<%BANNER%>

A Coorientational Approach for Measuring Organization-Public Relationships

Permanent Link: http://ufdc.ufl.edu/UFE0021250/00001

Material Information

Title: A Coorientational Approach for Measuring Organization-Public Relationships
Physical Description: 1 online resource (173 p.)
Language: english
Creator: Seltzer, Trenton C
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: coorientation, evaluation, maintenance, management, measurement, opr, organization, public, relationship, strategies
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: My study details the development of a new methodology for measuring and diagnosing organization-public relationships. Using a co-orientational framework, the Hon-Grunig (1999) relationship scales were adapted to measure not only the direct perspective of a stakeholder public, but also the direct perspective of the organization in the relationship. This represents a departure from current methods of measuring an organization-public relationship that typically only include the opinions of the public. My study goes a step further by including the meta-perspectives of each party as well (i.e., what one party believes the other party thinks about the relationship). By looking at the interaction between these four perspectives (publics direct perspective, publics meta-perspective, organizations direct perspective, and organizations meta-perspective), it is possible to determine what type of coorientational state best describes the relationship. Possible outcomes include true consensus (accurate perceptions of agreement between the parties), true dissensus (accurate perceptions of disagreement between the parties), false consensus (inaccurate perception of agreement between the parties) and false dissensus (inaccurate perception of disagreement between the parties). In my study, the organization-public relationship between the university police department and students living on campus at a major southwestern university is examined. The resulting coorientational measures indicated that while students were well aware that there was disagreement between them and the campus police, the police themselves were unaware of this disagreement and thought that students rated the relationship as strongly as the police. From the organizations viewpoint, the relationship was in a state of false consensus. The effect of additional variables, including interpersonal interaction, time in the relationship, exposure to relationship maintenance strategies, and media exposure, were also examined. While neither one-way nor two-way relationship maintenance strategies demonstrated any influence, the students perception of their interpersonal interactions with the campus police was a significant predictor of their ratings of the relationship. The study extends relationship management theory by developing a process for generating a comprehensive relationship construct. The findings suggest that two-way symmetrical strategies may not always be the ideal means for maintaining organization-public relationships and stress the important role that interpersonal interaction plays in maintaining organization-public relationships.
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 Trenton C Seltzer.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Mitrook, Michael A.
Local: Co-adviser: Roberts, Marilyn.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021250:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021250/00001

Material Information

Title: A Coorientational Approach for Measuring Organization-Public Relationships
Physical Description: 1 online resource (173 p.)
Language: english
Creator: Seltzer, Trenton C
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: coorientation, evaluation, maintenance, management, measurement, opr, organization, public, relationship, strategies
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: My study details the development of a new methodology for measuring and diagnosing organization-public relationships. Using a co-orientational framework, the Hon-Grunig (1999) relationship scales were adapted to measure not only the direct perspective of a stakeholder public, but also the direct perspective of the organization in the relationship. This represents a departure from current methods of measuring an organization-public relationship that typically only include the opinions of the public. My study goes a step further by including the meta-perspectives of each party as well (i.e., what one party believes the other party thinks about the relationship). By looking at the interaction between these four perspectives (publics direct perspective, publics meta-perspective, organizations direct perspective, and organizations meta-perspective), it is possible to determine what type of coorientational state best describes the relationship. Possible outcomes include true consensus (accurate perceptions of agreement between the parties), true dissensus (accurate perceptions of disagreement between the parties), false consensus (inaccurate perception of agreement between the parties) and false dissensus (inaccurate perception of disagreement between the parties). In my study, the organization-public relationship between the university police department and students living on campus at a major southwestern university is examined. The resulting coorientational measures indicated that while students were well aware that there was disagreement between them and the campus police, the police themselves were unaware of this disagreement and thought that students rated the relationship as strongly as the police. From the organizations viewpoint, the relationship was in a state of false consensus. The effect of additional variables, including interpersonal interaction, time in the relationship, exposure to relationship maintenance strategies, and media exposure, were also examined. While neither one-way nor two-way relationship maintenance strategies demonstrated any influence, the students perception of their interpersonal interactions with the campus police was a significant predictor of their ratings of the relationship. The study extends relationship management theory by developing a process for generating a comprehensive relationship construct. The findings suggest that two-way symmetrical strategies may not always be the ideal means for maintaining organization-public relationships and stress the important role that interpersonal interaction plays in maintaining organization-public relationships.
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 Trenton C Seltzer.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Mitrook, Michael A.
Local: Co-adviser: Roberts, Marilyn.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021250:00001


This item has the following downloads:


Full Text

PAGE 1

1 A COORIENTATIONAL APPROACH FOR MEASURING ORGANIZATION-PUBLIC RELATIONSHIPS By TRENT SELTZER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Trent Seltzer

PAGE 3

3 To Rick

PAGE 4

4 ACKNOWLEDGMENTS I thank my committee members for their enc ouragement and advice. I would especially like to thank my advisor and committee chair, Michael Mitrook. Dr. Mitrook has been a great teacher, mentor, and friend to me for many year s now. I appreciate all of his guidance and support. I thank my family and my friends for their l ove and kindness over the la st four years. Most of all, I owe the world to my wonderful, understanding (and incr edibly patient) wife, Laura. Thank you for putting up with me throughout this entire process. Finally, I would like to extend a very special thank you to my father. Every time life seemed too difficult and that I would never see the light at the end of this particular tunnel, I could always turn to the example that he set to find the motivation to keeping going. The way he lived his life in overcoming the challenges that were forced on him was all the inspiration I needed. I thank him for teaching me the meaning of the word “persistence.”

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .........9 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 INTRODUCTION................................................................................................................. .12 Purpose and Importance of Study...........................................................................................12 2 LITERATURE REVIEW.......................................................................................................14 The Relationship Perspective.................................................................................................14 Defining and Measuring Relationships..................................................................................22 Control Mutuality............................................................................................................25 Trust.......................................................................................................................... .......26 Satisfaction................................................................................................................... ...26 Commitment....................................................................................................................2 6 Exchange Relationships...................................................................................................27 Communal Relationships.................................................................................................27 Coorientation.................................................................................................................. ........28 Coorientational Approach....................................................................................................... 31 The Value of the Coorientational Approach...........................................................................34 Research Questions............................................................................................................. ....38 Hypotheses..................................................................................................................... .........39 3 METHODOLOGY.................................................................................................................4 4 Integrating the Coorienta tional Approach with Relationship Measurement..........................44 Adapting the Relational Dimensions fo r the Coorientational Approach.........................44 Sample......................................................................................................................... ....46 Survey Administration.....................................................................................................47 Data Analysis.................................................................................................................. .48 Pilot Study.................................................................................................................... ..........49 Method......................................................................................................................... ....50 Pilot Study Results..........................................................................................................51 Summing of scales...................................................................................................51 Scale Reliability.......................................................................................................52 Demographics...........................................................................................................52 Communication........................................................................................................53

PAGE 6

6 Agreement................................................................................................................53 Perceived Agreement...............................................................................................53 Accuracy...................................................................................................................54 Conclusions from Pilot Study..........................................................................................54 Evaluation of Pilot Study.................................................................................................57 Main Study..................................................................................................................... .........58 Organization & Stakeholder Public.................................................................................59 Mode of Survey Administration......................................................................................60 Sampling and Recruitment..............................................................................................64 Survey Instrument Construction......................................................................................68 Operationalization of Independent Variables..................................................................69 Party Membership....................................................................................................69 Relationship Maintenance Strategies.......................................................................69 Interpersonal Interaction..........................................................................................70 Time.........................................................................................................................71 Media Exposure........................................................................................................71 Demographics...........................................................................................................71 Operationalization of Dependent Variables....................................................................72 Direct Perspective....................................................................................................72 Meta-Perspective......................................................................................................72 Data Analysis Strategy....................................................................................................74 4 RESULTS...................................................................................................................... .........82 Preparing the Data for Analysis..............................................................................................82 Item-Missing Data...........................................................................................................83 Normality...................................................................................................................... ...85 Scale Reliability.............................................................................................................. .86 Sample Demographics............................................................................................................ 88 Student Sample................................................................................................................8 8 University Police Department Sample............................................................................90 Results of the Statistical Analysis...........................................................................................9 0 Results Related to the Research Questions.....................................................................91 Analyses Related to the Hypotheses................................................................................97 Calculation of Add itional Variables.........................................................................97 Multiple Regression Analysis..................................................................................99 5 DISCUSSION................................................................................................................... ....109 Evaluation of the Coorientational Approach........................................................................109 Factors Affecting the Relationshi p and Coorientation Measures.........................................113 Theoretical Im plications....................................................................................................... 119 Implications for the Measurement of Relationships......................................................119 Implications for Relations hip Management Theory......................................................121 Implications for Public Relations Practice...........................................................................127 Limitations.................................................................................................................... ........131 Suggestions for Future Research..........................................................................................137

PAGE 7

7 In Closing..................................................................................................................... .........143 APPENDIX A RELATIONSHIP MEASURES...........................................................................................145 B PILOT STUDY SURVEYS.................................................................................................146 Public Version................................................................................................................. .....146 C MAIN STUDY SURVEYS..................................................................................................151 Organization Version........................................................................................................... .151 Public Version................................................................................................................. .....154 D RECRUITMENT EMAILS..................................................................................................160 Organization Members.........................................................................................................16 0 Organization – Recontact.....................................................................................................16 1 Students....................................................................................................................... ..........162 LIST OF REFERENCES............................................................................................................. 164 BIOGRAPHICAL SKETCH.......................................................................................................173

PAGE 8

8 LIST OF TABLES Table page 3-1 Management and staff direct perspectives.........................................................................76 3-2 Management direct a nd meta-perspectives........................................................................77 3-3 Staff direct and meta-perspective.......................................................................................78 3-4 Management direct perspectiv e to staff meta-perspective.................................................79 3-5 Staff direct perspective to management meta-perspective.................................................80 4-1 Main study organization and public direct and meta-perspectives..................................106 4-2 Correlation matrix of independ ent and dependent variables...........................................107

PAGE 9

9 LIST OF FIGURES Figure page 2-1 The coorientational model of organization-public relationships.......................................43 3-1 Coorientation model of pilot study organization’s mana gement-staff relationship...........81 4-1 Coorientation model of main st udy organization-public relationship.............................108

PAGE 10

10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy A COORIENTATIONAL APPROACH FOR MEASURING ORGANIZATION-PUBLIC RELATIONSHIPS By Trent Seltzer August 2007 Chair: Michael A. Mitrook Cochair: Marilyn S. Roberts Major: Mass Communication My study details the development of a ne w methodology for measuring and diagnosing organization-public relationships. Using a coor ientational framework, the Hon-Grunig (1999) relationship scales were adapted to measure not only the direct perspective of a stakeholder public, but also the direct perspective of the orga nization in the relationship. This represents a departure from current methods of measuring an organization-public relati onship that typically only include the opinions of the public. My st udy goes a step furthe r by including the metaperspectives of each party as well (i.e., what one party believes the other party thinks about the relationship). By looking at the interaction between th ese four perspectives (public’s direct perspective, public’s meta-persp ective, organization’s direct pe rspective, and organization’s meta-perspective), it is possible to determine what type of coorientational state best describes the relationship. Possible outcomes include true consensus (accurate perceptions of agreement between the parties), true dissens us (accurate perceptions of di sagreement between the parties), false consensus (inaccurate perception of agreem ent between the parties) and false dissensus (inaccurate perception of disa greement between the parties).

PAGE 11

11 In my study, the organizationpublic relationship between the university police department and students living on campus at a major southw estern university is examined. The resulting coorientational measures indicated that while students were well aware that there was disagreement between them and the campus police, the police themselves were unaware of this disagreement and thought that students rated the re lationship as strongly as the police. From the organization’s viewpoint, the relationship was in a state of false consensus. The effect of additional variables, includi ng interpersonal interaction, time in the relationship, exposure to relati onship maintenance strategies, a nd media exposure, were also examined. While neither one-way nor two-way rela tionship maintenance strategies demonstrated any influence, the students’ per ception of their interpersonal interactions with the campus police was a significant predictor of thei r ratings of the relationship. The study extends relationship management th eory by developing a process for generating a comprehensive relationship construct. The findings suggest that two-way symmetrical strategies may not always be the ideal means for maintaining organization-public relationships and stress the important role that interpersona l interaction plays in maintaining organizationpublic relationships.

PAGE 12

12 CHAPTER 1 INTRODUCTION Previous public relations research has stresse d that the quality of the relationship between an organization and its publics is an indication of public relations effectiven ess (Dozier, Grunig, & Grunig, 1995). While it seems intuitive that p ublic relations should de monstrate its greatest impact on the organization-public relationship, earl y perspectives on the role of public relations within an organization did not always recognize this, focusing instead on one-way models of public relations. Perhaps due to the practitione r-focused research generated while the discipline was in its infancy (Ferguson, 1984), evaluating th e effectiveness of public relations activities consisted primarily of measuring the short-term, immediate results of a public relations program (“outputs”) or assessing the impact that the program had on a target audience (“outcomes”). While it is necessary for public relations profe ssionals to monitor these outputs and outcomes, focusing on these factors will only yield informa tion about the success of an individual public relations program (Hon & Grunig, 1999). To gauge the true effectiveness of public relations over time, a long-term perspective needed to be taken, requiring not only a new way of measuring the effect of public relations efforts, but a complete shift in the focus of public relations research and a new way of thinking about organi zations and their publics. Purpose and Importance of Study My study details the development of a rela tionship-centered method for measuring public relations effectiveness. It focuses on the imp act of public relations pr ogramming, organizational communication, and interpersona l communication between members of an organization and its stakeholders on the quality of the relationship between an organization and its publics by using established relationship measures within a co orientational framework. Unlike previous approaches, this proposal outlines a method that will attempt to include the perspectives and

PAGE 13

13 meta-perspectives of both parties (the organiza tion and the public) in an assessment of the organization-public relationship (OPR) by combini ng the coorientational approach advocated by Broom (1977) and Broom and Dozier (1990) with the relationship measures proposed by Hon and Grunig (1999). Applying these relationship meas ures within the coorientational framework will indicate the degree of agreement, accura te perception, and perception of perceived agreement (congruency) between organizations and their publics when assessing important relationship dimensions. This should provide more information about the shared perception of the state of the OPR that moves beyond measurem ent of perceptions of any one party in the relationship. Therefore, this study addresses the need to develop th e conceptualization of organization-public relationship measurement from a theoretical standpoint to advance relationship management theory. The study also a ddresses the practical needs of public relations practitioners by refining measurement tools that focus on long-term measures of relationships rather than the short-term outputs and outcome s as a means of demonstrating the value of managed organizational communication programs.

PAGE 14

14 CHAPTER 2 LITERATURE REVIEW The Relationship Perspective The development of the relationship perspective in public relations research is in many ways tied to the efforts of public relations pr actitioners and scholars to distinguish public relations as a profession and a th eoretical paradigm separate from similar disciplines such as marketing and advertising. One of the more recent and commonly accepted definitions of public relations defines the practice as “the manage ment function that establishes and maintains mutually beneficial relationships between an organization and the publics on whom its success or failure depends” (Cutlip, Center, & Broom, 2000, p. 6), where “publics” refers to “people who are somehow mutually involved or interdependent with these or ganizations” (Cutlip, Center, & Broom, 2000, p. 2). Using this definition, public relations can be differentiated somewhat from marketing and advertising. According to Cutlip, Center, a nd Broom (2000), whereas marketing typically focuses on a specific type of relationship that a business maintains with its customers (a very specific public) in which the focus is on the exch ange of resources, public relations is broader and includes the management of a wide variety of possible relati onships with a wide variety of stakeholder publics that can affect the organi zation – not just customers, but also the government, members of the community, activists, donors, etc. The function also differs from advertising, which from a public relations perspective, is but one tool that can be used for facilitating communication. Whereas marketing would use advertisi ng efforts to reach potential or current customers, public rela tions practitioners frequently use advertising as a form of controlled media to complement other forms of communication (e.g., press releases, press conferences, media tours, etc.).

PAGE 15

15 In an effort to further explicate the academ ic study of public relations from other fields, Ferguson (1984) proposed that public relations research should focus on finding a dominant paradigm, as well as a new unit of analysis, whic h public relations could call its own. She felt that the most promising candidate was the relations hip that exists between an organization and its publics an “organization-public re lationship,” or “OPR” for short. Ferguson’s content analysis of the journal articles in Public Relations Review over a nine-year period led her to propose that there was not only a need for further development of unique public relations theory, but that the study of relationships themselves should be the primary focus in public relations research. Ferguson states: If I were to want to put my public relations theory development eggs in one basket, this would be it. It is difficult to think of a ny other field where the primary emphasis is on the relationships between organiza tions, between organizations a nd one or more groupings in society, or, more generally, w ith society itself (p.16). Ferguson (1984) felt that focusing on relations hips as a paradigm for public relations would have several benefits. First, it would make the relationships themselves the focus of public relations research rather than the study of th e parties involved in the relationship. Second, making relationships a primary unit of analysis for studying public relatio ns would lead to the development of new methodologies that would be necessary to measure those relationships. Third, focusing on relationships would create a niche for public relations researchers and help legitimize the academic study of the practice. Ad ditionally, the investigation of relationships would allow research from other fields, such as organizational communication and interpersonal communication, to be applied within a public relations context. This relationship-focused perspective also fit naturally with the development of the twoway symmetrical communication model for public relations. Grunig (1984) and Grunig and Hunt (1984) proposed four models for the practi ce of public relations: press agentry, public

PAGE 16

16 information, two-way asymmetrical, and two-way symmetrical. While the press agentry (or pure publicity) and public information (or journalistic) approaches to th e practice of public relations are seen as one-way models in which the practi tioner communicates to a target public without feedback, the two-way models in corporate a feedback loop, but for different purposes. The twoway asymmetrical model, or scientific persuasion model, uses feedback to refine persuasive efforts; however, these efforts to persuade are ma de with organizational in terests in mind that do not necessarily represent the best interests of th e target audience. On the other hand, the two-way symmetrical model also uses feedback, but does so in an effort to increase understanding between an organization and a public to develop rela tionships that lead to mutual benefit (Grunig & Grunig, 1992). The two-way mo del has been proposed as bei ng an ethical model of public relations (Grunig, 2001) in that it sees communication as the end and not the means and that the interests of the public are balanced with those of the organization. While originally viewed as a model in which practitioners may have to sacrifi ce the interest of their organization in order to gain the cooperation of a stakeholder public a nd vice versa, the model has since been refined somewhat to acknowledge that it is a “mixed-motive” model in which both sides try to advance their interests and achieve their desired goals, but do so through compromise and negotiation so that a solution is reached that sa tisfies all the parties involved and leads to a mutually beneficial relationship (Grunig, 2001; Murphy, 1991). Th erefore, from the two-way symmetrical viewpoint, the purpose of public relations is to develop and manage relationships, not control public opinion through persuasion (Ehling, 1992). Two-way symmetrical communication has been identified as one of the ways in which organizations can practice “excellent” public rela tions (Dozier, Grunig, & Grunig, 1995; Grunig, Grunig, & Dozier, 2002). In 1985, the Internat ional Association of Business Communicators

PAGE 17

17 Research Foundation funded what has come to be known as the Excellence Study. The study qualitatively analyzed 327 organizations from the United States, Canada, and the United Kingdom to determine how the public relations function can make an organization more effective, what that contribution is worth, and what characteristics of the public relations function lead to more effective communica tion management, which in turn leads to organizational effectiveness (Doz ier, Grunig, & Grunig, 1995). The researchers determined that public relations leads to increa sed organizational effectiveness when public relations identifies strategic publics that ca n affect the organization and then he lps establish, build, and maintain long-term relationships with those publics. Th ese relationships should be based on trust and understanding, and employing the two-way symmetrical model of public relations is the best way to facilitate the development of these relationshi ps. By building healthy relationships with these strategic publics throug h effectively managed communication pr ograms, public relations helps to manage, minimize, or avoid conflict with thos e publics and thus allows the organization to pursue its objectives. Ferguson’s (1984) call to focus on relationships as a unit of analysis for public relations coupled with Dozier, Grunig, & Grunig’s ( 1995) acknowledgement of the importance of twoway communication in managing organization-public relationships that lead to organizational effectiveness provides the groundw ork for the relationship management perspective. Ledingham and Bruning (2000a) elaborated on the importance of such a perspective: The emergence of relationship management as a paradigm for public relations scholarship and practice calls into question the essence of p ublic relations what it is and what it does or should do, its function and value within the organizational structure and the greater society, and the benefits generated not only for sponsoring organizations but also for the publics those organizations serv e and the communities and soci eties in which they exist. The relationship paradigm also provides a fr amework in which to explore the linkage between public relations objectiv es and organizational goals, fo r constructing platforms for

PAGE 18

18 strategic planning and tactical implementati on, and approaching programmatic evaluation in ways understood and appreciated by th e ruling management group (p. xiii). While the relationship management perspective has been advocated as an attractive paradigm for public relations theory development, it has yet to become the dominant paradigm. Sallot, Lyon, Acosta-Alzuru, and Jones ( 2003) extended Ferguson’s (1984) research by conducting a content analysis of 748 articles in the three major public relations journals since their founding through 2000. While their results indi cated that the tota l amount of work contributing to public relations theory developmen t had increased signifi cantly from 4% to 20% since Ferguson’s study, not only had a dominant para digm failed to emerge, but only 10% of the theory development articles focused on organiza tion-public relationships, indicating the need for more research in this area. This is not to say that the call for further development of the relational perspective has been abandoned; several scholars have pursued the investigation of pub lic relationships and attempted to tie it to other areas such as measur ing public relations effectiveness (Sallot et al., 2003). Broom, Casey, and Ritchey (1997) addre ssed one of the major obstacles to the development of the relationship pe rspective when they stressed the need for a common definition of the term “relationship.” Thei r review of relationshi p theory in the fields of public relations, interpersonal relations, family relations, group dynamics, organizational relations, psychotherapy, and international re lations found a lack of a common definition of what exactly is meant by the term “relationship.” In their opini on, this presented a barrier that would prevent scholars from studying the actual relationship be tween an organization and its publics; this would result in a continued focus on indirect means of measurement that would only allow inferences about the relationship to be ma de (i.e., outputs and outcomes). Among their

PAGE 19

19 conclusions and suggestions for further study of relationships, Broom, Casey, and Ritchey (1997) made two important observations: The formation of relationships occurs when pa rties have perceptions and expectations of each other, when one or both parties need re sources from the other, when one or both parties perceive mutual threats from an uncerta in environment, and when there is either a legal or voluntary necessity to associate… Relationships are the dynamic results of the exchanges and reciprocity that manifest th emselves as the relationships develop and evolve, but they can be described at a given point in time (p. 95). Broom, Casey, and Ritchey (1997) also propos ed a model that could be used by those seeking to further develop the relationship pers pective. Their three-stage model acknowledged that there was a need to identify the anteced ents, relationship state, and consequences of organization-public relationships. Antecedents of relationships included perceptions, motives, needs, and behaviors of parties within the relationship. Consequences included the outputs of relationships that could affect the organizati on or the environment it which it operates; Broom, Casey, and Ritchey identified th ese as “goal achievement, dependency/loss of autonomy, and routine and institutionalized behavior” (p. 94). In th eir model, the relationship itself could act as an independent variable, a moderating variable, or a dependent variable as had been previously acknowledged by Ferguson (1984). For example, Ledingham, Bruning, and Wilson (1999) used relationship state as an intervening variable with time in the relationship acting as an antecedent condition and behavioral predispositions towa rds an organization as a consequence of relationship state. Perceptions of organization-pub lic relationships have been used to differentiate between members of a stakeholde r public in terms of loyalty (Ledingham and Bruning, 1998) and satisfaction (Ledingham, 2001; Ledingham and Bruning, 1998). Ki and Hon (2007) tested a model of organization-public relati onships in which they identified the effects of a stakeholder public’s perceptions of an orga nization-public relationshi p on their attitudes and

PAGE 20

20 behavioral intentions toward the organization; stronger evaluations of the relationship were demonstrated to lead to more positive attitude s which in turn led to more positive behaviors. Grunig and Huang (2000) also proposed a three-stage model of organization-public relationships that parallels the Broom, Casey, and Ritchey (1997) model. Grunig and Huang’s (2000) model includes situational antecedents maintenance strategies, and relationship outcomes. Here, situational antecedents describe the types of behavioral and situational factors that link publics and organizations to one another. Maintenance stra tegies include the efforts of the organization to develop and nurture its rela tionship with the public through symmetrical and asymmetrical communication efforts. Symmetrical strategies benefit both parties in the relationship and include positivity, disclosure, assurances of legitimacy, networking, integrative negotiation, cooperation, and sharing of task s. Asymmetrical stra tegies cater to the organization’s interests over those of the public s and include distributive negotiation, avoiding, contending, compromising, and accommodating. Re lationship outcomes include goal attainment and perceptions of relationship state. In his review of the literature on organi zation-public relationships, Ledingham (2003) summarized the relationship pers pective by suggesting a theory of relationship management that states “effectively managing organi zation-public relationships around common interests and shared goals over time results in mutual understanding and be nefit for interacting organizations and publics [emphasis added]” (p. 190). Leding ham believed that relationship management could serve as a useful framework for organi zing academic and applied endeavors in public relations. This perspect ive acknowledges that public relati ons is not solely a communication function, but uses communication st rategically in an effort to manage relationships (Ledingham & Bruning, 1998).

PAGE 21

21 Ledingham (2003) also summarized the major fi ndings of the research on the relational perspective by proposing a list of axioms of organization-public relationships. These axioms include: 1. Organization-public relations hips are transactional. 2. The relationships are dynamic; they change over time. 3. They are goal oriented. 4. Organization-public relationshi ps have antecedents and consequences and can be analyzed in terms of relationship quality, maintenance strategies, relationship type, and actors in the relationship. 5. These relationships are driv en by the perceived needs and wants of interacting organizations and publics. 6. The continuation of organizat ion-public relationships is dependent on the degree to which expectations are met. 7. Those expectations are expressed in intera ctions between organizations and publics. 8. Such relationships involve communicati on, but communication is not the sole instrument of relationship building. 9. These relationships are impact ed by relational hist ory, the nature of the transaction, the frequency of exchange, and reciprocity. 10. Organization-public relationshi ps can be described by type (personal, professional community, symbolic, and behavioral) inde pendent of the perceptions of those relationships. 11. The proper focus of the domain of pub lic relations is relationships, not communication. 12. Communication alone cannot su stain long-term relationshi ps in the absence of supportive organizational behavior. 13. Effective management of organizationpublic relationships supports mutual understanding and benefit. 14. The relationship perspective is applicable throughout the public relations process and with regard to all public re lations techniques (p.195).

