential Documents (Wanta, 1991; Wanta & Foote, 1994). The collective body of research on agendabuilding suggests that candidatecontrolled communication influence not just what topics are covered by

PAGE 69

69 the media (first level agendabuilding), but also how those topics are described by the media (secondlevel agendabuilding). In addition to the narrow focus on news releases and ads, largely absent from the agendabuilding literature is research which examines the relative influence of two different types of information subsidies controlled by the same candidate, or news source, in shaping news media coverage. A study by Kiousis, Kim, McDevitt, and Ostrowski, (2009) started to fill this void by examining the agendabuilding relationships of both candidate political advertising and news releases with news media content in nine statewide campaigns during the 2006 election season. Kiousis and colleagues (2009) found that news releases were more effective than ads at first level agendabuilding, while ads were more effective than news releases at secondlevel agendabuilding. In building upon the work of Kiousis et al. (2009), the current investigation into proxy contests will evaluate the relative influence of two different types of candidatecontrolled information subsidies (news releases and shareholder letters) for two sets of object agendas (issues and stakeholders). This study will probe for agendabuilding relationships among the candidates and media coverage for both types of information subsidies and object agendas. While the direction of influence from the news media to the public is well established, the direction of influence among news sources and the news media is less clear cut. The series of agendabuilding studies by Wanta regarding the presidential State of the Union addresses perhaps best illustrate the fluid relationship between news sources and media coverage. For example, Johnson, Wanta, Byrd, and Lee (1995) found somewhat surprisingly that Franklin Roosevelts addresses reacted to previous

PAGE 70

70 media coverage more than they influenced subsequent coverage. A related study revealed that addresses by Carter and Reagan were also influenced by the media agenda, while a Nixon address influenced the media (Wanta, Stephenson, Turk, & McCombs, 1989). A subsequent investigation by Wanta and Foote (1994), using the Weekly Compilation of Presidential Documents, found that on some issues Bush influenced the media, while on other issues the media influenced Bush (Wanta & Foote, 1994). Finally, on several issues, there was reciprocal influence detected between the two agendas (Wanta & Foote, 1994). The author is aware of only one previous study that assessed the direction of influence among sources and the news media in the setting of a corporate proxy contest. A study by Ragas et al. (2009) of the 2008 Yahoo! Carl Icahn proxy contest found that the information subsidies (news releases) of Yahoo!, the incumbent candidate, enjoyed some success in influencing media coverage, while the information subsidies of Icahn, the challenger candidate, did not. Reciprocal influence was detected among Yahoo! and media coverage at the first level (for issues), while at both the first and secondlevel (issues and candidate attributes), the media set the Icahn agenda. The current study will expand significantly upon this prior test (Ragas et al., 2009) by assessing the direction of influence among candidates and media for two sets of object agendas (issues and stakeholders) with two different types of subsidies (news releases and shareholder letters) and news media content (newspaper coverage and news wire coverage) during proxy contests. Candidate to Investor Agenda -S etting The media agenda is of particular interest to public relations practitioners since the news media serves as a potential conduit or bridge to accessing public opinion and the

PAGE 71

71 shaping of public attention and comprehension of an object be it a candidate, a corporation, an issue or a cause. In part due to the proliferation of the Internet and other new digital technologies, source controlled information subsidies have become even more broadly accessible and targetable to the public. Said another way, the high credibility media bridge to the public remains important, but is not the only way for sources, such as corporations, organizations, activist groups, and political candidates, to try to build the priorities of their publics. Sources also use these subsidies to attempt to shape public perceptions directly As asserted by Kiousis et al. (2006), the agendabuilding framework...is useful for examining the transfer of salience between organizations and several key stakeholder groups (in multiple directions), and not just media (p. 270). Empirical evidence (e.g. Golan, Kiousis, and McDaniel, 2007; Kiousis et al., 2006; Kiousis & Wu, 2008) behind this assertion has begun to accumulate. Kiousis (2006) and colleagues found solid evidence of first and secondlevel agendasetting relationships among candidate news releases and public opinion during the 2002 Florida gubernatorial campaign. In addition, research by Golan et al. (2007), also in a political campaign context, found mixed support for first and secondlevel agendasetting relationships between candidate ads and public opinion during the 2004 presidential election. Finally, in an international public relations setting, Kiousis and Wu (2008) found that public relations counsel played an important role in limiting the impact of negative news depictions on public salience and attitudes regarding foreign countries. In short, the early evidence suggests that in some circumstances there are significant positive linkages between the priorities in information subsidies and the priorities expressed by

PAGE 72

72 the public, although this linkage is generally weaker than the news mediato public relationship. The current investigation seeks to build upon this emerging body of work (e.g. Golan et al., 2007; Kiousis et al., 2006; Kiousis & Wu, 2008) by testing for agendasetting relationships among candidate information subsidies and investor opinion during proxy contests. Both types of information subsidies (news releases and shareholder letters) and object agendas (issues and stakeholders) will be used in these analyses. Consistent with the other hypotheses and research questions in this study examining agendasetting and agendabuilding relationships, the direction of influence in the candidateto investor opinion relationship will be examined for both types of subsidies and agendas In addition, the relative influence of news releases and shareholder letters in the candidateto investor opinion relationship will be evaluated for both types of agendas. Inter Candidate Agenda -S etting The new concept of inter candidate agendasetting ( Tedesco, 2005a, 2005b; Kiousis & Shields, 2008) represents a further extension of the agendasetting framework beyond its traditional mediato public relationship focus. Inter candidate agendasetting empirically evaluates the premise that political candidates not only use information subsidies to attempt to shape media coverage and public opinion, but also to set the agendas of their opponents. Said another way, in an effort to gain a strategic advantage, campaigns attempt to control the emphases of the public conversation that plays out in the dueling subsidies between candidates. As noted by Tedesco (2005b), campaigns battle each other to influence the content and tone of election coverage in terms advantageous to the campaigns policy platform (p. 188). Just as inter media

PAGE 73

73 agendasetting research has demonstrated that media outlets influence each other (Lopez Escobar et al., 1998; Reese & Danielian, 1989), and at times a particular media outlet may even set the overall media agenda (Gilberg, Eyal, M cCombs, & Nicholas, 1980), the initial inter candidate agendasetting research (e.g. Tedesco, 2005a, 2005b; Kiousis & Shields, 2008) has revealed that competing candidates influence each other during elections. Based on cross lagged correlational analyses, Tedesco found that there was generally mutual influence between the Bush and Kerry campaign agendas, as articulated in candidate news releases, during the 2004 presidential election (Tedesco, 2005b). However, during one month of the campaign, Kerry appears to have set the Bush agenda. Further evidence of inter candidate agendasetting was provided by an investigation of campaign agendas, again as expressed in news releases, during the 2004 Democratic presidential primary (Tedesco, 2005a). This study found that candidates share a converged issue agenda similar to how inter media agendasetting research has typically found that patterns of news coverage across media outlets are largely redundant (Tedesco, 2005a). No candidate during the 2004 Democratic pres idential primary consistently demonstrated agenda dominance over the other candidates, which may be because this was a primary that included multiple viable candidates (Tedesco, 2005a). Inter candidate agendasetting relationships have also been detected using political advertisements, blog posts, speeches, and issue platform statements to measure candidate agendas. Research by Golan et al. (2007) found that there was a significant first level association between the Bush and Kerry television ads during the

PAGE 74

74 2004 presidential election. At the secondlevel, Bush and Kerrys positive ads were also significantly associated (Golan et al., 2007). A separate analysis of the same campaign found that Bush and Kerrys blog posts were both highly correlated in terms o f issue salience (Sweetser, Golan, & Wanta, 2008). In addition, Bushs ads were significantly correlated with Kerrys blog posts (Sweetser et al., 2008). Cross lagged correlations between the two campaign blogs revealed that Kerrys blog posts may have exerted some influence on Bushs blog posts (Sweetser et al., 2008). An investigation of the Bush, Kerry, and Nader campaigns during the 2004 election found that first level inter candidate agendasetting relationships existed across candidate information subsidies, for example, between Bushs speeches and Kerrys news releases (Kiousis & Shields, 2008). At the secondlevel, significant substantive and affectivebased inter candidate relationships were detected for different types of information subsidies, as well as across them (Kiousis & Shields, 2008). As has been suggested, the relatively new concept of inter candidate agendasetting should be evaluated with additional competing actors and with different types of information subsidies (Kiousis & Shields, 2008; Tedesco, 2005a). All inter candidate agendasetting studies to date have been during political election campaigns. With this in mind, the current study provides a unique opportunity for theoretical development by exploring the applicability of inter -candidate agendasetting in a new setting (corporate proxy contests) and with new actors (incumbent and challenger corporate candidates). The current study will test for positive inter candidate agendasetting relationships between the proxy battle candidates for both sets of object agendas (issues and

PAGE 75

75 stakeholders) and for both types of subsidies (news releases and shareholder letters). Shareholder letters represent a subsidy type that is new to this research stream. Consequences: Attitudes and B ehavior Whi le most agendasetting research focuses on the cognitive aspects of the theory, scholars have also explored whether there may be attitudinal and behavioral consequences of the media setting the public agenda. A growing body of empirical evidence suggests t hat the agendasetting function of the press affects not only what the public thinks about (first level agendasetting) and how it thinks about a topic, issue, or candidate (secondlevel agendasetting), but, in some cases, even what the public thinks (and does) about that object Identified consequences of agenda setting include shifts in the criteria used to evaluate candidates (Iyengar & Kinder, 1987; Iyengar, 1991; Iyengar & Simon, 1993; Krosnick & Kinder, 1990) or the press (Kiousis, 2002), linkages w ith attitude strength (Kiousis, 2005; Kiousis & McCombs, 2004; Weaver 1984, 1991), shifts in public opinion (Kiousis, 2003; Page & Shapiro, 1992), purchase intention (Sutherland & Galloway, 1981), voter turnout (Kiousis & McDevitt, 2008), voting behavior (Roberts, 1994), and public policy actions on issues (Brewer & McCombs, 1996; Dearing & Rogers, 1996; Tan & Weaver, 2007). In short, scholarship into agendasetting consequences has turned Cohens classic statement (1963) on news media effects at least partially on its head. In certain situations, the media tells the public what to think about how to think about it and even what to think The most well known and well defined line of research into the consequences of agendasetting is media priming theory, which posits that the salient issues on the media agenda become a yardstick for evaluating political figures (Iyengar & Kinder, 1987). According to Iyengar and McGrady (2005), the more salient the issue, the

PAGE 76

76 greater the impact of opinions about that issue on political evaluations (p. 232). Priming provides an important rationale for why political campaigns exert so much effort in trying to raise the salience of issues in which they believe they enjoy a comparative advantage. Evidence for media priming of the publics evaluation of political figures has been found in both laboratory (Iyengar & Kinder, 1987) and real world settings (Iyengar & Simon, 1993; Krosnick & Kinder, 1990). A series of experiments conducted by Iyengar and Kinder (1987) found that when primed by television news stories on a particular issue, this issue weighed more heavily in the viewers assessment of presidential performance. Outside of the laboratory, studies using National Election Study data have found evidence of shifts in media attention priming voter evaluations of the president after the IranContra disclosure (Krosnick & Kinder, 1990) and the Persian Gulf War (Iyengar & Simon, 1993). As early as 1975, agendasetting scholars (e.g. Weaver, McCombs, & Spellman, 1975) reasoned that agendasetting might have an effect on the attitudes of the public. Moving beyond speculation to empirical investigation, there are indications that patterns of news coverage have attitudinal consequences for both political issues (Weaver, 1984, 1991) and political figures (Kiousis, 2003; Kiousis, 2005; Kiousis & McCombs, 2004). Using fall 1988 survey data, an examination of the federal budget deficit issue by Weaver (1991) found that increased issue salience was linked with stronger opinions and a greater likelihood of political participation, after a series of controls. An extensive analysis by Kiousis (2005) of news coverage and public opinion trends across five presidential elections (19801996) revealed that media salience of candidate attributes wa s strongly associated with the development of nonneutral attitudes by the public

PAGE 77

77 toward the candidates. Moving from an election context to an evaluation of a sitting president, Kiousis (2003) found that media attention to the Monica Lewinsky scandal was c orrelated with public shifts in perceived job approval and favorability of Clinton. For presidential favorability, there was a negative correlation with media attention. As coverage of the scandal increased, favorability decreased. On the other hand, a pos itive relationship approaching significance was found between coverage and job approval. Overall, a review of the research into the attitudinal consequences of agendasetting suggests that the media has a greater effect on attitude strength rather than attitude change, regarding objects in the public eye. Empirical investigations into the possible behavioral consequences of agendasetting have been relatively limited, although there have been several investigations into voter behavior and voter turnout. For example, a panel study conducted by Roberts (1992) successfully predicted voting behavior based on the level of voter concern among several salient political issues during the 1990 Texas gubernatorial election. Roberts (1992) was careful to note that research in other elections is necessary to validate her findings. More recently, a multi site, multi year panel study conducted by Kiousis and McDevitt (2008) into three 2002 gubernatorial elections and the 2004 presidential election found that agendasetti ng played a role in affecting voter turnout among youth. Besides at times influencing voting related behavior, agendasetting has also been shown to have an effect on public policy discussions and outcomes. While there often appears to be reciprocal mediato public policy influence at the national level (Tan & Weaver, 2007), at a local level, Brewer and McCombs (1996) found that an

PAGE 78

78 editorial campaign by a local newspaper appears to have affected the city budget in the year after the campaign. While agendasetting consequences havent been explored before in the context of a corporate proxy contest, a broader review of the literature yields mixed evidence of mediainduced behavioral outcomes in the corporate sphere. For example, several studies have explor ed whether the professional stock recommendations featured in prominent financial media outlets outperform the overall stock market. Sant and Zaman (1996) reviewed the performance of stocks recommended in Business Weeks Inside Wall Street column and fou nd that stocks with low securities analyst coverage mentioned positively in the column earned significant, positive abnormal returns. However, over a longer time period, the stock price return for these featured stocks was negative and significant (Sant & Zaman, 1996). An analysis by Huth and Maris (1992) of stock recommendations appearing in The Wall Street Journal s Heard on the Street column yielded similar findings. The short term stock price reaction to information in the column was positive and statistically significant, but, overall, not economically significant. Running counter to these two previous studies, Ferreira and Smith (2003) found that stocks recommendations made by guests on the Wall $treet Week with Louis Rukeyser television show generated significant positive abnormal returns over both the short and long term. What is unclear from these aforementioned studies is how much of the resulting stock price behavior is due to the stock picking prowess of the featured investment professionals and how much may be due to the prominent media attention the recommendations received. Huth and Maris (1992) perhaps sum up this conundrum

PAGE 79

79 best, noting that, apart from the value of security analysts recommendations, it is possible that investors will react to information in the column simply because it appears in print [italics added] (p. 28). Two recent studies by Pollock and colleagues (e.g. Pollock & Rindova, 2003; Pollock, Rindova, & Maggitti, 2008) more directly evaluate what effect media attention may have on stock price behavior. An analysis of U.S. initial public offerings (IPOs) by Pollock and Rindova (2003) concluded that the media influences investor choices about IPO firms. After taking into account a long list of controls, the volume of media coverage had a negative, diminishing relationship with stock underpricing and a positive, diminishing relationship with stock trading volume on the first day of trading (Pollock & Rindova, 2003). Additional analysis of IPOs by Pollock, Rindova and Maggitti (2008) again concluded that media coverage is significantly linked with stock market performance. Very few studies have attempted to evaluate whether the efforts of corporate actors to shape media coverage and investor opinion via financial advertising and public relations may also have attitudinal and behavioral consequences. At least two studies have come to conflicting conclusions on this topic. Bobinski and Ramirez (1994) found that financial advertising in U.S. financial daily The Wall Street Journal has the potential to increase stock trading volume, but only upon the initial appearance of the ad. Financial ads do not appear to affect a firms stock price, at least not in the short term. On the other hand, a review by Newman (1983) of financial communication during corporate takeover battles in the United Kingdom from 19581982 concludes that the press and financial public relations have a significant role to play in influencing investor

PAGE 80

80 opinion (p. 47). However, Newman (1983) cautions that the price and value of the bid remain the key determinants in deciding the outcome of a contested takeover. In an effort to contribute to the nascent theory building on agendabuilding and agendasetting consequences in the corporate sphere, the current study will probe for indications of a potential link age between the strength of candidate to media agendabuilding and candidate to investor agendasetting relationships in corporate proxy contests with shareholder voting behavior. More specifically this stud y will gauge whether in the majority of cases, the candidateto media agendabuilding relationship and direct candidateto investor agendasetting relationship is stronger for the candidate that win s the proxy fight than for the candidate that lose s the f ight Clearly, any linkages that are detected will not get directly at causality, but rather suggest that agendabuilding and agendasetting could be a contributing factor in a successful proxy fight outcome. This aspect of the study could help lay the groundwork for future research that concentrates exclusively on agendasetting consequences in this new setting. Agenda Setting in the Corporate S phere While most agendasetting and agendabuilding explorations into the corporate sphere and business news have taken place over the past decade, the genesis of research in this area goes back over 25 years. A study of the agendasetting consequences of financial media coverage found that the stocks of companies featured in Fortune magazine over a threeyear period outperformed the stock market during this span (Kieffer, 1983). In what would now be referred to as secondlevel agenda setting (affective attributes), the stocks receiving favorable coverage increased the most, but even stocks receiving neutral or negati ve coverage outperformed the overall stock market (Kieffer, 1983). Exploration of the agendabuilding and agendasetting process

PAGE 81

81 in the corporate realm then lay relatively dormant until research by Berger (2001) investigated attempts by the Business Roundtable, an elite business policy organization, to influence federal policy agendas. This study found that the associations information subsidies, such as lobbying and testimonies, did exert some influence on the policy agenda (Berger, 2001). A subsequent comprehensive review of the agendasetting, media framing, and issue management literature by Berger and colleagues (2002) concluded that agendasetting theory offers a potentially useful framework for assessing influence in the corporate sphere. These researchers assert that the construction or framing of issues and issue images by business political actors is a crucial area of research opportunity that has been largely neglected (Berger et al., 2002, p. 189). Berger et al. (2002) posit that corporate act ors, as active and powerful participants in the larger agendasetting environment, should receive greater scholarly attention: In this environment, economic producers and other actors attempt influence through multiple communication channels to gain the at tention and support of policy makers, media professionals and the public. Like the political science literature, however, many agendasetting studies do not examine the role and involvement of business actors, often dealing with the influences of media and public agendas on policy agendas and outcomes. (p. 173) Carroll and McCombs (2003) further explicated the concept of the corporate agenda and outlined the potential applicability of agendasetting and agendabuilding theory to the corporate sphere. They define the corporate agenda as the aggregate of discourse by a firm including attempts through public relations strategies to influence the media with a variety of information subsidies (Carroll & McCombs, 2003, p. 42). According to Carroll and McCombs (2003), the transfer of salience among sources, the media and the public is not limited to the public sphere:

PAGE 82

82 Although most of this research on the transfer of object salience from the media to the public has focused on public issues, public issues are not t he only objects that can be studied from the agendasetting perspective. During election campaigns, the candidates vying for political office are, in the abstract, a set of objects. The salience of each of these candidates frequently measured in political research as name recognition can be influenced by news coverage and political advertising. Here we extend the idea of the medias agendasetting influence to corporate actors. (p. 38) Several agendabuilding and agendasetting studies on corporate r eputation have empirically evaluated the propositions put further by Berger et al. (2002) and Carroll and McCombs (2003). For example, subsequent research by Carroll (2004) into the corporate reputations of U.S. companies revealed that increased media atte ntion to a corporation resulted in greater awareness of the firm among the public. In another test of agenda setting and corporate reputation, this time among major Dutch corporations, Meijer and Kleinnijenhuis (2006) found that the frequency of news on specific corporate issues influenced the salience of those corporate issues among the public. Finally, in a simultaneous investigation of agendabuilding and agendasetting effects among U.S. corporations, business media coverage, and public opinion of these corporations, Kiousis and colleagues (2007) identified a significant positive linkage between corporate salience in public relations messages and media coverage (agendabuilding) at both the first and secondlevel. However, Kiousis et al. (2007) did not find a link between corporate salience in media coverage and public perceptions of firm reputation (agendasetting). Two corporatefocused agendasetting studies in particular have helped lay the ground work for the current studys detailed investigation i nto agendasetting among dueling corporate actors, the financial media, and investor opinion during corporate proxy contests. The first study by Ohl et al. (1995) examined the potential agenda-

PAGE 83

83 building influence of corporate news releases on media coverage of a hostile takeover battle between two technology companies, MAI/Basic Four and Prime Computer. In this battle, Prime attempted to convince its shareholders to reject the hostile offer, while MAI/Basic tried to convince the Prime shareholders to accept it. This investigation found evidence that both companies influenced media coverage during the battle through the use of news releases (Ohl et al., 1995). While the presence of agendabuilding effects was found for both companies, Primes news releases were associated with articles that were longer and more favorable to its point of view. This latter point on the agreement in tone (affective attribute) between the Prime news releases and the ensuing media coverage of the takeover battle suggests evidence of a secondlevel agendabuilding relationship as well. This study did not attempt to link the corporate agendas or media coverage with investor opinion. The second study by Ragas and colleagues (2009) specifically evaluated the applicability of both the ag endabuilding and agendasetting framework to the highprofile 2008 corporate proxy battle between Yahoo! Inc. and activist investor Carl Icahn. At the first level, robust positive correlations were found between the issue agendas in both Yahoo! and Icahns news releases with the business news coverage of the contest. Robust positive correlations were also found between media coverage and investor opinion, as measured by discussion of the contest on the Yahoo! Finance message board. A significant positive linkage was also found between the Yahoo! issue agenda and investor opinion, but not between the Icahn issue agenda and opinion. At the secondlevel, evidence of agendabuilding and agendasetting was more mixed with only the Yahoo! framed images of Yahoo! and Icahn (as measured by a set of

PAGE 84

84 substantive candidate attributes) significantly associated with how these images were portrayed in media coverage and investor opinion. Finally, a series of cross lagged correlation analyses revealed that Yahoo!s news releases enjoyed some success in influencing media coverage, while Icahns did not. In short, when evaluated via an agendasetting and agendabuilding perspective, Yahoo!, the incumbent candidate, enjoyed more success with its communications than Icahn, the challenger. System -L evel Agenda Setting The Acapulco typology, which gained its name because it was first presented by Max McCombs at an annual meeting of the International Communication Association in Acapulco, Mexico, classifies agendasetting research approaches into four dimensions: competition, automaton, natural history, and cognitive portrait (McCombs, 1981, 2006). The initial Chapel Hill study (McCombs & Shaw, 1972) of the 1968 presidential election was an example of the competition perspective in which salience is measured at an aggregate system level for an array of issues competing for position on agendas. The automaton perspective also looks at an array of issues but shifts the analysis to a nonaggregated focus on the individual. The natural history perspective, on the other hand, analyzes at an aggregate level the shifts in salience of a single issue on agendas over an extended period of time. Winter and Eyals (1981) longitudinal study of agendasetting effects for the civil rights issue is a c lassic example of this perspective. While the systemlevel competition perspective is most often applied to an analysis of salience transfer within one event at a time, the data from timebounded related events can also be aggregated to form several new hi gher level agendas that give a true system level view across events

PAGE 85

85 For example, a recent study by Kiousis et al. (2007) adopted an aggregate systemlevel perspective for analyzing the transfer of salience among media content, public opinion (specifically corporate reputation), financial performance, and public relations materials for 28 large U.S. companies. This investigation found a significant positive relationship between the salience of companies in public relations materials, as measured by the number of news releases issued by each company, and news media content, as measured by the number of stories about each company (Kiousis et al., 2007). Specifically, significant linkages were found between the frequency of news releases issued and resulting s tories for both the New York Times and The Wall Street Journal These results confirm the basic agendabuilding hypothesis that increased object salience in public relations content stimulates increased attention to objects in news media content (Kiousis et al., 2007). This study did not find a linkage between the salience of companies in news stories with perceived corporate reputation, as measured by the rank order of these companies in the 2005 Harris Reputation Quotient survey of public opinion. Therefore, the basic agendasetting hypothesis was not supported. Interestingly, Kiousis and colleagues also found that increased media coverage in The Wall Street Journal was linked with greater company assets and weakly connected with greater profits (Kiousis et al., 2007). Neither of these relationships was found with the coverage in the New York Times (Kiousis et al, 2007). While the vast majority of hypotheses and research questions in this study will analyze the transfer of salience individually among proxy contests and then draw conclusions from the accumulated results, this study will also probe the proxy contests from a aggregate systemlevel agendasetting perspective. In this approach, the objects

PAGE 86

86 of interest switch from being the set of issues and stak eholders competing for attention across agendas in each proxy contest to rather the actual proxy contests themselves as a set of objects competing for attention across agendas within the larger corporate sphere. In other words, instead of evaluating the sy stemlevel salience of each company as was demonstrated by the Kiousis et al. (2007) study, the current study will evaluate the system level salience of each proxy contest based on the frequency of challenger and incumbent issued news releases and shareholder letters, financial news stories, and investor message board postings for each contest. More specifically, this investigation will probe for sourceto media agendabuilding linkages between the frequency of candidate issued news releases and the frequency of news stories for each media outlet, and the frequency of shareholder letters and the frequency of stories. Potential source topublic agendasetting relationships will also be assessed by comparing the frequency of shareholder letters with the frequency of message board postings and the frequency of news releases with the frequency of postings. Finally, potential mediato public agendasetting relationships will be evaluated by comparing the frequency of stories for each media outlet with the freque ncy of postings. Hypotheses and Research Questions Before proposing the specific hypotheses and research questions that will guide this investigation, Figure 24 provides a theoretical map of the overall agendabuilding and agendasetting process in the co ntext of corporate proxy contests. This map demonstrates how and where each set of hypotheses and research questions fit into the broader theoretical framework.

PAGE 87

87 Based on the logic of first and secondlevel agendabuilding and agendasetting evident in the reviewed literature, the following hypotheses and research questions are submitted: System Level Agenda Setting H1a. The salience of proxy contests in news releases will be positively related to the salience of proxy contests in media coverage. H1b. The salien ce of proxy contests in shareholder letters will be positively related to the salience of proxy contests in media coverage. H2a. The salience of proxy contests in news releases will be positively related to the salience of proxy contests in investor opinion. H2b. The salience of proxy contests in shareholder letters will be positively related to the salience of the proxy contest in investor opinion. H3. The salience of proxy contests in media coverage will be positively related to the salience of proxy contests in investor opinion. System Level Contingent Conditions RQ1a. What is the relationship between the salience of proxy contests in incumbent candidate news releases and media coverage before and after controlling for indicators of corporate financial performance? RQ1b. What is the relationship between the salience of proxy contests in incumbent candidate shareholder letters and media coverage before and after controlling for indicators of corporate financial performance? RQ2a. What is the relationship between the salience of proxy contests in challenger candidate news releases and media coverage before and after controlling for the activist campaign experience of challengers? RQ2a. What is the relationship between the salience of proxy contest s in challenger candidate shareholder letters and media coverage before and after controlling for the activist campaign experience of challengers? Candidate to Media Agenda Building H4a The salience of issues in news releases will be positively related to the salience of issues in media coverage. H4b The salience of issues in shareholder letters will be positively related to the salience of issues in media coverage.

PAGE 88

88 H5a The salience of stakeholders in news releases will be positively related to the sali ence of stakeholders in media coverage. H5b. The salience of stakeholders in shareholder letters will be positively related to the salience of stakeholders in media coverage. RQ3 Will news releases or shareholder letters be more strongly related with medi a coverage regarding issue salience? RQ4 Will news releases or shareholder letters be more strongly related with media coverage regarding stakeholder salience? RQ5a Will the salience of issues in news releases influence the salience of issues in media co verage? RQ5b Will the salience of issues in shareholder letters influence the salience of issues in media coverage? RQ6a Will the salience of stakeholders in news releases influence the salience of stakeholders in media coverage? RQ6b Will the salience of stakeholders in shareholder letters influence the salience of stakeholders in media coverage? Inter -C andidate Agenda Setting H6a. The salience of issues will be positively related in news releases among the competing candidates. H6b. The salience of is sues will be positively related in shareholder letters among the competing candidates. H7a. The salience of stakeholders will be positively related in news releases among the competing candidates. H7b. The salience of stakeholders will be positively related in shareholder letters among the competing candidates. Media to Investor (Traditional) Agenda Setting H8a. The salience of issues in media coverage will be positively related to the perceived salience of issues in investor opinion. H8b. The salience of stakeholders in media coverage will be positively related to the perceived salience of stakeholder in investor opinion. RQ7. Will the salience of issues in media coverage influence the perceived salience of issues in investor opinion?

PAGE 89

89 RQ8. Will the salie nce of stakeholders in media coverage influence the perceived salience of stakeholders in investor opinion? Candidate to Investor Agenda Setting H9 a. The salience of issues in news releases will be positively related to the perceived salience of issues in investor opinion. H9 b. The salience of issues in shareholder letters will be positively related to the perceived salience of issues in investor opinion. H10 a. The salience of stakeholders in news releases will be positively related to the perceived salience of stakeholders in investor opinion. H10 b. The salience of stakeholders in shareholder letters will be positively related to the perceived salience of stakeholders in investor opinion. RQ9. Will news releases or shareholder letters be more strongly related with investor opinion regarding issue salience? RQ10. Will news releases or shareholder letters be more strongly related with investor opinion regarding stakeholder salience? RQ11a. Will the salience of issues in news releases influence the perceived salience of issues in investor opinion? RQ11b. Will the salience of issues in shareholder letters influence the perceived salience of issues in investor opinion? RQ12a. Will the salience of stakeholders in news releases influence the perceived salience of stakeholders in investor opinion? RQ12b. Will the salience of stakeholders in shareholder letters influence the perceived salience of stakeholders in investor opinion? Consequences of Agenda Setting RQ13 Will the candidateto media agendabuilding relati onship be stronger for the candidates that win the proxy contest than for the candidates that lose the contest? RQ14 Will the candidateto investor agendasetting relationship be stronger for the candidates that win the proxy contest than for the candidat es that lose the contest?

PAGE 90

90 Figure 21. Timeline of shareholder a ctivism

PAGE 91

91 Figure 2-2 Corporate p roxy contests b y y ear

PAGE 92

92 Figure 2-3 First and secondlevel mediato public agendasetting

PAGE 93

93 Figure 2-4 Agenda building and agendasetting theoretical map during corporate proxy contests (with hypotheses and research questions)

PAGE 94

94 CHAPTER 3 METHOD Study Design Overview The pur pose of this study is to build theory that helps inform financial public relations and investor relations practice by testing for agenda building, agendasetting, and inter candidate agendasetting effects among competing corporate agendas (incumbent and challenger candidates), financial news media agendas, and investor opinion agendas in the new context of proxy contests. More specifica lly, this investigation tests for the transfer of salience among the aforementioned agendas by content analyzing candidatecontrolled news releases and shareholder letters, financial news stories, and investor message board postings to determine the salient issues and stakeholders in 25 U.S. corporate proxy contests held during the 20052009 proxy seasons. The content analyses conducted for determining the issue and stakeholder agendas in each proxy contest are linked by the use of a uniform time period for analysis (Neuendorf, 2002). Specifically, the time frame for the content analyzed in each proxy contest is the 30 day period directly before the annual meeting or special meeting date at which shareholders voted and decided the winner of the proxy contest For each contest, the content generated during the fifteen days immediately before the meeting date will fall under Time 2 and the content generated during the preceding fifteen days will fall under Time 1. The purposive sample selected for this study consists of the five largest proxy contests held each year over a fiveyear period (20052009) that were for representation or control of a companys board of directors and that went to a shareholder vote, known as going the distance (Laide & Mallea, 200 9). The corporate

PAGE 95

95 proxy contest database of FactSet SharkRepellent.net, a provider of comprehensive data on shareholder activism campaigns and corporate governance, provided the sample frame of proxy contests. A purposive sampling technique was appropriate for testing the full range of hypotheses and research questions submitted for this study, which required that the proxy contests were large enough to generate enough analyzable media content and resulting variance across the rank ordered agendas to be suf ficient for statistical analysis. Further, by only including in the sample the largest proxy contests that ultimately proceeded to a vote, rather than were settled or withdrawn, this design facilitates the testing of the research questions that focus on potential consequences of agenda setting in this case, voting behavior. Previous agendasetting research in a political election context has used study designs that probe for agendabuilding and agendasetting relationships within a series of election contests held over the same year (e.g. Kiousis et al., 2009), as well as for exploring agendasetting relationships among the same agendas over a multi year period (e.g. Kiousis, 2005). The current investigation combines aspects of these two designs, allowing for the testing of many of the hypotheses and research questions up to a maximum of twenty five times (the five largest proxy contests held each year over a continuous fiveyear period). The strength of this sampling approach and overall study design is that it provides for repeated internal replications of the same sets of hypotheses and research questions, thereby raising confidence in the findings and enhancing external validity. This design is arguably more robust than analyzing a series of contests over just a one year period or analyzing one contest per year for a multi year period.

PAGE 96

96 The candidatecontrolled news releases and shareholder letters were drawn from the U.S. S.E.C.s Electronic Data Gathering, Analysis, and Retrieval (EDGAR) system. All so licitation materials, such as news releases and shareholder letters, distributed during proxy contests are required to be filed with EDGAR. Financial news stories were drawn from the Factiva news database and Bloomberg Professional service, and investor message board postings were drawn from the message board archives on Yahoo! Finance, the largest financial news and research Web site in the U.S. (ComScore, 2008). Given the size and scope of this project, Diction 5.0, a computer assisted textual analysis (CATA) program, aided in the coding of the content for each of the content analyses. CATA has previously been used in agendasetting and agendabuilding studies that require the sorting and coding of large bodies of text using detailed coding protocols (e.g. Carroll, 2004; Dunn, 2006, 2009; Kiousis, 2005; Miller et al., 1998; Roberts & McCombs, 1994; Tedesco, 2001, 2005a, 2005b). A total of 215 candidate news releases, 94 candidate shareholder letters, 806 news stories, and 4,594 investor message board postings were collected and content analyzed across the 25 proxy contests. Specifically, the total number of issue mentions coded for across the contests was 3,787 in news releases, 1,893 in shareholder letters, 2,243 in news stories, and 1,042 in message board postings. The total number of stakeholder mentions coded was 8,292 in news releases, 3,673 in shareholder letters, 6,213 in news stories, and 1,523 in message board postings. The content analyses for each proxy contest were guided by the steps for conduct ing a content analysis outlined by Kaid and Wadsworth (1989). According to these two scholars, any application of content analytic procedures includes at least the

PAGE 97

97 following seven steps: 1) Formulate the hypotheses or research questions to be answered; 2) Select the sample to be analyzed; 3) Define the categories to be applied; 4) Outline the coding process and train the coders who will implement it; 5) Implement the coding process; 6) Determine reliability and validity; 7) Analyze the results of the coding process (Kaid & Wadsworth, 1989, p. 199). Pre Test According to Wimmer and Dominick (2006), the information collected from qualitative methods is often used to prepare a more elaborate quantitative analysis (p. 49). Following this approach, a qualitative pre test was conducted to aid in the development of the coding instrument used in the quantitative content analyses of the proxy contests. This pretest consisted of conducting a series of semi structured, in depth interviews with financial public relations professionals that are experienced in serving as advisors for both incumbent (corporations) and challenger candidates (activist shareholders) during proxy contests. The interview guide used for this pretest focused on three topics: 1) defining and describing the salient issues, 2) salient stakeholders, and 3) the most frequently used information subsidies during proxy fights (The assent script and interview guide is included as Appendix A and Appendix B, respectively). The qualitative data collected on these three topics during the interviews was then combined with information from prior studies on shareholder activism and proxy battles (e.g. Ashton Partners, 2005; Gillan & Starks, 2007; Morgan Joseph & Co. Inc., 2006; National Investor Relations Institute, 2007; Ramius LLC, 2008) to develop the coding categories that were used in the final instrument. Data gleaned during these interviews also assisted in selecting the business news media outlets that represented the media agenda during the content analyses of the proxy contests.

PAGE 98

98 Specifically, in depth interviews were conducted with 22 financial public relations professionals during February April 2009 regarding proxy contests. The initial group of interviewees was selected by contacting the top ten financial public relations firms in the U.S., as ranked by MergerMarket.com, based on the number of mergers and acquisitions (M&A) each firm had advised upon in 2008. (See Table 31 for MergerMarkets list of the top ten financial PR M&A advisors during 2008). The researcher is not aware of a list that ranks financial PR firms based on proxy contest advisory assignments, but discussions with multiple practitioners in this area suggested that MergerMarket.coms rankings would be a valid sample frame. Merger Market.com is owned by Pearson, the publisher of the Financial Times After contacting each of these financial public relations oriented firms, interviews were ultimately conducted with one or more principals at eight of the top ten ranked firms and all of the top three firms Additional interviews were conducted by snowballing from this group for the names of additional qualified practitioners. The 22 practitioners (17 male / 5 female) interviewed in total for this pretest were all in mid to senior -l evel positions in their organizations. This included holding titles such as founder, chief executive officer, president, managing director, vice president, and partner. In sum, the interviewees held several hundred years of experience working in financial public relations, investor relations corporation communication and crisis communication, and had advised the communication efforts of both incumbent and challenge candidates in hundreds of proxy battles and activist campaigns. Salient Issues While the issue platforms from one proxy contest to another may change, interviewees virtually all agreed that every salient issue during a contest will be framed

PAGE 99

99 and viewed through the lens of shareholder value creation. Said another way, when it comes to choosing one candidates nominees over another, or supporting one business strategy related proposal over another, shareholders are likely to choose the option that they think will create the most shareholder value and will be in their best economic interest Accord ing to the interviewees, in recent years, the salient issues in contests include: management and board of director changes, approving/blocking a merger or acquisition, business strategy changes, consideration of strategic alternatives, financial changes (initiating or increasing stock dividends, recapitalizations, stock buybacks, etc.), corporate governance changes, such as shareholder voting rights and the effectiveness of the board, executive compensation, and social responsibility initiatives. Another key aspect of many proxy contests, according to the interviewees, is defining and communicating the image, credibility, and track record of each candidate. This is not dissimilar to political election contests in which the candidates spend considerable time not only defining and debating the issues, but in trying to define their own image as well as the image of their rival. In a proxy contest, each side attempts to bolster their own image and thereby their issue positions, by citing previous accomplishments and experience, while also trying to negatively frame their rivals candidate image, thereby weakening their issue positions. In general, according to interviewees, the challenger candidate in a proxy contest is likely to go negative more quickly in t he hopes that the incumbent will go to the mat as well. The incumbent is generally more careful and restricted in the fight statements that it makes. Salient Stakeholders Not surprisingly, interviewees uniformly said that shareholders were always the m ost important stakeholder during a proxy contest. Interviewees generally divided

PAGE 100

100 shareholders into two groups institutional shareholders, such as mutual funds, hedge funds and pension funds, and individual investors, also known as retail investors. By and large, since institutional shareholders hold the most stock in a company, and thereby the most votes in a proxy contest, the two sides spend the most time and effort courting these institutional investors. According to the interviewees, the second most important stakeholder in a contest is often the proxy advisory firms, in particular ISS RiskMetrics and Glass Lewis. These consulting firms review and issue recommendations to institutional investors on how to vote in contests. Several interviewees said that the recommendations of these firms can influence 2025% of the total shareholder base. Prioritizing stakeholders after shareholders and the proxy advisors appears less clear cut and seems to depend on the particulars of the proxy contest. Other stakeho lders mentioned by interviewees include the financial press, stock analysts at Wall Street brokerage firms, employees, customers, suppliers, the local community and, in some cases, government regulators and politicians. On the topic of the financial press, The Wall Street Journal was unanimously named the most influential financial media outlet, followed by the New York Times and the business news wire services (Dow Jones News, Bloomberg News, Reuters and the AP). The Financial Times was also frequently mentioned. Business cable channel CNBC was acknowledged, but only for large, highprofile contests. The local hometown newspaper of the corporation embroiled in the contest was also mentioned as a media outlet that can be a factor. Business magazines were rarely mentioned. Blogs were noted as a growing source of coverage.

PAGE 101

101 While not perceived as being as important as the proxy advisory firms, interviewees suggested that Wall Street analysts were a stakeholder that each side attempts to influence since they can provide thirdparty endorsements. In addition, these analysts are often quoted by the financial press during a contest. Some interviewees stressed that the employees of the corporation embroiled in the proxy contest are an important, but at times overlooked stakeholder. Employees may hold stock in the company and it is also important that employee morale stays up during a contest. The support or lack of support of key customers and suppliers can also have an effect on a proxy contest outcome. Several interviewees also referenced situations in which lawmakers or government regulators became an influential stakeholder during a contest. Finally, labor unions were mentioned as being a key stakeholder at times, since they not only may represent a firms workforce, but may also influence an employee or retiree s stock holdings. Information S ubsidies According to the interviewees, proxy contests often consist of two parallel campaigns operating simultaneously At one level, a candidates issue platform is communica ted through mass mediated channels, while at another level the candidates message is communicated through interpersonal channels. At the mass mediated level, news releases and shareholder letters, informally known as fight letters, are most frequently used. Often these fight letters are not only distributed directly to shareholders by mail, but all or part of the letter is distributed as a news release. These candidateissued news releases are increasingly republished on financial news and information We b sites, such as Yahoo! Finance and Google Finance. As previously mentioned, all

PAGE 102

102 solicitation materials distributed during a contest, such as news releases and letters, are required to be filed with the S.E.C.s publicly accessible EDGAR database. Press br iefings and interviews with reporters are also a common occurrence during campaigns. According to the interviewees, fight Web sites are being used more often during contests. These microsites serve as an online repository for the candidates communication materials that is directly accessible by the media, shareholders, and other interested stakeholders. In isolated circumstances, the challenger candidate may release a full position paper or related indepth report explaining their issue positions to shar eholders and the media. Newspaper advertisements, which were once a staple of proxy contest communication efforts, are now in decline due to the high cost of creating and placing the ads and the rise of the Internet, which has made a candidates campaign materials much more widely accessible than in days past. While the mass mediated aspects of the campaign focus on reaching multiple stakeholders (institutional and individual shareholders, news media, analysts, employees, customers, etc.) the interpersonal aspect of the campaign generally targets only two stakeholders institutional investors and the proxy advisory firms. The interpersonal portion of the campaign consists of setting up a series of faceto face presentations, known as a road show, between the candidate and the companys major shareholders in their respective cities. The road show also often consists of meetings with the major proxy advisory firms. This approach may be supplemented by teleconferences with major shareholders. Interviewees also mentioned isolated examples, such as activist investor Bill Ackmans 2009 proxy fight with Target and Roy Disneys 2004 vote no campaign against then Walt Disney Company CEO Michael

PAGE 103

103 Eisner, in which dissidents have held their own open meetings/presentations before the corporations official annual shareholder meeting. The Ackman meeting was also available as a live webcast. Computer Assisted Textual Analysis As previously discussed, given the large quantity of content to be analyzed, it was necessary to use a computer assisted textual analysis (CATA) program to accomplish the coding for this investigation. A growing body of agendasetting scholarship (e.g. Carroll, 2004; Dunn, 2006, 2009; Kiousis, 2005; Miller et al., 1998; Roberts & McCombs, 1994; Tedesco, 2001, 2005a, 2005b) has employed CATA programs. In addition to minimizing the time needed for routine counting, CATA helps insure a greater degree of reproducibility, generates high reliability, and minimizes researcher bias (Kaid & Wadsworth, 1989). Berelson (1952) classically defined content analysis as a research technique for the objective systematic, and quantitative description [italics added] of the manifest content of communication (p. 18). Arguably, a well designed and executed large-scale CATA study may uphold Berelsons core tenets of this research method more effectively than a large study solely using human coding. On the downside, a perceived weakness of CATA is its difficulty in ascertaining context (Kaid & Wadsworth, 1989; Stevenson, 2000). With this in mind, to alleviate this limitation, agendasetting research using CATA has generally concentrated on coding for objects such as issues and candidates, which require less context than coding for substantive attributes or frames. Diction 5.0, a well established computer program for content analysis, was selected for this study. Developed over the past 30 years, Diction 5.0 is distributed by Roderick Hart, a rhetorician, professor and dean of the College of Communication at the

PAGE 104

104 Universit y of Texas at Austin, and Craig Carroll, an assistant professor of public relations in the School of Journalism and Mass Communication at the University of North Carolina at Chapel Hill (Neuendorf, 2002; DigiText, Inc., 2008). Diction has been used in over a hundred scholarly publications within the social sciences (see http://www.dictionsoftware.com for a complete list). Diction deploys approximately 10,000 search words in 33 lists, called dictionaries, to search a text for five standardized master variables: Certainty, Activity, Optimism, Realism, and Commonality. According to Hart (2000b), based on an analysis of some 20,000 verbal texts, virtually no stat istical relationships exists among the five variables, which means that each cluster sheds new and different light on the passage being examined (p. 45). In addition to offering these standard dictionaries Diction provides researchers with the ability t o create and deploy custom dictionaries based on the researchers own keyword lists. Diction can also generate a list of the high frequency words found in a text, which is helpful in getting the general flavor of a text, as well as in refining custom dic tionaries (For more background on Diction, see Hart, 2000b). The custom dictionary capability in Diction was extensively used in the content analyses conducted for assessing the transfer of salience among the various agendas during the corporate proxy contests. More specifically, custom dictionaries were developed and deployed to evaluate the presence of two sets of first level objects (issues and stakeholders) in the content analyzed (candidateissued news releases and shareholder letters, financial news media stories, and investor message board postings) to determine the agendas during each of the proxy contests.

