View
2
Download
0
Category
Preview:
Citation preview
1
RELATIONSHIP FRAMEWORK IN SPORT MANAGEMENT: HOW RELATIONSHIP QUALITY AFFECTS SPORT CONSUMPTION BEHAVIORS
By
YU KYOUM KIM
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2008
2
© 2008 Yu Kyoum Kim
3
To my wife, Hyun-Ok
4
ACKNOWLEDGMENTS
This dissertation benefited tremendously from my committee. I am truly honored that I
have learned from the best committee members. First of all, I would like to give many thanks my
advisor, Dr. Galen Trail. I am truly indebted for his advice, encouragement, tremendous support,
and opportunities he provided for last four years at the University of Florida. I cannot find word
to express my wholehearted appreciation for him. I remain in awe of Trail. Next, I would like to
give special thanks to Dr. Yong Jae Ko. He has been friendly, caring, supportive and helpful in
numerous ways. He guided me through tough times. I would also thank other committee
members, Drs. Lutz, Pennington-Gray, and Zhang, who have been really encouraging and
supportive. I am also very grateful that I have worked with the best colleagues at the University
of Florida and the College of Health and Human Performance. Most importantly, I deeply thank
my family and friends. I would like to propose a final toast to Hyun-Ok, my wife, friend, and
partner. I could not have overcome all the hurdles I have encountered throughout my doctoral
studies without her supports and sacrifices.
5
TABLE OF CONTENTS page
ACKNOWLEDGMENTS ...............................................................................................................4
LIST OF TABLES ...........................................................................................................................8
LIST OF FIGURES .......................................................................................................................10
ABSTRACT ...................................................................................................................................11
CHAPTER
1 INTRODUCTION ..................................................................................................................13
Significance of Spectator Sports .............................................................................................13 Need for a Relationship Paradigm ..........................................................................................14 Statement of Problem .............................................................................................................16 Purpose of the Study ...............................................................................................................18
2 LITERATURE REVIEW .......................................................................................................21
Relationship Marketing ..........................................................................................................21 Relationship Quality Construct ...............................................................................................23
Trust .................................................................................................................................23 Commitment ....................................................................................................................25 Relationship Satisfaction .................................................................................................25 Self-Connection ...............................................................................................................26 Love .................................................................................................................................27 Intimacy ...........................................................................................................................27 Reciprocity ......................................................................................................................28
Structural Nature of Relationship Quality ..............................................................................30 General Relationship Quality Factor Model ....................................................................30 Independent Factor Model ...............................................................................................31 Group Factor Model ........................................................................................................32 Second-Order Hierarchical Model ..................................................................................33 Modified Second-Order Hierarchical Model ...................................................................34
Predictive Value of Relationship Quality ...............................................................................35 Behavioral Intention ........................................................................................................35 Word of Mouth ................................................................................................................37 Media Consumption ........................................................................................................38 Licensed Products ............................................................................................................39 Attendance .......................................................................................................................39
6
Moderators of Relationship Quality’s Influence on Sport Consumption Behavior ...............40 Moderator Effects ............................................................................................................41 Moderators of Relationship Quality-Relationship Outcome Linkage .............................42
Product Category ......................................................................................................43 Psychographic Factors ..............................................................................................44
Summary .................................................................................................................................45
3 METHODOLOGY .................................................................................................................57
Participants and Procedures ....................................................................................................57 Instrumentation .......................................................................................................................59
Item Development ...........................................................................................................59 Relationship quality ..................................................................................................59 Relationship quality outcome variables ...................................................................60 Personality traits .......................................................................................................61 Demographics ...........................................................................................................62
Expert Review ........................................................................................................................62 Pilot Study ..............................................................................................................................63
Methods ...........................................................................................................................63 Participants and procedure .......................................................................................63 Instruments ...............................................................................................................63 Data analysis ............................................................................................................63
Results and discussion .....................................................................................................64 Data Analysis for Main Study ................................................................................................66
Descriptive Statistics .......................................................................................................67 Data Screening and Test of Assumption .........................................................................67 Measurement Model ........................................................................................................68 Structural Model ..............................................................................................................70 Moderating Effects ..........................................................................................................70
4 RESULTS ...............................................................................................................................83
Descriptive Statistics ..............................................................................................................83 Demographics ..................................................................................................................83 Relationship Quality Variables ........................................................................................83 Consumption Variables ...................................................................................................83 Relationship Style Variables ...........................................................................................84
Data Screening and Test of Assumptions for Structural Equation Modeling (SEM) .............84
7
Measurement Models ..............................................................................................................85 Relationship Quality (UF Football team) ........................................................................85
Validation of the measure ........................................................................................85 Structure of the relationship quality constructs ........................................................86
Relationship Quality (iPod) .............................................................................................87 Validation of the measure ........................................................................................87 Structure of the relationship quality constructs ........................................................89
Relationship Outcome (UF Football team) .....................................................................90 Relationship Outcome (iPod) ..........................................................................................91 Relationship Personality ..................................................................................................92
Structural Models ....................................................................................................................92 Moderating Effects .................................................................................................................93
Relationship Development ..............................................................................................93 Relationship Maintenance ...............................................................................................94
5 DISCUSSION .......................................................................................................................132
Validation of the Measures ...................................................................................................132 Relationship Quality Constructs (UF Football Team) ...................................................132 Relationship Quality Constructs (iPod) .........................................................................134 Sport Consumption Behaviors and Relationship Style .................................................135
Structural Nature of Relationship Quality ............................................................................136 UF Football Team ..........................................................................................................136 iPod ................................................................................................................................139 Sport Consumption Behaviors (UF Football) ................................................................139
Outcomes of Relationship Quality ........................................................................................140 Moderators of Relationship Quality-Consumption Association ...........................................141 Implications of the Research ................................................................................................142
Conceptual and Theoretical Implications ......................................................................142 Managerial Implications ................................................................................................144
Limitations and Future Directions ........................................................................................146 Summary ...............................................................................................................................147
LIST OF REFERENCES .............................................................................................................149
BIOGRAPHICAL SKETCH .......................................................................................................164
8
LIST OF TABLES
Table page 3-1 Relationship quality (UF Football Team) summary results for measurement model in
pilot study...........................................................................................................................72
3-2 Correlations among relationship quality constructs (UF Football Team) in pilot study ...73
3-3 Summary results for measurement model of relationship quality (iPod) in pilot study ....74
3-4 Correlations among relationship quality constructs (iPod) in pilot study ..........................75
3-5 Consumption behaviors (UF Football team) summary results for measurement model in pilot study ......................................................................................................................76
3-6 Correlations among sport consumption behaviors constructs in pilot study .....................77
3-7 Summary results for measurement model of purchase intention (iPod) in pilot study ......78
3-8 Correlations among consumption behaviors constructs (iPod) in pilot study ...................79
3-9 Summary results for measurement model of relationship style (Initial model) .................80
3-10 Summary results for measurement model of relationship style .........................................81
3-11 Correlations among relational personality constructs in pilot study ..................................82
4-1 Demographic characteristics of participants ......................................................................95
4-2 Descriptive statistics for relationship quality (UF Football) ..............................................96
4-3 Descriptive statistics for relationship quality (iPod) ..........................................................97
4-4 Descriptive statistics for relationship outcomes (UF Football) .........................................98
4-5 Descriptive statistics for relationship outcomes (iPod) .....................................................99
4-6 Descriptive statistics for consumption behaviors (iPod) .................................................100
4-7 Summary results for initial measurement model of relationship quality (UF Football, Seven-Factor Model) .......................................................................................................101
4-8 Summary results for measurement model of relationship quality (UF Football, Seven-Factor Model) .......................................................................................................102
4-9 Correlations among relationship quality constructs (UF Football) ..................................103
9
4-10 Summary results for measurement model of relationship quality (UF Football, Five-Factor Model)...................................................................................................................104
4-11 Correlations among relationship quality constructs (UF Football) ..................................105
4-12 Goodness of Fit indices and Χ2/df values for the hypothesized and alternative models (UF Football) ...................................................................................................................106
4-13 Summary results for initial measurement model of relationship quality (iPod, Seven-Factor Model)...................................................................................................................107
4-14 Summary results for measurement model of relationship quality (iPod, Seven-Factor Model) ..............................................................................................................................108
4-15 Correlations among relationship quality constructs (iPod) ..............................................109
4-16 Summary results for measurement model of relationship quality (iPod, Four-Factor Model) ..............................................................................................................................110
4-17 Correlations among relationship quality constructs (iPod) ..............................................111
4-18 Goodness of fit indices and Χ2/df values for the hypothesized and alternative models (UF Football team) ...........................................................................................................112
4-19 Summary results for measurement model of sport consumption behaviors (with Word of Mouth) ...............................................................................................................113
4-20 Correlations among consumption behaviors constructs ..................................................114
4-21 Summary results for measurement model of sport consumption behaviors ....................115
4-22 Correlations among sport consumption behaviors constructs .........................................116
4-23 Goodness of fit indices and Χ2/df values for the hypothesized and alternative models for sport consumption behaviors ......................................................................................117
4-24 Summary results for measurement model of consumption behaviors (iPod) with Word of Mouth ................................................................................................................118
4-25 Correlations among consumption behaviors constructs (iPod) with Word of Mouth .....119
4-26 Summary results for measurement model of purchase behavior (iPod) ..........................120
4-27 Summary results for measurement model of relationship style .......................................121
4-28 Correlation between relationship style constructs ...........................................................122
10
LIST OF FIGURES
Figure page 1-1 Conceptual framework of relationship quality and its effects on sport consumption
behaviors ............................................................................................................................20
2-1 Graphical comparison of intimacy and self-connection ....................................................47
2-2 General relationship quality factor model ..........................................................................48
2-3 Independent factor model ..................................................................................................49
2-4 Group factor model ............................................................................................................50
2-5 Second-order hierarchical model .......................................................................................51
2-6 Modified second-order hierarchical model ........................................................................52
2-7 Relationship between relationship quality and consumption behavior ..............................53
2-8 Moderating effects on the relationship quality-consumption behavior association ...........54
2-9 Relationship style ...............................................................................................................55
2-10 Relationship between motives and consumption behavior ................................................56
4-1 Second-order hierarchical model (UF Football) ..............................................................123
4-2 Second-order hierarchical model (iPod) ..........................................................................124
4-3 Second-order model for sport consumption behavior ......................................................125
4-4 Structural regression of relationship quality and sport consumption behaviors ..............126
4-5 Structural regression of relationship quality and sport consumption behaviors ..............127
4-6 Interaction effect of relationship development on relationship between relationship quality and sport consumption behaviors ........................................................................128
4-7 Interaction effect of relationship development on relationship between relationship quality and purchase intention for iPod ...........................................................................129
4-8 Interaction effect of relationship maintenance on relationship between relationship quality and sport consumption behaviors ........................................................................130
4-9 Interaction effect of relationship maintenance on relationship between relationship quality and purchase intention for iPod ...........................................................................131
11
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
RELATIONSHIP FRAMEWORK IN SPORT MANAGEMENT: HOW RELATIONSHIP
QUALITY AFFECTS SPORT CONSUMPTION BEHAVIORS
By
Yu Kyoum Kim
August 2008 Chair: Yong Jae Ko Major: Health and Human Performance
Relationship marketing is such an integral part of modern marketing including sport
marketing. Teams are striving to build a good relationship with their fans. The objective of this
dissertation is to provide a better understanding of the nature of the relationship between team
and sport consumers, and the impact of the relationship on various sport consumption behaviors.
Conceptual framework to investigate the research questions were developed based on the
relationship quality literature. Sport consumption behavior, relational personality traits, and
demographics surveys were conducted both online and face-to-face. Various statistical
techniques such as Confirmatory Factor Analysis (CFA), Structural Regression, and Multiple
Sample Structural Equation Modeling were employed for data analysis. A five factor model
including Trust, Commitment, Reciprocity, Self-Connection, and Relationship Satisfaction was
supported to best measure relationship quality between sport consumers and the UF Football
team. However, a four factor model incorporating Trust, Commitment, Reciprocity, and
Relationship Satisfaction was supported to best represent relationship quality for iPod. Regarding
the structural nature of relationship quality, results from both data referent to UF Football Team
and iPod provided support for a second-order hierarchical factor model. A sport consumption
12
behavior model, which consisted of Intention for Attendance, Media Consumption, and Licensed
Merchandise Consumption, was also best explained by a second-order hierarchical model. In
addition, relationship quality significantly influenced sport consumption behaviors related to the
UF Football team and purchase intentions for iPod. None of potential moderators influence the
relationship between relationship quality and its outcomes. This study extends sport management
literature by applying relationship marketing theories to the sport consumer behavior realm.
Researchers and sport industry practitioners should further examine the proposed relationship
quality model in this study.
13
CHAPTER 1 INTRODUCTION
Significance of Spectator Sports
The sport industry is a major segment of the North American economy. The size of sport
industry is estimated to be from $213 billion to $560 billion and it is one of the fastest growing
industries in the United States (Howard & Crompton, 2005). Sport spectating is one of the most
popular leisure activities and the spectator sport segment represents the largest proportion of the
sport industry. Enjoying spectator sports is a virtually ubiquitous phenomenon in North America
(Higgs & McKinley, 2005). Attending ones favorite teams’ games and supporting the team is an
important part of many American’s lives. It would be almost impossible to read the newspaper or
watch television without coming across some type of sport coverage. People are often more
interested in sport trivia than current economic or political issues. Not surprisingly, Street &
Smith’s Sports Business Journal (2007) reported the consumer spending on spectator sports in
the U. S. is estimated at approximately $33 billion a year.
More than 170 million Americans attend the games of the four major professional leagues,
which includes Major League Baseball, the National Basketball Association, the National
Football League, and the National Hockey League (ESPN, 2007) in the 2007 and 2006-2007
season. The National Football League NFL)’s broadcasting contract with the three major
networks of CBS, Fox, and NBC is worth more than $2 billion per year or about $70 million per
team (Badenhausen, Ozanian, & Settimi, 2007). In addition, the total market value for the 30
Major League Baseball teams and the 30 National Basketball Association (NBA) teams are
estimated to be $12.94 billion and $11.17 billion respectively in the 2007 fiscal year
(Badenhausen et al.). In college sports, the prosperity of spectator sport is the same. NCAA
Division I-A football and basketball attracted 37 million and 27 million spectators in the 2006
14
season (NCAA, 2007). The National Collegiate Athletic Association receives about $6 billion
over 11 years from their current contract with CBS for the exclusive broadcast of the NCAA
men’s Basketball Tournament. Average revenue for NCAA Division I schools are over $18
million per year (Fulks, 2005). In sum, the spectator sport segment is evidently a significant part
of the sport industry and North American industry.
Need for a Relationship Paradigm
Although sport organizations have been enjoying substantial benefits from the success of
the spectator sport segment for the last 30 years, sport organizations are recently experiencing a
number of significant changes in the sport business environment. Howard and Crompton (2005)
emphasized the following critical challenges that both professional and collegiate sport
organizations should cope with in the current sport industry: spiraling costs, a saturated market
place, economic disconnect, and emergence of new technology. While revenues have increased
substantially over the past few years, the cost of running a sport organization has gone up much
faster. The average salary in the NBA is more than $4 million a year and the cost of a new
stadium for NFL teams now exceeds $1 billion. In addition, average expenses for Division IA
programs are greater than 20 million. Competition for spectator dollars is more severe than ever
before. In North America, over 600 professional sport teams and 1,000 collegiate athletic
programs are vying with each other to attract spectators. Moreover, many working-class and
middles-class Americans are feeling marginalized from sport teams because the cost of going to
the games has rapidly risen and the traditional working-class and middle class fans cannot afford
the cost any more. Rapidly changing and developing technologies pose both opportunities and
threats to the sport organizations.
Facing these challenges, sport marketers find that a paradigm shift to relationship
marketing is increasingly necessary and relationship marketing has received considerable
15
attention in sport marketing practice due to the following reasons. First, under a saturated and
highly competitive market climate, sport marketers need to redirect their primary focus from
acquisition of new customers to maintenance of existing customers. Creating a new customer is
much more difficult and expensive than retaining a current customer (Fornell & Wernerfelt,
1987; Riechel & Sasser, 1990). The increased importance of customer retention is driving sport
marketers to embrace relationship marketing, which is mainly concerned with establishing a
long-term relationship with customers. It is noteworthy that the series of challenges, which sport
organizations confront now, are similar to those that were responsible for the development of
relationship marketing in many other U. S. industries. Sheth (2002) noted that excess capacity,
high material cost, and intensified competition on a global basis accounted for emergence of the
relationship marketing paradigm.
The second reason that a paradigm shift is necessary is that sport marketers can take
advantage of relationship marketing to repair damaged relationships with middle-class and
working-class fans. Sport organizations cannot afford losing middle-class and working-class fans
any longer. Although luxury seating has provided many teams with a primary source of income
recently, sport teams cannot rely on luxury seating for their sole revenue source because there are
only a finite number of people that have the economic capacity and willingness to pay for the
luxury suites. In addition, fan apathy will lead to declining overall attendance and TV ratings,
which eventually will result in the decrease of sponsorship, licensed products sales, broadcast
contracts, naming right deals, etc (Howard & Crompton, 2005). Relationship marketing
historically was associated with the efforts to nurture relationships with a few key exchange
partners (Hunt & Morgan, 1994). However, information technology (IT)-supported customer
contact techniques now enable sports marketers to develop some form of relationship not only
16
with a small number of affluent accounts but also with a large number of sport consumers. Direct
marketing and data base building techniques are readily available for sport marketers to develop
a customized relationship with many customers, while costing considerably less money than in
past years.
Finally, similar to a service, the sport product is produced, delivered, and consumed at the
same time (Gladden & Sutton, 2005). Therefore, the interaction between spectators and
constituents of sport teams is considered as part of the product (Aijo, 1996), which means that
development of a close relationship between spectators and the team is an integral component of
the marketing task. This characteristic makes relationship marketing a more suitable paradigm
for sport marketers. In sum, there is evidence that there is a growing need to use the relationship
paradigm in sport marketing to overcome the serious challenges confronting sport marketers.
Statement of Problem
Although the amount of the research on relationship marketing is continuously increasing
in various areas of study and demand for implementing relationship marketing is rapidly growing
in sport marketing practice, limited amount of research has investigated relationship marketing in
sport. While the current studies on relationship marketing that exist in the sport management
realm have yielded valuable insights (Bee & Kahle, 2006; Cousens, Babiak, & Bradish, 2006;
McDonald & Milne, 1997; Tower, Jago, & Deery, 2006), several areas of relationship marketing
research in sport management need to be expanded and improved.
First, previous relationship marketing research in sport management has not sufficiently
focused on discovering unique aspects of the team-sport consumer relationship compared to
relationships in the context of the business-to-business market, industrial market, or other
consumer markets. Given that a core value of research in applied areas including sport
management does not lie in only replicating the research findings from other disciplines but also
17
in finding a way to properly apply the findings to unique situations in the different areas, it will
be beneficial to investigate the unique aspects of relationship marketing in the sport consumer
market in addition to common features shared in both sport and general marketing contexts.
Next, the previous relationship marketing research in sport management has seldom considered
how individual’s psychographic and demographic characteristics influence effects of team-sport
consumer relationship on sport consumption behaviors. A better understanding of the
psychographic and demographic factors that change the association between team-sport
consumer relationship and sport consumption behavior will provide researchers with insights to
develop more comprehensive frameworks to explain sport consumption behavior within the
team-sport consumer relationship. Knowledge of the impact of the psychographic and
demographic factors will also enable practitioners to effectively develop a relationship marketing
strategy by segmenting their consumers. Finally, validity of the research findings from current
literature is questionable due, primarily, to the lack of the empirical evidence in support of the
models and results. Therefore, empirical research on relationship marketing in a sport context
will be necessary to advance the body of knowledge on relationship marketing. Among the
various academic and practical issues in relationship marketing, I am particularly interested in
relationship quality for this dissertation.
Relationship quality can be defined as an “Overall assessment of the strength of a
relationship, conceptualized as a composite or multidimensional construct capturing the different
but related facets of a relationship”(Palmatier, Dant, Grewal, & Evans, 2006, p138). The concept
of relationship quality was introduced by Crosby, Evans, and Cowles (1990) nearly two decades
ago based on Dwyer, Schurr, and Oh’s (1987) seminal article on relationships and considerable
effort has been devoted to investigating the various topics about relationship quality since then.
18
The relationship quality concept is an important research topic for three reasons. First,
relationship quality can provide insight into ways of distinguishing successful relationships from
unsuccessful ones. Second, knowledge of relationship quality helps identify what kinds of
problems exist in relationships and determine how those problems should be addressed. Finally,
careful examination of relationship quality can provide a valuable tool for evaluating the
relationship marketing effectiveness as previous research has shown that relationship quality is a
key predicator of company performance such as customer loyalty (De Wulf, Odekerken-
Schröder, & Iacobucci, 2001; Hennig-Thurau, Gwinner, & Gremler, 2002; Sirdeshmukh, Singh,
& Sabol, 2002), word of mouth (Hennig-Thurau et al., 2002; Reynolds & Beaty, 1999), and
expectation of continuity (Crosby et al., 1990; Doney & Cannon, 1997). It is reasonable to
believe that sport team and sport consumer also can benefit from a better understanding of the
relationship quality concept. Although there is a vast amount of relationship quality research, a
review of the extant work reveals some limitations on current literature. First, there seems to be
no consensus regarding the central constructs comprising relationship quality and the structural
nature of those constructs. In addition, very little attention has been paid to the issue of
relationship quality in sport consumer behavior contexts. Finally, there is no tested scale by
which both researchers and practitioners in sport management can measure the quality of team-
sport consumer relationship and evaluate the effects of relationship marketing programs.
Purpose of the Study
The general goal of this dissertation is to expand our knowledge of spectator sport
phenomenon beyond current boundaries by applying relationship theories to the team-sport
consumer context. Particularly, the objective of this dissertation is to provide a better
understanding of the nature of the relationship between team and sport consumers, and the
impact of the relationship on various sport consumption behaviors. The following research
19
questions are investigated throughout the dissertation: (1) What primary constructs can best
represent the quality of the relationship between team and sport consumers? (2) How are
relationship quality constructs structured and cognitively evaluated? (3) How much do the
relationship quality constructs influence sport consumption behaviors? (4) What behavioral
aspects of sport consumers are most influenced by the relationship quality? (5) To what extent is
the association between relationship quality and sport consumption behaviors different across
individuals and contexts? (6) What are the unique aspects of the team-sport consumer
relationship compared to the general firms-consumer relationship?