PAGE 22

22 Most of the research on organization-public relationships to date has focused on identifying dimensions that measure relationship state and the consequences of relationship state (such as attitude and behavior), creating a need for further exploration of the antecedents of relationships and the role of maintenance strategies in mana ging relationships. While various antecedent states of relationships have been proposed (Broom, Casey, & Ritchey, 2000; Grunig & Huang, 2000), only the influence of time in the relationship appears to have been empirically tested (Ledingham & Bruning, 1998; Ledingha m & Bruning, 2000b). Additionally, while relationship maintenance strategies have been suggested (Grunig & Huang, 2000; Hon & Grunig, 1999; Plowman, 1995), they have not been opera tionalized and tested. Ledingham’s (2003) list of axioms suggests that organization-public rela tionships can be driven by communication, the frequency of communication exchan ges, the nature of those exch anges, and the behaviors of parties in OPR; what is unclear is whether these factors should be classifi ed as antecedents to a relationship or maintenance strategies. What constitutes an antecedent, what constitutes a relationship state, and what constitutes an out come varies depending on which OPR model one uses. For the purposes of my study, the Gruni g and Huang (2000) model is used in which communication activities in the form of public re lations efforts and interpersonal communication are treated as maintenance strategies that preced e formation of perceptions of the organizationpublic relationship and the conse quences of those relationships. Defining and Measuring Relationships If the purpose of public relations is to manage communications effectively in order to build mutually beneficial relationships then one way of evaluating the effectiveness of organizational communication is to look at the effect it has on th e relationship itself. This raises the question of how to evaluate the quality of an organization-public relations hip and highlights the need for appropriate measures of an organization-public re lationship. Public relations researchers have

PAGE 23

23 sought to explicate the defining dimensions of re lationships within the literature in order to facilitate measurement of the relationship c onstruct. When Ferguson (1984) proposed that relationships should be the primary unit of analys is in public relations, she also suggested that there are several dichotomous dimensions that would help categorize relationships, including: dynamic/static, open/closed, and sa tisfactory/unsatisfactory. Fer guson also indicated that there are numerous other variables that would be us eful in describing relationships, including the degree to which each party feels it has control over the relationship, the amount of power possessed by each party in the relationship, percep tion of shared goals, as well as understanding, agreement and consensus. Many studies have sought to explicate th e relationship concept further through the identification of various dimensions of relations hips and testing of scales to measure these dimensions. To date, these efforts have developed in two directions. The first group of studies has moved toward th e development of relationship measures that classify the types of relationship that exists be tween an organization and a public. This stream of research begins with Ledingham, Bruning, T homlison, and Lesko (1997) that reviewed the literature from other disciplines and identif ied 17 dimensions including openness, trust, involvement, investment, and commitment. These fi ve dimensions were la ter operationalized by Ledingham and Bruning (1998) who conducted a su rvey of local telephone subscribers in territories newly opened to competition from ot her phone service providers as a testing ground for several proposed relationship dimensions They utilized Wood’s (1995) relational dimensions of trust, openness, involvement, inve stment and commitment. Wood had previously identified these dimensions as essential for su ccessful interpersonal relationships. Ledingham and Bruning concluded that an organization-pub lic relationship positively evaluated using these

PAGE 24

24 variables illustrated the value of a quality relati onship to an organization’s bottom-line and could be used to illustrate the e ffectiveness of public relations. Ledingham and Bruning (1998) then proposed that organization-public relationships could be grouped in to three categories: interpersonal relationships, community relati onships, and professional relationships. These categories were then used to develop a multi-item, multi-dimensional scale to measure organization-public relationship st ate (Bruning and Ledingham, 1999). Ledingham (2001) used the Bruning-Ledingham Relationship Scale to assess the public’s perceptions of relationship qual ity between community leaders and citizens in a suburb of a major Midwest metropolitan center. The Relations hip Scale is a multi-item, multi-dimensional scale that measures three types of relationshi ps (personal, professional and community) across eight different dimensions (trust, openness, i nvolvement, investment, commitment, reciprocity, mutual legitimacy, and mutual understanding). Ledingham provided additional support for the relational perspective and offe red several observations for managing organization-public relationships, including the need for identifying common points of interest between organizations and publics, as well as the conduc ting of longitudinal studies that examine how organizationpublic relationships change over time. Br uning and Galloway (2003) later expanded the Bruning-Ledingham Scale to include measures of two additional dimensions of personal relationships commitment (p ersonal and structural). This fits into the Broom, Casey, and R itchie (1997) model of organization-public relationships which identifies perceptions of relationships as an antecedent and posits relationship state as an interven ing variable prior to relationship outcomes. They believed that the concept of a relationship s hould be separate from the per ceptions of parties within the relationship. However, there is some disagreement as to whether a relationship can (or should) be

PAGE 25

25 understood outside the framework created by the pe rspectives of the pa rties involved in the relationship (Anderson, 1993; Duck, 1986; Laing, 1969; Thomlison, 2000). The second group of studies that focus on the development of organization-public relationship measures can be conceptualized w ithin the Grunig and Huang (2000) model as the consequences or outcomes of maintenance st rategies. Grunig, Grunig, and Ehling (1992) proposed reciprocity, trust, mutual legitimacy openness, mutual satisfaction, and mutual understanding as possible dimensions of an or ganization-public relationship. Huang (1997) reviewed the interpersonal comm unication literature and identified trust, control mutuality, commitment, and satisfaction as positive relatio nship outcomes. Hon and Grunig (1999) later developed quantitative measurement scales for these proposed dimensions of an organizationpublic relationship and included two more dime nsions: exchange relationships and communal relationships. They found the scales to be “good measures of perceptions of relationships, strong enough to be used in evaluating relationships” (Hon & Grunig, 1999, p. 5). Since then, these measures have been found to be reliable in other studies (e.g., Hon & Brunner, 2002; Huang, 2001; Jo, Hon, & Brunner, 2004; Ki & Hon, 2007; Kim, 2001b). The six Hon and Grunig (1999) relationship dimensions are described below, each accompanied by a sample Hon and Grunig measurement item: Control Mutuality Control mutuality encompasses the extent to wh ich the parties in the relationship agree as to who is authorized to exert pow er and control over one another. This construct recognizes that in many organization-public relationships, the or ganization typically has a more extensive pool of resources that will grant it a larger meas ure of control; however, a positive, healthy organization-public relationship will not be cont rolled purely by one party or the other – each party will be allowed to exercise some measure of power in the relationship. A sample item

PAGE 26

26 from the Hon and Grunig (1999) scale used to measure control mutuality reads, “This organization and people like me are atten tive to what each other say” (p. 4). Trust The trust dimension is actually made up of se veral other concepts which include integrity, dependability, and competence. Integrity is defi ned as the perception by one party that the other party in the relationship is “fair and just.” Dependability is the perception that a party will follow through on its promises and do what it claims it will do. Competence is the perception that a party in the relationship has the resour ces necessary to do what it claims it will do. A sample item from the Hon and Grunig (1999) sc ale used to measure trust reads, “This organization treats people like me fairly and justly” (p. 4). Satisfaction Parties perceive a relationship as satisfying when the expected benefits of being in the relationship exceed the costs of bein g in the relationship. Satisfac tion can also be achieved when the parties in the relationship feel that the other party is putting an adequate amount of effort into maintaining a positive relationship. A sample item from the Hon & Grunig (1999) scale used to measure satisfaction reads, “I am happy with this organization” (p. 4). Commitment The commitment dimension can be conceptua lized in two ways. Continuance commitment refers to the belief that a party feels the re lationship is worth maintaining through policy or action, while affective commitment refers more to the emotional energy that is expended in maintaining the relationship. A sample item from the Hon and Grunig (1999) scale used to measure commitment reads, “I feel that this organization is trying to maintain a long-term commitment to people like me” (p. 4).

PAGE 27

27 Exchange Relationships In exchange relationships, each party interact s with the other for the purpose of receiving something in return. Effort is expended on the relationship because benefits are either immediately realized or there is an expectation that the other party will recipr ocate at a later date. A sample item from the Hon and Grunig (1999) s cale used to measure ex change relationships reads, “Whenever this organization gives or o ffers something to people like me, it generally expects something in return” (p. 5). This type of relationship is mo re descriptive of the superficial relationships fostered by advertisi ng and marketing, not th e long-term, meaningful relationships sought through symmetr ic public relations channels. Communal Relationships Public relations efforts, specifically thos e that place a premium on developing healthy organization-public relationships, should be more concerned with fostering communal relationships. The parties in comm unal relationships provide benef its to one another just as in exchange relationships. The difference lies in the expectations that are associated with the granting of benefits. In an exchange relations hip, reciprocation is expected – not so in communal relationships. Benefits are granted becau se the parties in the relationship truly care about each other, not because they are expecting something in return. Relationships that are communal in nature should exhibit higher levels of trust, control mutability, continuance, and satisfaction than demonstrated by exchange re lationships. A sample item from the Hon and Grunig (1999) scale used to measure communal re lationships reads, “Thi s organization does not especially enjoy giving others aid” (p. 5). Recognizing a Western orientation in the sel ection of the dimensions used to define relationships, Huang (2001) developed an OPRA scale which used four of these dimensions (control mutuality, trust, satisfaction, and co mmitment) while adding a fifth dimension that

PAGE 28

28 reflected Eastern culture (f ace and favor). Favor (or renqing ) is a mode of conduct in which individuals stay in contact with influential parties, while face (or mianzi ) is more of a resource (somewhat similar to the Western concepts of ‘pride’ or ‘respect’) which can be exchanged between individuals as a means of securing favors. This resulted in a multi-item scale that could reliably be used to understand per ceptions of relationship quality. While studies have applied these relational di mensions in a variety of contexts (e.g., Cameron & McCollum, 1993; Huang, 2001; Ki & Hon, 2007) they are typically limited to measurement of the public’s pe rception of the OPR while igno ring the other party in the relationship – the organization. Despite many re searchers advocating its inclusion (Broom & Dozier, 1990; Hon & Grunig, 1999; Ledingha m, 2001, 2003; Ledingham & Bruning, 1998), the organization’s perceptions remain strangely ab sent from OPR research. One strategy for integrating both the public’s and the organizatio n’s perceptions of the relationship has been suggested numerous times in the literature, but seldom applied in measuring organization-public relationships. This method is the coorientational approach. Coorientation Again, Ferguson (1984) proved to be forw ard thinking in this regard. Among the dimensions that she suggested could be helpful in quantifying the nature of organization-public relationships, she listed mutuality of understanding, agreement, and consensus, and noted that the “coorientational measurement model should prove quite useful in conceptualizing relationship variables for this type of paradigm focus” (p. 17). The use of a coorientational approach has been proposed repeatedly in the public relations literature, however it wa s originally conceived of as being used to assess how two groups, fo r example, an organization and a stakeholder public, perceive an issue of mutual concern and not the relationship itself (Broom, 1977; Broom & Dozier, 1990; Grunig & Stamm, 1973; McLeod & Chaffee, 1973).

PAGE 29

29 Measuring relationships between two or more parties has been attempted previously in the interpersonal and mass communications fields. La ing, Phillipson, and Lee (1966) developed an Interpersonal Perception Method for assessing the perceptions of husband-wife dyads. Their measurement items focused on the relationship be tween the couple, not on perceptions of the spouses themselves. Their model “operates under the assumption that pair ed individuals sharing experiences over time come to develop ‘reciproc al perspectives’” (O’K eefe, 1973). This seems to indicate that one way of defining a relati onship is through assessment of how the parties involved in the relationship percei ve the relationsh ip itself. Laing’s (1969) Relational Perception Theo ry proposes that individuals within a relationship continuously influence one anothe r through their interac tions and that those interactions draw on three different perspectives. The first is the dire ct perspective that is what an individual in the relationship (Person A) thinks. The second pers pective is the meta-perspective and is what the individual (Person A) thinks the other individual in the relationship (Person B) thinks. The third perspective is the meta-meta-pe rspective and is what an individual (Person A) thinks the other individual (Person B) thinks that Person A thinks; that is, the meta-metaperspective is how a person thi nks their direct perspective is perceived by another. Laing proposed that the greater the degree of matching (i.e., accu racy) between these various perspectives of the parties in the relationship, the better those individuals would understand each other and feel that they were be ing understood. This in turn would lead to more positive, healthy relationships. With theoretical roots in Newcomb’s (1953) Symmetry Theory, coorientation proposes that the attitudes of two parties (A & B) toward an object (X) are influenced in large part by how they perceive each other’s attitudes toward the object. Symmetry theory posits that individuals

PAGE 30

30 will seek balance in their interactions, therefore “enabling two or more individuals to maintain simultaneous orientation toward one another and toward an object of communication” (Newcomb, 1953, p. 394). This is base d on the idea that individuals seek consistency or a tendency towards symmetry, in order to maintain an internal equilibrium and achieve balance. In order to achieve this balance, an individual may use communicative acts to change another individual’s attitudes or behavior s, or failing that, the individual may adjust their own attitudes and behaviors. For example, consider two individuals in an interpersonal relationship (A & B) and their attitudes towards the President (X). Albert (A) likes Betty (B), but dislikes the President (X). Betty likes Albert, but does not agree with him rega rding the President. Albert is then faced with an internal inconsistency: how can he like someone who disagrees with him regarding the President? Albert has several options. He can tr y to sway Betty to change her mind. This will result in a balanced equation where Albert and Betty like each other and they both dislike the President. However, suppose Betty refuses to ch ange her mind. In order to obtain symmetry, Albert must either change his attitude toward s Betty (disliking her because she doesn’t agree with him) or change his attitude towards the Pres ident so that it is now acceptable to like Betty because they agree on their att itudes towards the President. Previous work attempted to extend this interpersonal approach to communication between collectives (McLeod & Chaffee, 1973; Grunig & Stamm, 1973). McLeod & Chaffee reviewed interpersonal approaches to communication research including the coorientational model. Grunig & Stamm (1973) advocated a coorientat ion paradigm for communication between an organization and another social co llective (i.e., public). Part icularly striking is Grunig and Stamm’s observation that “if a researcher us es a coorientation paradigm…he can focus his

PAGE 31

31 attention on the relationship of sender to receiver in the communication system [emphasis added]” (p. 567). Again, these studies suggest important conc lusions about the rela tionship between two entities; most notably that the relationship is a construct separate from measures of attitudes towards issues, people, and other objects external to the relati onship. For instance, in our previous example of Albert and Betty, the state of their relati onship, as defined by levels of accuracy, agreement, and congruency between vi ewpoints, can be viewed as a construct independent of their assessments of the President. This is an important distinction for public relations researchers and practitione rs to keep in mind because it s uggests that two entities (e.g., an organization and stakeholder public) can hol d similar attitudes about an issue of common concern, and yet the relationship between them ca n still be considered a “poor” relationship (and vice versa). Also, the model again stresses the po int that it’s not enough to consider what one party in the relationship thinks, bu t that both parties have a say in defining the true nature of the relationship; the nature of the relationship rests on shared meaning. Using our simple interpersonal example above, Albert may like Betty, he may think that she likes him, and he may think that they have a healthy re lationship, but if Betty doesn’t lik e Albert, then their relationship is not healthy, regardle ss of Albert’s perceptions. Simply put, the saying that it takes two to tango is directly applicable to organizatio n-public relationship measurement. Using a coorientational approach to measure organizatio n-public relationships di rectly addresses this problem of shared meaning and perception. Coorientational Approach A coorientational approach to measuring the OPR includes four points of analysis: (1) the organization’s view of the relationship (the orga nization’s perspective), (2 ) the public’s view of the relationship (the public’s pers pective), (3) the organization’s estimate of the public’s view of

PAGE 32

32 the relationship (the organizati on’s meta-perspective), and (4 ) the public’s estimate of the organization’s view of the relationship (the public’s meta-perspective). The interaction between these perspectives and meta-perspectives creates three measures of coorientation. “Agreement” indicates the degree to which th e organization’s view matches th e public’s view of the OPR. “Accuracy” indicates the degree to which the organization correctly estimates the public’s viewpoint, and vice versa. “Per ceived agreement” (or “congruenc y”) is the degree to which the organization’s view matches its perception of the public’s viewpoint, and vice versa. The coorientational model presented in Figure 2-1 is adapted from Broom (1977) and Broom and Dozier (1990) and helps to clarify the linkages among these concepts. The coorientational model has been used in other fields such as interpersonal communication (e.g., O’Keefe, 1973; Purnine & Carey, 1999), employee communications (e.g. Jo & Shim, 2005), political science (e.g., Hesse, 1976), journalism (e.g., Jones, 1993) and environmental policy (e.g., Connelly & Knuth, 2002) but even in these instances it is usually limited to assessments of one side’s perceptions of the relationship or evaluations of an issue common to both parties. The model has also been applied to some degr ee within a public relations context, but its application has been either piecemeal (by ignori ng the organizational perspective) or focused on issue or attitude assessment rath er than relationships (e.g., Bowes & Stamm, 1975; Broom, 1977; Broom & Dozier, 1990; Grunig, 1972; Stegall & Sa nders, 1986, Stamm & Bowes, 1972). In their content analysis of published research on orga nization-public relations hips from 1985-2004, Ki and Shin (2006) found that only one of the 38 studies included in their analysis used coorientation as a theoretical framework.

PAGE 33

33 Cameron and McCollum (1993) used part of th e coorientational appr oach to investigate internal corporate cultures and perceptions of the relationship between employees and their company. However, Cameron and McCollum only used employee perceptions of the relationship. They did not look at upper management (i.e., the organization’s dominant coalition) perceptions of the relationship betw een themselves and their employees (i.e., an internal public). This is typical of the current state of incorporating the coorientational model into public relationship research. Study after st udy tiptoes around the c oorientational approach without utilizing the direct pers pectives of both the or ganization and its publics in measuring the relationship between them, much less assessi ng the meta-perspectives of each party. Hon and Brunner (2002) used the Hon and Gruni g relationship measures in a survey to measure the perceptions of college students toward their university. Following a partial coorientational approach, Hon and Brunner con ducted qualitative interviews with college administrators to obtain their direct perspec tives on the university-student relationship. The researchers found that satisfaction was the best indicator of relationshi p quality while control mutuality was the weakest. Notably, the meta-p erspectives of the respective parties in the relationships were not measured; therefore, m easures of accuracy, agreement, and congruency could not be assessed. Perhaps the closest that recent research has come to using relationship measures in a full coorientational framework are studies by Shin & Cameron (2005) and Christen (2005). In Shin & Cameron (2005) the source-reporter relationship between public relatio ns practitioners and journalists was analyzed using some of the H on-Grunig (1999) measures in addition to other measures of conflict. However, my study looks at two distinct publics, the members of which are not unified in any sense beyond their choi ce of profession (e.g., the public relations

PAGE 34

34 practitioners didn’t all work for the same firm ). Evaluations were based on one profession’s assessment of the other profession. The coorient ational approach of relationship measurement still has yet to be fully utilized within the cont ext of an organization-public relationship in which the public in question is a stak eholder in the organization and where the organization actively uses relationship maintenance strategies to mana ge the relationship. Christen (2005) considered the use of measures of power and trustworthiness in moderating willingness to negotiate between groups; however, her study did not use the Hon-Gr unig (1999) scales for measuring relationship quality. Additionally, the study used students in an experimental design in which they were to assume the identities of various groups and did not assess the perceptions of the organizations and their stakeholders directly. The Value of the Coorientational Approach Using a coorientational relationship measuremen t approach could have several benefits for the public relations profession. Not only can it be used as a measur e of public relations effectiveness, but it can also help support academics and practiti oners in their effort to prove the value of public relations to thei r organization’s management or to their clients. Heath (2001) notes that there are “cost-reduction paradigms” and “revenue-generation paradigms” that are used to justify the value of public relations. That is to say, one way of looking at the value of public relations is that it is not necessarily tied to the bottom-line, that there is not always an immediate financial benefit. Gr unig et al. (1992) state that eff ective public relations helps an organization by reducing the costs, stress, and strains experienced by the organization due to conflict, legislation, and litig ation. For instance, good crisis management programs may not generate revenue for an organization, but they can certainly help the organization prevent the loss of revenue by avoiding litigation. Public re lations programs can also help an organization weather tough times by building stro ng relationships with publics prior to a crisis occurring. In

PAGE 35

35 the relationship perspective, the bottom-line isn’t the end all be all indicator of the value of public relations – other units of analysis, such as outputs and outcomes are more often relied on, but as was noted earlier, these measures may not be the best barometers of the long-term relationship building goals of public relations. However, despite ongoing calls for better measures of public relations effectiveness, th e predominant means of assessing public relations impact continue to center on outcomes and outputs (see Xavier, Johnston, Patel, Watson, & Simmons, 2005 for a concise review of the literature on program evaluation). In this regard, the coorientat ional method may provide a bette r measure of effectiveness by going beyond simple outputs and outcomes and yi elding information about the relationship between the organization and the public itself, not only by evaluating the parties’ direct perceptions of the relatio nship, but by utilizing the meta-persp ectives of both parties to determine degrees of accuracy, agreement, and congruency. As has been noted in previous literature, outputs, although popular measures among practitioners, ca n be weak indicators of effectiveness. Just because a program obtained a lot of media placements doesn’t mean the placements had the desired effect. The coorientational relations hip measures also go beyond simple outcomes. Commonly utilized outcomes (such as awareness of an organizational offering) focus more on the results of a particular public relations pr ogram, while the coorientational relationship measures are a gauge of more long-term and stable results. Put another wa y, consistently positive program outcomes should eventually lead to positive relationships. But even this fails to capture the dual-sided nature of the relationship by fo cusing on the effects of program outcomes on the target audience’s perception of the relationship. The coorientational relationship measures provide a snapshot of the whole organizationpublic relationship in the form of shared perceptions and gaps in perception between th e organization’s and public’s perspectives and

PAGE 36

36 meta-perspectives, thereby facilitating a mo re thorough diagnosis of the relationship and suggesting possible prescriptive actions to improve the relationship if need be. By making comparisons between the four coorie ntational elements, it will be possible to gauge the state of the three c oorientational variables: accura cy, agreement, and perceived agreement (congruency) (Broom & Dozier, 1990) between the organizat ion’s and the public’s views of the relationship. Based on these comparisons, one can determine the type of coorientational relationship st ate that exists between the organization and the public. The possible coorientational states are true cons ensus (accurate perception of strong agreement between views), dissensus (accurate perception of strong disagreement between views), false consensus (inaccurate perception of agreement), or false conflict (inaccurate perception of disagreement) (Broom, 1977; Broom & Dozier, 1990). Being able to assess the coorientational vari ables of accuracy, agreement, and congruency, as well as the resulting coorientational relationship states of dissensus, co nsensus, false conflict, and false consensus, could be especially valuable to public relations prac titioners; possibly even more valuable than outcome measures of attitude or relational perception. This is particularly true for relationships that are found to be in a st ate of false consensus or false conflict in that it may be easier for public relations practitioners to increase perceptual accuracy between the parties in the relationship by correct ing these misperceptions than it is to achieve more traditional public relations objectives such as attitude change (Broom, 1977). For example, use of the coorientational met hod of relationship measurement might reveal that there is strong agreement between the true viewpoints of the organi zation and the public, but due to poor articulation of the or ganization’s viewpoint, the public inaccurately perceives that the organization’s views are incongruent with its own. In that case, a state of false conflict exists

PAGE 37

37 between the organization and its public. Armed w ith this information, public relations programs targeted at this public might highlight the areas of true agreement between the organization and the public in an effort to strengthen the accuracy and congruency dimensions of the relationship. Looking at the relationship between the OPR me asures and public re lations activities will indicate the effectiveness of these activities in actually changing the relationship that exists between the organization and the public. By look ing at individual dimensions, it will be possible to identify exactly which aspects of the relationshi p were affected by public relations efforts. For example, the public relations campaign may have increased trust between the organization and the public but failed to affect commitment. Again, this information can be used to ev aluate and refine ongoing and future public relations efforts. It would also be a means of demonstrating the return on investment in public relations to a client by illustrati ng that public relations activities influence not only the amount of earned media, etc., but more importantly, the im pact that public relations has on the actual relationship between an orga nization and its publics. Of course, this may not satisfy clients and managers who adopt the “revenue generating paradigm” and continue to define public relations effectiveness in terms of dollars. In this perspective, public relations s hould have an impact on the bottom-line business results of an organization, such as increasing sales. Even if the client or organizati onal management insist on adopting this viewpoint, then the coorientational approach to m easuring relationships still has value, especially if it is integrated into other methods of determining return on investment. For example, Kim (2001a) developed a two-stage model th at illustrated the imp act of public relations expenditures on corporate reputation, which in turn was correlated with market share. Coorientational relationship measures could be utilized in a similar fashion by determining the

PAGE 38

38 amount of money budgeted for pub lic relations program at T1, T2, T3, etc., linking expenditures to the coorientational relationship measures over time, and then in turn connecting this to market share at T1, T2, T3, etc. The relationship measures could also be utilized in market mix models. For example, Proctor & Gamble is using marketing mix tools to illustrate the value of public relations versus advertising (Neff, 2005). The c oorientational relationship measur es between an organization and publics segmented by geographic region and time c ould be incorporated into these models along with expenditures on public relations, advertis ing, direct marketing, and other marketing communication expenditures as well as sales data. Along similar lin es, overall metrics representing the strength of the relationship or mean score differe nces of the individual measures could be incorporated into an or ganization or firm’s existing return on investment tools, such as Ketchum’s ROI Lab (Rockland, 2005), in order to clarify the link between public relations expenditure, the effectiveness of public relations, and bottom-line business results. Research Questions The preceding discussion highlighted the valu e of integrating esta blished relationship measures within a coorientational framework. Measuring organization-pub lic relationships in this manner would move beyond commonly us ed measures of outputs and outcomes. Additionally, it would move beyond assessments of a public’s direct perceptions of the relationship by including the dir ect perspectives and meta-perspectives of both parties involved in the relationship. This w ould allow practitioners and researchers to measure public relationships in terms of the coorientational variables of accu racy, agreement, and perceived agreement (i.e., congruency). This in turn would allow the organization-pub lic relationship to be evaluated in terms of its coorientational relationship state and determine whether the relationship is in a state of false consensus, false conflict, consensus, or dissensus.

PAGE 39

39 Therefore, the first objectiv e of my study is to test a methodology for measuring an organization-public relationship in such a manne r. As noted in the literature review, a methodology for qualitative measurement of a real-w orld organization-public relationship within a fully coorientational framework has not b een implemented to date. Thus, the following research questions drive stage one of the study: RQ1: Can the coorientational framework be used in conjunction with existing organization-public relationship measures to measure an organizati on-public relationship? RQ2: Can the resulting coorientational meas ures of accuracy, agreement, and congruency be used to evaluate the state of the organization-pub lic relationship? Hypotheses Organization-public relationships do not form in a vacuum; as previous models have noted, antecedents (e.g., need, motivation, etc.) and relationship maintenance efforts can act to influence relationship states and outcomes (Br oom, Casey, & Ritchey, 2000; Grunig & Huang, 2000; Hon & Grunig, 1999). Time in the relationship has also been identified as a factor in perceptions of the re lationship (Ledingham & Bruning, 2000b; Ledingham, Bruning, & Wilson, 1999). Therefore, measurement and evaluation of th e direct and meta-persp ectives of the parties within the organization-p ublic relationship should be considered in light of the effect of these factors. Hon and Grunig (1999) acknowledged that on e of the benefits of using relationship measurement to evaluate public relations effec tiveness is that these measures provide an indication of public relations programming eff ectiveness over the long-term and are a better indicator than short-term outputs and outcomes that measure the impact of a specific, short-term program. Ledingham, Bruning, and Wilson (1999) f ound that time in a relationship influenced perceptions of the dimensions of an organiza tion-public relationship, concluding that building

PAGE 40

40 the long-term relationships that are desired by public relations practitioners demands a long-term commitment. Therefore: H1a: Longer time in the organization-public relationship will lead to more favorable ratings of the organizationpublic relation dimensions. While Ledingham, Bruning, & Wilson (1999) look ed at the effects of time on the direct perspectives of a relationship, my study will go a step further and look at the impact of time on the coorientational variables of agreement, accu racy, and congruency. Since longer time in the relationship provides more opportuni ties to interact and communicate with the other party in the relationship, individuals who have spent more tim e in the relationship have more experience with one another and should have a more accurate pe rception of the other party’s perspective. Therefore: H1b: Longer time in the organization-public re lationship will lead to a greater degree of accuracy between the meta-perspectives of one party in the organization-public relationship and the direct pers pective of the other party. Ledingham and Bruning (2000b) acknowledged th at “research supports the notion that management communication programs can influe nce perceptions of the organization-public relationship” (p.65). The relationship management perspective repeatedly focuses on the use of organizational communication to establish, main tain, and nurture relationships between an organization and its strategic publics (Grunig & Huang, 2000; Hon & Grunig, 1999; Ledingham, 2003; Ledingham & Bruning, 2000b). The effective use of relationship maintenance strategies should improve perceptions of the organization-p ublic relationship (Broom Casey, & Ritchey, 2000; Grunig & Huang, 2000; Hon & Grunig, 1999). Therefore: H2a: Exposure to relationship maintenance strate gies will lead to more favorable ratings of the organization-public re lationship dimensions. Not only should the effective use of manage d communication improve direct perceptions of the relationship, but it should also improve the degree of cooperation, reduce conflict, and

PAGE 41

41 help organizations and their publics understand each other better (Broom, 1977; Ehling, 1992; Hon & Grunig, 1999; Ledingham, 2003). Therefore: H2b: Exposure to relationship maintenance st rategies will lead to a greater degree of agreement, accuracy, and congruency between the perspectives of the parties in the organization-public relationship. Furthermore, excellence theory and relati onship management theory both propose that two-way symmetrical communication is the best way for public relations practitioners to improve organizational effectiven ess and build mutually benefici al relationships (Dozier, 1995; Dozier, Grunig, & Grunig, 1995; Grunig & Hua ng, 2000; Grunig, Grunig, & Dozier, 2002). Therefore: H2c: Exposure to relationship maintenance st rategies that utili ze two-way symmetrical communication will lead to more favorabl e evaluations of the organization-public relationship dimensions than exposure to rela tionship maintenance strategies that utilize one-way channels of communication. H2d: Exposure to relationship maintenance st rategies that utili ze two-way symmetrical communication will lead to a greater degree of agreement, accuracy, and congruency than exposure to relationship maintenance strate gies that utilize on e-way channels of communication. Maintenance strategies stemming from manage ment communication programs are not the only interaction that parties in the organizationpublic relationship will have with each other. Members of each party in the relationship could have direct experience with each other in the form of interpersonal communica tion as part of their behavi or, actions, and interactions. Relationships are based in part on expectations for behavior and whether the parties in the relationship meet those expectations or not (Littlejohn, 1995; Thomlison, 2000). The relationship management perspective also acknowledges that managed communication alone cannot maintain a relationship between the organiza tion and its publics; those messa ges have to be reinforced by behavior (Ledingham, 2003).

PAGE 42

42 H3a: Positive interpersonal interaction with the other party in the organization-public relationship will lead to favorable ratings of the organization-public relationship dimensions while negative interpersonal interaction will lead to unfavorable ratings of the organization-public relationship dimensions. H3b: Positive interpersonal interaction w ill lead to a greater degree of accuracy, agreement, and congruency between the perspe ctives of the parties in the organizationpublic relationship while negative interpersonal interaction will lead to a lower degree of agreement, accuracy, and congruency between the perspectives of the parties in the organization-public relationship. The next chapter discusses how a coorientat ional methodology for m easuring organizationpublic relationships (including both directa nd meta-perspectives) would be designed. The method and results for a pilot study are discusse d, followed by a review of the final methodology utilized in the main study.

PAGE 43

43 Figure 2-1. The coorientational model of organi zation-public relationships (adapted from Broom (1977) and Broom & Dozier (1990)).

PAGE 44

44 CHAPTER 3 METHODOLOGY This chapter outlines a methodology for includi ng both parties (the organization and the public) in an evaluation of the organization-public relationship by combining the coorientational methodology advocated by Broom (1977), Broom a nd Dozier (1990), and others (e.g., Hon & Grunig, 1999) with the relational dimension meas ures established by Hon and Grunig (1999). Applying these measures within the coorientatio nal model will yield assessments of accuracy, agreement, and congruency that can be used to a ssess the state of the rela tionship. The results of a pilot study investigat ing the use of a proposed coorientational methodology and measurement instrument are also discussed prior to a desc ription of the final methodology employed in the main study. Integrating the Coorientational Approach with Relationship Measurement The following section presents an overview of how the measurement of relationship dimensions can be integrated methodologically with the coorientational fr amework. It includes a discussion of modifications to existing relati onship measurement items so that they are appropriate for application within a coorientational approach. This is followed by a discussion of considerations that will need to be made regarding sample selec tion, survey administration, and initial data analysis. This broader discussion ad dresses how the coorientat ional procedure differs methodologically from current approaches for re lationship measurement and provides the basis for a more detailed description of the pilot st udy and main study methodologi es presented later in the chapter. Adapting the Relational Dimensions for the Coorientational Approach The first step in integrating the relationship measures into the coorientational framework is adapting Hon and Grunig’s (1999) relationship dimensions measurement scale so that the

PAGE 45

45 same items can be used when surveying both the public and the organization. This is achieved by restructuring the questions so that they are mo re objectively worded. For instance, the first “trust” measure reads, “This organi zation treats people like me fairly and justly.” While this is appropriate wording for a respondent from a public sample (the sample frequently used in current research), it would be inappropriate when presented to a member of an organization. For example, if one was interested in the relations hip between Organization A and Public B, the new objectively reworded trust measure should read, “Organization A treats Public B fairly and justly.” This statement would be appropr iate for use with respondents from both the organization sample and the public sample. A list of the modified scale items representing the four relationship dimensions is presented in Appendix A. Such objective wording would not only be consistent with the theory behind the coorientational approach, but also with the goal of explicating the relationship as a construct independent of the parties in the relationship. First, both symmetry theory and coorientation conceive of the interested part ies (A and B) evaluating another object (X), not necessarily each other. By rewording the statements, we make them objective statements about the relationship between A and B, not about A and B themselves – the “relationship” essentially becomes the “X” within the A-B-X framework. Secondly, by having the organization and the public evaluate objectively worded statements about the nature of the relationship, perceptions about the individual parties are explicated somewhat from the perceptions about the relationship between the parties. The reworded statements don’t ask a respondent how the organization or public treats the respondent per se, but now focuses on the dynamic between the organization and the public.