PAGE 105

105 Selection of Sample The corporate proxy contests selected for this study were drawn from a proprietary database of FactSet SharkRepellent.net, a provider of comprehensive data on U.S. shareholder activism and corporate governance. With annual revenue of over $600 million, FactSet Research Systems, Inc. is one of the largest providers of financial data and research solutions to investment professionals (FactSet, 2009). As previously referenced, the purposive sample selected for this study consisted of the five largest U.S. proxy contests held each year over a five year period (20052009) that were for representation or control of a companys board of directors, and that went to a shareholder vote, known as going the distance. According to Wimmer and Dominick (2006), nonprobability sampling, such as purposive samples, are frequently used in mass media research. A purposive sample includes subject s or elements selected for specific characteristics or qualities and eliminates those who fail to meet these criteria (Wimmer & Dominick, 2006, p. 92). A purposive sample was necessary for testing the full range of hypotheses and research questions submitted for this study, which required that the contests were large enough to generate enough analyzable media content and resulting variance across the rank ordered agendas to be sufficient for statistical analysis. Using a probability sampling technique would have significantly restricted the opportunities for analysis. The following criteria were used to select the proxy contests which comprise the studys sample. An initial search of the FactSet SharkRepellent.net database was conducted by year for a complete list of all U.S. proxy contests held for board representation or control. Other forms of shareholder activism, such as submitting stockholder proposals, were excluded. The criterion for determining the size of the

PAGE 106

106 proxy contest was based on the total market capitalization of the corporation that was the target of the proxy contest at the time of the announcement of the campaign, as listed in the FactSet SharkRepellent.net database. The criterion for determining that the contest went the distance was that the database showed the contest as having been resolved through a shareholder vote, rather than being withdrawn or settled. The corporation (incumbent) that was the target of the proxy contest instituted by the activist shareholder was required to be listed on a national stock exchange, such as the NASDAQ or New York Stock Exchange, thereby ensuring that the proxy contest solicitation materials (news releases, shareholder letters, proxy statements, and cards, etc.) would be on file with the S.E.C. (Listed companies are required to adhere to S.E.C. guidelines and keep their filings current). Unless the activist shareholder (challenger) was another listed corporation, the activist was required to own at least a 0.5% economic interest in the target corporation at the time of the campaign. This latter criterion excluded from the sample any contest attempts for board seats by a fringe shareholder that was unlikely to generate meaningful corporate, media, or investor attention. Ownership and stock exchange data for each contest was provided by the FactSet SharkRepellent.net database. Based on this criteria, the final sample for this investigation was comprised of the following 25 corporate proxy contests held between 20052009: for 2009, Target Corporation / Pershing Square Capital Management, NRG Energy Inc. / Exelon Corporation, Myers Industries Inc. / GAMCO Investors, PHH Corporation / Pennant Capital Management; for 2008, Conseco Inc. / Keith Long, Biogen Idec Inc. / Carl C. Icahn, CSX Corporation / TCI Fund Management, Micrel, Incorporated / Obrem Capital

PAGE 107

107 Management, International Rectifier / Vishay Intertechnology, Inc., Hexcel Corporation / OSS Capital Management; for 2007, Arrow International, Inc. / McNeil Trust, Atmel Corporation / George Perlegos, H&R Block, Inc. / Breeden Capital Management, Motorola, Inc. / Carl C. Icahn, Openwave Systems / Harbinger Capital; for 2006, Career Education Corporation / Steve Bostic, UbiquiTel Inc. / Deephaven Capital Management, Massey Energy / Third Point LLC, Motient Corporation / Highland Capital Management, GenCorp Inc. / Pirate Capital LLC; for 2006, Blockbuster Inc. / Carl C. Icahn, Exar Corporation / GWA Investments, Six Flags, Inc. / Daniel Snyder, BKF Capital Group, Inc. / Steel Partners II, and Computer Horizons Corp. / Crescendo Partners. Table 32 provides additional descriptive data on the proxy contests that served as the sample for this study. Using Target Corporation / Pershing Square Capital Management, the first proxy contest listed on the table as the example, this table shows that Target Corporation is the incumbent (corporation) in this proxy contest, while Pershing Square Capital Management is the challenger (activist shareholder) in this contest. The year listed of 2009 is the year in which this proxy contest was held. The meeting date is the date of the annual meeting or special meeting in which Target shareholders decided the outcome of the contest by voting on the candidates for the board of directors. As previously explained, the uniform time peri od of analysis for each contest contained in the sample was the thirty day period directly before the date each meeting was held. For Target Corporation / Pershing Square Capital Management, the meeting was held on May 28, 2009. Time 1 of April 21, 2009 May 12, 2009 represents the 15day period beginning a month before the meeting date at which shareholders voted, while Time 2 of May 13, 2009 May 27, 2009 represents the 15day period

PAGE 108

108 directly before the date of the shareholder vote. All data on this table was provided by the FactSet SharkRepellent.net database, except for the Time 1 and Time 2 information for each contest, which was calculated by the researcher. Previous empirical research (Winter & Eyal, 1981) has established four weeks or more as the optimum time span for examining agendasetting effects. With this in mind, the researcher initially set the time period for analysis in each contest as the twomonth period directly before the shareholder vote. This provided for adjacent Time 1 and Time 2 periods of 30 days each. A preliminary analysis revealed that these onemonth increments were not suitable for examining the direction of influence among the agendas. The vast majority of media content during a proxy contest is generated in the final month before the vote. Therefore, based on this preliminary analysis, the uniform time period of analysis chosen for the contests was the 30day period immediately before the meeting date with Time 1 and Time 2 set as adjacent 15day periods. Several prior studies (Lopez Escobar et al., 1998; Lee, Lancendorfer, & Lee, 2005; Roberts & McCombs, 1994; Roberts et al., 2002) have used adjacent time periods of less than a month apiece for examining agendabuilding and agendasetting influence using crosslags. McComb s and Roberts (1994) argue that the use of adjacent time periods of less than a month works against finding significant agendasetting effects and is therefore a more rigid theoretical test. News Releases and Shareholder Letters News releases and shareholder letters were the two types of information subsidies that were most frequently mentioned by financial public relations professionals during the pretest interviews regarding communication during proxy battles. Previous agendabuilding research has demonstrated the validity of using news releases as a source for

PAGE 109

109 identifying issue emphases and agendas in a political election context (Kiousis et al., 2006; Miller et al., 1998; Tedesco, 2001, 2005a, 2005b; Tedesco & McKinnon, 1998). The one prior agendabuild ing study conducted in a proxy contest context (Ragas et al., 2009) used news releases for determining the incumbent and challenger candidate agendas. By also incorporating shareholder letters into the current analysis, this investigation answers the call by Kiousis and colleagues (Kiousis et al., 2009; Kiousis & Shields, 2008) for more research that examines the relative influence of multiple types of information subsidies (not just news releases) on building media coverage and public opinion. Finally, the use of multiple information subsidy types enhances reliability (Chaffee, 1991). To construct the candidate agendas, the candidatecontrolled news releases and shareholder letters for each proxy contest were drawn from the S.E.C.s Electronic Data Gatheri ng, Analysis, and Retrieval (EDGAR) system. All solicitation materials and material information disclosed during proxy contests by either side, including news releases and shareholder letters, are required to be filed with EDGAR ( http://www.sec.gov/edgar.shtml ). According to the S.E.C., all companies, foreign and domestic, are required to file registration statements, periodic reports, and other forms electronically through EDGAR (U.S. Securities and Exchange C ommission, 2009, para. 1). For each proxy contest in the sample, a search was conducted of the EDGAR database using the name of the corporation (the incumbent) and the date of the annual meeting and proxy contest vote (see Table 32 for the meeting dates of each proxy battle in the sample). The S.E.C. Form DFAN14A and DEF14A filings provided by these search parameters that fell within the 30day period directly before the meeting date

PAGE 110

110 were then carefully reviewed. The news releases and shareholder letters found through this search were downloaded and assigned into one of four sets of source, date, and message typedenoted text files: 1) incumbent issued news releases, 2) incumbent issued shareholder letters, 3) challenger issued news releases or 4) challenger -issued shareholder letters. For each proxy contest, these sets of collected news releases and shareholder letters comprised the body of content for determining the issue and stakeholder agendas by time period (Time 1 and Time 2) for the information subsidies (news releases and shareholder letters) controlled by each candidate (incumbent and challenger). Each release and letter served as the unit of analysis. Using the Target Corporation / Pershing Square Capital Management proxy contest as an example, a search using the term Target Corporation was conducted from the EDGAR Company Search feature. This search provided a page with a complete reverse chronological list of all S.E.C. filings regarding Target. To limit the filing results to the time period relevant to the study, the date of the 2009 Target annual meeting (2009/05/28) was entered into the Prior To search box on the Target S.E.C. filings page. This search parameter resulted in a list of all Target filings made prior to May 28, 2009. The Form DFAN14A and DEF14A filings that fell within the 30day time period directly before the annual meeting (April 28 May 27, 2009) were carefully reviewed, and the news releases and shareholder letters found by this search were downloaded and assigned into t he relevant sets of text files for analysis. All solicitation materials distributed by challengers during proxy battles are required to be filed with the S.E.C. as Form DFAN14A filings, while all materials distributed by incumbents are required to be filed as Form DEF14A filings. Following this process resulted in a total of

PAGE 111

111 12 Target issued news releases, 12 Pershing Square Capital issued news releases, 2 Target issued shareholder letters, and 2 Pershing Square Capital issued shareholder letters. This same procedure was followed for the other 24 proxy contests in the sample. See Table 33 and Table 34 for a list of the news release and shareholder letters, respectively, issued in each contest by candidate and time period. The Wall Street Journal New York Times Financial Times and the major news wire services were the media outlets that were perceived by the pretest interviewees as being the most influential in shaping investor opinion during proxy contests. The Wall Street Journal New York Times and the wire services are all recognized as elite media sources in the agendasetting literature and are frequently selected to represent the media agenda (Kiousis, 2004b; Reese & Danielian, 1994; Rogers & Chang, 1991; Singer & Ludwig, 1987; Winter & Eyal, 198 1). The Financial Times has been acknowledged as a leading source of business journalism (McCombs, 2006) and was included in the one prior study of agendabuilding and agendasetting during a proxy battle (Ragas et al., 2009). Bloomberg News has not been included in prior agendabuilding and agendasetting research, but was identified as an influential business news source by the pretest interviewees. With over 2,300 professionals in 140 news bureaus around the world, Bloomberg is one of the worlds largest providers of business journalism (Bloomberg L.P., 2009). Bloomberg News is available through the Bloomberg Professional service and its stories are syndicated in over 450 newspapers worldwide (Bloomberg L.P., 2009). Taking this all into account, the following seven media outlets, consisting of three national newspapers and four news wire services, were selected for this investigation:

PAGE 112

112 The Wall Street Journal New York Times Financial Times (U.S. edition), Dow Jones News, Bloomberg News, the Associated Press, and Reuters News. The sample selected for the media agenda helps provide for a broad range of viewpoints from news gatherers on the proxy contests. Further, the use of multiple indicators for the media agenda enhances reliability (Chaffee, 1991). To construct the media agendas, searches of the Factiva database of news content was conducted using each of the media outlets as sources for the 25 proxy contests. Six of the seven media outlets selected to represent the media agenda are available through Factiva. Bloomberg News is not available through Factiva or Lexis Nexis and is only available in an abbreviated form via Bloomberg.com. A complete archive of Bloomberg News content is only available through the Bloomberg Professional service. Therefore, for each contest, searches for Bloomberg News stories were conducted through the advanced search feature of Bloomberg Professional. Following the approach of previous agendasetting investigations in the corporate sphere (e.g. Kiousis et al., 2007; Ragas et al., 2009), for each proxy contest, a search string with the names of the incumbent and the challenger candidate was used to search the headline and lead of all news stories appearing in these sources over the time period of the contest (the two month period directly before the annual meeting). These search criteria helped ensure that only news stories that were substantively concerned with the proxy contest would be included in the sample. Consistent with prior research (e.g. Kiousis et al., 2007; Ragas et al., 2009), only hardnews stories were included in the sample. Therefore, soft stories, such as letters to the editor and editorials, were excluded. News stories that were simply daily stock market recaps,

PAGE 113

113 highlights or summaries, which only briefly mentioned the proxy contest in passing, were also excluded. For each contest, news stories meeting these criteria were each downloaded from Factiva or Bloomberg Professional and assigned into one of seven sets of source, message type, and datedenoted text f iles for analysis. For each contest, these sets of news stories comprised the body of content for determining the issue and stakeholder media agendas in aggregate, as well as the individual agendas of the news media outlets, by time period (Time 1 and Time 2). Consistent with the releases and letters, each news story served as the unit of analysis. Using the Target Corporation / Pershing Square Capital Management proxy contest as an example, the search string Target AND Pershing Square was used to searc h the headline, lead paragraph, and first 100 words of news stories published by the seven media outlets during the contest time period (April 28 May 12, 2009). Soft news stories, recaps, and summaries were excluded from the sample. Following this proces s resulted in a total of four stories in The Wall Street Journal two stories in the New York Times five stories in the Financial Times 13 stories by Reuters, 21 stories by AP, 18 stories by Dow Jones, and 31 stories by Bloomberg about this contest. This same procedure was followed for the other 24 contests in the sample. See Table 35 for a complete list of the news story frequencies by media outlet and time for the other contests. Investor Opinion In the absence of public opinion polling date, agenda-se tting researchers have used the public discussion on electronic bulletin boards, also known as message boards, as a surrogate for the public agenda (Lee et al., 2005; Ragas et al., 2009; Roberts et al., 2002). Media framing studies have also used message boards as a

PAGE 114

114 source for determining public opinion on a topic (Constantinescu & Tedesco, 2007; Zhou & Moy, 2007). The stock discussion message boards on the Yahoo! Finance Web site were selected as the source for ascertaining the investor opinion agenda duri ng each proxy contest. With more than 18 million visitors each month, Yahoo! Finance was chosen because it ranks as the largest financial news and research site in the U.S. (ComScore, 2008). Yahoo! Finance provides message boards for the discussion of most publicly held U.S. firms. The site includes searchable archived message board posts dating back to 1999. To construct the investor opinion agendas, searches of the Yahoo! Finance stock message boards were conducted for each of the 25 proxy contests. Specifically, the name of the corporation (incumbent candidate) that was the target of the proxy contest was entered into the Get Quotes search box on the home page of Yahoo! Finance ( http://finance.yahoo.com/ ). From the resulting company page, the Message Board link for the company was accessed. The Advanced Search feature on the companys stock message board page was then used to search the message board for all postings that were made during the time period of the proxy contest (the onemonth period directl y before the date of the meeting at which shareholders voted). For each contest, the archived stock message board postings meeting these criteria were downloaded from the Yahoo! Finance Web site and assigned to source, message type and datedenoted text f iles for analysis. The sets of collected message board postings comprised the body of content for determining the issue and stakeholder agendas for investor opinion by time period (Time 1 and Time 2) for each contest.

PAGE 115

115 Consistent with the news releases, shareholder letters and news stories, each message board posting served as the unit of analysis. Using the Target Corporation / Pershing Square Capital Management proxy contest as an example, from the Yahoo! Finance home page, the name Target was entered into the Get Quotes search box. From the resulting Target Corporation company page on the Yahoo! Finance Web site, the Message Board link was accessed. A search string with a Start Date of April, 28, 2009 and an End Date of May 27, 2009 was then entered into the Advanced Search feature on the Target Corporation stock message board page. Following this process resulted in a total of 214 postings, which were each manually downloaded and prepared into text files for analysis. This same procedure was followed for the other 24 proxy contests in the sample. See Table 36 for a complete list of the message board posting frequencies for the other contests. Financial Performance and Experience Indicators The data on the financial performance of the target corporations (incumbents) and the activist campaign experience of the activist shareholders (challengers) was obtained from the S.E.C.s EDGAR system and the FactSet SharkRepellent.net database, respectively. Every year, publicly held corporations are required to file with the S.E.C. an annual report, known as a Form 10K filing, which provides a detailed review of the firms business and financial performance. For each incumbent in the sample, the market value, revenue, assets, and net income data was collected from the companys 10K report filed the reporting year before the date of the meeting (either an annual meeting or special meeting of shareholders) which decided the proxy contest in question. These and related financial indicators have been used in previous me dia effects studies in a corporate context (e.g. Kiousis et al., 2007; Pollock & Rindova,

PAGE 116

116 2003; Pollock et al., 2007). The introductory section to every 10K report includes the aggregate market value of the voting stock of the corporation. Within the 10-K filing, the revenue and net income data for the corporation is included in the Income Statement and the asset data is found in the Balance Sheet. For example, for the Target Corporation / Pershing Square Capital Management proxy contest, the meeting tied to the shareholder vote over the proxy battle was held on May 28, 2009. Using Target Corporation as the company name, a query of the EDGAR system was conducted. From the resulting Target Corporation page of S.E.C. filings, a second query was conducted limiting the results to only Form Type 10K. This search revealed that the annual report filed by Target Corporation that was closest to the meeting date in question was its 10K filing on March 13, 2009. A search of the relevant financial statements in th is filing revealed that the market value, assets, revenue, and net income for Target Corporation during reporting year 2008 were $33.6 billion, $17.5 billion, $65 billion, and $2.2 billion, respectively. This same procedure was followed for the other 24 proxy contests in the sample. See Table 37 for a complete list of the financial performance indicators of the incumbent firms in the other contests. These indicators will be used in the system level contingent condition research questions (RQ1a and RQ1b), which inquired about the relationship between the salience of proxy contests in incumbent information subsidies (news releases and shareholder letters) and media coverage before and after controlling for indicators of corporate financial performance.

PAGE 117

117 The SharkWatch feature within the FactSet SharkRepellent.net database was used to obtain the data on the activist campaign experience of the challenger candidate in each proxy contest. This feature provides searchable profiles of activist shareholders. Contain ed within this profile is the number of total campaigns the activist shareholder has waged. For example, for the Target Corporation / Pershing Square Capital Management contest, a search of this feature Table within the FactSet SharkRepellent.net database for the profile of Pershing Square Capital Management revealed that this activist shareholder had waged 38 campaigns. This same procedure was followed for the other 24 proxy contests in the sample. See Table 38 for a list of the activist campaign experience indicators of the challenger candidates in the other contests. This indicator will be used in the system level contingent condition research questions (RQ2a and RQ2b), which inquired about the relationship between the salience of proxy contests in challenger information subsidies (news releases and shareholder letters) and media coverage before and after controlling for the activist campaign experience of challengers. Coding Categor ies for Content The sets of content (news stories, news releases, shareholder letters, and message board postings) collected for each proxy contest were analyzed and coded using Diction 5.0, a computer assisted textual analysis (CATA) program. The content was coded for the frequency of issue and stakeholder mentions using the category operationalizations described in detail in the next section (The code sheet and code book is included as Appendix C and Appendix D, respectively). The frequency output files generated by the processing of the text files through Diction were then used to determine the ranks of the objects (issues or stakeholders) which comprised the various

PAGE 118

118 agendas in each contest. Consistent with prior agendasetting research, an agenda is defined as a set of objects or attributes that are communicated in a hierarchy of importance at a point in time (Dearing & Rogers, 1996). The hierarchy or rank order of the objects or attributes on the agenda was determined by the salience of each object or attribute. In keeping with prior research, salience was measured by the frequency of object and attribute mentions in the content analyzed (e.g. Kiousis, 2005; Miller et al., 1998, Tedesco, 2001). This study followed the process for developing and implementing a CATA coding scheme used by Tedesco (2001, 2005a, 2005b) in previous agenda setting and agendabuilding investigations. The first step in this process was to develop the coding categories such as a list of issues, based on theoretical considerations and prior research. The next step was to carefully identify and create unique lists of keywords that are used to represent and measure mentions of these categories in the content analyzed. Particular care was taken to only select keywords which were likely to appear in the proper context. These categories and word lists were added to Diction as custom dictionaries. After the text files were cleaned and prepared for processing, the files were then run through the word frequency count feature of the program. By comparing the raw frequency counts with the custom dictionary categories and word lists, this feature helped identify any additional categories or keywords that should be incorporated into the coding scheme. As recommended by Tedesco (2001), the researcher reviewed those terms that appeared more than ten times in a text. After the final coding scheme had been decided upon, the custom dictionaries were then uploaded to Diction and the text files were processed by the program. Finally, the resulting frequency outputs from

PAGE 119

119 the custom dictionaries and related data for each proxy contest were imported into the SPSS statistical software package for analysis. Before developing the coding scheme for the current study, the researcher contacted Tedesco (personal communication, May 19, 2009), who graciously shared in detail the steps he has used for successfully conducting several largescale agendasetting studies using a CATA program. As part of this learning process, the researcher requested and carefully reviewed the custom dictionary word list files used by Tedesco in several pri or CATA based agendasetting studies (e.g. Tedesco, 2001, 2005a, 2005b). The researcher also carefully reviewed a category and word list file provided by Hansen (personal communication, May 15, 2009) that was used in a large scale CATA based study on issue ownership theory (Petrocik, Benoit, & Hansen, 2004). Following the guidelines established by Tedesco (2001, 2005a, 2005b) and Petrocik et al. (2004), the initial list of categories and keywords were devised based upon a combination of the issues and stakeholders most frequently referenced in the pre test interviews, as well as a review of the previous research in this area (e.g. Ashton Partners, 2005; Gillan & Starks, 2007; Morgan Joseph & Co. Inc., 2006; National Investor Relations Institute, 2007; Ramius LLC, 2008). As in a previous CATA based agendasetting study (Kiousis, 2005), a conventional thesaurus was used to aid in the construction of the word lists. These categories and word lists were then refined after reviewing raw word frequency count lists generated from a preliminary analysis by the CATA program of the text files for each proxy contest in the sample. The final custom dictionaries were then uploaded to Diction and the text files were processed with the

PAGE 120

120 resulting frequency output files imported into the SPSS software package for statistical analysis. The following lists show the operationalizations of each issue and stakeholder category through the use of keyword lists. Specifically, the following issue categories (with the specific search keywords for each in parenthesis) were used to identify the issue agendas in the contests: Corporate Performance (underperform, underperformance, underperforming, underperformed, perform, performance, performed, performing, outperformance, outperform, outperformed, outperforming) Acquisition (acquisition, acquisitions, acquire, acquires, acquiring, acquired, acquirer, acquirers, acquirers, takeover, takeovers, take over, take overs, bid, bids, bidding, bidder, bidders, bidders, buyout, buyouts, buy out, bu youts) Compensation (compensation, compensate, compensating, compensated, pay, pays, paying, paid, payment, payments, say onpay, bonus, bonuses, salary, salaries, incentive, incentives, incentivize, incentivized, perk, perks) Transaction (transact, transaction, transactions, transactions, deal, deals, deals, agreement, agreements, agreements, arrangement, arrangements, arrangements) Experience (experience, inexperience, experienced, inexperienced, expertise, skill, skills, skilled, qualified, highly -q ualified, unqualified, qualification, qualifications, credential, credentials, credentialed, knowledge, knowledgeable, proficient, proficiency) Strategy (strategy, strategies, strategic, strategized, strategically) Merger (merge, merges, merger, mergers, mergers, merging, merged) Shareholder Value (value, valuing, valued, valuable) Stock Dividend (dividend, dividends) Joint Venture (JV, JVs, venture, alliance, alliances) Partnership (partnership, partnerships, partnering, partnered, partner, partners) Corporate Governance (governance, governing, governed, oversight, accountable, accountability, responsible, responsibility)

PAGE 121

121 Divestiture (divestiture, divestitures, divest, divesting, divested, spinoff, spinoffs, spin off, spin offs, spunoff, spun off, spli toff, splitoffs, split, splitting) Stock Repurchase (repurchase, repurchase, repurchases, repurchasing, repurchased, buyback, buybacks, buy back, buy -backs) Independent (independent, independence, autonomous, autonomy) Social Responsibility (CSR, social, socially, society, societal) Litigation (litigation, litigations, litigate, litigating, litigated, litigates, lawsuit, lawsuits, lawsuits, lawsuit, law suits, suit, suits, sue, suing, sued) The following stakeholder categories (with the specific searc h keywords for each category in parenthesis) were used to identify the various stakeholder agendas in the proxy contests: Management (C.E.O., CEO, management, managements, manage, manager, managers, managing, mismanaging, managed, mismanaged, mismanagement, managerial, leadership, leaderships, executive, executives, executives, executives) Regulator (regulator, regulators, regulators, regulatory, regulation, regulations) Board of Directors (BOD, B.O.D., director, directors, directors, directors, board, boards, boards, nominee, nominees, nominees, nominees, candidate, candidates, candidates, candidates) Shareholder (shareholder, shareholders, shareholders, shareholders, stockholder, stockholders, stockholders, stockholders, investor, investor s, investors, investors, owner, owners, owners, owners) Stakeholder (stakeholder, stakeholders, stakeholders, stakeholders) Analyst (analyst, analysts, analysts, analysts, researcher, researchers, researchers, researchers, forecaster, forecasters forecasters, forecasters) Union (union, unions, unions) Supplier (supplier, suppliers, suppliers vendor, vendors, vendors) Employee (employee, employees, employees, employees, worker, workers, workers, workers, workforce, workforces, staff, sta ffs)

PAGE 122

122 Community (community, communitys, communities, resident, residents, residents, citizen, citizens, citizens, neighbor, neighbors, neighbors) Retiree (retiree, retirees, retirees, retirees) Government (government, governmental, governments, gover nments, lawmaker, lawmakers, lawmakers, legislator, legislators, legislators, legislators, policymaker, policymakers, policymakers, policymakers) Customer (customer, customers, customers, customers, consumer, consumers, consumers, consumers, shop per, shoppers, shoppers, shoppers, guest, guests, guests, guests, client, clients, clients, clients, patient, patients, patients, patients, patron, patrons, patrons, patrons) Proxy Advisor (RiskMetrics, RiskMetrics, RiskMetricss, ISS, ISSs, Eg anJones, Egan Jones, Egan-Joness) News Media (press, news, media, medias, journalist, journalists, journalists, reporter, reporters, writer, writers, writers, columnist, columnists, commentator, commentators, commentators) Court (court, courts, courts, judge, judges, judges, magistrate, magistrates, magistrates) Intercoder Reliability Intercoder reliability tests have been conducted in several previous CATA based agendasetting studies to assess the reliability of the sample selection process or the coding protocol (Kiousis, 2005; Kiousis & McCombs, 2004; Roberts & McCombs, 1994). In the case of the current study, human judgment was involved in the decision to include or omit news stories and information subsidies (news releases and letters) from each proxy contest sample. As outlined previously, only hardnews stories that were substantively concerned with each proxy contest were to be included in the sample. Soft stories, such as letters to the editor and editorials, as well as daily stock marke t recaps, highlights or summaries, which only mentioned the proxy contest in passing, were to be excluded. Turning to the information subsidies, only candidate news releases and shareholder letters were to be included in each proxy contest sample.

PAGE 123

123 Other solicitation materials, such as investor presentations and conference call transcripts, memos to employees, advertisements, proxy statements, and position papers, were to be excluded. To assess the reliability of the sample selection process, three proxy co ntests (representing 12% of the 25 proxy contests examined in the study) were randomly selected and doublecoded to establish the extent of agreement. For each of these three proxy battles (Career Education Corporation / Steve Bostic, Six Flags, Inc. / Dan iel Snyder, and Computer Horizons Corp. / Crescendo Partners), a second independent, trained coder accessed Factiva and Bloomberg News (news stories), and S.E.C. EDGAR databases (news releases and shareholder letters) following the search criteria and sample selection protocol. Intercoder reliability for the stories in the Career Education Corporation / Steve Bostic contest was .99 using Holtsis (1969) coefficient of reliability and .88 based on Scotts pi (1955), which corrects for chance agreement. For t he Six Flags, Inc. / Daniel Snyder contest, the Holstis value was .99 and Scotts pi was 0.97. For the Computer Horizons Corp. / Crescendo Partners contest, they were 1.00 and 1.00. Turning to the information subsidies, for the Career Education Corporation / Steve Bostic contest, intercoder reliability according to Holtis formula was .94 and for Scotts pi was .89. For the subsidies in the Six Flags, Inc. / Daniel Snyder contest, the figures were .91 and .86. Finally, for the Computer Horizons Corp. / Crescendo Partners contest, they were .98 and .97. All of these values are above generally accepted levels of intercoder reliability (Neuendorf, 2002; Wimmer & Dominick, 2006). An intercoder

PAGE 124

124 reliability check was not performed on the message board data since all posts made during each sample time period was included in the final sample. Data Analysis Strategy Several statistical tests were used to analyze the data and evaluate the hypotheses and research questions that guided this investigation into mass commu nication during proxy contests. Specifically, the following statistics were used: Spearmans rho rank order correlations, cross lagged correlations along with the Rozelle Campbell baseline statistic, and partial correlations. Internal replications for each hypothesis and research question were conducted for each proxy contest in which there was sufficient data to provide the variance necessary for statistical testing. As a rule, agendas consisting of mentions in less than four issue or stakeholder categories were excluded from testing throughout this study. This approach to data analysis is consistent with prior agendabuilding research that has used a multi case design (Kiousis et al., 2009). Spearmans rho rank order correlations (e.g. McCombs & Bell, 1996; McCombs & Shaw, 1972), the nonparametric alternative to Pearsons product moment correlations, was the statistical test used for comparing the competing candidate, news media, and investor opinion agendas in each contest regarding issue and stakeholder salience (Hypotheses: H18, H10 12 and Research Questions: RQ34, RQ9 10). Nonparametric statistics, such as Spearmans rho, are recommended when the sample size is relatively small, such as in this investigation (e.g. McCall, 1994; Weaver, 1981). While correlations can establish the presence and strength of linkages among the various agendas, significant correlations do not alone prove causality.

PAGE 125

125 With this in mind, cross lagged correlations were used to evaluate the research questions which asked about the direction of influence among the first level agendas (Research Questions: RQ58, RQ11 12). Cross lagged correlations are an established technique within the agendasetting literature for exploring causal relationships using time ordered correlational data (Dunn, 2009; Kiousis, Mitrook, Popescu, Shields, & Seltzer, 2006; Lee et al., 2005; Lopez Escobar et al., 1998; Ragas et al., 2009; Roberts & McCombs, 1994; Sweetser et al., 2008; Tedesco, 2001, 2005a, 2005b). A cross lagged correlation allows for the examination of two competing propositions simultaneously. Cross lagged analysis assumes that if X causes Y more than Y causes X, and then P X1Y2 should be higher than P X2Y1 (Campbell & Kenny, 1999). The Rozelle Campbell (1969) baseline statistic, which is computed from the other four correlations present in two variable cross lags, was used to determine the significance of the cross lagged correlation results. The formula for computing the Rozelle Campbell baseline is: [( P X1Y1 + PX2Y2)/2]{[ P X1X2)2 + ( P Y1Y2)2]/2}1/2 The baseline represents the level of correlation to be expected on the basis of the autocorrelations and synchronous correlations alone (Lopez Escobar et al., 1998). In essence, this statistic provides a baseline value that represents the cross la gged correlation that would indicate no influence among the two variables of interest (Lopez Escobar et al., 1998). Figure 31 provides a graphical representation of the values in a two variable cross lag that were used to compute the RozelleCampbell baseline. When using two variable cross lagged correlations and the RozelleCampbell baseline, several conditions must be met to suggest that one agenda unidirectionally

PAGE 126

126 influenced the other. Using Figure 31 as the example, if the correlation between Variabl e One at Time One and Variable Two at Time Two ( P X1Y2) was higher than the correlation for Variable Two at Time One and Variable One at Time Two ( P Y1X2), this alone would not necessarily suggest that Variable One influenced Variable Two. An additional condition would still need to be met. The P X1Y2 correlation would need to exceed the Rozelle Campbell baseline value, while the P Y1X2 correlation would need to fall below the baseline, which suggests no influence. Only once these conditions have been met is there sufficient evidence to suggest that one agenda has influenced the priorities of another agenda, or in the parlance of agendabuilding and agendasetting theory, Variable One set the agenda of Variable Two. If both of the cross lagged values were abov e the baseline, this would indicate reciprocal influence Partial correlations were used to evaluate the research questions that probe for whether the corporate financial performance of incumbent candidates and the prior activist campaign experience of challenger candidates are contingent conditions of agendabuilding and agendasetting during proxy battles (Research Questions: RQ12). Contingent conditions are moderators, or conditions, that enhance or limit agendasetting effects. Partial correlations were also used to help determine which type of candidate information subsidy (news release or shareholder letter) was more strongly related with media coverage and investor opinion regarding issue and stakeholder salience (Research Questions: RQ34, RQ9 10). Partial correlations partial out or statistically control for, the presence of one or more variables that might be influencing the relationship among the two variables of interest. Partial correlations have been employed in a similar fashion in previous agendasetting scholarship (e.g. Kiousis,

PAGE 127

127 2003; Kiousis et al., 2009; Kiousis & McCombs, 2004; McCombs et al., 2000; Roberts & McCombs, 1994). Finally, for the supplemental analysis, in which the general distribution of the categorical data (issue and stakeholder mentions) in candidate information subsidies, news media content, and message board posts was examined across the contests, chi square test of independence and Fishers exact test was used. Agendasetting researchers (Golan & Wanta, 2001; Seltzer & Dittmore, 2009) have interpreted the presence of a nonsignificant chisquare value as evidence of agendasetting, on the basis that a nonsignificant value suggests that the proportions of attention to the object (such as an issue or stakeholder) in a comparison among sources of content (such as individual media outlets) are similar. Fishers exact test is a conservative alternative to the chi square test which is recommended for 2 x 2 contingency tables, especially when there are small cells or the data is very unequally distributed (Agresti & Finlay, 1997).

PAGE 128

128 Table 3 1. MergerMarket.com top ten financial PR M&A advisors for 2008 Rank Financial Number of Public Relations Firm Assignments 1 Kekst and Company 85 2 Abernathy MacGregor Group 67 3 Joele Frank Wilkinson Brimmer Katcher 41 4 Sard Verbinnen & Co. 39 5 FD Ashton Partners 30 6 Brunswick Group 27 7 Sloane & Company 21 8 Owen Blicksilver Public Relations Inc. 18 9 Integrated Corporate Relations Inc. (ICR) 11 10 Financial Relations Boar d 10 10 Edelman 10 Source: MergerMarket.com

PAGE 129

129 Table 3 2. Sample of 2005 2009 U.S. corporate proxy contests for study Proxy contest incumbent / challenger Year Meeting date Time 1 Time 2 Winner Target Corporation / Pershing Square 2009 5/28/200 9 4/28/2009 5/12/2009 5/13/2009 5/27/2009 Incumbent NRG Energy, Inc. / Exelon Corporation 2009 7/21/2009 6/21/2009 7/5/2009 7/6/2009 7/20/2009 Incumbent Myers Industries Inc. / GAMCO Investors 2009 4/30/2009 3/31/20094/14/2009 4/15/20094/29/2009 Incumbent PHH Corporation / Pennant Capital 2009 6/10/2009 5/11/2009 5/25/2009 5/26/2009 6/9/2009 Challenger Conseco, Inc. / Keith Long 2009 5/12/2009 4/12/2009 4/26/2009 4/27/2009 5/11/2009 Challenger Biogen Idec Inc. / Carl C. Icahn 2008 6/19/2008 5/20/2 008 6/3/2008 6/4/2008 6/18/2008 Incumbent CSX Corporation / TCI Fund Management 2008 6/15/2008 5/16/2008 5/30/2008 5/31/2006 6/14/2008 Challenger Micrel, Incorporated / Obrem Capital 2008 5/20/2008 4/20/2009 5/4/2009 5/5/2009 5/19/2009 Incumbent Intl. Rectifier / Vishay Intertechnology, Inc. 2008 10/10/2008 9/10/2008 9/24/2008 9/25/2008 10/9/2008 Incumbent Hexcel Corportion / OSS Capital Mgmt. 2008 5/8/2008 4/7/2008 4/21/2008 4/22/2008 5/7/2008 Incumbent Arrow International, Inc. / McNeil Trust 2007 9/20/2007 8/21/2007 9/4/2007 9/5/2007 9/19/2007 Incumbent Atmel Corporation / George Perlegos 2007 5/18/2007 4/18/2007 5/2/2007 5/3/2007 5/17/2007 Incumbent H&R Block, Inc. / Breeden Capital 2007 9/6/2007 8/7/2007 8/21/2007 8/22/2007 9/5/2007 Challenger Motorola, Inc. / Carl C. Icahn 2007 5/7/2007 4/7/2007 4/21/2007 4/22/2007 5/6/2007 Incumbent Openwave Systems / Harbinger Capital 2007 1/17/2007 12/18/2006 1/1/2007 1/2/2007 1/16/2007 Incumbent Career Education Corp. / Steve Bostic 2006 5/18/2006 4/18 /2006 5/2/2006 5/3/2006 5/17/2006 Incumbent UbiquiTel Inc. / Deephaven Capital 2006 6/27/2006 5/28/2006 6/11/2006 6/12/2006 6/26/2006 Incumbent Massey Energy / Third Point LLC 2006 5/16/2006 4/16/20064/30/2006 5/1/20065/15/2006 Challenger Motient Co rporation / Highland Capital 2006 7/12/2006 6/12/2006 6/26/2006 6/27/2006 7/11/2006 Incumbent GenCorp Inc. / Pirate Capital LLC 2006 3/31/2006 3/1/2006 3/15/2006 3/16/2006 3/30/2006 Challenger Blockbuster Inc. / Carl C. Icahn 2005 5/11/2005 4/11/2005 4/ 25/2005 4/26/2005 5/10/2005 Challenger Exar Corporation / GWA Investments 2005 10/27/2005 9/27/2005 10/11/2005 10/12/2005 10/26/205 Challenger Six Flags, Inc. / Daniel Snyder 2005 11/22/2005 10/23/200511/6/2005 11/7/200511/21/2005 Challenger BKF Capit al Group, Inc. / Steel Partners II 2005 6/23/2005 5/24/2005 6/7/2005 6/8/2005 6/22/2005 Challenger Computer Horizons / Crescendo Partners 2005 10/11/2005 9/11/2005 9/25/2005 9/26/2005 10/10/2005 Challenger Notes: Proxy contest year = year corporate proxy contest held, Meeting date = date of annual meeting or special meeting for proxy contest vote, Time 1 = 15day period preceding Time 2, Time 2 = 15day period directly before date of proxy contest vote. All proxy contest data provided by FactSet SharkRepellent.net database

PAGE 130

130 Table 3 3. Frequency of candidate news releases in proxy contests Incumbent Incumbent Challenger Challenger Aggregate Aggregate Aggregate releases releases releases releases incumbent challenger candidate Proxy contest inc umbent / challenger Time 1 Time 2 Time 1 Time 2 releases releases releases Target Corporation / Pershing Square 3 9 1 11 12 12 24 NRG Energy, Inc. / Exelon Corporation 1 4 3 3 5 6 11 Myers Industries Inc. / GAMCO Investors 0 1 0 1 1 1 2 PHH Corporati on / Pennant Capital 3 6 2 2 9 4 13 Conseco, Inc. / Keith Long 0 1 0 1 1 1 2 Biogen Idec Inc. / Carl C. Icahn 0 2 0 0 2 0 2 CSX Corporation / TCI Fund Management 1 3 1 2 4 3 7 Micrel, Incorporated / Obrem Capital 2 4 1 5 6 6 12 Intl. Rectifier / V ishay Intertechnology, Inc. 2 8 1 5 10 6 16 Hexcel Corportion / OSS Capital 0 2 2 2 2 4 6 Arrow International, Inc. / McNeil Trust 1 4 2 5 5 7 12 Atmel Corporation / George Perlegos 4 10 2 5 14 7 21 H&R Block, Inc. / Breeden Capital 1 1 1 3 2 4 6 M otorola, Inc. / Carl C. Icahn 1 3 0 4 4 4 8 Openwave Systems / Harbinger Capital 1 3 0 2 4 2 6 Career Education Corp. / Steve Bostic 1 5 2 10 6 12 18 UbiquiTel Inc. / Deephaven Capital 0 3 0 0 3 0 3 Massey Energy / Third Point LLC 0 1 0 1 1 1 2 Mot ient Corporation / Highland Capital 1 2 2 4 3 6 9 GenCorp Inc. / Pirate Capital LLC 0 2 0 3 2 3 5 Blockbuster Inc. / Carl C. Icahn 1 5 0 2 6 2 8 Exar Corporation / GWA Investments 1 0 0 2 1 2 3 Six Flags, Inc. / Daniel Snyder 1 4 0 4 5 4 9 BKF Capita l Group, Inc. / Steel Partners II 1 0 2 4 1 6 7 Computer Horizons / Crescendo Partners 1 4 3 1 5 4 9 Total 27 87 25 76 114 101 215 Notes: Releases = news releases. All news release frequency data provided by the S.E.C.'s EDGAR filing system