To answer these questions, first I developed a framework (Figure 1-1).The intent of the
framework is to offer a conceptual foundation for applying relationship theories to explain sport
consumption behaviors. This framework describes the primary components of relationship
quality and structure of relationship quality constructs, the relationship between relationship
quality constructs and sport consumption behaviors, and the role of potential moderators on the
relationship between relationship quality constructs and sport consumption behaviors. Next, I
designed a multi-phase and multi-method study to empirically examine the theoretical model of
relationship support for the theoretical framework while maintaining a focus on the development
of psychometrically enhanced instruments to measure relationship quality constructs, sport
consumption behavior constructs, and relational personality traits. Furthermore I applied recently
developed statistical techniques for analyzing data and testing the model.
20
Figure 1-1. A conceptual framework of relationship quality and its effects on sport consumption
behaviors
21
CHAPTER 2 LITERATURE REVIEW
I begin the following section with an overview of relationship marketing theory. The
definition of relationship marketing is then briefly discussed. Next, I develop a conceptual
framework to investigate the research questions by first reviewing the current literature on
relationship quality constructs and the structural nature of the relationship quality, then by
discussing the consequences of relationship quality, and finally by exploring potential
moderators of the relationship between relationship quality and its outcomes.
Relationship Marketing
Since Berry (1983) first introduced the term “relationship marketing” in the services
marketing area, relationship marketing has grown tremendously both in practice and academia.
The growth of relationship marketing is due to the general belief that relationship marketing
efforts can build stronger customer relationships that lead to improvement in seller performance
outcomes such as sales, market share, and profits (Crosby et al., 1990; Morgan & Hunt, 1984).
The domain of relationship marketing research has extended to the sub-disciplines of marketing.
These include business to business marketing (Doney, Barry, & Abratt, 2007; Dwyer et al.,
1987; Keep, Hollander, & Dickinson, 1998; Naude & Buttle, 2000), sales management (Boles,
Johnson, & Barksdale, 2000; Boorom, Goolsby, & Ramsey, 1998; Brashear, Boles, Bellenger,
Brooks, 2003; Smith & Barclay, 1997), brand management (Fournier, 1998; McAlexander,
Schouten, Koenig, 2002; Parvatiayar & Sheth, 2001; Smit, Bronner, & Tolboom, 2007) channel
relationship (Nicholson, Compeau, Sethi, 2001; Robicheaux & Coleman, 1994; Weitz & Jap,
1995), service marketing (Berry 1995; Evanschitzky, Iyer, Plassmann, Niessing, & Meffer, 2006;
Hennig-Thurau et al., 2002; Gronroos 1995; Gwinner, Gremler, & Bitner, 1998; Robert et al.,
2003), consumer marketing (Garbarino & Johnson, 1999; Odekereken-Schröder, De Wulf, &
22
Schumcher, 2003; O’Malley & Prothero, 2004; Sheth & Parvatiyar, 1995), and international
marketing (Bello & Gilliland, 1997; Sin, et al., 2005; Pan & Tse, 2000 ). Relationship marketing
also has been the focus of research in various industries such as banking (Liang & Wang, 2007;
Molina, Martin-Consuegra, & Esteban, 2007; Prince, 1989), information technology (IT)
(Eastlick, Lotz, & Warrington, 2006; Gruen, Osmonbekov, & Czaplewski, 2006; Sigala, 2006),
the automobile industry (De Hildebrand E Grisi & Ribeiro, 2004; Morgan & hunt, 1994), retail
business (Fullerton, 2005; Srinivasan & Moorman, 2005), health care (Paul, 1988; Naidu,
Parvatiyar, Sheth, & Westgate, 1999; Wrright & Taylor, 2004), advertising (Beltramini & Pitta,
1991; Davies & Palihawadana, 2006; So, 2005), hospitality (Essawy, 2007; Kim, 2006; Kim &
Cha, 2002), nonprofit organizations (Helen & Deborah, 2006; MacMillan, Money, Money, &
Downing, 2005; Simon & Colin, 2007), leisure (Álvarez, Martin, Casielles, 2007; Morais,
Dorsch, & Backman, 2004; Peters, 2004; Tseng & Wu, 2005), and the sport industry (Bee &
Kahle, 2006; McDonald & Milne, 1997; Tower et al., 2006). Overall, it seems that relationship
principles have essentially superseded the short-term exchange scheme in both marketing
research and practice (Palmtier et al., 2006).
Definitions of relationship marketing have been as varied as the disciplines and contexts in
which relationship marketing has been researched. Therefore, I briefly discuss the literature on
conceptualization of relationship marketing and present the definition of relationship marketing
on which I base our conceptual model of relationship quality. Berry (1983) suggested that,
“Relationship marketing is attracting, maintaining and--in multi-service organizations--
enhancing customer relationships” (Berry, 1983, p.25). Since then, many scholars have proposed
various definitions to capture the nature of relationship marketing (Gronroos, 1994; Kotler,
Bowen, Makens, 1996; Morgan & Hunt, 1984; Sheth & Parvatiyar, 2000). Although these
23
definitions vary in perspectives and approaches, they typically identify three fundamental aspects
of relationship marketing: process, purpose, and parties (Sheth & Parvatiyar). First, the
definitions emphasize the process aspect of relationship marketing and the prevailing idea is that
the process is characterized by establishment, enhancement, and maintenance of relationships.
Next, there is general agreement that the purpose of relationship marketing is to achieve benefit
for all parties involved in the relationship. Finally, by its very nature, relationship marketing has
entities who participate in relational exchanges with a focal firm and the nature of the
relationship differs by the type of partners. Morgan and Hunt suggested that there were ten types
of partners: (1) goods suppliers; (2) service providers; (3) competitors; (4) nonprofit
organizations; (5) government; (6) ultimate customers; (7) intermediate customers; (8) functional
departments; (9) employees; and (10) business units. For the purpose of the current research, I
focus on the ultimate customers, in our case, sport consumers, as relationship partners. Thus,
based on the previous literature, I propose the following: Relationship marketing to sport
consumers is a set of marketing activities to establish, enhance, and maintain a relationship with
sport consumers for the benefit of both sport team and sport consumers.
Relationship Quality Construct
Many researchers have offered various lists of relationship quality constructs. After closely
reviewing the literature pertaining to components of relationship quality I identified seven
constructs that are commonly claimed to capture the essential facets of relationship quality. The
constructs that I have included in my conceptual model are trust, commitment, satisfaction, love
(liking), intimacy, self-connection and reciprocity. I elaborate on each of these constructs below.
Trust
Trust typically has been considered as a critical component of a successful relationship
(Garbarino & Johnson, 1999; Dwyer, et al., 1987; Morgan & Hunt, 1994; Palmatier et al., 2006).
24
Anderson and Weitz (1989) defined trust as “one party’s belief that its needs will be fulfilled by
actions undertaken by the other party” (p. 312). Trust was also defined as “a willingness to rely
on an exchange partner in whom one has confidence” (Moorman, Deshpandé, & Zaltman, 1993,
p. 82). Both definitions emphasized that confidence is an essential part of trust. Morgan and Hunt
(1984) suggested that trust is based on a judgment that the relationship partner is reliable and has
high integrity and they defined trust as “confidence in an exchange partner’s reliability and
integrity” (p. 23). Trust has been broadly investigated in the social exchange literature and other
areas. Trust has been considered to be the foundation of cooperation (John, 1984; Nicholson et
al., 2001). Morgan and Hunt found that trust reduced opportunistic behavior and conflict in
relational exchanges. They also found that trust encouraged cooperative behavior. Berry (1995)
argued that trust can reduce customers’ perceived risk in service, which is inherent because it is
difficult for customers to determine quality of service before they experience it. In addition, trust
has been found to influence various seller performance objectives such as market share, sale, and
profit (Doney & Cannon, 1997; Reynolds and Beatty, 1999; Siguaw, Simpson, & Baker, 1998;
Palmatier, et al., 2006). Some researchers have tended to highlight types of trust that can be
found in person-to-person relationships such as employee-employer and salesperson-customer
relationships. However, Morgan and Hunt claimed that trust was critical to all types of relational
exchanges. Garbarino and Johnson(1999) also suggested that consumer’s trust could be put in an
organization as well as a person. They argued that the consumer’s trust in the organization is the
consumer’s confidence in the quality and reliability of service or in the product offered by an
organization in the same way as the consumer’s trust in an individual partner is the confidence in
quality and reliability of an action taken by individual partners. Therefore, rather than focusing
on trust in individuals, I investigate customers’ trust in a sport team, which reflects customers’
25
beliefs about the quality and reliability of various services provided by the team, following
Garbarino and Johnson’s approach.
Commitment
Like trust, commitment has been identified as a vital component of successful relationships
(Dwyer, et al., 1987; Garbarino & Johnson, 1999; Morgan & Hunt, 1994; Palmatier et al., 2006).
Commitment has generally been defined as “an exchange partner believing that an ongoing
relationship with another is so important as to warrant maximum efforts at maintaining it; that is,
the committed party believes that relationship is worth working on to ensure that it endures
indefinitely” (Morgan & Hunt,1984, p. 23). Levy and Weitz (2004) stated that commitment is
one of the major characteristics that differentiate relational partnerships from functional
relationships. Morgan and Hunt found that commitment had a positive influence on acquiescence
and cooperative behavior, but commitment had a negative influence on propensity to leave. Also,
it has been shown that strong commitment results in improvement of sales, market share, and
profits (Doney & Cannon, 1997; Reynolds and Beatty, 1999; Siguaw et al., 1998; Palmatier et
al., 2006).
Relationship Satisfaction
Satisfaction with the relationship has been regarded as an important measure of
relationship quality (Garbarino & Johnson, 1999; Odekerken-Schoröer et al., 2003; Palmatier et
al., 2006; Roberts et al., 2003). Relationship satisfaction can be defined as customers’ affective
or emotional state toward the relationship with a brand or firm based on the overall evaluation of
the relationship (Garbarino & Johnson; Odekerken-Schoröer et al.; Palmatier et al.; Roberts et
al.). Note that relationship satisfaction in our study reflects cumulative evaluation over the course
of relationship rather than transactional or encounter-specific evaluation. In addition, relationship
satisfaction differs from general satisfaction because the former exclusively refers to the
26
customer’s satisfaction with the relationship with a brand or firm but the latter describes the
customer’s satisfaction with the overall exchange. Odekerken-Schoröer et al. (2003) found that
the customers tended to be more satisfied with the relationship with a firm when the customers
perceived higher level of the firm’s customer retention orientation. Crosby et al. (1990)
suggested that relationship satisfaction resulted in high sales effectiveness and more future
interaction. In addition, relationship satisfaction has been found to positively influence sales,
market share, and profit (Palmatier et al., 2006).
Self-Connection
Self-connection has been frequently recognized as an essential indicator of the relationship
quality (Smit et al., 2007; Swaminathan, Page, & Gürhan-Canli, 2007; Thorbjørnsen,
Supphellen, Nysveen, & Pedersen, 2002). Fournier (1998) stated that self-connection is a
“relationship quality facet [that] reflects the degree to which the brand delivers on important
identity concerns, tasks, or themes, thereby expressing a significant aspect of self” (p. 364).
Strong self-connections guide customers to maintain the relationship through developing the
protective feelings of uniqueness and dependency (Drigotas & Rusbult, 1992). In addition, high
levels of self-connection encourage customers to stay in the relationship when they face adverse
circumstance (Lydon & Zanna, 1990). Self-connection to brand or organization parallels team
identification. Theoretical origin of both concepts can be traced back to the identity theory
(Stryker, 1968), which noted that individuals assume multiple roles (identities) that represent
who they are and these identities guide the individual’s behavior. Although team identification
has not been investigated within a relationship quality framework, it has been considered to be a
key factor to explain various sport consumer behaviors (Trail, Anderson, & Fink, 2005). Team
identification has been found to influence expectancies for event experience and outcome (Trail,
Fink, & Anderson, 2003), intention to attend games (Matsuoka, Chelladurai, & Harada, 2003),
27
and actual attendance (Laverie & Arnett, 2000). Moreover, Sutton, McDonald, Milne, and
Cimperman (1997) suggested that highly identified fans were less price-sensitive. Wann (2006)
claimed a higher level of identification with a local team positively influenced the psychological
health of fans.
Love
Several researchers have recognized that love is an important construct to understand the
nature of relationships (Fournier, 1998; Nicholson et al., 2001; Pawle & Cooper, 2006; Smit et
al., 2007). Fournier suggested that love is an emotional feeling that embraces warmth, affection,
passion, infatuation, and obsession. While some researchers suggested that a customer’s love of a
company or objects might not be wholly analogous to interpersonal love (Oliver, 1999), the
literature generally supported that a customer’s love of a company or objects and interpersonal
love are fundamentally similar in many contexts (Thompson, MacInnis & Park, 2005;
Thorbjørnsen et al., 2002). Love has been considered to be a strong motivator for developing and
maintaining human relationships (Sternberg, 1986). Previous research has shown that love
attenuated negative consequences of relationship problems (Rusbult, Verette, Whitney, Slovik, &
Lipkus, 1991), influenced judgment on attributing blame (Bradbury & Fincham, 1990), and
positively biased the perception of the partner (Murray, Holmes, & Griffin, 1996).
Intimacy
Several researchers have identified intimacy as a fundamental component of relationship
quality (Barnes, 1997; Smit et al., 2007; Fletcher, Simpson, & Thomas, 2000; Monga, 2002).
Much of the extant literature has focused on the intimacy within romantic relationships and the
term intimacy often refers to sexual feelings and physical contact, which were only experienced
in the context of romantic relationships (Gaia, 2002). However, many social psychologists have
also recognized and investigated non-sexual dimensions of intimacy, which can be widely found
28
in most relationships and they play a critical role in relationships (McAdams, 2000; Hook,
Gerstein, Detterich, & Gridley, 2003; Prager & Buhrmester, 1998). In the current study, I focus
on non-sexual dimensions of intimacy and conceptualize intimacy as the degree of familiarity,
closeness, and openness to relationship partners following Fournier (1998), who defines intimacy
in the customer-brand relationship context. My definition is also consistent with Sternberg’s
(1986) definition, which also highlights that closeness and openness are essential features that
constitute intimacy in various relationship contexts. Although intimacy might be interrelated
with self-connection, intimacy is different from self-connection in that intimacy is a concept
focusing on the distance between individuals and an organization while the focal point of the
self-connection concept is the extent of the overlap between an individual’s self and an
organization (Figure 2-1). Fournier emphasized that successful brand relationship was built on
the higher level of intimacy between relationship partners. In addition, intimacy has been
considered to foster continuity of relationship by influencing perceptions of relationship partners
(Murray et al., 1996), improving the effect of persuasive communication efforts, and facilitating
conflict resolution (Stern, 1997).
Reciprocity
A strong and successful relationship is also characterized by a high degree of perceived
reciprocity between relationship partners (De Wulf et al., 2001; Eyuboglu & Buja, 1993; Miller
& Kean, 1997; Schwartz, Trommsdorff, Albert, & Mayer, 2005; Uhl-Bien & Maslyn, 2003).
According to Gouldner (1960), reciprocity is the generalized moral norm guiding social
interaction among individuals and Gouldner stated that “the generalized norm of reciprocity
evokes obligations toward others on the basis of their past behavior (p. 170). The principle of
reciprocity states that when one benefits from another, the recipient should return the favor in
proportion to what the other has done for him or her. Until the recipient reciprocates the benefit
29
received from the donor, he or she is obliged or indebted to the giver (Gouldner). The importance
of reciprocity as a guiding principle of social behavior has long been recognized. Cicero
acknowledged, “There is no duty more indispensable than that of returning a kindness” (Goulder,
p. 161). Cialdini (1998) stated that the rule of reciprocity is one of the most widespread and vital
norms of human culture and society. He also emphasized that the development of various
relationships, transactions, and exchanges that are foundations of human society heavily depend
on this sense of obligation or indebtedness. He also argued that all members of society learned
from childhood that they should abide by the rule or they would be punished with serious social
disapproval. Consequently, the reciprocity rule often affects the decision to agree to another’s
request. Cialdini suggested that giving something to others would be a very useful tactic for
persuasion because of the three characteristics of the principle of reciprocation: (1) the principle
of reciprocity is exceptionally strong and it frequently dominates the impact of other factors that
generally influence the decision to comply with a request; (2) the principle is in effect even in a
situation that the initial favor was not invited by recipient, which limits an individual’s choice in
deciding whom he or she wants to owe and allows others to have the choice; (3) the principle can
stimulate unequal exchanges. An individual often willingly returns a considerably larger favor
than the one he or she first was given to be free from uncomfortable feelings of obligation or
indebtedness.
In general, reciprocity has been considered to be a key factor in predicting the duration and
stability of an exchange relationship (Larson, 1992). In addition, Gouldner (1960) noted that
reciprocity plays a critical role as a starting mechanism as well as stabilizing function.
Reciprocity has frequently been considered as a key variable of interest in channel relationships
(De Wulf et al., 2001). For example, Smith and Barclay (1997) found that perceived reciprocity
30
in channel members increase perceived task performance by selling partners, reduced barrier of
risk, and therefore motivated parties to continue the relationship. Uhl-Bien and Maslyn (2003)
investigated the reciprocity in manager-subordinate relationships and they reported that
perception of positive reciprocity in manager-subordinate work relationships and performance of
subordinates was positively associated. In the consumer behavior realm, Bagozzi (1995, p175)
highlighted reciprocity as “the core of marketing relationship” and stated that the principle of
reciprocity could apply to customer-firm relationships as well. He also emphasized that
relationship marketing research should further examine the psychological manifestation of
reciprocity and the way it serves its role in relational exchanges between consumer and firm.
When consumers perceived that they have a reciprocal relationship with a brand or store, they
respond to an unsatisfactory quality of product or service by collaborating and compromising
(Kaltcheva & Weitz, 1999). Miller and Kean (1997) found that in a rural community, reciprocity
was the strongest motivator for maintaining a relationship with local retailers. In a leisure
context, Morais et al. (2004) reported that tourists’ perceived reciprocity in tourist-provider
relationships encouraged the tourist to resist changing providers when they faced counter
persuasion.
Structural Nature of Relationship Quality
This multi-faceted array of relationship quality factors raises an essential question: How
are these constructs evaluated and structured? There might be several possible solutions for the
question. In the following section, several types of models that differ by restrictions on the
models will be discussed. After the theoretical discussion, four models were empirically tested.
General Relationship Quality Factor Model
Although I hypothesize that relationship quality incorporates the seven distinct
constructs, it is possible that relationship quality is one global factor that collapses across the
31
seven individual aspects: trust, commitment, satisfaction, love (liking), intimacy, self-
connection, and reciprocity. According to this explanation, the individual dimensions of
relationship quality do not exist as distinct conceptual constructs. This model is depicted in the
Figure 2-2. Although this model can be considered to be a starting point to empirically determine
how many common factors are measured by all the indicator variables in an exploratory way, the
model is not theoretically plausible because previous research generally identified the seven
factors as distinctive conceptual factors (Fletcher et al., 2000; Morgan & Hunt, 1984; Fournier,
1998; Roberts et al., 2003).
Independent Factor Model
The second possibility is that the seven constructs of relationship quality (trust,
commitment, satisfaction, love (liking), intimacy, self-connection, and reciprocity) are distinct
facets of relationship quality and they are completely independent. This model is illustrated in
Figure 2-3. This model and the general relationship quality factor model are at the opposite ends
of the spectrum in terms of uniqueness of the individual domains of relationship quality. The
former states that there is no distinct sub-domain of relationship quality and all the indicators are
completely accounted for by one global factor of relationship quality. However, the latter
suggests that each sub-domain of relationship quality represents completely unique aspects of
relationship quality. Although the independence or orthogonality of the underlying dimensions is
often assumed in certain types of classic exploratory factor analysis, similar to this independent
factor model, the model is not theoretically plausible. First, in social science, it is unrealistic to
hypothesize that all the theoretical constructs of consideration in any study are completely
uncorrelated. In addition, previous research has shown that the seven constructs of relationship
quality are correlated to each other (Garbarino & Johnson, 1999; Mogan & Hunt, 1984;
Nicholson et al., 2001; Stern, 1997; Uhl-Bien & Maslyn, 2003).
32
Group Factor Model
The third possibility is that the seven constructs of relationship quality reflect distinct
facets of relationship quality and they are correlated with each other to a certain extent.
However, no higher order relationship quality construct is specified. According to this model,
individual sub-domains of relationship quality represent more of a unique aspect of relationship
quality than common aspects. This model is shown in Figure 2-4. This type of model is a
reasonable explanation of factor structures when a specific structure that would account for the
relationship among the first order factors is not known (Kline, 2005; Rindskopf & Rose, 1988).
In addition, this type of model is particularly plausible if there is a theoretical justification that
the factors in the analysis are correlated with each other. For example, many researchers have
investigated motivational factors for sport consumption behavior and have developed scales to
measure those factors (Milne & McDonald, 1999; Trail & James, 2001; Wann, 1999). They
typically have specified that the factors are unconstrained in the confirmatory factor analysis and
the group factor model has been considered to explain the structure of motives best because the
motives have shown to be correlated with each other but a consistent structure that accounts for
relationship of the motives has not been found. The breadth and diversity of the motivational
factors identified in sport consumption behavior might account for the finding. The seven
constructs of relationship quality have been considered to be interrelated with each other in the
previous research (Garbarino & Johnson, 1999; Johnson & Grayson, 2005; Morgan & Hunt,
1984; Nicholson et al., 2001; Sin et al., 2005; Stern, 1997; Uhl-Bien & Maslyn, 2003).The
correlation between those constructs vary across the contexts but they typically ranged from .25
to .80. However, there is limited research concerning the measurement structure of those
constructs. Therefore, we propose that the group factor model, which specifies that all factors are
33
correlated with each other but are not represented by a higher order factor, is a competing model
to explain the structure of relationship quality constructs included in this study.