PAGE 46

46 The final survey will utilize the Hon and Grunig (1999) short scale of relationship measures for trust, satisfaction, control mutuality, and commitment. In the interest of parsimony, the final two dimensions, exchange relationships and communal relationsh ips, will be dropped. These two constructs are really types of relationships and are th erefore reflective of the other dimensions rather dimensions in and of themse lves. Also, given that each respondent will rate every statement twice (once to report their dir ect perspective and once to report their metaperspective), dropping the nine items making up these two dimensions actually shortens the survey by 18 questions. Their removal shoul d reduce the respondents’ perceived burden for completing the survey (Groves, Fowler, C ouper, Lepkowski, Singer, & Tourangeau, 2004). Sample Defining, identifying, and recr uiting representative samples for this procedure depends on the nature of the organization and the public(s) to be studied. In every application of the method, two sets of perspectives and meta-perspectives n eed to be represented: the organization and the public of interest. Depending on the size and structure of the organization, either a representative sample of the orga nization should be drawn or a full census. In most situations, a random sample of the public of interest would be drawn. To illustrate how the sampling process would be governed by the situation, two examples are provided. In the first example, suppose an organization was interested in the relationship between management and an internal public such as its employees. The organizational sample could be composed of upper-level managers wh ile the public sample could be composed of either all employees (in the case of a small comp any) or a random sample of employees (in the case of a large company). The sample could also be stratified to focus on different types of employees such as full-time, part-time, new hi res, etc. In the second example, suppose an organization was interested in the relationshi p between the organization as a whole and an

PAGE 47

47 external public, such as constituents. The organizational sample could be constructed to represent both management and lower-level em ployees (or focus solely on the dominant, decision-making coalition) while the public sa mple could be a constructed from a random sample of constituents (again, with the option of stratifying the sample to isolate specific subgroups among the stakeholder public). Survey Administration Alternately worded versions of the questionn aire should be administered to respondents from the public and the organizational sample. While the same objectively worded relationship items would be presented in both the public and the organization versions of the survey, each survey should include directions that are wo rded with the appropriate respondent type (organizational or public) in mind. Also, ther e would be some variation in the type of demographic questions used. For example, as king members of the organization how long they had been with the organization while asking me mbers of the public how long they had been customers, advocates, activists, etc. In both the organizational and public version of the surveys, the 21 items representing the four relationship dimensions are presented twice. In the first presentation, each respondent should indicate their personal response to each item (i.e., their direct perspective) In the second presentation, the respondent would be instructed to estimate how a member of the other party (either organization or public) would respond to the same item (i.e., their meta-perspective). Following Hon and Grunig (1999), the questionnaire should ask respondents to use a seven-point Likert scale to indicate the degree to which they agree or disagree with statements that measure the four dimensions of control mutuality, trus t, commitment, and satisfaction. While Hon and Grunig (1999) used a nine-point scale, studies of survey research response options have indicated that respondents find fiveor seven-point sc ales easier to utilize (Groves et al., 2004).

PAGE 48

48 This process would yield data for each of the four coorientational elements: (1) the public’s view of the relationship (i.e., th e public’s direct perspective), (2 ) the organization’s view of the relationship (i.e., the organizati on’s direct perspectiv e), (3) the public’s estimate of how the organization views the relationship (i.e., th e public’s meta-perspective), and (4) the organization’s estimate of how the public views the relationship (i.e., th e organization’s metaperspective). The surveys should include additional demogra phic measures and other items that gauge exposure to relationship maintenance strategies in the form of inte rnal or external organizational communication and public relations activities. Agai n, the exact nature of these measures would depend on the specific relationship under study. For example, if the relati onship being measured is between the organization’s management and its employees, then employees should be asked questions such as “how long have you worked for this organization?” and “how often do you read the company newsletter?” Researchers utilizing this met hod should work closely with the individual organization to craft a ppropriate items that answer the que stions that are of interest to the organization and capture the nature of th e specific relationship being evaluated. Data Analysis Data analysis begins by calculating mean scores of the four relational dimensions for each coorientational perspective (i.e., what are the mean relationship scores for the two direct perspectives and the two me ta-perspectives). Broom and Dozier (1990) recommended two methods for making comparisons between the coorient ational perspectives. The first consists of calculating correlation coefficients between mean sc ores for the public and the organization. The other method requires the calculation of differen ce scores (“d-scores”) between means for the public and the organization. A large d-score i ndicates little agreement and low accuracy; a small d-score indicates greater agreement and higher accuracy. While a combined approach can be

PAGE 49

49 utilized, Purnine and Carey (1999) felt that the correlational method produced superior results. However, Purnine and Carey’s study focuses on making comparisons between paired individuals who were married. Such an a pproach wouldn’t make sense with in the context of a single organization-public relationship. Since most prac tical applications of this method will only be focusing on one organization and one public of intere st (either internal or external), the best strategy would be to calculate mean scores for each relational dimension for both the organization and the public of interest and then compare the means for significant differences between the organization and the public. Aggreg ate level scores on the various relationship items could be used to determine trends in the measures over time. By making these comparisons between the f our coorientational elements, it will be possible to determine the degree of accuracy, ag reement, and perceived agreement (congruency) between the organization’s and the public’s views of the relationship. This interaction of the perceptions of the relationship between the partie s will yield a clearer picture of the relationship by indicating states of consensus, dissensus, false conflict, and false consensus that exist between an organization and the specific public in question. Pilot Study To begin an investigation into the use of the coorientational a pproach for measuring organization-public relationships, a pilot study was conducted in the spring of 2006. The study was conducted at a local health and fitness comp any that has 300 employees at two locations. The pilot study focused on measuring the relati onship between the organization’s management (representing the organizational directand meta-perspective s) and organization’s staff (representing an internal public’s di rectand meta-perspectives).

PAGE 50

50 Method Two samples were utilized for this pilot study: (a) a sample of 17 members of the organization’s management, and (b) and sample of 22 staff members to represent the view of an internal public. Each group was administered one of two versions of a short survey that asked questions regarding their percep tion of the relationship between management (the organization) and staff (an internal public). Managers recei ved the management version of the survey which asked for management’s perception (i.e., organizatio n’s direct perspective) of the relationship. The survey then asked the managers to estimate how they thought staff members perceived the relationship (i.e., organization’s me ta-perspective). Staff members received the staff version of the survey which asked for the staff members’ pe rception of the relationshi p (i.e., public’s direct perspective). The survey then asked the staff members to estimate how they thought management perceived the relationship (i.e., public’s meta-perspective). Each survey included 21 attitudinal statemen ts about the relationship between management and staff. These statements represented measures of trust, satisfaction, commitment, and influence and were drawn from previous studi es on organization-public relationships (Hon & Grunig, 1999). Respondents were asked to (a) indicate how st rongly they agree or disagree with each statement to capture their direct perspective, and (b) indicate how they thought the other party would respond to the same statements to captur e their meta-perspective. Responses were made on a seven-point Likert scale anchored by 1 (“st rongly disagree”) and 7 (“strongly agree”), with higher scores indicating stronger agreement. One question (the third control mutuality question) was reversed for scoring as in Hon and Grunig (1999). The survey concluded by asking demographic ques tions (gender, age, position, years at the organization, etc.). It also asked respondent s to estimate how many hours a week they spent

PAGE 51

51 communicating with either staff members (on the management survey) or managers (on the staff survey) face-to-face, online, and by phone. Copies of the survey instruments are presented in Appendix B. Pilot Study Results After collecting the surveys, it was necessary to remove one manager and one staff survey because these two respondents had received the wrong version of the survey (i.e., the manager received a staff survey and vice versa). This resulted in the respondents wondering why they were being asked to respond to the same ques tions twice due to wording of the survey instructions (e.g., the manager wa s asked for their perception and then instructed to estimate how a manager would respond). Therefore, these two cases were removed from the analysis. Four cases had missing data. Data was imputed using the mode; while one would normally use casewise deletion when using a larger sample, this particular sample was too small – dropping even four of the cases would re sult in a significant loss of data. Summing of scales. Responses to questions comprising each of the four relationship dimensions were averaged to gene rate the four relationship measures This produced four sets of data (one set for each perspective), with each set providing scores for each of the four relationship measures: 1. Staff’s perception of the relationship in term s of control mutuality, trust, commitment, and satisfaction (i.e., public ’s direct perspective). 2. Staff’s estimate of management’s perception of the relationship in terms of control mutuality, trust, commitment, and satisfaction (i.e., public’s meta-perspective). 3. Management’s perception of the relationship in terms of control mutuality, trust, commitment, and satisfaction (i.e., or ganization’s direct perspective). 4. Management’s estimate of staff’s perception of the relationship in terms of control mutuality, trust, commitment, and satisfact ion (i.e., organization’s meta-perspective).

PAGE 52

52 Figure 3-1 clarifies the relationship between these four sets of data an d presents a model of the management-staff relationship (i.e., th e organization-public relationship). Scale Reliability. Since the items composing the scales were reworded to make them more objective (and therefore, more appropriate for the coorientational approach) and because they are being applied outside of their usual cont ext in which only the pub lic’s direct perspective is measured, one of the purposes of the pilot stud y was to assess scale reliability. All scales had a Cronbach’s Alpha of .81 or better. Control mutu ality was .81, the trust scale was .94, the commitment scale was .94, and the satisfaction scale was .96. This is similar to the scale reliability reported for the original Hon and Gr unig (1999) study in which the trust scale had a Cronbach’s Alpha of .86, control mutuality was 87, commitment was .85, and satisfaction was. 89. Demographics. The average age for managers was 34.9 years ( SD =12.16), while the average age for staff members was 27.2 years ( SD =9.81). The overall average age for respondents was 30.5 years ( SD =11.41). In terms of gender, 51.4% were male while 48.6% were female. Looking at managers only, 56.3% were male and 43.8% were female. 47.6% of the staff members were male and 52.4% were female In terms of experience at the organization, 27% had spent less than one year at the organizat ion, 39% had been there for 1-5 years, 19% had been there for 6-10 years, and 16% had been ther e for more than 10 years. 37.5% of managers had been at the organization for 1-5 years, 31.3% had been there for 6-10 years, and 31.3% had been there for more than 10 years. 47.6% of the staff members had been there for less than one year, 38.1% had been there for 1-5 years, and 14.3% had been at the organization for six years or more.

PAGE 53

53 Communication. Overall, respondents spent 14.01 hours per week ( SD =13.16) engaged in face-to-face communication with other empl oyees. Respondents spent 2.28 hours a week ( SD =7.19) communicating online and 1.57 hours a week ( SD =2.90) communicating by phone. In total, employees spent 17.86 hours a week ( SD =20.12) communicating w ith other employees. Looking at managers only, th ey spent 25.3 hours a week ( SD =12.2) talking to staff face-to-face, 4.75 hours a week ( SD =10.58) communicating online, and 2.63 hours a week ( SD =4.05) communicating by phone. Looking at staff only, they spent 5.39 hours a week ( SD =4.5) communicating with managers, 0.39 hours a week ( SD =.58) communicating online, and 0.77 hours a week ( SD =1.13) communicating by phone. Agreement. A series of independent t -tests were conducted to compare management’s view of the OPR to staff’s view of the OPR (s ee Table 3-1). While management’s rating of each of the four relationship dimensions is always hi gher than the staff’s ratings, only management’s ratings of control mutuality and trust were significantly different. Management’s mean rating of control mutuality was significantly higher than staff’s mean rating of control mutuality. Management’s mean rating of trust was significantly higher than staff’s mean rating of trust. Perceived Agreement. Paired samples t -tests were conducted to compare management’s view of the OPR to how management estimated the staff’s view of the OPR (see Table 3-2). While management’s rating of each relationship dimension is alwa ys higher than the perceived staff view, only control mutuality, trust, a nd commitment were significantly different. Management’s mean rating of control mutuality was significantly higher than management’s estimate of the staff’s rating of control mutuality. Management’s mean rating of trust was significantly higher than their estimate of the staff’s rating. Management’s mean rating of commitment was significantly higher than their estimate of the staff’s rating.

PAGE 54

54 Next, paired samples t -tests were conducted to compare the staff’s view of the OPR with how staff estimated management’s view of the OPR (see Table 3-3). While staff consistently estimated that management would rate the OPR mo re favorably than staff would, only trust and commitment are significantly different. Staff’s mean rating of trust was significantly lower than their estimate of management’s rating. Sta ff’s mean rating of commitment was significantly lower than their estimate of management’s rating. Accuracy. A series of independent t -tests were conducted to compare the staff’s estimate of management view to management’s actual view of the OPR; this indicated how accurate the staff’s prediction of the management view wa s (see Table 3-4). There were no significant differences between the staff’s estimate of th e management view and the actual management view for three of the relationship dimensions; staff accurately predicted how management would rate the relationship in terms of trust, commitment and satisfaction. However, staff’s estimate of control mutuality was significantly lower than management’s actual rating of control mutuality. Next, a series of independent t -tests were conducted to compare management’s estimate of staff’s view of the OPR to staff’s actual view of the OPR; this indicated how accurate management’s prediction of the staff view wa s (see Table 3-5). There were no significant differences; thus management accurately predicte d how staff would rate the relationship on the four relationship dimensions. Conclusions from Pilot Study To clarify the results of the pilot study, it he lps to address the findings as they relate to each of the four relationship dimensions in turn. In regards to control mutuality, there was a lack of actual agreement between management and staff, with management scoring significantly higher on the control mutuality sc ale than staff. Management seemed to be aware of this; management estimated that staff would score lo wer. However, staff failed to recognize this

PAGE 55

55 disagreement. This difference in percepti on is reinforced by the accuracy with which management predicted staff’s actual control mutu ality score and staff’s inaccurate estimate of management’s score (staff’s estim ate of management’s control mutuality score was lower than management’s actual score). In essence, manage ment feels that the relationship has a greater degree of control mutuality than staff does, but on ly management seems to be aware of this; staff incorrectly perceived that the two parties agr ee that there is less control mutuality in the relationship (i.e., false consensus). Manageme nt could address this inaccurate perception by better communicating to staff that employees’ op inions and input are valued. Management should also seek to understand why staff is less like ly to agree that management listens to their concerns. There was a lack of agreement in regard to the dimension of trust, with management scoring significantly higher than staff. This tim e, both parties in the relationship were aware of this difference. Management accurately pr edicted that staff would score lower than management, while staff accurately estimated that management would rate the trust measures higher than staff. So, there appears to be a re al disagreement (dissensus) between management and staff in terms of the level of trust in the re lationship. If management isn’t already taking steps to communicate to employees that they can depend on management to treat them fairly, then steps need to be taken to do so. When looking at the dimension of commitment, th ere is evidence of a false conflict. While there is actually agreement between management and staff in regards to the level of commitment in the relationship, management wrongly perceives that staff rates the relationship lower in terms of commitment while staff incorrectly perceives that management rates the relationship higher. However, while statistically significant, this perceived difference is slight. Regardless,

PAGE 56

56 management should take action to not only incr ease the amount of actual commitment to the relationship, but also demonstrate to staff that the two parties are actually in agreement on this issue. The satisfaction dimension of the OPR provide s an example of a consensus – both sides agree on how satisfied they are with the relation ship and they accurately understand each other’s viewpoint. Management’s communication and polic y efforts don’t necessarily need to focus on clarifying misunderstandi ng regarding this dimension; effo rts can focus instead on increasing actual satisfaction with the relationship. In summary, the underlying relationship betw een the management and staff at this particular organization appears to be fairly healthy; all the means for the dimensions were above “5” indicating that on average, respondents at least “somewhat agreed” with the relationship measures. These relatively positive ratings should mitigate somewhat the findings of disagreement and false perceptions on some of the measures. While there were statistically significant differences in many cases, for practical purposes, many of the scores were still on the positive end of the rating scale. Nonethel ess, future communication efforts between management and staff should address some of the areas of misperception regarding the organization-public relationship. The results are interesting in that they provi de examples of each of the four possible coorientational states: consensu s, dissensus, false consensus, and false conflict. Control mutuality demonstrated a false consensus. Empl oyees believe that management and staff are in agreement on the amount of control mutuality in the relationship even though they really disagree. A true dissensus was evident on th e trust dimension with both sides accurately perceiving that management felt that the relations hip was more trusting than staff did. A false

PAGE 57

57 conflict existed between management and staff in regards to commitment. While both sides actually rated the commitment measures similarl y, they each incorrectly perceived that they disagreed. Finally, the satisfaction dimension pr ovides an example of true consensus. Both parties accurately perceived that they were in ag reement as to how satisfied they were with the relationship. It should be noted that there are distinct limitations on the conclusions drawn from this pilot study. Most notably, the sample used in the pilot study was a convenience sample and does not represent a random selection of managers and employees from the organization. Additionally, it was a relatively small sample. Another concern has to do with motivated misreporting on the part of staff re spondents. It is possible that st aff, due to fears of reprisals from management, might be motivated to give socially desirable res ponses and overstate the degree to which they agreed with the relatio nship measures. Due to these restrictions, generalizing these findings to the whole organizati on is not possible. However, the purpose of this pilot study was not to paint a detailed picture of the relationship at th is specific organization, but rather to investigate the feasibility of us ing a coorientational approach to measuring an organization-public relationship using the HonGrunig (1999) relationshi p measures and to uncover any potential problems with the methodol ogy prior to implementing a broader study. These observations will be addressed further in the following section. Evaluation of Pilot Study In moving forward with the main study, the mode of survey administration will need to be tailored to the organization that will participate in the main study. Due to concerns over motivated misreporting, it might be necessary to administer the instrument face-to-face if measuring the perceptions of an internal public. If not face-to-face, then mailing lists for these publics would need to be to be obtained so that the survey and a self-add ressed return envelope

PAGE 58

58 can be sent directly to thes e respondents; however, while th is might reduce the issue of motivated misreporting, it would certainly affect response rates and increase the amount of time necessary for data collection. A happy medium might be to have the surveys distributed within the organization via the organizational contact person, but include an envelope that the completed surveys can be sealed in prior to re turning them. Another op tion may be to develop an online version of the survey that can be taken anonymously. The pilot study did demonstrate that measur ing directand meta-perspectives of both a public and an organization within a coorientational framework is feas ible as long as care is taken in the administration of the two separate survey instruments to ensure that the correct instrument is provided to the correct type of respondent Also, the scales demonstrated good internal reliability despite the fact that the wording of the items was slightly modified to fit the circumstances of the particular application and th at they were administered not only to a public sample, but also to a sample from the organizati on. Finally, the result s of the pilot study were encouraging in that comparisons of the various perspectives made it possible to generate assessments of accuracy, agreement, and perceived agreement. Analysis of these coorientational measures made it possible to determine coorientational states that existed between the organization and the public in re gards to the relationship dimens ions of trust, satisfaction, commitment, and control mutuality. Main Study Having pilot tested an instrument that used the modified Hon-Grunig (1999) relationship measures within a coorientational framework, the cooperation of a client organization and access to a relevant stakeholder public was secured fo r the main study that was conducted in the spring of 2007.

PAGE 59

59 Organization & Stakeholder Public The organization that agreed to particip ate in the main study is a university police department for a large southeastern univer sity. The organization has approximately 130 employees including 90 sworn and 40 non-sworn offi cers and support staff. It serves a campus covering 2,000 acres with a student body of a pproximately 50,000 (average annual enrollment). This includes approximately 7,000 undergraduat e students living on-campus in residence halls and 1,000 married and graduate students living on -campus in family-housing villages. These oncampus students represent the primary stakehol der public for the organization (J. Holcomb, personal communication, Jan. 26, 2007) and are th e focus of the study. Access to these students was negotiated through the university’s housing a nd residence education office. The primary contact for the university police department was a captain in the community services division. The primary contact for the housing office was the coordinator of research programs and services operating out of the marketing a nd public relations department. While the use of student samples is typical ly frowned upon, within the context of organization-public relationship research analyzing the relationship between an organization (here, the university police de partment) and its primary stake holder public (here, on-campus students), the use of a student sa mple is justified and may be more telling than using a sample drawn from the general public. University police and students interact with each other on a regular basis; their decisions and behavior potentially have a direct effect on one another. Using a student sample that composes a stakeholder public that is involved in a direct relationship with an organization may represent an improvement over other organizati on-public relationship studies that examine the relationship between an organization and the general public who may only know the organization by their reputation and are not directly involved with the organization (Hung, 2006; Ki and Hon, 2007).

PAGE 60

60 Mode of Survey Administration The main study utilizes a Web-based survey instrument to measure perceptions of the organization-public relationship th at exists between the univers ity police department and the students living on-campus; these students are th e organization’s primary stakeholder public. The mode used in my study, the Internet, warrant s some discussion since the mode of survey administration can affect several aspects of the survey design, including the available sampling frame and sampling design, coverage of the ta rget population, survey nonresponse, and data quality, as well as the cost and time required to administer the survey (Groves et al., 2004). Couper’s (2000) review of Web-based survey tec hniques addresses several of these concerns from a survey error perspective, specifically noting that Internet surveys are potentially problematic when it comes to coverage error, no nresponse error, and measurement error. Several of these issues are addressed in turn. Regarding the sampling frame and target popula tion coverage issues, Couper (2000) points out that one of the greatest draw backs to the use of th e Internet to survey the general population is under-coverage of target popula tions. Coverage issues arise when there is a mismatch between the available sampling frame from which potenti al respondents are selected and the target population of interest (Groves et al., 2004). Ma ny members of the general population are not reached via Web-based surveys due to the lack of penetration of Internet availability. However, this drawback is minimized if the target popul ation is known to have Internet access (Couper, 2000). Aoki and Elasman (2000) state that “though th ere are still limitations to be overcome if the Web is used for general population survey, the Web will present adva ntages over traditional modes of data collection if it is used for specific populations that are known to be Internet savvy” (p. 3).

PAGE 61

61 Such is the case in my study where both populat ions of interest are known to have access to the Internet. All members of the university police department are assigned university e-mail accounts and have access to the Internet at headqu arters. Likewise, all students enrolled at the university are assigned a university e-mail account which is either their primary e-mail address or which can be forwarded to another e-mail address. Students are encouraged to check their university e-mail regularly as this is the primar y, official means through which the university communicates with them. Furthermore, the university requires that all students have a computer; even if they don’t have a personal computer at their place of residenc e, the university has numerous computer labs with Internet access available for student use. An additional coverage issue related to Inte rnet surveys is the availability of sampling frames (Groves et al., 2004). Even if the targ et population is known to have computer and Internet access, an adequate sampling frame may not be available from which to draw a sample. For my study, this is not a factor since list frames of e-mail addresses are available for both populations of interest. An e-ma il distribution list of all sworn and non-sworn officers currently employed at the university police department was made available through the community services division (J. Leffert, personal comm unication, March 28, 2007). Similarly, the university’s office of housing and residence education maintains an e-mail distribution list of all undergraduate and graduate student s currently living on campus that is updated every semester (Y. Zhang, personal communication, March 30, 2007). Another issue of major concern for We b-based surveys is survey nonresponse and potential nonresponse bias. Nonresponse occurs wh en subjects do not par ticipate in the survey because they are either unwilling or unable to co mplete the survey instrument. Consequently, nonresponse error results when respondents and nonrespondents differ significantly on survey

PAGE 62

62 statistics (Couper, 2000; Groves et al., 2004). Mo de can affect response and nonresponse rates due to the presence or absence of an interviewe r who can assist with the administration of the survey either by helping to ensure that res pondents receive the instrument, making sure the respondents understand the survey directions, and by encouraging respondents to complete the instrument (Groves et al.). Self-administere d questionnaires complete d by mail and the Web typically feature lower response ra tes than surveys that are admi nistered by an interviewer. Studies on response rates have placed the av erage mail survey response rates between 40% (Kerlinger, 1986) and 56% (Baruch, 1999), altho ugh Dillman (2000) suggests that by utilizing strategic design features, response rates can be improved to approximately 70%. While many studies have suggested ways to im prove mail survey response rates, few have looked at methods for increasing Web survey response rates. Response rates for Web-based surveys are typically low, with various published studies reporting respon se rates ranging from 15% (Porter & Whitcomb, 2003) to 41% for comp leted surveys (Couper, Traugott, & Lamias, 2001). One meta-analysis reported an average re sponse rate of 34% for Web surveys (Cook, Heath, & Thompson, 2000). While the conventi onal wisdom says that what works for mail should work for the Web, Couper (2000) disagr ees, suggesting that Web surveys often cannot use these techniques due to the technical difficu lties and confidentiality issues involved with Web use. In a meta-analysis of 68 Web surveys, Cook, Heath, and Thompson (2000) identified number of contacts, personalized contacts, and precontacts as methods for boosting the response rates of online surveys. Porter and Whitcomb’s (2003) investigated recruitment e-mail features on response rates for Web surveys and found that a deadline for completing the survey along with stressing how the respondents were able to provide specialized information not available from the general public could improve response rates. It should be noted that in their study, other

PAGE 63

63 design features that increased response rates in mail surveys, such as personalized salutations, did not have an effect on response rates for Web surveys. Couper (2000) also identified measurem ent error as a potential problem with Web surveys. Measurement error occurs when responses to the survey questions do not reflect the true value of the answers (Groves et al., 2004). Th is can be due to seve ral reasons, including satisficing (mental shortcuts that respondents engage in to simplify the survey response process), motivated misreporting (when the respondent delib erately provides an answer that does not reflect a true attitude or behavior for social de sirability purposes), bias ed or confusing question and response option wording, the format of respon se options, and question order. This is related in part to the lack of a traine d interviewer who can he lp the respondent by clarifying questions or providing motivation to complete the instrument. It should be noted that these problems are not unique to Web surveys, but can become an issu e for other self-adminis tered questionnaires as well. Another issue related to mode selection is data quality; specifically the completeness of the data. Groves et al. (2004) acknowledge that one advantage that computer-assisted selfadministered questionnaires have is that they have low missing data rates. In the present survey, the Internet survey was configured to prompt the respondent to complete all questions on any particular page of the survey be fore moving on to the next page. The Web may present some advantages to survey research in regard to two other factors – cost and time. Groves et al. (2004) state that the fixed co st of a Web-based survey can be close to zero; this was certainly the case in my study wher e the online survey service that was used cost only $20 for one month and allowed up to 1,000 re sponses per month (charging $0.05 for every

PAGE 64

64 response beyond that). Another advantage provide d by the use of a Web-based survey was the speed with which recruitment and data collec tion could take place. Couper (2000) observes: “Web surveys offer the research community enormous opportunities for low-cost self administered surveys using a wide variety of stimulus material (sound, image, video, etc.) that has heretofore simply not been available fo r it has been too costly to implement widely in interviewer-administered surveys” (p. 477). In selecting a mode of survey administrati on for my study, all of these considerations were weighed against one anothe r along with considering the t ype of organization and public under investigation and the type of access to these populations that was available. In the end, the coverage issue related to Intern et surveys was deemed not to be relevant since my study had access to list-based samples of high-coverage pop ulations. In an attempt to increase response rate, steps were taken such as stressing organi zational sponsorship in the recruitment e-mails; furthermore, if there was a low response rate, it did not necessarily mean that nonresponse bias was present. Additionally, the use of Internet surveys offered several advantages, such as ensuring completeness of the data, making a complex instrument easier to administer than it was when presented on paper, and reducing th e time required to collect the data. Sampling and Recruitment My study features a probability sampling technique that utilizes two list-based samples of high coverage populations (Couper, 2000) – a sample of member s of the university’s police department using a departmentally controlled li st of employee e-mail addr esses and a sample of currently enrolled students that live on campus drawn from a recently updated list of e-mail addresses maintained by the university housing o ffice. Similar to Couper, Traugott, and Lamias (1999), samples were drawn by contacting individua ls listed on the sampling frames with an invitation to participate in the survey sent via an e-mail that included a link to the survey Web site.