PAGE 131

131 Table 3 4. Frequency of candidate shareholder letters in proxy contests Incumbent Incumbent Challenger Challenger Aggregate Aggregate Aggregate letters letters letters letters incumbent challenger candidate Proxy contest incumbent / challenger Time 1 Time 2 Time 1 Time 2 letters letters letters Target Corporation / Pershing Square 1 1 1 1 2 2 4 NRG Energy, Inc. / Exelon Corporation 1 3 0 0 4 0 4 Myers Industries Inc. / GAMCO Investors 2 1 0 0 3 0 3 PHH Corporation / Pennant Capital 1 0 2 1 1 3 4 Conseco, Inc. / Keith Long 0 0 0 0 0 0 0 Biogen Idec Inc. / Carl C. Icahn 2 2 1 2 4 3 7 CSX Corporation / TCI Fund Management 2 4 0 0 6 0 6 Micrel, Incorporated / Obrem Capital 2 1 0 0 3 0 3 Intl. Rectifier / Vishay Intertechnology, Inc. 1 1 0 0 2 0 2 Hexcel Corportion / OSS Capital 1 0 0 0 1 0 1 Arrow International, Inc. / McNeil Trust 0 0 1 0 0 1 1 Atmel Corporation / George Perlegos 2 0 0 1 2 1 3 H&R Block, Inc. / Breeden Capital 3 2 1 1 5 2 7 Motorola, Inc. / Carl C. Icahn 2 1 2 2 3 4 7 Openwave Systems / Harbinger Capital 0 0 0 0 0 0 0 Career Education Corp. / Steve Bostic 1 3 2 3 4 5 9 UbiquiTel Inc. / Deephaven Capital 1 1 0 0 2 0 2 Massey Energy / Third Point LLC 1 2 1 1 3 2 5 Motient Corporation / Highland Capital 2 1 0 0 3 0 3 GenCorp Inc. / Pirate Capital LLC 0 1 0 2 1 2 3 Blockbuster Inc. / Carl C. Icahn 1 0 0 2 1 2 3 Exar Corporation / GWA Investments 1 1 1 1 2 2 4 Six Flags, Inc. / Daniel Snyder 2 2 1 0 4 1 5 BKF Capital Group, Inc. / Steel Partners II 1 1 1 2 2 3 5 Computer Horizons / Crescendo Partners 1 0 2 0 1 2 3 Total 31 28 16 19 59 35 94 Notes: Letters = shareholder letters. All shareholder letter frequency data provided by the S.E.C.'s EDGAR filing system

PAGE 132

132 Table 3 5. Financial media attention to proxy contests WSJ NYT FT RN AP DJ BN Aggregate stories stories stories stories stories stories stories media Proxy contest T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 coverage Target Corporation / Pershing Square 1 3 0 2 2 3 6 7 13 8 3 15 8 23 94 NRG En ergy, Inc. / Exelon Corporation 4 1 1 1 4 4 5 5 3 4 6 7 12 24 81 Myers Industries Inc. / GAMCO Investors 0 0 0 0 0 0 0 0 0 0 1 0 1 0 2 PHH Corporation / Pennant Capital 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 Conseco, Inc. / Keith Long 0 0 0 0 0 0 0 0 0 0 1 1 0 0 2 Biogen Idec Inc. / Carl C. Icahn 1 1 0 0 0 0 0 4 0 4 16 9 19 8 62 CSX Corporation / TCI Fund Management 1 1 0 0 2 4 4 6 0 1 5 7 7 24 62 Micrel, Incorporated / Obrem Capital 0 0 0 0 0 0 1 0 1 1 4 9 0 0 16 Intl. Rectifier / Vishay Intertechnology, Inc. 2 1 0 0 0 0 2 3 3 3 8 8 6 3 39 Hexcel Corportion / OSS Capital 0 0 0 0 0 0 0 0 1 0 3 3 0 0 7 Arrow International, Inc. / McNeil Trust 0 0 0 0 0 0 1 0 1 1 1 0 3 0 7 Atmel Corporation / George Perlegos 0 0 0 0 0 0 1 3 0 1 3 4 5 2 19 H&R Block, In c. / Breeden Capital 1 0 1 0 1 1 5 6 3 6 7 7 10 10 58 Motorola, Inc. / Carl C. Icahn 2 3 1 2 3 5 7 14 5 8 11 21 17 27 126 Openwave Systems / Harbinger Capital 0 0 0 0 0 0 2 3 2 3 3 4 3 2 22 Career Education Corp. / Steve Bostic 0 0 0 0 0 0 2 3 0 2 2 7 6 9 31 UbiquiTel Inc. / Deephaven Capital 0 0 0 0 0 0 0 1 0 0 0 4 0 0 5 Massey Energy / Third Point LLC 0 0 0 0 0 0 1 0 0 3 0 0 2 0 6 Motient Corporation / Highland Capital 0 0 0 0 0 0 0 3 0 0 2 2 0 1 8 GenCorp Inc. / Pirate Capital LLC 0 0 0 0 0 0 0 0 0 1 0 3 0 0 4 Blockbuster Inc. / Carl C. Icahn 2 2 1 3 0 0 8 16 9 11 10 14 8 12 96 Exar Corporation / GWA Investments 0 0 0 0 0 0 0 0 0 0 0 2 0 0 2 Six Flags, Inc. / Daniel Snyder 1 0 0 1 0 0 1 7 2 9 4 8 0 17 50 BKF Capital Group, Inc. / Steel Par tners II 0 0 0 1 0 0 0 1 1 0 2 0 0 0 5 Computer Horizons / Crescendo Partners 0 0 0 0 0 0 1 0 0 0 3 1 0 0 5 Notes: WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News, AP = Associated Press, DJ = Dow Jones News, BN = Bloomberg News, Aggregate media coverage = the combined media coverage across the seven media outlets, Time 1 (T1) = 15 day period preceding Time 2, Time 2 (T2) = 15day period directly before data of proxy contest vote

PAGE 133

133 Table 3 6. Investor attention on message boards to proxy contests Message board Message board Aggregate message Proxy contest incumbent / challenger posts Time 1 posts Time 2 board posts Target Corporation / Pershing Square 112 102 214 NRG Energy, Inc. / Exelon Corporation 98 1 70 268 Myers Industries Inc. / GAMCO Investors 6 3 9 PHH Corporation / Pennant Capital 4 4 8 Conseco, Inc. / Keith Long 954 875 1829 Biogen Idec Inc. / Carl C. Icahn 39 47 86 CSX Corporation / TCI Fund Management 98 111 209 Micrel, Incorporated / Obrem Capital 66 80 146 Intl. Rectifier / Vishay Intertechnology, Inc. 76 133 209 Hexcel Corportion / OSS Capital 45 29 74 Arrow International, Inc. / McNeil Trust n/a n/a n/a Atmel Corporation / George Perlegos n/a n/a n/a H&R Block, Inc. / Breede n Capital 427 409 836 Motorola, Inc. / Carl C. Icahn 24 3 27 Openwave Systems / Harbinger Capital 7 24 31 Career Education Corp. / Steve Bostic 83 294 377 UbiquiTel Inc. / Deephaven Capital n/a n/a n/a Massey Energy / Third Point LLC 65 62 127 Mo tient Corporation / Highland Capital n/a n/a n/a GenCorp Inc. / Pirate Capital LLC 3 4 7 Blockbuster Inc. / Carl C. Icahn n/a n/a n/a Exar Corporation / GWA Investments 1 2 3 Six Flags, Inc. / Daniel Snyder n/a n/a n/a BKF Capital Group, Inc. / Steel Partners II 9 12 21 Computer Horizons / Crescendo Partners 43 70 113 Total 2160 2434 4594 Notes: Time 1 = 15 day period preceding Time 2, Time 2 = 15 day period directly before date of proxy vote All message board post data provided by the searchable message board archives of Yahoo! Finance n/a = archived message board post data on the company was not available for this time period

PAGE 134

134 Table 3 7. Financial performance of incumbent candidates in proxy contests Proxy contest Incumbent candidate Ma rket value ($) Assets ($) Revenue ($) Net Income ($) Target Corporation / Pershing Square Target Corporation 33,663 17,488 64,948 2,214 NRG Energy, Inc. / Exelon Corporation NRG Energy, Inc. 10,001 24,808 6,885 1,188 Myers Industries Inc. / GAMCO Inv estors Myers Industries Inc. 287 569 868 44 PHH Corporation / Pennant Capital PHH Corporation 831 8,273 2,960 254 Conseco, Inc. / Keith Long Conseco, Inc. 1,800 28,770 4,190 1,127 Biogen Idec Inc. / Carl C. Icahn Biogen Idec Inc. 18,379 8,629 3, 172 638 CSX Corporation / TCI Fund Management CSX Corporation 19,700 25,534 10,030 1,336 Micrel, Incorporated / Obrem Capital Micrel, Incorporated 592 295 258 44 Intl. Rectifier / Vishay Intertechnology, Inc. International Rectifier 2,419 2,647 1,2 02 78 Hexcel Corportion / OSS Capital Hexcel Corportion 1,982 1,060 1,171 61 Arrow International, Inc. / McNeil Trust Arrow Intl, Inc. 902 697 482 56 Atmel Corporation / George Perlegos Atmel Corporation 999 1,818 1,671 15 H&R Block, Inc. / Breeden Capital H&R Block, Inc. 7,009 7,500 4,021 434 Motorola, Inc. / Carl C. Icahn Motorola, Inc. 49,200 38,593 42,879 3,661 Openwave Systems / Harbinger Capital Openwave Systems 1,600 919 412 5 Career Education Corp. / Steve Bostic Career Education Corp 3,552 1,495 1,855 234 UbiquiTel Inc. / Deephaven Capital UbiquiTel Inc. 657 563 423 48 Massey Energy / Third Point LLC Massey Energy 2,899 2,987 2,204 102 Motient Corporation / Highland Capital Motient Corporation 442 233 0 122 GenCorp Inc. / Pirate Capital LLC GenCorp Inc. 1,000 1,056 622 231 Blockbuster Inc. / Carl C. Icahn Blockbuster Inc. 508 3,995 6,053 1,249 Exar Corporation / GWA Investments Exar Corporation 595 503 57 5 Six Flags, Inc. / Daniel Snyder Six Flags, Inc. 662 3,642 880 465 BKF Capital Group, Inc. / Steel Partners II BKF Capital Group 178 162 123 2 Computer Horizons / Crescendo Partners Computer Horizons 100 159 214 25 Notes: Numbers in millions of dollars. All corporate financial and market value data provided by the annual 10K filing (preceding the proxy contest vote) of the incumbent candidate with the S.E.C.'s EDGAR system

PAGE 135

135 Table 3 8. Activist campaign experience of challenger candidates in proxy contests Proxy contest Challenger candidate Activist c ampaign Experience Target Corporation / Pershing Square Pershing Square Capital 38 NRG Energy, Inc. / Exelon Corporation Exelon Corporation 1 Myers Industries Inc. / GAMCO Investors GAMCO Investors 264 PHH Corporation / Pennant Capital Pennant Capi tal 5 Conseco, Inc. / Keith Long Keith Long 2 Biogen Idec Inc. / Carl C. Icahn Carl C. Icahn 77 CSX Corporation / TCI Fund Management TCI Fund Management 10 Micrel, Incorporated / Obrem Capital Obrem Capital 2 Intl. Rectifier / Vishay Intertechnol ogy, Inc. Vishay Intertechnology 1 Hexcel Corportion / OSS Capital OSS Capital Management 1 Arrow International, Inc. / McNeil Trust Robert W. Cruickshank 1 Atmel Corporation / George Perlegos George Perlegos 1 H&R Block, Inc. / Breeden Capital Bree den Capital 8 Motorola, Inc. / Carl C. Icahn Carl C. Icahn 77 Openwave Systems / Harbinger Capital Harbinger Capital 37 Career Education Corp. / Steve Bostic Steve Bostic 2 UbiquiTel Inc. / Deephaven Capital Deephaven Capital 2 Massey Energy / Th ird Point LLC Third Point LLC 44 Motient Corporation / Highland Capital Highland Capital 16 GenCorp Inc. / Pirate Capital LLC Pirate Capital 24 Blockbuster Inc. / Carl C. Icahn Carl C. Icahn 77 Exar Corporation / GWA Investments GWA Investments 3 Si x Flags, Inc. / Daniel Snyder Daniel Synder 1 BKF Capital Group, Inc. / Steel Partners II Steel Partners II 116 Computer Horizons / Crescendo Partners Crescendo Partners 22 Notes: Activist campaign experience = number of previous activist campaigns waged by the challenger candidate according to FactSet SharkRepellent.net database

PAGE 136

136 CHAPTER 4 RESULTS System Level Agenda Setting The first set of hypotheses probe for agendabuilding and agendasetting relationships at a systemlevel with each agenda consis ting of the salience of the proxy contests in the sample of contests by year. In other words, these hypotheses do not assess the transfer of issue or stakeholder salience at an individual level (within each contest), but rather test for the presence of pos itive relationships among the frequencies of candidate news releases generated, candidate shareholder letters generated, financial news stories written and investor message board postings made within each group of five contests held within the same calendar year. Tables 4 1 through 45 display these system level descriptive statistics and correlations by proxy contest year. H1a, which posited that the salience of proxy contests in new releases would be positively related with the salience of proxy contests in media coverage, was not supported. For incumbent candidate releases, statistically significant positive relationships were detected in six of 30 possible comparisons with the individual media outlets by year. An unexpected significant inverse relationship was found in one case. For challenger candidate releases, just three of 30 possible comparisons attained significance. Finally, aggregate candidate releases were significantly correlated with aggregate media coverage in none of five possible comparisons. H1b predicted that the salience of proxy contests in candidate shareholder letters would be positively related to the salience of proxy contests in media coverage. This hypothesis was also not supported. Only 10 of 60 comparisons (five for incumbent le tters and five for challenger letters) attained significance. Aggregate candidate

PAGE 137

137 shareholder letters were significantly correlated with aggregate media coverage in just one of five possible comparisons. Next, hypotheses H2a and H2b tested for the direct c andidateto investor opinion agendasetting relationship at a system level during proxy contests. Specifically, H2a predicted that the salience of proxy contests in candidate news releases would be positively related to the salience of proxy contests in investor opinion, as measured by investor message board posts. This hypothesis was not supported. While challenger candidate releases were significantly linked with investor opinion in one of two possible comparisons, incumbent releases were not linked in either of two possible comparisons. Finally, the aggregate candidate news releases were not significantly linked with investor opinion in either of the two possible comparisons. H2b, which expected the salience of proxy contests in candidate shareholder le tters to be positively associated with the salience of proxy contests in investor opinion, also was not supported. Out of two possible comparisons apiece, no significant linkages were detected for either incumbent or challenger letters. The aggregate candi date shareholder letters were also not significantly correlated with investor opinion in either of the two possible comparisons. Turning to the traditional agendasetting relationship at a system level, H3 predicted that the salience of proxy contests in media coverage would be positively correlated with investor opinion. The data did not support this hypothesis. None of 14 possible comparisons with the individual media outlets and investor opinion attained statistical significance, and neither of the two possible comparisons with the aggregate media coverage were significant.

PAGE 138

138 Overall, the results of the system level hypothesis testing indicate that the sheer amount of candidate information subsidies disseminated during a proxy fight does not stimulate the amount of raw media or investor attention that a contest receives. However, it is worth noting that these system level correlations did reveal evidence of inter media agendasetting bonds among the media outlets in their coverage of the contests. There was a fairly high level of agreement among the outlets in the allocation of attention, as measured by the frequency of stories each outlet generated per contest. Significant inter media linkages in the patterns of news coverage among the financial news gatherers were detected in 51 of 78 possible comparisons. System Level Contingent Conditions The systemlevel research questions examine the role of corporate financial performance (incumbent) and activist investor campaign experience (challenger) as potential contingent conditions or moderators, affecting the magnitude of candidateto news media agendabuilding relationships during proxy fights. With this in mind, RQ1a asked about the relationship between the salience of proxy contests in incumbent candidate news releases and media coverage before and after controlling for indicators of corporate financial performance (stock market value, assets, revenue and net income of the publicly held corporations that were the incumbent candidates in the contests). There were five bivariate correlations in which incumbent releases and coverage by individual media outlets was significantly linked. Partial correlations, controlling for the effect of each corporate financial indicator on the incumbent news release to media relationship at a system level during proxy fights, found the following: for market value, the before (bivariate) and after (multivariate) correlation values increased in four of five possible comparisons; for

PAGE 139

139 assets, the before and after values increased in three of five possible comparisons; for revenue, the values declined in four of five possible comparisons, and for net income, the values also declined in four of five possible comparisons. RQ1b inquired about the relationship between the salience of proxy contests in incumbent candidate shareholder letters and media coverage before and after controlling for indicators of corporate financial performance. There were five bivariate correlations in which incumbent shareholder letters and coverage by individual media outlets were significantly linked. As with RQ1a, bivariate and partial correlations were used to answer this question. The before and after correlations for market value rose in three of five possible comparisons; for assets, the value rose in t hree of four possible comparisons; for revenue, the value rose in three of five possible comparisons, and for net income, it declined in three of five possible comparisons. Taken as a whole, while undoubtedly exploratory in nature, these results indicate that all financial performance moderators of the incumbent candidateto media agendabuilding relationship do not behave the same. Based on the data, two of these indicators, stock market value and assets, tend to enhance the strength of incumbent candidate to media agendabuilding relationships (for both incumbent news releases and shareholder letters), while another indicator, net income, a gauge of a corporations profitability, tends to limit the magnitude of these associations (for both incumbent rel eases and letters). Finally, firm revenue (i.e. sales) as a contingent condition appears more complex, as it reduced agendabuilding linkages when the incumbent information subsidy was a release and increased it when the subsidy was a letter.

PAGE 140

140 RQ2a asked about the relationship between the salience of proxy contests in challenger candidate news releases and media coverage before and after controlling for the previous activist investor campaign experience of these challenger candidates. As with the previous question, bivariate and partial correlations were used to answer this inquiry. There were five bivariate correlations in which challenger releases and coverage by individual media outlets were significantly related. The before (bivariate) and after (multivariate) correlation values were unchanged in all three possible comparisons. RQ2b asked the same question regarding challenger shareholder letters. In this case, the before and after correlations values increased a small amount in four of five possible comp arisons. The data indicates that the more prior activist investor campaign experience a challenger candidate holds marginally enhances the amount of media attention that shareholder letters disseminated by the challenger receives. In other words, the medi a seems somewhat more likely to pay more attention to letters sent by veteran activist investors than by less experienced activists. Experience did not seem to have an impact on challenger releases and coverage. Placement of Issues and Stakeholders on Agendas Prior to examining the hypotheses and research questions at an individual contest level the general trends among the placement of the issues and stakeholders on the various agendas will be examined to provide a highlevel backdrop for the results. As shown on Table 46, the five issues that received the most combined media attention across the contests were mentions of acquisitions, shareholder value, partnerships, transactions, and corporate governance. The series of chi square tests on Table 47

PAGE 141

141 reve aled that there were some significant differences among the individual media outlets in the proportion of total attention allocated to certain issues. Table 46 also shows that the five issues which received the most investor attention across the proxy contests were shareholder value, compensation, acquisitions, transactions, and performance. Table 48 and Table 49 displays the top five most mentioned issues in the combined candidate news releases and shareholder letters, respectively. Within incumbent news releases, shareholder value, strategy, experience, corporate governance, and director independence were most mentioned, while within challenger releases, shareholder value, corporate governance, experience, director independence, and transactions were most emphasized. Turning to incumbent shareholder letters, shareholder value, experience, strategy, acquisitions, and director independence were the most salient. Within challenger letters, shareholder value was again most emphasized, followed by experience, corporate governance, performance, and director independence. Shifting to the examination of the placement of stakeholders on the various agendas, Table 410 shows that the five stakeholder groups which received the most combined media attention were mentions of the board of directors, shareholders, management, news media, and courts. As with issue salience, the series of chi square tests on Table 411 found that there were some significant differences between the individual media outlets in the proportion of total attention each focused on certain stakeholders. Table 410 also displays the allocation of investor attention across the contests to the various stakeholders. The top five most salient stakeholders for

PAGE 142

142 investors were the board of directors, s hareholders, management, news media, and analysts. The stakeholder emphases in the combined incumbent and challenger candidate news releases were also similar. As exhibited on Table 412, the top four stakeholders emphasized in both incumbent and challenger releases were the board of directors, shareholders, management, and proxy advisors. The next most salient stakeholder in incumbent releases was customers, while in challenger releases, it was employees. Finally, Table 413 displays the allocation of attention to stakeholders in candidate shareholder letters. These emphases were again similar in the combined incumbent and challenger shareholder letters, with the board of directors, management and shareholders again the three most emphasized stakeholders. The next most mentioned stakeholders in incumbent letters were customers and employees, while for challenger letters, employees and proxy advisors rounded out the top five most emphasized stakeholders. In addition to the just reviewed aggregate data, the d istribution of issue and stakeholder mentions in candidate news releases and shareholder letters for each of the 25 fights are displayed on Tables E 1 E25 of Appendix E. Issue mentions in news stories and investor message board posts for each of the fig hts are displayed on Tables F1 F25 of Appendix F, and stakeholder mentions in news stories and message board posts for each of the fights are displayed on Tables G 1 G25 of Appendix G. Candidate to Media Agenda Building The third set of hypotheses and research questions tests at an individual level (per contest) for the presence of significant linkages between the issue and stakeholder priorities found in candidate news releases and shareholder letters, with the emphases

PAGE 143

143 these issues and stakeholders receive in financial news media coverage. Agendasetting research that concentrates on the dynamic twoway exchange of salience among source controlled information subsidies and news media content is called agendabuilding (Berkowitz & Adams, 1990; Turk, 1985; Turk & Franklin, 1987). Tables 414 and 415 summarize the issue agendabuilding relationships for each of the contests, while Tables 416 and 417 summarize the stakeholder agendabuilding relationships. See Tables H 1 H25 in Appendix H for the individual candidateto news medium issue agendabuilding correlations within each contest and I 1 I25 in Appendix I for the individual candidate tonews medium stakeholder agendabuilding correlations. H4a, which predicted that the salience of issues in candidate news releases would be positively associated with the salience of issues in media coverage, received solid support. For incumbent candidate news releases, statistically significant correlations were observed in 40 of 67 possible correlations, with a median correlation value of .51. For challenger candidate releases, 38 of 63 possible correlations attained significance. The median correlation value was .50. H4b tested the same relationship using candidate shareholder letters. This hypothesis als o received solid support. Thirty seven of 62 possible correlations were significantly associated regarding the salience of issues in incumbent candidate letters with the salience of issues in media coverage. The median correlation value was .48. For the ch allenger candidate letters, 28 of 44 possible correlations were significant, with a median correlation value of .50. Hypotheses H5a and H5b examined the candidateto media relationship as it relates to stakeholder salience. H5a, which posited positive linkages between the

PAGE 144

144 salience of stakeholders in candidate news releases and media coverage, was not supported. Incumbent candidate news releases were only significantly associated with media coverage in 10 of 85 possible comparisons, with a median correlation value of .31. For challenger releases, only 17 of 82 possible comparisons attained significance. The median correlation value was .30. The evidence also did not support H5b, which tested the candidateto media stakeholder agendabuilding relationship with candidate shareholder letters. For incumbent shareholder letters, only 13 of 73 possible correlations attained significance, with a median value of .25. Finally, for challenger letters, 12 of the 59 possible comparisons were significant, and the median value was .33. RQ3 asked whether candidate news releases or shareholder letters were more strongly related with media coverage regarding issue salience. When examining the ratio of significant correlations to possible correlations for each information subsidy type, the data indicated that the overall strength of the linkage for releases (78/130) and letters (65/106) with media coverage was very similar. To more rigorously assess this question, partial correlations were computed to control for the impact of each subsidy type whenever significant bivariate correlations were found in a contest for both candidate news releases/media coverage and candidate shareholder letters/media coverage. This additional analysis revealed that candidate releases were more stro ngly linked with media coverage than letters in 10 of 15 possible multivariate comparisons. RQ4 inquired about the relative strength of these two candidate information subsidy types regarding linkages with stakeholder salience in media coverage. While both subsidy types were only weakly associated overall with the salience of

PAGE 145

145 stakeholders in media coverage, the ratio of significant correlations to possible correlations favored news releases (27/167) over shareholder letters (15/132). To further evaluate t his question, partial correlations were once again used to control for the impact of each subsidy type. Consistent with the findings of the bivariate comparisons, the partial correlations found that releases were more strongly associated with media coverage than letters in both of the two possible multivariate comparisons. Research questions RQ5a through RQ6b move beyond testing for candidateto media associations to probing for the direction of influence in these relationships. In other words, do the priorities articulated in candidate information subsidies generally lead to shifts in the priorities in media coverage or do subsidies instead generally respond to the priorities found in coverage? RQ5a asked whether the salience of issues in news releases influenced the salience of issues in media coverage. Figure 41 and 42 display the results of the cross lag correlation analyses for this question. For the newspaper coverage, out of six possible comparisons, the candidate news releases led, or set the newspaper agenda, in two cases, mutual influence was detected in three instances, and there was no evidence of influence between the two agendas in the other case. Candidate releases were less successful in unidirectionally shaping news wire coverage. Out of 19 possible comparisons, the cross lags revealed sixteen instances of reciprocal influence between the wire coverage and releases, and three cases in which the wire coverage led. Notably, there were no instances in which the releases led, or set the agenda of news wire coverage. RQ5b explored whether the salience of issues in candidate shareholder letters influenced the salience of issues in media coverage. Figures 43 and 4 4 provide the

PAGE 146

146 results of the crosslags for this question. For the shareholder let ter to newspaper coverage cross lags, the letters led the newspaper coverage in two of the three possible comparisons. Mutual influence was found in the other comparison. For the shareholder letter to wire coverage comparisons, out of 14 possible comparisons, mutual influence was exhibited in nine cases, letters led in three cases, and the wire coverage led in the remaining two instances. Overall, the data indicates that there is no dominant actor in the candidateto media relationship as it relates to infl uencing issue salience. While there are cases of candidate information subsidies leading media coverage and vice versa, in general, the giveandtake between these two actors during proxy fights seems fairly balanced. RQ6a and RQ6b investigate the direction of influence in the candidateto media relationship regarding stakeholder salience with candidate releases and letters, respectively. Figures 4 5 and 46 display the results of the cross lags for RQ6a, while Figures 47 and 48 display the results for RQ6b. Starting with candidate releases and newspaper coverage, out of five possible comparisons, releases led in two cases and coverage led in one case. Mutual influence was exhibited in the other two cases. For news wire coverage, out of 16 possible comparisons, there was evidence of mutual influence between the two agendas in nine cases, wire coverage led in five cases, and releases led in two cases. Turning to candidate letters and newspaper coverage, out of two possible comparisons, mutual influence was found in one case and coverage led in the other instance. Out of 11 possible comparisons, wire coverage led in eight cases and mutual influence was displayed in the three other cases. While the direction of influence in the candidate news release to media relationship for stakeholder salience

PAGE 147

147 appears fairly balanced, the data indicates that shareholder letters often react to the stakeholder priorities first established in media coverage. The media seems to lead this aspect of the relationship. Inter Candidate Agenda Setting The fourth set of hypotheses tests for the transfer of salience between the competing incumbent and challenger candidate agendas, a concept known as inter candidate agendasetting (Tedesco, 2005a, 2005b). Table 418 summarizes the inter candidate issue and stakeholder agendasetting relationships among the competing candidate information subsidies (candidatecontrolled news releases and shareholder letters) for each of the 25 corporate proxy contests. H6a, which predicted that the salience of issues in the competing candidate news releases would be positively related, received very strong support. As displayed for each proxy contest on Table 418, 18 of 20 possible comparisons attained statistical significance with a median correlation value of .71. H6b, which expected positive relationships between the salience of issues in the competing candidate shareholder letters, also received robust support. Specifically, 13 of 15 possible comparisons attained significance with a median correlation value of .66. The strongest inter candidate agendasetting relationship was found for H7a, which predicted that the salience of stakeholders would be positively related in the competing candidate news releases. This hypothesis was supported in all 21 pos sible comparisons, with a median correlation value of .84. Finally, H7b, which expected positive associations between the salience of stakeholders in the candidate shareholder letters, also received robust support. Significant associations were found in al l 12 possible comparisons, with a median correlation of .67.

PAGE 148

148 Media to Investor Agenda Setting The fifth set of hypotheses and research questions tests for a variation of the classic mass media to public opinion agendasetting relationship (i.e. McCombs & S haw, 1972). In this new context of proxy contests, the traditional agendasetting linkage is evaluated by testing for the transfer of issue and stakeholder salience among business news media coverage (media agenda) and investor opinion during these contest ed corporate elections. Table 419 summarizes the mediato investor issue and stakeholder agendasetting relationships in each of the contests. See Tables H 1 H25 in Appendix H (issue salience) and Tables I 1 I25 in Appendi x I (stakeholder salience) for the individual news medium to investor correlations in each fight. H8a predicted that the salience of issues in media coverage would be positively related to the perceived salience of issues in investor opinion. This hypothesis received mixed support, with 19 of 45 possible comparisons attaining statistical significance. The median correlation value was .37. Much stronger support was found for H8b, which expected a positive relationship between media coverage and investor opinion regarding the salience of stakeholders. This hypothesis was supported in 61 of 66 possible comparisons, with a median value of .71. Moving beyond establishing associations, the two research questions assessed the direction of influence in the mediato investor relationship. RQ7 asked whether the salience of issues in news media coverage influenced the perceived salience of issues in investor opinion. Figures 47 and 48 display the results of the cross lag correlation analyses for this question. For the news wire coverageto -i nvestor opinion cross lags, out of nine possible comparisons, mutual influence among the two agendas was detected in three cases, investors led in three cases and the news wires led in two

PAGE 149

149 cases. For the newspaper coverageto investor opinion cross lags, o ut of four possible comparisons, the newspapers led the investor agenda in two cases, investors led in one case and mutual influence was detected in one case. Based on these analyses, no dominant direction of influence was found in the mediato investor relationship regarding issue salience. RQ8 asked whether the salience of stakeholders in media coverage influenced the perceived salience of stakeholders in investor opinion. Figures 49 and 4 10 display the results of the crosslag analyses for this question. For the wire coverageto investor crosslags, out of eight possible comparisons, there was mutual influence in five cases and the wires led the investor agenda in three instances. For the newspaper coverageto investor opinion cross lags, out of four possible comparisons, the newspapers led in three cases and there was reciprocal influence in the other case. Notably, investors did not set the media agenda in any case. These results indicate that the media leads this relationship in terms of influencing stakeholder salience. Candidate to Investor Agenda Setting Looking beyond the influence of candidatecontrolled information subsidies on business media coverage, the sixth set of hypotheses and research questions extends recent research on potential direct sourceto public agendasetting linkages. Specifically, in the new context of corporate proxy contests, this hypothesized bond is evaluated by testing for the transfer of issue and stakeholder salience among candidate news releases and investor opinion, as well as between candidate shareholder letters and investor opinion. Tables 420 and 421 summarize the candidate to investor issue and stakeholder agendasetting relationships for each of the contests.

PAGE 150

150 H10a, which expected that the salience of issues in candidate news releases would be positively related to the perceived salience of issues in investor opinion, received mixed support. For the incumbent candidate news releases, seven of 16 possible correlations attained significance, with a median correlation value of .71. The challenger candidate releases were significantly associated with investor opinion in eight of 16 possible correlations, with a median value of .84. The next hypothesis, H10b, also received mixed support. The salience of issues in incumbent shareholder letters was significantly associated with the perceived salience of issues in investor opinion in seven of 15 possible correlations. The median value was .66. Finally, for challenger shareholder letters, five of nine possible correlations attained significance, with an almost identical median value of .67. H11a, which posited that the salience of stakeholders in candidate news releases and the perceived salience of stakeholders in investor opinion would be positively associated, was not supported. Only two of 17 possible correlations with the incumbent news releases attained significance, and the median correlation value was .23. Similarly, only one of 15 possible correlations with the challenger news releases was significant. The median correlation value was .30. H11b also was not supported. The salience of stakeholders in incumbent shareholder letters was only significantly related to the perceived salience of stakeholders in investor opinion in one out of 12 possible correlations. The me dian value was .29. For challenger letters, three of eight possible correlations were significantly associated, and the median value was .37.

PAGE 151

151 RQ9 asked whether news releases or shareholder letters were more strongly related with investor opinion regarding issue salience. When examining the ratio of significant correlations to possible correlations for each information subsidy type, the strength of linkages for letters (12/24) and releases (15/32) with investor opinion was almost identical. To more rigorously assess this question, partial correlations were computed to control for the impact of each subsidy type whenever significant bivariate correlations were found in a contest for both candidate news releases/investor opinion and candidate shareholder letters/investor opinion. In five of eight possible comparisons, this multivariate analysis revealed that stronger linkages existed between news releases and investor opinion than between shareholder letters and investor opinion. RQ10 inquired whether news releases or shareholder letters were more strongly linked with investor opinion regarding stakeholder salience. While both types of subsidies were only weakly associated with stakeholder salience in investor opinion, the ratio of significant correlations to possible correlations favored the letters (4/20), rather than the news releases (3/32). Partial correlation analyses were not performed since there were not any instances in which candidate releases and letters were both significantly linked with investor opinion. The next four research questions moved beyond testing for linkages to assess the direction of influence in the candidateto investor relationship during proxy fights. Specifically, RQ11a asked whether the salience of issues in news releases influenced the perceived salience of issues in investor opinion, while RQ11b asked the same question as it relates to shareholder letters. Figure 411 displays the results of the

PAGE 152

152 crosslag correlation analyses for RQ11a and Figure 412 exhibits the results for RQ11b. Starting with candidate releases, out of 11 possible comparisons, mutual influence was detected in six cases, releases led in three cases, and investor opinion led in two cases. Turning to candidate letters, out of nine possible comparisons, mutual influence was found in six cases, and candidate letters and investor opinion both led in two cases apiece. As it relates to issue salience influence in the candidateto investor relationship appears to be fairly balanced with no dominant direction of infl uence apparent. RQ12a and RQ12b asked whether the salience of stakeholders in new releases and shareholder letters, respectively, influenced the perceived salience of stakeholders in investor opinion. Figure 413 exhibits the results of the cross lag analyses for RQ12a and Figure 414 provides the results for RQ12b. For the candidate news release to investor opinion cross lags, out of 10 possible comparisons, mutual influence was identified in eight cases, and investor opinion led in the other two instances Turning to the shareholder letters, out of eight possible comparisons, the cross lag analyses revealed six cases of mutual influence, and one case apiece in which shareholder letters and investor opinion led. As was the case for issue salience, there appears to be no dominant direction of influence in the candidateto investor relationship as it relates to stakeholder salience. For both issue and stakeholder salience, reciprocal influence among candidate information subsidies and investor opinion was the norm during the contests. Consequences of Agenda Setting The agendasetting function of the mass media in the formation of public opinion is well established (for detailed overviews of the empirical support for this proposition, see

PAGE 153

153 Dearing & Rogers, 1996; McCombs, 2006; Wanta & Ghanem, 2007). In addition to shaping cognitions specifically the perceived importance of objects and the attributes of objects, some agendasetting researchers have suggested that there may be attitudinal and behavioral consequences to this process. With this in mind, the final set of research questions probes for potential agendabuilding and agendasetting consequences as it relates to shareholder voting (i.e. behavioral outcomes) in corporate proxy contests. Specifically the f inal two questions probed for potential linkages between the strength of the agenda building and agendasetting relationships within each contest, and the voting outcome of the contest. In other words, in general, do winning candidates in contests enjoy stronger candidateto news media agendabuilding relationships than the losers? Along the same vein, in general, do contest winners enjoy stronger candidateto investor opinion agendasetting relationships than losers? RQ13 inquired about whether the candid ate to media agendabuilding relationship would be stronger for the candidates that win the contest than for the candidates that lose the contest. The median correlation values that serve as the source for these comparisons may be found on Tables 414 417. Starting with issue salience in news releases, the contest winner enjoyed stronger agendabuilding bonds with media coverage in 12 of 18 possible comparisons. For issue salience in shareholder letters, the winner exhibited stronger agendabuilding link ages with media coverage in six of 11 possible comparisons. Turning to stakeholder salience, the contest winner produced stronger agendabuilding associations in 10 of 21 possible comparisons for releases, and in nine of 12 possible comparisons for letters. Except for the candidate news releases to media coverage stakeholder comparisons, the other three sets of

PAGE 154

154 comparisons found that winners enjoyed stronger agendabuilding linkages with the media in the majority of cases. RQ14 asked whether the candidateto investor agendasetting relationship would be stronger for the candidates that win the proxy contest than for the candidates that lose the contested corporate election. The median correlation values that serve as the source for these comparisons can be found on Table 420 and 421. For issue salience, contest winners displayed stronger agendasetting relationships than contest losers in seven of 14 possible comparisons with candidate news releases, and six of nine possible comparisons with candidate shareholder letters. As for stakeholder salience, contest winners enjoyed stronger agendasetting relationships than losers in seven of 14 possible comparisons with releases, and four of seven possible comparisons with letters. Therefore, the strength of the c andidate to investor agendasetting correlations appears to be consequential only for shareholder letters and not for news releases. Taken as a whole, the data provides tentative evidence that the c ampaign with the stronger candidateto media agendabuil ding relationship (as shaped with either news releases or shareholder letters) and direct candidateto investor agendasetting relationship (as shaped only with letters) does tend to win the proxy contest. Future research, specifically focused on potential outcomes of the agenda building and agendasetting processes, is needed to draw more definitive conclusions in this area.

PAGE 155

155 Table 4 1. System level descriptive statistics and correlations for 2009 proxy contests Variable Mean s.d. 1 2 3 4 5 1. Incumben t news releases 5.60 4.88 1.00 2. Challenger news releases 4.80 4.55 .90* 1.00 1.00 3. Aggregate news releases 10.40 9.13 1.00 .90* 4. Incumbent shareholder letters 2.00 1.58 .05 .36 .05 1.00 5. Challenger shareholder letters 1.00 1.41 .80 .4 6 .80 .34 1.00 6. Aggregate shareholder letters 3.00 1.73 .86* .86* .86* .45 .63 7. Wall Street Journal stories 1.80 2.49 .46 .80 .46 .67 .13 8. New York Times stories 0.80 1.10 .59 .89* .59 .58 .00 9. Financial Times stories 2.60 3.72 .46 .80 .46 .6 7 .13 10. Reuters News stories 4.60 6.39 .69 .92* .69 .45 .13 11. Associated Press stories 5.60 9.13 .69 .92* .69 .45 .13 12. Dow Jones News stories 7.00 7.97 .50 .76 .50 .21 .06 13. Bloomberg News stories 13.60 18.26 .29 .66 .29 .87* .29 14. Aggre gate news stories 36.00 47.24 .37 .68 .37 .46 .23 15. Investor message board postings 466.80 772.84 .36 .05 .36 .10 .67 16. Activist campaign experience 62.00 113.96 .05 .15 .05 .00 .22 17. Market value 9316.40 14170.87 .62 .82* .62 .10 .11 18. Assets 15981.60 11626.36 .10 .10 .10 .30 .34 19. Revenue 15970.20 27465.80 .62 .82* .62 .10 .11 20. Net income 395.40 1310.34 .62 .82* .62 .70 .11 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either una vailable or did not provide sufficient variance for statistical analysis.

PAGE 156

156 Table 4 1 Continued Variable 6 7 8 9 10 11 12 13 14 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbe nt shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder letters 1.00 7. Wall Street Journal stories .63 1.00 8. New York Times stories .65 .97** 1.00 9. Financial Times stories .63 1.00** .97** 1.00 10. Reuters News stories .63 .88* .97** .88* 1.00 11. Associated Press stories .63 .88* .97** .88* 1.00*** 1.00 12. Dow Jones News stories .34 .80 .89* .80 .92* .92* 1.00 13. Bloomberg News stories .52 .92* .89* .92* .80 .80 .66 1.00 14. Aggregate news stories .29 .80 .89* .80 .92* .92* .92* .82* 1.00 15. Investor message board postings .45 .34 .29 .34 .22 .22 .56 .21 .46 16. Activist campaign experience .11 .45 .29 .45 .11 .11 .31 .15 .05 17. Market value .45 .78 .87* .78 .89* .89* .98** .56 .82* 18. Assets .22 .34 .29 .34 .22 .22 .56 .05 .31 19. Revenue .45 .78 .87* .78 .89* .89* .98** .56 .82* 20. Net income .67 .78 .87* .78 .89* .89* .67 .87* .82* Notes: p < .05, ** p < .01, *** p < .001. All one tai led tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 157

157 Table 41 Continued 15 16 17 18 19 20 Variable 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder letters 7. Wall Street Journal stories 8. New York Times stories 9. Financial T imes stories 10. Reuters News stories 11. Associated Press stories 12. Dow Jones News stories 13. Bloomberg News stories 14. Aggregate news stories 1.00 15. Investor message board postings .60 1.00 16. Activis t campaign experience .50 .40 1.00 17. Market value .90* .80 .60 1.00 18. Assets .50 .40 1.00*** .60 1.00 19. Revenue .10 .20 .06 .20 .60 1.00 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailabl e or did not provide sufficient variance for statistical analysis.

PAGE 158

158 Table 4 2. System level descriptive statistics and correlations for 2008 proxy contests Variable Mean s.d. 1 2 3 4 5 1. Incumbent news releases 4.80 3.34 1.00 2. Challenger new s releases 3.80 2.49 .79 1.00 3. Aggregate news releases 8.60 5.46 .98** .87* 1.00 4. Incumbent shareholder letters 3.20 1.92 .05 .56 .20 1.00 5. Challenger shareholder letters 0.60 1.34 .54 .54 .71 .35 1.00 6. Aggregate shareholder letters 3 .80 2.59 .21 .67 .40 .90* .71 7. Wall Street Journal stories 1.40 1.34 .43 .08 .32 .26 .19 8. New York Times stories n/a n/a n/a n/a n/a n/a n/a 9. Financial Times stories 1.20 2.68 .00 .36 .00 .71 .25 10. Reuters News stories 4.00 3.94 .41 .21 .30 .70 .00 11. Associated Press stories 2.80 2.17 .50 .26 .36 .10 .36 12. Dow Jones News stories 14.40 6.95 .21 .15 .00 .30 .71 13. Bloomberg News stories 13.40 14.78 .08 .66 .21 .82 .36 14. Aggregate news stories 37.20 25.47 .03 .63 .21 .87* .54 15. Investor message board postings 136.00 69.22 .87* .41 .80 .30 .35 16. Activist campaign experience 18.20 33.09 .40 .76 .57 .87* .73 17. Market value 8614.40 9552.08 .21 .72 .30 .70 .35 18. Assets 7633.00 10527.92 .21 .72 .30 .70 .35 19. Revenue 3166.60 3981.47 .21 .72 .30 .70 .35 20. Net income 431.40 564.17 .21 .72 .30 .70 .35 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailable or did not provide sufficient variance f or statistical analysis.