Second-Order Hierarchical Model
The next possibility is that the relationship quality constructs are manifestations of a
global or higher order construct of relationship quality. This model suggests that each
relationship quality construct can be measured by items or observed variables corresponding to
one of the seven relationship quality constructs and these individual constructs reflecting
different aspects of relationship quality are indicators of the more general, higher-order latent
relationship quality construct. This model is depicted in Figure 2-5. This type of model can be
justified when people make evaluative judgments on different but related conceptual domains
consistently based on a broader conceptual node. For example, Trail and Robinson (2005)
proposed that an individual might be identified with seven different points of attachments such as
the players, the coach, the university, the sport, the level of the sport (e. g. college opposed to
professional), and the community, as well as the team. Later, Kim and Trail (2007) found
identification with those different points of attachment converged to an overall perception of
identification. Although there is limited research on the structure of relationship quality, a few
studies suggested that relationship quality might be a global construct that can be reflected by
various sub-dimensional factors. DeWulf et al. (2001) identified relationship satisfaction, trust,
and relationship commitment as primary relationship quality constructs and they are specific sub-
dimensions that make up more abstract relationship quality perception. Fournier (1994)
emphasized that behavioral interdependence, love/passion, personal commitment, self-concept
connection, intimacy, partner quality, and nostalgic connection are major facets of relationship
quality. Fournier also suggested that these facets are distinct but interrelated sub-dimensions that
homogeneously represented general relationship quality, which was a higher-order construct.
34
Similarly, Fletcher et al. (2000) reported that satisfaction, commitment, intimacy, trust, passion,
and love were critical components of perceived relationship quality and they were consistent
reflections of one’s overall attitude toward relationship. Based on the theoretical discussion
above, I determined that the second-order hierarchical model provides a plausible account for the
structure of the relationship quality constructs in our study.
Modified Second-Order Hierarchical Model
Another possibility is that the relationship quality constructs are manifestations of more
than two second-order constructs of relationship quality. This model suggests that first-order
relationship quality constructs tap not one but two or more second-order relationship quality sub-
dimensions. The second-dimensions are correlated but distinct. Thus, no third-order relationship
quality constructs are specified. This model is illustrated in Figure 2-6. This type of model can be
a tenable explanation when a psychological concept refers to a wide range of sub-dimensional
constructs and those sub-dimensions do not converge to single global dimensions but to two or
more higher-order dimensions (Kline, 2005; Rindskopf & Rose, 1988). For example, Oliver
(1997) suggested that attitudinal brand loyalty was divided into three distinct conceptual
categories: cognitive, affective, and cognitive domains of attitudinal brand loyalty. This is also
consistent with general acceptance of the idea that attitude is multi-dimensional rather than uni-
dimensional (Ostrom, 1969; Bagozzi & Burnkrant, 1980). Although there is limited research on
second-order dimensions that consist of overall relationship quality perception, Fournier (1994)
argued that relationship quality could be divided into two main categories (i.e., affective and
cognitive/behavioral domains), which is reflected by a lower-order latent factor. Similarly, Smit
et al. (2007) reported that brand relationship quality was a two-dimensional construct. The first
domain contained passionate attachment, intimacy, and connection, while the second domain
contained trust, personal commitment, and love. Based on the theoretical discussion above, I
35
propose the modified second-order model as a competing model. This model hypothesizes that
first-order relationship quality constructs tap two corresponding second-order constructs:
cognitive (i.e., trust, commitment, satisfaction, and self-connection) and affective (i.e., love,
intimacy, and reciprocity) dimensions.
Predictive Value of Relationship Quality
Although relationship quality constructs are valuable in their own right, a crucial issue in
the research of relationship quality is to examine predictive capacity of the relationship quality
constructs. How well do relationship quality constructs predict managerially relevant sport
consumption behavior? Literature reveals four behavioral aspects of interest in sport
management: Word of Mouth, Merchandise consumption, media consumption, and attendance.
After a note on the relationship between intention of behavior and actual behavior, I will present
a model to address the question about the relationship between relationship quality and sport
consumption behavior (Figure 2-7).
Behavioral Intention
The prediction of actual sport consumption behavior is of principal interest in sport
management. However, deciding which specific construct and measure should be used to best
predict the actual sport consumption behavior has been a major issue. Consumers’ self reported
intentions of behavior have been most frequently employed in academia and practice in
marketing. For example, most scholarly works on satisfaction have utilized repurchase intention
as the criterion variable and practitioners typically depend on purchase intention data to make
predictions about consumers’ initial purchase of new products or the repeat purchase of existing
products (Chandon, Morwitz, & Reinartz, 2005). Although it is generally admitted that
participants’ self-reported intentions do not always accurately forecast their future behavior and
the strength of relationship between self-reported intention and the actual behavior are
36
considerably influenced by several factors (Morwitz, Steckel, & Gupta, 2007), using intention as
a measure to predict or explain actual behavior can be justified by following theoretical and
practical reasons.
First, many theoretical frameworks of consumer behavior have conceptualized that
intention is a proximate psychological construct for actual behavior. For example, Fishbein and
Ajzen (1975, p. 368-369) noted, “if one wants to know whether or not an individual will perform
a given behavior, the simplest thing one can do is to ask the individual whether he intends to
perform that behavior.” Warshaw (1980) suggested that most conceptual frameworks of
consumer behavior characterized intention as a critical mediator between attitude and choice
behavior, which indicated that intention was a better psychological measure than belief and other
cognitive measures. In addition, Morwitz et al. (2007), using meta-analysis, found that intentions
are generally predictive of future behavior and the variation in predictive or explanatory power
of intention could be accounted for by types of products and the way that the data was collected.
Specifically, Morwitz et al. found the following:
Intentions are more correlated with purchases (1) for existing products than for new products; (2) for durable products than for non-durable products; (3) when respondents are asked to provide intentions to purchase specific brands or models than when they are asked to provide intentions to buy at the product category level; (4) when purchase levels are measured in terms of trial rated rather that total market sales; (5) for short time horizons than for long time horizons: and (6) when intentions are collected in a comparative mode than when they are collected monadically (p. 361).
Based on these findings, Morwitz et al. further suggested that intention could be a useful
predictor of future behavior if researchers have knowledge about the contextual factors that
affect the strength of the intention-behavior relationship and ability to properly interpret intention
scores in various situations based on the knowledge of those factors. In sport consumption
behavior context, Trail et al. (2005) suggested that behavioral intention is a preferable predictor
of actual sport consumption behavior. In addition, Trail, Anderson, and Lee (2006) examined the
37
relationship between spectators’ intention to attend a specific team’s games in preseason and
their actual attendance during the season. Their result showed that preseason attendance intention
accounted for 45% of variance in actual attendance, indicating intention to attend sporting events
is a significant predictor of actual attendance.
Utilizing intention as a proxy of behavior is also justified because it is a practical
alternative to actual behavior. Examining the relationship between variables of interest and
consumers’ self-reported behavioral intention is typically more time and cost efficient than
investigating relationships between those variables and the consumers’ future behavior because
the former does not require a longitudinal study. Moreover, there are various forecasting models
to convert intentions to actual behavior that are now easily available. For example, Jamieson and
Bass (1989) introduced several conversion schemes that researchers and marketers can employ to
predict actual purchase behavior from intentions scores. Using previous studies that have
information about the consumers’ purchase intention and then following their actual purchases,
Jamieson and Bass attained the conversion rates that were applied to measure purchase intentions
to estimate future purchases. In addition, Morwitz et al. (2007) described how the conversion
rates for intention scores can be modified by the characteristics of the studies. Therefore, current
studies focus on the relationship between relationship quality constructs and behavioral intention
in lieu of actual behavior.
Word of Mouth
Word of mouth refers to a behavior in which a consumer informally communicates
experience, evaluation, and recommendation of goods or services with other potential consumers
(Anderson, 1998). Word of mouth communication is a highly influential factor in consumers’
purchasing decisions and often more powerful than other promotional methods that can be used
by marketers mainly because personal communication is perceived as a more trustworthy and
38
dependable source than non-personal information (Hennig-Thurau et al., 2002; Zeithmal &
Bitner, 1996). While consumers’ product awareness is largely improved by mass media, in many
occasions word of mouth is more effective in influencing the actual purchasing decision (Bayus,
1985). Previous literature frequently refers to word of mouth as a key outcome of relationship
quality constructs. Palmatier et al. (2006) reported that relationship quality explained on average
37% of variance in word of mouth across 17 different empirical studies in a consumer products
context. In a service products context, Hennig-Thurau et al. (2002) found that more than 35% of
variance in word of mouth was accounted for by relationship quality. Thus, word of mouth can
be viewed as a critical outcome of team-spectator relationship quality.
Media Consumption
One of the unique characteristics of sport consumption is that sport consumers can
consume the product through the media such as television, radio, and internet. Development of
media has changed the way sport is consumed and the sport industry. The sport media segment is
the financial base of the sport industry. According to Howard and Crompton (2005), U.S.
consumers spent almost $13 billion on media sport consumption. Increasing media related
consumption of a team is essential for the success of the organization. Established and
marketable teams that can attract a large audience can enjoy substantial revenue from broadcast
rights. In addition, media consumption level is closely related to the teams’ sponsorship
solicitation and licensed product sales (Goff & Ashwell, 2005). Consequently, not only
professional organizations but also collegiate athletic programs have become more interested in
the level of media consumption. Although no research has directly investigated the association
between relationship quality and level of media consumption, previous research on relationship
quality generally suggests that relationship quality is a crucial predictor of behavioral
dependence (Fournier, 1998). That is, customers who perceived high quality relationship with a
39
brand or company do not only buy more products from the brand or the company but also
expand scope, diversity and frequency of brand-related or company-related activities. Hence, I
expect that the fans who perceive a higher level of relationship quality intend to consume sport
through media.
Licensed Products
Sport-related licensed products are any and all products bearing the name or logo of a
sports team, which manufactures use, sell, and offer for sale through licensing contract with the
league or team. Retail sales of sport-related licensed products in the U.S. and Canada were
estimated to be worth $13.23 billion in 2005 (“Licensing Letter Survey”, 2006). The sale of
licensed products has been a major area of interest for sport managers and marketers due to the
following reasons. First, licensing is major revenue source for teams and leagues. In the 2005
fiscal year, licensed apparel sales for the National Football League (NFL) totaled $1.58 billion
(Brochstein, 2006). Next, licensing enables teams and leagues to enhance awareness and interest
as well. No research has empirically examined the relationship between relationship quality and
level of licensed product consumption. However, it has been shown that a higher level of
relationship quality led to positive attitude toward brand extension (Park, Kim, & Kim, 2002),
implying that consumers who perceive good relationship quality are more likely to buy products
which use the same brand name. In addition, it has been well documented that fans use team-
logoed items to support their teams or to show their affiliation with the teams (Cialdini et al.,
1976). In line with these research findings, I propose that fans who perceive a higher level of
relationship quality will intend to buy more licensed products.
Attendance
Increasing attendance is a primary objective of sport managers and marketers. Ticket sales
account for approximately 20% to 50% of the total revenues for professional teams in Major
40
League Baseball, the National Football League, the National Basketball League, and National
Hockey League (Badenhausen et al., 2007). Even higher proportions of total revenue are
generated from ticket sales for the teams at the collegiate level and minor leagues across various
types of sport (Howard & Crompton, 2005). In addition, the revenue from the sale of on-site
game day concessions, merchandising, and parking, which was $8.84 billion a year in the U.S., is
also contingent on attendance (Broughton, Lee, & Netheny, 1999). Accordingly, attendance or
intention of attendance has been the most frequently employed outcome variable in sport
management research. No researcher has investigated how relationship quality influences
attendance (or intention to attend a sporting event). However, previous literature on relationship
quality typically indicated that relationship quality had positive influence on consumer’s
intention to purchase. Hennig-Thurau and Klee (1997) suggested that relationship quality is a
predictor of repeated purchase behaviors. Some empirical evidence has been found as well.
Palmatier et al. (2006) reported that relationship quality accounted for an average of 52% of
variance in purchase intention in the 50 empirical studies in the consumer products context.
Moreover, Fournier (1994) argued that brand relationship quality is a better predictor of purchase
intention than brand attitude and satisfaction because brand relationship quality explained 61%
of variance in purchase intention, whereas brand attitude and satisfaction explained 37% and
52% of variance in purchase intention respectively. Derived from these findings, I expect that
fans who perceive a higher level of relationship quality intend to attend more games.
Moderators of Relationship Quality’s Influence on Sport Consumption Behavior
Once the relationship between the relationship quality and the outcome variables is
examined, the next question drawing our attention is how the direction and association of the
relationship changes across many different situations. For whom or in which situation is the
relationship between relationship quality and hypothesized outcome stronger or weaker? To
41
address this issue, the author will identify and empirically examine the influence of potential
moderators on the relationship between relationship quality and sport consumption behaviors. In
reviewing the literature, we focus on product category as well as psychographics and
demographics of sport consumers, which might alter the relationship quality-purchase intention
linkage.
Moderator Effects
A moderator is a third variable that changes the nature of the relationship between a
predictor and an outcome (Baron & Kenny, 1986). The moderator effect is also called an
interaction by which the effect of a predictor variable on an outcome variable changes as the
value of a third variable changes. Discovering important moderators of relationships between
predictors and outcome variables has been a vital theoretical and empirical issue in many social
and behavioral science disciplines (Aguinis, Boik, & Pierce, 2001; Judd, McClelland, &
Culhane, 1995). In addition, Cohen, Cohen, West, and Aiken (2003) stated that identification of
moderators is at the heart of theory in social science. Despite this continuing importance of
interaction effects, empirical research on hypothesized moderators has been limited. Such
scarcity of interaction effects application is not due to a lack of relevant research questions that
require examining interaction effects. Rather, this is due to the difficulties for the applied
researchers to properly implement the statistical technique to test the interaction effects (Frazier,
Tix, & Barron, 2004; Marsh, Wen, Hau, 2004; Rigdon, Schumacker, & Wothke, 1998). When
both the predictor variable and potential moderator are categorical variables, researchers can
easily test interaction effects with traditional analysis of variance (ANOVA) procedure. When
both predictor and potential mediator are continuous variables, regression procedures are
preferred over using ANOVA with an artificially dichotomized variable (Cohen et al., 2003).
When a predictor variable is a latent continuous variable and a potential moderator is an
42
observed categorical variable, a comparison of a model with cross-group equality constraints to a
model without cross-groups equality using structural equation modeling (SEM) can provide a
test of interaction effects (Klein, 2005). When each predicator and potential moderator is a latent
continuous variable, interaction effects can be tested by including the latent product variable in
the original model and testing if the latent product variable is statistically significant (Algina &
Moulder, 2001; Jöreskog, 2000; Kenny & Judd, 1984). Despite the widespread use of SEM
because of its advantages over traditional statistical procedures that assume no measurement
errors, there have been limited applications of SEM to test interaction effects of latent variables
in the sport management area. In particular, there is a lack of research on latent continuous
variable interactions in sport management despite its theoretical and practical importance.
Therefore, better understanding of methods to identify potential moderators and test an
interaction effect using SEM will make both a theoretical and practical contribution.
Moderators of Relationship Quality-Relationship Outcome Linkage
As mentioned in the earlier section of this chapter, academicians and practitioners
generally suggested that relationship quality positively influences consumers’ purchase
intentions and purchases (Hennig-Thurau & Klee, 1997; Palmatier et al., 2006, Fournier, 1994).
However, researchers also recognize that the strength of the associations varies with several
moderating factors. Anderson and Narus (1991) suggested that exchanges vary across a
continuum between pure transactional to pure relational. That is, relationships are not equally
important in all exchanges and therefore, the influence of the relationship between consumers
and sellers on outcomes differs by the importance of the relationships for those exchanges.
Palmatier et al. identified three contexts in which the relationship is more critical than in other
situations for a successful exchange. They stated that the customer-seller relationship is more
important (1) for service exchanges than for product-based exchanges; (2) for exchanges
43
between channel partners than for seller-customer exchanges (3) for business or industrial
markets than for consumer markets. Moreover, they found that relationship quality influences
seller objective performance significantly more for service, channels, and business market
exchanges than product-based, seller-customer, and consumer market exchanges. In addition,
Fournier noted that personality traits such as relationship style, interpersonal orientation, and
relationship value centrality moderated the strength of brand-consumer relationship. Fournier
further added category involvement, gender, age, and education as moderators. However, there is
limited research that empirically examined the effects of these hypothesized moderators on the
relationship between relationship quality and behavioral outcomes including purchase intentions.
Furthermore there is a lack of research that has explored the potential moderators of relationship
quality-behavioral outcome relationship in sport consumer behavior contexts. Therefore, an
objective of our study is to identify and empirically examine the potential factors that might
moderate the relationship between relationship quality and sport consumption behaviors. Based
on the literature, I identified the following potential moderators of relationship quality-sport
consumption behavior associations. This potentially moderating relationship is depicted in
Figure 2-8
Product Category
I choose classification of products as a potential moderator by which the impact of
relationship quality on sport consumption behaviors might differ. I differentiate the spectator
sport product from typical manufactured goods like toothpaste, computers, automobiles, etc.
Gladden and Sutton (2005) stated that spectator sport products are different from manufactured
goods in the following ways: (1) spectator sport is less tangible; (2) spectator sport is more
subjective; (3) spectator sport is more experiential; (4) spectator sport is publicly consumed and
spectator satisfaction is influenced by social facilitation; (5) the quality of spectator sport is less
44
consistent, less predictable, and less controllable; and (6) consumers are more involved in the
production. These aspects are consistent with the characteristics that distinguish services from
manufactured products (Zeithmal, Parasuraman, & Berry, 1985). According to Palmatier et al.
(2007), the inherent interactions in the production and consumption of a service might make the
consumer-seller relationship more important for services than manufactured products. In
addition, the intangibility, inconsistency, and unpredictability of the service offerings make
relationship quality a more critical criterion for consumers because evaluation of other aspects is
often ambiguous. Thus, I expect that the association between relationship quality and
consumption behaviors, behavioral intentions in my case, will be stronger for spectator sport
than general manufactured products.
Psychographic Factors
To reflect differences among consumers in psychological characteristics which govern the
association between relationship quality and sport consumption behaviors, I choose a number of
psychographic factors as potential moderators. Operating on Fournier’s (1994) suggestion that an
individual’s brand relationship is consistently influenced by the personality traits that affect the
individual’s interpersonal relationships, I propose relational personality traits as potential
moderators of relationship quality-sport consumption behavior associations. Following Fournier,
I will include and measure relationship styles, relationship drive, and general interpersonal
orientation in the current study. Based on Mathews (1986), Fournier identified three primary
relationship styles: independent, discerning, and acquisitive. Individuals who possess the
independent relationship style easily start a relationship but maintain a distance and simply
discontinue the relationship as the environment around their lives change. Individuals who
possess the discerning relationship style selectively and slowly develop new relationships and
maintain only a few strong relationships, but relationships are not easily influenced by the
45
individual’s life circumstances. Finally, the individual with an acquisitive relationship style is
readily able to initiate new relationships, but also values the development of strong relationships
and the retention of relationships. Consequently, these individuals tend to enlarge the number of
relationships throughout their life-span.
Reviewing Mathew’s (1986) descriptions of three relationship styles reveals that the
classification is based on two main characteristics: relationship development style and
relationship maintenance style. Figure 2-9 portrays Mathew’s relationship categories in a two
dimensional format. The horizontal dimension represents relationship development style. People
who score high on this dimension typically are not afraid of interaction with new people, easily
start a new relationship, quickly develop a relationship. The vertical dimension is based on
relationship maintenance style. People who are rated high on this dimension do not easily end
any relationship and tend to maintain long and strong relationships with others. Upper-right box,
upper-left box, and lower-right box in the matrix represent acquisitive, discerning, independent
relationship style respectively. Fournier (1994) found that relationship style, relationship drive,
and interpersonal orientation influenced the perceived importance of brand-consumer
relationships to the consumers, implying that those relational personality traits moderated the
impact of relationship quality on behavioral outcome. Similarly, I expect that individuals with
different personality traits that influence the interpersonal relationships carry those same
personality traits into the spectator-team relationship domain. Therefore, relationship style will
moderate associations between relationship quality and sport consumption behaviors.
Summary
In the preceding section, a considerable amount of the literature on relationship
marketing and relationship quality has been reviewed. Relationship marketing has grown
tremendously both in practice and academia. However, definitions of relationship marketing
46
have been varied across disciplines and contexts. Therefore, I briefly discussed the definitions of
relationship marketing and proposed a definition of relationship marketing in the sport
consumption context. Among the various academic and practical issues in relationship
marketing, I focus on relationship quality in this study. I selected seven constructs (i.e., trust,
commitment, relationship satisfaction, self-connection, love, intimacy, and reciprocity) as critical
components of relationship quality and discussed possible measurement models (i.e., general
relationship quality factor model, independent factor model, group factor model, second-order
hierarchical model, and modified second-order hierarchical model). Media consumption,
licensed-product consumption, and attendance were discussed as outcomes of relationship
quality. Motives were introduced as criterion explanatory variables to evaluate the predictive
value of relationship quality on those outcome variables. Furthermore, I identified product
category and relational personality traits as potential moderators of the association between
relationship quality and consumption behavior.
47
Figure 2-1. Graphical comparison of intimacy and self-connection
48
Figure 2-2. General relationship quality factor model
49
Figure 2-3. Independent factor model
50
Figure 2-4. Group factor model
51
Figure 2-5. Second-order hierarchical model
52
Figure 2-6. Modified second-order hierarchical model
53
Figure 2-7. Relationship between relationship quality and consumption behavior
54
Figure 2-8. Moderating effects on the relationship quality-consumption behavior association
55
Figure 2-9. Relationship style
56
Figure 2-10. Relationship between motives and consumption behavior
57
CHAPTER 3 METHODOLOGY
To obtain empirical data on relationship quality, sport consumption behavior, relational
personality traits, and demographics surveys were conducted both online and face-to-face. Data
analysis was performed using various statistical techniques such as Confirmatory Factor
Analysis (CFA), Structural Regression, and Multiple Sample Structural Equation Modeling. This
chapter presents the methodology used in this study in the following order: (1) Participants and
procedures (2) Instrumentation (3) Pilot study (4) Data analysis.