PAGE 65

65 The recruitment e-mails stated that the partic ipation of the respondent was desired in order to better understand the relations hip that existed between the un iversity police department and students living on campus. It inco rporated several features that had been shown to improve response rates for Internet survey s as well as features that were known to increase response rates for mail surveys (Cook, Heath, & Thompson, 2000; Couper, 2000; Porter & Whitcomb, 2003; Groves et al., 2004). These features incl uded a deadline to complete the survey, acknowledgement of organizational sponsorship from the university, and an assurance that the data was not being collected for marketing pu rposes. Since the e-mail to members of the university police department was distributed internally via a lis tserv, a personalized contact could not be included. The e-mail that students received did include a personalized contact of sorts in that the e-mail was addressed to their individual university email accounts and not to a university listserv. This coul d have been taken a step fu rther by personalizing the actual salutation within the e-mail itself with the re cipient’s name, but the technical difficulties involved with this precluded it from being done in a satisfactory manner; that is to say, there would have been the possibility that the automated survey softwa re would incorrectly format the respondent’s name and make it appear an improperly addressed spam e-mail. Manually personalizing the recruitment e-mails would have been problematic as well due to the sheer number of e-mails that were being sent. Also, since the list I was working from was a computer generated database provided by the housing o ffice, any incorrectly formatted names on the sampling frame list would have been carried over to the recruitment e-mails anyway, regardless of whether they were entered manually or by using the survey software. Each survey also included an informed consent statement as required by the university’s institutional review board. This statement about the respondent’s rights and the risks involved

PAGE 66

66 with participation in the study must be included whenever human respondents are used. The recruitment e-mail included an acknowledgment that clicking on the link to the survey Web site constituted consent. See Appendix D for copies of the initial and follow-up recruitment e-mails used with each sample. On the morning that the surveys were opene d up for online access, th e first organizational recruitment e-mail was sent to all sworn and non-sworn officers ( N =130) employed by the university police department. A similar recruitment email tailored for the public sample was sent to a random sample of 1,235 students living on ca mpus. A systematic random sample with a random start point (determined by using a rando m number generator) was used to draw 1,235 student e-mails from the list provided by the university housing office including 1,035 undergraduate students living on campus in residence halls and 200 graduate students living in family housing villages. The initial sample si ze was determined in discussions with housing office officials who initially approved that num ber for contact. Estimating a response rate of approximately 30% (Cook, Heath, & Thompson, 2000), it was believed that this sample would yield an adequate number of completed survey s to make the results generalizable to the population of 8,175 students living on campus (b ased on a 95% confid ence level with a confidence interval of +/5%, 367 completed su rveys were needed (Austin & Pinkleton, 2006)). For the organizational sample, multiple c ontacts were made over the two week data collection period to improve the response ra te (Cook, Heath, & Thompson, 2000; Couper, Traugott, & Lamais, 2001; Groves et al., 2004; Porter & Whitcomb 2003). Members of the university police department were contacted agai n three business days after the initial contact and again five business days following the second c ontact. This is similar to the strategy used by Couper, Traugott, and Lamais (2001), who acknowle dged that one of the major advantages of

PAGE 67

67 Web-based surveys was the speed with which follow -up contacts could be made in an effort to increase response rates. The sampling strategy used with the student sa mple had to be adjusted during the course of the study. There was an incredibly poor res ponse rate after the fi rst attempt to recruit respondents from the initial approved sample of 1,235 students. Fearing that the low response rate would lead to a small sample size that would prove inadequate for statistical analysis, approval was sought and obtained from the univers ity housing office to open up the survey to all students living on campus. Another recruitment email was sent to the remaining 6,940 students on the university housing office’s e-mail list. Although longer data collecti on periods have been shown to increase response rate (Groves et al., 2004; Heberlein & Baumgartner, 1978), the online surveys in my study were only kept open for two weeks. This was due in part be cause the survey needed to be administered to the organization and public simultaneously to ensure that all the responden ts would be evaluating the organization-public relationship at the same point in time. It was also hoped that this would control for any artifacts create d by real world events happening at different points in time. Additionally, due to the speed with which responses are received after contact, it was believed anyone that was going to respond to the survey w ould do so relatively soon after receiving the recruitment e-mail (Couper, Traugott, & Lamias, 2001). At the end of the two week period, 56 memb ers of the university police department had accessed the survey and completed the first section measuring direct perspectives of the relationship, yielding a “click-thr ough rate” of 43.1% (typically, Internet surveys have reported click through rates anywhere from 1%-30% (Wimmer & Dominick, 2002)). Forty-four respondents completed the entire survey, resulting in a 33.8% re sponse rate for the organization

PAGE 68

68 sample, which was similar to the average Web su rvey response rate reported by Cook, Heath, and Thompson (2000). For the student sample, 703 respondents acce ssed the survey and completed the first section, yielding a click-through rate of 8.6%. The final completion rate was 5.7% with 468 respondents completing the entire survey. While the final valid sample size exceeded the 367 responses that were deemed necessary for stat istical analysis, the low response rate raised concerns about the pote ntial for nonresponse bias among the student sample. Survey Instrument Construction After clicking on the link in the recruitmen t e-mail, respondents were directed to the survey Web site featuring one of two versions of the survey; students were directed to the publicversion of the survey, while university poli ce department members were directed to the organization-version of the survey. The basic st ructure of both surveys was similar, with the landing page featuring a greeting along with a short blurb stressing the importance of the study and the importance of the answers the respondent s would provide. The remainder of the online survey was divided into four sections. Secti on 1 asked respondents to provide their direct perspective of the rela tionship. Section 2 asked respondents for their meta-perspective of the relationship. Section 3 included questions about exposure to rela tionship maintenance strategies and the type of interpersonal interaction the responde nt had with the other party. Section 4 concluded the survey by request demographic in formation that included how long the respondent had been involved in the organization-public rela tionship. Copies of both survey instruments are included in Appendix C. Each section of the survey was presented on one online page. Peytchev, Couper, McCabe, and Crawford (2006) had previously investigated the effects of using paging versus scrolling to view and complete online survey s; they determined that item nonresponse was slightly higher for

PAGE 69

69 surveys that used scroll ing arguing that one drawb ack to scrolling is that it is not possible to retain information from partially completed inte rviews. For my study, respondents were required to provide an answer to all the questions on the pa ge before they could proceed to the next page; if they attempted to continue to the next pa ge without answering a question, a pop-up dialogue box would appear and ask them to answer all the questions on the page before continuing. This was done in an effort to cut down on item-missing da ta and facilitate the re tention of data from partial interviews. Respondents were not prevented from returning to a previous page in order to change an answer. Previous research has shown that this helps to reduce social desirability effects (Richman, Kiesler, Weisband, & Drasgow, 1999). Add itionally, since an actua l list of student emails was provided by the housing office in a database format, it was possible to manage access for each individual respondent using the survey Web site’s list management tool. Thus, if a student was to stop working on the survey due to a timed-out Internet connection, they could return to the site and continue taking the survey from the point they left off; again, it was hoped that this would help increase the completion rate and reduce the amount of item-missing data. Since the police department survey s were distributed through a lis tserv, it was not possible to track those surveys individually; therefore, this function was not availa ble for the respondents completing the organization-version of the survey. Operationalization of Independent Variables Party Membership. Each survey was coded for 1 if the person completing the survey was a member of the police department (i.e., the orga nization), or 2 if they were a student living on campus (i.e., the public). Relationship Maintenance Strategies. Section 2 on the public survey included measures of exposure to organizational relationship maintena nce strategies that reli ed on either one-way or

PAGE 70

70 two-way communication. Students were asked a bout exposure to one-way strategies using a checklist item that asked whether they have vis ited the police department’s Web site, had read one of their brochures, read one of their news letters, read a press release issued by the department, or listened to a speech by a member of the department. Each item was coded 0 if the respondent had not been exposed to it and 1 if they had. Exposure to two-way relationship maintenance strategies that involved engagement in services or programs offered by the organization was also measured. Students were asked about their participation in these programs on a check list item. These presentations, crime-prevention classes, educational seminars, and other pr ograms are offered on a regular basis by the organization and exhibit many of the features of two-way symmetrical maintenance strategies including access, positivity, shar ing of tasks, networking, and integration (Grunig & Huang, 2001; Hon & Grunig, 1999). Interpersonal Interaction. The amount of interpersonal cont act that the parties had with one another was measured with an item that aske d how many hours in an average day that they spent communicating with a member of the othe r party face-to-face, online, and on the phone. Responses were indicated on a drop-down menu w ith response options th at ranged from 0-24 hours. Students were also asked how many hours in a an average week that they communicated directly with police department members; again, responses we re indicated on a drop-down menu with response options ranging from 0-24 hours a nd included a “more than 24 hours” option. Interpersonal interaction was also included in a checklist item on the student survey and included whether or not a student had called the university police department, wrote an e-mail or letter to the organization, visited the police depa rtment’s offices, or talked face to face with a member of the department. These items were also coded 1 if present and 0 if absent.

PAGE 71

71 Respondents were also asked to indicate how they would characteri ze these inte rpersonal interactions by indicatin g valence on a seven-point Likert scale that ranged from 1 (“very negative”) to 7 (“very positive”). Time. Time in the relationship was assessed diffe rently on each survey. Members of the police department were asked on the organization version of the survey to indicate how many years they had worked in their current position, how many years they ha d worked at the police department in all, and how many years that they had worked at the university. Each response was indicated on a drop-down menu with response options ranging from 1-20 and included “less than one year” and “more than 20 years” options. Students were asked on the public version of the survey to indicate how many years they had been a student at the university and how many years they lived in their current location. Responses to both questions were indicated on a drop-down menu with response options ranging from 1-7 years and that included “less than one year” and “more than seven year” options. Media Exposure. Students were also asked on the public version of the survey if they had ever seen or read stories about the university police department in the media. “Yes” responses were coded as 1 and “no” responses were coded as 0. They were also asked about how they thought the police department was portrayed in these stories. Responses were indicated on a seven-point Likert scale ranging from 1 (“ very negative”) to 7 (“very positive”). Demographics. Students were asked to indicate their gender (“male” coded as 0, “female” coded as 1), their age using a drop-down menu ranging from 18-64, and their class standing at the university (coded 1-5 for “freshman,” “sopho more,” “junior,” “senior,” and “graduate student” respectively). They were also asked wh ether they lived on or off campus, whether they had ever lived on campus, whether they drive or park a car on campus, and whether they had

PAGE 72

72 ever been cited by the university police department for any reason. Responses to these questions were coded 1 for “yes” and 0 for “no.” Members of the university police department we re also asked to indicate their gender and their age using the same response options as abov e. They were also asked to indicate whether they were a sworn (coded as 1) or non-sworn officer (coded as 2). An open-ended response box was also included so that they could enter the ti tle of their current position at the university police department. Operationalization of Dependent Variables. Direct Perspective. Each respondent’s direct persp ective of the four relationship dimensions of trust, control mutuality, satisfac tion, and commitment was measured using the 21item Hon-Grunig (1999) relationship scale. Each ite m was slightly reworded as in the pilot study so that they were objective in nature and applicable to the specific organization-public relationship under investigation; th e reworded items are presented in the survey instruments in Appendix C. As in the pilot study, the responde nts indicated their degree of agreement or disagreement with each item on a seven-point Li kert scale ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). One advantage of using th e online survey to present the relationship scale items was that the survey software would random ize the order of the items for each respondent, thus controlling for question order effects. Meta-Perspective. After responding to the relationship items measuring the respondent’s direct perspective, they were taken to the next page of the survey were they were asked to provide their estimates of the other party’s view of the relationship (i.e ., the respondent’s metaperspective). This was measured using the sa me 21-item relationship measures as before. Responses were measured using a seven-point Likert scale. For the organization version of the survey, the scale ranged from 1 (“a student w ould strongly disagree”) to 7 (“a student would

PAGE 73

73 strongly agree”); for the student ve rsion of the survey, the scale ra nged from 1 (“a member of the university police department woul d strongly disagree”) to 7 (“a me mber of the university police department would strongly agree”). One Final Variable – A Real World Event As mentioned previously, the da ta collection period was limited to two weeks in part to avoid the possibility of real wo rld events occurring that could have an effect on the manner in which respondents perceived the organization-pub lic relationship. Originally, the objective was to ensure that members of the organization and members of the public were evaluating the relationship simultaneously; for example, the resear cher did not want to survey the organization at Time 1 and then later survey the public at Time 3 if some event that could affect the relationship occurred at Time 2. Sadly, such an event did take place during the course of the study. Six days into the data collection period for my study, a gunman shot a nd killed 33 students a nd faculty on the campus of Virginia Tech University; the event rece ived heavy media coverage for several weeks afterward. Obviously, this raised the issue of campus security at universi ties around the nation, and the university that was the site for data collection in my study was no exception. The day following the shootings, university administrators sent an e-mail via the campus listserv to every student, faculty member, and staff member at the university. In the e-mail, the university expressed their sorrow over the shooting, recognized the possibl e security concerns of the campus community, and stressed that the universit y police department was prepared to handle a similar event should it ever occur on their campus. To account for this event, an additional independe nt variable related to the influence of real world events is included in the analysis. Surv eys received prior to the distribution of the

PAGE 74

74 university’s e-mailed statement regarding the shoo ting were coded as “0” and surveys received after the e-mail was sent were coded as “1.” Includ ing this variable in the analysis may indicate the influence of this real world event on the salience of the relations hip to respondents and whether it effected respondent’s perceptions of the organization-public relationship between students and the university police department. It should also be noted that two weeks before data collection started, a city police officer was killed by a drunk driver during the celebra tion following a university sporting event. Since this event did not involve a university police offi cer, it was not included as a real world cue in my study. Even if students did not draw a distinc tion between city and university police officers, the incident still took place well before data coll ection started, so respondents would have prior knowledge of the event before taking the surv ey, unlike the school shooting incident, which could have affected perceptions of re spondents during the course of the study. Data Analysis Strategy After the reversed control mutuality item was recoded, Cronbach’s Alpha was calculated for each relationship measure to ensure the scale items representing that dimension were reliable. The scale items representing each relationship meas ure were then averaged to obtain the mean trust, satisfaction, control mutuality, and co mmitment scores for each respondent’s direct perspective and meta-perspective. An overall relationship measure was also calculated by averaging all 21 scale items for each respondent’s direct perspective and meta-perspective. D-scores were then calculated for each re spondent by subtracting th e average relationship ratings of their individual dir ect perspective and meta-perspectives from the average direct perspective ratings for the other party as a whole. This procedure generated two sets of d-scores: (a) one set of d-scores that indicated accura cy between the respondent’s individual metaperspective and the other party’s overall direct perspective, and (b) one set of d-scores that

PAGE 75

75 indicated agreement between the respondent’s direct perspective and the other party’s overall direct perspective. Congruency was calcula ted by finding the difference between each respondent’s own perspective and th eir own meta-perspective. The ab solute value of all d-scores was then calculated so that they were positive. Smaller d-score values in dicate stronger levels of agreement, accuracy, and congruency while larger d-score values indicate weaker levels of agreement, accuracy, and congruency. To address the research questions regarding the type of relationship states that exist between the organization and the public, repeated measures multiple analysis of variance (MANOVA) was used. Follow-up ANOVAs were also used to provide additional information regarding the interaction of the various perspectives. To address the impact of the independent variables (time, maintenance strategies, interp ersonal interaction) on the dependent variables (relationship measures, coorientation measures), multiple linear regression analysis was used.

PAGE 76

76 Table 3-1. Management and staff direct perspectives. Management Staff Relationship dimension M SD M SD t ( df = 35) Control mutuality 5.96 0 .65 5.11 0 .9 3.17 ** Trust 6.31 0 .3 5.37 1.1 3.31 ** Commitment 5.66 0 .65 5.3 1.4 0 .949 Satisfaction 5.52 1.1 5.26 1.3 0 .671 p <= .05 ** p <= .01 *** p <= .001

PAGE 77

77 Table 3-2. Management direct and meta-perspectives. Management direct perspective Management metaperspective Relationship dimension M SD M SD t ( df =15) Control mutuality 5.96 0 .65 5.01 0 .96 3.54 ** Trust 6.31 0 .3 5.6 0 .99 2.91 Commitment 5.66 0 .65 5.15 1.1 2.26 Satisfaction 5.53 1.1 5.38 0 .85 0 .722 p <= .05 ** p <= .01 *** p <= .001

PAGE 78

78 Table 3-3. Staff direct and meta-perspective. Staff direct perspective Staff metaperspective Relationship dimension M SD M SD t ( df =20) Control mutuality 5.11 0 .90 5.36 0 .97 -1.76 Trust 5.37 1.1 5.85 1.31 -3.55 ** Commitment 5.3 1.4 5.7 1.36 -2.22 Satisfaction 5.26 1.28 5.53 1.28 -1.67 p <= .05 ** p <= .01 *** p <= .001

PAGE 79

79 Table 3-4. Management direct pers pective to staff meta-perspective. Management direct perspective Staff metaperspective Relationship dimension M SD M SD t ( df = 35) Control mutuality 5.96 0 .65 5.36 0 .97 2.14 Trust 6.31 0 .30 5.85 1.3 1.38 Commitment 5.66 0 .65 5.70 1.36 -.114 Satisfaction 5.53 1.10 5.53 1.28 -.021 p <= .05 ** p <= .01 *** p <= .001

PAGE 80

80 Table 3-5. Staff direct perspectiv e to management meta-perspective. Management metaperspective Staff direct perspective Relationship dimension M SD M SD t ( df = 35) Control mutuality 5.01 0 .96 5.11 0 .90 -.330 Trust 5.60 0 .99 5.37 1.1 0 .661 Commitment 5.15 1.1 5.30 1.39 -.365 Satisfaction 5.38 0 .85 5.26 1.28 0 .319 p <= .05 ** p <= .01 *** p <= .001

PAGE 81

81 Figure 3-1. Coorientation model of pilot st udy organization’s manageme nt-staff relationship. Note : Values with the same superscripts indicate means that there are significantly different ( p < .05). In the model, CM = control mutuality, T = trust, C = commitment, and S = satisfaction.

PAGE 82

82 CHAPTER 4 RESULTS The following chapter details the findings of the multivariate procedures that were used to analyze the data for my study. Fi rst, I discuss the examination and cleaning of the data set, including checks for data entry errors, missing data, constructi on of the summated relationship scales, and assessments of scale reliability. I then present the demographics of the public and organization sub-samples. Then, I present the results of the multiple analysis of variance and analysis of variance procedur es used to explore the research questions regarding the coorientational approach to measuring relationships. Finally, I detail the results of the multiple linear regression analyses used to test the hypoth eses regarding the effect of the independent variables on the relationship and coorientation measures. Preparing the Data for Analysis Once access to the online surveys closed at the end of the data collect ion period, the data sets were downloaded into Excel for examination and cleaning. The responses on each variable were reviewed to ensure that no impossible va lues were entered (e.g., an “8” entered as a response to a seven-point Like rt scale). For the question re garding exposure to two-way relationship maintenance strategies, when the “o ther program” checklist item was selected, the open-ended response was reviewed to see if it was categorizable in to one of the other designated checklist items. For example, if a respondent did not indicate having attended a presentation on safety or crime prevention, and then checked “other” and indicated that they had attended a bicycle safety presentation, then the item was r ecoded from “other” to “attended an educational presentation on personal safety or crime prevention.” Since the data set downloaded from the survey Web site did not have skipped and missing item codes included, they had to be entered into the data set. For items missing due to a skip

PAGE 83

83 pattern within the survey, a code of “88” was en tered. For items that were missing due to item nonresponse, a code of “99” was entered. Next, the reversed control mutuality item on both the public and organization versions of survey had to be recoded. This was done for both relationship scales (direct perspective and meta-perspective scales). Once this was done, the public data set and the organization data set were combined and imported into SPSS. There, th e variable attributes we re assigned (e.g., name, scale, missing value designators, etc.). Item-Missing Data Item-missing data is when information is missing for some items on an observation that has provided data on other items; this happens when a respondent answers some questions and then fails to answer others, either due to skippi ng the item or terminating the survey (Groves et al., 2004; Hair, Anderson, Tatham, & Black,1998). Missing data is a common problem in survey research, with studies in some fields reporting that 50% or more of the respondents in their surveys have missing data (Acock, 2005). Exam ining the data set for my study, 29.9% of observations had item-missing data after completing the first section of the survey; that is to say, the respondent terminated the surv ey after completing the first “p age” of the survey online. To assess the randomness of the o ccurrence of missing da ta in the data set, a new variable, “missing,” was created and coded 0 if the respondent did not termin ate after the first section and coded 1 if the respondent did terminate after the first section. A multiple analysis of variance (MANOVA) was calculated to exam ine whether the observations th at had missing data or not had an effect on the direct perspective ratings of the relationship dimensions. No significant effect was found ( F (4, 752) = .959, p > .05). None of the direct perspective ratings of the relationship dimensions were significantly in fluenced by whether or not the respondent terminated the survey (and thus had mi ssing data on the rest of the survey).

PAGE 84

84 Additionally, a chi-square test of independence was used to examine the difference in the proportion of respondents who had mi ssing data before receiving the e-mail from the university administrators regarding the Vi rginia Tech shooting (.31) to the proportion of respondents who had missing data and completed the survey after the e-mail was distributed (.26). No significant difference was found (chi-square (1) = 1.543, p > .05). A chi-square test of independence was also calculated to compare the frequency of students who had missing data to the frequency of university police department members who had missing data. In this case, a significant interaction was found (chi-square(1) =4.213, p < .05). Students were more likely to have missing data (30.9%) than members of the organization (17.9%). Although significant, coupled with th e results of the MANO VA and the other chisquare test, this would seem to indicate that ther e is not an underlying pa ttern to the item-missing data; it comes as no surprise that students would be more likely to terminate the survey process earlier than the university police department sa mple. It was assumed that the data was missing completely at random (MCAR); that is to say, that the observed ratings of the direct perspective relationship measures (from section 1 of the survey ) were a truly random sample of all the direct perspective measures (observed and unobserved) w ith no systematic pattern present that would lead to classifying the missing data as missing at random (MAR) (Hair et al., 1998). The next decision was how to handle the missing data. Although there are methods available for imputing data, it was felt that due to the high level of cases missing data, that imputation (especially imputation of mean values ) would distort the sample statistics (Acock, 2005; Hair et al., 1998; Groves, 2004) which can crea te a serious problem if the data is being analyzed using multivariate techniques (K alton & Kasprzyk, 1982). Other commonly used options include casewise (also referred to as listw ise) deletion or pairwi se deletion. Casewise

PAGE 85

85 deletion removes observations from the data se t that have item-missing data on any variable, while pairwise only omits observations that are missing data on the variables used in any given calculation. Thus, the advantage of pa irwise deletion is that it uses all the data that is available in the data set and is thought to be acceptable when the data is assumed to be MCAR (Acock, 2005). Therefore, all the multivaria te procedures used in data analysis for my study excluded missing data pairwise unless otherwise specified. Normality One of the assumptions for the use of M ANOVA is that the dependent variables are normally distributed. First, an assessment of normality of the dependent variables using the Shapiro-Wilks test of normality was conducted, as this was the most appropriate test given the sample size (Hair et al., 1998). The student sample and the organization sample were evaluated separately since they would be compared to one another in MANOVA a nd had different sample sizes. The test was significant for all the dir ect perspective measures of the relationship, indicating that sample di stribution of those dependent variable s were different from that of the normal population distribution. For the meta-persp ective measures, the test of the organization sample was not significant, but th e test of the student sample was significant, again, indicating a non-normal distribution of those me asures within that sample. Next, histograms, box plots, and Q-Q plots of the distribution of th e dependent variables within the public and organization samples were evaluated. The histogra ms and box plots for the public sample appeared acceptable with the excep tion of the meta-perspective dimensions, which were slightly skewed. The plots for the organizati on sample were also skewed with the exception of the plots for the meta-perspective dimensi ons. Next, the univariate P-P plots for each dependent measure were examined. The direct pe rspective measures were normally distributed while the meta-perspective plots were skewed.

PAGE 86

86 Several data transformations, including s quare-root and log transformations, were attempted to improve the data, but after another review of the various plots, the transformation appeared to only make the data increasing abnorma l. The next step was to look for outliers that could be acting as leverage points within the da ta. Box plots of each dependent variable were created for the organization and student samples. Visual examination identified a few outliers within the student sample, but they were not ex treme cases and the same case did not generate outliers on multiple variables. On the other hand, examination of the box plots for the organization sample identified four cases that were extreme outliers on multiple variables. The decision was made to remove these cases from the analysis. The normality assessment was run again; this ti me the Shapiro-Wilks test statistic was not significant for the dependent variables within th e organization sample – a marked improvement over the previous test. It was important that th e dependent variables were normally distributed within the organization sample as the cells that they would compose during the multiple analysis of variance calculations would be small and thus more susceptible to the violation of the normality assumption. Even though the student sample still demonstrated evidence of some violations of this assumption, this problem may be mitigated to some degree by the much larger sample size for students (Hair et al., 1998). PP plots were generated once more for the distribution of the dependent variables within ea ch sample. This time, the plots were deemed acceptable for the organization sample. Despite the significant Shapiro-Wilks test statistic, the PP plots also looked fairly normal, although more so for the direct perspective relationship than the meta-perspective relationship measures. Scale Reliability Since the Hon-Grunig relationship scales have been repeatedly validated, several times within a university relationship setting, scale reliability was assessed using Cronbach’s Alpha, a

PAGE 87

87 measure of internal consistency of the items th at compose a summated scale (Hair et al., 1998). Using SPSS, scale reliability was assessed for each set of items composing the various dimensions on the Hon-Grunig relationship scale, plus an additional analysis that looked at the reliability of all 21-items as a whole to indicate total relationship perception. The lower limit for an acceptable Cronbach’s Alpha is .70 (.60 in explor atory studies is also ac ceptable) (Hair et al., 1998). In addition to Cronbach’s Alpha, the item to tal correlations were also examined to ensure that they were greater than .50, and the inter-item correlations were examined to ensure that they were greater than .30. The Alpha for direct perspective of control mutuality was .858; removing the third control mutuality item improved reliability to .910. Sin ce the third control mutuality item had inter-item and item-total correlations below the acceptable le vels, it was dropped from the scale. The Alpha for the direct perspective of commitment was .897. The Alpha for the direct perspective of satisfaction was .94. The Alpha for the overall di rect perspective relationship measure using all 21-items was .971 (.974 after removing the third control mutuality item). The Alpha for the meta-perspective of control mutuality was .84; removing the third control mutuality item improved this to .937. Again, since the interitem and item total correlations were below the recommended levels, this item was removed. The Alpha for the meta-perspective of trust was .959, for the meta-perspective of commitment, it was .907, and for the meta-perspective of satisfaction, it was .939. The Alpha for the meta -perspective relationship measure using all 21items was .974 (.979 after removing the third control mutuality item). Based on the results of the reliability tests, the third control mutuality item was removed for both the direct perspective a nd meta-perspective scales. The remaining 20 items were used to create the summated scales for each direct and me ta-perspective relationship dimension, as well

PAGE 88

88 as the final direct and meta-perspective overall relationship measures. Suspecting that multicollinearity among the independent variables may become an issue, it was felt that having one overall measure of the direct and meta-perspective of the re lationship might prove useful. Sample Demographics Student Sample The average age of the students composing the “public” sample was 20.53 years old ( SD = 4.54). The average respondent had been a student at the university 1.82 years ( SD = 1.26). In terms of gender, 45.1% were male and 54.9% we re female. For class standing, 35.9% indicated that they were freshman, 34.2% were sophomores, 13.2% were juniors, 7.3% were seniors, and 9.4% were graduate students. These demographics of the student sample are comparable to the demographic information regarding students liv ing on campus that was available from the housing office and may help mitigate some of the concerns regarding the low response rate. Some other additional points of interest include d how many students within the sample drove or parked a car on campus (73.5%) and how many had been cited by the university police (52.4%). According to the university police department, a pproximately 85-90% of the citations that are issued to students are for traffic tickets (J Holcomb, personal communication, May 31, 2007), which will come as no surprise to anyone who has spent any amount of time on a large university campus. Also, 83.6% of the responden ts indicated that they had not received the e-mail regarding the Virginia Tech shooting pr ior to taking the survey. In regards to exposure to two-way relations hip maintenance strategies, 8.9% indicated that they had attended the “SCOPE” program (a program designed to build the partnership between the university police, th e housing office, and the students living on campus). The rape awareness and defense program (“RAD”) had been attended by 13.3% of the respondents. 21.4% indicated that they had attended some other educational presentation on crime prevention or

PAGE 89

89 safety besides SCOPE or RAD, while 2.3% indica ted that they had participated in some other program offered by the university police depart ment. All together, 36.7% of the student respondents had been exposed to at least one of these two-way re lationship maintenance strategies. For one-way relationship mainte nance strategy exposure, 50.6% said that they had visited the university police department’s Web site, 55. 9% indicated that they had read a brochure distributed by the university police department, 20. 6% said they had read one of the university police department’s newsletters, and 32% indicated that they had read a press release issued by the university police. Over 32.4% said that they had listened to a speech given by a member of the university police department. All together, 84. 1% of the respondents in the student sample had been exposed to at least one of the one-way relationship maintenance strategies. In terms of interpersonal interaction, the average amount of weekly interpersonal contact with the university police department was .43 hours a week ( SD =1.56). 33.9% of the students surveyed indicated that they had called the uni versity police, and 10.4% said that they had emailed the university police. Over 65.9% said that they had talked face-to -face with a member of the university police. All together, 75.6% of th e respondents in the student sample indicated having had some form of interpersonal interaction with the university police department. On average, students indicated that the perceived valence of their interpersonal interaction with members of the university police department was 4.18 ( SD =1.30) on a seven-point scale ranging from 1 (“very negative) to 7 (“very positive”). Finally, 78.3% of the sample indicated that th ey had seen or read a news story about the university police department in the media. The av erage valence of the portrayal of the university

PAGE 90

90 police department in these stories was perceived to be 4.91 ( SD = 1.30) on a seven-point scale where 1 indicated very negative and 7 indicated very positive. University Police Department Sample For the organization sample, the average age of the university police department respondents was 38.5 years old ( SD =9.23). In terms of gender, 72.7% were male and 27.3% were female. Over 77.3% were sworn officers and 22.7% were non-sworn officers. Respondents indicated that they had spent an average of 4.7 years ( SD =4.42) in their current position, 10.2 years ( SD =6.35) with the university police de partment in all, and 10.8 years ( SD =6.30) working at the university all together. University pol ice department members reported that they had interpersonal contact with students for 3.37 hours ( SD =2.53) a day on average, and indicated that the valence of this interpersonal in teraction was an average of 5.5 ( SD =1.09) on a seven-point scale where 1 indicated that the interaction was perceived as bei ng very negative and 7 indicated that the interaction was perceived as being very positive. Results of the Statistical Analysis The reporting of the results from the statistica l analyses is broken into two sections. The first section addresses statistical analyses that investigate the research questions regarding the efficacy of the coorientational approach and its use in evaluating an organization-public relationship. The analyses and results for this section focus on a macro-view of the overall organization-public relationship and utilize bo th the student (public) and university police (organization) samples. This addresses one of the primary goals of my study, which was to investigate the use of the coorientational appr oach for providing a snapshot of the overall organization-public relationship by looking at the interplay of the different perspectives involved in the relationship. MANOVA and ANOVA were the pr imary statistical proce dures used in this regard. The second section addresse s analyses and results related to the hypotheses regarding the

PAGE 91

91 impact of relationship maintenance strategies, time in the relationship, and interpersonal interaction on individual assessm ents of the organization-public relationship. Therefore, that section focuses on a micro-view of the relationship and utilizes the student sample. Multiple linear regression analysis is the primary statis tical procedure used for hypothesis testing. Results Related to the Research Questions To explore the research questions and begi n an analysis of the organization-public relationship as a whole, means were generated for each of the relationship measures (control mutuality, trust, commitment, satisfaction, and the overall relationship measure) for each of the four perspectives (public dire ct perspective, public meta-per spective, organization direct perspective, and organization me ta-perspective). Table 4-1 pres ents the means and standard deviations of the relationship dimensions and the overall rating of th e relationship for each perspective. The primary statistical procedure used to investigate the research questions was multiple analysis of variance (MANOVA) MANOVA is a “dependence technique that measures the differences for two or more metric dependent va riables based on a set of categorical (non-metric) variables acting as independent vari ables” (Hair et al., p. 326) and is therefore appropriate for the present investigation since the relationship measures are metric and the independent measures (party membership and perspective) are ca tegorical. Whereas independent and paired t -tests were used in the pilot study, MANOVA offers an adva ntage in that it not only accounts for the higher probability of making a Type I error due to multip le comparisons, but it is also more powerful than univariate tests at finding di fferences between dependent variables in instances where there is multicollinearity among them. The repeated measures version of MANOVA is utilized here due to the exploratory nature of the study and du e to the treatment of the “perspective” factor; there may be some question as to whether indivi duals’ assessments of the direct perspective of

PAGE 92

92 the relationship are independent of their assessment of the meta-perspective of the relationship. One of the main assumptions of MANOVA is that dependent measure observations are independent, but the use of repeated measur es MANOVA can account for the possibility of dependence (Hair et al., 1998). There are several considerations for the use of MANOVA. First, an adequate sample size is necessary; a minimum of 20 observations per cell (the groups created by the various factor interactions) is recommended. My study meets th is requirement. Normality of the dependent variables is also important and has been addressed pr eviously. Despite the fact that there was some evidence of non-normality for some of the dependent measures, the violation of this assumption has minimal impact if a large sample is used and if the differences are the result of skewness and not outliers, which is the case here. MANOVA is very sensitive to the presence of outliers; as was discussed earlier, the few outliers that were evident in the organization sample were identified and removed. Another consideration for MANOVA is that there should be little multicollinearity among the dependent variables because the use of redundant dependent measures can reduce the ability of the procedure to identify significant differe nces. A Pearson correlation of all the dependent measures of the relationship dimensions was calcula ted. The direct perspective measures were all strongly correlated with each other, with correlations ranging from .846 to .949 ( p < .001 for all the correlations). The meta-perspective measures were also strongly correlated with one another, with correlation coefficients ranging from .839 to .954 ( p < .001 for all the correlations). There was moderate correlation between the direct pe rspective and meta-perspective measures, with coefficients ranging from.425 to .591 ( p < .001 for all the correlations). Thus, there was further justification for combining all of the relations hip measurement items together to create the

PAGE 93

93 overall direct and meta-perspective relationship measures (“overall relationship”). Since there was still some interest in the differences between the various dimensions themselves for the purpose of relationship diagnos is, two groups of MANOVAs would be conducted, one that utilized the four relationship dimensions and one that used just the direct and meta-perspective overall relationship measures as dependent variables. To assess the estimation and fit of the MANOV A models, Pillai’s Trace was selected as a test statistic since it is robust when violations of some of the assumptions of MANOVA are made and is recommended when there are unequal cell sizes; however, it still maintains adequate statistical power (Hair et al., 1998). Whereas this test statisti c assesses the overall MANOVA model, it is still necessary to assess the diffe rences between groups durin g post hoc analysis. For that purpose, the Scheffe method was used as it is the most conservati ve (Hair et al., 1998). A repeated measures MANOVA was calcula ted comparing the effects of party membership (organization, public), the within-sub jects repeated measure of perspective (direct perspective, meta-perspective), and whether the respondents received the e-mail from the administration regarding the sc hool shooting (yes, no) on the relationship ratings of control mutuality, trust, commitment, and satisfaction. The main effect of receiving the e-mail ( F (4,523)=.388, p > .05), the party membership by e-mail interaction ( F (4,523)=.799, p > .05), the perspective by party member ship by e-mail interaction ( F (4, 523)=.398, p > .05) and the perspective by e-mail interaction ( F (4, 523)=.339, p > .05) were all not significant. The perspective by party membership interaction was significant ( F (4,523)=18.71, p < .01), as were the main effects for perspective ( F (4,523)=4.75, p < .01) and party membership ( F (4, 523)=7.22, p < .01).