PAGE 159

159 Table 42 Continued Variable 6 7 8 9 10 11 12 13 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder l etters 1.00 6. Aggregate shareholder letters .26 1.00 7. Wall Street Journal stories n/a n/a n/a 8. New York Times stories .35 .19 n/a 1.00 9. Financial Times stories .50 .79 n/a .71 1.00 10. Reuters News stories .15 .65 n/a .5 4 .21 1.00 11. Associated Press stories .60 .58 n/a .35 .30 .87* 1.00 12. Dow Jones News stories .72 .65 n/a .73 .87* .03 .31 1.00 13. Bloomberg News stories .87* .65 n/a .54 .82* .24 .56 .95** 14. Aggregate news stories .10 .74 n/a .35 .80 .46 .30 .41 15. Investor message board postings .98** .08 n/a .36 .36 .03 .46 .66 16. Activist campaign experience .60 .63 n/a .71 .80 .05 .20 .98** 17. Market value .60 .63 n/a .71 .80 .05 .20 .98** 18. Assets .60 .63 n/a .71 .80 .05 .20 .98** 19. Reven ue .60 .63 n/a .71 .80 .05 .20 .98** Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 160

160 Table 42 Continued Variable 14 15 16 17 18 19 20 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder letters 7. Wall Street Journal s tories 8. New York Times stories 9. Financial Times stories 10. Reuters News stories 11. Associated Press stories 12. Dow Jones News stories 13. Bloomberg News stories 14. Aggregate news stories 1.00 15. Investor message board postings .41 1.00 16. Activist campaign experience .79 .10 1.00 17. Market value .87* .30 .56 1.00 18. Assets .87* .30 .10 1.00*** 1.00 19. Revenue .87* .30 .56 1.00*** 1.00*** 1.00 Variable .87* .30 56 1.00*** 1.00*** 1.00*** 1.00 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 161

161 Table 4 3. System level descriptive statistics and corr elations for 2007 proxy contests Variable Mean s.d. 1 2 3 4 5 1. Incumbent news releases 5.80 4.71 1.00 2. Challenger news releases 4.80 2.17 .73 1.00 3. Aggregate news releases 10.60 6.31 .92* .89* 1.00 4. Incumbent shareholder letters 2.0 0 2.12 .55 .08 .92* 1.00 5. Challenger shareholder letters 1.60 1.52 .37 .11 .00 .82* 1.00 6. Aggregate shareholder letters 3.60 3.29 .40 .11 .03 .95** .95** 7. Wall Street Journal stories 1.20 2.17 .63 .30 .34 .80 .92* 8. New York Times stori es 0.80 1.30 .63 .30 .34 .80 .92* 9. Financial Times stories 2.00 3.46 .63 .30 .34 .80 .92* 10. Reuters News stories 8.40 7.93 .72 .63 .56 .72 .67 11. Associated Press stories 6.00 5.00 .82* .63 .67 .56 .67 12. Dow Jones News stories 12.20 11.99 .61 .46 .40 .82* .76 13. Bloomberg News stories 15.80 17.11 .46 .26 .21 .87* .82* 14. Aggregate news stories 46.50 48.40 .72 .63 .56 .72 .67 15. Investor message board postings 298.00 465.93 n/a n/a n/a n/a n/a 16. Activist campaign exp erience 24.80 32.74 .63 .81* .63 .29 .37 17. Market value 11942.00 20982.62 .72 .63 .56 .72 .67 18. Assets 9905.40 16276.67 .46 .26 .21 .87* .82* 19. Revenue 9893.00 18497.39 .31 .11 .05 .87* .98** 20. Net income 660.60 1689.13 .46 .37 .56 .21 .36 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 162

162 Table 4 3 Continued Variable 6 7 8 9 10 11 1. Incumbent news release s 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder letters 1.00 7. Wall Street Journal stories .86* 1.00 8. New York Times stories .86* 1.00*** 1.00 9. Financial Times stories .86* 1.00*** 1.00*** 1.00 10. Reuters News stories .67 .89* .89* .89* 1.00 11. Associated Press stories .56 .89* .89* .89* .90* 1.00 12. Dow Jones News stories .79 .92* .92* .92* .98 ** .82* 13. Bloomberg News stories .87* .89* .89* .89* .90* .70 14. Aggregate news stories .67 .89* .89* .89* 1.00*** .90* 15. Investor message board postings n/a n/a n/a n/a n/a n/a 16. Activist campaign experience .26 .69 .69 .69 .87* .87* 17. Mark et value .67 .89* .89* .89* 1.00*** .90* 18. Assets .87* .89* .89* .89* .90* .70 19. Revenue .98** .89* .89* .89* .70 .60 20. Net income .10 .22 .22 .22 .00 .10 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests.

PAGE 163

163 Table 4 3 Continued Variable 12 13 14 15 16 17 18 19 20 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggr egate shareholder letters 7. Wall Street Journal stories 8. New York Times stories 9. Financial Times stories 10. Reuters News stories 11. Associated Press stories 12. Dow Jones News stories 1.00 13. Bloomberg News stories .98** 1.00 14. Aggregate news stories .98** 1.00*** 1.00 15. Investor message board postings n/a n/a n/a n/a 16. Activist campaign experience .76 .62 .87* n/a 1.00 17. Market value .98** .90* 1 .00*** n/a .87* 1.00 18. Assets .98** 1.00*** .90* n/a .62 .90* 1.00 19. Revenue .82* .90* .70 n/a .36 .70 .90* 1.00 20. Net income .05 .10 .00 n/a .15 .00 .10 .30 1.00 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests.

PAGE 164

164 Table 4 4. System level descriptive statistics and correlations for 2006 proxy contests Variable Mean s.d. 1 2 3 4 5 1. Incumbent news releases 3.00 1.87 1.00 2. Challenger news releases 4.40 4.83 .56 1.00 3. Aggregate news releases 7.40 6.50 .82* .90* 1.00 4. Incumbent shareholder letters 2.60 1.14 .50 .62 .46 1.00 5. Challenger shareholder letters 1.80 2.05 .11 .53 .32 .43 1.00 6. Aggregate shareholder letters 4.40 2.79 .13 .67 .36 .76 .87* 7. Wall Street Journal stories n/a n/a n/a n/a n/a n/a n/a 8. New York Times stories n/a n/a n/a n/a n/a n/a n/a 9. Financial Times stories n/a n/a n/a n/a n/a n/a n/a 10. Reuters News stories 2.00 2.00 .76 .67 .67 .92* .22 11. Associated Press stories 1.20 1.30 .37 .15 .21 .40 .81* 12. Dow Jones News stories 4.00 3.24 1.00*** .56 .82* .50 .11 13. Bloomberg News stories 3.60 6.43 .29 .62 .36 .95** .65 14. Aggregate news stories 10.80 11.39 .62 .70 .06 .98** .32 15. Investor message board postings 170.33 188.80 n/a n/a n/a n/a n/a 16. Activist campaign experience 17.60 17.52 .92* .21 .56 .26 .08 17. Market value 1710.00 1416.64 .05 .30 .10 .46 .95** 18. Assets 1266.80 1074.67 .36 .00 .30 .31 .79 19. Revenue 1020.80 955.81 .36 .00 .30 .31 .79 20. Net income 34.60 180.0 9 .62 .10 .20 .67 .32 Notes: p < .05, ** p < .01, *** p < .001. All onetailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 165

165 Table 4 4 Continued Variable 6 7 8 9 10 11 12 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder letters 1.00 7. Wall Street Journal stories n/a n/a 8. New York Times stories n/a n/a n/a 9. Financial Times stories n/a n/a n/a n/a 10. Reuters News stories .53 n/a n/a n/a 1.00 11. Associated Press stories .82* n/a n/a n/a .03 1.00 12. Dow Jones News stories .13 n/a n/a n/a .76 .37 1.0 0 13. Bloomberg News stories .92* n/a n/a n/a .76 .66 .29 14. Aggregate news stories .67 n/a n/a n/a .98** .21 .62 15. Investor message board postings n/a n/a n/a n/a n/a n/a n/a 16. Activist campaign experience .16 n/a n/a n/a .55 .50 .92* 17. Mar ket value .82* n/a n/a n/a .21 .87* .05 18. Assets .72 n/a n/a n/a .05 .98** .36 19. Revenue .72 n/a n/a n/a .05 .98** .36 20. Net income .36 n/a n/a n/a .67 .21 .62 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 166

166 Table 4 4 Continued (2) Variable 13 14 15 16 17 18 19 20 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder letters 7. Wall Street Journal stories 8. New York Times stories 9. Financial Times stories 10. Reuters New s stories 11. Associated Press stories 12. Dow Jones News stories 13. Bloomberg News stories 1.00 14. Aggregate news stories .87* 1.00 15. Investor message board postings n/a n/a n/a 16. Activist campaign ex perience .03 .36 n/a 1.00 17. Market value .67 .30 n/a .05 1.00 18. Assets .56 .10 n/a .41 .90* 1.00 19. Revenue .56 .10 n/a .41 .90* 1.00*** 1.00 20. Net income .56 .60 n/a .67 .50 .30 .30 1.00 Notes: p < .05, ** p < .01, *** p < .001 All one tailed tests.

PAGE 167

167 Table 4 5. System level descriptive statistics and correlations for 2005 proxy contests Variable Mean s.d. 1 2 3 4 5 1. Incumbent news releases 3.60 2.41 1.00 2. Challenger news releases 3.60 1.67 .36 1.00 3. Aggreg ate news releases 7.20 2.49 .65 .27 1.00 4. Incumbent shareholder letters 2.00 1.23 .47 .31 .08 1.00 5. Challenger shareholder letters 2.00 0.71 .47 .35 .57 .35 1.00 6. Aggregate shareholder letters 4.00 1.00 .58 .58 .16 .92* .00 7. Wall Stre et Journal stories 1.00 1.73 .83* .41 .34 .06 .50 8. New York Times stories 1.20 1.64 .56 .03 .11 .03 .00 9. Financial Times stories n/a n/a n/a n/a n/a n/a n/a 10. Reuters News stories 6.80 10.13 .87* .08 .55 .16 .34 11. Associated Press stori es 6.40 8.91 .65 .11 .24 .05 .23 12. Dow Jones News stories 8.60 9.58 .97** .35 .63 .27 .57 13. Bloomberg News stories 7.40 10.19 .83* .41 .34 .06 .50 14. Aggregate news stories 31.60 41.16 .87* .08 .55 .16 .34 15. Investor message board pos tings 45.67 59.00 n/a n/a n/a n/a n/a 16. Activist campaign experience 43.80 50.71 .05 .32 .31 .53 .89* 17. Market value 408.60 254.60 .00 .37 .10 .74 .67 18. Assets 1692.20 1950.07 .53 .58 .05 .16 .45 19. Revenue 1465.40 2585.60 .95** .16 67 .26 .45 20. Net income 347.20 541.79 .95** .16 .67 .26 .45 Notes: p < .05, ** p < .01, *** p < .001. All onetailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 168

168 Table 4 5 Continued Variable 6 7 8 9 10 11 12 1. Incumbent news releases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder let ters 1.00 7. Wall Street Journal stories .18 1.00 8. New York Times stories .08 .83* 1.00 9. Financial Times stories n/a n/a n/a n/a 10. Reuters News stories .16 .92* .87* n/a 1.00 11. Associated Press stories .08 .92* .97** n/a 92* 1.00 12. Dow Jones News stories .41 .92* .65 n/a .92* .76 1.00 13. Bloomberg News stories .18 1.00*** .83* n/a .92* .92* .92* 14. Aggregate news stories .16 .92* .87* n/a 1.00*** .92* .92* 15. Investor message board postings n/a n/a n/a n/a n/ a n/a n/a 16. Activist campaign experience .16 .11 .37 n/a .10 .15 .15 17. Market value .47 .45 .21 n/a .15 .36 .21 18. Assets .00 .89* .74 n/a .67 .82* .67 19. Revenue .32 .89* .74 n/a .98** .82* .98** 20. Net income .32 .89* .74 n/a .98** 82* .98** Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 169

169 Table 4 5 Continued Variable 13 14 15 16 17 18 19 20 1. Incumbent news r eleases 2. Challenger news releases 3. Aggregate news releases 4. Incumbent shareholder letters 5. Challenger shareholder letters 6. Aggregate shareholder letters 7. Wall Street Journal stories 8. New York Times stories 9. Financial Times stories 10. Reuters News stories 11. Associated Press stories 12. Dow Jones News stories 13. Bloomberg News stories 1.00 14. Aggregate news stories 92* 1.00 15. Investor message board postings n/a n/a n/a 16. Activist campaign experience .11 .10 n/a 1.00 17. Market value .45 .15 n/a .70 1.00 18. Assets .89* .67 n/a .20 .70 1.00 19. Revenue .89* .98** n/a .00 .10 .60 1.00 20. Net income .89** .98** n/a .00 .10 .60 1.00*** 1.00 Notes: p < .05, ** p < .01, *** p < .001. All one tailed tests. n/a = Data was either unavailable or did not provide sufficient variance for statistical analysis.

PAGE 170

170 Table 4 6. Aggregate m edia attention and investor attention to issues during proxy contests Aggregate media attention Investor attention Significance level Issue Mentions n % n % Fisher's exact test Acquisition 638 28.4% 154 14.8% p < .001 Compensation 78 3.5% 175 16.8% p < .001 Corporate governance 174 7.8% 34 3.3% p < .001 Director independence 104 4.6% 19 1.8% p < .001 Divestiture 26 1.2% 17 1.6% n.s. Experience 155 6.9% 57 5.5% n.s. Joint Venture 1 0.0% 6 0.6% p < .01 Litigation 88 3.9% 23 2.2% p < .05 Merger 14 0.6% 45 4.3% p < .001 Partnership 204 9.1% 22 2.1% p < .001 Performance 94 4.2% 63 6.0% p < .05 Shareholder value 301 13.4% 185 17.8% p < .001 Social responsibility 1 0.0% 3 0.3% n.s. Stock dividend 13 0.6% 31 3.0% p < .001 Stock repurchase 66 2.9% 26 2.5% n.s. Strategy 106 4.7% 54 5.2% n.s. Transaction 180 8.0% 128 12.3% p < .001 Notes: Fisher's exact test is a conservative alternative to the chi square test for a 2 x 2 contingency table. df = 1 for all tests. Total number of financial news stories = 806. Total number of investor message board posts = 4594. Media coverage of contests without corresponding message board post data was excluded from this comparison. There is a significant association ( r s = .65, p < .001) between the aggr egate media attention and investor attention.

PAGE 171

1 71 Table 4 7. Media attention to issues during proxy contests across individual media outlets WSJ NYT FT RN Issue mentions n % n % n % n % Acquisition 53 30.5% 14 26.9% 51 32.5% 122 28.7% C ompensation 11 6.3% 7 13.5% 8 5.1% 41 9.6% Corporate governance 16 9.2% 2 3.8% 7 4.5% 32 7.5% Director independence 6 3.4% 0 0.0% 9 5.7% 22 5.2% Divestiture 3 1.7% 1 1.9% 2 1.3% 8 1.9% Experience 8 4.6% 3 5.8% 12 7.6% 27 6.4% Joint vent ure 0 0.0% 0 0.0% 0 0.0% 1 0.2% Litigation 8 4.6% 0 0.0% 0 0.0% 12 2.8% Merger 3 1.7% 3 5.8% 1 0.6% 2 0.5% Partnership 5 2.9% 2 3.8% 6 3.8% 21 4.9% Performance 8 4.6% 5 9.6% 9 5.7% 20 4.7% Shareholder value 9 5.2% 6 11.5% 22 14.0% 42 9.9% Social responsibility 0 0.0% 0 0.0% 0 0.0% 0 0.0% Stock dividend 8 4.6% 0 0.0% 3 1.9% 4 0.9% Stock repurchase 5 2.9% 0 0.0% 4 2.5% 10 2.4% Strategy 9 5.2% 7 13.5% 4 2.5% 33 7.8% Transaction 22 12.6% 2 3.8% 19 12.1% 28 6.6% No tes: Total number of financial news stories = 806, small cells = over 20% of the cells have an expected count of less than 5.

PAGE 172

172 Table 4 7 Continued AP DJN BN Significance level Issue mentions n % n % n % chi square test Acquisition 73 18.2% 13 3 22.2% 340 28.6% 2 = 27.9, df = 6, p < .001 Compensation 40 10.0% 65 10.9% 127 10.7% 2 = 8.68, df = 6, n.s. Corporate governance 32 8.0% 43 7.2% 63 5.3% 2 = 9.22, df = 6, n.s. Director independence 20 5.0% 20 3.3% 37 3.1% 2 = 9.35, df = 6, n.s. Divestiture 7 1.7% 16 2.7% 3 0.3% small cells Experience 30 7.5% 23 3.8% 66 5.6% 2 = 8.17, df = 6, n.s. Joint venture 0 0.0% 0 0.0% 0 0.0% small cells Litigation 5 1.2% 19 3.2% 56 4.7% 2 = 20.85, df = 6, p < .01 Merger 3 0.7% 6 1.0% 3 0.3% small cells Partnership 25 6.2% 62 10.4% 92 7.7% 2 = 21.87, df = 6, p < .001 Performance 10 2.5% 17 2.8% 54 4.5% 2 = 11.23, df = 6, n.s. Shareholder value 47 11.7% 67 11.2% 170 14.3% 2 = 18.45, df = 6, p < .01 Social respons ibility 0 0.0% 1 0.2% 0 0.0% small cells Stock dividend 20 5.0% 22 3.7% 8 0.7% small cells Stock repurchase 12 3.0% 5 0.8% 30 2.5% 2 = 8.59, df = 6, n.s. Strategy 50 12.5% 45 7.5% 64 5.4% 2 = 32.49, df = 6, p < .001 Transaction 27 6.7% 55 9.2% 76 6.4% 2 = 17.44, df = 6, p < .01 Notes: Total number of financial news stories = 806, small cells = over 20% of the cells have an expected count of less than 5.

PAGE 173

173 Table 4 8. Candidate attention to issues in news releases during during proxy contests. Incumbent news releases Challenger news releases Significance level Issue mentions n % n % Fisher's exact test Acquisition 107 5.9% 80 4.1% p < .05 Compensation 108 5.9% 121 6.2% n.s. Corporate governance 185 10.1% 258 13 .1% p < .01 Director independence 182 10.0% 249 12.7% p < .01 Divestiture 11 0.6% 19 1.0% n.s. Experience 246 13.5% 257 13.1% n.s. Joint venture 4 0.2% 6 0.3% n.s. Litigation 34 1.9% 28 1.4% n.s. Merger 78 4.3% 32 1.6% p < .001 Par tnership 39 2.1% 81 4.1% p < .001 Performance 101 5.5% 133 6.8% n.s. Shareholder value 294 16.1% 418 21.3% p < .001 Social responsibility 2 0.1% 0 0.0% n.s. Stock dividend 21 1.2% 5 0.3% p < .001 Stock repurchase 34 1.9% 16 0.8% p < .01 Strategy 255 14.0% 100 5.1% p < .001 Transaction 124 6.8% 159 8.1% n.s. Notes: Fisher's exact test is a conservative alternative to the chi square test for a 2 x 2 contingency table. df = 1 for all tests. Total number of incumbent news releases = 114. Total number of challenger news releases = 107. There is a significant positive association ( r s = .91, p < .001) between the issues mentions in the combined incumbent and challenger candidate news releases across the proxy contests.

PAGE 174

174 Table 4 9. Candidate attention to issues in shareholder letters during proxy contests. Incumbent shareholder letters Challenger shareholder letters Significance level Issue mentions n % n % Fisher's exact test Acquisition 110 9.7% 23 3.0% p < .001 Com pensation 58 5.1% 68 9.0% p < .001 Corporate governance 63 5.6% 95 12.5% p < .001 Director independence 76 6.7% 65 8.6% n.s. Divestiture 9 0.8% 17 2.2% p < .05 Experience 178 15.7% 118 15.6% n.s. Joint venture 0 0.0% 3 0.4% n.s. Litigation 15 1.3 % 6 0.8% n.s. Merger 18 1.6% 30 4.0% p < .01 Partnership 64 5.6% 40 5.3% n.s. Performance 45 4.0% 71 9.4% p < .001 Shareholder value 255 22.5% 145 19.1% n.s. Social responsibility 0 0.0% 1 0.1% n.s. Stock dividend 8 0.7% 4 0.5% n.s. Stock repurcha se 13 1.1% 17 2.2% n.s. Strategy 150 13.2% 27 3.6% p < .001 Transaction 73 6.4% 28 3.7% p < .01 Notes: Fisher's exact test is a conservative alternative to the chi square test for a 2 x 2 contingency table. df = 1 for all tests. Total number of in cumbent shareholder letters = 59. Total number of challenger shareholder letters = 35. There is a significant positive association ( r s = .75, p < .001) between the issues mentions in the combined incumbent and challenger candidate shareholder letters acro ss the proxy contests.

PAGE 175

175 Table 4 10. Aggregate media attention and investor attention to stakeholders during proxy contests Aggregate media attention Investor attention Significance level Stakeholder mentions n % n % Fisher's exact test Analyst 121 1.9% 77 5.1% p < .001 Board of directors 2379 38.3% 445 29.2% p < .001 Community 1 0.0% 8 0.5% p < .001 Court 216 3.5% 23 1.5% p < .001 Customer 92 1.5% 68 4.5% p < .001 Employee 70 1.1% 61 4.0% p < .001 Government 40 0.6% 23 1.5% p < .001 Man agement 877 14.1% 292 19.2% p < .001 News media 480 7.7% 171 11.2% p < .001 Proxy advisor 183 2.9% 11 0.7% p < .001 Regulator 68 1.1% 7 0.5% p < .05 Retiree 0 0.0% 0 0.0% n.s. Shareholder 1668 26.8% 318 20.9% p < .001 Stakeholder 1 0.0% 1 0.1% n.s. Supplier 3 0.0% 7 0.5% p < .001 Union 14 0.2% 11 0.7% p < .01 Notes: Fisher's exact test is a conservative alternative to the chi square test for a 2 x 2 contingency table. df = 1 for all tests. Total number of financial news stories = 806. Total nu mber of investor message board posts = 4594. Media coverage of contests without corresponding message board post data was excluded from this comparison. There is a significant association ( r s = .84, p < .001) between the aggregate media attention and inv estor attention.

PAGE 176

176 Table 4 11. Media attention to stakeholders during proxy contests across individual media outlets WSJ NYT FT RN Stakeholder mentions n % n % n % n % Analyst 11 2.6% 0 0.0% 2 0.7% 28 2.1% Board of directors 161 37.6% 48 35.0% 131 43.8% 451 34.6% Community 0 0.0% 0 0.0% 0 0.0% 0 0.0% Court 6 1.4% 0 0.0% 10 3.3% 48 3.7% Customer 22 5.1% 2 1.5% 7 2.3% 21 1.6% Employee 8 1.9% 3 2.2% 2 0.7% 2 0.2% Government 2 0.5% 0 0.0% 11 3.7% 2 0.2% Managemen t 62 14.5% 18 13.1% 37 12.4% 197 15.1% News media 13 3.0% 12 8.8% 10 3.3% 104 8.0% Proxy advisor 5 1.2% 0 0.0% 2 0.7% 49 3.8% Regulator 2 0.5% 0 0.0% 4 1.3% 19 1.5% Retiree 0 0.0% 0 0.0% 0 0.0% 0 0.0% Shareholder 135 31.5% 54 39.4% 82 27.4% 382 29.3% Stakeholder 0 0.0% 0 0.0% 0 0.0% 0 0.0% Supplier 1 0.2% 0 0.0% 1 0.3% 0 0.0% Union 0 0.0% 0 0.0% 0 0.0% 0 0.0% Notes: Total number of financial news stories = 806, small cells = over 20% of the cells have an expected count of less than 5.

PAGE 177

177 Table 4 11 Continued AP DJN BN Significance level Stakeholder mentions n % n % n % chi square test Analyst 73 18.2% 133 22.2% 340 28.6% 2 = 27.9, df = 6, p < .001 Board of directors 40 10.0% 65 10.9% 127 10.7% 2 = 8.68, df = 6, n.s. Community 32 8.0% 43 7.2% 63 5.3% 2 = 9.22, df = 6, n.s. Court 20 5.0% 20 3.3% 37 3.1% 2 = 9.35, df = 6, n.s. Customer 7 1.7% 16 2.7% 3 0.3 % small cells Employee 30 7.5% 23 3.8% 66 5.6% 2 = 8.17, df = 6, n.s. Government 0 0.0% 0 0.0% 0 0.0% small cells Management 5 1.2% 19 3.2% 56 4.7% 2 = 20.85, df = 6, p < .01 News media 3 0.7% 6 1.0% 3 0.3% small cells Proxy advisor 25 6.2% 62 10.4% 92 7.7% 2 = 21.87, df = 6, p < .001 Regulator 10 2.5% 17 2.8% 54 4.5% 2 = 11.23, df = 6, n.s. Retiree 47 11.7% 67 11.2% 170 14.3% 2 = 18.45, df = 6, p < .01 Shareholder 0 0.0% 1 0.2% 0 0.0% small cells Stakeholder 20 5.0% 22 3.7% 8 0.7% small cells Supplier 12 3.0% 5 0.8% 30 2.5% 2 = 8.59, df = 6, n.s. Union 50 12.5% 45 7.5% 64 5.4% 2 = 32.49, df = 6, p < .001 Stakeholder mentions 27 6.7% 55 9.2% 76 6.4% 2 = 17.44, df = 6, p < .01

PAGE 178

178 Tabl e 4 12. Candidate attention to stakeholders in news releases during proxy contests. Incumbent news releases Challenger news releases Significance level Stakeholder Mentions n % n % Fisher's exact test Analyst 21 0.5% 33 0.8% n.s. Boa rd of directors 1914 47.7% 1872 43.8% p < .001 Community 5 0.1% 12 0.3% n.s. Court 69 1.7% 32 0.7% p < .001 Customer 77 1.9% 40 0.9% p < .001 Employee 59 1.5% 63 1.5% n.s. Government 4 0.1% 6 0.1% n.s. Management 513 12.8% 617 14.4% p < .05 News media 23 0.6% 32 0.7% n.s. Other 0 0.0% 1 0.0% n.s. Proxy advisor 169 4.2% 208 4.9% n.s. Regulator 19 0.5% 50 1.2% p < .001 Retiree 0 0.0% 0 0.0% n.s. Shareholder 1139 28.4% 1310 30.6% p < .05 Supplier 2 0.0% 2 0.0% n.s. Union 0 0.0% 0 0.0% n.s. Notes: Fisher's exact test is a conservative alternative to the chi square test for a 2 x 2 contingency table. df = 1 for all tests. Total number of incumbent news releases = 114. Total number of challenger news releases = 107. There is a significant positive association (rs = .94, p < .001) between the issues mentions in the combined incumbent and challenger candidate news releases across the proxy contests.

PAGE 179

179 Table 4 13. Candidate attention to stakeholders in sharehol der letters during proxy contests. Incumbent shareholder letters Challenger shareholder letters Significance level Stakeholder Mentions n % n % Fisher's exact test Analyst 5 0.2% 12 0.8% p < .05 Board of directors 1074 49.7% 653 43.2% p < .001 Community 3 0.1% 7 0.5% n.s. Court 27 1.2% 3 0.2% p < .001 Customer 48 2.2% 25 1.7% n.s. Employee 35 1.6% 58 3.8% p < .001 Government 17 0.8% 1 0.1% p < .001 Management 265 12.3% 327 21.7% p < .001 News media 13 0.6% 9 0.6% n.s. Other 0 0.0% 0 0.0% n.s. Proxy advisor 28 1.3% 39 2.6% p < .01 Regulator 17 0.8% 13 0.9% n.s. Retiree 0 0.0% 0 0.0% n.s. Shareholder 624 28.8% 361 23.9% p < .001 Supplier 6 0.3% 2 0.1% n.s. Union 1 0.0% 0 0.0% n.s. Notes: Fisher's exact test is a conservative alternative to the chisquare test for a 2 x 2 contingency table. df = 1 for all tests. Total number of incumbent shareholder letters = 59. Total number of challenger shareholder letters = 35. There is a significant positive association (rs = .89, p < .001) between the stakeholder mentions in the combined incumbent and challenger candidate shareholder letters across the proxy contests.

PAGE 180

180 Table 4 14. Summary of candidate news releases media issue agenda building relationships in proxy contests Corpora te proxy contest Incumbent releases Media Challenger releases Media Incumbent/Challenger Median correlation Correlation ratio Median correlation Correlation ratio Target .69*** 6 / 7 .67** 6 / 7 NRG Energy .39 3 / 7 .61** 6 / 7 Myers Industries PHH / Pennant Capital Conseco / Keith Long Biogen Idec .37 1 / 4 CSX Corporation / TCI Fund .22 0 / 6 .03 0 / 6 Micrel .49* 1 / 2 .53* 2 / 2 International Rectifier .78*** 2 / 2 .71*** 2 / 2 Hexcel .87*** 1 / 1 .37 0 / 1 Arrow Intl. .62** 1 / 1 .55* 1 / 1 Atmel .67** 2 / 3 .63** 3 / 3 H&R Block / Breeden Capit al .21 2 / 5 .53* 3 / 5 Motorola .65** 7 / 7 .29 1 / 7 Openwave .48* 2 / 4 .03 2 / 4 Career Education Bostic .59** 3 / 3 .66** 3 / 3 UbiquiTel Massey Energy / Third Point .46* 1 / 2 .16 0 / 2 Motient .70*** 1 / 1 .78*** 1 / 1 GenCorp / Pirate Capital Blockbuster / Carl Icahn .41* 3 / 5 .11 1 / 5 Exar / GWA Investments Six Flags / Daniel Snyder .59** 4 / 5 .82** ** 5 / 5 BKF Capital / Steel Partners II .36 0/ 1 .67** 1 / 1 Computer Horizons/ Crescendo .36 0 / 1 .70*** 1 / 1 Total 40 / 67 38 / 63 Notes: Releases = candidate news releases, Median Correlation = The median candidate news media correlation coefficient value for each proxy contest. Correlation Ratio = The proportion of possible candidate news media correlations per contest that attain statistical significance. p < .05, ** p <.01, *** p <.00 1. All one tailed tests.

PAGE 181

181 Table 4 15. Summary of candidate shareholder letters media issue agenda building relationships in proxy contests Corporate proxy contest Incumbent letters Media Challenger letters Media Incumbent/Challenger Median c orrelation Correlation ratio Median correlation Correlation ratio Target .10 0 / 7 .01 0 / 7 NRG Energy .45* 4 / 7 Myers Industries PHH / Pennant Capital Conseco / Keith Long Biogen Idec .51* 4 / 4 .48* 3 / 4 CSX Corpora tion / TCI Fund .02 0 / 6 Micrel .64** 2 / 2 International Rectifier .86*** 2 / 2 Hexcel .75*** 1 / 1 Arrow Intl. .20 0 / 1 Atmel .53* 2 / 3 .63** 3 / 3 H&R Block / Breeden Capital .15 1 / 5 .57** 4 / 5 Motorola .61** 7 / 7 .46* 7 / 7 Openwave Career Education .60** 3 / 3 .69*** 2 / 3 UbiquiTel Massey Energy / Third Point .65** 2 / 2 .73*** 2 / 2 Motient .68*** 1 / 1 GenCorp / Pirate Capital Blockbuster / Carl Icahn .39 2 / 5 .01 0 / 5 Exar / GWA Investments Six Flags / Daniel Sn yder .75*** 5 / 5 .62** 5 / 5 BKF Capital / Steel Partners II .55* 1 / 1 .52* 1 / 1 Computer Horizons / Crescendo .36 0 / 1 .53* 1 / 1 Total 37 / 62 28 / 44 Notes: Letters = candidate shareholder letters, Median correlation = The median can didate news media correlation coefficient value for each corporate proxy contest. Correlation ratio = The proportion of possible candidate news media correlations per contest that attain statistical significance. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 182

182 Table 4 16. Summary of candidate news releases media stakeholder agenda building relationships in proxy contests Corporate proxy contest Incumbent releases Media Challenger releases Media Incumbent/Challenger Median correlation Correlation ratio Median correlation Correlation ratio Target .41 1 / 7 .32 0 / 7 NRG Energy .32 1 / 7 .14 0 / 7 Myers Industries PHH / Pennant Cap ital Conseco / Keith Long .13 0 / 1 .09 0 / 1 Biogen Idec .23 0 / 4 CSX Corporation / TCI Fund .29 0 / 6 .37 2 / 6 Micrel .33 0 / 3 .39 1 / 3 International Rectifier .13 0 / 4 .30 0 / 4 Hexcel .30 0 / 2 .55* 2 / 2 Arrow Intl. .07 0 / 2 .09 0 / 2 Atmel .24 0 / 4 .36 1 / 4 H&R Block / Breeden Capital .43* 3 / 6 .46* 4 / 6 Motorola 22 0 / 7 .30 0 / 7 Openw ave 24 0 / 4 .04 0 / 4 Career Education 13 0 / 4 .20 0 / 4 UbiquiTel Massey Energy / Third Point 39 0 / 3 .46* 2 / 3 Motient 30 0 / 3 .39 1 / 3 GenCorp / Pirate Capital 42 1 / 2 .34 0 / 2 Blockbuster / Carl Icahn 41 1 / 6 .44* 3 / 6 Exar / GWA Investments 02 0 / 1 .13 0 / 1 Six Flags / Daniel Snyder 38 2 / 5 .20 0 / 5 BKF Capital / Steel Partners II 19 1 / 3 .27 1 / 3 Computer Horizons / Crescendo .08 0 / 1 .23 0 / 2 Total 10 / 85 17 / 82 Notes: Releases = candidate news releases, Median Correlation = The median candidate news media correlation coefficient value for each corporate proxy contest. Correlation Ratio = The proportion of possible candidatenews media correlations per contest that attain statistical significance. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 183

183 Table 4 17. Summary of candidate shareholder letters medi a stakeholder agenda building relationships in proxy contests Corporate proxy contest Incumbent letters Media Challenger letters Media Incumbent/Challenger Median correlation Correlation ratio Median correlation Correlation ratio Target hing Square .07 0 / 7 .32 1 / 7 NRG Energy .32 3 / 7 Myers Industries PHH / Pennant Capital Conseco / Keith Long Biogen Idec .37 1 / 4 .31 1 / 4 CSX Corporation / TCI Fund 3 4 1 / 6 Micrel 33 0 / 3 International Rectifier 18 0 / 4 Hexcel 18 0 / 2 Arrow Intl. .09 0 / 2 Atmel 54* 3 / 4 .40 1 / 4 H&R Block / Bre eden Capital 26 0 / 6 .48* 4/ 6 Motorola 16 0 / 7 .31 0 / 7 Openwave Career Education 25 0 / 4 .04 0 / 4 UbiquiTel Massey Energy / Third Point .35 0 / 3 Motient .47* 2 / 3 GenCorp / Pirate Capital .32 0 / 2 .28 0 / 2 Blockbuster / Carl Icahn .31 1 / 6 .44* 3 /6 Exar / GWA Investments .13 0 / 1 .14 0 / 1 Six Flags / Daniel Snyder .33 1 / 5 .34 1 / 5 BKF Capital / Steel Partners II .23 1 / 3 .33 1 / 3 Computer Horizons / Crescendo .03 0 / 2 .29 0 / 2 Total 13 / 73 12 / 59 Notes: letters = Candidate shareholder letters, Median correlation = The median candidate news m edia correlation coefficient value for each corporate proxy contest. Correlation ratio = The proportion of possible candidatenews media correlations per contest that attain statistical significance. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 184

184 Table 4 18. Summary of intercandidate issue and stakeholder agenda setting relationships in proxy contests Corporate proxy contest Issue salience Stakeholder salience Incumbent/Challenger Releases Releas es Letters Letters Releases Releases Letters Letters Target .92*** .33 .89*** NRG Energy .66** .52* Myers Industries PHH / Pennant Capital .91*** .54* .76*** .64** Conseco / Keith Long .98*** Biogen Idec .72*** .6 5** CSX Corporation / TCI Fund .54* .90*** Micrel .85*** .84*** International Rectifier .86*** .65** Hexcel .60** .62** Arrow International .72*** .81*** Atmel .84*** .66** .75*** .79*** H&R Block / Breeden Capital .76*** .63** .99*** .61** Motorola .14 .48* .84*** .88*** Openwave .84*** .85*** Career Education .73*** .80*** .86*** .76*** UbiquiTel Massey Energy / Third Point .45* .55* .90*** Motient .69*** .83*** GenCorp / Pirate Capital .58** .74*** .85*** Blockbuster / Carl Icahn .24 .30 .85*** .58 *** Exar / GWA Investments .62** .72*** .83*** .62** Six Flags / Daniel Snyder .64** .74*** .78*** .58** BKF Capital / Steel Partners II .63** .74*** .90*** .89*** Computer Horizons / Crescendo .79*** .75*** .73*** .69** Mean .67 .62 .81 .71 Median .71 .66 .84 .67 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 185

185 Table 4 19. Summary of medi a investor issue and stakeholder agenda setting relationships in proxy contests Corporate proxy contest Media Investor issue Media Investor stakeholder Incumbent/Challenger Median correlation Correlation ratio Median correlation Correlation ratio Target / Pershing Square .43* 4 / 7 .74*** 7 / 7 NRG Energy .81*** 6 / 7 .72*** 7 / 7 Myers Industries .81*** 1 / 1 PHH / Pennant Capital Conseco / Keith Long .51* 1 / 1 Biogen Idec .36 2 / 4 67** 4 / 4 CSX Corporation / TCI Fund .04 0 / 6 .73*** 6 / 6 Micrel .55* 1 / 2 .72*** 3 / 3 International Rectifier .55* 2 / 2 .52* 3 / 4 Hexcel .36 0 / 1 .81*** 2 / 2 Arrow International l Trust Atmel H&R Block / Breeden Capital .22 2 / 5 .76*** 6 / 6 Motorola .52* 5 / 7 Openwave .12 0 / 4 .49* 3 / 4 Career Education .38 1 / 3 .68** 4 / 4 UbiquiTel Massey Energy / Third Point .31 0 / 2 .73*** 3 / 3 Motient GenCorp / Pirate Capital .80*** 2 / 2 Blockbuster / Carl Icahn Exar / GWA Investments Six Flags / Daniel Snyder BKF Capital / Steel Partners II .31 0 / 1 .57* 2 / 3 Computer Horizons / Crescendo .55* 1 / 1 .58** 2 / 2 Total 19 / 45 61 / 66 Notes: Median Correlation = The median news media investor opinion co rrelation coefficient value for each proxy contest. Correlation Ratio = The proportion of possible news media investor opinion correlations per contest that attain statistical significance. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 186

186 Table 4 20. Summary of candidate investor issue agenda setting relationships in proxy contests Corporate proxy contest Incumbent PR Investor issue salience Challenger PR Investor issue salience Incu mbent/Challenger Releases Posts Letters Posts Releases Posts Letters Posts Target / Pershing Square .54* .19 .54* NRG Energy .41 .64** .67** Myers Industries PHH / Pennant Capital .14 .34 .13 .06 Conseco / Keith Long .18 Biogen Idec .35 .45* .77*** CSX Corporati on / TCI Fund .61** .28 .27 Micrel .73*** .82*** .78*** International Rectifier .67** .58** .57** Hexcel .60** .76*** .69*** Arrow International Atmel gos H&R Block / Breeden Capital .00 .02 .02 .37 Motorola Openwave .39 .29 Career Education .54* .54* .83*** .74*** UbiquiTel Massey Energ y / Third Point .12 .17 .10 .56** Motient GenCorp / Pirate Capital .00 .24 .09 .09 Blockbuster / Carl Icahn Exar / GWA Investments .23 .38 .34 .41 Six Flags / Daniel Snyder BKF Capital / Steel Partners II .08 .39 .58** .52* Computer Horizons / Crescendo .66** .55* .73*** .72*** Mean .38 .37 .41 .46 Median .40 .39 .44 .52 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters Posts = investor me ssage board posts, p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 187

187 Table 4 21. Summary of candidate investor stakeholder agenda setting relationships in proxy contests Corporate proxy contest Incum bent PR Investor stakeholder salience Challenger PR Investor stakeholder salience Incumbent/Challenger Releases Posts Letters Posts Releases Posts Letters Posts Target .35 .34 NRG Energy .43* .28 .05 Myers Industries .00 PHH / Pennant Capital .29 .39 .30 .49* Conseco / Keith Long .10 .04 Biogen Idec .13 .27 .40 CSX Corporation / TCI Fund .31 .33 .39 Micrel .47* .34 .39 Internat ional Rectifier .05 .06 .41 Hexcel .21 .30 .60** Arrow International Atmel H&R Block / Breeden Capital .39 .26 .41 .57* Motorola .42 .33 .30 .46* Openwave .01 .05 Career Education .10 .18 .17 .05 UbiquiTel Massey Energy / Third Point .23 .15 .34 Motient GenCorp / Pirate Capital .39 46* .24 .34 Blockbuster / Carl Icahn Exar / GWA Investments Six Flags / Daniel Snyder BKF Capital / Steel Partners II Computer Horizons / Crescendo .14 .28 .24 .33 Mean .24 .29 .27 .37 Median .23 .29 .3 0 .37 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters Posts = investor message board posts, p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 188

188 A B Figure 4-1. Cross lag correlations for issue salience among candidate news releases and newspaper coverage. A) Blockbuster releases and newspaper coverage, B) CSX releases and newspaper coverage, C) TCI releases and newspaper coverage, D) Motorola releases and newspaper coverage, E) Exelon releases and newspaper coverage, F) Target releases and newspaper coverage.

PAGE 189

189 C D E Figure 41 Continued

PAGE 190

190 F Figure 41 Continued

PAGE 191

191 A B Figure 42. Cross lag correlations for issue salience among candidate news releases and news wire coverage. A) Arrow releases and wire coverage, B) Atmel releases and wire coverage, C) Perlegos releases and wire coverage, D) Blockbuster releases and wire coverage, E) Career Education releases and wire coverage, F) Bostic releases and wire coverage, G) CSX releases and wire coverage, H) TCI releases and wire coverage, I) Breeden releases and wire coverage, J) OSS releases and wire coverage, K) Intl. Rectifier releases and wire coverage, L) Micrel releases and wire coverage, M) Motient release s and wire coverage, N) Obrem releases and wire coverage, O) Highland releases and wire coverage, P) Motorola releases and wire coverage, Q) Exelon releases and wire coverage, R) Openwave releases and wire coverage, S) Six Flags releases and wire coverage, T) Target releases and wire coverage.

PAGE 192

192 C D E Figure 42 Continued

PAGE 193

193 F G H Figure 42 Continued

PAGE 194

194 I J K Figure 42 Continued

PAGE 195

195 L N Figure 42 Continued M

PAGE 196

196 P Q Figure 42 Continued O

PAGE 197

197 R S T Figure 42 Continued

PAGE 198

198 U Table 42 Continued

PAGE 199

199 A B C Figure 43. Cross lag correlations for issue salience among candidate shareholder letters and newspaper coverage. A) Motorola letters and newspaper coverage, B) Icahn letters and newspaper coverage, C) Exelon letters and newspaper coverage.

PAGE 200

200 A B Figure 44. Cross lag correlations for issue salience among candidate shareholder letters and news wire coverage. A) Biogen letters and wire coverage, B) Icahn letters and wire coverage, C) Career Education letters and wire coverage, D) Bostic letters and wire coverage, E) H&R Block letters and wire coverage, F) Breeden letters and wire coverage, G) Intl. Rectifier letters and wire coverage, H) Third Point letters and wire coverage, I) Micrel letters and wire coverage, J) Motient letters and wire coverage, K) Motorola letters and wire coverage, L) Icahn letters and wire coverage, M) Exelon letters and wire coverage, N) Six Flags letters and wire coverage.

PAGE 201

201 C D E Figure 44 Continued

PAGE 202

202 F G H Table 44 Continued

PAGE 203

203 I J K Table 44 Continued

PAGE 204

204 L M N Figure 44 Continued

PAGE 205

205 A B Figure 45. Cross lag correlations for stakeholder salience among candidate news releases and newspaper coverage. A) Blockbuster releases and newspaper coverage, B) CSX releases and newspaper coverage, C) Motorola releases and newspaper coverage, D) Exelon releases and newspaper coverage, E) Target releases and newspaper coverage

PAGE 206

206 C D E Figure 45 Continued .

PAGE 207

207 A B Figure 46. Cross lag correlations for stakeholder salience among candidate news releases and news wire coverage. A) Arrow releases and wire coverage, B) Perlegos releases and wire coverage, C) Steel releases and wire coverage, D) Blockbuster releases and wire coverage, E) Career Education releases and wire coverage, F) Bostic releases and wire coverage, G) CSX releases and wire coverage, H) OSS releases and wire coverage, I) Micrel releases and wire coverage, J) Obrem releases and wire coverage, K) Highland releases and wire coverage, L) Motorola releases and wire coverage, M) Exelon releases and wire coverage, N) Openwave releases and wire coverage, O) Target releases and wire coverage.