Participants and Procedures
The target population for this study was people 18 years of age or older who were affiliated
with the University of Florida. Potential respondents were selected using the judgmental
sampling method. This method is a type of non-probability sampling in which researchers select
a sample to be observed based on the researchers’ knowledge and judgment about the
population, its elements, and the purpose of the study. This type of sampling method is
considered to be a valid alternative to probability samplings when it is unrealistic to obtain a
truly random sample (which is true on many occasions in social science) and when the
researchers reasonably believe that the chosen sample is representative of the entire population
based on knowledge of the population (Babbie, 2007). Mail, telephone, face-to-face, and online
modes are the most frequently used methods in social science research. Each mode has its own
strengths and weaknesses in coverage, non-response rate, measurement quality, and cost. Mixed-
mode design is increasingly used as a way to reduce effects or bias of data collection modes on
the survey results while balancing cost (Groves, Fowler, Couper, Lepkowski, Singer, &
Tourangeau, 2004). The data were collected combining two major survey modes: face-to-face
self-administered and online self-administered.
58
Participants in face-to-face surveys were recruited via visiting undergraduate and graduate
classes, dining areas, and recreational sport facilities on campuses. After agreeing to participate
in the survey, participants were given a brief explanation of the purpose of the study and
instructions on filling out the survey. Then participants were asked to complete the questionnaire
about demographics, relationship quality, relational personality traits, and several behavioral
aspects of sport consumption behavior. In compliance with Institutional Review Board’s (IRB)
protocol, informed consent process was ensured. This process involved providing information on
planned procedure in language appropriate to the level of understanding of the participants and
then requesting voluntary participation in accordance with the Code of Federal Regulation. It
took approximately 15 minutes for a respondent to complete a survey. Data were collected from
356 people who participated in face-to-face self-administered survey. Of these, 51 surveys were
unusable, leaving a total of 305 usable responses.
Online survey participants were recruited by sending an email that contained an invitation
to participate in the online survey and a link to an Internet website on which the survey
questionnaire was posted. Email lists were obtained through class and college List-Serves and
university homepages. The purpose of the study, description of planned procedure in language
appropriate to the level of understanding of the participants, request of voluntary participation in
accordance with the Code of Federal Regulation, and brief instructions for completion of the
survey, was included in the first part of the questionnaire. Informed consent was obtained by the
respondents reading the cover letter and choosing to fill out the survey posted on the Web page.
The survey was closed one week after the invitation was sent. Participants received a thank you
notice after they completed the survey. The data was stored on a server provided by an online
survey company and the data was downloaded at the end of data collection. E-mail was sent to
59
2100 email addresses. Of these 2100 email addresses, 23 e-mails were returned as undeliverable,
leaving 2077 effective email addresses. A total of 254 responded, for an effective response rate
of 12%. Of these, 63 surveys were unusable, leaving 191 usable responses. Thus, a total of 496
were useable across both the face-to-face and web surveys.
Instrumentation
The questionnaire was comprised of four main parts: relationship quality constructs,
relational personality traits, relationship quality outcomes, and demographics. Items in each part
were randomly placed in order to avoid response bias from order effect. The instrumentation
process included the following steps: (1) item selection and modification, (2) expert review, and
(3) pilot study.
Item Development
Relying on previous literature, items for measuring seven relationship quality constructs,
four relationship outcomes, four relational personality trait constructs, and demographics were
selected and modified.
Relationship quality
The first part measuring relationship quality consisted of 7 subscales (Trust, Commitment,
Relationship Satisfaction, Self-connection, Intimacy, Love, and Reciprocity) represented by a
total of 52 items. Although most of the measures for the relationship quality constructs examined
in this study are available in the current literature, the items used to measure the constructs were
inconsistent across studies. The items that were believed to be most appropriate for this study
and had shown good psychometric properties were initially selected. Then, the selected items
were modified to suit the spectator sport context. Two versions of the relationship quality scale
were included in the survey. One version was designed to measure consumers’ perceived
60
relationship quality regarding a sport team (i.e., the UF Football team) and the other one focused
on a brand in the manufactured product industry (i.e., iPod).
Three items from Fletcher et al. (2000), 1 item from MacMillan et al. (2005), and 1 item
Morgan and Hunt (1994), respectively, were selected and those items were modified to measure
Trust. To measure Commitment, 3 items were chosen from Fletcher et al. and 2 items were
chosen from Fournier (1994). The items were revised. Three items from Fletcher et al. and 2
items from Spake, Beatty, Brockman, and Crutchfield (2003) were selected and the items were
reworded to measure Relationship Satisfaction. Four items from Fournier were adopted and then
those were modified to measure Self-connection. Three items from Fletcher et al., 3 items from
Fournier, and 3 items from Spake et al. were selected and those were reworded to measure
Intimacy. Two items from Fournier and 3 items from Fletcher et al. were chosen and modified to
measure the Love factor. In order to measure Reciprocity, 5 items from Uhl-Bien and Maslyn
(2003) were modified and 1 item was generated based on the literature on reciprocity.
The number of response categories can influence answers (Groves, et al., 2004). When too
few options are presented, the rating scale often cannot differentiate among participants with
different underlying judgments. When too many options are presented, participants might not be
able to accurately discern the difference between categories. Response theory literature generally
suggests that seven scale points is the best compromise (Krosnick & Fabrigar, 1997). Therefore,
all items were answered on a 7-point Likert-type scale. Point 1 on this scale demonstrated strong
disagreement with a statement, point 7 indicated strong agreement, and point 4 indicated neither
agreement nor disagreement.
Relationship quality outcome variables
The second part measuring relationship quality outcome variables included 4 subscales
with 12 items (Sport Media Consumption, Licensed Product Consumption, Attendance, and
61
Purchase). Measures for these constructs were readily obtainable from the current literature.
However, items used to measure the constructs had varied considerably across studies.
Therefore, items that were considered to be most suitable for the purpose of this study and have
had good psychometric properties in past research were initially selected. Then, the items were
reworded to be appropriate to the context of this study.
Two items from Johnson and Grayson (2005), 2 items from Kwon, Trail, and James
(2007), and 1 item from Fournier (1994) were chosen and the items were revised to measure
Purchase and Attendance Intention. In order to measure Licensed Product Purchase, 3 items from
Lee and Trail (2007) were adopted. Two items from Fink et al. (2002) and 1 item from Trail et
al. (2005) were chosen and those were modified to measure Media Consumption. Despite
widespread use of sport media consumption intention constructs in both academic research and
practice, of the studies to date that investigated sport consumption behavior through media, few
seem to use multiple indicators. Multiple-indicator measures of a construct are preferable
because any single-indicator measure is easily influenced by measurement error and using
multiple indicators tends to reduce the influence of the measurement error (Kline, 2005).
Therefore, a multiple-item measure of sport media consumption intention were developed and
used for this study. All of the relationship quality outcome items were measured using a 7-point
Likert-type scale response format ranging from strongly disagree (1) to strongly agree (7).
Personality traits
The fourth part measuring relational personality traits was comprised of 2 subscales
(Relationship Development Style and Relationship Maintenance Style) with 12 items. Mathews
(1986) identified three main relationship styles: independent, discerning, and acquisitive
relationship style. Following Mathews’ distinction, Fournier (1994) developed a scale to measure
relationship style. However, Fournier’s scale showed inadequate psychometric properties. For
62
this study, a scale was developed to measure relationship style based on Mathews’ criteria. One
item from Fournier (1994) was selected and 5 items were generated to measure Relationship
Development Style based on Mathews’ (1986) framework. Similarly, 2 items from Fournier were
selected and 4 items were created to measure Relationship Maintenance Style according to
Mathews.
Demographics
Items measuring demographic characteristics of participants were also included in the
questionnaire. The questions measured gender, age, and ethnicity.
Expert Review
The items were reviewed by a five-judge panel of scholars who have expertise on both the
concepts measured and methodological issues associated with survey research. The experts were
asked to review the survey items to evaluate whether their content was suitable for measuring the
intended constructs and also to identify if items contained the following problems commonly
associated with survey questionnaires (Grasser, Kennedy, Wiemer-Hastings, & Ottai, 1999): (1)
complex syntax, (2) working memory overload, (3) vague or ambiguous noun phrases, (4)
unfamiliar technical terms, (5) vague or imprecise predicate or relative term, (6) misleading or
incorrect presupposition, (7) unclear question category, (8) amalgamation of more than one
question category, (9) mismatch between the question category and the answer options, (10)
difficult to recall information, (11) respondent unlikely to know answer, and (12) unclear
question purpose. The survey questionnaire was finalized for a pilot study after adding new items
that were necessary but previously omitted, and refining or eliminating problematic items based
on the results of the expert review.
63
Pilot Study
A pilot study was conducted before the main study for the following purposes (Groves et
al., 2004): (1) to test appropriateness of the instruments; (2) to assess variability in outcome
measures that assisted in determining the sample size for main study; and (3) to identify potential
problems associated with proposed data analysis technique.
Methods
Participants and procedure
A total of 154 students enrolled in sport activity classes participated in the study. The
sample consisted of 51% male and 49% female. The average age of the participants was 21 years
old (M = 20.52, SD = 2.93) and slightly more than 50% of participants were white/non-Hispanic.
Face-to-face self-administered mode was utilized to collect the data. Standard survey procedure
was followed in accordance with IRB protocol. It took approximately 15 minutes for a
respondent to complete a questionnaire.
Instruments
The questionnaire consisted of four main parts (relationship quality constructs, relational
personality traits, relationship quality outcomes, and demographics) with 23 subscales and 117
items for the pilot study.
Data analysis
To evaluate the measurement models for relationship quality constructs, relationship
quality outcomes, and relational personality traits, five separate confirmatory factor analyses
were conducted on each group of factors using Mplus 5.1. Among various specialized software
packages for SEM, Mplus was used in both the pilot study and main study for the following
reasons: (1) Mplus offers several options to handle categorical (including Likert-type scale) and
non-normal data; (2) Mplus incorporates a model-based imputation method to manage missing
64
data (3) Mplus can model both continuous latent variables and categorical latent variables; (4)
Mplus can analyze multilevel SEM with complex sample data; (5) Mplus can correctly analyze a
correlation matrix using constrained estimation methods; (6) Mplus provides factor scores; (7)
Mplus offers extensive Monte Carlo facilities both for data generation and data analysis; and (8)
Syntax of Mplus is comparatively straightforward.
Results and discussion
Table 3-1 displays the loadings, Cronbach’s alpha, and AVE values of relationship quality
constructs for the UF Football team. The model fit the data poorly (Χ2/df = 984.26/506 = 1.95,
RMSEA = .10, CFI = .80, SRMR = .09, WRMR = 1.12). Five items were dropped based on the
assessment of factor loadings and theoretical relevance after the initial CFA. The revised model
showed improved fit (Χ2/df = 654.209/356 = 1.84, RMSEA = .09, CFI = .85, SRMR = .08,
WRMR = 0.98). The remaining items for the relationship quality factors showed adequate
reliability values in terms of Cronbach’s alpha values (α = .79 to .89) and Average Variance
Explained (AVE) values (.49 to .69) in the pilot study. Pairwise Χ2difference tests showed that
all correlations between factors were significantly different from 1.0, providing evidence for
discriminant validity. However, squared correlations between some pairs of factors were greater
than the AVE score of either factor (Table 3-2), indicating that people could not distinguish those
factors although those factors were theoretically different. Therefore, discriminant validity of
these factors was reexamined with larger sample in the main study.
Table 3-3 shows the loadings, Cronbach’s alpha, and AVE values of relationship quality
constructs for iPod. The model had poor fit for data (Χ2/df = 1020.80/506 = 2.02, RMSEA = .10,
CFI = .80, SRMR = .12, WRMR = 1.51). Five items were dropped after an initial CFA. The
remaining items showed good reliability values with Cronbach’s alpha values ranging from .78
to .92 and AVE values ranging from .53 to .72. Pairwise Χ2difference tests showed that all
65
correlations between factors were significantly different from 1.0, indicating discriminant
validity. However, squared correlations between some pairs of factors were greater than the AVE
score of either factor (Table 3-4), suggesting that people could not distinguish those factors
although those factors were theoretically different. Therefore, discriminant validity of these
factors was reexamined with a larger sample in the main study.
The measurement model for sport consumption behaviors fit the data well (Χ2/df =
70.56/59 = 1.19, RMSEA = .04, CFI = .99, SRMR = .04, WRMR = 0.50). Cronbach’s alpha
coefficients for sport consumption behavior subscales ranged from .87 to .95 and AVE values
ranged from .64 to .87, indicating good internal consistency and construct reliability (Table 3-5).
Table 3-6 displays the correlations between sport consumption behaviors factors. Although
pairwise Χ2difference tests showed that all correlations between factors were significantly
different from 1.0, the squared correlation between Word of Mouth and Media (r2 = .74) was
greater than AVE value for the Word of Mouth factor (AVE = .64). Therefore, discriminant
validity of the two factors was reexamined in the main study.
The measurement model for consumption behaviors (iPod) had good fit for data (Χ2/df =
21.68/13 = 1.67, RMSEA = .08, CFI = .98, SRMR = .03, WRMR = 0.40).Word of Mouth (α =
.88 and AVE = .66) and Purchase Intention (α = .93 and AVE = .82) also had good internal
consistency and good construct reliability (Table 3-7). Pairwise Χ2difference tests showed that all
correlations between factors were significantly different from 1.0. However, the squared
correlation between Word of Mouth and Purchase Intention (r2 = .81) was greater than AVE
value for the Word of Mouth (AVE = .66) factor (Table 3-8). Therefore, discriminant of the two
factors was reexamined in the main study.
66
Four items for relationship style subscales were dropped after an initial CFA. Table 3-9
displays the loadings, Cronbach’s alpha, and AVE values of two relationship style constructs.
The model fit the data poorly (Χ2/df = 110.78/53 = 2.09, RMSEA = .10, CFI = .87, SRMR = .12,
WRMR = 1.45). The revised model achieved good fit for the data (Χ2/df = 30.99/19 = 1.63,
RMSEA = .08, CFI = .97, SRMR = .06, WRMR = 0.81). Relationship Development (α = .88 and
AVE = .65) and Relationship Maintenance (α = .69 and AVE = .55) subscales showed adequate
internal consistency (Table 3-10). Pairwise Χ2difference test showed that the correlation between
Relationship Development and Relationship maintenance was significantly different from 1.0
and AVE values of both factor were greater than squared correlation (r2 = 21) between the two
factors (Table 3-11).
Through the pilot study a total 14 items were dropped based on the assessment of
psychometric properties and theoretical relevance of those items. The instrument for the main
study had 96 items: 30 items for relationship quality of the UF Football team, 30 items for
relationship quality of iPod, 20 items for relationship quality outcomes, and 6 items for
relationship style.
Data Analysis for Main Study
Data analysis was performed in four stages. First, descriptive statistics for the variables
used in the study was obtained. Second, data from the survey was screened and the critical
assumptions underlying the statistical techniques used in study were tested. Third, measurement
models for the constructs were analyzed followed by a structural model for relationship quality
and relationship quality outcomes. Finally, interaction effects of potential moderators were
examined.
67
Descriptive Statistics
Various descriptive statistics of the variables used in this study such as measures of
central tendency (e. g., mean, mode, median, etc.) and measures of variability (e. g., range,
variance, standard deviation, etc.) were obtained using SPSS 15.0 to describe the basic
characteristics of the data in this study.
Data Screening and Test of Assumption
Prior to the main analyses, all the variables were examined using various SPSS programs
and Mplus programs for accuracy of data entry, outliers, and fit between the characteristics of the
data and the critical assumptions of various SEM techniques used in this study.
Outliers in the variables were evaluated using extreme values output from the Explore
analysis. Elimination of case or variable, transformation, and score alteration is typically
considered to reduce the influence of outliers based on the nature of the outlier. Normality of the
observed variables was assessed through examination of histogram and summary descriptive
statistics using SPSS Descriptives. Multivariate normality was tested using Mardia’s (1985)
multivariate skewness and kurtosis coefficients and normalized estimates of the coefficients,
which were available through PRELIS 2.80. If Mardia’s Normalized Coefficient of both
skewness and kurtosis is significant, multivariate non-normality can be inferred. When the
normality assumption is violated enough to cause problems with the SEM analysis, hypothesized
models can be estimated with maximum likelihood estimation but tested with Satorra-Bentler
scaled chi-square statistic (SB χ2, 1994), which has been shown to perform reliably with medium
sample sizes (100 <N < 500) under condition of nonnormality (Bentler & Yuan, 1999; Curran,
West, & Finch, 1996; Hu, Bentler, & Kano, 1992). Accordingly, model fit indices depend on chi-
square statistic should be adjusted based on SB χ2.
68
Linearity of the observed variables was assessed by examining randomly selected pairs of
scatterplots using SPSS Graphs because it was not practical to examine all pairwise scatterplots
to evaluate linearity. When linearity assumptions are violated, data transformation can be used as
a remedy. The determinant of the input matrix was inspected to identify multicollinearity and
singularity in the data. The determinant of the input matrix is available in SPSS Factor Analysis
when it is requested. When serious multicollinearity or singularity occurs, the variables causing
the multicollinearity or singularity should be deleted or combined into a new composite variable.
Measurement Model
Separate confirmatory factor analyses were performed on five groups of constructs (i.e.,
relationship quality constructs for UF Football team and iPod, relationship outcomes for UF
football team and iPod, and relational personality traits) using Mplus 5.1 to evaluate the
measurement models. Goodness of fit indices used to evaluate overall fit of the model in the
current study were the CFI in conjunction with standard root mean squared residual (SRMR)
following Hu and Bentler (1999). A cutoff-value close to .95 or higher for CFI in combination
with a cutoff value close to (less than) .09 for SRMR was recommended (Hu & Bentler).
Additionally, the root-mean-square error of approximation (RMSEA), which is thought to reduce
problems with incremental fit indices (e.g., CFI) and absolute fit indices (e.g., GFI) according to
Brown and Cudeck (1992), was used. RMSEA values of less than .06 indicate good fit (Hu &
Bentler), values of .08 or less would represent reasonable fit and values higher than .10 indicate
poor fit (Brown & Cudeck). Furthermore, the weighted mean square residual (WRMR), which is
more appropriate with categorical data or non-normal continuous data (Muthén & Muthén,
2006), was used. WRMR values under 1.0 indicate good fit and smaller values demonstrate
better fit (Yu & Muthén, 2002).
69
The discrepancy matrix was also examined in conjunction with modification indices (MI)
statistics since clear misspecification cannot be discovered by the indexed fit of the composite
structural model and it is impossible to decide which elements of the composite hypothesis can
be considered satisfactory from the global goodness of fit indices alone (McDonald & Ho, 2002).
In addition, internal consistency values (Cronbach’s alpha coefficients) were utilized to examine
how well the items measuring a specific subscale were correlated with each other. Values greater
than .70 are considered to be adequate (Nunnally & Bernstein, 1994). Average Variance
Explained (AVE) values were employed to evaluate how well the items on a specific subscale
collectively explained the underlying construct’s variance. AVE values above .50 indicate that
the composite reliability of the construct is adequate (Fornell & Larcker, 1981). Discriminant
validity for each of the factors was tested through the procedure that involves χ2 difference test
between a model in which two individual factors are constrained to be 1.0 (i.e., the two factors
are perfectly correlated) and a model where the correlation between two factors are freely
estimated. If the unrestricted model without the unity constraints fits the model fit significantly
better, it might be inferred that the two factors are distinct in the population. AVE values were
also used to evaluate discriminant validity of the constructs. Results from the each analysis
above were considered collectively in reaching a final decision regarding which items and
factors to retain and which to eliminate.
To better understand how the relationship quality constructs were evaluated and structured,
the four models (general relation quality factor model, independent factor model, group factor
model, and second-order hierarchical model) discussed in the previous chapter were empirically
tested.
70
Structural Model
To examine the relationship between relationship quality constructs and sport
consumption behaviors, a structural regression (SR) model, which incorporated the five
relationship quality constructs (Trust, Commitment, Relationship Satisfaction, Self Connection,
and Reciprocity) and the three consumption behavior constructs (Media Consumption Intention,
Licensed-Product Consumption Intention, and Attendance Intention), was conducted using
Mplus 5.1
Moderating Effects
Two basic techniques have been developed to model moderating effects with continuous
observed variables using SEM. One approach, which is referred to as the product indicant
technique, was introduced by Kenny and Judd (1984). This approach generates a latent
interaction variable by multiplying pairs of observed variables and then incorporates the new
latent interaction variable in the structural model. Although this technique has provided a
valuable means to conduct structural equation interaction models, the application of the
technique to sport management research has been limited. The dearth of such application might
be due to the fact that (1) it is difficult for applied researchers to properly specify the nonlinear
constraints in the matrices and (2) multiplying of all pairs of observed variables to form latent
interaction variables possibly results in a too large and impractical model. Jöreskog (2000)
suggested a technique utilizing latent variable scores to test latent variable interactions in a
structural equation model. This method alleviated the problems listed above. Especially, using
this method can considerably reduce the number of indicators included in the model.
Schumacker (2002) compared the new approach to the product indicant approach and reported
that the new method yielded satisfactory parameter estimation.
71
Relationship Quality was the independent variable and Sport Consumption Behaviors was
the dependent variable in the main relationship. Interaction effects of Relationship Development
and Relationship Maintenance which were continuous latent variables, were tested following
Jöreskog’s (2000) latent score approach. Latent scores of the second-order Relationship Quality
construct, the second-order Sport Consumption Behaviors construct, and two relationship style
constructs were computed and saved through CFA using Mplus 4.21. Then, latent interaction
variables were created by multiplying the latent variable scores of Relationship Quality and each
relationship style construct. Lastly, multiple path analyses with two independent variables
(Relationship Quality and one of the relational style constructs), one latent product variable, and
Sport Consumption Behaviors were performed. Interaction effects can be inferred if a path
coefficient for the direct effect of the product variable on the dependent variable is statistically
significant.