PAGE 94

94 The between-subject effects of party membership on control ( F (1,526)=9.737, p < .01), trust ( F (1,526)=15.35, p < .01), commitment ( F (1,526)=13.25, p < .01), and satisfaction ( F (1,526)=21.71, p < .01) were all significant. The public rated all four relationship dimensions significantly lower than the organization. The within-subject effects of pe rspective on control mutuality ( F (1, 526)=1.46, p > .05), trust ( F (1, 526)=.180, p > .05), and commitment ( F (1, 526)=1.59, p > .05) were not significant, while the effect of perspectiv e on satisfaction was significant ( F (1, 526)=4.93, p < .05). The direct perspective evaluation of satisfaction wa s significantly lower than the meta-perspective evaluation of satisfaction. Within-subject effects of the interaction of perspective by party membership on control mutuality ( F (1,526)=55.08, p < .01), trust ( F (1,526)=63.45, p < .01), commitment ( F (1,526)=33.06, p < .01), and satisfaction ( F (1,526)=23.67, p < .01) were all significant. The meta-perspectives and direct perspectives of the public and th e organization differed significantly on all four re lationship dimensions. Based on the results of these significance tests, the factor related to whether or not the respondents had received an e-mail from the admi nistration regarding the Virginia Tech shooting was removed from further analysis so as not to include too many non-si gnificant factors in the MANOVA (Hair et al., 1998). A repeated measures MANOVA was calculated with party membership (organization or public) as the between-subjects factor and perspectiv e (direct or meta-perspective) as the withinsubjects factor to see their effect on the responde nt ratings of the relationship dimensions of control mutuality, trust, commitment, and satisf action. The main effects of party membership

PAGE 95

95 ( F (4, 525)=9.49, p < .01), the main effect of perspective ( F (4, 525)=5.41, p < .01), and the interaction of party memb ership x perspective ( F (4, 525)=22.75, p < .01) were all significant. The between-subjects main effect of party membership on control mutuality ( F (1, 528)=13.73, p < .01), trust ( F (1, 528)=20.20, p < .01), commitment ( F (1, 528)=17.85, p < .01), and satisfaction ( F (1, 528)=29.55, p < .01) were all significant. The organization rated all four relationship dimensions higher than the public did. The within-subjects main effects of perspective on control mutuality ( F (1, 528)=2.39, p > .05), trust ( F (1, 528)=.117, p > .05), and commitment ( F (1, 528)=1.82, p > .05) were not significant. There was a significant main effect of perspective on satisfaction ( F (1, 528)=5.70, p < .05). The direct perspective ra ting of satisfaction was lower than the meta-perspective rating. The within-subjects effects of the interaction of perspective and party on control mutuality ( F (1, 528)=68.79, p < .01), trust ( F (1, 528)=77.75, p < .01), commitment ( F (1, 528)=42.85, p < .01), and satisfaction ( F (1, 528)=29.98, p < .01) were all signifi cant. The directand metaperspectives of the organization and the direct and meta-perspectives of the public differed significantly on the four relationship dimensions. A repeated measures ANOVA was conducted with party membership (organization, public) as the between-subjects factor and perspe ctive as the within-subjects factor (direct perspective, meta-perspective) to see their effect on respondents’ ratings of the overall relationship (i.e., using the overa ll relationship measure that in cluded all 20 items). Here, the main effect of party membership ( F (1, 528)=22.69, p < .01) and the inter action of perspective with party membership ( F (1, 528)=62.29, p < .01) were both significan t. The main effect of perspective ( F (1, 528)=1.51, p > .05) was not significant. The public rated the overall

PAGE 96

96 relationship lower than the organization did. Furthermore, direct perspectives and metaperspectives of the relations hip differed among the parties. To untangle the differences between groups th at resulted from the within-subjects factor of perspective interacting with the between-sub ject factor of party membership, additional MANOVA and ANOVA tests needed to be conducted in such a manner so that the interaction of perspectives and parties and the resulting eff ect on ratings of the individual relationship dimensions and the overall relati onship rating could be clarified through post hoc testing of the individual group means. To facilitate such an analys is, the data set had to be restructured so the direct perspectives and meta-perspectives of eac h party in the relationship would be separated into four groups. Whereas the preceding analysis treated the direct and meta-perspectives evaluation of each dependent variable as repeated measures, the following analysis treats them as independent measures. Using SPSS, the data set was restructured to create a new independent variable representing the perspec tive by party interaction. The ne w variable, “perspective,” could now be treated as a factor with four levels: (a ) public’s direct perspe ctive, (b) public’s metaperspective, (c) organization’s direct perspective, and (d) or ganization’s meta-perspective. A MANOVA was conducted to see the effects of the new perspective variable on the four relationship dimensions of control, trust, comm itment, and satisfaction. The main effect of perspective ( F (12,3837)=22.131, p < .01) was significant. Follow-up univariate ANOVAs indicated that control mutuality ( F (3, 1280)=73.99, p < .01), trust ( F (3, 1280)=73.99, p < .01), commitment ( F (3, 1280)=73.99, p < .01), and satisfaction ( F (3, 1280)=73.99, p < .01) differed significantly between the four pers pectives. Post hoc analysis usi ng Scheffe’s test indicated that there were significant differences between the pu blic’s direct perspec tive, the public’s metaperspective, and the organization’s direct perspe ctive in regards to c ontrol mutuality, trust,

PAGE 97

97 commitment, and satisfaction. The organization’ s meta-perspective was significantly different from the public’s direct pers pective for all four relations hip dimensions, and was only significantly different from the organization’s dire ct perspective in regards to trust. Table 4-1 clarifies the significant differences between the four perspectives. A one-way ANOVA was conducted to compare the overall relationshi p rating within the four different perspectives. A significant difference was found among the perspectives ( F (3,1280)=62.88, p < .01). Scheffe’s test was used for pos t hoc analysis to identify differences between the individual group perspectiv es. The public’s dire ct perspective ( M =3.96, SD =1.28), the public’s meta-perspective ( M =4.83, SD =1.36), and the organization’s direct perspective ( M =5.57, SD =.658) were all significantly different from each other. The organization’s metaperspective of the overall relationship ( M =4.96, SD =.990) was only significantly different from the public’s direct perspective of the overall relationship. Figure 4-1 presents a model of the organization-public relationship that helps to clarify the interplay between these different perspectives and indicates the pattern of significant differenc es between each perspective’s ratings of the relationship measures. Analyses Related to the Hypotheses Calculation of Additional Variables. Prior to running the multiple linear regression analysis, new dependent and independent variab les needed to be gene rated, including the creation of several dichotomous independent variab les that could be included in the analysis along with the other metric independent variables. First, the dependent measures of coorientation needed to be generated. This included the calculation of d-scor es for each individual student. This was accomplished by utilizing the mean re lationship measures for the university police department direct and meta-perspectives of th e overall relationship and finding the difference between the organization’s overall rating and the student’s individual rating. To obtain the d-

PAGE 98

98 score for agreement, the organization’s mean di rect perspective overall relationship rating was subtracted from the student’s individual direct perspective overall relationship rating. To calculate the d-score for accuracy, the organizati on’s mean direct perspective relationship rating was subtracted from the student’s individual me ta-perspective relationship rating. To calculate the d-score for congruency, the student’s indivi dual direct perspectiv e relationship rating was subtracted from their own meta-perspective relations hip rating. The absolute value of all three dscores was used to remove negative values; reca ll that the smaller the d-score is (in absolute value), the greater the degree of agreement, accuracy, and congruency of the various perspectives. Next, interpersonal interaction valence wa s recoded so that “very negative” was -3, “negative” was -2, and so on, with “very positiv e” being coded as 3. Exposure to interpersonal interaction channels (i.e., whether the respondent had called, emailed, written, visited, or talked to a member of the university police department) was also code d as a dichotomous independent variable, “interpersonal interacti on,” were 0 indicated no exposure to interpersonal interaction and 1 indicated that the respondent had been ex posed to interpersonal interaction with the university police. Exposure to oneand two-way relationship ma intenance strategies were also coded as dichotomous independent variables where 0 indicat ed no exposure and 1 indicated exposure. If a respondent had visited the organization’s Web site read a brochure, read a newsletter, read a press release, or attended a speech, then the on e-way relationship maintenance strategy exposure variable was coded as 1. If the respondent had participated in a ny of the organization’s community relationship or educational programs, then the exposure to two-way relationship maintenance strategies variable was coded as 1.

PAGE 99

99 Finally, the valence of the news story variable was recoded similar to the interpersonal interaction valence variable, where “very negative” was coded as -3, “negative” was coded as -2, and so on. Multiple Regression Analysis. Multiple regression analysis is “a statistical technique that can be used to analyze the relationship between a single dependent (cri terion) variable and several independent (predi ctor) variables” (Hair et al., p. 148). The procedure can also handle non-metric independent variables through the use of dichotomous dummy variable coding, as was done in this analysis (Hair et al., 1998). Thus, one of the advantages of using multiple linear regression to test the hypotheses in my study is that it can acco mmodate both the metric and nonmetric independent variables that are of interest here. Multiple regression also allows the size and direction of the effect of each independent variable on the dependent variables relative to one another to be assessed; th is will allow several of the hypotheses to be addressed. Several considerations need to be kept in mind for the use of multiple regression analysis. One consideration is sample size; it is recommen ded that 20 observations per dependent variable be included in the analysis. When employing stepwise regression, which is used here, the recommendation is that 50 cases per dependent variable be used (Hair et al., 1998). Since this analysis examines five dependent variables, a minimum sample of 250 observations is needed; this is one reason why this analysis focuses on the student sample only. Another issue is multicollinearity of the indepe ndent variables. When using multiple linear regression, it is desirable th at the dependent variables are correlated with the independent variables, but that the independent variables are not correlated with each other (Hair et al., 1998). While including additional individual variab les can increase the overall amount of variance explained by the model (i.e., the coefficient of determination), as multicollinearity of the

PAGE 100

100 independent variables increases, the unique variance in the depende nt variable explained by an individual independent variable is reduced. A co rrelation matrix of all independent variables was generated to investigate the presence of multicol linearity among the independent variables. None of the correlations were above .90, which is sugges ted as an indicator of hi gh collinearity (Hair et al., 1998). In fact, the highest correlation of one independent variable to another independent variable was for interpersonal interac tion valence and news story valence ( r (472) = .334, p < .01) (See Table 4-2 for a correlation matrix of the independent and dependent variables included in the analysis). Another way of assessing multicollin earity is to look at the tolerance values and variance inflation factor (VIF) values gene rated by SPSS when conducting multiple regression analyses. Small tolerance values (<.10) or larg e VIF values (> 10) indicate the presence of collinearity. These values were examined for all the multiple linear regression models generated in this analysis; in no instance did they indicate that collinearity was present. Prior to running the multiple regression analyses examining the impact of the independent variables on each dependent variab le, the variable selection method had to be picked. Although there are a vari ety of sequential search tech niques that can be used, the decision was made to use stepwise regression as the method for entering independent variables into the model since the study is exploratory and there was no theory that suggested an order for entering the variables into the model. One of th e primary drawbacks of using stepwise regression is that it is sensitive to multicollinearity, but as indicated by the previous analysis of the independent variables, this did not appear to be a problem. A series of multiple linear regression e quations were calculated using the following dependent variables: (a) direct pe rspective of the overall relations hip, (b) meta-perspective of the overall relationship, (c) agreement between subj ects’ direct perspective and the organization’s

PAGE 101

101 meta-perspective, (d) accuracy of the subjects’ meta-perspective in estimating the organization’s direct perspective, and (e) the congruency of the subjects’ dir ect perspective with their own meta-perspective. For each equation calculated, stepwise regression was used to enter the following independent variables: (a ) years in the relationship, (b) issued a citation, (c) one-way relationship maintenance strategy exposure, (d ) two-way maintenance strategy exposure, (e) interpersonal interaction, (f) interpersonal interaction valence, and (g) news story valence. The stepwise multiple linear regression looking at the effects of the independent variables on subjects’ direct perspective of the overall relationship gene rated a significant regression equation ( F (3, 464)=251.99, p < .01), with an adjusted R2 of .617. Interper sonal interaction valence, news story valence, and interpersona l interaction were all significant predictors. Subjects’ predicted direct pers pective rating of the relations hip is equal to 3.98 + .545 (interpersonal inte raction valence) +.255 (news story valenc e) .403 (interpersonal interaction). Subjects’ direct perspective rating of the relati onship increased .545 for each one unit increase in interpersonal interaction valence and increase d .255 for each one unit increase in news story valence. Subjects who had inte rpersonal interaction with members of the university police department rated the relationship .403 lower than those who hadn’t had interpersonal interaction with members of the university police department. Years in the relationship, issued a citation, one-way maintenance strategy exposure, and two-way maintenance strategy exposure were excluded from the model and were not significant pr edictors. Plots of the regression standardized residuals using a histogram, the normal P-P plot, and a scatter plot were examined and revealed no violations of the assumptions for multiple regression analysis. The stepwise multiple linear regression examining the effects of the independent variables on subjects’ meta-perspective of the overall re lationship generated a significant regression

PAGE 102

102 equation ( F (3, 464)=58.69, p < .01), with an adjusted R2 of .270. Interper sonal interaction valence, news story valence, and years in th e relationship were all significant predictors. Subjects’ predicted meta-perspective rating of the relationship is equal to 4.75 + .351 (interpersonal inte raction valence) +.249 (news story valenc e) .088 (years in the relationship). Subjects’ meta-perspective rating of the relati onship increased .351 for each one unit increase in interpersonal interaction valence, increased .270 for each one unit in crease in news story valence, and decreased .088 for each year they had been in the relationship. Issued a citation, one-way maintenance strategy exposure, two-way mainte nance strategy exposure, and interpersonal interaction were excluded from the model and were not significant predictors. Plots of the regression standardized residuals using a histogram, the normal P-P plot, and a scatter plot were examined and revealed no violations of the assumptions for multiple regression analysis with the exception of some slight heteroscedasticity. The stepwise multiple linear regression looking at the effects of the independent variables on agreement between subjects’ direct persp ective of the overall relationship with the organization’s direct perspective of the overa ll relationship generated a significant regression equation ( F (3, 464)=218.19, p < .01), with an adjusted R2 of .583. Interper sonal interaction valence, news story valence, and interpersona l interaction were all significant predictors. Subjects’ predicted agreement d-score for the ov erall relationship is equal to 1.61 .496 (interpersonal intera ction valence) .172 (news story valenc e) + .414 (interpersonal interaction). Subjects’ agreement d-score regarding the relati onship decreased .496 for each one unit increase in interpersonal interaction valence and decrea sed .172 for each one unit increase in news story valence. Subjects who had inte rpersonal interaction with members of the university police department had agreement d-scores that we re .414 higher than those who hadn’t had

PAGE 103

103 interpersonal interaction with members of the university police department. Keep in mind that for d-scores, smaller scores indicate greater agreement; therefore, a negative regression coefficient indicates that the independent variable actually had a “positive” effect and vice versa. Years in the relationship, having been issued a citation, one-way maintenance strategy exposure, and two-way maintenance strategy exposure we re excluded from the model and were not significant predictors. Plots of the regression standard ized residuals usin g a histogram, the normal P-P plot, and a scatter plot were examined and revealed no violations of the assumptions for multiple regression analysis. The stepwise multiple linear regression looking at the effects of the independent variables on the accuracy of subjects’ meta-perspective of the overall relationship generated a significant regression equation ( F (3, 464)=54.82, p < .01), with an adjusted R2 of .257. Interpersonal interaction valence, news story valence, and interpersonal interaction were all significant predictors. Subjects’ predicted accuracy d-score for the overall re lationship is equal to 1.08 .292 (interpersonal interaction va lence) .136 (news story vale nce) + .254 (interpersonal interaction). Subjects’ accuracy d-score regard ing the relationship decreased .292 for each one unit increase in interpersonal interaction valen ce and decreased .136 for each one unit increase in news story valence. Subjects who had interperso nal interaction with members of the university police department had accuracy d-scores that were .254 higher than those who had not had interpersonal interaction with members of th e university police department. Years in the relationship, having been issued a citation, one -way maintenance strategy exposure, and two-way maintenance strategy exposure were excluded from the model and were not significant predictors. Plots of the regressi on standardized residua ls using a histogram, the normal P-P plot, and a scatter plot were examined and revealed no violations of the assumptions for multiple

PAGE 104

104 regression analysis with the exception of some s light heteroscedasticity as with the multiple linear regression looking at meta-perspective. A stepwise multiple linear regression was calculated to examine the effects of the independent variables on the congruency of s ubjects’ direct perspective of the overall relationship with their own meta-perspective of the overall relationship. A significant regression equation was generated ( F (1,466)=52.63, p < .01), with an adjusted R2 of .10. Only interpersonal valence was found to be a significant predictor; all other independent variables were excluded from the model. Subjects’ predicted congruency d-score for the overall relationship is 1.05 .229 (interpersonal intera ction valence). Subjects’ congrue ncy d-score regarding the overall relationship decreased .229 for each one unit increase in interpers onal interaction valence. Plots of the regression standardized residuals using a histogram, the normal P-P plot, and a scatter plot were examined. The histogram of the standardized regression residuals showed some very slight positive skew. The P-P is close to normal. Examina tion of the scatterplot of the residuals against the predicted dependent values for congruency showed some evidence of heteroscedasticity. It should be noted that simple linear regressi ons were also computed to confirm the results of the stepwise linear regression s. Independent variables that were not significant in the stepwise models were also not significant wh en using simple linear regression. In all of the models, exposure to one-way re lationship maintenance strategies and exposure to two-way symmetrical relationship maintenance strategies were not retained as significant predictors of any of the dependent variables m easuring relationship perc eption or coorientation of the subjects’ perspectives. A series of multiple linear regressions were calculated using specified models in which only the two relationshi p maintenance strategy variables were entered as independent variables – none of the models assessing the effect of these variables on the

PAGE 105

105 dependent variables were significant. Also, the correlation matrix between the independent and dependent variables (see Table 4-2) showed that neither exposure to one-way strategies nor exposure to two-way strategies were significantl y correlated with any of the relationship or coorientation measures. A partia l correlation was conducted to examine the relationships among exposure to one-way strategies, exposure to tw o-way strategies, direct perspective of the relationship, meta-perspective of the relationship, agreement, accuracy, and congruency while controlling for interpersonal interaction, interp ersonal interaction vale nce, and news story valence. Exposure to one-way strategies and exposure to two-way strategies were not significantly correlated with any of the relationship measures in the partial correlation model. Finally, a MANOVA was calculated to see the effects of expos ure to one-way strategies and exposure to two-way strate gies on the direct perspectiv e and meta-perspective of the relationship, as well as their effect on the th ree coorientation measures The main effects of exposure to two-way strategies ( F (5, 464)=.836, p > .05) and the interac tion of exposure to oneway and exposure to two-way strategies ( F (5, 464)=.832, p > .05) were not significant. The main effect of exposure to one-way strategies ( F (5, 464)=2.29, p < .05) was significant. Looking at the univariate tests, the effect of one-way strategies on direct pe rspective of the relationship ( F (1, 468)=.948, p > .05), meta-perspective of the relationship ( F (1, 468)=2.15, p > .05), agreement ( F (1, 468)=.258, p > .05), and congruency ( F (1, 468)=.181, p > .05) were not significant. The effect of exposure to one-way strategies on accuracy ( F (1, 468)=4.360, p < .05) was significant. Students who had been exposed to one-way strate gies demonstrated a greater degree of accuracy between their meta-perspective and the university police department’s direct perspective.

PAGE 106

106 Table 4-1. Main study organization a nd public direct and meta-perspectives. Organization direct perspective Organization meta perspective Public direct perspective Public meta perspective Relationship dimension M SD M SD M SD M SD Control mutuality 5.37ae .722 4.61c 1.16 3.72acd 1.32 4.80de 1.48 Trust 6.15abe .595 5.23bc 1.04 4.43acd 1.29 5.28de 1.42 Commitment 5.35ae .888 4.87c 1.09 3.91acd 1.37 4.70de 1.36 Satisfaction 5.27ae .860 5.00c 0 .977 3.65acd 1.46 4.45de 1.48 Overall relationship 5.57ae .658 4.96c 0 .990 3.96acd 1.28 4.83de 1.36 Note : Values with the same superscripts indicate m eans that there are significantly different on a given relationship measure ( p < .05).

PAGE 107

107 Table 4-2. Correlation matrix of i ndependent and dependent variables. 1 2 3 4 5 6 7 8 9 10 11 1. Time 2. Cited .068 3. One-way strategies -.034 .009 4. Two-way strategies -.043 -.009 .162** 5. Interpersonal interaction .105* .226** .145** .114* 6. Interpersonal valence .018 -.118* -.093* .160** -.002 7. News valence .042 -.079 .134** .027 .032 .334** 8. Relationship (direct) .011 -.162** .044 .058 -.129** .743** .458** 9. Relationship (meta) -.066 -.116* .048 .064 -.070 .475** .353** .579** 10. Agreement -.029 .177** -.021 -.053 .150** -.731** -.400* -.962** -.542** 11. Accuracy .042 .152** -.077 -.062 .099* -.480** -.295** -.519** -.856** .558** 12. Congruency -.082 .076 -.013 -.018 .081 -.319** -.114* -.477** .402** .472** -.308** p < .05 ** p < .01

PAGE 108

108 108 Figure 4-1. Coorientation model of main study organization-public relationship. Note : Values with the same superscripts indicate m eans that there are significantly different on a given relationship measure ( p < .05). In the model, CM = c ontrol mutuality, T = trust, C = commitment, S = satisfaction, and R = overall relationship.

PAGE 109

109 CHAPTER 5 DISCUSSION This chapter presents an interpretation of the findings from the main study and a discussion of the implications of these findings for public relations and organiza tion-public relationship research. First, the integration of the relationshi p measures into the coorientational framework is discussed from a methodological and measurement pe rspective. This is followed by a discussion of the results of the hypothesis tests and the eff ectiveness of the various independent variables on the relationship and coorientation measures. Next, the theoretical and pract ical implications of the study are discussed. The chapter concludes w ith an acknowledgement of the limitations of the study and suggestions for future research regarding organization-public relationships. Evaluation of the Coorientational Approach The first research question asked if the c oorientational framework could be used in conjunction with the existing organization-publ ic relationship measures to measure an organization-public relationship. The answer appears to be yes; it was fairly simple to integrate the Hon-Grunig (1999) relationship m easures into the coorientational approach that has long been advocated by Broom (1977), Broom & Dozi er (1990), and others (Hon & Grunig, 1999). Despite the fact that the relationship measur ement scales were now being used with an organization (instead of just a public) to measure di rect perspectives and were also being used to assess meta-perspectives, they still maintained good reliability and provided adequate measures of how the parties perceived the relationship. Even though the coorientation framework was originally suggested to measure organization and public attitudes a bout an issue, it appears to provide an ideal framework in which to examin e how these parties perceive the organizationpublic relationship as well. From a methodologi cal perspective, the scales were easy to administer and despite the fact that including the ratings of the meta-perspective in essence

PAGE 110

110 doubles the length of the relationship measurem ent instrument, over 70% of the respondents managed to complete the assessments of both direct perspective and meta-perspective. The added complexity of requiring two samples (one from the organization and one from a stakeholder public) may be trouble some in practical applications but can be achieved if the organization in question is cooperative and th ere is support for applying coorientational measurement within the organi zation’s dominant coalition. The use of the coorientational approach not only made it possible to move beyond simple measurement of the public’s direct perspective to measurement of the organization’s direct perspective, but also facilitates measurement of the meta-perspectives that are necessary to understand the true state of the re lationship that exists between an organiza tion and a stakeholder public. In my study, having both th e direct perspective of the st udents and the university police department, as well as their meta-perspectives of each other, not only made it possible to generate individual measures of agreement, accu racy, and congruency between a respondent and the other party in the relationshi p, but it also made it possible to examine the organization-public relationship at the macro-level in order to unders tand what coorientationa l state the relationship was in. This relates to the second research question which asked if the resu lting coorientational measures of accuracy, agreement, and congruency could be used to evaluate the state of the organization-public relationship. Again, the answer is yes; using coorient ational measurement is a marked improvement over the one-way measures of the public’s direct perspective that are typically used because it reveals what the shared perspective of the relationship is between the organization and the public. In my study, the organi zation-public relationshi p that exists between the students and the university police department is more accurately portrayed by using the

PAGE 111

111 coorientational approach than by simply relyin g on the one-way approach that measures public opinion. In this case, the public’s direct perspective of the relationshi p was slightly negative, both overall and in regards to each of the four indi vidual relationship dimensions with the exception of trust, which the public rated slightly favorable From the organization’s direct perspective, the relationship is viewed somewhat positively, both overall and with respect to individual relationship dimensions, especially in regards to trust. Based on th e results of the MANOVA tests, a true state of disagreem ent exists between the direct perspective of the university police department and the direct pers pective of students liv ing on campus, with the university police department viewing the relationshi p more positively than students. Where the coorientational measurement rea lly reveals its utility in diagnosing the organization-public relationship is in the next stage of analysis, where the meta-perspectives of both parties are examined. In this case, the me ta-perspective of the students (i.e., how they thought the university police department views the relationship) was signif icantly different from the students’ own direct perspe ctive, with the students believing that the university police department would rate the relationship higher than students, which was indeed the case. Therefore, there was a perceived disagreement be tween the students’ direct perspective and the students’ meta-perspective. Coupled with the meas ures from the direct perspective, it becomes clear that students are aware that there is a state of disagreement between the two groups. The university police perceive the relationship as being more positive than students and the students recognize this. In fact, th e disagreement is even larger than students believe it to be. By looking at the accuracy dimension of the relationship, the student meta-perspective is significantly different from the organization’s di rect perspective; students know that the university police rate

PAGE 112

112 the relationship more positively than students, they just don’t understand how much more positively the university police department rates the re lationship. To put it another way, there is a gap, and the students correctly s ee the gap, they just underestimate how large the gap is. Taken altogether, from the students’ viewpoint, a state of true disagreement exists between them and the university police department. What becomes even more interesting is the disconnect between the public’s view and the organization’s view of the relationship. The meta -perspective of the university police department (i.e., how they think the students view the relatio nship) is not significantly different from their own direct perspective of the relationship. That is to say, the university police department perceives agreement between themselves and the st udents in how they all view the relationship. This is also reflected in the significant diffe rence between the organization’s meta-perspective and the students’ direct pers pective; the university police department overestimates how favorably students will view the relationship. So, from the university police department perspective, students and police bo th view the relationship favorably; however, there is actually disagreement between them regarding the state of the relationship. From the organization’s viewpoint, a state of false consensus exists. So, the students rate the relationship more ne gatively than the orga nization, the students realize that the organization ra tes the relationship more positiv ely (they just underestimate how much more positively), but the organization does not realize this. The organization believes the relationship is healthy and that students would agree; meanwhile, students are well aware of the difference of opinion. Therein lies the benefit of the coorientational approach – the relationship is not defined by one party’s view, but rather rest s on the shared perceptions of all parties in the relationship. To use an interpersonal example, th e situation that exists here would be analogous

PAGE 113

113 to a husband and wife dyad in which one partner believes the marriage is healthy while the other partner thinks it is not and al so is aware that their spouse has no idea that they are in disagreement. This situation could result in on e partner feeling misunde rstood while the other partner, blissfully unaware of the problem, could be in for a rude awakening. It should be noted that these patterns of in teraction are those observed for the particular organization-public relationship in question. Th e samples used in the study represent the members of a unique relationship between a sp ecific university police department and the students living on campus at that university. The direct and meta -perspectives expressed by these individuals resulted in the types of interactions observed here and will not necessarily appear in other types of organiza tion-public relationships. Factors Affecting the Relationsh ip and Coorientation Measures The hypotheses proposed that several independ ent variables would have an effect on the way respondents evaluated their relationship with the organization and how they believed the organization perceived the relations hip. Also, it was believed that these factors would have an effect on the degree of accuracy, agreement, and congruency between the respondents and the organization. Each group of hypotheses is discussed in turn. Hypothesis 1a proposed that longer time in th e organization-public re lationship would lead to more favorable ratings of the organization-pub lic relationship dimensions, while hypothesis 1b proposed that longer time in the organization-pub lic relationship would le ad to a greater degree of accuracy between the meta-perspectives of one party in the organization-public relationship and the direct perspective of the other party. Based on the multiple regression models, time in the relationship was not a significant fa ctor in predicting direct pers pectives of the relationship, agreement, accuracy, or congruency. It did ha ve a significant negative effect on the metaperspective of the respondents. Th e longer that the student had been in the relationship, the more

PAGE 114

114 negatively they believed that members of th e university police depart ment would rate the relationship. However, the weak regression coeffici ent suggests that the practical effect of time in the relationship is minor. Neither hypothesis wa s supported. It should be noted, however, that most of the respondents were firstand second-year students at the university; if more thirdand fourth-year students had been included, time ma y have proved to be a significant factor. Hypothesis 2a proposed that exposure to rela tionship maintenance strategies would lead to more favorable ratings of the organization-pub lic relationship dimensions, while hypothesis 2b proposed that these strategies would lead to a greater degree of agreement, accuracy, and congruence between the perspectives of the parties in the organi zation-public relationship. Based on the multiple regression models, neither one-way nor two-way relationship maintenance strategies were significant factors in predicting any of the relationship or coorientation measures. Additionally, in the MANOVA model, the main effects for both of these factors were not significant; students who had been exposed to one-way or two-way strategies did not rate the relationship significantly different than those who had not been e xposed to those strategies, nor did students exposed to one or two-way strategi es exhibit a greater degree of agreement or congruency. As for accuracy, two-way maintena nce strategy exposure had no effect; however, one-way strategy exposure did have an effect – students who had been exposed to one-way strategies exhibited a greater degree of accur acy in estimating the organization’s direct perspective. These findings also address hypot hesis 2c and 2d which proposed that two-way maintenance strategies would be more effective th an one-way strategies in regards to having a positive effect on the relationship and coorienta tion measures. Again, it would appear that neither was a significant factor, with the exceptio n of the positive effect on accuracy generated by one-way strategies. In summary, none of these hypotheses were supported.