PAGE 208

208 C D E Table 46 Continued

PAGE 209

209 F G H Table 46 Continued

PAGE 210

210 I J K Table 46 Cont

PAGE 211

211 L M N Table 46 Continued

PAGE 212

212 O Table 46 Continued

PAGE 213

213 A B Figure 47. Cross lag correlations for issue salience among newspaper coverage and investor opinion. A) Career Education Bostic newspaper coverage and inv e stor opinion, B) CSX TCI newspaper coverage and investor opinion, C) NRG Exelon newspaper coverage and investor opinion, D) Target Pershing Square newspaper coverage and investor opinion.

PAGE 214

214 C D Figure 47 Continued

PAGE 215

215 A B Figure 48. Cross lag correlations for issue salience among news wire coverage and investor opinion. A) Biogen Icahn wire coverage and investor opinion, B) Career Education Bostic wire coverage and investor opinio n, C) CSX TCI wire coverage and investor opinion, D) H&R Block Breeden wire coverage and investor opinion, E ) Massey Third Point wire coverage and investor opinion, F ) Micrel Obrem wire coverage and investor opinion, G ) NRG Exelon wire coverage and investor opinion, H) Openwave Harbinger wire coverage and investor opinion, I ) Target Pershing Square wire coverage and investor opinion.

PAGE 216

216 C D E Figure 48 Continued

PAGE 217

217 F G H Figure 48 Continued

PAGE 218

218 I Figure 48 Continued

PAGE 219

219 A B Figure 49. Cross lag correlations for stakeholder salience among newspaper coverage and investor opinion. A) Career Education B ostic newspaper coverage and investor opinion, B) CSX TCI newspaper coverage and investor opinion, C) NRG Exelon newspaper coverage and investor opinion, D) Target Pershing Square newspaper coverage and investor opinion.

PAGE 220

220 C D Figure 49 Continu ed

PAGE 221

221 A B Figure 410. Cross lag correlations for stakeholder salience among newspaper coverage and investor opinion. A) Biogen Icahn wire coverage and investor opinion, B) CSX TCI wire coverage and investor opinion, C) H&R Block Breeden wire coverage and investor opinion, D) Hexcel OSS wire coverage and investor opinion, E) Massey Third Point wire coverage and investor opinion, F) Micrel Obrem wire coverage and investor opinion, G) NRG Exelon wire coverage and investor opinion, H) Openwav e Harbinger wire coverage and investor opinion, I) Target Pershing wire coverage and investor opinion.

PAGE 222

222 C D E Figure 410 Continued

PAGE 223

223 F G H Figure 410 Continued

PAGE 224

224 I Figure 410 Continued

PAGE 225

225 A B Figure 411. Cross lag correlations for i ssue salience among candidate news releases and investor opinion. A) Target releases and investor opinion, B) Exelon releases and investor opinion, C) CSX releases and investor opinion, D) TCI releases and investor opinion, E) Micrel releases and investor opinion, F) Obrem releases and investor opinion, G) Breeden releases and investor opinion, H) Openwave releases and investor opinion, I) Career Education and investor opinion, J) Bostic releases and investor opinion, K) Computer Horizons releases and investor opinion.

PAGE 226

226 C D E Figure 411 Continued

PAGE 227

227 F G H Figure 411 Continued

PAGE 228

228 I J K Figure 411 Continued

PAGE 229

229 A B Figure 412. Cross lag correlations for issue salience among candidate shareholder letters and investor opinion. A) NRG letters and investor opinion, B) Biogen letters and investor opinion, C) Icahn letters and investor opinion, D) Micrel letters and investor opinion, E) H&R Block letters and investor opinion, F) Breeden letters and investor opinion, G) Career Education letters and investor opinion, H) Bostic letters and investor opinion, I) Third Point letters and investor opinion.

PAGE 230

230 C D E Figure 412 Continued

PAGE 231

231 F G H Figure 412 Continued

PAGE 232

232 I Figure 412 Continued

PAGE 233

233 A B Figure 413. Cross lag correlations for stake holder salience among candidate news releases and investor opinion. A) Target releases and investor opinion, B) Exelon releases and investor opinion, C) CSX releases and investor opinion, D) Micrel releases and investor opinion, E) Obrem releases and inves tor opinion, F) OSS releases and investor opinion, G) Openwave releases and investor opinion, H) Career Education releases and investor opinion, I) Bostic releases and investor opinion, J) Crescendo releases and investor opinion.

PAGE 234

234 C D E Figure 413 C ontinued

PAGE 235

235 F G H Figure 413 Continued

PAGE 236

236 I J Figure 413 Continued

PAGE 237

237 A B Figure 414. Cross lag correlations for stakeholder salience among candidate shareholder letters and investor opinion. A) Biogen letters and investor opinion, B) Icahn letters and investor opinion, C) CSX letters and investor opinion, D) Micrel letters and investor opinion, E) Breeden letters and investor opinion, F) Career Education letters and investor opinion, G) Bostic letters and investor opinion, H) Third Point letters and investor opinion.

PAGE 238

238 C D E Figure 414 Continued

PAGE 239

239 F G H Figure 414 Continued

PAGE 240

240 CHAPTER 5 DISCUSSION Summary of Findings The days of large shareholders passively sitting back and supporting the questionable strategies of a seemingly underper forming corporate management and board of directors are over. The rise of shareholder activism in the U.S. over the past decade has resulted in a record number of corporate proxy contests, the most expensive and aggressive form of shareholder activism, being held annually (Laide & Mallea, 2009). These political style election campaigns typically feature dueling financial public relations messages between the targeted corporation (incumbent candidate) and activist shareholder group (challenger candidate) as they try to persuade shareholders to support their nominee or nominees for election to the board of directors (Carney & Jorden, 1993; Marken, 2005; Wilcox, 2004). As in a political election setting, these efforts include each campaign trying to shape media coverage of the contest to their advantage (Bagli, 1997; Cheney, 2001; A. Davis, 2002; L. Davis, 1989; Newman, 1983). The purpose of this study is to help build public relations theory and inform the practice of investor relations and financial public relations in this burgeoning domain by examining the role of mass mediatedinfluence during a contested corporate election from an agendabuilding and agendasetting perspective (e.g. McCombs & Shaw, 1972). Specifically, this investigation extensively analyzed the transfer of issue and stakeholder salience among the information subsidies (news releases and shareholder letters) of the competing campaigns, financial news media coverage (business oriented newspaper and news wire stories), and investor opinion (investor stock message board postings) in the 25 largest U.S. proxy contests for board representation or control that

PAGE 241

241 went to a shareholder vote over a fiveyear period (20052009). Figure 51 displays the theoretical map with the hypotheses and research questions which guided this study, while Table 51 summarizes the findings for each set of hypotheses and research questions. Nearly a century ago, Walter Lippmann (1922), the intellectual forefather of agendasetting theory, described the news media as being like the beam of a searchlight that moves restlessly about, bringing one episode and then another out of darkness into vision (p. 364). Applying this analogy to business media coverage during proxy contests, this study found little to no indication th at the sheer number of information subsidies disseminated by campaigns during contests influenced how much attention the financial press decided to shine on a contest. Rather, the system level analysis portion of this study revealed that the frequency of stories written about a proxy contest was significantly correlated with the size of the firm (incumbent candidate) under assault from the activist shareholder group (challenger candidate). In general, the data indicated that the larger the targeted corporation, as measured by its annual revenue, and, to a lesser extent, stock market capitalization, the more raw media attention the proxy fight received. While financial public relations efforts did not seem to materially influence how brightly the media searchlight decided to shine overall on a contest this study found solid evidence that these efforts could influence how sharply the specific issues within each contest were brought into focus by the press. Said another way, other factors, perhaps traditional journalistic news values and cross checking among outlets, rather than financial public relations efforts, seemed to dictate whether or not a media outlet

PAGE 242

242 decided to cover a proxy contest and how much raw media attention the fight received. (Proxy figh ts involving large firms would arguably have greater news value to a business news mediums audience than events involving smaller companies.) Upon deciding to cover a contest, it appears that business journalists did then turn to the issue salience cues embedded in the news releases and shareholder letters disseminated by the dueling campaigns for some guidance in prioritizing contest issues in their coverage. These salience cues work both ways, though, as this study found that campaigns generally responde d to rather than influenced the importance accorded specific stakeholders by journalists in coverage. The candidateto media agendabuilding portion of this study found that there was overall a significant correspondence during the contests between the issues highlighted in media coverage and those issues emphasized in candidatecontrolled information subsidies (incumbent news releases Mdn correlation = .51; challenger news releases Mdn correlation = .50; incumbent shareholder letters Mdn correlation = .48; challenger shareholder letters Mdn correlation = .50). Partial correlations revealed that releases were more strongly linked with media coverage than letters for issue salience. The crosslagged correlation analyses found that there was generally mutu al influence in the candidateto media relationship during contests regarding issue salience. It is worth noting that campaigns seemed to have more success in unidirectionally influencing issue coverage in newspapers, rather than with news wire services. Unlike the medium strength issue salience findings, there was overall only a weak correspondence between the stakeholder groups highlighted in media coverage and those groups emphasized in candidatecontrolled information subsidies (incumbent releases Mdn

PAGE 243

243 correlation = .31; challenger releases Mdn correlation = .30; incumbent letters Mdn correlation = .25; challenger letters Mdn correlation = .33). The cross lag analyses of this relationship found that candidate letters in particular tended to respond to the stakeholder group emphases first established in news media content. Gershon Kekst, the founder of Kekst & Co., one of the largest financial public relations firms in the U.S., was once asked if he would rather be on offense (challenger) or defense (incumbent) during a corporate takeover battle (Kaback, 1999). Kekst responded by saying that, when youre working on the defense, you have a significant public relations advantage (Kaback, 1999, para. 76). While this may be true during a hostile merger or acquisition situation, the data in this study did not indicate that media coverage consistently favored the agenda of one side more than the other during the proxy contests. The strength of the bond between the issue priorities found in media coverage and incumbent information subsidies (for both releases and letters), and between coverage and challenger subsidies during contests was virtually identical. Turning to stakeholder salience, neither incumbent nor challenger subsidies were overall significantly link ed with coverage. In short, at least among the largest U.S. proxy fights in recent years that went to a vote, neither side seemed to enjoy a noticeably tighter linkage with the press. In a political election setting, researchers have noted that candidates at times prefer to talk past each other and not engage in a back and forth exchange (Iyengar & McGrady, 2007; Simon, 2002). This does not appear to be the case during large contested corporate elections in the U.S., as the inter candidate agendasetting portion of this study found that candidates generally placed similar emphases on both the

PAGE 244

244 issues and stakeholders mentioned in their information subsidies. Overall, there was a significant linkage during the proxy contests between the salience of issues in the competing candidate news releases ( Mdn correlation = .71), as well as in the competing candidate shareholder letters ( Mdn correlation = .66). This linkage was even stronger among the competing candidate releases ( Mdn correlation = .84) and letters ( Mdn correlation = .67) for stakeholder salience. Of course, while strong inter candidate agendasetting linkages were found at the first level for issue and stakeholder salience, the dueling campaigns could conceivably differ at the secondlevel in how they used attributes to frame these objects. Turning to the mediato investor agendasetting portion of the study, this study found mixed support overall for the classic agendasetting hypothesis, which predicted a close correspondence during contests between the issue priorities in media coverage and the priorities discussed in investor message board postings ( Mdn correlation = .37). A much stronger linkage was detected between the stakeholder priorities found in coverage and expressed by investors ( Mdn correl ation = .71). The cross lag analyses of this relationship found that the direction of influence was generally balanced between the media and investors for issue salience, but that the media clearly led for stakeholder salience. Notably, investors did not set the media agenda for stakeholder salience during any fight. This finding dovetails with the stakeholder salience portion of the agendabuilding results, which found that the media tends to lead this aspect of the candidateto media relationship. Among the three actors (media, investors, candidates), the media seems to play the lead role in determining the importance of stakeholders during contests.

PAGE 245

245 The candidateto investor agendasetting portion of the study revealed that the issue priorities found in candidate information subsidies were overall moderately associated with the priorities expressed in investor discussion, while the stakeholder priorities among the subsides and investors were generally not significantly associated. The strength of association between candidate releases and investor opinion (incumbent releases Mdn correlation = .40; challenger releases Mdn correlation = .44) and candidate letters and investor opinion (incumbent letters Mdn correlation = .39; challenger letters Mdn correlati on = .52) for issue salience was similar. Analyzing these relationships by candidate type, challenger letters overall enjoyed a stronger relationship with investor opinion than did incumbent letters, while with candidate news releases, the strength of the association by candidate type was very similar. Consistent with what was found in the candidateto media relationships for stakeholder salience, investors generally did not seem to take meaningful stakeholder salience cues from the priorities embedded in the candidate releases and letters (incumbent releases Mdn correlation = .23; challenger releases Mdn correlation = .30; incumbent letters Mdn correlation = .29; challenger letters Mdn correlation = .37). For both issue and stakeholder salience, the cross -l ag correlation analyses overall revealed no dominant direction of influence in the candidateto investor relationship. A somewhat unexpected finding of this study was that a closer bond was detected between the issue priorities expressed in candidate information subsidies and investor discussion, than between financial media coverage and investor discussion (a variant of the traditional agendasetting hypothesis). In political election settings, agendasetting researchers have generally found that the news media plays a more influential role in

PAGE 246

246 shaping public perceptions than the campaigns do directly themselves (McCombs, 2006). One possible explanation for the stronger candidateto investor issue linkage ( Mdn correlation value range of .39 to .52) than media to investor issue linkage ( Mdn correlation = .37) may because the investor opinion indicator was based upon investor discussion on online stock message boards. These message board participants would fit the definition of an active informationseeking p ublic that was likely following the information flow on the proxy contest very closely, including directly reading the information subsidies provided online by the dueling candidates (either on the campaign Web sites or aggregated by major business news Web sites). It may be the financial press is more influential in affecting issue salience during contests among less active publics, such as lower involvement investors, customers, suppliers, retirees, and even some employees, for which the press serves as t he leading information source on the contests. This conjecture clearly should be evaluated empirically in the future. The finding that the issue salience expressed in investor discussion was linked more closely with candidate subsidies than with media coverage is also consistent with the situational theory of publics (Grunig, 1997; Grunig & Hunt, 1984). A public high in problem recognition and level of involvement that feels unconstrained tends to engage in active, informationseeking behavior, whereas a lo wer problem recognition, less involved, more constrained public tends to engage in more passive, informationprocessing behavior. With this in mind, interpersonal and organizational communication (i.e. information provided by the source directly to the public) is generally most impactful among higher involvement publics, and media relations and advertising is often used for reaching lower involvement publics (Smith, 2009). According to J. Grunig, L. Grunig,

PAGE 247

247 and Dozier (2006), some people do indeed learn from the media in some situations, but the media are not the solution to every public relations problem. Rather, public relations is a process in which organizations must communicate with publics in different ways in different situations (pp. 2223). In the situation of a proxy fight, active publics likely do not ignore the media agenda, but direct communication efforts by the campaigns, particularly through interpersonal channels that provide opportunities for enhanced dialogue (meetings and conference calls), may play the greatest role in affecting cognitions, attitudes, and behavior s. The final section of the analysis explored potential consequences of agendasetting during contested corporate elections. According to Iyengar and McGrady (2005), in a political election setting, all else being equal, candidates who manage the media more effectively are the candidates who win on Election Day (p. 225). While the influence of news media may be more muted in corporate elections, the agendasetting consequence s portion of the study did find at an individual contest level that the candidate with the stronger candidateto media agenda building and candidateto investor agendasetting relationship won more often than not. This finding in no way implies causality, but suggests that shaping media coverage and investor opinion may at least be a contributing factor in a successful proxy fight outcome. Previous research indicates that voting decisions during fights and hostile takeovers are driven for the most part by economic considerations with financial public relations playing a supporting role in communicating the economic value, or supposed lack of value, to a firms shareholders (A. Davis, 2002, Newman, 1983).

PAGE 248

248 While only shareholders directly pull the voting lever on a proxy contest outcome, it is important to keep in mind that there are other key publics that may be influenced by the financial public relations efforts and media coverage that they are exposed to during a contest. Practitioners have suggested that employee morale may rise or fall, customer and supplier attitudes may shift, and governmental or judicial opinions may be influenced as a result of the war of words between campaigns that plays out in part through the media (Bagli, 1997; Cheney, 1983; L. Davis, 1989; Kaback, 1999; Marken, 2005). In other words, analyzing shareholder voting results may only tell part of the story in terms of gauging the potential consequences of agenda building and agendasetting during a proxy contest. Future research into influence at a mass mediatedlevel during proxy fights should attempt to measure the opinions and attitudes of these other key publics to see if there is indeed empirical evidence to support these supposed linkages Theoretical Contributions This study makes a variety of important contributions towards advancing knowledge within agendabuilding and agendasetting theory, as well as within the financial public relations and investor relations literature as a whole. Starting with the theory specific contributions, a recurring criterion in assessing the overall value of a theory is in gauging its theoretical scope or generality (Chafee & Berger, 1987; Littlejohn, 2002; Shoemaker et al., 2004). A theory high in scope can be applied to a variety of different settings. There are several hundred studies which have demonstrated agendabuilding and agendasetting effects within the public sphere, often during political election campaigns (McCombs, 2006; Dearing & Rogers, 1996). A smaller body of work has identified agenda building effects within the corporate sphere.

PAGE 249

249 This current study gets directly at evaluating the theoretical scope of this perspective by testing for agendabuilding and agendasetting effects within an entirely new setting corporate proxy contests. The overall support this investigation found for the core agendabuilding and agendasetting propositions the transfer of salience between the source agenda and the media agenda, and between the media agenda and the public agenda, respectively sugges ts that this perspectives theoretical scope and applicability extends well beyond its traditional context of political communication. In addition to exploring agendabuilding and agendasetting theory within a new setting this investigation also helps introduce new actors a new information subsidy type and new sources of business journalism into the literature. While corporations and social activists have received some attention in previous agendabuilding and agendasetting scholarship, this is one of the first studies to focus on the shareholder activist as a key actor in the corporate agendabuilding and agendasetting process. Further, the inclusion in this study of candidate shareholder letters (known during proxy contests as fight letters) int roduces a new corporate information subsidy type into the literature. Finally, given that most prior agendabuilding and agendasetting research has been within the public sphere, there has been extensive research into the effects of media content generated by general interest news organizations, such as the major metropolitan newspapers and the evening news programs. Business news organizations have received only minimal attention in the media effects literature. This studys inclusion of a mix of both established and underexplored business journalism sources helps provide needed diversification in this area.

PAGE 250

250 This study is also unique in that it simultaneously explores multiple stages of agendasetting theory basic agendasetting effects, contingent cond itions, sources of the media agenda, and consequences of agendasetting effects all within one investigation. This approach provides for a more comprehensive examination of the agendabuilding and agendasetting process and a potential tighter integration of the theorys components (McCombs, 2005; Tan & Weaver, 2007). Most prior research concentrates on exploring either the source to media relationship or mediato public relationship, but not these two stages in tandem along with an assessment of potential moderators and attitudinal or behavioral consequences of these processes The multi stage analysis approach undertaken in this study is akin to taking several steps back to fully take in a large painting, rather than standing up close to the canvas and only observing a portion of it. The insight that the financial press appears to be the dominant actor in the candidatemediainvestor triumvirate in establishing stakeholder salience during contests could not be gleaned unless this examination had undertaken a multiple stage approach. Hopefully, this study will encourage future research that examines multiple stages simultaneously. In addition to demonstrating the presence of first level agendabuilding and agendasetting effects using the traditional issue agenda construct, this investigation introduces a new construct the stakeholder agenda into this research tradition. In the abstract, an agenda consists of a set of objects communicated in a hierarchy of importance at a point in time (Dearing & Rogers, 1996). Often, these objects have been operationalized as issues, but the agendasetting metaphor of a set of objects competing for attention need not be confined to issues (McCombs & Shaw, 1993). Far

PAGE 251

251 more than issues are discussed and emphasized or deemphasized by public relations professionals, journalists and the public on a daily basis. An organizations various stakeholder groups represent another set of objects that receive varying levels of attention in organizationprovided information subsidies, media coverage of the organization, and public discussion of the organization. In general, this study found evidence of a transfer of stakeholder salience between the media agenda and investor agenda during contests, as well between the information subsidies of the competing campaigns. The stakeholder construct provides a novel new way to test for first level effects in a diverse range of organizational settings, since every organization, regardless of its industry type, generally has multiple stakeholders or constituents. Besides corporate agendasetting, this construct could have applicability in political, crisis, non profit, and public information/social action contexts. At its core, the agenda building and agendasetting framework traces the format ion and transfer of salience from one agenda to another. Tracking and understanding the sources which affect the perceived importance (salience) of an issue is directly applicable to adjacent public relations theory areas which predict the degree to which publics might become active based in part on problem recognition and concern, such as the situational theory of publics (Grunig & Hunt, 1984) and issues management (Hallahan, 2001). A conscious effort by public relations scholars to better integrate these seemingly complimentary research streams could help advance public relations theory as a whole. Other areas of communication research have made such attempts. Within the health communication field, the stages of change model (Propochaska, DiClemente, & Norcross, 1993; Slater, 1999) has been used to integrate

PAGE 252

252 various media effects theories, including agendasetting, in an attempt to provide an overarching framework for practitioners to use in guiding campaigns. A similarly ambitious effort within the public relations field to organize complimentary, but often isolated, public relations theories under a cohesive, general framework would at the very least spark healthy debate among scholars in various theoretical silos. Finally, this study makes a general contribution to advancing scholarship in financial public relations and investor relations as a whole. A review of the extant literature revealed that the scholarship on financial public relations and investor relations is relatively sparse, and tends to be atheoretical. In many ways, the state of research in these specialty areas of public relations is similar to the general state of public relations research 25 years ago (Ferguson, 1984; Sallot et al., 2003). This current investigation introduces two established theoretical frameworks in agendabuilding and agendasetting into these understudied areas. Exploratory research free of theoretical constraints has its benefits, but to truly build a cohesive body of knowledge within these areas that will provide situation specific prediction and explanation, it is likely necessary that scholars coalesce their efforts around existing or new theoretical perspectives. T heories of selective presentation and audience perc eption, such as agendabuilding, media framing, and m edia priming, are but one theoretical path that researchers can follow in analyzing aspects of the investor relations and financial public relations functions. What is most important is that future scholarship in these areas embrace theory.

PAGE 253

253 Practical Implications Implications for Communication Professionals This study places empirical weight behind the assertions of practitioners that investor relations and financial public relations efforts can play a significant role in shaping details of media coverage during a proxy contest or hostile takeover attempt. Overall, there was a correspondence during fights between the issue priorities emphasized in candidateprovided information subsidies and those found in news media content. Media sociologist Gans (1979) has described the source journalist relationship as being like a dance in which, more often than not, sources do the leading (p. 116). This was not the case, though, during the proxy fights, as reciprocal influence among the candidate and media agendas was the norm for issue priorities and the media not the candidates often led in building stakeholder priorities. Financial public relations professionals that approach media relations during fights with a mere oneway placement of messages mentality are likely to be disappointed. Even an expert dancer used to leading can wear out their welcome if they dont also listen to the concerns of their partner. This research also helps inform the specific media relations activities of financial public relations and investor relations professionals during proxy contests. While some issues, such as discussion of acquisitions and shareholder value, ranked high across the board in media coverage, other issues seemed to resonate more with particular m edia outlets. For example, the Wall Street Journal mentioned the issue of corporate governance more often than other outlets, while The New York Times seemed to be more interested in executive compensation than other outlets. The AP and The New York Times placed more emphasis than other business news gatherers on the issue of

PAGE 254

254 corporate strategy. Finally, Bloomberg News and the Financial Times placed more focus than their peers on the issue of management and board experience. Turning to the mentions of stakeholders in media coverage, as with the issue mentions, the emphases across media outlets were overall similar, but there were some individual differences in the coverage. For example, Bloomberg News and the Financial Times placed greater emphasis on the court than other outlets, the Wall Street Journal and AP placed greater emphasis on customers, and the news wire services devoted more attention to the proxy advisors than the newspapers. This data into the presentation of issues and stakeholders in media coverage during proxy fights can be used by practitioners to provide insights into which issues and which stakeholders a particular outlet is likely to be most interested in during a fight. With this information in hand, practitioners can better formulate story pitches towards a particular outlet and better brief management in advance of an interview with a reporter from a particular outlet. In the Internet age, candidatecontrolled information subsidies are now widely accessible to the general public through campaign Web sites (Tedesco, 2005b). In a corporate election setting, campaign news releases, shareholder letters (i.e. fight letters), and other communication materials are not only available on specialized fight Web sites, but are generally picked up in their entiret y on highly trafficked financial news Web sites, such as Yahoo! Finance and Google Finance. Journalists are no longer the only intended audience for campaign news releases. Interestingly, the data in this current study indicates that the issue priorities expressed in investor discussion during contests was more strongly linked with the priorities exhibited in candidate news

PAGE 255

255 releases than in candidate shareholder letters (perhaps because these releases tend to be shorter and more widely accessible than the l etters) For financial public relations and investor relations professionals, this finding underlines that campaign news releases during contests are likely to shape the perceptions of multiple audiences (i.e. reporters and editors, as well as shareholders and other stakeholders) and, therefore, should be written accordingly. While the news release is but one information subsidy type of many in a professional communicators toolbox, it is important to recognize that, at least within the corporate sphere and business news arena, the potential audience reach and impact of a news release is arguably greater than it has ever been. Implications for Business Journalists This study also has at least two implications for business editors and reporters. On the plus s ide, this investigation found that business news media coverage of proxy contests, at least as it related to corporate issues and st akeholders, appeared balanced. Overall, n either side during the contests enjoyed a noticeably stronger bond with the issue a nd stakeholder priorities exhibited in media coverage. This balance augurs well for the functioning of corporate democracy. T his study also found a high degree of similarity in coverage of the issues and stakeholders across the different media outlets duri ng the fights. While this redundancy in coverage is not necessarily a bad thing, and crosschecking is a natural part of the news gathering process, business journalists should be aware of this pattern and make sure that they are not inadvertently conformi ng to the priorities set by other outlets. On the one hand, a converged business media agenda helps focus the attention of corporate publics on a similar set of concerns, which in turn could lead to consensus building. On the flipside, this same redundancy risks suppressing issues and stakeholders that deserve more attention.

PAGE 256

256 Implications for Financial Regulators Finally, this study also has a specific implication for the regulators of the U.S. financial markets, namely the Securities and Exchange Commissi on (S.E.C.). This analysis of mass communication during major proxy contests indicates that the proxy mechanism has generally been successful in forcing publicly held firms and disgruntled shareholders to publicly address and debate a similar set of concer ns In general, the data revealed that there was a fairly high level of agreement among the actors (competing campaigns, investors, and media) across the contests regarding which issues were most important. While efforts by the S.E.C. to lower the cost and burden of shareholder initiated proposals and director nominations is good for corporate democracy, new disclosure regulations should not become so onerous that they inadvertently discourage a free flow of communication from campaigns to shareholders and media. Limitations As with any study, there are several limitations in particular that should be taken into account when evaluating the results of this investigation and weighing the conclusions. A brief discussion is provided for each of the primary limitations identified by the investigator. While the use of multiple internal replications for most of the hypotheses and research questions helps enhance the external validity of the findings, these results should not necessarily be generalized to all proxy contests held. Given that the hypotheses and research questions in this study necessitated the use of a purposive sample of the largest U.S. proxy contests that went the distance (proceeded to a shareholder vote) over a multi year period, these findings are not necessarily applicable

PAGE 257

257 to U.S.based contests that were settled before going to a shareholder vote. In addition, the relationships identified in the results may not be applicable to U.S. based contests that do go the distance, but instead involve s malland medium sized firms. Finally, given that this studys sample frame consists of solely U.S. based publicly held companies, these results should not be generalized to contested corporate elections held in other countries. This investigation made use of multiple indicators whenever possible to enhance reliability (Chaffee, 1991). For example, the priorities on the media agenda were measured using seven media outlets (newspapers and news wire services) and the priorities on the candidate agendas were measured using two information subsidy types (news releases and shareholder letters). However, the priorities on the investor agenda were based on only one source of investor opinion, archived investor message board posts from Yahoo! Finance, since other surrogate sources of investor opinion were not available for the full fiveyear sample time period. While Yahoo! Finance is the largest U.S. financial news and research Web site (ComScore, 2008), the use of parallel posts from other discussion sites would have enhanced this measures reliability. On a related note, the priorities articulated within investor message board posts are not necessarily representative of investor opinion as a whole for a proxy fight. The individuals posting to message boards are by the nature of their actions an active public. Their priorities could be different than investors who are less engaged in following and discussing the fight. Further, this surrogate indicator of the investor agenda may be more representative of the opinions of vocal individual investors than of institutional investors, whom may be less likely to actively participate in discussions on

PAGE 258

258 stock message boards. These caveats should be taken into account when assessing the mediato investor and candidateto invest or agendasetting results. As previously discussed, a series of major print news media outlets were selected to represent the media agenda since practitioners indicated during the pretest that the business oriented national newspapers and wire services are among the most influential sources during proxy contests. In addition, transcripts over the full fiveyear sample period were not readily available for business oriented cable news outlets, such as CNBC, Bloomberg Television, or FOX Business. The exclusion of cable news coverage from the study may have masked the influence on and from these outlets during proxy fights. However, this limitation is alleviated somewhat by the literature on inter media agendasetting, which has generally found that the patterns of news coverage across different types of outlets (print vs. broadcast) are similar (Boyle, 2001; Dearing & Rogers, 1996; Golan, 2006; Lopez Escobar et al., 1998; McCombs, 2006; McCombs & Shaw, 1976; Reese & Danielian, 1994; Roberts & McCombs, 1994). Turning to the data analysis, this study went beyond simply testing for associations among the hierarchies of importance on the various agendas to attempting to gauge the direction of influence in these two variable relationships. Given that the vast maj ority of content for most of the contests was generated the month before the shareholder vote, this relatively short time frame would not facilitate a more advanced timeseries analysis using statistical techniques such as an autoregressive integrated moving average (ARIMA) model or a Granger analysis. Therefore, the use of a two wave (T1 and T2) crosslagged correlational analysis was appropriate. Interpretation of the cross lags was strengthened through the use of the RozelleCampbell baseline statistic.

PAGE 259

259 Future Research While this study likely represents the most extensive analysis of the role of mass mediated influence among the major actors in corporate proxy contests to date, it still only scratches the surface regarding knowledge accumulation in this area. There are many ways to build upon and extend this research by further exploring proxy contests from an agendabuilding, agendasetting, media framing and media priming perspective, as well as by exploring contests using alternate mass communication a nd public relations theoretical lenses An analysis of proxy contests at a different level of communication, such as organizational communication or interpersonal communication, could also yield additional theoretical and practical insights (Chaffee & Berger, 1987). This study extensively tested for and found evidence of the transfer of salience at the first level (object salience) among the information subsidies of corporate candidates, news media coverage, and the online discussion of investors during proxy contests. A next logical step in this research program is to assess the transfer of salience at the secondlevel (attribute salience) among these same actors during contests (McCombs & Evatt, 1995; McCombs et al., 2000). The set of substantive candida te image attributes frequently used by agendasetting researchers during political election contests could be adapted and used in a corporate proxy fight context (Kiousis et al, 2006; McCombs et al., 1997; McCombs et al., 2000). It is also important to as sess affective attribute agendasetting in future research in this area to understand the rule of tone in framing issues and stakeholders. For example, two competing campaigns may emphasize similar issues, but could place very different emphases on the sub stantive and affective attributes used to define and describe these issues.

PAGE 260

260 This study focused on the largest proxy contests held in the U.S. over a fiveyear period (20052009) that went to a shareholder vote. Replications are needed that test the same agenda building and agendasetting relationships in U.S. proxy contests involving small and medium sized companies. Similarly, replications should also be conducted in international settings. A hallmark of agendasetting research in the public sphere over the years has been replications conducted in many countries around the world, thereby providing a stronger empirical foundation across time and place for the theorys core propositions (McCombs et al., 1997). Western Europe, South Korea, and Japan have all seen a rise in shareholder activism and proxy contests in recent years, and could serve as settings for replications of the current investigation. This investigation adds to recent agenda building research (e.g. Kiousis et al., 2009) which compares the r elative influence of different types of source controlled information subsidies in shaping media coverage during a campaign. While this study focused on comparing the relative influence of candidate news releases and shareholder letters during proxy contests, there is several additional information subsidies, which should warrant attention in future research. Based on the pretest interviews with financial public relations practitioners, additional subsidies used by campaigns during proxy fights to shape coverage and investor opinion include advertisements, white papers, slide presentations, proxy statements, brochure mailers, executive memos, interview transcripts, conference call transcripts and specialized fight Web sites. There are also additional actors involved in proxy contests that should be explored in future research. Two intercessory publics of interest to publicly held corporations and

PAGE 261

261 their investor relations and financial public relations professionals are Wall Street stock analysts and the proxy advisory firms, such as ISS RiskMetrics, Glass Lewis, and Proxy Governance. Wall Street analyst reports are read by investors, and analysts are frequently quoted by the financial press. Corporations perceive analysts and the financial press as influencing their stock prices (Laskin, 2006, 2009). What role does the priorities exhibited in analyst reports and during interviews with reporters play in potentially influencing the priorities in media coverage and investor opinion? Similarly, what role do the reports issued and interviews given by proxy advisory firms play in potentially shaping coverage and opinion? Proxy advisory firms recommend to professional investors, such as mutual fund and pension fund managers, how to vote on proxy matters, such as contested board seats. Interestingly, similar to political campaigns, the candidates during proxy fights selectively incorporate into their news releases and letters quotes from analysts and advisors which are favorable to their issue positions. Future research should attempt to measure investor opinion of the perceived importance of issues and stakeholders during proxy fights using probability based telephone or Web survey data, rather than surrogate measures of the investor agenda. A difficulty with collec ting investor opinion data via survey is that it requires the cooperation of the corporation to gain access to its shareholder roster at the time of the contest or soon thereafter In the continued absence of conventional investor opinion data, researchers should explore additional ways, besides stock message boards such as Yahoo! Finance, to compile a surrogate investor agenda. New online sources of searchable and timestamped investor discussion include Stock Twits, a Twitter based

PAGE 262

262 platform for investors to discuss stocks, and SeekingAlpha, a network of blogs primarily written by professional investors. Future research could trace possible inter media influence among investor discussion on message boards, blogs, and Twitter. Are the priorities articulated by investors on the different online platforms similar? If so, does discussion on one platform tend to drive the discussion on other platforms? These questions could be evaluated by company size, industry, and country of origin. Supplemental analysis in this study revealed evidence of inter media agendasetting among the financial news media outlets at a system level during contests. This is consistent with inter media agendasetting research of topics in the public sphere, which has found that the patterns of news coverage from one news medium to the next tend to be convergent rather than divergent (Boyle, 2001; Dearing & Rogers, 1996; Golan, 2006; Lopez Escobar et al., 1998; McCombs, 2006; McCombs & Shaw, 1976; Reese & Danielian, 1994; Roberts & McCombs, 1994). Replications of this finding are needed in additional corporate settings. Assuming there is redundancy in the salience accorded topics across the financial press, the next question becomes, which business news medium is most influential in setting the overall business news agenda? In the public sphere, The New York Times has often played this role (Danielian & Reese, 1989; Mazur, 1987; Ploughman, 1984; Reese & Danielian, 1994). In the current study, practitioners indicated during the pretest intervie ws that The Wall Street Journal was the most influential outlet in the corporate sphere, followed by The New York Times Besides national newspapers and wire services, future research in this area should incorporate the major business magazines ( Forbes Fo rtune Business Week The

PAGE 263

263 Economist ), cable business news networks (CNBC, FOX Business, Bloomberg Television), and business news Web sites (TheStreet.com and MarketWatch.com). An important point in the evolution of agenda setting theory was the identification of contingent conditions, most notably need for orientation (Weaver, 1977, 1980), which strengthen or weaken the effect of the mass media on public opinion. A general criticism of media effects theories, including agendasetting, is that researchers s pend too much time demonstrating evidence of the effect (the what) and not enough time focusing on the when and the why questions (Nabi & Oliver, 2009). Agendasetting has long been viewed as being fairly strong in its predictive power, but weaker in providing explanation (Chaffee & Berger, 1987). With this in mind, a greater emphasis by agendabuilding scholars in identifying and testing potential moderators of the source to journalist relationship could help address this criticism and enhance the theorys overall value. The current study contributed to this aspect of the theory by exploring the role of financial performance and activist investor experience as potential moderators during proxy fights. Borrowing from previous findings in the agenda-set ting literature, potential moderators of the source to journalist relationship in the corporate sphere may include the perceived favorability or credibility of the source (Wanta, 1991; Wanta & Hu, 1994). The situational theory of publics (Grunig, 1997; Grunig & Hunt, 1984), which predicts the communication behavior of publics and segments them based on this behavior, should be explored as a potential contingent condition that helps explain the magnitude of agendasetting effects. According to situational t heory, three antecedents (problem recognition, constraint recognition, and level of involvement) influence whether a public will engage in active, informationseeking or more passive, information-

PAGE 264

264 processing behavior on an issue or set of issues (Grunig, 1997; Grunig & Hunt, 1984). Based on situational theory and within the context of agendasetting in the corporate sphere, the agendasetting effect of the financial press should be stronger among more passive stakeholders, in which the press is a primary source of information about a firm, and weaker among more active stakeholders, in which the press is but one of a variety of sources. Examining the role of the communication behavior of a stakeholder (based on the situational theory variables) as a potential moderator of sourceto public and mediato public agendasetting relationships would seem to provide an opportunity to both advance needed research on contingent conditions and take a step towards theoretical integration among these two well established research traditions. Looking beyond agendabuilding and agendasetting, there are a variety of other theoretical frameworks that could provide future guidance on understanding the role of mass communication during proxy contests. In addition to the situational theory of publics (Grunig & Hunt, 1984), several public relations theories worth considering include excellence theory, dialogic communication theory, and contingency theory. Excellence theory argues that activism provides the impetus for excellent public relations (Grunig, Grunig, & Dozier, 2006, p. 56). Does shareholder activism indeed stimulate investor relations departments to become excellent? Dialogic communication explores the relationshipbuilding potential of organizational Web sites (Kent & Taylor, 1998). How is this theorys dialogic communication principles applied to fight Web sites produced by challenger and incumbent candidates during proxy contests? Finally, contingency theory encompasses a variety of variables which predict whether a n organization will take a stance of advocacy or accommodation with its public relations

PAGE 265

265 efforts (Cancel, Cameron, Sallot, & Mitrook, 1997). How predictive are these variables of the stance that a corporation will take when communicating with an activist i nvestor? Each of these theories would seem to offer promising routes for future research. Finally, future research should analyze the communicative elements of proxy contests at additional levels of analysis, such as organizational and interpersonal (Chaff ee & Berger, 1987). The pretest interviews with practitioners indicated that financial public relations is conducted during contests and activist investor campaigns at both a mass mediated and more interpersonal level. In addition to mass mediated efforts campaigns conduct road shows consisting of teleconferences and inperson meetings with small groups of institutional shareholders in major financial centers. Future research should attempt to develop a model that delineates at each level of communicati on (mass communication, organizational, interpersonal), which information subsidies (news releases, shareholder letters, press interviews, fight Web sites, conference calls, inperson meetings, etc.) are used to reach whom (institutional investors, individual investors, stock analysts, financial reporters, customers, employees, suppliers, etc.) at what stage of the proxy fight or activist campaign (beginning, middle, end). Such a model could both help inform the practice of financial public relations and investor relations during investor activism as well as spark new domainspecific theory building that incorporates multiple levels and elements.

PAGE 266

266 Table 5 1. Summary of findings for hypotheses and research questions Set of Hypotheses/Research Questions Find ings System level agendasetting (H1aH3) Hypothetized system level agendabuilding and agendasetting relationships not supported System level contingent conditions (RQ1a RQ2b) Weak evidence of incumbent financial size indicators as agenda building mode rators Candidateto media agendabuilding (H4aRQ6b) Issue salience Modest support found with the direction of influence generally balanced; releases stronger Stakeholder salience Weak support found with influence generally flowing from media coverage to the candidates Inter candidate agenda setting (H6a H7b) Issue salience Strong support found with both news releases and shareholder letters; releases stronger Stakeholder salience Robust support found with both news releases and shareholder letters; releases stronger Media to investor agenda setting (H8a RQ8) Issue salience Modest support found with the direction of influence generally balanced Stakeholder salience Strong support found with influence generally flowing from media coverage to inves tors Candidate to investor agenda setting (H9a RQ12b) Issue salience Modest support found with the direction of influence generally balanced; releases stronger Stakeholder salience Generally not supported with no dominant direction of influence; letter s stronger Consequences of agenda setting (RQ13 RQ14) Candidate to media agenda building Candidate with stronger agenda building linkages tends to win (with releases and letters) Candidate to investor agenda setting Candidate with stronger agenda setting linkages tends to win (only for lettters not releases)

PAGE 267

267 Figure 51. Agendabuilding and agendasetting theoretical map during corporate proxy contests (with hypotheses and research questions)

PAGE 268

268 APPENDIX A ASSENT SCRIPT Assent Script I am a PhD s tudent at the University of Floridas College of Journalism and Communications. My supervisor is Dr. Spiro Kiousis, chair of the Department of Public Relations. I hope you will agree to participate in this academic study, which explores financial PR durin g corporate proxy contests. Specifically, the purpose of this research is to assess your perceptions on the following topics stakeholders, issues and information subsidies during proxy battles. It is important you know that your participation is completely voluntary. If you agree to participate, you may decline to answer any question that you dont wish to answer. The length of the interview will depend on how much you, the participant, wish to say. However, I anticipate the interview will take betw een 20 and 30 minutes. I believe there is no more than minimal social, psychological or economic risk to you, the participant. You may also discontinue your participation at any time without consequence. There is no compensation for participation and no other direct benefit to you for participating. However, I believe this resulting research may provide you with a useful industrywide perspective of current proxy contest communication practices among your peer group. Your name and identity will r emain confidential to the extent provided by law, as well as in the papers resulting from this research, which I may submit to academic conferences or publications. I intend to make audio recordings of the interviews for this research. Only I will have access to the recordings. After the interviews have been transcribed, the recordings will be erased. If you have any questions about this study, I may be contacted at: mragas@ufl.edu or my supervisor, Dr. Kiousis at: skiousis@jou.ufl.edu. For questions regarding your rights as a research participant, you may contact the IRB at (352) 392-0433 or P.O. Box 112250, Gainesville, Florida 32611.

PAGE 269

269 APPENDIX B INTERVIEW GUIDE Question Guide Stakeholders Issues Information Subsidies Interviewee Backgr ound What is your full job title? What are your primary responsibilities in your current position? How many years have you worked in financial public relations and/or corporate communications? Roughly how many proxy contests and activist investor campaigns have you been involved in? Prioritizing Stakeholders In your opinion, generally what are the priority stakeholders for your client in a proxy contest? How would you rank these stakeholders in order of importance to your client in a proxy contest? How important are shareholders (both institutional and individual) in a proxy contest? How important is the business and financial news media in a proxy contest? Why? Which news media do you believe are the most influential in a proxy contest? How important are the Wall Street analysts/industry analysts in a proxy contest? Why? How important are the proxy advisory firms in a proxy contest? Has their influence changed? Making the Case: The Issues In your opinion, generally what are the most salient issues that are communicated in proxy contests? How would you rank these salient issues in order of importance in a typical proxy contest? How similar are the issues that are communicated in one proxy contest compared to the next?