72
Table 3-1. Summary results for measurement model of relationship quality (UF Football Team) in pilot study
Factors and items λ α AVE Trust .793 0.49
I trust the brand 0.718 I can count on the brand 0.661 The brand has integrity 0.694 The brand is reliable 0.735
Commitment .890 0.69 I am dedicated to the brand 0.796 I am faithful to the brand in spirit 0.744 I am devoted to the brand 0.918 I am committed to the brand 0.862
Relationship Satisfaction .858 0.55 My relationship with the brand is favorable 0.853 I am pleased with the relationship that I have with the brand 0.813 I am happy with my relationship with the brand 0.734 I am satisfied with my relationship with the brand 0.534
Self-Connection .855 0.61 The brand's image and my self-image are similar in a lot of ways 0.771 The brand and I have a lot in common 0.788 The brand reminds me of who I am 0.713 The brand is part of me 0.836
Love .891 0.62 I love this brand 0.696 I am passionate about this brand 0.893 I adore the brand 0.782 I am emotionally attached to the brand 0.760
Intimacy .808 0.49 I am very close to the brand 0.827 I am very familiar with the brand 0.587 I know a lot about the brand 0.688 I feel as though I really understand the brand 0.666
Reciprocity .833 0.55 The brand unfailingly pays me back when I do something extra for it 0.705 The brand constantly returns the favor when I do something good for it 0.752
The brand places my needs above its own needs 0.744 The brand gives me back equivalently what I have given them 0.711 The brand pays attention to what I get relative to what I give them 0.807
73
Table 3-2. Correlations among relationship quality constructs (UF Football Team) in pilot study 1 2 3 4 5 6 7 1. Trust 1.00 2. Commitment 0.77 1.00 3. Reciprocity 0.62 .50 1.00 4. Self-Connection 0.67 .89 .69 1.00 5. Love 0.81 .98 .45 .86 1.00 6. Intimacy 0.89 .87 .56 .83 .86 1.00 7. Satisfaction 0.92 .84 .45 .70 .87 .88 1.00
74
Table 3-3. Summary results for measurement model of relationship quality (iPod) in pilot study Factors and items λ α AVE
Trust .847 0.57 I trust the brand 0.798 I can count on the brand 0.881 The brand has integrity 0.482 The brand is reliable 0.793
Commitment .864 0.60 I am dedicated to the brand 0.798 I am faithful to the brand in spirit 0.728 I am devoted to the brand 0.765 I am committed to the brand 0.817
Relationship Satisfaction .921 0.72 My relationship with the brand is favorable 0.831 I am pleased with the relationship that I have with the brand 0.891 I am happy with my relationship with the brand 0.857 I am satisfied with my relationship with the brand 0.808
Self-Connection .854 0.60 The brand's image and my self-image are similar in a lot of ways 0.806 The brand and I have a lot in common 0.741 The brand reminds me of who I am 0.747 The brand is part of me 0.799
Love .892 0.61 I love this brand 0.662 I am passionate about this brand 0.860 I adore the brand 0.831 I am emotionally attached to the brand 0.749
Intimacy .776 0.53 I am very close to the brand 0.380 I am very familiar with the brand 0.821 I know a lot about the brand 0.914 I feel as though I really understand the brand 0.679
Reciprocity .857 0.58 The brand unfailingly pays me back when I do something extra for it 0.816 The brand constantly returns the favor when I do something good for it 0.804
The brand places my needs above its own needs 0.709 The brand gives me back equivalently what I have given them 0.687 The brand pays attention to what I get relative to what I give them 0.793
75
Table 3-4. Correlations among relationship quality constructs (iPod) in pilot study 1 2 3 4 5 6 7 1. Trust 1.00 2. Commitment .76 1.00 3. Reciprocity .52 .75 1.00 4. Self-Connection .60 .94 .71 1.00 5. Love .76 .98 .60 .82 1.00 6. Intimacy .39 .52 .40 .65 .36 1.00 7. Satisfaction .93 .78 .53 .63 .79 .51 1.00
76
Table 3-5. Summary results for measurement model of consumption behaviors (UF Football team) in pilot study
Factors and items λ α AVE Word of Mouth .877 0.64
I will tell other people about how good this brand is 0.750 I will encourage my friends and relatives to buy this brand 0.846 I will go out of my way to say positive things about this brand 0.738 I will recommend this brand whenever anyone seeks my advice 0.87
Attendance .951 0.87 I intend to attend the Gators Football team’s game(s) 0.907 The likelihood that I will attend the Gators Football team’s game(s) in the future is high 0.941
I will attend the Gators Football team’s game(s) in the future 0.949 Media .939 0.85
I will watch or listen to the Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 0.924
I will support the Gators Football team by watching or listening to Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 0.958
I will track the news on the Gators Football team through the media (e.g., TV, Internet, Radio, etc.) 0.875
Merchandise .950 0.87 In the future, purchasing Gators Football team licensed merchandise is something I plan to do 0.942
I am likely to purchase Gators Football team’s licensed merchandise in the future 0.900
In the future, I intend to purchase licensed merchandise representing the Gators Football team 0.950
77
Table 3-6. Correlations among sport consumption behaviors constructs in pilot study 1 2 3 4 1. Word of Mouth 1.00 2. Attendance .54 1.00 3. Media .50 .86 1.00 4. Merchandise .63 .66 .65 1.00
78
Table 3-7. Summary results for measurement model of purchase intention (iPod) in pilot study Factors and items λ α AVE Word of Mouth-iPod .883 .66
I will tell other people about how good this brand is 0.669 I will encourage my friends and relatives to buy this brand 0.888 I will go out of my way to say positive things about this brand 0.788 I will recommend this brand whenever anyone seeks my advice 0.879
Purchase Intention-iPod .928 .82 I will buy this brand’s products in the future 0.869 The likelihood that I will buy this brand in the future is high 0.928 I intend to purchase products from this brand 0.912
79
Table 3-8. Correlations among consumption behaviors constructs (iPod) in pilot study 1 2 1. Word of Mouth 1.00 2. Purchase .90 1.00
80
Table 3-9 Summary results for measurement model of relationship style (Initial model) Factors and items λ α AVE Relationship Development .881 .56
I start a new relationship casually rather than selectively 0.606 It usually takes only short time for me to make a new friend 0.754 I can easily make new friends 0.820 It seems that I develop a new relationship more easily than most people I know 0.796
I feel comfortable with meeting new people 0.838 It seems that I move through my life collecting new relationships all along the way 0.665
Relationship Maintenance .687 .38 I have a strong relationship with most of my friends 0.561 It is difficult for me to end any relationship with another 0.164 I continue a relationship with others for a long time 0.720 I can see that my core group of friends will stay close over the course of my life 0.811
It seems that I have more friends than most people I know 0.306 In general, relationships between my friends and me are very close 0.818
81
Table 3-10. Summary results for measurement model of relationship style Factors and items λ α AVE
Relationship Development .881 0.65 It usually takes only short time for me to make a new friend 0.761 I can easily make new friends 0.854 It seems that I develop a new relationship more easily than most people I know 0.790
I feel comfortable with meeting new people 0.810 Relationship Maintenance .687 0.55
I have a strong relationship with most of my friends 0.547 I continue a relationship with others for a long time 0.711 I can see that my core group of friends will stay close over the course of my life 0.837
In general, relationships between my friends and me are very close 0.822
82
Table 3-11. Correlations among relational personality constructs in pilot study 1 2 1. Relationship Development 1.00 2. Relationship Maintenance .46 1.00
83
CHAPTER 4 RESULTS
The results of the study are presented in the following order: (1) Descriptive statistics (2)
Data screening and test of assumptions (3) Measurement models (4) Structural models (5)
Moderating effects.
Descriptive Statistics
Demographics
Demographic characteristics of participants (N = 496) are depicted in Table 4-1. The
majority of the participants were women (64%). The average age of the participants was 25 years
old (M = 24.84, SD = 9.35) and 61% of participants were white/non-Hispanic.
Relationship Quality Variables
Descriptive statistics for relationship quality variables are presented in Tables 4-2 and 4-
3. The means of the relationship quality items for the UF Football team ranged from 2.75 to 5.32.
Standard deviations ranged from 1.25 to 1.85. The items for Love, Commitment, and
Relationship Satisfaction factor had the highest means on the 7-point Likert type scale. The items
for Reciprocity and Self-Connection had the lowest means. The item “I love the Gators Football
team” had the highest mean (M = 5.53, SD = 1.65) and the item “The Gators Football team pays
attention to what I get relative to what I give them” had the lowest mean (M = 2.75, SD = 1.46).
Consumption Variables
Table 4-4 and Table 4-5 display the descriptive statistics for relationship outcomes.
Means of all the items for UF Football were above 5.00 (4.0 mid-point) and ranged from 5.39
(SD = 1.56) for the item “I will track the news on the Gators Football team through the media
(e.g., TV, Internet, Radio, etc.)” to 5.91 (SD = 1.41) for the item “I will support the Gators
Football team by watching or listening to Gators Football team’s game(s) through the media
84
(e.g., TV, Internet, Radio, etc.).” Standard deviations for the items ranged from 1.41 to 1.68. The
means of items measuring Purchase Intention for iPod ranged from 4.53 (SD = 1.74) for the item
“I intend to buy iPod products” to 4.64 (SD = 1.63) for the item “The likelihood that I will buy
iPod products in the future is high.” Standard deviations ranged from 1.73 to 1.77.
Relationship Style Variables
Descriptive statistics for relationship style items are shown in Table 4-6. The item “In
general, relationships between my friends and me are very close” had the highest mean (M =
5.71, SD = 1.15) and the item “It seems that I develop a new relationship more easily than most
people I know” had the lowest mean (M = 4.40, SD = 1.48). Standard deviations for the items
ranged from 1.15 to 1.48.
Data Screening and Test of Assumptions for Structural Equation Modeling (SEM)
No standardized score for any variable was above 3.29 and no standardized score for any
variable was below -3.29, which were the suggested cut-off values for potential outliers
(Tabachnick & Fidell, 2007). There was evidence that both univariate and multivariate normality
assumptions for observed variables were violated. Distributions for fifty three out of seventy
observed variables were significantly skewed at p < .01 and the distributions for 39 of 70
variables showed significant kurtosis at p < .01. Moreover, Mardia’s Normalized Coefficient of
both skewness (z = 41.69) and kurtosis (z = 14.13) was significant, p < .01. For dealing with the
non-normality, Satorra-Bentler scaling method (SB χ2, 1994) was used for the SEM analyses in
the current study. Consequently, model fit indices depending on chi-square statistic were
adjusted based on SB χ2. All randomly selected pairs of variables appeared to be linearly related.
The determinants of all the matrices used in this study were much larger than 0, indicating there
was no extreme multicollinearity or singularity in those matrices.
85
Measurement Models
Relationship Quality (UF Football team)
Validation of the measure
The measurement model (Figure 2-4), which specified seven latent factors to be correlated
with each other, was tested. Six items were dropped based on the results from the initial CFA,
leaving 23 observed variables for 7 latent factors. Factor loadings, theoretical relevance and
parsimoniousness of the model were considered collectively in reaching a final decision
regarding which items to retain and which to eliminate. Table 4-7 shows factor loadings,
Cronbach’s alpha coefficients, and AVE values for the initial CFA. After the modification, the
model showed mediocre fit (Χ2/df = 812.61/209 = 3.89, RMSEA = .08, CFI = .92, SRMR = .06,
WRMR = 1.78). Cronbach’s alpha coefficients for relationship quality factors ranged from .79
for Trust to .93 for Commitment (Table 4-8), indicating good internal consistency according to
the suggested cut-off values of .70 (Nunnally, 1978). The AVE values ranged from .55 for
Relationship Satisfaction to .81 for Commitment, indicating good construct reliability. Pairwise
Χ2difference tests showed that all correlations between factors were significantly different from
1.0, providing evidence for discriminant validity.
However, squared correlation between Commitment and Love (r2 =.94) was greater than
the AVE score of both factors (.81 and .66 respectively). Table 4-9 displays the correlations
between relationship quality factors. Observed variables for Intimacy and Love had the largest
standardized residuals in the discrepancy matrix, which indicated the relationships between those
variables and other variables were misspecified. In addition, MI statistics suggested that the
model could be improved most if the constraint that fixed paths from items for Intimacy to other
factors to be zero were freely estimated (MI > 20), implying that variances in the items were
86
accounted for by more than one factor. Due to the inadequate psychometric properties of the
Love and Intimacy scales, those two factors were dropped from the model.
The revised measurement model achieved good fit for data (Χ2/df = 319.65/94 = 3.40,
RMSEA = .07, CFI = .95, SRMR = .04, WRMR = 1.24). Factor loadings, Cronbach’s alpha
coefficients, and AVE are reported in Table 4-10. The internal consistency measures for all the
factors were greater than .70 with the lowest of .79 for Relationship Satisfaction and the highest
of .93 for Commitment. The AVE values for all the factors were also greater than .50 with the
lowest of .55 for Relationship Satisfaction and the highest of .81 for Commitment. Pairwise
Χ2difference tests showed that all correlations between factors were significantly different from
1.0. In addition, no squared correlation between two factors was greater than the AVE score of
either factor with the exception of the squared correlation between Trust and Self-Connection (r2
=.67), which was greater than AVE value for Trust and Self-Connection (AVE = .66 for both) by
only third decimal place, indicating discriminant validity. Correlations among latent factors are
reported in Table 4-11. Highly intercorrelated but still distinct factor structure provided
preliminary evidence that the first-order factors converge into a higher-order relationship quality
factor.
Structure of the relationship quality constructs
In testing the proposed second-order hierarchical model, comparisons were made with
three alternative models discussed in the previous chapter. The modified second-order factor
model was not tested because two factors (i.e., Love and Intimacy), which were hypothesized to
tap into a second-order latent variable, Affective Relationship Quality, were dropped after the
construct validation procedure. Therefore, the hierarchical model specifying two second-order
latent factors was no longer a plausible model. Model fit information for the second-order
hierarchical model and the alternative models is displayed in Table 4-12. The hypothesized
87
second-order hierarchical factor model (Figure 2-5) fit the data well (Χ2/df = 380.43/99 = 3.84,
RMSEA = .05, CFI = .94, SRMR = .05, WRMR = 1.54). Two alternative models did not pass the
desired criteria for fit. The independent factor model that specified the relationship quality
factors to be completely uncorrelated with each other had poor fit (Χ2/df = 1435.81/104 = 13.81,
RMSEA = .16, CFI = .69, SRMR = .38, WRMR = 11.08). In addition, the general factor model
that hypothesized a global relationship factor that collapsed across all 16 indicators fit the data
poorly (Χ2/df = 1080.09/104 = 10.39, RMSEA = .14, CFI = .77, SRMR = .09, WRMR = 2.46).
The group factor (measurement) model that allowed the correlation for all pairs of first-
order factors to be freely estimated achieved good fit as noted above (Χ2/df = 319.65/94 = 3.40,
RMSEA = .07, CFI = .95, SRMR = .04, WRMR = 1.24). Figure 4-1 depicted the relationship
between first-order relationship quality factors and the second-order relationship quality factor.
Loadings for the first-order factors on the second order factors were significantly different from
zero in every case and all standardized loadings were greater than .70 ranging from .71 for
Reciprocity to .92 for Self-Connection.
Relationship Quality (iPod)
Validation of the measure
The measurement (Figure 2-5) model, which hypothesized seven latent factors to be
correlated with each other, was tested. Six items were dropped based on the results from the
initial CFA, leaving 23 observed variables for 7 latent factors. A final decision regarding which
items to retain and which to eliminate were made by taking into account the factor loadings,
theoretical relevance and parsimoniousness collectively. Table 4-13 displays factor loadings,
Cronbach’s alpha coefficients, and AVE values for the initial CFA. After the modification, the
model showed marginal fit (Χ2/df = 1023.75/209 = 4.90, RMSEA = .09, CFI = .89, SRMR = .06,
WRMR = 1.85). All subscales showed good internal consistency with Cronbach’s alpha
88
coefficients ranging from .81 for Self-Connection to .90 for Commitment (Table 4-14). All
subscales had good construct reliability with the AVE values ranged from .60 for Trust to .74 for
Commitment. Pairwise Χ2difference tests showed that all correlations between factors were
significantly different from 1.0.
However, the squared correlation between Commitment and Love (r2 =.90) was greater
than the AVE score of both factors (.81 and .67 respectively). In addition, squared correlation
between Commitment and Intimacy (r =.92) was also greater than the AVE score of both factors
(.81 and .61 respectively). The correlations between relationship quality factors are presented in
Table 4-15. Observed variables measuring Self-Connection and Intimacy contributed to the
largest standardized residuals in the discrepancy matrix, indicating the relationship between
those variables and other variables was misspecified. In addition, on the basis of the MI statistics,
the model could be improved most if the paths from items for Self-Connection and Intimacy to
other factors were freely estimated (MI > 20), which implied that more than one factor accounted
for variance in the items. Therefore, Love, Intimacy, and Self-Connection were dropped from the
model due to the inadequate psychometric properties of the subscales.
The final model fit data well (Χ2/df = 188.76/59 = 3.20, RMSEA = .07, CFI = .96, SRMR
= .05, WRMR = 1.38). Factor loadings, Cronbach’s alpha coefficients, and AVE values are
presented in Table 4-16. The internal consistency measures for all the factors exceeded .70 with
the lowest of .80 for Trust and the highest of .90 for Commitment. The AVE values for all the
factors also exceeded .50 with the lowest of .60 for Trust and the highest of .75 for Commitment.
Pairwise Χ2difference tests showed that all correlations between factors are significantly different
from 1.0 and no squared correlation between two factors was greater than the AVE score of
either factor except the square correlation of Trust and Commitment (r2 =.75) and Trust and
89
Relationship Satisfaction (r2 =.75). Correlations among latent variables are reported in Table 4-
17.
Structure of the relationship quality constructs
Table 4-18 shows model fit information for the second-order hierarchical model and the
alternative models. The modified second-order factor model was not tested because three factors
(i.e., Love, Intimacy, and Self-Connection) were dropped after the construct validation
procedure. Therefore, the hierarchical model specifying two second-order latent factors was no
longer a plausible model. The hypothesized second-order hierarchical factor model achieved
good fit (Χ2/df = 200.53/61 = 3.29, RMSEA = .07, CFI = .96, SRMR = .05, WRMR = 1.48). Two
alternative models showed inadequate fit. The independent factor model that set the relationship
quality factors to be completely uncorrelated with each other fit data poorly (Χ2/df = 1049.03/65
= 16.14, RMSEA = .18, CFI = .71, SRMR = .36, WRMR = 10.44). Similarly, the general factor
model that specified a global relationship factor that collapsed across all 16 indicators had poor
fit as well (Χ2/df = 790.10/65 = 12.16, RMSEA = .15, CFI = .78, SRMR = .10, WRMR = 2.82).
The group factor (measurement) model that let the correlations for all pairs of first-order
factors to be freely estimated showed good fit as noted above (Χ2/df = 188.76/59 = 3.20, RMSEA
= .07, CFI = .96, SRMR = .05, WRMR = 1.38). Both the second-order hierarchical factor model
and the group factor model had satisfactory fit. The relationship between first-order relationship
quality factors and the second-order relationship quality factor is displayed in Figure 4-2.
Loadings for the first-order factors on the second order factors were significantly larger than zero
in all cases. With the exception of Reciprocity (λ = .61), all standardized loadings exceeded .70
ranging from .61 for Reciprocity to .96 for Trust.
90
Relationship Outcome (UF Football team)
Table 4-19 displays factor loadings, Cronbach’s alpha coefficients, and AVE values for
the initial CFA for sport consumption behaviors. The model showed good fit (Χ2/df = 149.14/48
= 3.11, RMSEA = .07, CFI = .98, SRMR = .03, WRMR = 0.76). The subscales had good internal
consistency and good construct reliability with Cronbach’s alpha ranging from .82 to 96 and the
AVE values ranging from .62 to .89. Pairwise Χ2difference tests showed that all correlations
between factors are significantly different from 1.0. However, the AVE value for Word of Mouth
(AVE = .62) was smaller than squared correlation between Word of Mouth and all other factors
(Table 4-20), indicating lack of discriminant validity of the Word of Mouth factor with the other
factors in the model. Therefore, Word of Mouth was dropped from the model. Factor loadings,
Cronbach’s alpha coefficients, and AVE values from the CFA on sport consumption behaviors
are presented in Table 4-21. The revised model fit data very well (Χ2/df = 42.52/24 = 1.77,
RMSEA = .04, CFI = .99, SRMR = .02, WRMR = 0.40). The results from the CFA also
indicated good internal consistency for observed variables with Cronbach’s alpha coefficients
ranging from .89 for Media to .96 for Attendance and the AVE values ranging from .74 for
Media to .89 for Attendance. Pairwise Χ2difference tests showed that all correlations between
factors were significantly different from 1.0 and no squared correlation between two factors was
greater than the AVE score of either factor providing an evidence of discriminant validity for the
items. Table 4-22 shows correlations among latent factors. It can be inferred that the first-order
factors converge into a higher-order factor from the highly intercorrelated but still distinct factor
structure in conjunction with the theoretical justification discussed in the earlier chapter.
Model fit information for the second-order hierarchical model and the alternative models
are reported in 4-23. Hypothesized second-order hierarchical factor model showed the good fit
for data (Χ2/df = 42.52/24 = 1.77, RMSEA = .04, CFI = .99, SRMR = .02, WRMR = 0.40). Two
91
alternative models demonstrated inadequate fit based on goodness of fit indices and Χ2/df values.
The independent factor model that fixed the correlation among the three sport consumption
variables to be zero had poor fit (Χ2/df = 544.88/27 = 20.18, RMSEA = .20, CFI = .84, SRMR =
.42, WRMR = 9.27). Likewise, the general factor model that hypothesized 9 indicators directly
measuring global sport consumption behavior factor fit data poorly (Χ2/df = 958.03/27 = 35.48,
RMSEA = .26, CFI = .71, SRMR = .10, WRMR = 1.94).
The group factor (measurement) model that was specified to freely estimate the correlation
for all pairs of first-order factors achieved good fit as noted above (Χ2/df = 42.52/24 = 1.77,
RMSEA = .04, CFI = .99, SRMR = .02, WRMR = 0.40). Figure 4-3 depicts the relationship
between three first-order latent factors and the second-order sport consumption behavior factor.
Loadings for the first-order factors on the second order factors were significantly larger than zero
in all cases and all the standardized loadings were greater than .70 with the lowest of .78 for
Attendance and the highest of .89 for Merchandise.
Relationship Outcome (iPod)
Table 4-24 shows factor loadings, Cronbach’s alpha coefficients, and AVE values for the
initial CFA for consumption behaviors. The model showed good fit (Χ2/df = 10.87/8 = 1.35,
RMSEA = .03, CFI = .99, SRMR = .01, WRMR = 0.25). Word of Mouth (α = .89 and AVE =
.74) and Purchase Intention (α = .95 and AVE = .86) subscales showed adequate internal
consistency and construct reliability. Pairwise Χ2difference test showed that correlations between
Word of Mouth and Purchase Intention are significantly different from 1.0. However, AVE value
for Word of Mouth (AVE = .74) was smaller than squared correlation (r2 = .76) between Word
of Mouth and Purchase Intention (Table 4-25), indicating lack of discriminant validity of the
Word of Mouth factor. Therefore, Word of Mouth was dropped from the model. The
measurement model consisted of a single factor (Purchase Intention), which was a saturated
92
model, and thus fit the data perfectly (Χ2/df = 0.00/0 = 0.00, RMSEA = .00, CFI = 1.00, SRMR =
0.00, WRMR = 0.00). Cronbach’s alpha value for items measuring Purchase Intention was .95
and AVE value was .86, indicating good internal consistency and construct reliability (Table 4-
26).