PAGE 115

115 This outcome is surprising given the litera ture that suggests that effectively managed, two-way symmetrical communicat ion should lead to mutual understanding and more positive relationships between the organization and its stakeholders; however, this is not the case in the organization-public relationship in my study. The lack of effectiveness of two-way strategies could be explained as a result of either a methodological problem relating to the operationalization of the relationshi p maintenance strategy measures or it could be a result of the way that these purportedly twoway strategies are actually im plemented in practice. The operationalization issue will be addressed during th e discussion of the study’s limitations, so for now the focus will be on the implementation issue. The community services unit w ithin the university police depa rtment is responsible for many of the two-way relationship maintenance strategies that the department utilizes. Its mission is to facilitate the relationship with the co mmunity the police department serves. The unit organizes, maintains, and promotes crime prevention and community-oriented programs to achieve this goal. The question then becomes wh ether these programs actually meet this goal in their execution. For example, one of the twoway strategies, SCOPE, focuses on developing a crime prevention partnership between students, housing officials, and the university police department and as it is conceived, displays many of the features of a two-way maintenance strategy. However, in its implementation, it could be doing more to provide students with information about crime and safety issues rath er than actually focusing on relationship building between the students and the uni versity police department. Meanwhile, the one-way relationship maintenance strategies (e.g., using brochures, newsletters, etc. to communicate with stake holders) also focus on providing information on crime prevention and safety, not information n ecessarily about the orga nization’s relationship

PAGE 116

116 with students. The tactics that implement one-way relationship maintenance strategies seem to provide information to students that they can use in correctly aligning their meta-perspective with the organizational perspec tive; these tools help students understand how the organization views their relationship, and also may explain on e reason why they realize that the university police department perceives the relationship more favorably than students. However, these oneway tactics do nothing to provide feedback to th e organization about the students’ more negative views of the relationship. Additionally, they do not improve the relationship so as to bring the student direct perspective in alignment with the organization’s direct perspective. The ineffectiveness of two-way relationship bu ilding strategies is reflected at the macrolevel when one considers the lack of accuracy between the university police department’s metaperspective with the students’ direct perspec tive and vice versa. Also, based on the congruency between their direct and meta-perspectives, the university police depart ment does not see that there is disagreement about the na ture of the relationship between them and the students, perhaps due in part to the supposed two-way symmetrical relationship building strategies being executed in a one-way asymmetrical manner. Thus, while a two-way maintenance strategy might be geared toward improving a relationship, both in its operationalization from a theoretical perspective and in its planning fr om a strategic management perspective, the way these strategies are implemented may not in fact be two-way symmetrical at all. This then becomes an operationalization problem for rela tionship theory researchers and an implementation issue for practitioners. Hypothesis 3a proposed that posit ive interpersonal interaction with the other party in the organization-public relationship wo uld lead to favorable rating s of the organization-public relationship dimensions while ne gative interpersonal interacti on would lead to unfavorable

PAGE 117

117 ratings of the organization-public relationship dimensions. Similarly, hypothesis 3b suggested that positive interpersonal interaction would lead to a greater degree of accuracy, agreement, and congruency while negative interper sonal interaction woul d lead to a lower degree of agreement, accuracy, and congruency. Based on the multiple regression models, both interpersonal contact between students and members of the university police department as well as the positive or negative valence of the contact we re significant predictors of es timates of respondents’ direct and meta-perspectives of the relationship, agre ement, and accuracy, although the variables had different types of effects. Exposure to interpersonal inte raction with members of the university police department had a negative effect on students’ direct perspect ive, agreement, and accuracy. This suggests that when students have direct experience with memb ers of the university police department, it leads them to perceive the relationship with the university police more negatively. This in turn leads to lower levels of agreement with the university police department perspective which feels that the relationship with students is favorable. Furthermore, the interpersonal interaction could also lead students to believe that the univers ity police evaluate the relationshi p more negatively as well. It could be that the nature of thes e interactions that bring students and university police into contact is frequently one with negative overtones, such as when police respond to a noise complaint or when a student is pulled over for speeding on cam pus. The adversarial nature of these encounters could cause the student to look negatively on th e experience. Meanwhile, the demeanor of the officer within the context of the encounter, when they are “all business” a nd just doing their job, may cause the student to think that the indivi dual member of the university police department also looks negatively on their relationship with students.

PAGE 118

118 However, the valence of interpersonal interaction between students and the university police had a positive relationship with all the rela tionship and coorientation measures. Increased positive interaction with the university police led students to report increasingly positive evaluations of their direct pe rspective, their meta-perspec tive, agreement, accuracy, and congruency. The way a student felt about the di rect experience with the university police influenced not only how the student perceived the relationship, but also how the student felt that the university police perceived the relationship. If it was a positive experience, then this improved not only the student’s di rect perspective, but also the student’s meta-perspective, which in turn increased agreement, accuracy, and perceived accuracy as well since more positive student direct and meta-perspectives were now being brought into line with the generally favorable view of the relationship held by the university police department. Overall, these findings support hypothesis 3a and 3b. Although no formal hypothesis had been proposed regarding media exposure, it should be noted that the perceived valence of news stories about the univer sity police department was also a significant factor predicting respondents’ estimates of the relationship measures and the coorientation measures, with the exception of congruency. The more positively students perceived news coverage of the university police department to be the more favorable were their direct and meta-perspectives of the relationship. Additionally, the more positive news coverage was perceived, the higher was the students’ degree of agreement and accuracy. Thus, it appears that positive news coverage brought the students’ view of the re lationship into alignment with the favorable organizational view. Also, positive news coverage was apparently better at helping students’ accurately understand how the university police departme nt viewed the relationship.

PAGE 119

119 Theoretical Implications The findings from my study have theoretical im plications for both the efforts to develop adequate measurement of organization-public re lationships and also implications for the relational perspective and rela tionship management theory. Implications for the Measurement of Relationships The findings of my study have several implications releva nt to the ongoing efforts to develop adequate measures of organization-public relationships. First and foremost, my study serves as another example of th e reliability of the Hon-Grunig (1999) relationship measures. The scales have been shown to be a reliable tool for measuring the direct perspective of publics in several situations (Hon & Gruni g, 1999; Huang, 2001; Jo, Hon, & Brunner, 2004), including a university setting (Hon & Brunner, 2002; Ki & Hon, 2007). My study extends the ap plication of the relationship scale by utilizing these measur es to assess not only the public’s direct perspective, but also by using them to measur e the organizational perspective as well as the organizational and public meta-perspectives. In my study, the scales were found to be reliable when used to measure both the direct perspe ctive of the relationship and also the metaperspective of th e relationship. As has been noted previously, measurement of the organization-public relationship by using the coorientational framework to asse ss both the organization’s direct and metaperspective as well as the public’s direct and meta-perspective, provides for a more comprehensive organization-public relationship cons truct. By considering all four perspectives within the organization-public relationship, res earchers attempting to use relationships as a unit of analysis will be getting the whole picture of the relationship, rather than the one-sided view that is provided by only measuring th e direct perspective of the public.

PAGE 120

120 In my study, limiting measurement to the students that make up the organization’s primary stakeholders would only reveal that the students viewed the relationship in a somewhat negative light. This does not adequately describe the st ate of this organizationpublic relationship. To understand the organization-public relationship fully, one also has to consider the organizational view of the relationship; in this case, the university police departme nt viewed it as fairly positive. Even though both party’s direct perspectives are now considered, this still does not fully capture the nature of this relationship because it does not address the degree of understanding between the two groups. Here, the students understand that the university pol ice do not view the relationship the same way as they do; however the university police department fails to understand that the students do not share the organization’s favorable view. If strong relationships are built on mutu al understanding and agreemen t, then the use of the coorientational approach provides a means for act ually measuring this dynamic, which current measurement of organization-public relationships fails to do through only examining one side (and one perspective) of the rela tionship. Thus, measures of th e relationship derived through the coorientational approach could be considered more valid measurements of organization-public relationships valid measures that are necessary for the continued develo pment of the relational perspective. The findings of my study also serve to illustra te why the perceptions of the parties within the organization-public relations hip should not be removed from assessments of relationship states in favor of more objective measures as some have suggested (Broom, Casey, & Ritchey, 1998, 2000). Even though it has been ge nerally accepted that relationships can be measured apart from the perceptions of partie s in the relationship (Ledingham, 2003), these objective measures may not necessarily be better measures of the relationship.

PAGE 121

121 For example, if objective measures were to be used in measuring the university policestudent relationship, one might consider usi ng the number of citati ons issued during the semester, the number of calls to the university police department received in a month, the number of students requesting to use the universit y police’s gun storage service, etc. – concrete measures of interaction and co mmunication. However, an obj ective measure that one would think would color this relationship, such as bei ng issued a citation, did not significantly affect students’ perceptions of the relationship in my study. What really did matter were their perceptions of those interpersonal interactions (as an affect on students’ direct perspective) and the st udents’ and university police’s shared view of the relationship as expressed thr ough their direct and meta-perspect ives of the relationship, as well as whether and to what degree those perspec tives were in agreement or disagreement. In short, perception seems to matter more than th e reality of a situation, especially when considering an intangible construct such as the relationship between two en tities. To again use an interpersonal example, a married couple’s re lationship could be quan tified using concrete, objective measures such as the am ount of time that they spend toge ther in an average week or how often they call one another, but they ma y still secretly loathe each other. For an organization-public relationship, the situa tional antecedent conditions and communication linkages between the two parties may point to a positive relationship due to their frequency and interdependence, but if one or bot h of the parties percei ve the relationship to be negative, then those objective measures would seem to have missed the mark. Implications for Relationship Management Theory The application of the coor ientational methodology for m easuring organization-public relationships that was used in my study and th e findings that this procedure generated have implications for relationship management theory Most importantly, by refining measurement of

PAGE 122

122 the organization-public relationshi p construct, my study helps to develop means for measuring a phenomenon that is the primary unit of analysis for the relational perspective. Additionally, the findings stress the importance of two-way symm etrical communication’s role in relationship management and that two-way symmetrical relati onship maintenance strategies may be distinct from the two-way interpersonal communication that takes place between members of an organization and public. The findings also unders core how managed communication efforts can be undercut by negative interperso nal interactions between members of an organization and the public and that the length of time someone is in the organization-public relationship may not necessarily lead to positive per ceptions of the relationship. La stly, the study represents yet another step in the continued effort to integr ate interpersonal theories into the relational perspective and public relations research. The study contributes to public relations theory development primarily by continuing the extension of knowledge regarding the relational perspective. It further refines the organizationpublic relationship construct by including the coor ientational measures that should be used to measure the organization-public relationship. The continued development of valid, reliable, and descriptive relationship measures is crucial to the advancement of the relational perspective and relationship management theory. If the relationship is indeed going to be the unit of analysis for public relations research and if the relational pers pective is going to be proffered as a dominant paradigm for the discipline, then every effort needs to be made to ensure that the measurement of that “unit” is adequate. Therefore, there is a need to measure the whole relationship that results from the shared perspectives of all the parties involved and not just use one side of that equation to represent the entire construct. The relationa l perspective and relations hip management theory

PAGE 123

123 should be developed around a comprehensive unit of analysis, and using a coorientational approach to measuring organization-public relations hips is a positive step in that direction. The study also provides further justification fo r the relational perspective by stressing that the strategic management of organization-public relationships requires that public relations establishes and maintains effective relationships between organizations and publics, not just establishing and maintaining relationships with publics. In their bounda ry-spanning role, public relations practitioners use managed communication to establish a relationship with a public, but they also have the capacity to look inward and help build a relationship between the public and the organization itself. The use of the two-way symmetrical model of public relations has been repeatedly proposed as the best way to build th ese mutually beneficial relationships; however, the press agentry model has persis ted to be the most frequently used model in public relations (Grunig & Grunig, 1992). It has already been proposed in the preceding discussion that the organization in the university po lice-student relationship, despite having programs that appear to be two-way relationship maintenance strategies that are intended to build and strengthen relationships, may in fact be one -way in their execution, primarily providing information to external publics about issues and events in lieu of focusing on building consensus between members of the organization and st akeholders. This would be consis tent with studies that have identified the public information model as the most prevalent of the four models among state and federal government agencies (e.g., Pollack, 1984; Turk, 1982). This reliance on one-way channels of communication and dependence on a pub lic information model instead of a two-way symmetrical approach may be one reason why the or ganization’s view of the relationship is one of false consensus. If the organization had actua lly been utilizing a tw o-way symmetrical public

PAGE 124

124 relations strategy, then it may have been more in tune with the students’ perspective of the relationship. There is also the distinct possibility that the organization was in fact practicing two-way symmetrical communication, but this two-way stra tegy simply failed to have an effect on the public’s perception of the relatio nship. Although the two-way symmetrical model is frequently proposed as the ideal way to practice excellent pu blic relations, alternative viewpoints do exist. For instance, the contingency theory of accommo dation proposes that there is a continuum of public relations stances that an organization can adopt ranging from one of pure advocacy to one of pure accommodation. This continuum may be mo re representative of the actual practice of public relations in that it allows for flexibility in how an organizati on responds to an everchanging and unpredictable environment (Cancel Cameron, Sallot, & Mitrook, 1997; Cancel, Mitrook, & Cameron, 1999). Contingency theory pr ovides a set of predisposing and situational factors that influence where on the continuum th e organization places itself in response to a given public within the context of a given situation. The theo ry suggests that the two-way symmetrical model may not be the best way (or ev en the most ethical or legal way) to practice public relations in every situation. In the case of the university police department, it is possible that situational and pr edisposing factors related to the sp ecific dynamics operating within the organization-public relationship may require the use of a strategy other than two-way symmetrical communication to have a positive effect on the relationship. The lack of effectiveness of the organizati on’s one-way and two-way strategies is even more striking in light of the significant affect that interpersonal communication had on perceptions of the re lationship. While exposure to the one-way and two-way maintenance strategies (acknowledging that the two-way strategies may su ffer from operationalization and/or

PAGE 125

125 actual implementation issues) did little to influen ce how students rated their relationship with the university police, the students who indicated that they had negative interpersonal interactions with the university police had more negative perceptions of th e relationship. This supports relationship management’s conten tion that healthy organization-publ ic relationships can not be maintained without supportive be haviors (Ledingham, 2003). It also helps to draw a distinction between the types of two-way co mmunication that takes place be tween representatives of an organization who interact with a public with in the context of a communication exchange involving managed communication, such as the SC OPE program in my study, and the type of two-way communication that takes place between a representative of the organization and the public that occurs outside of t hose contexts in unmanaged situa tions, such as when a university police officer chastises a student who jaywalks on campus. This would seem to dispute the idea that interpersonal communication is analogous to two-way symmetrical public relations strategies (Stafford & Canary, 1991). Also, due to the relative influence of interpersonal communication and lack of influence of the re lationship maintenance strategies, the findings reinforce the importance for those studying public relations, advertising, marketing, and other forms of managed communication to continue to be mindful of the role that interpersonal communication plays within the mass communication world. Research studies seeking to evaluate the effectiveness of managed communication should be sure to consider how personal and interpersonal experi ence interacts with mediated communication. Another finding that has implications for the relational perspective is that the length of time that respondents had been in the organizati on-public relationship had little effect on their perceptions of the relationship and the degree of accuracy, agreement, or congruency that those perceptions had with the organizational perspective. There seems to be this assumption in the

PAGE 126

126 relationship management literatur e that longer time in the rela tionship will lead to favorable outcomes, such as a greater degree of understand ing, agreement, and congruency. However, the findings from my study indicate that time may not always lead to positive relationships; more time in the relationship does not seem to have an appreciable affect on the way the public views its relationship with the organization. In fact, it is quite easy to conceive of the possibility that more time in the relationship could lead to increas ingly negative perceptions of the relationship if longer time in the relationship increases the likel ihood of having a negative experience with the organization. In the student-university police rela tionship, the longer an in dividual is a student at the university, the more likely they have had an en counter with the university police – this in turn may provide more instances to expose the indi vidual to negative information about the organization from direct experi ence, news coverage, word-of-m outh, etc. Additionally, in a relationship such as the one that exists be tween students and university police where the organization is unaware that the public has a nega tive perception of the rela tionship, if the public sees that this is the case and that things do not change over time, it may lead to feelings of resentment by the public that they are not only being understood by the organization, but that the organization has continued to turn a blind eye to the situation. The use of the coorientational approach to measure relationships also helps to advance relationship management and public relations th eory in general thr ough the integration of theories from other disciplines – one of the bene fits that Ferguson (1984) suggested the relational perspective would provide. The use of a coorientational approach not only serves to improve the relationship construct as previously discussed, but also provides a ve hicle for the continued integration of interpersonal theories and mode ls into the relational perspective, including coorientation, theories of balance and consistenc y, and interpersonal perc eption. The integration

PAGE 127

127 of coorientation and relationship measurement to measure an organization-public relationship continues the evolution of a pr ocess whereby interpersonal comm unication theories originally designed to describe one-on-one relationships are being applied to group-to-group relationships. This process started with theories of consistenc y such as Heider’s bala nce theory (1946, 1958), which looked at how one person organized his or he r internal attitudes to ward people and objects in relation to each other within the individua l’s own cognitive structure. Then Newcomb’s (1953) symmetry theory included a second indi vidual and focused on how two people orient toward each other while consider ing their attitudes toward another object. These concepts were then developed for group to group applications by McLeod and Chaffee (1973) and Grunig and Stamm (1973). Next, Broom (1977), Broom and Do zier (1990), and others applied this framework to organization-public interactions in considering an issue of mutual concern to the organization and stakeholders. My study takes th is one step further by considering how two groups orient toward each other, not about an issue, but about the actual relationship between them. This reinforces Thomlison’s (2000) acknowledgement about the importance of interpersonal communication theories in public relations research due to the fact that relationships are the common core of both inte rpersonal and public relations research. Implications for Public Relations Practice In addition to its implications for public re lations theory, the findings of my study have several implications for public rela tions practitioners. First, the c oorientational approach utilized here to measure a real-world organization-public relationship can be employed by practitioners as a useful measurement tool that can be used to move beyond the outputs and outcomes that are typically used to assess the effects of public relations programming. Second, the findings stress the importance of practitioners fu lfilling a boundary-spanning role no t just with stakeholders, but between stakeholders and the organization’ s dominant coalition. Additionally, the

PAGE 128

128 coorientational approach suggests the importance of agreement and understanding as goals for public relations practitioners. Finally, the findings illu strate the role that interpersonal interaction plays in organizational communication. Obviously, one of the main implications of my study for practitioners is that it provides a measurement tool that is in many ways an im provement over existing measures of organizationpublic relationships. Since the advantages of using coorientational measurement over measurement of the public’s direct perspective of the relationship alone have been addressed in previous sections, the focus here will be on the benefits of using coorie ntational measurement to replace or complement other more traditional meas ures of public relations effectiveness such as measuring outputs (e.g., number of press releas es distributed) and outcomes (e.g., attitude change). While these measures are indicators of what a particular public relations program produces and what those outputs achieve in the sh ort-term with the spec ific target publics of those programs, the coorientationa l relationship measures can help practitioners identify whether the cumulative effect of those program effort s are achieving long-term public relations goals (building mutually beneficial relationships) and organizationa l goals (creating a favorable environment in which the organization is free to fulfill its mission). The findings also reinforce the importance of practitioners acting in a boundary-spanning role (Cutlip, Center, & Broom, 2000) between the organization and the public, as well as the importance of the practitioner’s role in the tw o-way symmetrical model of public relations in which the practitioner negotiates with both the or ganization’s dominant coalition and the public to establish mutual understanding and benefit. Subscrib ing to one-way models of public relations which use one-way communication (such as public information and press agentry) may fail to achieve the mutual understanding and agreement that characterize healthy relationships by

PAGE 129

129 ignoring the boundary-spanning role. In the student -university relationship, a reliance on oneway communication and possibly ineffective two-wa y maintenance strategies failed to improve the level of agreement and understanding between th e organization and its stakeholders. This is not only because the student perspective of the relationship was somewhat negative, but also because members of the organization have an incorre ct perception of the stude nts’ perspective. If the two-way symmetrical model of public relatio ns was truly being implemented, along with the appropriate two-way relationship maintenance strategies, then the public relations function – by virtue of using two-way managed communication channels – might have made the organization aware of the students’ somewhat negative view of the relationship, and consequently, the organization’s meta-perspective would have demonstrated a greater degree of accuracy in estimating the students’ direct perspective. The alignment and adjustment of the not onl y the public’s but also the organization’s perspectives to foster agreement and understand ing may be a more appropriate goal for some public relations programs. Broom (1977) proposed that this might be one of the benefits of using a coorientational approach in regards to or ganization and public pe rceptions of issues, particularly when a lack of understand ing exists between the two parties: More interesting and more susceptible to ch ange through public relations efforts are the states of false consensus and pluralistic ignorance [i.e. fals e dissensus]… When the actual states of agreement-disagreement are inaccurate ly assessed, those involved act on the basis of their misconceptions, or in the context of this discussion, based on what they erroneously believe to be the other group’s definition of the issue (p.111). Broom goes on to state: Since mutual understanding and accuracy are often the necessary conditions for appropriate and meaningful corporate-public in teraction on an issue, changes in these two variables may be the most important goals of public relations. And given the difficulties of changing the deeply held attributes often a ssociated with important issues, changes in mutual understanding and accuracy may be a mo re realistic goal of public relations (p. 117).

PAGE 130

130 In the organization-public relationship in my study, one of the first steps to improving the students’ view of the relationship may be for the organization’s public relations managers to design internal communication vehicles target ed at the members of the university police department in order to adjust the organization’s meta-p erspective on the issue and to let students know that the organization is in tu rn aware of students’ perception of the relationship. To lay this groundwork for improving the relationship, effectiv e two-way strategies that foster open and honest dialogue between the parties in the relationship need to be implemented. This in turn may eventually lead to improvement in the public’s perspective regarding the relationship. Again, standard measures of public relations effectiveness do not assess these types of interactions; the coorientational approach used in my study does, and can be used to suggest strategies for improving the relationship from both sides by using two-way symmetrical communication targeted not only at the public, but also at the organization, as well as building the necessary channels of communication between the two partie s that will lead to increased agreement and understanding. Another implication of the findings for practi tioners is the importance that interpersonal interaction played in shaping the public’s pe rspectives regarding the organization-public relationship. As has been discussed previous ly, while the one-way and two-way relationship maintenance strategies employed by the university police department appeared to have no effect on the students’ perspectives of th e relationship, interpersonal intera ction did have an effect. This reinforces the importance of employees as both target internal audiences for public relations efforts and the role that they play as amba ssadors for the organization. The rank and file members of the organization, not just management, are the fa ce of an organization; the interaction that the public has with members of an organization at the individual, one-on-one,

PAGE 131

131 interpersonal level can undermine the effo rts of professional communicators at the organizational, managed communication level. Th is stresses the importance that relationship management theory places on the need for s upportive behaviors to reinforce the efforts of managed communication programs in building a nd maintaining healthy relationships with stakeholders (Ledingham, 2003). It also suggests th at public relations pract itioners need to look inward to design employee/organizational communication programs that will ensure that members of the organization understand the importa nce of their role in building organizationpublic relationships and provide them w ith the knowledge an d tools to do so. Limitations Like most research studies, the present inve stigation is not without its limitations. The areas of greatest concern regard limitations due to the survey mode, operationalization of the maintenance strategy measures, the difficulties relate d to the measurement of attitudes, the nature of the specific organization-public relationship ex amined, and drawbacks tied to the multivariate analysis. The use of an onlin e survey may have contribute d to unit and item nonresponse problems. The operationalization of the two-wa y relationship maintenance strategies was problematic; it has been touched on briefly in the preceding discussion and is elaborated on further here. Additionally, there are several pr oblems tied to the measurement of attitudes that are common within survey research, such as the potential reporting of “non-attitudes” and differences in respondent response style. Also, th e police department and student samples used in the study represent only one speci fic type of organization-public relationship. Finally, the multivariate techniques used to analyze the data are not the most sophisticated methods available. The study has several limitations due to the selection of the mode that was used to administer the survey. An online survey did provide many advantages, such as affordability,

PAGE 132

132 speed in the recruitment and data collection process, high covera ge of a sample known to have Internet access and for which there was an adequa te sampling frame, and management tools that provided the ability to track data such as wh en the survey was take n and how long it took to complete. However, the mode of survey admini stration may have contributed to a low response rate and missing data issues as a result of ite m-non response. The response rate for the student sample was fairly low, even for an Internet survey, and it may be that students use e-mail so much in the course of conducting th eir academic and social relations hips that they simply do not have the time or patience to respond to every request from the university for information, especially if the request is coming from a resear cher at the university that they have no prior relationship with. Plus, the prevalence of spam phishing, and other junk e-mail may lead many recipients of e-mails, not just students, to dele te unsolicited e-mail, even if it is for legitimate research purposes, clearly identifies its purpose, assures them that it is not an attempt to sell them something, and it has official sponsorship from university organizations. Additional causes of unit nonresponse related to mode could include incorrect or abandoned email addresses; slow connection, incompatible browsers, or other technical issues which made it difficult or impossible to access the survey; and the lack of incentives. Offering incentives may have done much to improve the response rate, but due to th e limited resources available for my study, this was simply not an option and in fact was one of the reasons for choosing an online survey as the data collection method. A final problem that was the result of survey mode (at least in part) was the form that itemmissing data took. If respondents terminated the survey by closing the browser window prior to clicking the “continue to the next page” link at th e bottom of any given “page” (i.e., screen), then any data that had been entered on that page was lost. Therefore, there was no item-missing data

PAGE 133

133 within any completed page, but only item-missing data for the entire pa ge. Whereas any data present on an uncompleted pen-and-paper survey c ould be retrieved, a missing page of the online survey was lost forever and irretr ievable. However, the online survey could still be considered as providing an advantage in that if a respondent decided to term inate the survey early, any completed pages were submitted; if the recipient of a mail survey decides to terminate early, the entire instrument may wind up in th e trash. In the end, the net resu lt of the low response rate and the item-missing data is that the survey result s for the student sample should be interpreted cautiously, especially when attempting to generalize from the sample of students who completed the survey to the larger populati on of all students living on campus. Another limitation to the study which has been alluded to previously in the discussion is the operationalization of the relationship maintenance strategy measures, particularly the twoway strategies. Again, one explanation for th e lack of any effects from two-way strategy exposure could simply be that the organization’ s implementation of the strategy may not have actually reflected the true nature of a two-way relationship maintenance strategy. However, it must be acknowledged that the opera tionalization of the measure itself could be responsible, or it could be a combination of these two factors. This could present a problem for researchers desiring to study the effectiveness of these types of strategies on organizati on-public relationships, and needs to be addressed with as much enthusiasm as has been exhibited for developing the relationship measures themselves. Whereas quantifying the content of tactics that facilitate one-way communication is fa irly straightforward, such as when press releases are content analyzed as part of agenda bu ilding or framing studies, measurement of two-way strategies may prove to be more elusive. How does or should one measure a community relations program? How can this be accomplished in a reliable manner