PAGE 270

270 Do you expect that the most salient issues in proxy contests will change much in the future? Information Subsidies What types of information subsidies (explain the term) do you generally use in a proxy contest? Can you give specific examples of the effective use of such information subsidies in a proxy contest? What information subsidies do you believe are generally the most effective? Least effective? In the future, what information subsidies do you expect will be used more in proxy contests? Less? Futur e of Proxy Contest Communications How do you see proxy contests and activist campaign communications changing in the future? How has the Internet changed the nature of proxy contest / activist campaign communications? Have any other technologies or eve nts significantly changed communications in proxy contests? In the future, do you expect the number of contests /campaigns each year to increase or decline?

PAGE 271

271 APPENDIX C CODE SHEET Code Sheet for Proxy Contest Content 1. Proxy Contest ___ (1) Target Corporation / Pershing Square Capital Mgmt (2) NRG Energy, Inc. / Exelon Corporation (3) Myers Industries Inc. / GAMCO Investors (4) PHH Corporation / Pennant Capital Management (5) Conseco, Inc. / Keith Long (6) Biogen Idec Inc. / Carl C. Icahn (7) CSX Corporation / TCI Fund Management (8) Micrel, Incorporated / Obrem Capital Management (9) International Rectifier / Vishay Intertechnology, Inc. (10) Hexcel Corporation / OSS Capital Management (11) Arrow International, Inc. / McNe il Trust (12) Atmel Corporation / George Perlegos (13) H&R Block, Inc. / Breeden Capital Management (14) Motorola, Inc. / Carl C. Icahn (15) Openwave Systems / Harbinger Capital (16) Career Education Corporation / Steve Bostic (17) UbiquiTel Inc. / Deephaven Capital Management (18) Massey Energy Company / Third Point LLC (19) Motient Corporation / Highland Capital Mgmt (20) GenCorp Inc. / Pirate Capital LLC (21) Blockbuster Inc. / Carl C. Icahn (22) Exar Corporation / GWA Investments (23) Six Flags, Inc. / Daniel Snyder (24) BKF Capital Group, Inc. / Steel Partners II (25) Computer Horizons Corp. / Crescendo Partners 2. Date ___ ___ /___ ___/ ___ ___ 3. Time Period ___ (1) Time 1 (2) Time 2 4. Item ID ___ ___ ___ ___ 5. Item Title _____________________________________________________ 6. Item Source ___ (1) Incumbent News Release

PAGE 272

272 (2) Incumbent Shareholder Letter (3) Challenger News Release (4) Challenger Shareholder Letter (5) The Wall Street Journal (6) New York Times (7) Financial Times (8) Dow Jones News (9) Bloomberg News (10) The Associated Press (11) Reuters News (12) Yahoo! Finance message board posts 7. Issue Categories (frequencies based on word list mentionssee code book ): a. Corporate Performance ___ b. Acquisition ___ c. Compensation ___ d. Transaction ___ e. Experience ___ f. Strategy ___ g. Merger ___ h. Shareholder Value ___ i. Stock Dividend ___ j. Joint Venture ___ k. Partnerships ___ l. Corporate Governance ___ m. Divestiture ___ n. Stock Repurchase ___ o. Ind ependent ___ p. Social Responsibility ___

PAGE 273

273 8. Stakeholder Categories (frequencies based on word list mentionssee code book): a. Management ___ b. Regulator ___ c. Board of Directors ___ d. Shareholder ___ e. Stakeholder ___ f. Analyst ___ g. Union ___ h. Supplier ___ i. Employee ___ j. Community ___ k. Retiree ___ l. Government ___ m. Customer ___ n. Proxy Advisor ___ o. News Media ___

PAGE 274

274 Code Sheet for Financial Performance Data 1. Incumbent Candidate ___ (1) Target Co rporation (2) NRG Energy, Inc. (3) Myers Industries Inc. (4) PHH Corporation (5) Conseco, Inc. (6) Biogen Idec Inc. (7) CSX Corporation (8) Micrel, Incorporated (9) International Rectifier (10) Hexcel Corporation (11) Arrow International, Inc. (1 2) Atmel Corporation (13) H&R Block, Inc. (14) Motorola, Inc. (15) Openwave Systems (16) Career Education Corporation (17) UbiquiTel Inc. (18) Massey Energy Company 19) Motient Corporation (20) GenCorp Inc. (21) Blockbuster Inc. (22) Exar Corporati on (23) Six Flags, Inc. (24) BKF Capital Group, Inc. (25) Computer Horizons Corp. 2. Market Value ($) __________________________ 3. Assets ($) __________________________ 4. Revenue ($) __________________________ 5. Net Income ($) __________________________

PAGE 275

275 APPENDIX D CODE BOOK Code Book for Proxy Contest Content 1. Proxy contest: Select and record from the list on the code sheet the number (125) listed for the proxy contest which pertains to the item (news story, news release, shareholder letter, or message board posting) you will be analyzing. 2. Date: Record the date of the item in the space provided on the code sheet following the format MM/DD/YY. So for example, December 1, 1977, would be recorded 12/01/77. 3. Time Period: Items that were published during the 15-day period beginning a month before the meeting date to determine the proxy contest should be assigned a (1) for Time 1, while items published during the 15-day period directly before the meeting should be assigned a (2) for Time 2. The Time 1 and Time 2 periods for each contest are as follows: Target Corporation / Pershing Square Capital Management Time 1: 4/28/2009 5/12/2009 Time 2: 5/13/2009/27/2009 NRG Energy, Inc. / Exelon Corporation Time 1: 6/21/2009 7/5/2009 Time 2: 7/6/2009 7/20/2009 Myers Industries Inc. / GAMCO Investors Time 1: 3/31/2009 4/14/2009 Time 2: 4/15/2009 4/29/2009 PHH Corporation / Pennant Capital Management Time 1: 5/11/2009 5/25/2009 Time 2: 5/26/2009 6/9/2009 Conseco, Inc. / Keith Long Time 1: 4/1 2/2009 4/26/2009 Time 2: 4/27/209 5/11/2009 Biogen Idec Inc. / Carl C. Icahn Time 1: 5/20/2008 6/3/2008 Time 2: 6/4/2008 6/18/2008 CSX Corporation / TCI Fund Management Time 1: 5/16/2008 5/30/2008 Time 2: 5/31/2006 6/14/2008 Micrel Incorpo rated / Obrem Capital Management Time 1: 4/20/2009 5/4/2009 Time 2: 5/5/2009 5/19/2009 International Rectifier / Vishay Intertechnology, Inc. Time 1: 9/10/2008 9/24/2008 Time 2: 9/25/2008 10/9/2008 Hexcel Corporation / OSS Capital Management Ti me 1: 4/7/2008 4/21/2008 Time 2: 4/22/2008 5/7/2008

PAGE 276

276 Arrow International, Inc. / McNeil Trust Time 1: 8/21/2007 9/4/2007 Time 2: 9/5/2007 9/19/2007 Atmel Corporation / George Perlegos Time 4/18/2007 5/2/2007 Time 2: 5/3/2007 5/17/2007 H&R Block, Inc. / Breeden Capital Management Time 1: 8/7/2007 8/21/2007 Time 2: 8/22/2007 9/5/2007 Motorola, Inc. / Carl C. Icahn Time 1: 4/7/2007 4/21/2007 Time 2: 4/22/2007 5/6/2007 Openwave Systems / Harbinger Capital Time 1: 12/18/2006 1/1/2 007 Time 2: 1/2/20071/16/2007 Career Education Corporation / Steve Bostic Time 1: 4/18/2006 5/2/2006 Time 2: 5/3/2006 5/17/2006 UbiquiTel Inc. / Deephaven Capital Management Time 1: 5/28/2006 6/11/2006 Time 2: 6/12/2006 6/26/2006 Massey Energ y Company / ThirdPoint LLC Time 1: 4/16/2006 4/30/2006 Time 2: 5/1/2006 5/15/2006 Motient Corporation / Highland Capital Management Time 1: 6/12/2006 6/26/2006 Time 2: 6/27/2006 7/11/2006 GenCorp Inc. / Pirate Capital LLC Time 1: 3/1/2006 3/15/2006 Time 2: 3/16/2006 3/30/2006 Blockbuster Inc. / Carl C. Icahn Time 1: 4/11/2005 4/25/2005 Time 2: 4/26/2005 5/10/2005 Exar Corporation / GWA Investments Time 1: 9/27/2005 10/11/2005 Time 2: 10/12/2005 10/26/2005 Six Flags, Inc. / Daniel Snyder Time 1: 10/23/2005 11/6/2005 Time 2: 11/7/2005 11/21/2005 BKF Capital Group, Inc. / Steel Partners II Time 1: 5/24/2005 6/7/2005 Time 2: 6/8/2005 6/22/2005 Computer Horizons Corp. / Crescendo Partners Time 1: 9/11/2005 9/25/2005 Time 2 : 9/26/2005 10/10/2005

PAGE 277

277 4. Item ID: Each item (news story, news release, shareholder letter, or message board posting) should be assigned a unique identification number starting with 0001. This unique number for each item should be recorded in the item ID space provided. 5. Item Title: Record an abbreviated title of the item (such as the headline of a news story, news release, shareholder letter or message board posting) in the space provided. 6. Item Source: Record the originating source of the item: (1 ) incumbent-issued news release (2) incumbent issued shareholder letter, (3) challenger issued news release (4) challenger issued shareholder letter (5) The Wall Street Journal (6) New York Times (7) Financial Times (8) Dow Jones News, (9) Bloomberg News, (10) The Associated Press, (11) Reuters News, or (12) Yahoo! Finance message board posts. The following list shows the incumbent and challenger candidates for each of the proxy contests. For each contest, the incumbent candidate is listed first and the challenger is listed second: Incumbent / Challenger: Target Corporation / Pershing Square Capital Management NRG Energy, Inc. / Exelon Corporation Myers Industries Inc. / GAMCO Investors PHH Corporation / Pennant Capital Management Conseco, Inc. / Keith Long Biogen Idec Inc. / Carl C. Icahn CSX Corporation / TCI Fund Management Micrel Incorporated / Obrem Capital Management International Rectifier / Vishay Intertechnology, Inc. Hexcel Corporation / OSS Capital Management Arrow International, Inc. / McNeil Trust Atmel Corporation / George Perlegos H&R Block, Inc. / Breeden Capital Management Motorola, Inc. / Carl C. Icahn Openwave Systems / Harbinger Capital Career Education Corporation / Steve Bostic UbiquiTel Inc. / Deephaven Capital Management Massey Energy Company / ThirdPoint LLC Motient Corporation / Highland Capital Management GenCorp Inc. / Pirate Capital LLC Blockbuster Inc. / Carl C. Icahn Exar Corporation / GWA Investments Six Flags, Inc. / Daniel Snyder BKF Capital Group, Inc. / Stee l Partners II Computer Horizons Corp. / Crescendo Partners 7. Issue Categories : The unit of analysis is the story, news release, shareholder letter or message board post. Using the below word lists (custom dictionaries), each item will be coded for the following proxy contest issues. Please note that, consistent with previous

PAGE 278

278 CATA agenda -setting research, this coding scheme is based on frequency counts of each issue mention in a unit rather than a simple binary (present/absent) coding scheme: Corporate Per formance (underperform, underperformance, underperforming, underperformed, perform, performance, performed, performing, outperformance, outperform, outperformed, outperforming) Acquisition (acquisition, acquisitions, acquire, acquires, acquiring, acquired, acquirer, acquirers, acquirers, takeover, takeovers, takeover, takeovers, bid, bids, bidding, bidder, bidders, bidders, buyout, buyouts, buy out, buy outs) Compensation (compensation, compensate, compensating, compensated, pay, pays, paying, paid, payment, payments, say onpay, bonus, bonuses, salary, salaries, incentive, incentives, incentivize, incentivized, perk, perks) Transaction (transact, transaction, transactions, transactions, deal, deals, deals, agreement, agreements, agreements, arrangem ent, arrangements, arrangements) Experience (experience, inexperience, experienced, inexperienced, expertise, skill, skills, skilled, qualified, highly qualified, unqualified, qualification, qualifications, credential, credentials, credentialed, knowledge, knowledgeable, proficient, proficiency) Strategy (strategy, strategies, strategic, strategized, strategically) Merger (merge, merges, merger, mergers, mergers, merging, merged) Shareholder Value (value, valuing, valued, valuable) Stock Dividend (dividend, dividends) Joint Venture (JV, JVs, venture, alliance, alliances) Partnership (partnership, partnerships, partnering, partnered, partner, partners) Corporate Governance (governance, governing, governed, oversight, accountable, accountability, respons ible, responsibility) Divestiture (divestiture, divestitures, divest, divesting, divested, spinoff, spinoffs, spinoff, spin offs, spunoff, spun off, split off, splitoffs, split, splitting) Stock Repurchase (repurchase, repurchase, repurchases, repurchas ing, repurchased, buyback, buybacks, buy back, buy -backs)

PAGE 279

279 Independent (independent, independence, autonomous, autonomy) Social Responsibility (CSR, social, socially, society, societal) Litigation (litigation, litigations, litigate, litigating, litigated, litigates, lawsuit, lawsuits, lawsuits, lawsuit, law suits, suit, suits, sue, suing, sued) 8. Stakeholder Categories : The unit of analysis is the story, news release, shareholder letter or message board post. Using the below word lists (custom dictionaries ), each item will be coded for the following proxy contest issues. Please note that, consistent with previous CATA agenda -setting research, this coding scheme is based on frequency counts of each issue mention in a unit rather than a simple binary (present /absent) coding scheme: Management (C.E.O., CEO, management, managements, manage, manager, managers, managing, mismanaging, managed, mismanaged, mismanagement, managerial, leadership, leaderships, executive, executives, executives, executives) Regu lator (regulator, regulators, regulators, regulatory, regulation, regulations) Board of Directors (BOD, B.O.D., director, directors, directors, directors, board, boards, boards, nominee, nominees, nominees, nominees, candidate, candidates, candidate s, candidates) Shareholder (shareholder, shareholders, shareholders, shareholders, stockholder, stockholders, stockholders, stockholders, investor, investors, investors, investors, owner, owners, owners, owners) Stakeholder (stakeholder, stakeholders, stakeholders, stakeholders) Analyst (analyst, analysts, analysts, analysts, researcher, researchers, researchers, researchers, forecaster, forecasters, forecasters, forecasters) Union (union, unions, unions) Supplier (supplier, suppliers, suppliers vendor, vendors, vendors) Employee (employee, employees, employees, employees, worker, workers, workers, workers, workforce, workforces, staff, staffs) Community (community, communitys, communities, resident, residents, residents, citizen, citizens, citizens, neighbor, neighbors, neighbors) Retiree (retiree, retirees, retirees, retirees)

PAGE 280

280 Government (government, governmental, governments, governments, lawmaker, lawmakers, lawmakers, legislator, legislators, legislators, legislators, policymaker, policymakers, policymakers, policymakers) Customer (customer, customers, customers, customers, consumer, consumers, consumers, consumers, shopper, shoppers, shoppers, shoppers, guest, guests, guests, guests, client, clients, client s, clients, patient, patients, patients, patients, patron, patrons, patrons, patrons) Proxy Advisor (RiskMetrics, RiskMetrics, RiskMetricss, ISS, ISSs, Egan Jones, EganJones, Egan-Joness) News Media (press, news, media, medias, journalist, jour nalists, journalists, reporter, reporters, writer, writers, writers, columnist, columnists, commentator, commentators, commentators) Court (court, courts, courts, judge, judges, judges, magistrate, magistrates, magistrates)

PAGE 281

281 Code Book for Financial Performance Data 1. Incumbent Candidate: Select and record from the list on the code sheet the number (1-25) listed for the incumbent candidate which you will be analyzing. (1) Target Corporation (2) NRG Energy, Inc. (3) Myers Industries Inc. (4) PHH Co rporation (5) Conseco, Inc. (6) Biogen Idec Inc. (7) CSX Corporation (8) Micrel, Incorporated (9) International Rectifier (10) Hexcel Corporation (11) Arrow International, Inc. (12) Atmel Corporation (13) H&R Block, Inc. (14) Motorola, Inc. (15) Openwave Systems (16) Career Education Corporation (17) UbiquiTel Inc. (18) Massey Energy Company (19) Motient Corporation (20) GenCorp Inc. (21) Blockbuster Inc. (22) Exar Corporation (23) Six Flags, Inc. (24) BKF Capital Group, Inc. (25) Computer Horizons Corp. PLEASE NOTE: The 10 K annual report of interest for financial data categories 25 below is the report filed for the reporting year immediately before the date of the meeting at which the proxy contest vote is held. For example, for the Target Corporation / Pershing Square Capital Management proxy contest, the meeting tied to the shareholder vote over the proxy contest was held by Target on May 28, 2009. A search of the S.E.C.s EDGAR database reveals that the annual report filed by Target that was the closest to the meeting date in question was the 10K filing it made with the S.E.C. on March, 13, 2009. This is the 10K report of interest for retrieving the corporate financial performance data. The proxy contest meeting dates for each incumbent (corporation) are listed below: (1) Target Corporation (5/28/2009) (2) NRG Energy, Inc. (7/21/2009) (3) Myers Industries Inc. (4/30/2009) (4) PHH Corporation (6/10/2009)

PAGE 282

282 (5) Conseco, Inc. (5/12/2009) (6) Biogen Idec Inc. (6/19/2008) (7) CSX Corporati on (6/15/2008) (8) Micrel, Incorporated (5/20/2008) (9) International Rectifier (10/10/2008) (10) Hexcel Corporation (5/8/2008) (11) Arrow International, Inc. (9/10/2007) (12) Atmel Corporation (5/18/2007) (13) H&R Block, Inc. (9/6/2007) (14) Motorola, Inc. (5/7/2007) (15) Openwave Systems (1/17/2007) (16) Career Education Corporation (5/18/2006) (17) UbiquiTel Inc. (6/27/2006) (18) Massey Energy Company (5/16/2006) (19) Motient Corporation (7/12/2006) (20) GenCorp Inc. (3/31/2006) (21) Blockbuster Inc. (5/11/2005) (22) Exar Corporation (10/27/2005) (23) Six Flags, Inc. (11/22/2005) (24) BKF Capital Group, Inc. (6/23/2005) (25) Computer Horizons Corp. (10/11/2005) 2. Market Value ($): The market value for the incumbent (corporation) is listed in its 10K report filed with the Securities and Exchange Commission (S.E.C.). The 10K report of interest is the report filed the reporting year before the date of the meeting. The aggregate market value of the voting stock of the corporation is listed in the introductory section of the report. Record the market value in the space provided on the code sheet. 3. Assets ($): The assets for the incumbent (corporation) are listed in its 10K report filed with the Securities and Exchange Commission (S.E.C.). The 10K report of interest is the report filed the reporting year before the date of the meeting. Company assets are listed in the Balance Sheet section of the report. Record the assets in the space provided on the code sheet. 4. Revenue ($): The revenue for the incumbent ( corporation) is listed in its 10 K report filed with the Securities and Exchange Commission (S.E.C.). The 10K report of interest is the report filed the reporting year before the date of the meeting. Company revenue is listed in the Income Statement section of the report. Record revenue in the space provided on the code sheet. Net Income ($): The net income for the incumbent (corporation) is listed in its 10 K report filed with the Securities and Exchange Commission (S.E.C.). The 10-K report of interest is the report filed the reporting year before the date of the meeting. Company net income is listed in the

PAGE 283

283 Income Statement section of the report. Record net income in the space provided on the code sheet.

PAGE 284

284 APPENDIX E INFORMATION SUBSIDIE S CONTEST FREQUENCY TABLES Table E1. Candidate issue and stakeholder mentions during the 2005 Computer Horizons Crescendo Partners proxy contest. Computer Horizons PR Crescendo Partners PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 4) (n = 1) (n = 0) (n = 3) (n = 1) (n = 2) (n = 0) Issue categories Acquisition 0 5 8 3 1 3 Compensation 0 3 0 1 1 3 Corporate governance 0 0 0 0 0 4 Directo r independence 0 2 0 0 2 0 Divestiture 0 2 1 2 2 2 Experience 0 5 4 8 2 10 Joint venture 0 0 0 0 0 0 Litigation 0 0 0 0 0 0 Merger 0 2 2 9 4 9 Partnership 0 2 0 4 1 2 Performance 0 2 1 1 1 12 Shareholder va lue 1 19 12 25 17 13 Social responsibility 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 Stock repurchase 0 2 0 2 1 1 Strategy 0 12 5 5 6 7 Transaction 0 1 0 1 4 2 Stakeholder categories Analyst 0 0 0 8 4 6 B oard of directors 0 64 33 47 27 33 Community 0 0 0 0 0 0 Court 0 0 0 0 0 0 Customer 0 2 5 3 0 0 Employee 0 2 5 11 0 5 Government 0 2 5 0 0 0 Management 0 18 10 8 1 15 News media 0 0 0 0 0 0 Other 0 0 0 0 0 0 Proxy advisor 0 11 0 0 3 0 Regulator 0 0 0 0 0 0 Retiree 0 0 0 0 0 0 Shareholder 4 52 16 35 26 20 Supplier 0 1 2 0 0 0 Union 0 0 0 0 0 0 Notes: Time 1 = September 11 September 25, 2005, Time 2 = September 26 October 10, 2005 n = number of news releases or shareholder letters PR = Public Relations

PAGE 285

285 Table E2. Candidate issue and stakeholder mentions during the 2005 BKF Capital Crescendo Steel Partners proxy contest. BKF Capital Group Inc. PR Steel Partners II PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 0) (n = 1) (n = 1) (n = 2) (n = 4) (n = 1) (n = 2) Issue categories Acquisition 0 0 1 1 0 1 0 Compensation 6 6 12 3 26 5 13 Corporate governance 7 6 1 12 34 8 8 Director independence 3 3 1 6 10 2 1 Divestiture 0 0 0 0 0 0 0 Experience 0 0 4 1 2 1 1 Joint venture 0 0 0 0 0 1 0 Litigation 0 0 0 0 0 0 0 Merger 0 0 2 0 0 0 0 Partnership 0 21 12 11 21 7 1 Performance 1 1 2 2 24 2 10 Shareholder value 8 7 16 3 2 7 1 Social responsibility 0 0 0 0 0 0 0 Stock dividend 1 0 1 0 0 2 0 Stock repurchase 0 0 0 0 0 0 0 Stra tegy 3 3 17 1 2 1 0 Transaction 1 0 8 0 7 1 1 Stakeholder categories Analyst 0 0 1 0 0 0 0 Board of directors 24 14 60 42 117 21 28 Community 1 1 0 0 0 0 0 Court 0 0 0 0 0 0 0 Customer 3 2 9 1 4 3 3 Employee 4 2 4 0 19 8 16 Government 0 0 0 0 0 0 0 Management 14 14 30 4 11 1 7 News media 0 0 0 1 2 0 0 Other 0 0 0 0 0 0 0 Proxy advisor 4 0 0 15 5 0 0 Regulator 0 0 0 0 0 0 0 Retiree 0 0 0 0 0 0 0 Shareholder 26 14 65 22 61 20 11 Su pplier 0 0 0 0 0 0 0 Notes: Time 1 = May 24 June 7, 2005, Time 2 = June 8 June 22, 2005 n = number of news releases or shareholder letters PR = Public Relations

PAGE 286

286 Table E3. Candidate issue and stakeholder mentions during the 2005 Six Flags Daniel Snyder proxy contest. Six Flags, Inc. PR Daniel Snyder PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 4) (n = 2) (n = 2) ( n = 0) (n = 4) (n = 1) (n = 0) Issue categories Acquisition 1 14 1 11 14 2 Compensation 1 3 0 5 8 9 Corporate governance 0 0 0 0 3 2 Director independence 1 0 0 0 1 0 Divestiture 0 0 0 0 0 0 Experience 1 2 1 0 1 2 Joint venture 0 0 0 0 0 0 Litigation 0 0 0 0 0 0 Merger 0 1 0 1 0 0 Partnership 0 0 1 0 0 0 Performance 0 0 0 1 16 2 Shareholder value 2 5 5 6 14 10 Social responsibility 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 Stock repurchase 0 0 0 0 0 0 Strategy 1 0 3 4 2 1 Transaction 1 5 2 5 4 3 Stakeholder categories Analyst 0 0 0 0 0 0 Board of directors 11 15 4 12 42 14 Community 0 0 0 0 0 0 Court 0 0 0 0 0 0 Customer 0 2 0 2 0 0 Employee 0 0 0 0 2 2 Government 0 0 0 0 0 0 Management 4 12 1 3 46 21 News media 0 2 0 1 0 1 Other 0 0 0 0 0 0 Proxy advisor 0 11 0 6 8 0 Regulator 0 0 0 0 0 0 Retiree 0 0 0 0 0 0 Shareholder 9 42 10 24 54 16 Supplier 0 0 0 0 0 1 Union 0 0 0 0 0 0 Notes: Time 1 = October 23 November 6, 2005, Time 2 = November 7 November 21, 2005 n = number of news releases or shareholder letters PR = Public Relations

PAGE 287

287 T able E4. Candidate issue and stakeholder mentions during the 2005 Exar Corporation GWA Investments proxy contest. Exar Corporation PR GWA Investments PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Ti me 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 0) (n = 1) (n = 1) (n = 0) (n = 2) (n = 1) (n = 1) Issue categories Acquisition 5 1 1 1 1 0 Compensation 3 3 9 3 8 1 Corporate governance 14 5 7 6 8 1 Director independence 2 3 4 6 15 0 Divestiture 0 0 0 0 0 0 Experience 12 15 19 6 9 5 Joint venture 3 0 0 0 2 0 Litigation 0 0 0 0 0 0 Merger 0 0 0 0 0 0 Partnership 3 0 0 3 1 0 Performance 3 0 1 2 5 0 Shareholder value 6 8 7 3 5 0 Social responsibility 1 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 Stock repurchase 2 2 2 0 0 0 Strategy 10 5 7 0 1 0 Transaction 3 0 1 1 1 0 Stakeholder categories Analyst 0 0 0 0 0 0 Board of directors 62 53 73 46 55 8 Community 0 0 0 0 1 0 Court 0 1 1 0 0 0 Customer 3 0 0 0 0 0 Employee 0 0 0 0 4 0 Government 0 0 0 0 0 0 Management 15 11 15 8 27 3 News media 0 0 0 0 1 0 Other 0 0 0 0 0 0 Proxy adviso r 1 0 0 8 1 0 Regulator 0 0 0 0 0 0 Retiree 0 0 0 0 0 0 Shareholder 25 24 29 38 23 6 Supplier 1 0 0 0 0 0 Union 0 0 0 0 0 0 Notes: Time 1 = September 27 October 11, 2005, Time 2 = October 12 October 26, 2005 n = number of news releases or shareholder letters PR = Public Relations

PAGE 288

288 Table E5. Candidate issue and stakeholder mentions during the 2005 Blockbuster Inc. Carl Icahn proxy contest. Blockbuster, Inc. PR Carl C. Icahn PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 5) (n = 1) (n = 0) (n = 0) (n = 2) (n = 0) (n = 2) Issue categories Acquisition 0 0 0 0 0 Compensation 10 1 3 4 0 Corporate governance 1 0 0 4 2 Director independence 1 2 2 0 0 Divestiture 1 1 0 0 0 Experience 1 4 6 2 2 Joint venture 0 0 0 0 0 Litigation 0 0 0 0 0 Merger 0 0 0 0 0 Partnership 0 0 0 0 0 Performance 0 0 0 2 0 Shareholder value 2 5 2 1 2 Social responsibility 0 0 0 0 0 Stock dividend 8 2 2 0 0 Stock repurchase 1 0 0 0 0 Strategy 9 11 10 0 0 Transaction 0 0 1 1 2 St akeholder categories Analyst 0 0 0 0 0 Board of directors 5 46 20 47 34 Community 0 0 0 0 0 Court 0 0 0 0 0 Customer 1 1 3 0 0 Employee 2 2 2 0 0 Government 0 0 0 0 0 Management 1 20 2 17 13 News media 0 1 0 5 5 Other 0 0 0 0 0 Proxy advisor 0 11 0 5 8 Regulator 0 0 0 0 0 Retiree 0 0 0 0 0 Shareholder 9 37 8 24 15 Supplier 0 0 0 0 0 Union 0 0 0 0 0 Notes: Time 1 = Ap ril 11 April 25, 2005, Time 2 = April 26 May 10, 2005 n = number of news releases or shareholder letters PR = Public Relations

PAGE 289

289 Table E6. Candidate issue and stakeholder mentions during the 2006 GenCorp Inc. Pirate Capital proxy contest. GenCorp, Inc. PR Pirate Capital LLC PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 0) (n = 2) (n = 0) (n = 1) (n = 0) (n = 3) (n = 0) (n = 2) Iss ue categories Acquisition 3 0 8 5 Compensation 0 0 0 2 Corporate governance 1 3 10 18 Director independence 0 1 3 1 Divestiture 0 1 2 5 Experience 0 6 8 14 Joint venture 0 0 0 0 Litigation 3 0 3 0 Merger 0 0 1 1 Partnership 0 0 4 7 Performance 0 1 1 2 Shareholder value 0 7 7 18 Social responsibility 0 0 0 0 Stock dividend 0 0 0 0 Stock repurchase 0 0 0 0 Strategy 0 8 0 3 Transaction 0 0 3 6 Stakeholder categories Analyst 0 0 1 2 Board of directors 5 27 46 64 Community 0 0 5 5 Court 4 0 4 0 Customer 0 0 0 0 Employee 0 0 0 0 Government 0 0 0 0 Management 1 6 10 18 News media 0 0 0 0 Other 0 0 0 0 Proxy advisor 0 2 12 4 Regulator 0 0 0 0 Retiree 0 0 0 0 Shareholder 5 12 46 42 Suppl ier 0 0 0 0 Union 0 0 0 0 Notes: Time 1 = March 1 March 15, 2006, Time 2 = March 16 March 30, 2006 n = number of news releases or shareholder letters PR = Public Relations

PAGE 290

290 Table E7. Candidate issue and stakeholder mentions during the 2006 Motient Corporation Highland Capital proxy contest. Motient Corporation PR Highland Capital Management PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 2) (n = 2) (n = 1) (n = 2) (n = 4) (n = 0) (n = 0) Issue categories Acquisition 1 0 1 0 2 1 Compensation 1 2 0 0 1 9 Corporate governance 0 0 2 1 4 15 Director independence 0 3 3 0 5 12 Divestiture 0 0 1 0 0 0 Experience 0 3 6 2 6 19 Joint venture 0 0 0 0 0 0 Litigation 0 9 4 0 5 0 Merger 0 0 0 0 2 0 Partnership 0 0 3 0 2 3 Performance 0 0 2 0 1 1 Shareholder value 0 3 25 3 15 22 Social re sponsibility 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 Stock repurchase 0 1 0 0 0 1 Strategy 2 5 15 1 5 6 Transaction 9 2 34 2 35 32 Stakeholder categories Analyst 0 0 0 0 0 0 Board of directors 0 17 13 13 47 84 Community 0 0 0 0 0 0 Court 0 5 1 0 3 4 Customer 0 0 0 0 0 0 Employee 2 0 0 0 0 0 Government 0 0 0 0 0 0 Management 0 3 7 3 21 51 News media 0 0 0 0 1 0 Other 0 0 0 0 0 0 Proxy advisor 0 7 0 0 0 12 Regulator 0 0 0 0 1 1 Retiree 0 0 0 0 0 0 Shareholder 1 14 33 5 35 77 Supplier 0 0 0 0 0 0 Union 0 0 0 0 0 0 Notes: Time 1 = June 12 = June 26, 2006, Time 2 = June 27 July 11, 2006 n = number of news releases or shareholder letters PR = Public Relations

PAGE 291

291 Table E8. Candidate issue and stakeholder mentions during the 2006 Massey Energy Third Point LLC proxy contest. Massey Energy Company PR Third Point LLC PR News releases Shareholder letter s News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 0) (n = 1) (n = 1) (n = 2) (n = 0) (n = 1) (n = 1) (n = 1) Issue categories Acquisition 0 0 0 0 0 0 Compensation 3 0 2 0 11 0 C orporate governance 0 0 0 1 2 2 Director independence 5 0 2 1 0 0 Divestiture 0 0 0 0 0 0 Experience 3 0 2 1 1 2 Joint venture 0 0 0 0 0 0 Litigation 0 0 0 0 1 0 Merger 0 0 0 0 0 0 Partnership 0 0 0 0 0 0 P erformance 1 0 0 2 4 2 Shareholder value 0 0 3 0 4 3 Social responsibility 0 0 0 0 0 0 Stock dividend 0 0 0 0 1 0 Stock repurchase 2 0 2 0 6 6 Strategy 0 0 2 1 1 1 Transaction 0 0 0 0 2 0 Stakeholder categories Analyst 0 0 0 0 0 3 Board of directors 12 9 24 1 32 8 Community 0 0 0 0 0 0 Court 0 0 0 0 0 0 Customer 0 0 0 0 0 0 Employee 1 0 0 0 1 2 Government 0 0 0 0 0 0 Management 1 1 2 1 33 0 News media 0 0 0 0 0 0 Other 0 0 0 0 0 0 Proxy advisor 4 0 0 5 0 6 Regulator 0 0 0 0 0 0 Retiree 0 0 0 0 0 0 Shareholder 14 3 14 6 19 15 Supplier 0 0 0 0 0 1 Union 0 0 0 0 0 0 Notes: Time 1 = April 16 April 30, 2006, Time 2 = May 1 May 15, 2006 n = number of news releases or shareholder letters PR = Public Relations

PAGE 292

292 Table E9. Candidate issue and stakeholder mentions during the 2006 UbiquiTel Inc. Deephaven Capital proxy contest. UbiquiTel Inc. PR Deephaven Capital Management PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 0) (n = 3) (n = 1) (n = 1) (n = 0) (n = 0) (n = 0) (n = 0) Issue catego ries Acquisition 2 0 0 Compensation 0 0 0 Corporate governance 8 0 0 Director independence 2 0 0 Divestiture 0 0 0 Experience 0 0 0 Joint venture 0 0 0 Litigation 1 0 0 Merger 8 2 1 Partnership 0 0 0 Performance 0 0 0 Shareholder value 1 0 0 Social responsibility 0 0 0 Stock dividend 0 0 0 Stock repurchase 0 0 0 Strate gy 2 0 0 Transaction 5 0 0 Stakeholder categories Analyst 0 0 0 Board of directors 13 2 2 Community 0 0 0 Court 0 0 0 Customer 0 0 0 Employee 0 0 0 Government 0 0 0 Management 2 0 0 News media 0 0 0 Other 0 0 0 Proxy advisor 0 0 0 Regulator 0 0 0 Retiree 0 0 0 Shareholder 13 3 3 Supplier 0 0 0 Union 0 0 0 Notes: Time 1 = May 7 May 21, 2006, Time 2 = May 22 June 5, 2006 n = number of news releases or shareholder letters PR = Public Relations

PAGE 293

293 Table E10. Candidate issue and stakeholder mentions during the 2006 UbuiTel Inc. Deephaven Capital proxy contest. Career Education Corporation PR Steve Bostic PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 5) (n = 1) (n = 3) (n = 2) (n = 10) (n = 2) (n = 3) Issue categories Acquisition 1 4 3 3 2 10 0 2 Compensation 0 0 1 0 0 6 0 1 Corporate governance 1 4 4 6 7 21 3 1 Director independence 1 0 2 1 3 8 2 0 Divestiture 0 0 0 0 0 0 0 0 Exper ience 9 2 7 4 5 21 3 0 Joint venture 0 0 0 0 0 0 0 0 Litigation 1 7 0 2 7 8 5 0 Merger 0 1 1 1 0 0 0 0 Partnership 0 0 1 0 2 1 0 0 Performance 2 0 0 0 1 5 0 0 Shareholder value 7 2 6 3 9 37 8 7 Social responsibility 0 1 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 0 0 Stock repurchase 1 0 1 0 0 0 0 0 Strategy 0 0 1 1 1 0 0 0 Transaction 1 0 0 0 0 2 0 0 Stakeholder categories Analyst 0 0 0 0 0 2 0 0 Board of directors 25 29 27 26 33 103 14 5 Community 0 0 0 0 1 1 1 0 Co urt 2 10 0 3 3 2 3 0 Customer 0 4 0 0 3 1 3 0 Employee 3 5 2 0 8 12 8 0 Government 0 1 0 1 0 2 0 0 Management 10 10 12 10 37 61 29 5 News media 0 5 0 5 0 6 0 0 Other 0 0 0 0 0 0 0 0 Proxy advisor 0 0 0 0 0 9 0 0 Regulator 3 6 5 5 13 23 1 1 1 Retiree 0 0 0 0 0 0 0 0 Shareholder 11 34 16 24 27 104 16 6 Supplier 0 0 0 0 0 0 0 0 Union 0 0 0 0 0 0 0 0 Notes: Time 1 = April 18 May 2, 2006, Time 2 = May 3 May 17, 2006 n = number of news releases or shareholder letters PR = Public Relations

PAGE 294

294 Table E11. Candidate issue and stakeholder mentions during the 2007 OpenWave Systems Deephaven Capital proxy contest. Openwave Systems PR Harbinger Capital PR News releases Shareholder letters News releases Shareh older letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 3) (n = 0) (n = 0) (n = 0) (n = 2) (n = 0) (n = 0) Issue categories Acquisition 0 1 3 Compensation 0 5 5 Corporate governance 0 3 4 Director independence 2 9 7 Divestiture 0 1 0 Experience 1 27 11 Joint venture 0 1 0 Litigation 0 0 0 Merger 0 0 0 Partnership 2 2 0 Performance 0 3 6 Shareholder value 2 23 10 Social responsibility 0 0 0 Stock dividend 0 0 0 Stock repurchase 0 5 1 Strategy 1 36 7 Transaction 0 2 3 Stakeholder categories Analyst 0 0 0 Board of directors 5 134 52 Community 0 0 2 Court 0 0 0 Customer 1 14 6 Employee 0 2 3 Government 0 0 0 Management 3 37 17 News media 0 2 0 O ther 0 0 0 Proxy advisor 0 2 6 Regulator 0 0 0 Retiree 0 0 0 Shareholder 9 62 48 Supplier 0 0 1 Union 0 0 0 Notes: Time 1 = December 18 January 1, 2007, Time 2 = Januar y 2 January 16, 2007 n = number of news releases or shareholder letters PR = Public Relations

PAGE 295

295 Table E12. Candidate issue and stakeholder mentions during the 2007 Motorola, Inc. Carl Icahn proxy contest. Motorola, Inc. PR Carl C Icahn PR News Releases Shareholder Letters News Releases Shareholder Letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 3) (n = 2) (n = 1) (n = 0) (n = 4) (n = 2) (n = 2) Issue categories Acquisition 4 5 3 1 0 0 0 Compensation 0 0 0 0 0 1 1 Corporate governance 0 3 2 0 3 7 10 Director independence 0 3 5 3 0 1 0 Divestiture 0 0 0 0 2 0 0 Experience 1 4 16 3 0 9 0 Joint venture 0 0 0 0 0 0 0 Litigation 0 0 0 0 0 0 0 Merger 0 2 0 0 0 0 0 Partnership 0 0 0 1 0 1 0 Performance 3 3 5 0 4 4 3 Shareholder value 0 3 10 0 5 8 12 Social responsibility 0 0 0 0 0 1 0 Stock dividend 0 0 1 0 0 0 0 Stock repurchase 5 0 1 2 0 1 0 Strategy 3 1 4 0 0 0 0 Transac tion 0 6 0 0 0 1 0 Stakeholder categories Analyst 0 0 0 1 0 0 0 Board of directors 0 42 69 37 15 45 19 Community 0 0 0 1 0 0 0 Court 0 0 1 0 0 0 0 Customer 1 1 4 0 0 0 4 Employee 2 0 2 5 0 0 2 Government 0 0 0 0 0 0 0 Management 1 3 9 2 5 14 11 News media 0 0 0 0 0 0 0 Other 0 0 0 0 0 0 0 Proxy advisor 0 3 10 2 1 0 2 Regulator 0 1 2 0 0 0 0 Retiree 0 0 0 0 0 0 0 Shareholder 1 28 32 10 21 18 23 Supplier 0 0 0 0 0 0 0 Union 0 0 0 0 0 0 0 Notes: Time 1 = April 7 April 21, 2007, Time 2 = April 22 May 6, 2007 n = number of news releases or shareholder letters PR = Public Relations

PAGE 296

296 Table E13. Candidate issue and stakeholder mentions during the 2007 H&R Bl ock, Inc. Breeden Capital proxy contest. H&R Block, Inc. PR Breeden Capital Management PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 1) (n = 3) (n = 2) (n = 1) (n = 3) (n = 1) (n = 1) Issue categories Acquisition 0 0 0 0 0 0 0 0 Compensation 0 0 0 0 0 0 2 0 Corporate governance 2 0 0 0 7 5 6 0 Director independence 0 1 4 0 0 5 11 2 Divestiture 0 1 0 0 2 0 0 0 Experience 0 1 1 1 2 2 2 0 Joint venture 0 0 0 0 0 0 0 0 Litigation 0 0 0 0 0 0 0 0 Merger 0 0 0 0 0 0 0 0 Partnership 0 2 7 4 4 7 10 4 Performance 0 0 0 0 6 5 12 1 Shareholder value 0 2 1 2 4 2 14 1 Social responsibility 0 0 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 1 0 Stock repurchase 0 0 0 0 2 0 1 0 Strategy 0 2 3 2 2 0 3 0 Transaction 0 0 0 0 0 0 6 0 Stakeholder Categories Analyst 0 0 0 0 0 0 0 0 Board of directors 12 7 29 15 15 20 49 8 Community 0 0 0 0 0 0 0 0 Court 0 0 0 0 0 0 0 0 Customer 0 0 0 0 0 0 3 0 Employee 0 0 0 0 0 0 3 0 Government 0 0 1 0 0 0 0 0 Management 1 2 2 1 13 6 21 0 News media 0 0 0 0 0 0 0 1 Other 0 0 0 0 0 0 0 0 Proxy advisor 0 2 0 0 0 2 0 5 Regulator 0 0 1 0 0 0 1 0 Retiree 0 0 0 0 0 0 0 0 Shareholder 6 7 12 6 9 6 23 5 Supplier 0 0 0 0 0 0 0 0 Union 0 0 0 0 0 0 0 0 Notes: Time 1 = August 7 August 21, 2007, Time 2 = August 22 September 5, 2007 n = number of news releases or shareholder letters PR = Public Relation s

PAGE 297

297 Table E14. Candidate issue and stakeholder mentions during the 2007 Atmel Corporation George Perlegos proxy contest. Arrow International PR McNeil Trust PR News releases Shareholder letters News releases Shareholder letters Time 1 T ime 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 4) (n = 0) (n = 0) (n = 2) (n = 0) (n = 1) (n = 0) Issue categories Acquisition 0 2 0 0 Compensation 1 2 0 0 Corporate governance 0 4 3 0 Director i ndependence 1 2 1 0 Divestiture 0 0 0 1 Experience 1 1 2 2 Joint venture 0 0 0 0 Litigation 0 0 0 0 Merger 16 45 11 20 Partnership 0 0 0 0 Performance 0 1 1 4 Shareholder val ue 0 0 0 0 Social responsibility 0 0 0 0 Stock dividend 0 0 0 0 Stock repurchase 0 0 0 0 Strategy 1 7 0 4 Transaction 8 39 5 3 Stakeholder categories Analyst 0 0 0 0 Board o f directors 15 79 7 29 Community 0 0 0 0 Court 0 0 0 0 Customer 0 4 0 3 Employee 0 0 0 0 Government 0 0 0 0 Management 1 6 1 4 News media 0 0 0 0 Other 0 0 0 0 Prox y advisor 0 3 0 0 Regulator 0 2 1 0 Retiree 0 0 0 0 Shareholder 17 41 10 14 Supplier 0 0 0 0 Union 0 0 0 0 Notes: Time 1 = August 21 September 4, 2007, Time 2 = September 5 September 19, 200 7 n = number of news releases or shareholder letters PR = Public Relations