Relationship Personality
Two items were dropped based on the results from the initial CFA, leaving 6 observed
variables for 2 latent factors. The measurement model for relationship style, which consisted of
the Relationship Development and Relationship Maintenance factors, yielded a good fit (Χ2/df =
9.12/8 = 1.14, RMSEA = .02, CFI = 0.99, SRMR = 0.01, WRMR = 0.32). Cronbach’s alpha
coefficients and AVE values for Relationship Development (α = .88 and AVE = .72) and
Relationship Maintenance (α = .84 and AVE = .67) were greater than widely used cut-off
criteria, indicating that items measuring relationship style had good reliability (Table 4-27).
Pairwise Χ2difference tests showed that correlation between Relationship Development and
Relationship Maintenance was significantly different from 1.0 and the squared correlation
between the two factors was greater than the AVE value for either factor, providing evidence for
discriminate validity. The correlation between the two factors was .44 in the sample (Table 4-
28).
Structural Models
The hypothesized model that examined the relationship between relationship quality and
sport consumption for the UF Football team is depicted in Figure 4-4. The model fit the data well
(Χ2/df = 712.66/266 = 2.68, RMSEA = .06, CFI = 0.95, SRMR = 0.06, WRMR = 1.79). The
second-order Relationship Quality factor significantly influenced the second-order Sport
Consumption Behavior factor (standardized γ = .82, z = 13.00). Relationship Quality explained
67 % of the variance in Sport Consumption Behaviors.
93
The hypothesized model that examined the relationship between relationship quality and
purchase intention for iPod is in Figure 4-5. The model had good fit (Χ2/df = 307.34/99 = 3.10,
RMSEA = .07, CFI = 0.96, SRMR = 0.05, WRMR = 1.53). The path coefficient between the
second-order Relationship Quality factor and the second-order Sport Consumption Behavior
factor was significant (standardized γ = .80, z = 12.50) and Relationship Quality explained 64 %
of the variance in Purchase Intention.
Moderating Effects
Relationship Development
The model that represented the hypothesized interaction effect of Relationship
Development on the relationship between Relationship Quality and Sport Consumption
Behaviors is depicted in Figure 4-6. The model yielded satisfactory fit (Χ2/df = 887.98/364 =
2.43, RMSEA = .05, CFI = 0.95, SRMR = 0.06, WRMR = 1.68). The path coefficient from the
product term of Relationship Quality and Relationship Development to Sport Consumption
behaviors was not significant (standardized γ = -.02, z = -0.50) and the product term explained
less than 1% of variance in Sport Consumption Behaviors, which indicated no significant
interaction effect.
The model that examined the interaction effect of Relationship Development on the
relationship between Relationship Quality and Purchase Intention for iPod is presented in Figure
4-7. The model achieved good fit for (Χ2/df = 424.16/161 = 2.63, RMSEA = .06, CFI = 0.96,
SRMR = 0.05, WRMR = 1.41). The path coefficient from the product term of Relationship
Quality and Relationship Development to Purchase Intention was not significant (standardized γ
= -.06, z = -1.1) and the product term explained less than 1% of variance in Sport Consumption
Behaviors. Therefore, no significant interaction effect was found.
94
Relationship Maintenance
Figure 4-8 shows the model that specified the interaction effect of Relationship
Maintenance on the relationship between Relationship Quality and Sport Consumption
Behaviors. The model had good fit (Χ2/df = 846.08/364 = 2.32, RMSEA = .05, CFI = 0.95,
SRMR = 0.06, WRMR = 1.66). The product term of Relationship Quality and Relationship
Maintenance did not significantly influence Sport Consumption Behaviors (standardized γ = -
.03, z = -1.07) and less than 1% of variance in Sport Consumption Behaviors was explained by
the product term, indicating no significant interaction effect.
Figure 4-9 depicts the model that investigated the interaction effect of Relationship
Maintenance on the relationship between Relationship Quality and Purchase Intention for iPod.
The model fit data well (Χ2/df = 354.71/161 = 2.20, RMSEA = .05, CFI = 0.96, SRMR = 0.05,
WRMR = 1.39). The product term of Relationship Quality and Relation Maintenance did not
significantly influence Purchase Intention (standardized γ = -.09, z = -1.70) and the product term
accounted for less than 1% of variance in Purchase Intention. Therefore, no significant
interaction effect was inferred.
95
Table 4-1. Demographic characteristics of participants Variable Group n % Gender Male 180 36.29 Female 316 63.71 Age 18-21 269 55.24 22-25 110 22.59 26-30 42 8.62 30+ 66 13.55 Ethnicity American Indian/Alaskan Native 5 1.02 Asian 30 6.10 Black 36 7.32 Hawaiian/Pacific Islander 5 1.02 Hispanic/Non-White 20 4.07 White/Hispanic 83 16.87 White/Non-Hispanic 304 61.79 Other 9 1.83
96
Table 4-2. Descriptive statistics for relationship quality (UF Football) Factors and items M SD Trust I can count on the Gators Football team 4.39 1.52 I trust the Gators Football team 4.63 1.25 The Gators Football team is reliable 4.14 1.45 Commitment I am committed to the Gators Football team 4.75 1.80 I am devoted to the Gators Football team 4.85 1.78 I am dedicated to the Gators Football team 4.83 1.77 Reciprocity The Gators Football team unfailingly pays me back when I do something extra for it 3.27 1.48 The Gators Football team gives me back equivalently what I have given them 3.61 1.58 The Gators Football team constantly returns the favor when I do something good for it 3.15 1.42 The Gators Football team pays attention to what I get relative to what I give them 2.75 1.46 Self-Connection The Gators Football team’s image and my self-image are similar in a lot of ways 3.31 1.50 The Gators Football team is part of me 4.00 1.85 The Gators Football team and I have a lot in common 3.38 1.57 Love I adore the Gators Football team 4.81 1.70 I am passionate about the Gators Football team 5.11 1.65 I am emotionally attached to the Gators Football team 4.32 1.84 I love the Gators Football team 5.32 1.65 Intimacy I am very familiar with the Gators Football team 5.03 1.62 I know a lot about the Gators Football team 4.97 1.56 I am very close to the Gators Football team 3.80 1.73 Relationship Satisfaction I am pleased with the relationship that I have with the Gators Football team 4.78 1.33 My relationship with the Gators Football team is favorable 4.90 1.48 I am satisfied with my relationship with the Gators Football team 4.76 1.34
97
Table 4-3. Descriptive statistics for relationship quality (iPod) Factors and items M SD Trust iPod is reliable 5.26 1.26 I trust iPod 4.08 1.62 I can count on iPod 4.27 1.59 Commitment I am committed to iPod 4.15 1.79 I am devoted to iPod 3.47 1.76 I am dedicated to iPod 3.22 1.75 Reciprocity iPod pays attention to what I get relative to what I give them 3.01 1.53 iPod constantly returns the favor when I do something good for it 2.60 1.43 iPod unfailingly pays me back when I do something extra for it 2.55 1.42 iPod gives me back equivalently what I have given them 2.99 1.50 Self-Connection iPod is part of me 2.89 1.75 iPod and I have a lot in common 2.65 1.52 iPod’s image and my self-image are similar in a lot of ways 2.61 1.49 Love I love iPod 4.08 1.78 I am passionate about iPod 3.31 1.71 I adore iPod 3.33 1.78 I am emotionally attached to iPod 2.68 1.64 Intimacy I am very close to iPod 3.57 1.80 I am very familiar with iPod 4.70 1.72 I know a lot about iPod 4.18 1.77 Relationship Satisfaction I am pleased with the relationship that I have with iPod 4.46 1.47 I am satisfied with my relationship with iPod 4.13 1.61 My relationship with iPod is favorable 4.01 1.51
98
Table 4-4. Descriptive statistics for relationship outcomes (UF Football) Factors and items M SD Attendance I intend to attend the Gators Football team’s game(s) 5.55 1.76 The likelihood that I will attend the Gators Football team’s game(s) in the future is high 5.67 1.77 I will attend the Gators Football team’s game(s) in the future 5.63 1.70 Media I will track the news on the Gators Football team through the media (e.g., TV, Internet, Radio, etc.) 5.39 1.56 I will watch or listen to the Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 5.77 1.45 I will support the Gators Football team by watching or listening to Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 5.91 1.41 Merchandise I am likely to purchase Gators Football team’s licensed merchandise in the future 5.46 1.63 In the future, purchasing Gators Football team licensed merchandise is something I plan to do 5.47 1.66 In the future, I intend to purchase licensed merchandise representing the Gators Football team 5.45 1.68
99
Table 4-5. Descriptive statistics for relationship outcomes (iPod) Factors and items M SD Purchase Intention-iPod The likelihood that I will buy iPod products in the future is high 4.64 1.77 I intend to buy iPod products 4.53 1.74 I will purchase iPod products in the future 4.46 1.73
100
Table 4-6. Descriptive statistics for consumption behaviors (iPod) Factors and items M SD Relationship Development It seems that I develop a new relationship more easily than most people I know 4.40 1.48 I can easily make new friends 5.32 1.32 It usually takes only short time for me to make a new friend 4.94 1.44 Relationship Maintenance I can see that my core group of friends will stay close over the course of my life 5.32 1.39 I have a strong relationship with most of my friends 5.69 1.19 In general, relationships between my friends and me are very close 5.71 1.15
101
Table 4-7. Summary results for initial measurement model of relationship quality (UF Football, Seven-Factor Model)
Factors and items λ α AVE Trust .848 0.59The Gators Football team has integrity .603 I trust the Gators Football team .840 The Gators Football team is reliable .773 I can count on the Gators Football team .827 Commitment .924 0.76I am committed to the Gators Football team .912 I am devoted to the Gators Football team .906 I am dedicated to the Gators Football team .878 I am faithful to the Gators Football team in spirit .774 Reciprocity .843 0.51The Gators Football team unfailingly pays me back when I do something extra for it .654
The Gators Football team places my needs above its own needs .574 The Gators Football team gives me back equivalently what I have given them .744
The Gators Football team constantly returns the favor when I do something good for it .849
The Gators Football team pays attention to what I get relative to what I give them .779
Self-Connection .877 0.64The Gators Football team reminds me of who I am .749 The Gators Football team’s image and my self-image are similar in a lot of ways .759
The Gators Football team is part of me .828 The Gators Football team and I have a lot in common .851 Love .908 0.71I adore the Gators Football team .803 I am passionate about the Gators Football team .898 I am emotionally attached to the Gators Football team .840 I love the Gators Football team .838 Intimacy .876 0.64I am very familiar with the Gators Football team .836 I know a lot about the Gators Football team .814 I feel as though I really understand the Gators Football team .788 I am very close to the Gators Football team .773 Relationship Satisfaction .813 0.52I am happy with my relationship with the Gators Football team .632 I am pleased with the relationship that I have with the Gators Football team .724
My relationship with the Gators Football team is favorable .781 I am satisfied with my relationship with the Gators Football team .735
102
Table 4-8. Summary results for measurement model of relationship quality (UF Football, Seven-Factor Model)
Factors and items λ α AVE Trust .856 0.66 I can count on the Gators Football team 0.823 I trust the Gators Football team 0.773 The Gators Football team is reliable 0.843 Commitment .927 0.81 I am committed to the Gators Football team 0.915 I am devoted to the Gators Football team 0.911 I am dedicated to the Gators Football team 0.874 Reciprocity .839 0.58 The Gators Football team unfailingly pays me back when I do something extra for it 0.654
The Gators Football team gives me back equivalently what I have given them 0.743
The Gators Football team constantly returns the favor when I do something good for it 0.858
The Gators Football team pays attention to what I get relative to what I give them 0.767
Self-Connection .849 0.66 The Gators Football team’s image and my self-image are similar in a lot of ways 0.738
The Gators Football team is part of me 0.851 The Gators Football team and I have a lot in common 0.850 Love .909 0.72 I adore the Gators Football team 0.802 I am passionate about the Gators Football team 0.899 I am emotionally attached to the Gators Football team 0.842 I love the Gators Football team 0.839 Intimacy .845 0.67 I am very familiar with the Gators Football team 0.876 I know a lot about the Gators Football team 0.850 I am very close to the Gators Football team 0.731 Relationship Satisfaction .792 0.55 I am pleased with the relationship that I have with the Gators Football team 0.684
My relationship with the Gators Football team is favorable 0.819 I am satisfied with my relationship with the Gators Football team 0.711
103
Table 4-9. Correlations among relationship quality constructs (UF Football) 1 2 3 4 5 6 7 1. Trust 1.00 2. Commitment 0.77 1.00 3. Reciprocity 0.69 .49 1.00 4. Self-Connection 0.81 .80 .73 1.00 5. Love 0.77 .97 .50 .80 1.00 6. Intimacy 0.60 .80 .40 .70 .81 1.00 7. Satisfaction 0.79 .78 .58 .72 .73 .70 1.00
104
Table 4-10. Summary results for measurement model of relationship quality (UF Football, Five-Factor Model)
Factors and items λ α AVE Trust .856 0.66 I can count on the Gators Football team 0.817 I trust the Gators Football team 0.774 The Gators Football team is reliable 0.847 Commitment .927 0.81 I am committed to the Gators Football team 0.900 I am devoted to the Gators Football team 0.913 I am dedicated to the Gators Football team 0.887 Reciprocity .839 0.57 The Gators Football team unfailingly pays me back when I do something extra for it 0.653
The Gators Football team gives me back equivalently what I have given them 0.743
The Gators Football team constantly returns the favor when I do something good for it 0.857
The Gators Football team pays attention to what I get relative to what I give them 0.768
Self-Connection .849 0.66 The Gators Football team’s image and my self-image are similar in a lot of ways 0.739
The Gators Football team is part of me 0.851 The Gators Football team and I have a lot in common 0.849 Relationship Satisfaction .792 0.55 I am pleased with the relationship that I have with the Gators Football team 0.676
My relationship with the Gators Football team is favorable 0.828 I am satisfied with my relationship with the Gators Football team 0.702
105
Table 4-11. Correlations among relationship quality constructs (UF Football) 1 2 3 4 5 1. Trust 1.00 2. Commitment .78 1.00 3. Reciprocity .69 .48 1.00 4. Self-Connection .82 .80 .73 1.00 5. Satisfaction .78 .79 .57 .73 1.00
106
Table. 4-12. Goodness of Fit indices and Χ2/df values for the hypothesized and alternative models (UF Football)
Model RMSEA CFI SRMR WRMR Χ2/df General .14 .77 .09 2.46 1080.09/104 = 10.39 Independent .16 .69 .38 11.08 1435.81/104 = 13.81 Group .07 .95 .04 1.24 319.68/94 = 3.40 Second-Order .08 .94 .05 1.54 380.43/99 = 3.84
107
Table 4-13. Summary results for initial measurement model of relationship quality (iPod, Seven-Factor Model)
Factors and items λ α AVE Trust .824 0.55 iPod is reliable .618 I trust iPod .808 iPod has integrity .653 I can count on iPod .871 Commitment .904 0.71 I am committed to iPod .807 I am devoted to iPod .850 I am dedicated to iPod .891 I am faithful to iPod in spirit .811 Reciprocity .879 0.60 iPod pays attention to what I get relative to what I give them .676 iPod constantly returns the favor when I do something good for it .858 iPod unfailingly pays me back when I do something extra for it .780 iPod gives me back equivalently what I have given them .794 iPod places my needs above its own needs .755 Self-Connection .860 0.62 iPod is part of me .725 iPod and I have a lot in common .828 iPod reminds me of who I am .790 iPod’s image and my self-image are similar in a lot of ways .813 Love .889 0.67 I love iPod .774 I am passionate about iPod .865 I adore iPod .845 I am emotionally attached to iPod .791 Intimacy .864 0.61 I am very close to iPod .821 I feel as though I really understand iPod .706 I am very familiar with iPod .796 I know a lot about iPod .791 Relationship Satisfaction .872 0.63 I am pleased with the relationship that I have with iPod .750 I am satisfied with my relationship with iPod .797 My relationship with iPod is favorable .848 I am happy with my relationship with iPod .768
108
Table 4-14. Summary results for measurement model of relationship quality (iPod, Seven-Factor Model)
Factors and items λ α AVE Trust .809 0.60 iPod is reliable 0.627 I trust iPod 0.805 I can count on iPod 0.874 Commitment .898 0.74 I am committed to iPod 0.839 I am devoted to iPod 0.870 I am dedicated to iPod 0.878 Reciprocity .862 0.62 iPod pays attention to what I get relative to what I give them 0.690 iPod constantly returns the favor when I do something good for it 0.884 iPod unfailingly pays me back when I do something extra for it 0.788 iPod gives me back equivalently what I have given them 0.775 Self-Connection .807 0.61 iPod is part of me 0.742 iPod and I have a lot in common 0.816 iPod’s image and my self-image are similar in a lot of ways 0.774 Love .889 0.67 I love iPod 0.785 I am passionate about iPod 0.865 I adore iPod 0.843 I am emotionally attached to iPod 0.782 Intimacy .843 0.61 I am very close to iPod 0.861 I am very familiar with iPod 0.736 I know a lot about iPod 0.730 Relationship Satisfaction .829 0.61 I am pleased with the relationship that I have with iPod 0.775 I am satisfied with my relationship with iPod 0.755 My relationship with iPod is favorable 0.819
109
Table 4-15. Correlations among relationship quality constructs (iPod) 1 2 3 4 5 6 7 1. Trust 1.00 2. Commitment .87 1.00 3. Reciprocity .54 .56 1.00 4. Self-Connection .72 .82 .84 1.00 5. Love .85 .95 .62 .88 1.00 6. Intimacy .84 .96 .46 .74 .89 1.00 7. Satisfaction .87 .80 .61 .70 .79 .78 1.00
110
Table 4-16. Summary results for measurement model of relationship quality (iPod, Four-Factor Model)
Factors and items λ α AVE Trust .809 0.60 iPod is reliable 0.626 I trust iPod 0.803 I can count on iPod 0.876 Commitment .898 0.75 I am committed to iPod 0.851 I am devoted to iPod 0.881 I am dedicated to iPod 0.86 Reciprocity .862 0.62 iPod pays attention to what I get relative to what I give them 0.694 iPod constantly returns the favor when I do something good for it 0.875 iPod unfailingly pays me back when I do something extra for it 0.788 iPod gives me back equivalently what I have given them 0.783 Relationship Satisfaction .829 0.61 I am pleased with the relationship that I have with iPod 0.77 I am satisfied with my relationship with iPod 0.757 My relationship with iPod is favorable 0.824
111
Table 4-17. Correlations among relationship quality constructs (iPod) 1 2 3 4 1. Trust 1.00 2. Commitment .87 1.00 3. Reciprocity .54 .55 1.00 4. Satisfaction .87 .78 .62 1.00
112
Table. 4-18. Goodness of fit indices and Χ2/df values for the hypothesized and alternative models (UF Football team)
Model RMSEA CFI SRMR WRMR Χ2/df General .15 .78 .10 2.82 790.10/65 = 12.16 Independent .18 .71 .36 10.44 1049.03/65 = 16.14 Group .07 .96 .05 1.38 188.76/59 = 3.20 Second-Order .07 .96 .05 1.48 200.53/61 = 3.29
113
Table. 4-19. Summary results for measurement model of sport consumption behaviors (with Word of Mouth)
Factors and items λ α AVE Word of Mouth .823 0.62 I will encourage my friends and relatives to attend the Gators Football team’s game(s) 0.724
I will recommend the Gators Football team whenever anyone seeks my advice 0.737
I will tell other people about how good Gator Football team is 0.895 Attendance .959 0.89 I intend to attend the Gators Football team’s game(s) 0.918 The likelihood that I will attend the Gators Football team’s game(s) in the future is high 0.971
I will attend the Gators Football team’s game(s) in the future 0.940 Media .890 0.74 I will track the news on the Gators Football team through the media (e.g., TV, Internet, Radio, etc.) 0.763
I will watch or listen to the Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 0.883
I will support the Gators Football team by watching or listening to Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 0.928
Merchandise .952 0.87 I am likely to purchase Gators Football team’s licensed merchandise in the future 0.911
In the future, purchasing Gators Football team licensed merchandise is something I plan to do 0.937
In the future, I intend to purchase licensed merchandise representing the Gators Football team 0.949
114
Table 4-20. Correlations among consumption behaviors constructs 1 2 3 4 1. Word of Mouth 1.00 2. Attendance .80 1.00 3. Media .87 .67 1.00 4. Merchandise .84 .70 .76 1.00
115
Table. 4-21. Summary results for measurement model of sport consumption behaviors Factors and items λ α AVE
Attendance .959 0.89 I intend to attend the Gators Football team’s game(s) 0.915 The likelihood that I will attend the Gators Football team’s game(s) in the future is high 0.974
I will attend the Gators Football team’s game(s) in the future 0.938 Media .890 0.74 I will track the news on the Gators Football team through the media (e.g., TV, Internet, Radio, etc.) 0.759
I will watch or listen to the Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 0.886
I will support the Gators Football team by watching or listening to Gators Football team’s game(s) through the media (e.g., TV, Internet, Radio, etc.) 0.928
Merchandise .952 0.87 I am likely to purchase Gators Football team’s licensed merchandise in the future 0.911
In the future, purchasing Gators Football team licensed merchandise is something I plan to do 0.938
In the future, I intend to purchase licensed merchandise representing the Gators Football team 0.948
116
Table 4-22. Correlations among sport consumption behaviors constructs
1 2 3 1. Attendance 1.00 2. Media .66 1.00 3. Merchandise .70 .76 1.00
117
Table. 4-23. Goodness of fit indices and Χ2/df values for the hypothesized and alternative models for sport consumption behaviors
Model RMSEA CFI SRMR WRMR Χ2/df General .26 .71 .10 1.94 958.03/27 = 35.48 Independent .20 .84 .42 9.27 544.88/27 = 20.18 Group .04 .99 .02 .40 42.52/24 = 1.77 Second-Order .04 .99 .02 .40 42.52/24 = 1.77
118
Table 4-24. Summary results for measurement model of consumption behaviors (iPod) with Word of Mouth
Factors and items λ α AVE Word of Mouth-iPod .893 .74
I will recommend iPod whenever anyone seeks my advice 0.810 I will tell other people about how good iPod is 0.826 I will encourage my friends and relatives to buy iPod’s product(s) 0.934
Purchase Intention-iPod .950 .86 The likelihood that I will buy iPod products in the future is high 0.911 I intend to buy iPod products 0.943 I will purchase iPod products in the future 0.931
119
Table 4-25. Correlations among consumption behaviors constructs (iPod) with Word of Mouth 1 2 1. Word of Mouth 1.00 2. Purchase .87 1.00
120
Table 4-26. Summary results for measurement model of purchase behavior (iPod) Factors and items λ α AVE
Purchase Intention-iPod .95 .86 The likelihood that I will buy iPod products in the future is high 0.912 I intend to buy iPod products 0.938 I will purchase iPod products in the future 0.936
121
Table 4-27. Summary results for measurement model of relationship style Factors and items λ α AVE
Relationship Development .879 .72 It seems that I develop a new relationship more easily than most people I know 0.742
I can easily make new friends 0.920 It usually takes only short time for me to make a new friend 0.876 Relationship Maintenance .842 .67 I can see that my core group of friends will stay close over the course of my life 0.671
I have a strong relationship with most of my friends 0.870 In general, relationships between my friends and me are very close 0.892
122
Table 4-28. Correlation between relationship style constructs 1 2 1. Relationship Development 1.00 2. Relationship Maintenance .44 1.00
123
Figure 4-1. Second-order hierarchical model (UF Football)
124
Figure 4-2. Second-order hierarchical model (iPod)
125
Figure 4-3. Second-order model for sport consumption behavior
126
Figure 4-4. Structural regression of relationship quality and sport consumption behaviors
127
Figure 4-5. Structural regression of relationship quality and sport consumption behaviors
128
Figure 4-6. Interaction effect of relationship development on relationship between relationship
quality and sport consumption behaviors
129
Figure 4-7. Interaction effect of relationship development on relationship between relationship
quality and purchase intention for iPod
130
Figure 4-8. Interaction effect of relationship maintenance on relationship between relationship
quality and sport consumption behaviors
131
Figure 4-9. Interaction effect of relationship maintenance on relationship between relationship
quality and purchase intention for iPod
132
CHAPTER 5 DISCUSSION
The primary purpose of this dissertation was to provide a theoretically-based and
empirically tested account of the relationship quality between team and sport consumers and the
role of relationship quality in sport consumption behaviors. To achieve these goals, first,
measurement scales for relationship quality, consumption behavior, and relationship personality
were developed relying on literature on relationship quality across various contexts. Second, the
measurement scales were tested and validated through assessment of essential psychometric
properties of the scales based on results from multiple CFAs. Third, the structural nature of the
relationship quality and consumption behaviors constructs were explored through model
comparisons. Next, the relationship between relationship quality and consumption behaviors was
examined by conducting structural regression (SR) models. Lastly, the potential moderating
effects of relationship personality on relationship quality-consumption behaviors relationship
were tested with SR models that incorporated interaction terms. This section will begin with
discussion on the results from the various analyses throughout the dissertation. Next, conceptual
and theoretical contributions of the research findings from this study will be discussed and then
managerial implication of relationship framework in sport context will be followed. Lastly,
limitations of the study will be addressed as well as suggestions for future research.