PAGE 134

134 from one organization-public relationship study to the next? Although relationship maintenance strategies are suggested as im portant factors in moderating pe rceptions of organization-public relationships (Broom, Casey, & Ritchey, 1998; Grunig & Huang, 2000; Hon & Grunig, 1999), they are only described in loose terms. St acks and Watson (2007) suggest some hypothetical models for potentially measuring two-way communi cation, but do not offer any firm conclusions as to the best way to operationalize measures of relational variables for use in quantitative analyses. Furthermore, even if these strategies can be clearly identified and operationalized, if the organization does not implement the two-way strategy in the manner it was designed, it may only have the effect of a one-way strategy. Another limitation of my study is a problem i nherent in much of survey research – the difficulty in attempting to measure attitudes. Seve ral aspects of attempting to measure attitudes can result in measuremen t error wherein the true measure of the respondents’ attitude is not assessed. Among the factors that co uld contribute to these repor ting errors include the use of different estimation processes by different respondents when res ponding to an attitudinal rating scale such as the ones used on the survey in my study. Another problem with attitudinal rating scales is that some respondents utilize an extreme response style in which they consistently use the extreme ends of a multi-ite m attitudinal scale (e.g., always selecting “strongly disagree” or “strongly agree” on a Likert scale). Other re spondents may exhibit an acquiescence response style in that they are predisposed to indicat e agreement on attitudinal scales (e.g., always selecting “strongly agree” on a Likert scale). Yet another problem related to measuremen t of attitudes is the possibility that the respondents really do not have well-formed attitude s at all. There is some debate within the survey research literature over how and if re spondents form and report “non-attitudes.” Converse

PAGE 135

135 (1964, 1970) believed that the general public di d not have well-formed attitudes about many issues as evidenced by the lack of ideological cons traint exhibited in their attitudes; when asked to report their attitudes in rega rds to a particular issue, Conve rse believed that respondents simply made up attitudes, thus attitudinal items really meas ure “non-attitudes.” Zaller and Feldman (1992) believed that respondents hold a number of considerations (orientations or predispositions toward issues), some of them conflicting, at any given moment, and particular considerations can be made more salient through the framing of issues or word choice in a survey. Holbrook, Green, and Krosnick (2003) i ndicated that respondents engaged in satisficing behaviors in order to reduce the amount of effort required to answer a particular question; the nature of the question, the motivation of the respondent, and the difficulty of the task were all factors that could lead to satisfi cing. So, if respondents are not re porting their true attitudes, if they are unable to report their true attitudes, or if they ar e satisficing and only reporting nonattitudes, then the answers derived from the surv ey may not reflect the true attitudes held by the population and therefore the results would lack validity. This c ould certainly be an issue in my study, as it is possible that many students did no t have well defined attitu des or considerations about their relationship with the university police department. Sudde nly confronted with a survey requesting those attitudes, the respondents may have exhibited satisficing behaviors to some degree or simply “made up” and reported attitudes on the spot. An additional problem for my study was that it examined only one specific organizationpublic relationship. The characte ristics of the relationship betw een the police department and university students and the nature of that relationship may have le d to the observed results. Given that the university police department enjoys a large measure of authority and power in comparison to the university students living on ca mpus and that the median age of the student

PAGE 136

136 respondents was below the legal drinking age, it is not surprising that there was the possibility for a contentious relationship between these two groups. Other t ypes of relationships featuring different organizations and publics may not exhib it the same patterns of interactions among the various group perspectives. Also, independent va riables that were significant predictors of relationship outcomes in this study may not prove to be significant in studi es that utilize other types of relationships. For instance, while twoway relationship maintenance strategies did not have a significant effect on ratings of the relationship dimensions in this study, they may prove to be effective in other police-student relations hips or in other forms of organization-public relationships, such as the relationship between a nonprofit organization and the community in which it operates. Other limitations of my study are related to th e types of statistical analyses that were conducted to investigate the re search questions a nd test the hypotheses. While MANOVA and multiple linear regression are certainly well established and widely used statistical tools for conducting multivariate analysis, there are certainly more advanced tools that could have been used. For example, structural equation modeling (SEM), which is a “melding of factor analysis and path analysis into one comprehensive stat istical methodology” (Kaplan, 2000, p. 3), could have been helpful in developing a model of th e complex pattern of interactions among the numerous independent and dependent variables in my study. Another poten tial problem with the statistical analysis is that so me of the assumptions for MANOVA, such as the normality of the dependent variables, may have b een violated; these issues and e fforts to account for them have already been detailed in the results section. One fi nal limitation related to the data analysis is the limited explanatory power of some of the models. The multiple linear regression equations that were calculated for the effect of the independent variables on the depende nt variables of meta-

PAGE 137

137 perspective, accuracy, and congruenc y had fairly low adjusted coefficients of determination and consequently do not explain a lot of the varian ce observed in those dependent measures. Again, this not only means that caution should be take n in interpreting the results related to these dependent variables, but it also suggests that additional independent variables exist that were not included in the models. The need to identify these pot ential variables leads to the next part of this discussion – suggestions for future research. Suggestions for Future Research The need to identify additional variables is bu t one avenue public relations researchers can pursue to extend their understand ing of organization-public re lationships and the relational perspective. Consideration of the uses and gratific ations that individuals seek from relationships could be explored as possible antecedent variable s, as could the influence of opinion leaders and real world cues. New ways of studying relationships need to be considered as well, including a shift from examining individual perceptions of re lationships and to studyi ng the characteristics of the organizations and the publics themselves th at lead to healthier relationships. The investigation of time as a factor in the organiza tion-public relationship need s to be continued, but by taking a long-term approach and by examining the relationship outcomes as a series of effects. New methodologies besides survey res earch, such as qualitative methods, could be utilized as well. Finally, relati onship management theory can bene fit from integration with other mass communications theories, specifically, th e agenda-setting processes and contingency theory. Including additional antecedent conditions would provide a promising starting point. Both Broom, Casey, and Ritchey’s (1998) and Gruni g and Huang’s (2000) models for organizing relationship theory development suggest the pres ence of antecedent states that can affect the organization-public relationship. While they seem to focus more on classifying the types of

PAGE 138

138 possible linkages between organizations and publ ics (e.g., organization affects public, public affects organization, etc.), it may be that individual antecedents pl ay a role as well; therefore, drawing on uses and gratifications theory may prove fruitful. Uses and gratifications theory seeks to understand why individuals choose between various media and what needs are met through their use (Rubin, 1994; Ruggiero, 2000; Severi n & Tankard, 2001). Uses and gratifications departs from traditional media effects theories which tend to focus on what media does to individuals by asking instead what individuals do with media and why certain people gravitate to particular media. In a similar vein, organizatio n-public relationship rese arch could address why certain individuals gravitate to ward particular relationships and develop a typology of motives, needs, and antecedents that drive this process, similar to the typologies developed by uses and gratifications researchers. Another possible variable that could aff ect perceptions of the organization-public relationship is the influence of opinion leaders on individuals and parties within the relationship. Through interpersonal communication, persuasion, or by virtue of their expertise or authority, certain influentials may be able to sway the individual and group perceptions of an organizationpublic relationship in a manner sim ilar to the two-step flow pro cess in which media messages are consumed first by opinion leaders who then pass on these messages to others who look to these opinion leaders for information (Katz, 1957). Th is idea could be extended even further by incorporating the diffusion-innova tion process (Rogers, 1995); here, the spread of new ideas and practices throughout the population is affected by characteristics of the innovation such as its relative advantage over existing ideas or practices, its compatibility with the beliefs and needs of adopters, its complexity, the degree to which it can be tested before adoption, and degree to which the benefits of adoption can be seen by others. The decision to invest oneself in a

PAGE 139

139 relationship and the spread of id eas and perceptions of the rela tionship could be affected by a process similar to the diffusion-innovation process, especially if opinion l eaders are acting as the engine for the diffusion of those ideas. Yet another variable th at could prove to be important is the influence of real world events on perceptions of the relationship. Even though the Virginia Tech shoot ing did not exhibit a significant effect on perceptions of the organization-public relati onship in my study, it does not seem unreasonable to suggest that in other situations with other organizations and publics, a real world cue could have a signifi cant effect on the organizationpublic relationship by increasing the salience of the relationship, certain dimensi ons of the relationship, or various situational antecedents of the relationship, thus making them more relevant to individuals when judging the relationship. The obtrusiveness or unobtrusiveness of the event (i.e., whether the individual has direct experience with the event) could moderate the influence of a real world cue, similar to how issue obtrusiveness moderates the ag enda-setting effect (Gonzenbach, 1996). Moving from possible independent variables th at might influence the organization-public relationship measures when viewed as dependent variables, additional research can focus on using measures of the organiza tion-public relationshi p to understand how perceptions of the relationship affect the outcomes and consequences of relationship state. For example, Ki and Hon’s (2007) study investigating the influence of perceptions of the re lationship measures on resulting attitudes and behavioral intentions toward an organization could be extended by including the coorientational measures as indepe ndent variables. Also, more work needs to be done to tie the perception of relationship meas ures and the coorientation measures to the objective relationship measures that have been advocated by Broom, Casey, and Ritchey (1998, 2000).

PAGE 140

140 Another way the study of organization-public relationships could be extended is by moving from investigations of single orga nization-public relationships in which the observational level is the individu al member of the organization or public and instead looking at multiple organization-public relationships in which observations are made at the organization and public level. By studying multiple organiza tion-public relationships including different types of organizations, different types of public s, and different types of relationships – the characteristics of the organizations and the publics themselves can be examined at the macroorganizational level. In this manner, factors th at vary from organization to organization and public to public can be studied as independent variables and not just what drives individual perceptions of these relationships. Replicating my study with multiple organizatio ns would most likely be easier said than done given the difficulty in securing the cooperation of just one organization to participate in my study, much less recruiting respondents to com pose multiple stakeholder samples. This may suggest the use of other methodol ogies besides surveys to inve stigate the organization public relationship. Focus groups, in-depth interviews, experiments, even field observations could all prove to be informative avenues for exploring the complexities of multiple, unique organizationpublic relationships. Another consideration for future research of organization-public relationships is to conduct more longitudinal studies and to consider the ordering of relationship outcomes. Since the management and cultivation of relationships takes a long-term perspective, organizationpublic relationships need to be studied over a period of years and not just at one point in time. Even though time did not prove to be a factor in influencing percepti ons of the organizationpublic relationship in my study, it could be that students are not involved in a relationship with

PAGE 141

141 the university police department long enough fo r time to have emerged as a factor. A few organization-public relationship studies have begun the invest igation of time’s impact on relationship perception (Ledingham & Bruning, 2000b; Ledingham, Bruning, & Wilson, 1999), but this avenue of research needs to be contin ued; especially if the additional coorientation dimensions are to be considered. Additionally, the manner in which individuals within an organization-public relationship form perceptions of the various relational dimensions should be considered. For example, over ti me, do perceptions of trust deve lop prior to the formation of perceptions of satisfaction? While some studies have begun to investigate the intra-relationship of these relational dimensions th rough assessments of public direct perspectives (e.g., Jo, 2003; Ki & Hon, 2007), this line of inquiry needs to be continued and extended to the coorientational approach by examining the ordering of relati onships outcomes within both organization and public meta-perspectives as well as organizational direct perspectives. Another suggestion for future research aimed at further developing the study of organization-public relationships, and one that could benefit from including the constructs of agreement, congruency, and accuracy between the various direct and meta-perspectives within an organization-public relationship, would be an e ffort to integrate the re lational perspective and relationship management into the agenda-setting processes. Agenda -setting theory asserts that increased media concern and emphasis on certain issues over others will raise the salience of those issues for the public; in other words, incr eased salience of issues on the media agenda will correspond with increased salience of the same issues on the public agenda (Kiousis & McCombs, 2004; Lopez-Escobar, Llamas, & McCo mbs, 1998). The agenda-setting concept has been extended to the agenda-building process in which the media, policy makers, news sources, organizations, and the public collaborate to sh ape the media agenda (Lang & Lang, 1983). Not

PAGE 142

142 only can the salience of issues be influenced, but the salie nce of attributes of those issues can be influenced by actors in the agenda-building / agenda-setting process through framing. Framing focuses on the selection and salience of specific object attributes during both agenda setting and agenda building (Ghanem, 1997; Lancedorfer & Lee, 2003; McCombs, Llamas, Lopez-Escobar, & Rey, 1997; McCombs & Ghanem, 2001; Ohl, Pi ncus, Rimmer, & Harrison, 1995; Tedesco, 2001). The integration of agenda setting, agenda building, and framing into the theory of relationship management could pr ove beneficial to public relati ons researchers and would place added emphasis on the use of coorientational measur ement. Public relations can be considered in some ways the strategic management of frames; this frame management could be used to align the organization and public per ceptions of an issue, which would lead to the mutual understanding necessary to facil itate healthy relationships. Th e agenda-building and agendasetting model could possibly add a frame feedba ck loop in which public relations practitioners read and interpret the salience and framing of issues and then communicate them to the dominant coalition within the organization. By acting as a facilitator of frame alignment, the public relations practitioner would be helping organiza tions and publics understand one another in order to build mutually beneficial relationships. The practitioner could even reframe the public agenda in a manner that would help the dominant co alition understand organi zational stakeholders’ needs and concerns, and thereby serve to move both the organization and the public toward the “win-win” zone proposed by the two-way symmet rical model of public relations practice. In addition to the agenda-setting processes, the integration of relationship management theory with the contingency theory of accommo dation would seem to be a natural fit. As suggested earlier, the lack of ef fectiveness of the two-way rela tionship maintenance strategies

PAGE 143

143 employed by the university police department ma y not be due to poor operationalization or implementation, but could be the result of sele cting the wrong strategy to address the public relations problem at hand. Incorporating the predisposing and situa tional variables from contingency theory as antecedents in the orga nization-public relationship model would provide one way of integrating the two theories. Additio nally, other types of relationship maintenance strategies besides two-way symmetrical strategies should be included in the model to see which are more effective at impr oving the organization-public re lationship under different circumstances. In Closing My study began with the proposition that the cu rrent conceptualizati on of “relationships” and “organization-public relationships” found in the field of public rela tions as well as the methods for measuring and evaluating these constr ucts needed refinement. The definitions and measurement of these concepts need to be furt her developed from both a theoretical standpoint and from a practitioner standpoint. In terms of theoretical use, the lack of an adequate definition of organization-public relationships and the limitations inherent in the methods currently used to measure those relationships rest ricts relationship theory deve lopment. From a practitioner standpoint, current measures of public relations e ffectiveness that focus on short-term program outputs and target public outcomes often fail to demonstrate the value of public relations effectiveness in establishing and maintaining th e long-term mutually be neficial relationships with publics that should be the ultimate objective of effective pub lic relations practice. Measures that can fully capture the quality and nature of the relationships between an organization and its publics could be used to justify the value of public relations to an organization by demonstrating public relations effectiveness. However, measur ement of organization-pub lic relationships has continued to focus on assessments of the public’s direct perspective only.

PAGE 144

144 Public relations theories, such as relationship management, as well as interpersonal communication theories, suggest that the percepti ons of both parties in the organization-public relationship are critical to understanding the na ture of the relationship between them. By only incorporating the public’s direct perspective, the curre nt conceptualization of the organizationpublic relationship fails to acknowle dge the perspectives of the ot her party in the relationship – the organization. My study represents an e ffort to advance theoretical and methodological development in the area of relationship measur ement. The proposition that relationships should be the unit of analysis in pub lic relations theory demands the continued development of new methodologies to measure relationships to ensure that the unit of analys is is being properly operationalized. The coorientational approach for measuring organization-public relati onships detailed in my study represents another step in the effort to improve the measurement of the organizationpublic relationships that comprise this key unit of analysis for public relations research. The findings from my study have serv ed to illustrate how the coor ientational measures that are generated by this methodology can be used to ga in a more comprehensive understanding of the true state of an organization-public relationship.

PAGE 145

145 APPENDIX A RELATIONSHIP MEASURES Control Mutuality: 1. The organization and public are a ttentive to what each other say. 2. The organization believes the opinions of the public are legitimate. 3. In dealing with the public, the organizati on has a tendency to throw its weight around ( reversed ). 4. The organization really listens to what the public has to say. 5. The organization gives the public enough sa y in the decision-making process. Trust: 6. The organization treats the public fairly and justly. 7. Whenever the organization makes important deci sions, I know it will be concerned about the public. 8. The organization can be relied on to keeps its promises. 9. I believe that the organizati on takes the opinions of the pub lic into account when making decisions. 10. I feel very confident about the organization’s capabilities. 11. The organization has the ability to accomplish what it says it will do. Commitment: 12. I feel that the organization is trying to ma intain a long-term commitment to the public. 13. I can see that the organization wants to maintain a relationship with the public. 14. There is a long-lasti ng bond between the orga nization and the public. 15. Compared to other organizations, the public va lues its relationship with the organization more. 16. The public would rather work w ith the organization than not. Satisfaction: 17. The public is happy with the organization. 18. Both the organization and the public benefit from the relationship. 19. Most members of the public are happy in their interactions with the organization. 20. Generally speaking, the public is pleased with the relationship the organization has established with the public. 21. Most members of the public enj oy dealing with the organization.

PAGE 146

146 APPENDIX B PILOT STUDY SURVEYS Public Version SECTION I. The following items are statements about the relationship that exists between [ORGANIZATION] management and staff. For each item, you will be asked to state whether you personally agree or disagree with the statemen t (indicate in the appropriate column to the right of each statement). Indicate agreement or disagreement with each st atement by circling a number between 1 and 7, where 1 means “I strongly disagre e” and 7 means “I strongly agree.” 1. Management and employees are atte ntive to what each other say. 2. Management believes the opinions of employees are legitimate. 3. In dealing with employees, management has a tendency to throw its weight around. 4. Management really listens to what employees have to say. 5. Management gives employees enough sa y in the decision-making process. 6. Management treats employees fairly and justly. 7. Whenever management makes important deci sions, I know it will be concerned about employees. 8. Management can be relied on to keeps its promises. 9. I believe that management takes the opini ons of employees into account when making decisions. 10. I feel very confident about management’s capabilities. 11. Management has the ability to acco mplish what it says it will do. 12. I feel that management is trying to main tain a long-term commitment to employees. 13. I can see that management wants to ma intain a relationship with employees. 14. There is a long-lasting bond betw een management and employees. 15. Compared to other organizations, employees value their relationship with management more. 16. Employees would rather work with management than not. 17. Employees are happy with management. 18. Both management and employees benefit from the relationship. 19. Most employees are happy in thei r interactions with management. 20. Generally speaking, employees are pleased with the relationship management has established with staff. 21. Most employees enjoy dealing with management. SECTION II. You will now be asked to estimate how uppe r management at [O RGANIZATION] would respond to the same statements (indicate in th e appropriate column to the right of each statement). In your opinion, how would a typical member of [O RGANIZATION]’s upper management respond to these statements?

PAGE 147

147 Indicate agreement or disagreement with each st atement by circling a number between 1 and 7, where 1 means “I strongly disagre e” and 7 means “I strongly agree.” 1. Management and employees are atte ntive to what each other say. 2. Management believes the opinions of employees are legitimate. 3. In dealing with employees, management has a tendency to throw its weight around. 4. Management really listens to what employees have to say. 5. Management gives employees enough sa y in the decision-making process. 6. Management treats employees fairly and justly. 7. Whenever management makes important deci sions, I know it will be concerned about employees. 8. Management can be relied on to keeps its promises. 9. I believe that management takes the opini ons of employees into account when making decisions. 10. I feel very confident about management’s capabilities. 11. Management has the ability to acco mplish what it says it will do. 12. I feel that management is trying to main tain a long-term commitment to employees. 13. I can see that management wants to ma intain a relationship with employees. 14. There is a long-lasting bond betw een management and employees. 15. Compared to other organizations, employees value their relationship with management more. 16. Employees would rather work with management than not. 17. Employees are happy with management. 18. Both management and employees benefit from the relationship. 19. Most employees are happy in thei r interactions with management. 20. Generally speaking, employees are pleased with the relationship management has established with staff. 21. Most employees enjoy dealing with management. SECTION III. The following questions ask for some general de mographic information. Please circle the appropriate response or fill in the blank as needed 1. Your gender ( circle one ): (a) MALE (b) FEMALE 2. How old are you? 3. What is your position at [ORGANIZATION]? How many years have you worked in that position? 4. Please estimate the number of hour s you spend in a typical work week involved in face-to-

PAGE 148

148 face interactions with management ( please round to the nearest hour ). 5. Please estimate the number of hours you spend in a typical wo rk week involved in online interactions (e.g., e-mail, chat, etc.) with management ( please round to the nearest hour ). 6. Please estimate the number of hours you spend in a typical work week involved in interactions with management via the telephone ( please round to the nearest hour ). 7. How many hours do you work at [ORGAN IZATION] during a typical week? 8. How many years have you worked at [ORGANIZATION] in all ( circle one )? (a) LESS THAN A YEAR (b) 1-5 YEARS (c) 6-10 YEARS (d) MORE THAN 10 YEARS Organization SECTION I. The following items are statements about the relationship that exists between [ORGANIZATION] management and staff. For each item, you will be asked to state whether you personally agree or disagree with the statemen t (indicate in the appropriate column to the right of each statement). Indicate agreement or disagreement with each st atement by circling a number between 1 and 7, where 1 means “I strongly disagre e” and 7 means “I strongly agree.” 1. Management and employees are atte ntive to what each other say. 2. Management believes the opinions of employees are legitimate. 3. In dealing with employees, management has a tendency to throw its weight around. 4. Management really listens to what employees have to say. 5. Management gives employees enough sa y in the decision-making process. 6. Management treats employees fairly and justly. 7. Whenever management makes important deci sions, I know it will be concerned about employees. 8. Management can be relied on to keeps its promises. 9. I believe that management takes the opini ons of employees into account when making decisions. 10. I feel very confident about management’s capabilities. 11. Management has the ability to acco mplish what it says it will do. 12. I feel that management is trying to main tain a long-term commitment to employees. 13. I can see that management wants to ma intain a relationship with employees. 14. There is a long-lasting bond betw een management and employees.

PAGE 149

149 15. Compared to other organizations, employees value their relationship with management more. 16. Employees would rather work with management than not. 17. Employees are happy with management. 18. Both management and employees benefit from the relationship. 19. Most employees are happy in thei r interactions with management. 20. Generally speaking, employees are pleased with the relationship management has established with staff. 21. Most employees enjoy dealing with management. SECTION II. You will now be asked to estimate how staff at [ORGANIZATION] woul d respond to the same statements (indicate in the appropriate column to the right of each statement). In your opinion, how would a typical [ORGANIZATION] employee respond to these statements? Indicate agreement or disagreement with each st atement by circling a number between 1 and 7, where 1 means “I strongly disagre e” and 7 means “I strongly agree.” 1. Management and employees are atte ntive to what each other say. 2. Management believes the opinions of employees are legitimate. 3. In dealing with employees, management has a tendency to throw its weight around. 4. Management really listens to what employees have to say. 5. Management gives employees enough sa y in the decision-making process. 6. Management treats employees fairly and justly. 7. Whenever management makes important deci sions, I know it will be concerned about employees. 8. Management can be relied on to keeps its promises. 9. I believe that management takes the opinions of employees into account when making decisions. 10. I feel very confident about management’s capabilities. 11. Management has the ability to acco mplish what it says it will do. 12. I feel that management is trying to main tain a long-term commitment to employees. 13. I can see that management wants to ma intain a relationship with employees. 14. There is a long-lasting bond betw een management and employees. 15. Compared to other organizations, employees value their relationship with management more. 16. Employees would rather work with management than not. 17. Employees are happy with management. 18. Both management and employees benefit from the relationship. 19. Most employees are happy in thei r interactions with management. 20. Generally speaking, employees are pleased with the relationship management has established with staff. 21. Most employees enjoy dealing with management. SECTION III.

PAGE 150

150 The following questions ask for some general de mographic information. Please circle the appropriate response or fill in the blank as needed. 1. Your gender (circle one): (a) MALE (b) FEMALE 2. How old are you? 3. What is your position at [ORG ANIZATION]? How many years have you worked in that position? 4. Please estimate the number of hours you spend in a typical wo rk week involved in faceto-face interactions with staff (p lease round to the nearest hour). 5. Please estimate the number of hours you spend in a typical wo rk week involved in online interactions (e.g., e-mail, chat, etc.) with staff (please round to the nearest hour). 6. Please estimate the number of hours you spend in a typical work week involved in interactions with staff via the telepho ne (please round to the nearest hour). 7. How many employees do you currently supervise? 8. How many years have you worked at [O RGANIZATION] in all (circle one)? (a) LESS THAN A YEAR (b) 1-5 YEARS (c) 6-10 YEARS (d) MORE THAN 10 YEARS

PAGE 151

151 APPENDIX C MAIN STUDY SURVEYS Organization Version SECTION 1: Below are several statements about the relations hip that exists between the university police department (UPD) and university students. Please indicate whether you agree or disagree with each statement. Answer to the right of each statement on the scale ranging from 1 to 7, wher e 1 means “I strongly disagree” and 7 means “I strongly agree." [Respondent answers by clicking on 7-point Likert scale with following response options: 1. strongly disagree, 2. disagree, 3. somewhat disagr ee, 4. neither agree nor disagree, 5. somewhat agree, 6. agree, and 7. strongly agree.] 1. UPD and students are attentive to what each other say. 2. UPD believes the opinions of students are legitimate. 3. In dealing with students, UPD has a tendency to throw its weight around. 4. UPD really listens to what students have to say. 5. UPD gives students enough say in its decision-making process. 6. UPD treats students fairly and justly. 7. Whenever UPD makes important decisions, I know it will be concerned about students. 8. UPD can be relied on to keep its promises. 9. I believe that UPD takes the opinions of st udents into account when making decisions. 10. I feel very confident about UPD’s capabilities. 11. UPD has the ability to accomp lish what it says it will do. 12. I feel that UPD is trying to maintain a long-term commitment to students. 13. I can see that UPD wants to maintain a relationship with students. 14. There is a long-lasting bond between UPD and students. 15. Compared to other organizations, students value their relationship with UPD more. 16. Students would rather work with UPD than not. 17. Students are happy with UPD. 18. Both UPD and students benefit from the relationship. 19. Most students are happy in th eir interactions with UPD. 20. Generally speaking, students are pleased with the relationship UPD has established with students. 21. Most students enjoy dealing with UPD. Please click “next” to continue to the next section. [Respondent clicks “Next” to contin ue to next page of survey.] SECTION 2:

PAGE 152

152 Thank you for continuing with the survey. Your responses are valuab le and appreciated. In this section, please estimate how students would respond to the same set of statements. In your opinion, how would a university stud ent respond to these statements? Indicate your estimate to the right of each stat ement on the scale ranging from 1 to 7, where 1 means “A student would strongly disagree” and 7 means “A student would strongly agree." [Respondent answers by clicking on 7-point Likert scale with following response options: 1. strongly disagree, 2. disagree, 3. somewhat disagr ee, 4. neither agree nor disagree, 5. somewhat agree, 6. agree, and 7. strongly agree.] 1. UPD and students are attentive to what each other say. 2. UPD believes the opinions of students are legitimate. 3. In dealing with students, UPD has a tendency to throw its weight around. 4. UPD really listens to what students have to say. 5. UPD gives students enough say in its decision-making process. 6. UPD treats students fairly and justly. 7. Whenever UPD makes important decisions, I know it will be concerned about students. 8. UPD can be relied on to keep its promises. 9. I believe that UPD takes the opinions of st udents into account when making decisions. 10. I feel very confident about UPD’s capabilities. 11. UPD has the ability to accomp lish what it says it will do. 12. I feel that UPD is trying to maintain a long-term commitment to students. 13. I can see that UPD wants to maintain a relationship with students. 14. There is a long-lasting bond between UPD and students. 15. Compared to other organizations, students value their relationship with UPD more. 16. Students would rather work with UPD than not. 17. Students are happy with UPD. 18. Both UPD and students benefit from the relationship. 19. Most students are happy in th eir interactions with UPD. 20. Generally speaking, students are pleased with the relationship UPD has established with students. 21. Most students enjoy dealing with UPD. Please click “next” to continue to the next section. [Respondent clicks “Next” to contin ue to next page of survey.] SECTION 3: We’re almost done with the survey. Below are a few questions about th e type of contact you have with university students. 1. Which of the following UPD programs have you actively participated in? Please check all that apply.

PAGE 153

153 Presentation to students for the SCOPE program Presentation to students for the Rape Aggression Defense (RAD) program An educational presentation for students on person al safety or crime prevention (other than SCOPE or RAD) Other (please specify) [An open-ended respons e box is included for “other” responses.] 2. How many hours do you spend in an average work day communicating with students? This includes communicating face-to-f ace, online (e.g., via e-mail), and/or on the telephone? [Respondent answers using drop down menu ranging from 0 – 24.] 3. On a scale of 1 to 7, where 1 is “very negative” and 7 is “very positive,” in general, how would you characterize these inte ractions with students? [Respondent answers by clicking on 7-point Likert scale with the following response options: 1. very negative, 2. negative, 3. somewhat ne gative, 4. neutral, 5. somewhat positive, 6. positive, and 7. very positive] Please click “next” to continue to the next section. [Respondent clicks “Next” to contin ue to next page of survey.] SECTION 4: Thank you for your patience. Please answer th e following general background questions to conclude the survey. 1. Are you male or female? [Respondent answers by clicking “male” or “female.”] 2. How old are you? [Respondent answers using drop down menu ranging from 18 – 64 plus an “older than 64” option.] 3. Are you a sworn or non-sworn officer? [Respondent answers by clicking “sworn officer” or “nonsworn officer.”] 4. What is your current position at UPD? [Respondent answers in an open-ended response box.]

PAGE 154

154 5. How many years have you worked in your current position? [Respondent answers using drop down menu ranging from 1 – 20 plus “less than a year” and “more than 20 years” options.] 6. How many years have you worked at UPD in all? [Respondent answers using drop down menu ranging from 1 – 20 plus “less than a year” and “more than 20 years” options.] 7. How many years have you worked at the university in all ? [Respondent answers using drop down menu ranging from 1 – 20 plus “less than a year” and “more than 20 years” options.] Please click “next” to continue to the next section. [Respondent clicks “Next” to contin ue to next page of survey.] FINISHED! This concludes the survey. Thank you fo r your time and your participation. If you have any questions or comments regard ing this study, please feel free to contact the primary researcher at tseltzer@jou.ufl.edu. [Clicking on “Done” redir ects respondent to the UPD Web site homepage.] Public Version SECTION 1: Below are several statements about the relations hip that exists between the university police department (UPD) and university students. Please indicate whether you agree or disagree with each statement. Answer to the right of each statement on the scale ranging from 1 to 7, wher e 1 means “I strongly disagree” and 7 means “I strongly agree." [Respondent answers by clicking on 7-point Likert scale with following response options: 1. strongly disagree, 2. disagree, 3. somewhat disagr ee, 4. neither agree nor disagree, 5. somewhat agree, 6. agree, and 7. strongly agree.] 1. UPD and students are attentive to what each other say. 2. UPD believes the opinions of students are legitimate. 3. In dealing with students, UPD has a tendency to throw its weight around.