PAGE 298

298 Table E15. Candidate issue and stakeholder mentions during the 2007 Arrow International, Inc. McNeil Trust proxy contest. Arrow International PR McNeil Trus t PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 4) (n = 0) (n = 0) (n = 2) (n = 0) (n = 1) (n = 0) Issue categories Acquisition 0 2 0 0 Compensation 1 2 0 0 Corporate governance 0 4 3 0 Director independence 1 2 1 0 Divestiture 0 0 0 1 Experience 1 1 2 2 Joint venture 0 0 0 0 Litigation 0 0 0 0 Merger 16 45 1 1 20 Partnership 0 0 0 0 Performance 0 1 1 4 Shareholder value 0 0 0 0 Social responsibility 0 0 0 0 Stock dividend 0 0 0 0 Stock repurchase 0 0 0 0 Strategy 1 7 0 4 Transaction 8 39 5 3 Stakeholder categories Analyst 0 0 0 0 Board of directors 15 79 7 29 Community 0 0 0 0 Court 0 0 0 0 Customer 0 4 0 3 Employee 0 0 0 0 Government 0 0 0 0 Management 1 6 1 4 News media 0 0 0 0 Other 0 0 0 0 Proxy advisor 0 3 0 0 Regulator 0 2 1 0 Retiree 0 0 0 0 Shareholder 17 41 10 14 Supplier 0 0 0 0 Union 0 0 0 0 No tes: Time 1 = August 21 September 4, 2007, Time 2 = September 5 September 19, 2007 n = number of news releases or shareholder letters PR = Public Relations

PAGE 299

299 Table E16. Candidate issue and stakeholder mentions during the 2008 Hexcel Corporation OSS Capital proxy contest. Hexcel Corporation PR OSS Capital Management PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 0) (n = 2) (n = 1) (n = 0 ) (n = 2) (n = 2) (n = 0) (n = 0) Issue categories Acquisition 0 0 0 0 Compensation 0 1 0 0 Corporate governance 11 8 0 0 Director independence 5 5 2 1 Divestiture 0 0 0 0 Experience 5 22 8 6 Joint venture 0 0 0 6 Litigation 0 0 0 0 Merger 0 0 0 0 Partnership 2 2 1 2 Performance 2 5 4 1 Shareholder value 3 6 8 4 Social responsibility 0 0 0 0 Stock dividend 0 0 0 0 Stock repurchase 0 0 0 0 Strategy 4 5 0 2 Transaction 0 0 0 0 Stakeholder categories Analyst 0 0 0 0 Board of directors 29 41 20 26 Community 0 0 0 0 Court 0 0 0 0 Customer 0 5 1 0 Employee 0 3 0 0 Government 0 1 0 0 Management 4 4 14 6 News media 0 0 0 1 Other 0 0 0 0 Proxy advisor 7 2 0 0 Regulator 0 1 0 0 Retiree 0 0 0 0 S hareholder 15 9 19 13 Supplier 0 1 0 1 Union 0 0 0 0 Notes: Time 1 = April 7 April 21, 2008, Time 2 = April 22 May 7, 2008 n = number of news releases or shareholder letters PR = Public Relations

PAGE 300

300 T able E17. Candidate issue and stakeholder mentions during the 2008 International Rectifier Vishay Intertechnology, Inc. proxy contest. International Rectifier PR Vishay Intertechnology, Inc. PR News releases Shareholder letters News releases Sh areholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 2) (n = 8) (n = 1) (n = 1) (n = 1) (n = 5) (n = 0) (n = 0) Issue categories Acquisition 4 31 9 3 1 8 Compensation 1 14 3 0 0 9 Corporate governanc e 3 11 1 1 0 1 Director independence 2 19 5 5 0 24 Divestiture 0 0 0 0 0 0 Experience 3 9 5 3 0 9 Joint venture 0 0 0 0 0 0 Litigation 2 9 8 0 0 0 Merger 0 2 0 0 0 0 Partnership 0 0 0 0 0 0 Performance 0 3 1 0 0 0 Shareholder value 8 32 10 6 1 10 Social responsibility 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 Stock repurchase 0 0 0 0 0 0 Strategy 10 28 11 1 0 1 Transaction 0 14 2 3 1 4 Stakeholder categories Analyst 0 4 0 2 0 2 Board of directors 21 226 52 15 5 98 Community 0 1 0 0 0 0 Court 0 0 0 0 0 0 Customer 0 1 1 0 0 0 Employee 0 2 0 0 0 0 Government 0 0 0 0 0 2 Management 11 34 6 0 0 4 News media 0 0 0 0 0 2 Other 0 0 0 0 0 0 Proxy advisor 0 3 0 0 0 23 Regulator 0 3 1 0 0 0 Retiree 0 0 0 0 0 0 Shareholder 18 101 20 3 4 64 Supplier 0 0 0 0 0 0 Union 0 0 0 0 0 0 Notes: Time 1 = September 10 September 24, 2008, Time 2 = September 25 October 9, 2008 n = number of news releases or shareholder letters PR = Public Relations

PAGE 301

301 Table E18. Candidate issue and stakeholder mentions during the 2008 Micrel, Inc. Obrem Capital proxy contest. Micr el, Incorporated PR Obrem Capital Management PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 2) (n = 4) (n = 2) (n = 1) (n = 1) (n = 5) (n = 0) (n = 0) Issue categories Acquisition 0 2 19 0 1 1 Compensation 3 0 0 0 0 0 Corporate governance 0 14 2 0 7 15 Director independence 1 15 3 1 5 8 Divestiture 0 0 0 0 0 0 Experience 8 25 13 5 7 13 Joint venture 0 0 0 0 0 0 Litigation 1 0 0 0 0 0 Merger 0 0 1 0 0 0 Partnership 3 2 1 1 0 1 Performance 2 11 2 0 8 7 Shareholder value 13 13 11 5 14 20 Social responsibility 0 0 0 0 0 0 Stock dividend 9 1 3 1 1 0 Stock repurchase 5 1 0 1 0 0 Strategy 13 9 3 6 1 11 Transaction 1 2 1 0 0 1 Stakeholder categories Analyst 1 3 0 0 0 15 Board of directors 30 70 35 11 49 81 Community 0 0 0 0 0 1 Court 0 0 0 0 0 0 Customer 4 2 3 1 0 1 Employee 4 4 0 0 0 0 Government 0 0 0 0 0 0 Management 12 12 4 2 21 69 News media 1 0 0 0 1 1 Other 0 0 0 0 0 0 Proxy advisor 0 11 1 0 0 8 Regulator 0 0 0 0 0 0 Retiree 0 0 0 0 0 0 Shareholder 25 58 21 13 31 81 Supplier 0 0 0 0 0 0 Union 0 0 0 0 0 0 Notes: Time 1 = April 20 May 4, 2009, Time 2 = May 5 May 19, 2008 n = number of news releases or shareholder letters PR = Public Relations

PAGE 302

302 Table E19. Candidate is sue and stakeholder mentions during the 2008 CSX Corporation TCI Fund Management proxy contest. CSX Corporation PR TCI Fund Management PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Tim e 2 Time 1 Time 2 (n = 1) (n = 3) (n = 2) (n = 4) (n = 1) (n = 2) (n = 0) (n = 0) Issue categories Acquisition 0 2 0 0 0 0 Compensation 0 1 0 0 0 1 Corporate governance 0 1 0 0 0 0 Director independence 0 1 0 0 1 0 D ivestiture 0 0 0 0 0 0 Experience 8 4 6 0 8 5 Joint venture 0 0 0 0 0 0 Litigation 0 0 0 0 0 1 Merger 0 0 0 0 0 0 Partnership 4 6 3 0 1 2 Performance 1 3 3 0 1 3 Shareholder value 4 13 4 0 2 1 Social respon sibility 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 Stock repurchase 0 1 0 0 2 0 Strategy 0 19 0 0 0 0 Transaction 0 1 0 0 0 0 Stakeholder categories Analyst 0 0 0 0 0 0 Board of directors 27 34 25 44 25 23 Community 0 1 0 1 0 0 Court 0 35 0 20 0 9 Customer 1 4 1 5 0 1 Employee 1 2 1 4 0 2 Government 0 1 0 4 0 0 Management 7 12 7 18 9 10 News media 0 0 0 1 0 0 Other 0 0 0 0 0 0 Proxy advisor 0 0 0 1 0 0 Regulator 1 1 1 0 0 0 Retiree 0 0 0 0 0 0 Shareholder 8 12 34 32 10 12 Supplier 0 0 0 0 0 0 Union 0 0 0 1 0 0 Notes: Time 1 = May 16 May 30, 2008, Time 2 = May 31 June 14, 2008 n = number of news releases or sha reholder letters PR = Public Relations

PAGE 303

303 Table E20. Candidate issue and stakeholder mentions during the 2008 Biogen Idec, Inc. Carl Icahn proxy contest. Biogen Idec, Inc. PR Carl C. Icahn PR News releases Shareholder letters News r eleases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 0) (n = 2) (n = 2) (n = 2) (n = 0) (n = 0) (n = 1) (n = 2) Issue Categories Acquisition 4 4 17 8 36 Compensation 0 2 1 0 2 Corporate governance 7 0 4 2 0 Director independence 2 2 3 0 0 Divestiture 0 0 0 0 0 Experience 0 2 3 3 1 Joint venture 0 0 0 0 0 Litigation 0 0 0 0 3 Merger 1 2 2 0 1 Partnership 0 2 3 5 1 Performan ce 5 4 7 1 0 Shareholder value 6 8 15 4 2 Social responsibility 0 0 0 0 0 Stock dividend 0 0 0 0 0 Stock repurchase 0 0 0 0 0 Strategy 4 6 11 1 3 Transaction 2 4 3 0 10 Stakeholder categories Analyst 2 0 1 1 0 Board of directors 59 24 52 27 47 Community 0 0 0 0 0 Court 0 0 0 0 4 Customer 3 0 1 0 1 Employee 0 0 0 4 3 Government 0 0 0 0 0 Management 7 3 10 5 7 News media 0 1 2 0 0 Other 0 0 0 0 0 Proxy advisor 17 0 4 0 0 Regulator 0 0 0 0 0 Retiree 0 0 0 0 0 Shareholder 28 15 20 8 7 Supplier 0 0 0 0 0 Union 0 0 0 0 0 Notes: Time 1 = May 20 June 3, 2008, Time 2 = June 4 0 June 18, 2008 n = number of news releases or shareholder letters PR = Public Relations

PAGE 304

304 Table E21. Candidate issue and stakeholder mentions during the 2009 Conseco, Inc. Keith Long proxy contest. Conseco, Inc. PR Ke ith Long PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 0) (n = 1) (n = 0) (n = 0) (n = 0) (n = 1) (n = 0) (n = 0) Issue categories Acquisition 0 0 Compensation 0 0 Corporate governance 1 2 Director independence 0 0 Divestiture 0 0 Experience 2 4 Joint venture 0 0 Litigation 0 0 Merger 0 0 Partnership 0 2 Performance 0 3 Shareholder value 0 1 Social responsibility 0 0 Stock dividend 0 0 Stock repurchase 0 0 Strategy 0 0 Transactio n 0 0 Stakeholder categories Analyst 0 0 Board of directors 9 10 Community 0 0 Court 0 0 Customer 0 0 Employee 0 0 Government 0 0 Management 2 3 News media 0 0 Other 0 0 Proxy advisor 2 12 Regulator 0 0 Retiree 0 0 Shareholder 3 8 Supplier 0 0 Union 0 0 Notes : Time 1 = April 12 April 26, 2009, Time 2 = April 27 May 11, 2009 n = number of news releases or shareholder letters PR = Public Relations

PAGE 305

305 Table E22. Candidate issue and stakeholder mentions during the 2009 PHH Corporation P ennant Capital proxy contest. PHH Corporation PR Pennant Capital Management PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 3) (n = 6) (n = 1) (n = 0) (n = 2) (n = 2) (n = 2) (n = 1) Issue categories Acquisition 2 2 2 1 0 0 1 Compensation 0 2 0 4 0 6 2 Corporate governance 3 10 3 3 0 2 1 Director independence 0 9 0 11 4 13 9 Divestiture 1 0 1 0 0 0 0 Experience 7 7 7 6 2 26 1 2 Joint venture 0 0 0 0 0 0 0 Litigation 0 0 0 0 0 0 0 Merger 0 0 0 0 0 0 0 Partnership 0 0 0 0 0 1 1 Performance 3 8 3 3 2 1 5 Shareholder value 7 2 7 15 1 12 9 Social responsibility 0 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 0 0 Stock repurchase 0 0 0 0 0 0 0 Strategy 4 1 4 1 0 1 2 Transaction 1 1 1 1 0 0 0 Stakeholder categories Analyst 0 0 0 0 0 0 0 Board of directors 80 110 79 68 12 67 40 Community 0 0 0 0 0 0 0 Court 0 0 0 0 0 0 0 Customer 1 6 1 11 0 2 2 Employee 0 0 0 4 0 3 0 Government 0 0 0 0 0 0 1 Management 39 27 35 44 9 38 40 News media 1 4 1 3 0 0 0 Other 0 0 0 1 0 0 0 Proxy advisor 0 22 0 0 0 0 1 Regulator 1 0 1 2 0 0 0 Retiree 0 0 0 0 0 0 0 Shareholder 24 48 20 27 9 17 12 Supplier 0 0 0 0 0 0 0 Union 0 0 0 0 0 0 0 Notes: Time 1 = May 11 May 25, 2009, Time 2 = May 26 June 9, 2009 n = number of news releases or shareholder letters PR = Public Relations

PAGE 306

306 Table E23. Candidate issue and stakeholder mentions during the 2009 Myers Industries, Inc. GAMCO Investors proxy contest. Myers Industries, Inc. PR GAMCO Investors PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 0) (n = 1) (n = 2) (n = 1) (n = 0) (n = 0) (n = 0) (n = 0) Issue categories Acquisition 0 0 0 Compensation 2 0 5 Corporate governance 1 0 2 Directo r independence 1 0 4 Divestiture 0 0 1 Experience 0 0 2 Joint venture 0 0 0 Litigation 0 0 0 Merger 0 0 0 Partnership 0 0 0 Performance 0 0 0 Shareholder valu e 0 0 7 Social responsibility 0 0 0 Stock dividend 0 0 5 Stock repurchase 0 0 0 Strategy 0 0 13 Transaction 0 0 0 Stakeholder categories Analyst 0 0 0 Board of directors 6 4 32 Community 0 0 0 Court 0 0 0 Customer 1 0 1 Employee 0 0 2 Government 0 0 0 Management 0 0 6 News media 0 0 0 Other 0 0 0 Proxy a dvisor 1 0 0 Regulator 1 0 2 Retiree 0 0 0 Shareholder 3 6 16 Supplier 0 0 0 Union 0 0 0 Notes: Time 1 = March 31 April 14, 2009, Time 2 = April 15 April 29, 2009 n = number of news releases or shareholder letters PR = Public Relations

PAGE 307

307 Table E24. Candidate issue and stakeholder mentions during the 2009 NRG Energy, Inc. Exelon Corporation proxy contest. NRG Energy, Inc. PR Exelon Corporation PR News r eleases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 1) (n = 4) (n = 1) (n = 3) (n = 3) (n = 3) (n = 0) (n = 0) Issue categories Acquisition 1 10 7 6 9 10 Compe nsation 0 7 4 1 0 4 Corporate governance 0 7 0 0 0 0 Director independence 0 7 2 0 8 16 Divestiture 0 3 0 0 0 0 Experience 0 13 3 0 3 5 Joint venture 0 0 0 0 0 0 Litigation 0 0 0 0 4 0 Merger 0 0 0 0 2 2 Par tnership 0 1 0 1 0 0 Performance 0 8 2 2 0 4 Shareholder value 0 49 21 18 34 68 Social responsibility 0 0 0 0 0 0 Stock dividend 0 0 0 0 0 2 Stock repurchase 0 4 0 2 0 0 Strategy 0 11 1 0 3 2 Transaction 0 7 1 2 8 21 Stakeholder categories Analyst 0 2 0 0 0 0 Board of directors 4 100 23 19 32 49 Community 0 0 0 0 0 1 Court 0 0 0 0 7 0 Customer 0 1 0 0 0 0 Employee 0 0 0 0 0 0 Government 0 0 0 0 0 0 Mana gement 0 15 1 3 8 9 News media 1 1 0 1 1 2 Other 0 0 0 0 0 0 Proxy advisor 0 7 0 0 0 0 Regulator 0 0 0 0 5 4 Retiree 0 0 0 0 0 0 Shareholder 2 66 9 15 43 56 Supplier 0 0 0 1 0 0 Union 0 0 0 0 0 0 Notes : Time 1 = June 21 July 5, 2009, Time 2 = July 6 July 20, 2009 n = number of news releases or shareholder letters PR = Public Relations

PAGE 308

308 Table E25. Candidate issue and stakeholder mentions during the 2009 Target Corporation Pershing Square Capital proxy contest. Target Corporation PR Pershing Square Capital PR News releases Shareholder letters News releases Shareholder letters Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 Time 1 Time 2 (n = 3) (n = 9) (n = 1) (n = 1 ) (n = 1) (n = 11) (n = 1) (n = 1) Issue categories Acquisition 1 0 0 0 0 3 0 0 Compensation 7 3 0 0 0 28 0 1 Corporate governance 8 25 0 0 0 69 0 0 Director independence 8 15 0 0 0 35 0 0 Divestiture 0 1 0 4 0 2 0 7 Experience 15 21 0 0 0 47 0 0 Joint venture 0 0 0 0 0 0 0 0 Litigation 0 0 0 0 0 0 0 0 Merger 0 0 0 0 0 1 0 0 Partnership 5 0 1 0 0 9 0 0 Performance 7 4 0 0 0 6 0 0 Shareholder value 10 12 0 0 0 28 0 0 Social responsibility 0 0 0 0 0 0 0 0 Stock dividen d 0 0 0 0 0 2 0 0 Stock repurchase 4 0 0 0 0 0 0 0 Strategy 11 16 1 0 0 29 0 0 Transaction 5 5 2 0 0 23 0 0 Stakeholder categories Analyst 0 1 0 0 0 1 0 0 Board of directors 77 189 0 0 4 426 0 12 Community 0 1 0 0 0 1 0 0 Court 0 0 0 0 0 0 0 0 Customer 14 1 0 0 0 5 0 0 Employee 0 0 0 0 0 2 0 0 Government 0 0 0 0 0 2 0 0 Management 23 37 2 1 2 47 1 1 News media 1 1 0 0 0 4 0 0 Other 0 0 0 0 0 0 0 0 Proxy advisor 1 9 0 0 0 46 0 0 Regulator 0 0 0 0 0 0 0 0 Retiree 0 0 0 0 0 0 0 0 Shareholder 34 105 0 6 3 206 2 18 Supplier 0 0 0 0 0 0 0 0 Union 0 0 0 0 0 0 0 0 Notes: Time 1 = April 28 May 12, 2009, Time 2 = May 13 May 27, 2009 n = number of news releases or shareholder letters PR = Public Relations

PAGE 309

309 APPENDIX F MEDIA AND INVESTORS CONTEST FREQUENCY TA BLES (ISSUE SALIENCE)

PAGE 310

310 Table F 1. Media and investor issue mentions during the 2009 Target Corporation Pershing Square Capital Management proxy contest

PAGE 311

311 Table F-2 Media and investor issue mentions during the 2009 NRG Energy, Inc. Exelon Corporation proxy contes t

PAGE 312

312 Table F 3. Media and investor issue mentions during the 2009 Myers Industries, Inc. GAMCO Investors proxy contest

PAGE 313

313 Table F -4. Media and investor issue ment ions during the 2009 PHH Corporation Pennant Capital Management proxy contest

PAGE 314

314 Table F-5 Media and investor issue mentions during the 2009 Conseco, Inc. Keith Long proxy contest

PAGE 315

315 Table F 6. Media and investor issue mentions during the 2008 Biogen Idec, Inc. Carl C. Icahn proxy contest

PAGE 316

316 Table F 7. Media and investor issue mentions during the 2008 CSX Corporation TCI Fund Management proxy contest

PAGE 317

317 Table F 8. Media and investor issue mentions during the 2008 Micrel, Incorporated Obrem Capital Management proxy contest

PAGE 318

318 Table F -9 Media and investor issue mentions during the 2008 International Rectifier Vishay Intertechnology, Inc. proxy contest

PAGE 319

319 Table F 10 Media and investor issue mentions during the 2008 Hexcel Corporation OSS Capital Management proxy contest

PAGE 320

320 Table F 11. Media and investor issue mentions during the 2007 Arrow International, Inc. McNeil Trust proxy contest

PAGE 321

321 Table F 12. Media and investor issue mentions during the 2007 Atmel Corporation George Perlegos proxy contest

PAGE 322

322 Table F 13. Media and investor issue mentions during the 2007 H&R Block, Inc. Breeden Capital Management proxy contest

PAGE 323

323 Table F14 Media and investor issue mentions during the 2007 Motorola, Inc. Carl C. Icahn proxy contest

PAGE 324

324 T able F15 Media and investor issue mentions during the 2007 Openwave Systems Harbinger Capital proxy contest

PAGE 325

325 Table F16 Media and investor issue mentions during the 2006 Career Education Corporation Steve Bostic proxy contest

PAGE 326

326 Table F17 Media and investor issue mentions during the 2006 UbiquiTel Inc. Deephaven Capital Management proxy contest

PAGE 327

327 Table F18 Media and investor issue mentions during the 2006 Massey Energy Company Third Point LLC proxy contest

PAGE 328

328 Table F19 Media and investor issue mentions during the 2006 Motient Corporation Highland Capital Management proxy contest

PAGE 329

329 Table F20 Media and investor issue mentions during the 2006 GenCorp Inc. Pirate Capital, LLC proxy contest

PAGE 330

330 Table F21 Media and investor issue mentions during the 2005 Blockbuster Inc. Carl C. Icahn proxy contest

PAGE 331

331 Table F22 Media and investor issue mentions during the 2005 Exar Corporation GWA Investments proxy contest

PAGE 332

332 Table F23 Media and investor issue mentions during the 2005 Six Flags, Inc. Daniel Snyder proxy contest

PAGE 333

333 Table F24 Media and investor issue mentions during the 2005 BKF Capital Group, Inc. Steel Partners II proxy contest

PAGE 334

334 Table F-25 Media and investor issue mentions during the 2005 Computer Horizons Crescendo proxy contest

PAGE 335

335 APPENDIX G MEDIA AND INVESTORS CONTEST FREQUENCY TA BLES (STAKEHOLDER SA LIENCE) Table G1. Media and investor stakeholder mentions during the 2009 Target Corporation Pershing Square Capital Management proxy contest

PAGE 336

336 Table G-2 Media and investor stakeholder mentions during the 2009 NRG Energy, Inc. Exelon Corporation proxy contest

PAGE 337

337 Table G-3 Media and investor stakeholder mentions during the 2009 Myers Industries, Inc. GAMCO Investors proxy contest

PAGE 338

338 Table G-4 Media and investor stakeholder mentions during the 2009 PHH Corporation Pennant Capital Management proxy contest

PAGE 339

339 Table G-5 Media and investor issue mentions during the 2009 Conseco, Inc. Keith Long proxy contest

PAGE 340

340 Table G6. Me dia and investor issue mentions during the 2008 Biogen Idec, Inc. Carl C. Icahn proxy contest

PAGE 341

341 Table G7. Media and investor issue mentions during the 2008 CSX Corporation TCI Fund Management proxy contest

PAGE 342

342 Table G8. Media and investor issue mentions during the 2008 Micrel, Incorporated Obrem Capital Management proxy contest

PAGE 343

343 Table G-9 Media and investor issue mentions during the 2008 International Rectifier Vishay Intertechnology, Inc. proxy contest

PAGE 344

344 Table G10 Media and investor issue mentions during the 2008 Hexcel Corporation OSS Capital Management proxy contest

PAGE 345

345 Table G11. Media and investor issue mentions during the 2007 Arrow International, Inc. McNeil Trust proxy contest

PAGE 346

346 Table G12. Media and investor issue mentions during the 2007 Atmel Corporation George Perlegos proxy contest

PAGE 347

347 Table G13. Media and investor issue mentions during the 2007 H&R Block, Inc. Breeden Capital Management proxy contest

PAGE 348

348 Table G14 Media and investor issue mentions during the 2007 Motorola, Inc. Carl C. Icahn proxy contest

PAGE 349

349 Table G15 Media and investor issue mentions during the 2007 Openwave Systems Harbinger Capital proxy contest

PAGE 350

350 Table G16 Media and investor issue mentions during the 2006 Career Education Corporat ion Steve Bostic proxy contest

PAGE 351

351 Table G17. Media and investor issue mentions during the 2006 UbiquiTel Inc. Deephaven Capital Management proxy contest

PAGE 352

352 Table G18 Media and investor issue mentions during the 2006 Massey Energy Company Thi rd Point LLC proxy contest

PAGE 353

353 Table G19 Media and investor issue mentions during the 2006 Motient Corporation Highland Capital Management proxy contest

PAGE 354

354 Table G20 Media and investor issue mentions during the 2006 GenCorp Inc. Pirate Capital, LLC proxy contest

PAGE 355

355 Table G21 Media and investor issue mentions during the 2005 Blockbuster Inc. Carl C. Icahn proxy contest

PAGE 356

356 Table G22 Media and investor issue mentions during the 2005 Exar Corporation GWA Investments proxy contest

PAGE 357

357 Ta ble G23 Media and investor issue mentions during the 2005 Six Flags, Inc. Daniel Snyder proxy contest

PAGE 358

358 Table G24 Media and investor issue mentions during the 2005 BKF Capital Group, Inc. Steel Partners II proxy contest

PAGE 359

359 Table G25 Media and investor issue mentions during the 2005 Computer Horizons Crescendo proxy contest

PAGE 360

360 APPENDIX H CONTEST CORRELATION TABLES (ISSUE SALIEN CE)

PAGE 361

361 Table H 1. Issue agenda building and agenda setting relationships during the 2009 Target Pershing Square prox y contest WSJ NYT FT RN AP DJN BN Target Corporation PR (incumbent) Releases .69*** .51* .88*** .51* .92*** .39 .76*** Letters .10 22 .12 .00 .27 38 .10 Pershing Square PR (challenger) Releases .67** .46* .79*** .41 .89*** .45* .79*** Letters .36 .40 .13 .19 .01 .22 .11 Investor salience .32 .49* .71*** .23 .43* .20 .55* Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 362

362 Table H 2. Issue agenda building and agenda setting relationships during the 2009 NRG Energy Exelon proxy contest WSJ NYT FT RN AP DJN BN NRG Ene rgy PR (incumbent) Releases .26 .08 .39 .46* .43* .32 55* Letters .33 .05 .57** .40 .49* .45* 48* Exelon PR (challenger) Releases .48* .22 .64** .63** .61** .61** .67** Letters Investor salience .62** .46* .66* .34 .49* 76*** 48* Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloom berg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 363

363 Table H 3. Issue agenda building and agenda setting relationships during the 2009 Myers GAMCO proxy contest WSJ NYT FT RN AP DJN BN Myers Industrie s PR (incumbent) Releases Letters GAMCO PR (challenger) Releases Letters Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wal l Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 364

364 Table H 4. Issue agenda building and agenda setting relationships during the 2009 PHH Pennant Capital proxy contest WSJ NYT FT RN AP DJN BN PHH Corporation PR (incumbent) Releases Letters Pennant Capital PR (chall enger) Releases Letters Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 365

365 Table H 5. Issue agenda building and agenda setting relationships durin g the 2009 Conseco Keith Long Capital proxy contest WSJ NYT FT RN AP DJN BN Conseco PR (incumbent) Releases Letters Keith Long PR (challenger) Releases Letters In vestor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 366

366 Table H 6. Issue agenda building and agenda setting relationships during the 2008 Biogen Idec Carl Icahn proxy contest WSJ NYT FT RN AP DJ N BN Biogen Idec PR (incumbent) Releases .11 .40 .57** .34 Letters .43* .51* .58** .51* Carl Icahn PR (challenger) Releases Letters .45* .29 .51* .62** Investor salience .20 .26 .57** .45 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News enotes winner of the proxy contest. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 367

367 Table H 7. Issue agenda building and agenda setting relationships during the 2008 CSX Corp. TCI proxy contest WSJ NYT FT RN AP DJN BN CSX PR ( incumbent) Releases .03 .24 .40 .33 .20 .15 Letters .00 .15 .02 .12 .03 .08 TCI Fund PR (challenger) Releases .20 .09 .02 .18 .03 .14 Letters Investor salience .12 .01 .03 .17 .04 .11 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 368

368 Table H 8. Issue agend a building and agenda setting relationships during the 2008 Micrel Obrem Capital proxy contest WSJ NYT FT RN AP DJN BN Micrel PR (incumbent) Releases .61** .36 Letters .79*** .48* Obrem Capital PR (challenger) Releases .61** .45* Letters Investor salience .72*** .37 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financia l Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All onetailed tests.

PAGE 369

369 Table H 9. Issue agenda building and agenda setting relationships during the 2008 International Rectifier Vishay proxy contest WSJ NYT FT RN AP DJN BN Intl Rectifier PR (incumbent) Releases .77*** .79*** Letters .77*** .69*** Vishay Intertechnology PR (challenger) Releases .72*** .69*** Letters Investor salience .49* .61** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 370

370 Table H 10. Issue agenda building and agenda setting relationships during the 2008 Hexcel OSS Capital proxy contest WSJ NYT FT RN AP DJN BN Hexcel PR (incumbent) Releases .87*** Letters .75*** OSS Capital PR (challenger) Releases .37 Letters Investor salience .36 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01 *** p <.001. All one tailed tests.

PAGE 371

371 Table H 11. Issue agenda building and agenda setting relationships during the 2007 Arrow McNeil Trust proxy contest WSJ NYT FT RN AP DJN BN Arrow PR (incumbent) Releases .62** Le tters McNeil Trust PR (challenger) Releases .55* Letters .20 Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 372

372 Table H 12. Issue agenda building and agenda setting relationships during the 2007 Atmel George Perlegos proxy contest WSJ NYT FT RN AP DJN BN Atmel PR (incumbent) Releases .39 .67** .72*** Letters .35 .68*** .53* George Perlegos PR (challenger) Releases .66** .63** .58** Letters .74*** .63** .48* Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News e proxy contest. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 373

373 Table H 13. Issue agenda building and agenda setting relationships during the 20 07 H&R Block Breeden Capital proxy c ontest WSJ NYT FT RN AP DJN BN H&R Block PR (incumbent) Releases .20 .09 .86*** .59** .21 Letters .05 .09 .73*** .39 .15 Breeden Capital PR (challenger) Releases .53* .38 .88*** .82*** .39 Letters .52* .41 .78*** .68*** .57** Investor salience .06 .42* .09 .22 .57** Notes: PR = public relations, Releases = news releases, Letters = shareholder lette rs WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed test s.

PAGE 374

374 Table H 14. Issue agenda building and agenda setting relationships during the 2007 Motorola Carl Icahn proxy contest WSJ NYT FT RN AP DJN BN Motorola PR (incumbent) Releases .82*** .59** .72*** .50* .70*** .65** .58** Letter s .55* .69*** .75*** .67** .61** .51* .59** Carl Icahn PR (challenger) Releases .23 .41 .29 .50* .07 .17 .34 Letters .45* .44* .63** .53* .44* .46* .63** Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 375

375 Table H 15. Issue agenda building and agenda setting relationships during the 2007 Openwave Harbinger Capital proxy contest WSJ NYT FT RN AP DJN BN Openwave Systems PR (incumbent) Releases .28 .68*** .71*** .16 Letters Harbinger Capital PR (challenger) Releases .01 .47* .51* .01 Letters Investor salience .16 .13 .07 .34 N otes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News enotes winner of the proxy contest. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 376

376 Table H 16. Issue agenda building and agenda setting relationships during the 2006 Career Education Steve Bostic proxy contest WSJ NYT FT RN AP DJN BN Career Education PR (incumbent) Releases .59** .53* .79*** Letters .48* .60** .75*** Steve Bostic PR (challenger) Releases .66** .61** .84*** Letters .41 .69*** .81*** Investor salience .3 8 .29 .65** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 377

377 Table H 17. Issue agenda building and agenda setting relationships during the 2006 UbiquiTel Deephaven Capital proxy contest WSJ NYT FT RN AP DJN BN UbiquiTel PR (incumbent) Releases Letters Deephaven Capital PR (challenger) Releases Letters Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 378

378 Table H 18. Issue agenda building and agenda setting relationships during the 2006 Massey Energy Third Point proxy contest WSJ NYT FT RN AP DJN BN Massey Energy PR (incumbent) Releases .33 .59** Letters .79*** .51* Third Point PR (challenger) Releases .25 .06 Letters .74*** .71*** Investor salience .25 .37 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NY T = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 379

379 Table H 19. Issue agenda building and agenda setting relationships during the 2006 Motient Highland Capital proxy contest WSJ NYT FT RN AP DJN BN Motient PR (incumbent) Releases .70*** Letters .68*** Hi ghland Capital PR (challenger) Releases .78*** Letters Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New Yo rk Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 380

380 Table H 20. Issue agenda building and agenda setting relationships during the 2006 GenCorp Pirate Capital proxy contest W SJ NYT FT RN AP DJN BN GenCorp PR (incumbent) Releases Letters Pirate Capital PR (challenger) Releases Letters Investor salience Notes: PR = public relat ions, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News winner of the proxy contest. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 381

381 Table H 21. Issue agenda building and agenda setting relationships during the 2005 Blockbuster Carl Icahn proxy contest WSJ NYT FT RN AP DJN BN Block buster PR (incumbent) Releases .35 .45* .54* .41* .39 Letters .37 .52* .39 .37 .59** Carl Icahn PR (challenger) Releases .02 .32 .04 .11 .49* Letters .01 .06 .26 .16 .19 Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journ al, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 382

382 Table H 22. Issue agenda building and agenda setting relationships during the 2005 Exar GWA Investments proxy contest WSJ NYT FT RN AP DJN BN Exar PR (incumbent) Releases Letters GWA PR (challenge r) Releases Letters Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, R N = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 383

383 Table H 23. Issue agenda building and agenda setting relationships during the 2005 Six Flags Daniel Snyder proxy contest WSJ NYT FT RN AP DJN BN Six Flags PR (incumbent) Releases .69*** .37 .59** .46* .72*** Letters .83*** .42* .75*** .48* .85*** Daniel Snyder PR (ch allenger) Releases .86*** .82*** .76*** .84*** .81*** Letters .80*** .59** .66** .62** .77*** Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Stre et Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one taile d tests.

PAGE 384

384 Table H 24. Issue agenda building and agenda setting relationships during the 2005 BKF Capital Steel Partners proxy contest WSJ NYT FT RN AP DJN BN BKF Capital PR (incumbent) Releases .36 Letters .55* Steel Partners PR (challenger) Releases .67** Letters .52* Investor salience .31 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 385

385 T able G 25. Issue agenda building and agenda setting relationships during the 2005 Computer Horizons Crescendo Partners proxy contest WSJ NYT FT RN AP DJN BN Computer Horizons PR (incumbent) Releases .36 Letters .36 Crescendo Partners PR (challenger) Releases .70*** Letters .53* Investor salience .55* Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 386

386 APPENDIX I CONTEST CORRELATION TABLES (STAKEHOLDER SALIENCE) Table I 1. Stakeholder agenda building and agenda setting relation ships during the 2009 Target Pershing Square proxy contest WSJ NYT FT RN AP DJN BN Target Corporation PR (incumbent) Releases .42 .23 .43* .37 .42 .41 .39 Letters .07 .03 .07 .02 .07 .09 .06 Pershing Square PR (challenger) Releases .28 .33 .30 .36 .29 .32 .39 Letters .32 .33 .32 .26 .32 .35 .29 Investor salience .74*** .73*** .75*** .74*** .73*** .72*** .75*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 387

387 Table I 2. Stakeholder agenda building and agenda setting relationships during the 2009 NRG Energy Exelon proxy contest WSJ NYT FT RN AP DJN BN NRG Energy PR (incumbent) Releases .48* .32 .35 .29 .32 .28 .33 Letters .32 .40 .53* .23 .14 .42 .22 Exelon PR (challenger) Releases .12 .28 .15 .12 .09 .23 .14 Letters Investor salienc e .92*** .72*** .79*** .69*** .77*** .71*** .71*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All onetailed tests.

PAGE 388

388 Table I 3. Stakeholder agenda building and agenda setting relationships during the 2009 Myers GAMCO proxy contest WSJ NYT FT RN AP DJN BN Myers Industries PR (incumbent) Releases .00 Letters GAMCO PR (challenger) Releases Letters Investor salience .81*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, p <.01, *** p <.001. All one tailed tests.

PAGE 389

389 Table I 4. Stakeholder agenda building and agenda setting relationships during the 2009 PHH Pennant Capital proxy contest WSJ NYT FT RN AP DJN BN PHH Corporation PR (incumbent) Release s Letters Pennant Capital PR (challenger) Releases Letters Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 390

390 Table I 5. Stakeholder agenda building and agen da setting relationships during the 2009 Conseco Keith Long Capital proxy contest WSJ NYT FT RN AP DJN BN Conseco PR (incumbent) Releases .13 Letters Keith Long PR (challenger) Releases .09 Letters Investor salience .51* Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associ ated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 391

391 Table I 6. Stakeholder agenda building and agenda setting relationships during the 2008 Biogen Idec Carl Icahn proxy contest WSJ NYT FT RN AP DJN BN Biogen Idec PR (incumbent) Releases .34 .22 .11 .24 Letters .48* .38 .24 .35 Carl Icahn PR (challenger) Releases Letters .46* .33 .28 .10 Investor salience .66** .68** .81*** .52* Notes : PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News of the proxy contest. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 392

392 Table I 7. Stakeholder agenda building and agenda setting relationships during the 2008 CSX Corp. TCI proxy contest WSJ NYT FT RN AP DJN BN CSX PR (incumbent) Releases .42 .14 .29 .28 .37 .23 Letters .45* .21 .32 .28 .39 .36 TCI Fund PR (challenger) Releases .55* .13 .38 .35 .45* .35 Letters Investor salience .72*** .66** .72*** .76*** .73*** .77*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wal l Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 393

393 Table I 8. Stakeholder agenda building and agenda setting relationships during the 2008 Micrel Obrem Capital proxy contest WSJ NYT FT RN AP DJN BN Micrel PR (incumbent) Releases .38 .33 .22 Letters .37 .33 .14 Obrem Capital PR (challenger) Releases .46* .39 .34 Letters Investor salience .85*** .71*** .72*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 394

394 Table I 9. Stakeholder agenda building and agenda setting relationships during the 2008 International Rectifier Vishay proxy contest WSJ NYT FT RN A P DJN BN International Rectifier PR (incumbent) Releases .18 .17 .08 .02 Letters .30 .36 .24 .12 Vishay Intertechnology PR (challenger) Releases .35 .30 .30 .21 Letters Investor salience .53* .61** .51* .35 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomb erg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 395

395 Table I 10. Stakeholder agenda building and agenda setting relationships during the 2008 Hexcel OSS Capital proxy contest WSJ NYT FT RN AP DJN BN H excel PR (incumbent) Releases .30 .29 Letters .17 .18 OSS Capital PR (challenger) Releases .56* .53* Letters Investor salience .85*** .76*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 396

396 Table I 11. Stakeholder agenda building and agenda setting relationships during the 2007 Arrow McNeil Trust proxy contest WSJ NYT FT RN AP DJN BN Arrow PR (incumbent) Releases .02 .12 Letters McNeil Trust PR (challenger) Releases .09 .00 Letters .13 .05 Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareho lder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 397

397 Table I 12. Stakeholder agenda building and agenda setting relationships during the 2007 Atmel George Perlegos proxy contest WSJ NYT FT RN AP DJN BN Atmel PR (incumbent) Releases .15 .17 .31 .41 Letters .54* .55* .53* .42 George Perlegos PR (challenger) Releases .35 .37 .43* .34 Letters .39 .40 .47* .39 Investor salience No tes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News notes winner of the proxy contest. p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 398

398 Table I 13. Stakeholder agenda building and agenda setting relationships during the 2007 H&R Block Breeden Capital proxy contest WSJ NYT FT RN AP DJN BN H&R Block PR (incumbent) Releases .44* .53* .46* .35 .41 .37 Letters .34 .25 .22 .28 .31 .14 Breeden Capital PR (challenger) Releases .48* .57* .50* .39 .43* .41 Letters .42 .50* .51* .25 .49* .46* Investor salience .77*** .61** .81*** .63** .75*** .88*** Notes: PR = public relations, Releases = news releases, Letters = shareh older letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p < .001. All one tailed tests.

PAGE 399

399 Table I 14. Stakeholder agenda building and agenda setting relationships during the 2007 Motorola Carl Icahn proxy contest WSJ NYT FT RN AP DJN BN Motorola PR (incumbent) Releases .27 .21 .34 .21 .33 .22 .20 Letters .18 .07 .22 .12 .27 .16 .08 Carl Icahn PR (challenger) Releases .13 .33 .31 .30 .35 .25 .27 Letters .38 .25 .39 .28 .36 .31 .27 Investor salience .52* .55* .78*** .40 .70*** .43* .35 Notes: PR = public relations, Release s = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 400

400 Table I 15. Stakeholder agenda building and agenda setting relationships during the 2007 Openwave Harbinger capital proxy contest WSJ NYT FT RN AP DJN BN Openwave Systems PR (incumbent) Releases .21 .38 .21 .26 Letters Harbinger Capital PR (challenger) Releases .01 .12 .03 .06 Letters Investor salience .24 .50* .48* .49* Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 401

401 Table I 16. Stakeholder agenda building and agenda setting relationships during the 2006 Career Education Steve Bostic proxy contest W SJ NYT FT RN AP DJN BN Career Education PR (incumbent) Releases .14 .32 .12 .10 Letters .24 .41 .25 .21 Steve Bostic PR (challenger) Releases .22 .26 .17 .06 Letters .02 .15 .10 .15 Investor salience .73*** .72*** .64** .62** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 402

402 Table I 17. Stakeholder agenda building and agenda setting relationships during the 2006 UbiquiTel Deephaven Capital proxy contest WSJ NYT FT RN AP DJN BN UbiquiTel PR (incumbent) Releases Letters Deephaven Capital PR (challenger) Releases Letters Investor salience Notes: PR = public relations, Relea ses = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 403

403 Table I 18. Stakeholder agenda building and agenda setting relationships during the 2006 Massey Energy Third Point proxy contest WSJ NYT FT RN AP DJN BN Massey Energy PR (incu mbent) Releases .39 .18 .42 Letters Third Point PR (challenger) Releases .46* .22 .49* Letters .35 .17 .39 Investor salience .71*** .73*** .74*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed t ests.