Validation of the Measures
Relationship Quality Constructs (UF Football Team)
A main goal of this dissertation was to investigate the following research question: How
should relationship quality between a team and its sport consumers be conceptualized and
measured? This objective has been largely achieved through developing a scale that measures the
relationship quality between team and sport consumer and initially validating the scale. The
133
results from this study provide evidence that the final scale (after modifications) possesses
adequate psychometric properties: (1) Content validity was established by the review of
literature, expert review, and a test of content validity; (2) Internal consistency values for all
constructs were greater than a widely accepted cutoff criterion; (3) Construct reliability was
evidenced by high AVE values for all constructs; and (4) Significant results from the tests of the
difference from unity for all pairs of constructs suggested adequate discriminant validity.
The literature on relationship quality suggested that there are seven relation quality
dimensions: Trust, Commitment, Reciprocity, Self-Connection, Relationship Satisfaction, Love,
and Intimacy. However, the empirical results from the data referent to the UF Football team
provide support for a five-factor model consisting of Trust, Commitment, Reciprocity, Self-
Connection, and Relationship Satisfaction. Love and Intimacy were initially regarded as distinct
dimensions of relationship quality but the results indicated that the two factors lacked
discriminant validity. This result is inconsistent with previous research, which typically viewed
Love and Intimacy as distinct concepts (Barnes, 1997; Fletcher, Simpson, & Thomas, 2000;
Fournier, 1994; Monga, 2002; Nicholson et al., 2001; Pawle & Cooper, 2006; Smit et al., 2007)
One primary explanation for the lack of discriminant validity of the two factors lies in a
gap between semantic or theoretical distinction made by researchers and actual distinction made
by respondents. This discrepancy might occur because the respondents were not sufficiently
involved in the evaluation process to be able to differentiate the items purported to measure
distinct concepts or because the respondents were not capable of making sophisticated
distinctions between those concepts as researchers were. If the discrepancy resulted from the lack
of sufficient attention or involvement of participants, it can be reduced by improved survey
methods. For example, the number of questions in the survey can be reduced or incentives can be
134
offered to participants. Then, discriminant validity of the construct should be reexamined. If the
discrepancy stemmed from the respondents’ inherent incapability to differentiate the concepts, it
can be concluded that those constructs are redundant in explaining relationship quality although
they might be representing theoretically distinct concepts. Therefore, more empirical studies are
needed to identify the source of the weak or poor discriminant validity for some factors (e.g.,
Love and Intimacy) proposed in this study and determine the factors that are essential to measure
the relationship quality between teams and sport consumers.
Relationship Quality Constructs (iPod)
The results from the data referent to iPod indicate that the proposed scale for relationship
quality possessed acceptable content validity, internal consistency, and construct reliability.
Initially, a seven-factor model, which incorporated Trust, Commitment, Reciprocity, Self-
Connection, Relationship Satisfaction, Love, and Intimacy, was proposed. However, the
empirical results from the iPod data support a four-factor model, which exclude Love, Intimacy,
and Self-Connection; leaving Trust, Commitment, Reciprocity, and Relationship Satisfaction.
Similar to the data specific to the UF Football team, respondents did not differentiate the items
purported to measure Love, Intimacy, and Self-Connection constructs from items intended to
measure the other constructs.
This result is inconsistent with previous studies, which regarded Love, Intimacy, and Self-
Connection as distinct concepts (Barnes, 1997; Fletcher, Simpson, & Thomas, 2000; Fournier,
1994; Monga, 2002; Nicholson et al., 2001; Pawle & Cooper, 2006; Smit et al., 2007). Again,
this lack of discriminant validity of the three factors can be explained by the gap between a
theoretical distinction made by the researcher and the actual distinction made by the respondents.
This disagreement might be caused by the respondents’ lack of involvement or an inherent
incapability to distinguish the concepts. Therefore, the source of the discrepancy should be
135
investigated through more empirical studies to determine the essential factors to explain the
nature of the relationship quality.
Interestingly, examination of modification indices (MI) reveals that the respondents had
greater difficulty distinguishing the items intended to measure Self-Connection from the items
purported to measure the other constructs in the iPod case compared to the UF Football team
case. In addition, the overall mean score for three Self-Connection items in the iPod version was
2.77, which was noticeably lower than the overall mean score for three Self-Connection items (M
= 3.56) in the UF Football team version. Taken together, these results suggest that the Self-
Connection concept was more applicable in explaining the relationship between the UF Football
team and sport consumers than the relationship between iPod and consumers. This finding is
consistent with the previous research that argued that identification with teams, which paralleled
with Self-Connection in this study, was one of main characteristics differentiating sport
consumption behaviors from general consumption behaviors (Cialdini, 1976; Sloan, 1989;
Gladden & Sutton, 2008; Madrigal, 1995, 2003; Trail et al. 2003; Trail et al., 2005; Wann &
Branscombe, 1993). However, only one brand (i.e., UF Football and iPod) from each product
category (i.e., sport and general manufactured products) was investigated in this study.
Therefore, the finding should be interpreted with caution due to its limited generalizability.
Sport Consumption Behaviors and Relationship Style
After the initial CFA, the AVE value for Word of Mouth was smaller than the squared
correlation between Word of Mouth and all other factors, indicating the lack of discriminant
validity of the Word of Mouth factor with the other factors in the model. The lack of
discriminant validity and the parsimoniousness of the model were considered collectively in
reaching a final decision to eliminate the Word of Mouth construct. The results from this study
indicate that the final scale for sport consumption behaviors possessed good content validity,
136
internal consistency, construct reliability, and discriminant validity. This result augments the
previous attempts (Fink et al., 20022005; Lee & Trail, 2007Trail et al., 2005) in developing
scales to measure sport consumption behaviors.
The scale measuring relationship style initially consisted of 12 items; however, 6 items
were dropped based on the assessment of factor loadings and theoretical relevance throughout
the pilot study and the initial CFA in the main study. The final scale for relationship style, which
consisted of the Relationship Development and Relationship Maintenance factors, had good
content validity, internal consistency, construct reliability, and discriminant validity.
Structural Nature of Relationship Quality
UF Football Team
With the scale developed to measure individual components of relationship quality (Trust,
Commitment, Reciprocity, Self-Connection, and Relationship Satisfaction) in hand, the
structural nature of the constructs was explored by comparing and testing four models (general
relationship quality factor model, independent factor model, group factor model, and second-
order hierarchical model), which differed as to how the individual relationship quality constructs
were associated with each other. Although a modified second-order factor model was initially
considered, it was eliminated because two factors (i.e., Love and Intimacy), which were
hypothesized to tap into a second-order latent variable, Affective Relationship Quality, were
dropped after the construct validation procedure. Therefore, the hierarchical model specifying
two second-order latent factors was no longer a tenable model.
The results from the data referent to UF Football team clearly do not support the general
relationship quality factor model, which hypothesized that distinct relationship quality facets did
not exist, but a global relationship quality factor should represent all individual factors. This is
consistent with previous research, which typically identified Trust, Commitment, Reciprocity,
137
Self-Connection, and Relationship Satisfaction as distinct dimensions of relationship quality
(Fletcher et al., 2000; Morgan & Hunt, 1984; Fournier, 1998; Roberts et al., 2003). Therefore,
the rejection of the general relationship quality factor model is both theoretically and empirically
justified. The results also do not provide support for the independent factor model, which
proposed that all the relationship quality dimensions were completely independent. This result is
in line with previous research findings, which reported that the five relationship dimensions were
correlated with each other (Garbarino & Johnson, 1999; Mogan & Hunt, 1984; Nicholson et al.,
2001; Stern, 1997; Uhl-Bien & Maslyn, 2003). Hence, the independent factor model is excluded
from further consideration.
The group factor model, which specify that relationship quality constructs were related to
each other but had no higher-order relationship quality construct, fit the data well. This result is
consistent with previous research, which viewed that individual relationship quality constructs
were related but still distinct. The second-order hierarchical model, which was a major focus of
this study, posited that first-order latent relationship quality factors were indicators measuring
underlying higher-order relationship quality factor. The model also fit the data well. The result
supports the notion that people make evaluative judgments on individual relationship quality
constructs, which were related but different domains, consistently based on a higher-order factor
of overall relationship quality (DeWulf et al., 2001; Fletcher et al., 2000; Fournier, 1994; Lages,
Lages, & Lages, 2005).
Although the second-order hierarchical factor model and the group factor model fit the
data adequately, the second-order model was accepted as the most tenable model for the
following reasons. First, the high intercorrelations between the different aspects of relationship
quality supported the notion that these individual constructs are indicators of a more general,
138
higher-order latent relationship-quality construct. Second, the structural part of the group factor
model was saturated and always fits the data perfectly. The saturated model is less desirable
because it cannot be rejected by data and therefore, there is no way to empirically test or confirm
the plausibility of the model (Raykov & Marcoulides, 2000). Next, although one primary interest
of this study was to understand the structural nature of the relationship quality constructs, the
group factor model does little to explain how the relationship quality constructs are structured.
Finally, the hierarchical nature of the relationship quality constructs naturally cause
multicollinearity between individual factors. The multicollinearity can make it difficult to
evaluate the contribution of each factor as a relationship quality facet in predicting consumption
behavior in the ensuing structural regression analysis. When multicollinearity is detected, it is
recommended to use a composite measure or scale (Tabachnick and Fidell, 2007). Therefore, the
second-order factor model specifying a second-order factor, which is a composite measure of
first-order relationship quality factors, is preferable in addressing the problem associated with the
multicollinearity between first order factors.
Although the second-order hierarchical model was chosen over the group factor model for
the ensuing SR analyses in this study, it should be noted that the two models are rather
complementary than competitive. In fact, the group factor model only specifies that the first
order constructs are correlated with each other to some extent. Due to the unconstrained nature of
the structural specification of the model, the group factor model does not either empirically or
theoretically contradict the second-order hierarchical model, which attempts to provide more
detailed explanation for the correlation between the first order factors. Thus, the group factor
model could be considered as a basic model to evaluate how well the individual latent constructs
are measured by observed variable and initially explore how those constructs are structurally
139
related with each other when a specific structure that would account for the relationships among
the first order factors are not clearly determined (Kline, 2005; Rindskopf & Rose, 1988). With
the basic model in hand, plausible structural specifications based on theory (e.g., second-order
hierarchical structure in this study) can be investigated to explain the correlations among first
order factors in a more specific and parsimonious way than the broader specification for a
general factor model would do.
iPod
The results from the data referent to iPod clearly did not support the general relationship
quality factor model and independent factor model. This result is consistent with previous
research, which viewed that the four relationship quality constructs (i.e., Trust, Commitment,
Reciprocity, Relationship Satisfaction) were related but distinct (Fletcher et al., 2000; Morgan &
Hunt, 1984; Fournier, 1998; Nicholson et al., 2001; Roberts et al., 2003; Stern, 1997; Uhl-Bien
& Maslyn, 2003). Both the second-order hierarchical factor model and the group factor model fit
the data well. However, the second-order model was selected as the most tenable model for the
same reasons discussed in the previous section.
Sport Consumption Behaviors (UF Football)
Similar to relationship quality, it is evidenced that the general relationship quality factor
model and independent factor model did not fit the data well. This finding is in line with the
previous research, which regarded the attendance, media consumption, and licensed merchandise
purchase as related but distinct aspects of sport consumption behavior (Fink et al., 2002; Garcio-
Harrolle, 2007; Trail et al. 2003; Trail et al., 2005). Both the second-order hierarchical factor
model and the group factor model yielded the same predicted correlations and have equal fit
statistics including goodness of fit indices and model chi-square value. Therefore, it cannot be
determined which model should be preferred over the other based on the model fit. In the current
140
study, the second-order model was chosen as the most suitable model over group factor model
for the same tenets given as a basis for the decision regarding which model to be accepted among
the alternative models for relationship quality.
Outcomes of Relationship Quality
One of the main objectives of this dissertation is to answer the following question: Does
relationship quality significantly and meaningfully explain sport consumption behaviors? To
achieve this objective, first the measurement scale for relationship quality was developed and
tested as discussed in the earlier section. Then, the predictive capacity of relationship quality was
assessed using intentions for three sport consumer behaviors of interest (i.e. Attendance, Media
Consumption, and Licensed Merchandise Purchase) as dependent variables. As expected, results
from the SR model (for the data referent to UF Football team) indicated that the second-order
latent Relationship Quality factor significantly influenced the second-order latent Sport
Consumption Behaviors factor, explaining 67% of its variance. This result was consistent with
the Fournier’s (1994) finding that relationship quality was a major predictor of behavioral
dependence. Fournier found that customers who perceived a high quality relationship with a
brand or company did not only purchase more products from the brand or the company but also
expand scope, diversity and frequency of brand-related or company-related activities. This
behavioral dependence might explain the finding from the current study that showed the fans
who perceived a higher level of relationship quality intended to consume sport through media.
Next, the result is also in line with the Park et al.’s (2002) finding that a higher level of
relationship quality resulted in positive attitude toward brand extension. That is, consumers who
perceived good relationship quality were more likely to buy products which used the same brand
name. Accordingly, the current study showed that sport consumers who perceived good
relationship quality intended to buy more team licensed products. Finally, the result from this
141
study supports the author’s proposition that sport consumers who perceive a higher level of
relationship quality intend to attend more games. This result is in agreement with previous
research, which found that relationship quality was a salient predictor of various aspects of
consumption behaviors. Hennig-Thurau and Klee (1997) suggested that relationship quality
significantly influenced repeat purchase behaviors. Some empirical evidence has been reported
as well (Reynolds & Beaty, 1999). Palmatier et al. (2006) found that relationship quality
explained an average of 52% of variance in purchase intention in a meta-analysis using 50
empirical studies in the consumer products context. In addition, Fournier (1994) suggested that
brand relationship quality was a superior predictor of purchase intention to brand attitude and
satisfaction because brand relationship quality accounted for 61% of variance in purchase
intention, while brand attitude and satisfaction accounted for 37% and 52% of variance in
purchase intention.
The relationship between relationship quality for iPod and Purchase Intention was also
assessed. Similar to the results from the data referent to the UF Football team, Purchase Intention
for iPod was significantly affected by the second-order latent Relationship Quality factor and a
large proportion of the variance (64%) in Purchase Intention was accounted for by Relationship
Quality. This result confirms previous literature, which found a strong association between
relationship quality and various consumption behaviors of interest (Crosby et al., 1990; De Wulf
et al., 2001; Doney & Cannon, 1997; Hennig-Thurau et al., 2002; Sirdeshmukh et al, 2002;
Palmatier et al., 2006; Reynolds & Beaty, 1999).
Moderators of Relationship Quality-Consumption Association
Once a significant relationship between relationship quality and consumer behavior
intentions had been found, the following question was raised: Does the nature of association
between relationship quality and intentions differ by the psychographic characteristics of
142
consumers? Relationship Development style and Relationship Maintenance style were selected
as potential moderators and the influence of the two constructs on the linkage between
relationship quality and consumption behaviors intention was investigated.
The strength of the associations between relationship quality and sport consumption
behaviors intention did not vary with the proposed moderators. This is inconsistent with
Fournier’s (1994) finding that relational personality traits moderated the influence of relationship
quality on behavioral outcomes. The lack of an interaction effect might provide an evidence that
relationship quality is an important predictor of sport consumption behavior regardless the
psychographic characteristics of sport consumers. However, the finding could be sample specific
and should be interpreted with caution. Therefore, the finding from this study should be
replicated with samples varying in demographic, psychographic, and socioeconomic
characteristics. Similarly, the association between the Relationship Quality and Purchase
Intention was not moderated by Relationship Maintenance and Relationship Development.
Implications of the Research
The present research has both academic and practical implications. In this section,
conceptual and theoretical implications are discussed. Then managerial implications follow.
Conceptual and Theoretical Implications
In this dissertation, a relationship quality framework for sport consumption behavior was
proposed for a better understanding of the relationship between sport consumers and sport teams.
This study makes a contribution to the current literature in a number of ways.
First, the author investigated the nature of relationship quality perceived by sport
consumers and developed a conceptual model consisting of five critical constructs in sport
consumption context: Trust, Commitment, Reciprocity, Self-Connection, and Relationship
Satisfaction. There are few studies that incorporate the constructs under relationship quality
143
framework. Therefore, the development of the conceptual model will help researchers
understand the nature of the relationship between sport consumers and the team with more
comprehensive ideas of relationship quality. Moreover, this study extended sport management
literature by applying relationship marketing theories to the sport consumer behavior realm. Both
relationship marketing and sport management research can benefit from the validation of the
current knowledge about relationship marketing within sport consumption contexts and the
integration of new research findings from this study.
Second, this study provides an empirical examination of the relationship quality
framework in the sport consumption context. While the current studies on relationship marketing
that exist in the sport management area have advanced the conceptual understandings of
relationship marketing (Bee & Kahle, 2006; Cousens et al., 2006; McDonald & Milne, 1997;
Tower et al., 2006), few studies empirically examine relationship marketing theories applied to
the relationship between sport organizations and their relationship partners. This study collected
empirical evidence for what were previously only assumptions suggesting that the relationship
metaphor was applicable to sport consumer behaviors and suggesting that relationship quality
was a critical predictor of sports consumption behaviors. These empirical findings extend our
understanding of relationship marketing in a sport consumption context beyond a mere argument
or grounded theory.
Next, this study attempts to develop and test a scale to measure relationship quality
perceived by sport consumers. A review of the extant work reveals that there are numerous
scales to measure relationship quality in various contexts. However, no scale was developed to
measure the quality of sport consumer-team relationship. A relationship quality scale for sport
consumption behavior was developed through a rigorous scale development process including an
144
extensive review of literature, expert review, a content validity check, and a pilot study. Then,
the scale was initially validated, showing that the scale exhibited good content validity, internal
consistency, and construct reliability. A psychometrically sound scale developed in this study
will help researchers move forward in their understanding of the relationship between sport
consumers and teams in particular.
Managerial Implications
Sport managers are now more interested in developing and maintaining relationships with
their consumers. In addition, sport consumers are willing to engage in relationship with teams.
However, relationship marketing practices in sport teams are still rudimentary. The findings from
this dissertation also have some managerial implications.
For sport managers, the findings from this study validate the widely-held assumption in
practice that good relationships with sport consumers is a critical factor for a successful sport
business. Managerial decisions based on the allocation of resources for relationship marketing
depends on evidence of its capability to yield meaningful performance outcomes. Sport
managers need to know the payoff to be obtained from the investment on cultivating the
relationship with their consumers is valuable. This study showed that when sport consumers
perceive that they have a good relationship with a sport team, they are more willing to attend
games, buy team licensed merchandise, and consume sport contents related to the team through
media. Moreover, the strength of the association between relationship quality and sport
consumption behaviors was substantial. In sum, the findings from this study demonstrate the
value of establishing good relationships with sport consumers, which are crucial factors in
managerial decision making, and therefore justify the considerable efforts to build and maintain
strong consumer relationships.