PAGE 155

155 4. UPD really listens to what students have to say. 5. UPD gives students enough say in its decision-making process. 6. UPD treats students fairly and justly. 7. Whenever UPD makes important decisions, I know it will be concerned about students. 8. UPD can be relied on to keep its promises. 9. I believe that UPD takes the opinions of st udents into account when making decisions. 10. I feel very confident about UPD’s capabilities. 11. UPD has the ability to accomp lish what it says it will do. 12. I feel that UPD is trying to maintain a long-term commitment to students. 13. I can see that UPD wants to maintain a relationship with students. 14. There is a long-lasting bond between UPD and students. 15. Compared to other organizations, students value their relationship with UPD more. 16. Students would rather work with UPD than not. 17. Students are happy with UPD. 18. Both UPD and students benefit from the relationship. 19. Most students are happy in th eir interactions with UPD. 20. Generally speaking, students are pleased with the relationship UPD has established with students. 21. Most students enjoy dealing with UPD. Please click “next” to continue to the next section. [Respondent clicks “Next” to contin ue to next page of survey.] SECTION 2: Thank you for continuing with the survey. Your responses are valuab le and appreciated. In this section, please estima te how members of the UPD woul d respond to the same set of statements. In your opinion, how would a member of the UPD respond to these statements? Indicate your estimate to the right of each stat ement on the scale ranging from 1 to 7, where 1 means “A member of the UPD would strongly disagree” and 7 means “A member of the UPD would strongly agree." [Respondent answers by clicking on 7-point Likert scale with following response options: 1. strongly disagree, 2. disagree, 3. somewhat disagr ee, 4. neither agree nor disagree, 5. somewhat agree, 6. agree, and 7. strongly agree.] 1. UPD and students are attentive to what each other say. 2. UPD believes the opinions of students are legitimate. 3. In dealing with students, UPD has a tendency to throw its weight around. 4. UPD really listens to what students have to say. 5. UPD gives students enough say in its decision-making process. 6. UPD treats students fairly and justly. 7. Whenever UPD makes important decisions, I know it will be concerned about students.

PAGE 156

156 8. UPD can be relied on to keep its promises. 9. I believe that UPD takes the opinions of st udents into account when making decisions. 10. I feel very confident about UPD’s capabilities. 11. UPD has the ability to accomp lish what it says it will do. 12. I feel that UPD is trying to maintain a long-term commitment to students. 13. I can see that UPD wants to maintain a relationship with students. 14. There is a long-lasting bond between UPD and students. 15. Compared to other organizations, students value their relationship with UPD more. 16. Students would rather work with UPD than not. 17. Students are happy with UPD. 18. Both UPD and students benefit from the relationship. 19. Most students are happy in th eir interactions with UPD. 20. Generally speaking, students are pleased with the relationship UPD has established with students. 21. Most students enjoy dealing with UPD. Please click “next” to continue to the next section. [Respondent clicks “Next” to contin ue to next page of survey.] SECTION 3: We’re almost done with the survey. Below are a few questions about th e type of contact you have with members of the UPD. 1. Which of the following UPD services or program s have you used or participated in? Please check all that apply. Presentation to students for the Student Community Oriented Police Effort (SCOPE) program Presentation or class for the Rape Aggression Defense (RAD) program An educational presentation on personal safety or crime prevention (other than SCOPE or RAD) Other (please specify) [An open-ended respons e box is included for “other” responses.] 2. How many hours do you spend in an average day communicating with members of the UPD? This includes communicating face-to-face, onlin e (e.g., via e-mail), and/or on the telephone? [Respondent answers using drop down menu ranging from 0 – 24.] 3. How many hours do you spend in an average week communicating with members of the UPD? This includes communicating face-to-face, online (e.g., via e-mail), and/or on the telephone?

PAGE 157

157 [Respondent answers using drop down menu rangi ng from 0 – 24, plus a “more than 24 hours” option.] 4. On a scale of 1 to 7, where 1 is “very negative” and 7 is “very positive,” in general, how would you characterize thes e interactions with members of the UPD? [Respondent answers by clicking on 7-point Likert scale with following response options: 1. very negative, 2. negative, 3. somewhat negative, 4. neutral, 5. somewhat positive, 6. positive, and 7. very positive] 5. Have you done any of the following ? Please check all that apply. Visited the UPD Web site Read a UPD brochure (e.g., UPD Safe Campus brochure) Read a UPD newsletter (e.g., UPD Advocate Newsletter) Read a press release issued by UPD Listened to a speech by a UPD member Called UPD Wrote an e-mail or letter to UPD Visited UPD’s offices Talked face-to-face with a member of the UPD 6. Have you ever seen or read stories about UPD in the media (e.g., on the Internet, TV, radio, newspaper, etc.)? [Respondent answers by clicking on “yes” or “no.” If “no,” skip question 6 and continue to the next section; if “yes,” continue to question 6.] 7. In general, how do you think UPD is portrayed in these stories? Pleas e indicate your answer on a scale of 1 to 7, where 1 is “very negative” and 7 is “very positive.” [Respondent answers by clicking on 7-point Likert scale with following response options: 1. very negative, 2. negative, 3. somewhat negative, 4. neutral, 5. somewhat positive, 6. positive, and 7. very positive.] Please click “next” to continue to the next section. [Respondent clicks “Next” to contin ue to next page of survey.]

PAGE 158

158 SECTION 4: Thank you for your patience. Please answer th e following general background questions to conclude the survey. 1. Are you male or female? [Respondent answers by clicking “male” or “female.”] 2. How old are you? [Respondent answers using drop-down menu rangi ng from 18 – 64 plus an “older than 64” option.] 3. What is your class standing? [Respondent answers using drop-down me nu with the following response options: “freshman,” “sophomore,” “junior,” “s enior,” and “graduate student.”] 4. How many years have you been a student at the university? [Respondent answers using drop down menu ranging from 1 – 7 plus “less than a year” and “more than 7 years” options.] 5. Do you currently live on campus or off campus? [Respondent answers by clicking “on campus” or “off campus.”] 6. How many years have you lived in your current location? [Respondent answers using drop down menu ranging from 1 – 7 plus “less than a year” and “more than 7 years” options.] 7. Have you ever lived on campus? [Respondent answers by clicking “yes” or “no.”] 8. Do you ever drive on campus and/or park a car on campus? [Respondent answers by clicking “yes” or “no.”] 9. Have you ever been cited by UPD for any reason? [Respondent answers by clicking “yes” or “no.”] Please click “next” to continue to the next section.

PAGE 159

159 [Respondent clicks “Next” to contin ue to next page of survey.] FINISHED! This concludes the survey. Thank you fo r your time and your participation. If you have any questions regarding this study, please feel free to contact the primary researcher at tseltzer@jou.ufl.edu. [Clicking on “Done” redir ects respondent to the UPD Web site homepage.]

PAGE 160

160 APPENDIX D RECRUITMENT EMAILS Organization Members Dear UPD Employees, I would like to invite you to pa rticipate in a research study that I am conducting in cooperation with the university police department. This study examines the relationship that exists between UPD and university students. Your participation is critical to helping us understand the rela tionship between these two groups. The survey is not being conducted for marketi ng purposes; no attempt will be made to sell you anything. Please read the following information carefully be fore you decide to participate in this study: The purpose of this study is to understand the relationship between UPD and students at the university. You will be asked to complete a 10-mi nute online survey that asks for your opinions about this relationship. If you agree to participat e in this study, your answers will be kept confidential. No attempt will be made to link you to your responses. There are no anticipated risks or benefits to you if you agree to participate in this survey. Your participation is voluntary. You are free to withdraw your cons ent to participate at any time simply by closing your browser window. You may also skip any questions you do not wish to answer. If you have any questions or concerns regarding the study, please contact the primary researcher, Trent Seltzer, at tseltzer@jou.uf l.edu, or his supervising professor, Dr. Michael Mitrook, at mmitrook@jou.ufl.edu. Should you have any ques tions regarding your rights as a research participant, please contact the UF Institutiona l Review Board (IRB-02) at (352) 392-0433 or irb2@ufl.edu. The IRB protocol number for this study is #2007-U-322. If you chose to participate, please click on the fo llowing link or cut and paste the URL into your browser window to continue to the survey Web site: http://www.surveymonkey.com/s.asp?u=199163452591 The survey is open until April 25, 2007. Thank you for your cooperation. Thank you for your cooperation. Best regards,

PAGE 161

161 Trent Seltzer Primary Researcher Doctoral Candidate University of Florida College of Journalism & Communication Department of Public Relations Organization – Recontact Dear UPD Employees, Last week you received an email asking for your participation in a re search study that I am conducting in cooperation with the univ ersity police department (UPD). This study examines the relationship that exists between UPD and univers ity students. I would lik e to ask you again to please take a few minutes to participate in this study. Your participation is critical to helping us understand the rela tionship between these two groups. The survey is not being conducted for marketi ng purposes; no attempt will be made to sell you anything. Please read the following information carefully be fore you decide to participate in this study: The purpose of this study is to understand the relationship between UPD and students at the university. You will be asked to complete a 10-mi nute online survey that asks for your opinions about this relationship. If you agree to participat e in this study, your answers will be kept confidential. No attempt will be made to link you to your responses. There are no anticipated risks or benefits to you if you agree to participate in this survey. Your participation is voluntary. You are free to withdraw your cons ent to participate at any time simply by closing your browser window. You may also skip any questions you do not wish to answer. If you have any questions or concerns regarding the study, please contact the primary researcher, Trent Seltzer, at tseltzer@jou.uf l.edu, or his supervising professor, Dr. Michael Mitrook, at mmitrook@jou.ufl.edu. Should you have any ques tions regarding your rights as a research participant, please contact the UF Institutiona l Review Board (IRB-02) at (352) 392-0433 or irb2@ufl.edu. The IRB protocol number for this study is #2007-U-322. If you chose to participate, please click on the fo llowing link or cut and paste the URL into your browser window to continue to the survey Web site: http://www.surveymonkey.com/s.asp?u=199163452591 The survey is open until April 25, 2007. Thank you for your cooperation.

PAGE 162

162 Thank you for your cooperation. Best regards, Trent Seltzer Primary Researcher Doctoral Candidate University of Florida College of Journalism & Communication Department of Public Relations Students Dear university student, I would like to invite you to pa rticipate in a research study that I am conducting in cooperation with the university police department (UPD) and the Department of Housing & Residence Education. This study examines the relationship th at exists between UPD and university students. Your participation will help further our unders tanding of the relationship between these two groups. The survey is not being conducted for ma rketing purposes; no attempt will be made to sell you anything. Please read the following information carefully be fore you decide to participate in this study: The purpose of this study is to understand the relationship between UPD and students. You will be asked to complete a 10-minute online surv ey that asks for your opinions about this relationship. If you agree to participat e in this study, your answers will be kept confidential. No attempt will be made to link you to your responses. There are no anticipated risks or benefits to you if you agree to participate in this survey. Your participation is voluntary. You are free to withdraw your cons ent to participate at any time simply by closing your browser window. You may also skip any questions you do not wish to answer. If you have any questions or concerns regarding the study, please contact the primary researcher, Trent Seltzer, at tseltzer@jou.uf l.edu, or his supervising professor, Dr. Michael Mitrook, at mmitrook@jou.ufl.edu. Should you have any ques tions regarding your rights as a research participant, please contact the UF Institutiona l Review Board (IRB-02) at (352) 392-0433 or irb2@ufl.edu. The IRB protocol number for this study is #2007-U-322. If you chose to participate, please click on the fo llowing link or cut and paste the URL into your browser window to continue to the survey Web site:

PAGE 163

163 http://www.surveymonkey.com/s.asp?u=53993640846 The survey is open until April 25, 2007. Thank you for your cooperation. Best regards, Trent Seltzer Primary Researcher Doctoral Candidate University of Florida College of Journalism & Communication Department of Public Relations Please note: If you do not wish to receive further emails from us, please click the link below, and you will be automatically removed from our mailing list. [RemoveLink]

PAGE 164

164 LIST OF REFERENCES Acock, A. C. (2005). Working with missing values. Journal of Marriage and Family, 67 1012– 1028. Andersen, P. A. (1993). Cognitive schemata in pe rsonal relationships. In S. W. Duck (Ed.), Individuals in relationships (pp. 1-29). Newbury Park, CA: Sage. Aoki, K., & Elasmar, M. (2000, May). Opportunities and challenges of conducting Web surveys: Results of a field experiment Paper presented at the a nnual meeting of American Association for Public Opinion Research, Portland, Oregon. Austin, E. W., & Pinkleton, B. E. (2006). Strategic public relations management: Planning and managing effective communication programs (2nd Ed.). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Baruch, Y. (1999). Response rates in academic studies—a comparative analysis. Human Relations 52 421-434. Bowes, J. E., & Stamm, K. R. (1975). Ev aluating communication with public agencies. Public Relations Review, 1 23-37. Broom, G. M. (1977). Coorientatio nal measurement of public issues. Public Relations Review, 3, 110-119. Broom, G. M., Casey, S., & Ritchey, J. (1997) Toward a concept and theory of organizationpublic relationships. Journal of Public Relations Research, 9 83-98. Broom, G. M., Casey, S., & Ritchey, J. ( 2000). Concept and theory of organization-public relationships. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to the st udy and practice of public relations (pp. 322). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Broom, G. M., & Dozier, D. M. (1990). Using research in public relations: Applications to program management. Englewood Cliffs, NJ: Prentice-Hall. Bruning, S. D., & Galloway, T. (2003). Expandi ng the organization-public relationship scale: Exploring the role that stru ctural and personal commitment play in organization-public relationships. Public Relations Review, 29, 309-319. Bruning, S. D., & Ledingham, J. A. (1999). Relationships between organizations and publics: Development of a multi-dimensional orga nization-public relationship scale. Public Relations Review, 25 157-170. Cameron, G. T., & McCollum, T. (1993). Competing corporate cultures: A multi-method, cultural analysis of the role of internal communication. Journal of Public Relations Research, 5 217-250.

PAGE 165

165 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, 31-63. Cancel, A. E., Mitrook, M. A., & Cameron, G. T. (1999). Testing the c ontingency theory of accommodation in public relations. Public Relations Review, 25 171-197. Christen, C. T. (2005). The utility of coorient ational variables as predictors of willingness to negotiate. Journalism & Mass Communication Quarterly, 85 7-24. Connelly, N. A., & Knuth, B. A. (2002). Using the coorientation model to compare community leaders’ and local residents’ views about Hudson River ecosystem restoration. Society and Natural Resources, 15 933-948. Converse, P. (1964). Nature of belief systems in mass publics. Ideology and Discontent, pp. 227-245. Converse, P. (1970). Attitudes and non-attitudes: Continuation of a dialogue. Quantitative Analysis of Social Problems pp.168-180. Cook, C., Heath, F., & Thompson, R. (2000). A meta-analysis of response rates in webor Internet-based surveys. Educational and Psychological Measurement, 60 821-836. Couper, M. P. (2000). Web surveys: A review of issues and approaches. Public Opinion Quarterly, 64 464-494. Couper, M. P., Traugott, M. W., Lamias, M. J. (1999). Effective survey administration on the web. Paper presented at the Midwest Asso ciation for Public Opinion Research, November. Couper, M. P., Traugott, M. W., Lamias, M. J. (2001). Web survey design and administration. Public Opinion Quarterly, 65 230-253. Cutlip, S. M., Center, A. H., & Broom, G. M. (2000). Effective public relations Upper Saddle River, NJ: Prentice Hall. Dillman, D. (2000). Mail and Internet surveys: The total design method (2nd Ed.). New York: Wiley. Dozier, D. M. (1995). Manager’s guide to excellence in public relations and communication management. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Dozier, D. M., Grunig, L. A., & Grunig, J. E. (1995). Manager’s guide to excellence in public relations and communication management. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

PAGE 166

166 Duck, S. W. (1986). Human relationships: An introduc tion of social psychology. Beverly Hills, CA: Sage. Ehling, W. P. (1992). Estimating the value of public relations a nd communication to an organization. In J. E. Grunig, D. M. Dozier, W. P. Ehling, L. A., Grunig, F. C. Repper, & J. White (Eds.), Excellence in public relations and communication management (pp. 617638). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Ferguson, M. A. (1984, August). Building theory in public relations: Interorganizational relationships as a public relations paradigm Paper presented to the Public Relations Division, Association for Education in Journalism and Mass Communication Annual Convention, Gainesville, FL. Ghanem, S. (1997). Filling in the tapestry: The se cond level of agenda-setting. In M. McCombs, D. L. Shaw, & D. Weaver (Eds.), Communication and democracy (pp. 3–14). Mahwah, NJ: Lawrence Erlbaum Associates. Gonzenbach, W. J. (1996). The media, the president, and public opinion: A longitudinal analysis of the drug issue, 1984-1991. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Groves, R. M., Fowler, F. J., Couper, M. P., Lepkowski, J. M., Singer, E., & Tourangeau, R. Survey Methodology Hoboken, NJ: John Wiley & Sons, Inc. Grunig, J. E. (1972). Communication in community decisions on the problems of the poor. Journal of Communication, 22 5-25. Grunig, J. E. (1984). Organizations, envir onments, and models of public relations. Public Relations Research & Education 1 6-29. Grunig, J. E. (2001). Two-way symmetrical public re lations: Past, present, and future. In R. L. Heath (Ed.), Handbook of public relations (pp. 11-30). Thousand Oaks, CA: Sage. Grunig, J. E., Dozier, D. M., Ehling, W. P., Gr unig, L. A., Repper, F. C., & White, J. (1992). Excellence in public relations and communication management. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Grunig, J. E., & Grunig, L. A. (1992). Models of public relations and communication. In J. E. Grunig, D. M. Dozier, W. P. Ehling, L. A ., Grunig, F. C. Repper, & J. White (Eds.), Excellence in public relations and communication management (pp. 285-325). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Grunig, L. A., Grunig, J. E., & Dozier, D. M. (2002). Excellent public rela tions and effective organizations. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

PAGE 167

167 Grunig, L. A., Grunig, J. E., & Ehling, W. P. (1992) What is an effective organization. In J. E. Grunig, D. M. Dozier, W. P. Ehling, L. A ., Grunig, F. C. Repper, & J. White (Eds.), Excellence in public relations and communication management (pp.65-90). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Grunig, J. E., & Huang, Y. (2000). From organiza tional effectiveness to relationship indicators: Antecedents of relationships, public relationship strategies, and relationship outcomes. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to the study and practice of public relations (pp. 23-53). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Grunig, J. E., & Hunt, T. (1984). Managing public relations. New York: Holt, Rinehart & Winston. Grunig, J. E., & Stamm, K. R. (1973). Co mmunication and coorientat ion of collectives. American Behavioral Scientist, 16 567-591. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5th Ed.). Englewood Cliffs, NJ: Prentice-Hall. Heath, R. L. (2001). Shifting foundations: Public relations as relationship building. In R. L. Heath (Ed.), Handbook of public relations (pp. 1–10). Thousand Oaks, CA: Sage Heberlein, T., & Baumgartner, R. (1978). Factors affecting response rates to mailed questionnaires: a quantitative analys is of the published literature. American Sociological Review, 43 pp. 447-462. Hesse, M. B. (1976). A coorientation study of Wisc onsin state senators and their constituencies. Journalism Quarterly, 53 626-633, 660. Holbrook, A. L., Green, M. C., & Krosnick, J. A. (2003). Telephone versus face-to-face interviewing of national probability sample s with long questionnaires: Comparisons of respondent satisficing and social desirability response bias. Public Opinion Quarterly, 67, 79-125. Hon, L. C., & Brunner, B. (2002). Measuri ng public relationships among students and administrators at the University of Florida. Journal of Communication Management, 6 227-238. Hon, L. C., & Grunig, J. E. (1999). Guidelines for measuring relationships in public relations. Gainesville, FL: Institute for Public Relations. Huang, Y. H. (1997). Public relations strategies, relational outcomes, and conflict management strategies. Unpublished doctoral dissertation, Univ ersity of Maryland, College Park.

PAGE 168

168 Huang, Y. (1999). The effects of public relations strategies on conflict management. Paper presented at the annual confer ence of the International Co mmunication Association, San Francisco, California. Huang, Y. (2001). OPRA: A cross-cultural multiple-item scale for measuring organizationpublic relationships. Journal of Public Relations Research, 13 61-90. Hung, C. F. (2006). Toward the theory of relati onship management in public relations: How to cultivate quality relationships? In E. L. Toth (Ed.), The future of excellence in public relations and communication management: Challenges to the next generation (pp.443476). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Jo, S. (2003). Measurement of organization-public relationships: Validation of measurement using a manufacturer-retailer relationship. Unpublished doctoral dissertation, University of Florida, Gainesville. Jo, S., Hon, L. C., & Brunner, B. R. (2004). Organization-public rela tionships: Measurement validation in a university setting. Journal of Communication Management, 9, 14-27. Jo, S., & Shim, S. W. (2005). Paradigm sh ift of employee communica tion: The effect of management communication on trusting relationships. Public Relations Review, 31 277280. Jones, R. W. (1993). Coorientation of a news staff and its audience. Communication Reports, 6 41-46. Kalton, G., & Kasprzyk, D. (1982). Imputing for missing survey responses. Proceedings of the Survey Research Methods Section, American Statistical Association, pp. 22-31. Kaplan, D. (2000). Structural equation modelin g: Foundations and extensions. Thousand Oaks, CA: Sage. Katz, E. (1957). The two-step flow of communi cation: An up-to-date report of an hypothesis. Public Opinion Quarterly, 21 61-78. Kerlinger, F. N. (1986). Foundations of behavioral research (3rd Ed.). New York: Holt, Rinehart & Winston. Ki, E., & Hon, L. C. (2007). Testing the linka ges among the organization-public relationship and attitude and behavioral intentions. Journal of Public Relations Research, 19 1-23. Ki, E. J., & Shin, J. H. (2006). Status of or ganization-public relations hip research from an analysis of published articles, 1985-2004. Public Relations Review, 32 194-195. Kim, Y. (2001a). Measuring the economic value of public relations. Journal of Public Relations Research, 13 3-26.

PAGE 169

169 Kim, Y. (2001b). Searching for the organizatio n-public relationship: A valid and reliable instrument. Journalism & Mass Communication Quarterly, 78, 799-815. Kiousis, S., & McCombs, M. (2004). Agenda-setti ng effects and attitude strength: Political figures during the 1996 pr esidential election. Communication Research, 31, 36–57. Laing, R. D. (1969). Self and Others. London: Tavistock. Laing, R. D., Phillipson, H., & Lee, A. R. (1966). Interpersonal perception: A theory and a method of research New York: Springer. Lancendorfer, K. M., & Lee, B. (2003). Agenda building and the media: A content analysis of the relationships between the medi a in the 2002 Michigan governor’s race Paper presented at the annual meeting of the Asso ciation for Education in Journalism and Mass Communication, Kansas City, MO. Lang, G. E., & Lang, K. (1983). The battle for public opinion: Th e president, the press, and the polls during Watergate. New York: Columbia University Press. Ledingham, J. A. (2001). Government-community relationships: extending the relational theory of public relations. Public Relations Review, 27 285-295. Ledingham, J. A. (2003). Exp licating relationship management as a general theory of public relations. Journal of Public Relations Research, 15 181-198. Ledingham, J. A., & Bruning, S. D. (1998). Relationship management in public relations: Dimensions of an organization-public relationship. Public Relations Review, 24 55-65. Ledingham, J. A., & Bruning, S. D. (2000a). Public relations as relationship management: A relational approach to the study and practice of public relations Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Ledingham, J. A., & Bruning, S. D. (2000b). A longitudinal study of organization-public relationship dimensions: Defining the role of communication in practice of relationship management. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to the st udy and practice of public relations (pp. 5569). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Ledingham, J. A., Bruning, S. D., Thomlison, T. D ., & Lesko, C. (1997). The transferability of interpersonal relationship dimensions into an organizational setting. Academy of Managerial Communications Journal, 1 23-43. Ledingham, J. A., Bruning, S. D., & Wilson, L. J. (1999). Time as an indicator of the perceptions and behaviors of members of a key public: Monitoring and predicting organization-public relationships. Journal of Public Relations Research, 11 167-183.

PAGE 170

170 Littlejohn, S. W. (1995). Theories of human communication (5th Ed.). Belmont, CA: Wadsworth. Lopez-Escobar, E., Llamas, J. P., & McCombs, M. E. (1998). Agenda setting and community consensus: First and second level effects. International Journal of Public Opinion Research, 10, 335–348. McCombs, M., & Ghanem, S. (2001). The converge nce of agenda-setting and framing. In S. Reese, O. Gandy, & A. Grant (Eds.), Framing public life: Perspectives on media and our understanding of the social world (pp. 67-81). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. McCombs, M., Llamas, J. P., Lopez-Escobar, E., & Rey, F. (1997). Candidate images in Spanish elections: Second-level agenda-setting effects. Journalism & Mass Communication Quarterly, 74, 703–717. McLeod, J. M., & Chaffee, S. H. (1973). Inte rpersonal approaches to communication research. American Behavioral Scientist, 16 469-499. Murphy, P. (1991). The limits of symmetry: A game theory approach to symmetric and asymmetrical public relations. Public Relations Research Annual, 3 115-131. Neff, J. (2005, Nov. 7). Bottom line on PR: It works, says P & G. Advertising Age, 76 1. Newcomb, T. M. (1953). An approach to the study of communicative acts. Psychological Review, 60 393-404. Ohl, C. M., Pincus, J. D., Rimmer, T., & Harri son, D. (1995). Agenda-building role of news releases in corporate takeovers. Public Relations Review, 21, 89–101. O’Keefe, G. J. (1973). Coorienta tion variables in a family study. American Behavioral Scientist, 16 513-536. Peytchev, A., Couper, M. P., McCabe, S. E., & Crawford, S. D. (2006). Web survey design: Paging versus scrolling. Public Opinion Quarterly, 70 596-607. Plowman, K. D. (1995). Congruence between public relati ons and conflict resolution: Negotiating in the organization. Unpublished doctoral dissertation. University of Maryland, College Park. Pollack, E. J. (1984). An organizational analysis of four pub lic relations models in the federal government Unpublished master’s thesis, Univer sity of Maryland, College Park. Porter, S. R., & Whitcomb, M. E. (2003). The im pact of contact type on web survey response rates. Public Opinion Quarterly, 67 579-588.

PAGE 171

171 Purnine, D. M., & Carey, M. (1999). Dyadic coorientation: Reexamination of a method for studying interpersonal communication. Archives of Sexual Behavior, 28 45-62. Richman, W., Kiesler, S., Weisband, S., & Dras gow, F. (1999). A meta-analytic study of social desirability distortion in computer-adm inistered questionnaires, traditional questionnaires, and interviews. Journal of Applied Psychology, 84 754-775. Rockland, D. (2005). Is ROI fo r public relatio ns realistic? Public Relations Tactics, 12 12. Rogers, E. (1995). Diffusion of Innovations (4th Ed.). New York: Free Press. Rubin, A. M. (1994). Media uses and effects: A us es-and-gratifications perspe ctive. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 417-436). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Ruggeriero, T. E. (2000). Uses a nd gratifications theory in the 21st century. Mass Communication and Society, 3 3-37. Sallot, L.M., Lyon, L. J., Acosta-Alzuru, C., & Jones, K. O. (2003). From aardvark to zebra: A new millennium analysis of theory developmen t in public relations academic journals. Journal of Public Relations Research, 15 27-90. Severin, W. J., & Tankard, J. W., Jr. (2001). Communication theories: Origins, methods, and uses in the mass media New York, New York: Addison Wesley Longman. Shin, J., & Cameron, G. T. (2005). Different sides of the same coin: Mixed views of public relations practitioners and journalists for strategic conflict management. Journalism & Mass Communication Quarterly, 82 318-338. Stacks, D. W., & Watson, M. L. (2007). Two-wa y communication based on quantitative research and measurement. In, The future of excellence in public relations and communication management (E. L. Toth, Ed.). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Stafford, L., & Canary, D. J. (1991). Maintenanc e strategies and romantic relationship type, gender, and relational characteristics. Journal of Social and Personal Relationships, 8 217-242. Stamm, K. R., & Bowes, J. E. (1972). Comm unication during an environmental decision. Journal of Environmental Education, 3 49-56. Stegall, S. K., & Sanders, K. P. (1986). Coorie ntation of PR practitione rs and news personnel in education news. Journalism Quarterly, 63 341-352. Tedesco, J. C. (2001). Issue and strategy age nda-setting in the 2000 pr esidential primaries. American Behavioral Scientist, 44, 2048–2067.

PAGE 172

172 Thomlison, T. D. (2000). An interpersonal primer w ith implications for public relations. In J. A. Ledingham & S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to the study and practice of public relations (pp. 177-203). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Turk, J. V. (1985). Public relations in stat e government: A typology of management styles. Journalism Quarterly, 62 304-315. Wimmer, R. D., & Dominick, J. R. (2002). Mass Media Research: An introduction. Belmont, CA: Wadsworth. Wood, J. T. (1995). Relational communication Belmont, CA: Wadsworth. Xavier, R., Johnston, K., Patel, A., Watson, T., & Simmons, P. (2005). Using evaluation techniques and performance claims to demons trate public relations impact: An Australian perspective. Public Relations Review, 31 417-424. Zaller, J., & Feldman, S. (1992). A simple theory of the survey response: Answering questions versus revealing preferences. American Journal of Political Science, 36 (3), 579-616.

PAGE 173

173 BIOGRAPHICAL SKETCH Trent Seltzer is a fourth year doctoral candidate specializing in public relations research at the University of Florida’s College of Jour nalism and Communications. In December 2000, he was awarded his master’s degree in communicati on from the University of Central Florida in Orlando, Florida. He was awarded his bachelor’s degree in political scienc e from the University of Florida in December 1995. Mr. Seltzer has over eleven years of professi onal and teaching experi ence. In 2005, he was selected as the recipient of the Ketchum Excelle nce in Public Relations Research Award from the Institute for Public Relations. Mr. Seltzer has been published in the Journal of Public Relations Research and Public Relations Re view. He has presented papers at numerous conferences, including the annual conferences of the International Communication Association and the Association for Education in Journalism and Mass Communication. Mr. Seltzer is a native of Flor ida and has lived there most of his life. After graduation, he and his wife plan to move to Lubbock, Texas, where Mr. Seltzer has accepted a position as an assistant professor in public relations at Texas Tech University’s College of Mass Communications.