PAGE 404

404 Table I 19. Stakeholder agenda building and agenda setting relationships during the 2006 Motient Highland Capital proxy contest WSJ NYT FT RN AP DJN BN Motient PR (incumbent) Releases .30 .24 .41 Letters .47* .41 .57* Highland Capital PR (challenger) Releases .39 .35 .51* Letters Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Stree t Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 405

405 Table I 20. Stakeholder agenda building and agenda setting relationships during the 2006 GenCorp Pirate Capital proxy contest WSJ NYT FT RN AP DJN BN GenCorp PR (incumbent) Releases .47* .37 Letters .23 .41 Pirate Capital PR (challenger) Releases .29 .39 Letters .22 .34 Investor salience .76*** .8 3*** Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 406

406 Table I 21. Stakeholder agenda building and agenda setting relationships during the 2005 Blockbuster Carl Icahn proxy contest WSJ NYT FT RN AP DJN BN Blockbuster PR (incumbent) Releases .37 .43* .41 .41 .40 .40 Letters .46* .41 .23 .28 .30 .31 Carl Icahn PR (challenger) Releases .25 .47* .49* .45* .42 .40 Letters .24 .45* .49* .45* .41 .39 Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 407

407 Table I 22. Stakeholder agenda building and agenda setting relationships during the 2005 Exar GWA Investments proxy contest WSJ NYT FT RN AP DJN BN Exar PR (incumbent) Releases .02 Letters .13 GWA PR (challenger) Releases .13 Letters .14 Investor salience Notes: PR = public relatio ns, Releases = news releases, Letters = shareholder letters AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 408

408 Table I 23. Stakeholder agenda building and agenda setting relationships during the 2005 Six Flags Daniel Snyder proxy contest WSJ NYT FT RN AP DJN BN Six Flags PR (incumbent) Releases .38 .21 .29 .50* .45* Letters .33 .17 .24 .47* .42 Daniel Snyder PR (challenger) Releases .27 .12 .19 .37 .20 Letters .43* .24 .32 .34 .35 Investor salience Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Jou rnal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed test s.

PAGE 409

409 Table I 24. Stakeholder agenda building and agenda setting relationships during the 2005 BKF Capital Steel Partners proxy contest WSJ NYT FT RN AP DJN BN BKF Capital PR (incumbent) Releases .51* .15 .19 Letters .55* .15 .23 Steel Partners PR (challenger) Releases .49* .25 .27 Letters .67** .23 .33 Investor salience .68** .57* .38 Notes: PR = public relations, Releases = news releases, Letters = shareholder letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed t ests.

PAGE 410

410 Table I 25. Stakeholder agenda building and agenda setting relationships during the 2005 Computer Horizons Crescendo Partners proxy c ontest WSJ NYT FT RN AP DJN BN Computer Horizons PR (incumbent) Releases .08 .08 Letters .03 .03 Crescendo Partners PR (challenger) Releases .24 .22 Letters .30 .28 Investor salience .57* .58** Notes: PR = public relations, Releases = news releases, Letters = shareholde r letters WSJ = Wall Street Journal, NYT = New York Times, FT = Financial Times, RN = Reuters News AP = Associated Press, DJN = Dow Jones News, BN = Bloomberg News p < .05, ** p <.01, *** p <.001. All one tailed tests.

PAGE 411

411 LIST OF REFERENCES Agresti, A., & Finlay, B. (1997). Statistical methods for the social sciences (3rd ed.). Upper Saddle River, NJ: P rentice Hall. Aranow, E. R., & Einhorn, H. A. (1956). Corporate proxy contests: Expenses of management and insurgents. Cornell Law Review 42, 4 27. Ashton Partners (2005). The new crisis: Shareholder activism [Report]. Atwater, T., Salwen, M. B., & Ande rson, R. B. (1985). Interpersonal discussion as a potential barrier to agendasetting. Newspaper Research Journal 6 (4), 3743. Bagley, C. E., & Savage, D. W. (2006). Managers and the legal environment: Strategies for the 21st century (5th ed.). Mason, OH: Thomson West. Bagli, C. V. (1997, March 2). Goodbye, takeover pain: The spin doctor is in. New York Times Behr, R. L., & Iyengar, S. (1985). Television news, real world cues, and changes in the public agenda. Public Opinion Quarterly 49(1), 3857. Ber elson, B. (1952). Content analysis in communication research New York: Hafner. Berger, B. K. (2001). Private issues and public policy: Locating the corporate agenda in agendasetting theory. Journal of Public Relations Research 13(2), 91 126. Berger, B. K., Hertog, J. K., & Park, D. (2002). The political role and influence of business organizations: A communication perspective. In W. B. Gudykunst (Ed.), Communication yearbook 26 (pp. 160 200). Mahwah, NJ: Erlbaum. Berkowitz, D., & Adams, D. B. (1990). Information subsidy and agendabuilding in local television news. Journalism Quarterly 67(4), 723731. Bloomberg, L.P. (2009). About Bloomberg. Retrieved August 1, 2009, from http://about.bloomberg.com/news.html Blyskal, J., & Blyskal, M. (1985). PR: How the public relations industry writes the news New York: Morrow. Bobinski, G. S., & Ramirez, G. G. (1994). Advertising to investors: The effects of financial relations advertising on stock volume and price. Journal of Advertising, 23(4), 13 28. Bollen, L. H., Hassink, H. F., de Lange, R. K., Buijl, S. D. (2008). Best practices in managing investor relations Websites: Directions for future research. Journal of Information Systems 22(2), 171 194.

PAGE 412

412 Botan, C. H., & Hazelton, V. (2006). Public relations in a new age. In C. H. Botan & V. Hazelton (Eds.), Public relations theory II (pp. 1 18). Mahwah, NJ: Erlbaum. Boyle, T. P. (2001). Intermedia agenda setting in the 1996 presidential election. Journalism & Mass Communication Quarterly 78(1), 26 44. Brewer, M., & McCombs, M. (1996). Setting the community agenda. Journalism & Mass Communication Quarterly, 73(1), 7 16. Broom, G. M. (2006). An opensystem approach to building theory in public relations. Journal of Public Relations Research 18(2), 141 150. Br uck, C. (1989). The predators ball: The inside story of Drexel Burnham and the rise of the junk bond raiders New York: Penguin Books. Buddenbaum, J. M. (2001). The media, religion, and public opinion: toward a unified theory of cultural influence. In D. A. Stout & J. M. Buddenbaum (Eds.), Religion and popular culture: Studies in the interaction of worldviews (pp. 19 38). Ames, IA: Iowa State University Press. Cameron, G. T. (1996). Memory of investor relations messages: An informationprocessing study of Grunigs situational theory. Journal of Public Relations Research 4 (1), 45 60. Cameron, G. T., Sallot, L., & Curtin, P. A. (1997). Public relations and the production of news: A critical review and a theoretical framework. In B. Burleson (Ed.), Communic ation Yearbook 20 (pp. 111 155). Newbury Park, CA: Sage. Campbell, C., Gillan, S., & Niden, C. (1999). Current perspectives on shareholder proposals: The 1997 proxy season. Financial Management 28(1), 89 98. Cancel, A. E., Cameron, G. T., Sallot, L. M., & Mitrook, M. A. (1997). It depends: A contingency theory of accommodation in public relations. Journal of Public Relations Research, 9 (1), 3163. Carney, A. & Jorden, A. (1993). Prepare for business related crises. Public Relations Journal 49(8), 34 35. Carroll, C. E. (2004). How the mass media influence perceptions of corporate reputation: Exploring agendasetting effects within business news coverage. Unpublished doctoral dissertation, University of Texas, Austin. Carroll, C. E., & McCombs, M. (2003) Agenda setting effects of business news on the publics images and opinions about major corporations. Corporate Reputation Review 16(1), 36 46. Chaffee, S. H. (1991). Communication concepts 1: Explication. Newbury Park, CA: Sage.

PAGE 413

413 Chaffee, S. H., & Be rger, C. R. (1987). What communication scientists do. In C. R. Berger & S. H. Chaffee (Eds.), Handbook of communication science (pp. 99122). Newbury Park, CA: Sage. Cheney, R. E. (1983). PR to the rescue in takeover battles. Public Relations Quarterly 28(1), 23 26. Chiang, C. Y. (1995). Bridging and closing the gap of our society: Social function of media agenda setting, a case study in Taiwan Unpublished masters thesis, University of Texas, Austin. Clarke, G., & Murray, L. W. (2000). Investor relations: perceptions of the annual statement. Corporate Communications: An International Journal 5 (3), 144151. Clarke, K. (2004). An introduction to financial public relations. In A. Theaker (Ed.), The public relations handbook (2nd ed., pp. 205217). London: Routledge. Cohen, B. (1963). The press and foreign policy Princeton, NJ: Princeton University Press. Cole, B. M. (Ed.). (2004). The new investor relations: Expert perspectives on the state of the art New York: Bloomberg Press. Collis, T. (2004). Fin ancial public relations. In A. Gregory (Ed.), Public relations in practice (pp. 59 75). London: Kogan Page. ComScore. (2008, July 9). Yahoo! Finance ranks as top financial news & research site in the U.S. with more than 18 million visitors in May, according to Comscore. Retrieved July 3, 2009, from http://www.comscore.com/Press_Events/Press_Releases/ 2008/07/Yahoo!_Finance_Top_Financial_News_and_Research_Site_in_US Cone, K. R., Fischel, D. R., Pelnar, G. J., & Ross, D. J. (2004). The HewlettPackard merger: A case study. In B. M. Cole (Ed.), The new investor relations: Expert perspectives on the state of the art (pp. 169179). New York: Bloomberg Press. Constantinescu, A. R., & Tedesco, J. C. (2007). Framing a kidnapping Journalism Studies 8 (3),444464. Crosby, J. (2009, May 15). Sides split in proxy fight over Target board. Star Tribune. Retrieved June 25, 2009, from http://www.startribune.com/business/45449572.html?elr=KArksUUUU Cutlip, S. M. (1962, May 26). Third of newspapers content PR inspired. Editor & Publisher 26, 68. Cutlip, S. M., Center, A. H., & Broom, G. M. (1999). Effective Public Relations (7th ed.). Englewood Cliffs, NJ: Pre ntice Hall.

PAGE 414

414 Danielian, L., & Reese, S. (1989). A closer look at intermedia influences on agendasetting: The cocaine issue of 1986. In P. Shoemaker (Ed.), Communication campaigns about drugs: Government, media and the public (pp. 4766). Hillsdale, NJ: Erlbaum. Dasgupta, S., & Nanda, V. (1997). Tender offers, proxy contests, and largeshareholder activism. Journal of Economics & Management Strategy 6 (4), 787 820. Davidoff, S. M. (2009, March 31). Targets challenge. New York Times Retrieved June 25, 2009, from http://dealbook.blogs.nytimes.com/2009/03/31/targets challenge/ Davis, A. (2002). Public relations democracy: Public relations, politics and the mass media in Britain. Manchester, U.K.: Manchester University Press. Davis, A. (2006). The role of the mass media in investor relations. Journal of Communication Management 10 (1), 7 17. Davis, J. (1995, March). The art of spin. Management Today 72. Davis, L. J. (1989, September 24). Imagemakers in takeover land. New York Times Magazine. Dearing, J. W., & Rogers, E. M. (1996). Agenda setting Thousand Oaks, CA: Sage. de Bakker, F. G. A., & den Hond, F. (2008) Activists influence tactics and corporate policies. Business Communication Quarterly 71(1), 107 111. Deephouse, D. L. (2000). Media reputation as a strategic resource: An integration of mass communication and resource based theories. Journal of Management 26(6), 10911112. Demers, D. P., Craff, D., Choi, Y., & Pessin, B. M. (1989). Issue obtrusiveness and the agendasetting effects of n ational network news. Communication Research 16(6), 739 812. DigiText, Inc. (2008). Diction 5.0: About Retrieved July 6, 2009, fromhttp://www.dictionsoftware.com/index.php?option=com_content&task=view&i d=14&Itemid=33 Dolphin, R. R. (2003). Approaches to investor relations: Implementation in the British context. Journal of Marketing Communications 9 (1), 29 43. Dolphin, R. R. (2004). The strategic role of investor relations. Corporate Communications: An International Journal 9 (1), 2542. Dolphin, R. R., & Fan, Y. (2000). Is corporate communications a strategic function? Management Decision, 38(2), 99 107.

PAGE 415

415 Dunn, S. W. (2006). Candidate and media agenda setting in the 2005 Virginia gubernatorial election. Unpublished masters thesis, Virginia Polytechni c Institute and State University, Blacksburg, VA. Dunn, S. W. (2009). Candidate and media agenda setting in the 2005 Virginia gubernatorial election. Journal of Communication, 59(3), 635652. Emerson, F. D., & Latcham, F. C. (1953). Proxy contests: A study in shareholder sovereignty. California Law Review 41(3), 393 438. Evatt, D. S., & Bell, T. (1995, August). The political dance: A test of agendasetting effects in the 1994 Texas gubernatorial election. Paper presented at the annual meeting of the Association for Education in Journalism and Mass Communication, Washington, DC. FactSet Research Systems. (2009). FactSet at a glance Retrieved July 26, 2009, from http://www.factset.com/about/overview Farragher, E. J., Kleinman, R., & Bazaz, M. S. (1994). Do investor relations make a difference? The Quarterly Review of Economics and Finance, 34 (4), 405412. Ferguson, M. (1984, August). Building theory in public relations: Interorganizational relationships as public relations paradigms Paper pres ented at the annual meeting of the Association for Education in Journalism and Mass Communication, Gainesville, FL. Ferreira, E. J., & Smith, S. D. (2003). Wall $treet Week: Information or entertainment? Financial Analysts Journal 59(1), 4553. Fortunato, J. A (2000). Public relations strategies for creating mass media content: A case study of the National Basketball Association. Public Relations Review 26(4), 481 497. Funkhauser, G. R. (1973). The issues of the sixties: An exploratory study in the dynamics of public opinion. Public Opinion Quarterly 37 (1), 62 75. Gandy, O. H. (1982). Beyond agenda setting: Information subsidies and public policy Norwood, NJ: Ablex. Gans, H. (1979). Deciding whats news: A study of CBS Evening News, NBC Nightly News, News week and Time. New York: Pantheon. Ghorpade, S. (1986). Agenda setting: A test of advertising's neglected function. Journal of Advertising Research, 26(4), 23 27. Gillan, S. L., & Starks, L. T. (1998). A survey of shareholder activism: Motivation and empirical evidence. Contemporary Finance Digest 2 (3), 2 38.

PAGE 416

416 Gillan, S., & Starks, L. (2007). The evolution of shareholder activism in the United States. Journal of Applied Corporate Finance 19(1), 55 73. Gilberg, S., Eyal, C., McCombs, M. E., & Nicholas, D. (1980). The State of the Union address and the press agenda. Journalism Quarterly 57 (4), 584 588. Golan, G. (2006). Inter media agenda setting and global news coverage: Assessing the influence of the New York Times on three television evening news pr ograms. Journalism Studies 7 (2), 323334. Golan, G. J., Kiousis, S. K., & McDaniel, M. L. (2007). Secondlevel agenda setting and political advertising, Journalism Studies, 8(3), 432443. Gower, K. K. (2006). Public relations research at the crossroads Journal of Public Relations Research 18 (2), 177 190. Grunig, J. E. (1997), A situational theory of publics: Conceptual history, recent challenges, and new research. In D. Moss, T. MacManus, & D. Vercic (Eds.), Public relations research: An international perspective (pp. 3 48). London: International Thomson Business Press. Grunig, J. E. (2006). Furnishing the edifice: Ongoing research on public relations as a strategic management function. Journal of Public Relations Research, 18 (2), 151176. Grunig, J. E., Grunig, L. A., & Dozier, D. M. (2006). The excellence theory. In C. H. Botan & V. Hazelton (Eds.), Public Relations Theory II (pp. 2162). Mahwah, NJ: Erlbaum. Grunig, J. E., & Hunt, T. (1984). Managing public relations New York: Holt, Rinehart & Win son. Hallahan, K. (2001). The dynamics of issue activation and response: An issues processes model. Journal of Public Relations Research 13, 27 59. Hart, R. P. (2000a). Diction 5.0: The text analysis program Austin, TX: DigiText, Inc. Hart, R. P. (200 0b). Redeveloping Diction: Theoretical considerations. In M. D. West (Ed.), Theory, method, and practice in computer content analysis (pp. 43 60). Westport, CT: Ablex. Hockerts, K., & Moir, L. (2004). Communicating corporate social responsibility to inves tors: The changing role of the investor relations function. Journal of Business Ethics, 52(1), 85 98. Holsti, O. (1969). Content analysis for the social sciences and humanities Reading, MA: AddisonWesley.

PAGE 417

417 Holton, G. A. (2006). Investor suffrage movement. Financial Analysts Journal 62 (6), 15 20. Hon, L. C. (1998). Demonstrating effectiveness in public relations: Goals, objectives and evaluation. Journal of Public Relations Research 10 (2), 103 135. Hong, Y., & Cho, S. (2008, May). Who should represent?: Source credibility of the IR and the PR spokesperson and the affiliated company Paper presented at the annual meeting of the International Communication Association, Montreal, Canada. Hong, Y., & Kiousis, S. (2007, May). Relationship maintenance with financial publics: An analysis of investor relations on small companies Web sites Paper presented at the annual meeting of the International Communication Association, San Francisco. Hong, Y., & Ki, E. (2006). How do public relations practitioners perceive investor relations? An exploratory study. Corporate Communications: An International Journal 12(2), 199 213. Huth, W. L., & Maris, B. A. (1992). Large and small firm stock price response to Heard on the Street recommendations. Journal of Accounting, Auditing & Finance 7 (1), 27 44. Ikenberry, D., & Lakonishok (1993). Corporate governance through the proxy contest: Evidence and implications. Journal of Business 66(3), 405 435. Iyengar, S. (1991). Is anyone responsible? Chicago: University of Chic ago Press. Iyengar, S., & Kinder, D. R. (1987). News that matters: Television and American opinion Chicago: University of Chicago Press. Iyengar, S., & Simon, A. F. (2000). New perspectives and evidence on political communication and campaign effects. An nual Review of Psychology 51(1), 149 169. Iyengar, S., & McGrady, J. (2005). Mass media and political persuasion. In T. C. Brock & M. C. Green (Eds.), Persuasion: Psychological insights and perceptions (2nd ed., pp. 225248). Thousand Oaks, CA: Sage. Iy engar, S., & McGrady, J. A. (2006). Media politics: A citizens guide. New York: Norton. Iyengar, S., & Simon, A. (1993). News coverage of the gulf crisis and public opinion: A study of agendasetting, priming and framing. Communication Research 20 (3), 365383.

PAGE 418

418 Johnson, T. J., Wanta, W., Byrd, J. T., & Lee, C. (1995). Exploring FDRs relationship with the press: A historical agendasetting study. Political Communication 12(2), 157172. Kaid, L. L. (1976). Newspaper treatment of a candidates news relea ses. Journalism Quarterly 53(1), 135137. Kaid, L. L., & Wadsworth, A. J. (1989). Content analysis. In P. Emmert & L. L. Barker (Eds.), Measurement of communication behavior (pp. 197217). New York: Longman. Kent, M. L., & Taylor, M. (1998). Building dialogic relationships through the World Wide Web. Public Relations Review 24(3), 321 334. Kieffer, N. (1983). Agendasetting and corporate communication issues: Can the mass media influence corporate stock prices? Unpublished masters thesis, Syracuse Univ ersity, Syracuse, NY. Kim, Y. (2001). Measuring the economic value of public relations. Journal of Public Relations Research 13 (1), 3 26. Kiousis, S. (2002). Killing the messenger: An exploration of presidential newspaper coverage and public confidence in the press. Journalism Studies 3 (4), 557572. Kiousis, S. (2003). Job approval and favorability: The impact of media attention to the Monica Lewinsky scandal on public opinion of President Bill Clinton. Mass Communication & Society 6 (4), 435 451. Kio usis, S. (2004a, November). Before and after: Examining the precursors and outcomes of agendasetting Paper presented at the annual meeting of the National Communication Association, Chicago. Kiousis, S. (2004b). Explicating media salience: A factor analysis of New York Times issue coverage during the 2000 U.S. presidential election. Journal of Communication, 54(1), 71 87. Kiousis, S. (2005). Compelling arguments and attitude strength: Exploring the impact of secondlevel agenda setting on public opinion of presidential candidate images. The Harvard International Journal of Press/Politics 10(3), 3 27. Kiousis, S., Kim, S., McDevitt, M., & Ostrowski, A. (2009). Competing for attention: Information subsidy influence in agenda building during election campaigns. Journalism & Mass Communication Quarterly 86(3), 545 562. Kiousis, S., & McCombs, M. (2004). Agendasetting effects and attitude strength: Political figures during the 1996 presidential election. Communication Research 31(1), 36 57.

PAGE 419

419 Kiousis, S., & McDevitt, M. (2008). Agenda setting in civic development: Effects of curricula and issue importance on youth voter turnout. Communication Research 35(4), 481 502. Kiousis, S., Mitrook, M., Popescu, C., Shields, A., Seltzer, T. (2006, June). First and secondlevel agenda building and agenda setting: terrorism, the president and the media. Paper presented at the annual meeting of the International Communication Association, Dresden, Germany. Kiousis, S., Mitrook, M., Wu, X., Seltzer, T. (2006). First and secondlevel agendabuilding and agendasetting effects: Exploring the linkages among candidate news releases, media coverage, and public opinion during the 2002 Florida gubernatorial election. Journal of Public Relations Research 18(3), 265 285. Kio usis, S., Popescu C., & Mitrook, M. (2007). Understanding influence on corporate reputation: An examination of public relations efforts, media coverage, public opinion, and financial performance from an agendabuilding and agendasetting perspective. Journ al of Public Relations Research, 19(2), 147 165. Kiousis, S., & Shields, A. (2008). Intercandidate agendasetting in presidential elections: Issue and attribute agendas in the 2004 campaign. Public Relations Review 34(4), 325 330. Kiousis, S., & Wu, X. ( 2008). International agendabuilding and agendasetting: Exploring the influence of public relations counsel on US news media and public perceptions of foreign nations. International Communication Gazette, 70 (1), 58 75. Klapper, J. (1960). The effects of mass communication. New York: Free Press. Klein, A., & Zur, E. (2009). Entrepreneurial shareholder activism: Hedge funds and other private investors. The Journal of Finance 64(1), 187229. Kaback, H. (1999). Gershon Kekst: Master of the saykhl business Directors & Boards 23(2), 16 28. Krosnick, J. A., & Kinder, D. R. (1990). Altering the foundations of support for the president through priming. American Political Science Review 84 (2), 497 512. Laide, J., & Mallea, J. (2009, January 23). 2008 year end review: New highs sets for unsolicited M&A and proxy fights. SharkRepellent.net Retrieved March 3, 2009, from https://www.sharkrepellent.net/request?an=dt.getPage&st=1&pg=/pub/rs_200901 22.html&rnd=643214

PAGE 420

420 Lancendorfer, K. M., & Lee, B. (2003, August). Agenda building and the media: A content analysis of the relationships between the media in the 2002 Michigan governors race Paper presented at the annual meeting of the Association for Education in Journalism and Mass Communication, Kansas City, MO. Laskin, A. V. (2006). Investor relations practices at Fortune 500 companies: An exploratory study. Public Relations Review 32 (1), 69 70. Laskin, A. V. (2009). A descriptive account of the investor relations profession: A national study. Journal of Business Communication, 46(2), 208 233. Lasorsa, D. L., & Wanta, W. (1990). Effects of personal, interpersonal and media experiences on issue saliences. Journalism Quarterly 67 (4). 804 813. Lazarsfeld, P., Berelson, B., & Gaudet, H. (1944). The peoples choice New York: Columbia University Press. Lee, B., Lancendorfer, K. M., & Lee, J. K. (2005). Agendasetting and the Internet: The intermedia influence of Internet bulletin boards on newspaper coverage of the 2000 general election in South Korea. Asian Journal of Communication, 15(1), 5771. Lippmann, W. (1922). Public opinion New York: MacMillan. Littlejohn, S. W. (2002). Theory in the process of inquiry. In S. W. Littlejohn, Theories of human communication (pp 18 33). Belmont, CA: Wadsworth. Lopez Escobar, E., Llamas, J. P., McCombs, M., & Lennon, F. R. (1998). Two levels of agenda setting among advertising and news in the 1995 Spanish elections Political Communication 15 (2), 225 238. Lynch, S. N. (2009, October 3). Activists, take note: SEC delays a proxy vote. The Wall Street Journal Retrieved October 3, 2009, fromhttp://online.wsj.com/article/SB125450519477659915.html MacKenzie Partners. ( 2008). Proxy services. Retrieved June 25, 2009, from http://www.mackenziepartners.com/proxy -services.html#solution Manheim, J. B. (1986). A model of agenda dynamics. In M. L. McLaughlin (Ed.), Communication yearbook 10 (pp. 499 516). Beverly Hills, CA: Sag e. Manne, H. G. (1965). Mergers and the market for corporate control. The Journal of Political Economy 73(2), 110120. Marens, R. (2002). Inventing corporate governance: The midcentury emergence of shareholder activism. Journal of Business and Management 8(4), 365389.

PAGE 421

421 Marken, G. A. (2005). Good year for a merger. Public Relations Quarterly 50(1), 21 22. Marston, C. (1996). The organization of the investor relations function by large UK quoted companies. Omega, 24(4), 477488. Matthes, J. (2005). The need for orientation towards news media: Revising and validating a classic concept. International Journal of Public Opinion Research, 18(4), 422 444. McClure, R. D., & Patterson, T. E. (1976). Print vs. network news. Journal of Communication, 26(2), 2328. McCombs, M. E. (1977). Newspapers vs. television: Mass communication effects across time. In D. L. Shaw & M. E. McCombs (Eds.), The emergence of American political issues: The agendasetting function of the press (pp. 89105). St. Paul, MN: West. McCombs, M. (1981). The agendasetting approach. In D. D. Nimmo & K. R. Sanders (Eds.), The handbook of political communication (pp. 75110). Beverly Hills, CA: Sage. McCombs, M. (2005). A look at agendasetting: past, present and future. Journalism Stu dies 6 (4), 543 557. McCombs, M. (2006). Setting the agenda: The mass media and public opinion. Malden, MA: Polity Press. McCombs, M. E., & Bell, T. (1996). The agendasetting role of mass communication. In M. Salwen & D. Stacks (Eds.), An integrated appr oach to communication theory and research (pp. 93110). Mahwah, NJ: Erlbaum. McCombs, M., & Evatt, D. (1995). Los temas y los aspectos: Exolorando una nueva dimension de la agenda setting. [Objects and attributes: Exploring a new dimension of agenda setting]. Communicacion y Sociedad, 8 (1), 7 32. McCombs, M., Llamas, J. P., Lopez Escobar, E., & Rey, F. (1997). Candidate images in Spanish elections: Secondlevel agendasetting effects. Journalism & Mass Communication Quarterly 74(4), 703717. McCombs, M ., Lopez Escobar, E., & Llamas, J. P. (2000). Setting the agenda of attributes in the1996 Spanish general election. Journal of Communication, 50(2), 7793. McCombs, M. E., & Shaw, D. L. (1972). The agendasetting function of mass media. Public Opinion Quarterly 36(2), 176187. McCombs, M. E., & Shaw, D. L. (1976). Setting the political agenda: Structuring the unseen environment. Journal of Communication, 26(2), 18 22.

PAGE 422

422 McCombs, M. E., & Shaw, D. L. (1993). The evolution of agendasetting research: Twen ty five years in the marketplace of ideas. Journal of Communication, 43(2), 5868. McCombs, M. E., & Weaver, D. H. (1973). Voters need for orientation and use of mass communication. Paper presented at the annual meeting of the International Communication Association, Montreal, Canada. McLeod, J. M., Becker, L. B., & Byrnes, J. E. (1974). Another look at the agendasetting function of the press. Communication Research 1 (2), 131 166. Meijer, M., & Kleinnijenhuis, J. (2006). Issue news and corporate reputation: Applying the theories of agenda setting and issue ownership in the field of business communication. Journal of Communication, 56 (3), 543 559. Miller, A. N. (1986). Predicting the winners of corporate takeovers. American Business Review 54 61. Mil ler, E. (1991). Investor relations. In P. Lesly (Ed.), Leslys handbook of public relations and communications (4th ed., 164 213). New York: Amacom. Miller, M. M., Andsager, J. L., & Reichert, B. P. (1998). Framing the presidents in presidential primaries: issues and images in press releases and news coverage. Journalism & Mass Communication Quarterly 75(2), 312 324. Minow, N. (2002, July 29). The gadfly shortage: Why didnt more shareholders make trouble before the corporate scandals? Slate Retrieved J une 23, 2009, from http://www.slate.com/?id=2068578 Monks, R. A. G., & Minow, N. (2008). Corporate governance (4th ed.). West Sussex, England: John Wiley & Sons. Morgan Joseph & Co. Inc. (2006). Management in an era of shareholder activism [Report]. Nabi, R. L., & Oliver, M. B. (2009). Mass media effects. In C. R. Berger, M. E. Roloff, & D. R. Roskos Ewoldsen (Eds.), The handbook of communication science (2nd ed., 255272). Thousand Oaks, CA: Sage. National Investor Relations Institute. (2003). Mission and goals Retrieved June 21, 2009, from http://www.niri.org/about/mission.cfm National Investor Relations Institute (2007). NIRI activist investor survey engage for success [Report].

PAGE 423

423 Neuendorf, K. A. (2002). The content analysis guidebook Thousand Oaks, CA: Sage. Newman, K. (1983). Financial communications and the contested take over bid, 19581982. International Journal of Advertising, 2 (1), 47 68. N ewsom, D., Scott, A., & Turk, J. (1993). This is PR: The realities of public relations Belmont, CA: Wadsworth. Niedziolka, D. A. (2007). Investor relations in Poland: An evaluation of the state of affairs from empirical studies. Public Relations Review, 33(4), 433436. Ohl, C. M., Pincus, J. D., Rimmer, T., & Harrison, D. (1995). Agenda building role of news releases in corporate takeovers. Public Relations Review 21 (2), 89 101. Palter, R. N., Rehm, W., & Shih, J. (2008, April). Communicating with the right investors. McKinsey Quarterly Page, B., & Shapiro, R. Y. (1992). The rational public: Fifty years of trends in Americans policy preferences Chicago: University of Chicago Press. Palmgreen, P., & Clarke, P. (1977). Agendasetting with local and national issues. Communication Research 4 (4), 435 452. Perloff, R. M. (1998). Political communication: Politics, press and public in America Mahwah, NJ: Erlbaum. Petersen, B. K., & Martin, H. J. (1996). CEO perceptions of investor relations as a public r elations function: An exploratory study. Journal of Public Relations Research 8 (3), 173 209. Petrocik, J. R., Benoit, W. L., & Hansen, G. J. (2004). Issue ownership and presidential campaigning, 19522000. Political Science Quarterly 118 (4), 599626. P ollock, T. G., Rindova, V. P. (2003). Media legitimation effects in the market for initial public offerings. Academy of Management Journal 46 (5), 631 642. Pollock, T. G., Rindova, V. P., & Maggitti, P. G. (2008). Market watch: Information and availability cascades among the media and investors in the U.S. IPO market. Academy of Management Journal 51(2), 335 358. Pound, J. (1992). Raiders, targets, and politics: The history and future of American corporate control. Journal of Applied Corporate Finance 5 (3), 6 18. Press lets its prejudice show in PR coverage how papers bite the hand that feeds them (1994, July 1). PR Week 2.

PAGE 424

424 Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1993). In search of how people change: Applications to addictive behaviors. Journal of Addictions Nursing, 5 (1), 216. Ragas, M., Kim, J., & Lim, H (2009). Agendabuilding and agendasetting in the corporate sphere: Analyzing influence in the 2008 Yahoo! Icahn proxy contest Paper presented at the annual meeting of the Annual Association for Education in Journalism and Mass Communication, Boston. Ragas, M. W., & Roberts, M. S. (2009). Agenda setting and agenda melding in an age of horizontal and vertical media: A new theoretical lens for virtual brand communities. Journali sm & Mass Communication Quarterly 86(1), 45 64. Ramius LLC (2008). Ramius white paper: The case for activist strategies [Report]. Rao, H., & Sivakumar, K. (1999). Institutional sources of boundary spanning structures: The establishment of investor relations departments in the Fortune 500 Industrials. Organization Science, 10(1), 27 42. Regester, M. (1990). Why investor relations is growing in importance and requires special skills. International Public Relations Review 13(1), 4 7. Reese, S. D., & Dan ielian, L. H. (1994). Intermedia influence and the drug issue: Converging on cocaine. In P. Shoemaker (Ed.), Communication campaigns against drugs: Government, media and the public (pp. 4766). Hillsdale, NJ: Erlbaum. Roberts, M. S. (1992). Predicting voter behavior via the agendasetting tradition. Journalism Quarterly 69(4), 878 892. Roberts, M., & McCombs, M. (1994). Agenda setting and political advertising: Origins of thenews agenda. Political Communication, 11 (3), 249262. Roberts, M., Wanta, W., & Dzwo, T. (2002). Agenda setting and issue salience online. Communication Research 29(4), 452 465. Rodriguez, R. (2000) University professors as communication media: agendasetting of studentsand professors Unpublished doctoral dissertation, Complutense University, Madrid,Spain. Rogers, E. M., & Chang, S. (1991). Media coverage of technology issues: Ethiopian drought of 1984, AIDS, Challenger, and Chernobyl. In L. Wilkens & P. Patterson (Eds.), Risky business: Communicating issues of science, risk, and public policy (pp. 7596). New York: Greenwood. Rogers, E. M., & Dearing, J. W. (1988). Agendasetting research: Where has it been? Where is it going? In J. A. Anderson (Ed.), Communication yearbook 11 (pp. 555 594). Newbury Park, CA: Sage.

PAGE 425

425 Rogers, E. M., & Dearing, J. W. (2000). Agendasetting research: Where has it been, where is it going? In D. A. Graber (Ed.), Media Power in Politics (pp. 6885). Washington, DC: CQ Press. Sallot, L. M., Lyon, L. J., Acosta Alzuru, C., & Jones, K. O. (2003). From aardvar k to zebra: A new millennium analysis of theory development in public relations academic journals. Journal of Public Relations Research 15(1), 27 90. Sant, R., & Zaman, M. A. (1996). Market reaction to Business Week Inside Wall Street column: A self ful filling prophecy. Journal of Banking & Finance 20(4), 617643. Sauerhaft, S. (1997). The role of public relations in mergers and acquisitions. In. C. L. Caywood (Ed.), The handbook of strategic public relations & integrated communications (pp. 301310). New York: McGraw Hill. Scott, W. (1955). Reliability of content analysis: The case of nominal scale coding. Public Opinion Quarterly 17, 321 325. Seltzer, T., & Dittmore, S. W. (2009). Down, set, frame: Secondlevel agenda building and the NFL network carriage dispute. International Journal of Sport Communication, 2 (3), 340359. Severin, W. J., & Tankard, J. W. (2001). Communication theories: Origins, methods, and uses in the mass media (5th ed.). New York: Addison Wesley Longman. Shaw, D. L., & Marti n, S. E. (1992). The function of mass media agenda setting. Journalism Quarterly, 69(4), 902 920. Shaw, D., & McCombs, M. (1977). The emergence of American political issues St. Paul, MN: West. Sheafer, T. (2007). How to evaluate it: The role of story eva luative tone in agenda setting and priming. Journal of Communication, 57 (1), 21 39. Shoemaker, P. J., Tankard, J. W., & Lasorsa, D. L. (2004). How to build social science theories Thousand Oaks, CA: Sage. Sigal, L. V. (1973). Reporters and officials: The organization and politics of newsmaking. Lexington, MA: Heath. Silver, D. (2004). The IR PR nexus. In B. M. Cole (Ed.), The new investor relations: Expert perspectives on the state of the art (pp. 149 167). New York: Bloomberg Press. Simon, A. F. (200 2). The winning edge: Candidate behavior, campaign discourse, and democracy. New York: Cambridge University Press.

PAGE 426

426 Slater, M. D. (1999). Integrating application of media effects, persuasion, and behavior change theories to communication campaigns: A stag es of change framework. Health Communication, 11(4), 335 354. Smith, R. D. (2009). Strategic planning for public relations (3rd ed.). New York: Routledge. Stevenson, R. L. (2000). In praise of dumb clerks: Computer assisted content analysis. In M. D. West (Ed.), Theory, method, and practice in computer content analysis (pp. 112). Westport, CT: Ablex. Sutherland, M., & Galloway, J. (1981). Role of advertising: Persuasion or agenda setting? Journal of Advertising Research, 21(5), 2529. Sweetser, K. D., Golan, G. J., & Wanta, W. (2008). Intermedia agenda setting in television, advertising, and blogs during the 2004 election. Mass Communication and Society 11 (2), 197 216. Talner, L. (1983). The origins of shareholder activism Investor Responsibility Research Center, Inc., Washington D.C. Takeshita, T. (1997). Exploring the medias role in defining reality: From issueagenda setting to attribute agenda setting. In M. McCombs, D. Shaw, & D. Weaver (Eds.), Communication and democracy: Exploring the intellect ual frontiers in agendasetting theory (pp. 1527). Mahwah, NJ: Erlbaum. Tan, Y., & Weaver, D. H. (2007). Agendasetting effects among the media, the public and Congress, 19462004. Journalism & Mass Communication Quarterly 84(4), 729744. Tedesco, J. C (2001). Issue and strategy agendasetting in the 2000 presidential primaries. American Behavioral Scientist 44 (12), 20482067. Tedesco, J. C. (2005a). Intercandidate agenda setting in the 2004 Democratic presidential primary. American Behavioral Scient ist 49(1), 92 113. Tedesco, J. C. (2005b). Issue and strategy agenda setting in the 2004 presidential election:Exploring the candidatejournalist relationship. Journalism Studies 6 (2), 187201. Tedesco, J. C., & McKinnon, L. M. (1998, November). Agenda building and presidential campaign press releases: A comparison of print and online publications Paper presented at the annual meeting of the National Communication Association, New York. Thompson, T. A., & Davis, G. F. (1997). The politics of corporate control and the future of shareholder activism in the United States. Corporate Governance, 5 (3), 152 159.

PAGE 427

4 27 Turk, J. V. (1985). Information subsidies and influence. Public Relations Review 11(3), 1025. Turk, J. V., & Franklin, B. (1987). Information subsidies: Agendasetting traditions. Public Relations Review 13(4), 29 41. Useem, M. (1996). The new investor capitalism New York: Basic Books. U.S. Securities and Exchange Commission. (2009). Filings & forms Retrieved July 14, 2009, from http://www.sec.gov/edgar.shtml Vogel, D. (1983). Trends in shareholder activism: 19701982. California Management Review 25(3), 68 87. Wanta, W. (1991). Presidential approval ratings as a variable in the agendabuilding process. Journalism Quarterly 68 (4), 672 679. Wanta, W. (1997). The messenger and the message: Differences across news media. In M. McCombs, D. Shaw, & D. Weaver (Eds.), Communication and democracy: Exploring the intellectual frontiers in agendasetting theory (pp. 137 157). Mahwah, NJ: Erlbaum. Wanta, W., & Foote, J. (1994). The president news media relationship: A time series analysis of agendasetting. Journal of Broadcasting & Electronic Media, 38 (4), 437449. Wanta, W., & Ghanem, S. (2007). Effect s of agenda setting. In R. W. Preiss, B. M. Gayle, N. Burrell, M. Allen, & J. Bryant (Eds.), Mass media effects research: Advances through metaanalysis (pp. 37 52). Mahwah, NJ: Erlbaum. Wanta, W., & Hu, Y. (1994). The effects of credibility, reliance and exposure on media agendasetting: A path analysis model. Journalism Quarterly 71(1), 90 98. Wanta, W., Stephenson, M. A, Turk, J. V., & McCombs, M. (1989). How Presidents State of Union talk influenced news media agendas. Journalism Quarterly 66 (3), 5 37541. Wanta, W., & Wu, Y. (1992). Interpersonal communication and the agendasetting process. Journalism Quarterly 69 (4), 847 855. Weaver, D.H. (1977). Political issues and voter need for orientation. In D. L. Shaw & M. E. McCombs (Eds.), The emergence of American political issues: The agendasetting function of the press (pp. 107119). St. Paul, MN: West. Weaver, D.H. (1980). Audience need for orientation and media effects. Communication Research, 7(3), 361 376.

PAGE 428

428 Weaver, D. H. (1984). Media agenda-setting and public opinion: Is there a link? In R. N. Bostrom (Ed.), Communication Yearbook 8. Beverly Hills, CA: Sage. Weaver, D. H. (1991). Issue salience and public opinion: Are there consequences of agendasetting? International Journal of Public Opinion Research 3 (1), 53 68. Weaver, D., Graber, D., McCombs, M., Eyal, C. (1981). Media agenda setting in a presidential election: issues, images and interest Westport, CT: Greenwood. Weaver, D. H., McCombs, M. E., & Spellman, C. (1975). Watergate and the media: A case study of agendasetting. American Politics Quarterly 3 (4), 458 472. Westbrook, J. (2009, May 20). SEC weighs 1 percent threshold for board nominations. Bloomberg. Retrieved June 28, 2009, from http://www.bloomberg.com/apps/news?pid=20670001&sid=aTsdyGHJu59M Wilcox, D. L., Ault, P. H., & Agee, W. K. (1995). Public relations strategies and tactics (4th ed.). New York: Harper Collins. Wilcox, J. C. (2004). The art of winning proxy wars. In B. M. Cole (Ed.), The new investor relations: Expert perspectives on the state of the art (pp. 149 167). New York: Bloomberg Press. Wimmer, R. D., & Dominick, J. R. (2006). Mass media research: An introduction (8th ed.). Boston: Wadsworth. Winter, J. P. (1981). Contingent conditions in the agendasetting pr ocess. Mass communication review yearbook 2 (pp. 235243). Beverly Hills, CA: Sage. Winter, J., & Eyal, C. (1981). Agenda setting for the civil rights issue. Public Opinion Quarterly 45(3), 376383. Younglai, R. (2009, May 15). SEC mulls awarding more power to nominate directors. Reuters .Retrieved May 16, 2009, from http://www.reuters.com/article/ousiv/idUSTRE54E5G120090515 Zhou, Y., & Moy, P. (2007). Parsing framing processes: The interplay between online public opinion and media coverage. Journal of Communications 57 (1), 79 98. Zucker, H. (1978). The variable nature of news media influence. In B. Ruben (Ed.), Communication yearbook 2. New Brunswick, NJ: Transaction.

PAGE 429

429 BIOGRAPHICAL SKETCH Matthew W. Ragas was born in Metairie, Louisiana. He attended the University of Central Florida in Orlando, where he earned a Master of Science in m anagement and a Bachelor of Science in business admi nistration with an emphasis in marketing. Bef ore returning to school to pursue a Doctor of Philosophy in mass c ommunication at the University of Florida, Matt gained a decade of experience working in industry, primarily in corporate communication and business journalism roles. During this t ime he authored two trade books with Crown Business helped start two successful online media companies, and consulted on communication strategy in the U.S. and abroad. Matt links theory and practice in his research and teaching. His interdisciplinary mass media -e ffects focused research examines the interplay among financial public relations efforts (information subsidies), business journalism coverage, and investor opinion from an agendabuilding/setting, media framing, and media priming perspective. His secondary research interests include political and brand communications. While he remains a Gator at heart, Matt is now an Assistant Professor of Public Relations in the College of Communication at DePaul University. He looks forward to a productive career in ac adem ia and exploring all that the Windy City has to offer He is staying out of the Cubs and White Sox debate by remaining a Tampa Bay Rays fan. For additional information or to contact Matt please visit: http://www.mattragas.com or e mail him at: matt.ragas@gmail.com