145
The relationship quality framework and the developed scale for relationship quality
constructs can serve a number of essential purposes in the sport management context. First, using
the instrument, sport managers can identify level of quality of relationship with the consumers
and develop relationship management strategies based on the information. Next, the modified
measure developed in this study, which demonstrated good reliability and validity, can be a
useful tool to appraise the effectiveness of the relationship marketing campaign. Measuring
effectiveness of marketing campaigns is essential for sport marketer to understand how well their
marketing programs are performing in terms of achieving the marketing objectives and what
adjustments need to be made to enhance performance. Although measuring the effectiveness of a
marketing campaign is difficult and expensive, the benefits obtained from the efforts are
typically much greater than these costs. By storing and tracking relationship quality regularly
with the aid of the now readily available database management system, sport managers can
determine whether their relationship marketing actions are effectively enhancing or, instead,
worsening the relationships. Furthermore, the proposed scale, which consisted of multiple sub-
components of relationship quality, provides a diagnostic tool to discover which aspects of the
relationship are damaged such that remedial actions should be taken. For example, although
reciprocity was found to strongly affect the intention for sport consumption behaviors in this
study, the participants of this study did not perceive their relationship with the UF football team
to be reciprocal. This is the type of information on which sport marketers and managers of the
team need to focus to improve the relationship with the sport consumers and eventually increase
the consumption of the team’s products.
Lastly, this also provides sport managers with essential insights for human resource
management. Due to the nature of the sport product as a service, the interactions between
146
employees and sport consumers play a major role in determining the quality of the teams’
relationship with their customers. For this reason, when hiring personnel, managers need to
consider if the candidates have the capability to properly and socially interact with their
consumers. By incorporating the relationship marketing framework in training programs, the
managers can help staff understand the importance of the relationship with the consumers and
perform the activities related to relationship development. In addition, managers need to keep
motivating their employees to actively engage in the process to develop and maintain the
relationship with consumers.
Limitations and Future Directions
Although this dissertation has provided valuable insight into understanding relationship
quality, there are some limitations that should be considered for future research. The first
limitation is related to the sample used in this study. Although data was not collected entirely
from students, the majority of the participants in this study were college students. This might
limit the generalizability of the findings from this study. In addition, the context of this study, a
college football team, might also limit the generalizability of the findings. Therefore, the
generalizability of the findings could be improved by using broader and wider sampling frames
in various sport contexts (e.g., professional football and women’s basketball) for future studies.
Second, cross-sectional data were utilized in this research. Although the causal relationship was
hypothesized based on theory, the time sequence of the relationship between relationship quality
and intentions for consumption behaviors cannot be confirmed by the data used in this study.
Therefore, longitudinal studies can provide stronger evidence for the model developed and tested
in this research. Next, the scale developed and initially validated in this study requires further
refinement. The scale exposed a problem with some constructs regarding discriminant validity as
mentioned in earlier sections. In addition, relationship quality constructs included in this study
147
might not be the full range of possible components. Hence, further empirical testing with a more
comprehensive set of distinct relationship quality constructs will help fully understand the
relationship between sport consumers and teams.
The extant literature and research findings from this study identify several interesting
avenues for the future research. These avenues of inquiry provide sport management researchers
ample opportunities. These include, but are not limited to, investigation of the following
questions:
• Is there a sequential order among relationship quality constructs?
• If the sequential order exists, how are the relationship quality constructs are grouped and arranged in the hierarchy
• What are the most effective strategies to improve relationship quality?
• What are potential mediators intervening relationship between relationship quality and its outcomes?
• What are the antecedents and other outcomes of relationship quality?
• How does the nature of the linkage between relationship quality and consumption behaviors vary for people across different cultures and countries?
Summary
In summary, a five factor model including Trust, Commitment, Reciprocity, Self-
Connection, and Relationship Satisfaction was supported to best measure relationship quality
between sport consumers and the UF Football team. However, a four factor model incorporating,
Trust, Commitment, Reciprocity, and Relationship Satisfaction was supported to best represent
relationship quality for iPod. Regarding the structural nature of relationship quality, results from
both data referent to UF Football Team and iPod provided support for a second-order
hierarchical factor model. A sport consumption behavior model, which consisted of Intention for
Attendance, Media Consumption, and Licensed Merchandise Consumption, was also best
148
explained by a second-order hierarchical model. In addition, relationship quality significantly
influenced sport consumption behaviors related to the UF Football team and purchase intentions
for iPod. None of potential moderators influence the relationship between relationship quality
and its outcomes. Finally, researchers and sport industry practitioners should further examine the
proposed relationship quality model in this study.
149
LIST OF REFERENCES
Adams, J. M., & Warren, H. J. (1997). The conceptualization of marital commitment: An integrative analysis. Journal of Personality and Social Psychology, 72, 1177-1196.
Aguinis, H. Boik, R. J., & Pierce, C. A. (2001). A generalized solution for approximating the power to detect effects of categorical moderator variables using multiple regression. Organizational Research Methods, 4, 291-323.
Algina, J., & Moulder, B. C. (2001). A note on estimating the Jöreskog-Yang model for latent variable interaction using LISREL 8.3. Structural Equation Modeling, 8, 40-52.
Álvarez, L. S., Martin, A. D., & Casielles, R. V. (2007). Relationship marketing and information and communication technologies: Analysis of retail travel agencies. Journal of Travel Research, 45, 453-463.
Anderson, E. W. (1998). Customer satisfaction and word of mouth. Journal of Service Research, 1, 5-17.
Anderson, J. C., & Narus, J. A. (1991). Partnering as a focused market strategy. California Management Review, 33, 95-113.
Anderson, E. & Weitz, B. (1989). Determinants of continuity in conventional industrial dyads. Marketing Science, 8, 310-323.
Babakus, E., Ferguson, C. E., & Jöreskog, K. G. (1987). The sensitivity of confirmatory maximum likelihood factor analysis to violations of measurement scale and distributional assumptions. Journal of Marketing Research, 24, 22-28.
Babbie, E. (2007). The practice of social research (11th ed.). Belmont, CA: Thomson
Badenhausen, K., Ozanian, M. K., & Settimi, C. (2007). The Business of Baseball. http://www.forbes.com/2007/04/19/business-baseball-valuations-07mlbcz_ kb_0419baseball_land.html
Bagozzi, R. P. (1995). Reflections on relationship marketing in consumer markets. Journal of the Academy of Marketing Science, 27, 50-57.
Bagozzi, R. P. (1978). The construct validity of the affective, behavioral, and cognitive components of attitude by analysis of covariance structures. Multivariate Behavioral Research, 13, 9-31.
Bagozzi, R. P., & Burnkrant, R. E. (1980). Single component versus multicomponent models of attitude: Some cautions ant contingencies for their use. Advances in Consumer Research, 7, 339-344.
150
Barnes, J. (1997). Closeness, strength, and satisfaction: Examining the nature of relationships between providers of financial services and their retail customers. Psychology & Marketing, 14, 765-790.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.
Bayus, B. L. (1985). Word of mouth: The indirect effects of marketing efforts. Journal of Advertising Research, 25, 31-39.
Bello, D. C., & Gilliland, G. L. (1997). The effects of output controls, process controls and flexibility on export channel performance. Journal of Marketing, 61, 22-38.
Beltramini, R. F., & Pitta, D. A. (1991). Underlying dimensions and communications strategies of the advertising agency-client relationship. International Journal of Advertising, 10, 151-159.
Bentler, P. M., & Yuan, K. (1999). Structural equation modeling with small samples: Test statistics. Multivariate Behavioral Research, 34, 181-197.
Bee, C. C., & Kahle, L. R. (2006). Relationship marketing in sports: A functional approach. Sport Marketing Quarterly, 2006, 15, 102-110.
Berry, L. L. (1983). Relationship marketing. In Berry, L. L., Shostack, L. K., & Upah. (Eds.), Emerging Perspectives on Service Marketing (pp. 25-58). Chicago, IL: American Marketing Association.
Berry, L. L. (1995). Relationship marketing of services- Growing interest, emerging perspectives. Journal of the Academy of Marketing Science, 23, 236-245.
Brochstein, B. (2006). Licensing letter’s sports licensing report. New York: EPM Communications.
Boles, J. S., Johnson, J. T., & Barksdale, H. C., Jr. (2000). How sales people build quality relationships: A replication and extension. Journal of Business Research, 48, 75-81.
Boorom, M. L., Goolsby, J. R., & Ramsey, R. P. (1998). Relational communication traits and their effect on adaptiveness and sales performance. Journal of the Academy of Marketing Science, 26, 16-30.
Bradbury, T., & Fincham, F. (1990). Attributions in Marriage: Review and critique. Psychological Bulletin, 107, 3-33.
Brashear, T., Boles, J. D., Bellenger, D. N., & Brooks, C. M. (2003). An empirical test of trust-building process and outcomes in sales manager-salesperson relationships. Journal of Academy of Marketing Science, 31, 189-200.
151
Broughton, B., Lee, J., & Netheny, R. (1999, December 20-26). The question: How big is the U.S. sports industry? SportsBusiness Journal, 23-29.
Browne, M. W., & Cudeck, R. (1992). Alternatives ways of assessing model fit. Sociological Methods and Research, 21, 230-258.
Carroll, B. A., & Ahuvia, A. C. (2006). Some antecedents and outcomes of brand love. Marketing Letters, 17, 79-89.
Chandon, P., Morwitz, V. G., & Reinartz, W. J. (2005). Do intentions really predict behavior? Self-generated validity effects in survey research. Journal of Marketing, 69, 1-14.
Cialdini, R. B. (1998). Influence: The psychology of persuasion (Rev. ed.). New York: Quill.
Cohen, J. Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Erlbaum.
Coleman, R. (1983). The continuing significance of social class to marketing. Journal of Consumer Research, 10, 265–280.
Cousens, L., & Babiak, K., & Bradish, C. H. (2006). Beyond Sponsorship: Re-framing corporate-sport relationships. Sport Management Review, 9, 1-23.
Crosby, L., Evans, K. R., & Cowles, D. (1990). Relationship quality in service selling: An interpersonal influence perspective. Journal of Marketing, 54, 68-81.
Curran, P. S., West, S. G. & Finch, J. F. (1996). The robustness of test statistics to nonnormality and specification error in confirmatory factor analysis. Psychological Methods, 1, 16-29.
Davies, M. A. P., & Palihawadana, D. (2006). Developing a model of tolerance in client-agency relationships in advertising. International Journal of Advertising, 25, 381–407.
De Wulf, K., Odekerken-Schröder, & Iacobucci, D. (2001). Investments in consumer relationships: A cross-country and cross-industry exploration. Journal of Marketing, 61, 35-51.
De Hildebrand E Grisi, C. C. & Ribeiro, A. H. (2004). Supplier-manufacturer relationships in the Brazilian auto industry: An exploration of distinctive elements. The Journal of Business & Industrial Marketing, 19, 415-420.
Doney, P. M., Barry, J. M., & Abratt, R. (2007). Trust determinants and outcomes in global B2B services. European Journal of Marketing, 41, 1096-1116.
Doney, P. M., & Cannon, J. P. (1997). An examination of the nature of trust in buyer-seller relationships. Journal of Marketing, 61, 35-51.
Drigotas, S. M., & Rusbult, C. E. (1992). Should I stay or should I go? A dependence model of breakups. Journal of Personality and Social Psychology, 62, 62-87.
152
Dwyer, F. R., Schurr, P. H., & Oh, S. (1987). Developing buyer-seller relationships. Journal of Marketing, 51, 11-27.
Eastlick, M. A., Lotz, S. L., & Warrington, P. (2006). Understanding online B-to-C relationships: An integrated model of privacy concerns, trust, and commitment. Journal of Business Research, 59, 877-886.
Essawy, M. (2007). The current state of internet marketing of UK-based multi-unit hotel brands: Does it allow for customer relationship building? International Journal of Hospitality & Tourism Administration, 8, 89-106.
ESPN. NFL attendance 2007. (n. d.). Retrieved January, 29, 2008, from http://sports.espn.go.com/nfl/attendance
Evanshchitzky, H., Iyer, G. R., Plassmann, H., Niessing, J., & Meffert, H. (2006). The relative strength of affective commitment in securing loyalty in service relationships. Journal of Business Research, 59, 1207-1213.
Eyuboglu, N., & Buja, A. (1993). Dynamics of channel negotiations: Contention and reciprocity. Psychology & Marketing, 10, 47-65.
Fink, J. S., Trail, G. T., & Anderson, D. F. (2002). An Examination of team identification: Which motives are most salient to its existence? International Sports Journal. 6, 195-207.
Finney, S. J., & DiStefano, C. (2006). Non-normal and categorical data in structural equation modeling. In G. R. Hancock, & R. O. Mueller (Eds.), Structural equation modeling: A second course (pp. 269-314). Greenwich, CT: Information Age Publishing.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior. Reading, MA: Addison-Wesley.
Fletcher, G. J. O., Simpson, J. A., & Thomas, G. (2000). The measurement of perceived relationship quality components: A confirmatory factor analytic approach. Personality and Social Psychology Bulletin, 26, 340-354.
Fornell, C., & Larcker, D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50.
Fornell, C., & Wernerfelt, B. (1987). Defensive marketing strategy by customer complaint management: A theoretical analysis. Journal of Marketing Research, 24, 337-346.
Fournier, S. M. (1998). Consumers and their brands: Developing relationship theory in consumer research. Journal of Consumer Research. 24, 343-373.
Fournier, S. M. (1996). A consumer-brand relationship framework for strategic brand management. (Doctoral dissertation, University of Florida, 1994). Dissertation Abstracts International, 56, 4473.
153
Frazier, P. A., Tix, A. P., & Barron, K. E. (2004). Testing moderator and mediator effects in counseling psychology. Journal of Counseling Psychology, 51, 115-134.
Fulks, L. D. (2005). Revenues/Expenses: 2002-2003 NCAA revenues and expenses of divisions I and II intercollegiate athletics programs report. Indianapolis, IN: National Collegiate Athletic Association.
Fullerton, G. (2005). The service quality-loyalty relationship in relationship in retail services: Does commitment matter? Journal of Retailing and Consumer Services, 12, 99-111.
Funk, D. C., Mahony, D. F., Nakazawa, M., & Hirakawa, S. (2001). Development of the sport interest inventory (SII): Implications for measuring unique consumer motives at team sporting events. International Journal of Sports Marketing & Sponsorship, 3, 291-316.
Funk, D. C., Ridinger, L. L., & Moorman, A. M. (2003). Understanding consumer support: Extending the sport interest inventory (SII) to examine individual differences among women’s professional sport consumers. Sport Management Review, 6, 1-32.
Gaia, A. C. (2002). Understanding emotional intimacy: A review of conceptualization, assessment and the role of gender. Internal Social Review, 77, 151-170.
Garbarino, E., & Johnson, M. (1999). The different roles of satisfaction, trust, and commitment in customer relationships. Journal of Marketing, 63, 70-87.
Gladden, J. M., & Sutton, W. A. (2005). Marketing principles applied to sport management. In H. P. Masteralexis, C. A. Barr, & M. A. Hums (Eds.), Principles and practices of sport management (pp. 360-381). Sudbury, MA: Jones and Bartlett Publishers.
Goff, B., & Ashwell, T. (2005). Sport broadcasting. In H. P. Masteralexis, C. A. Barr, & M. A. Hums (Eds.), Principles and practices of sport management (pp. 360-381). Sudbury, MA: Jones and Bartlett Publishers.
Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25, 161-178.
Graesser, A. C., Kennedy, T., Wiemer-Hastings, P., & Ottati, V. (1999). The use of computational cognitive models to improve questions on surveys and questionnaires. In M. G. Sirken, D. J. Hermann, S. Schechter, N. Schwarz, J. M. Tanur, & R. Tourangeau (Eds.), Cognition and survey methods research (pp. 199-216). New York: Wiley.
Green, S. B., Akey, T. M., Fleming, K. K., Hershberger, S. L., & Marquis, J. G. (1997). Effect of the number of scale points on chi-square fit indices in confirmatory factor analysis. Structural Equation Modeling, 4, 108-120.
Gronroos, C. (1994), “Quo vadis marketing? Toward a relationship marketing paradigm. Journal of Marketing Management. 10, 347-360.
154
Gronroos, C. (1995). Relationship marketing: The strategy continuum. Journal of the Academy of Marketing Science, 23, 252-254.
Groves, R. M., Fowler, F. J., Couper, M. P., Lepkowski, J. M., Singer, E., & Tourangeau, R. (2004). Survey methodologoy. Hoboken, NJ: John Wiley & Sons.
Gruen, T. W., Osmonbekov, T., & Czaplewski, A. J. (2006). eWOM: The impact of customer-to customer online know-how exchange on customer value and loyalty. Journal of Business Research, 59, 449-456.
Gwinner, K. P., Gremler, D. D., & Bitner, M. J. (1998). Relational benefits in service industries: The customer’s perspective. Journal of the Academy of Marketing Science, 26, 101-104.
Harris, D.V. (1973). Involvement in sport: A somatopsychic rationale for physical activity. Philadelphia: Lea & Febiger.
Harrison-Walker, L. J. (2001). The measurement of word-of-mouth communication and an investigation of service quality and customer commitment as potential antecedents. Journal of Service Research, 4, 60-75.
Hellen, B., & Deborah, F. (2006). Developing relationship marketing in the voluntary sector. Journal of Nonprofit & Public Sector Marketing, 15, 151-174.
Hennig-Thurau, T., Gwinner, K. P., & Gremler, D. D. (2002). Understanding relationship marketing outcomes: An integration of relational benefits and relationship quality. Journal of Service Research, 4, 230-247.
Higgs, C., & McKinley, B. (2005). Why sport management matters. In A. Gillentine, & R. B. Crow (Eds.), Foundations of Sport Management (pp. 11-20). Morgantown, WM: Fitness Information Technology.
Hook, M. K., Gerstein, L. H., Detterich, L. & Gridley, B. (2003). How close are we? Measuring intimacy and examining gender differences. Journal of Counseling & Development, 81, 462-472.
Howard, D. R., & Crompton, J. L. (2005). Financing Sport (2nd ed.). Morgan Town, WV: Fitness Information Technology.
Hoyer, W. D., & MacInnis, D. J. (2007). Consumer Behavior (4th ed.). New York: Houghton Mifflin Company.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.
Hunt, S. D., & Morgan, R. M. (1994). Relationship marketing in the era of network competition. Marketing Management, 3, 18-28.
155
Hutchinson, S. R., & Olmos, A. (1998). Behavior of descriptive fit indexes in confirmatory factor analysis using ordered categorical data. Structural Equation Modeling, 5, 344-364.
James, J. D., & Ridinger, L. L. (2002). Female and male sport fans: A comparison of sport consumption motives. Journal of Sport Behavior, 25, 260-278.
James, J. D., Trail, G. T., Wann, D. L., Zhang, J. J. Funk, D. C. (2006, May). Bringing parsimony to the study of sport consumer motivations: Development of the big 5. Paper presented at the annual conference of North American Society for Sport Management, Kansas City, KS.
Jamieson, L. F., & Bass, F. M. (1989). Adjusting stated purchase intention measures to predict trial purchase of new products: A comparison of models and methods. Journal of Marketing Research, 26, 336-345.
John, G. (1984). An empirical investigation of some antecedents of opportunism in a marketing channel. Journal of Marketing Research, 21, 278-289.
Johnson, D., & Grayson, K. (2005). Cognitive and affective trust in service relationships. Journal of Business Research, 58, 500-507.
Jöreskog, K. G. (2000). Latent variable scores and their use. Lincolnwood, IL: Scientific Software International, Inc. Retrieved July 1, 2007, from http://www.ssicentral.com/lisrel/techdocs/lvscores.pdf
Judd, C. M., McClelland, G. H. & Culhane, S. E. (1995). Data analysis: Continuing issues in everyday analysis of psychological data. Annual Review of Psychology, 46, 433-465.
Kahle, L. R., Kambara, K. M., & Rose, G. M. (1996). A functional model of fan attendance motivations for college football. Sport Marketing Quarterly, 5, 51-60.
Kaltcheva, V., & Weitz, B. (1999). The effects of brand-consumer relationships upon consumers’ attributions and reactions. Advances in Consumer Research, 26, 455-462.
Keep, W. W., Hollander, S. C., & Dickinson, R. (1998). Forces impinging on long-term business-to-business relationships in the United States: An historical perspective. Journal of Marketing, 62, 31-45.
Kenny, D. A., & Judd, C. M. (1984). Estimating the nonlinear and interactive effects of latent variables. Psychological Bulletin, 96, 201-210.
Kim, B. Y. (2006). The impact of supplier development on financial performance in the restaurant industry. International Journal of Hospitality and Tourism Administration, 7, 81-103.
Kim, W. G., & Cha, Y. (2002). Antecedents and consequence of relationship quality in hotel industry. Hospitality Management, 21, 328-338.
156
Kim, Y. K., & Trail, G. T. (2006, May). Constraints and motivators: A new model to explain sport consumer behavior. Paper presented at the annual conference of North American Society for Sport Management, Kansas City, KS.
Kim, Y. K., & Trail, G. T. (2007, November). Constraints and motivators: A test of the hierarchical model of constraints and motivators. Paper presented at the International Conference on Sport and Entertainment Business, Columbia, SC.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York: Guilford Press.
Kotler, P., Bowen, J., & Makens, J. (1996). Marketing for hospitality and tourism. Englewood Cliffs, NJ: Prentice-Hall.
Krosnick, J., and Fabriagar, L. (1997). Designing rating scales for effective measurement in surveys. In L. Lyberg, P. Biemer, M. Collins, E. de Leeuw, E. Dippo, N. Schwarz, & D. Trewin (Eds.), Survey measurement and process quality (pp. 141-164). New York: Wiley.
Kwon, H., Trail, G. T., & James, J. D. (2007). The mediating role of perceived value: Team identification and purchase intention of team-licensed apparel. Journal of Sport Management, 21, 540-554.
Lacey, R., Suh, J., & Morgan, R. M. (2007). Differential effects of preferential treatment levels on relational outcomes. Journal of Service Research, 9, 241-256.
Larson, A. (1992). Network dyads in entrepreneurial settings: A study of the governance of exchange relationships. Administrative Science Quarterly, 37, 76-103.
Laverie, D. A., & Arnett, D. B. (2000). Factors affecting fan attendance: The influence of identity salience and satisfaction. Journal of Leisure Research, 32, 225–246.
Levy, M., & Weitz, B. A. (2004). Retailing Management (5th ed.). Chicago: Richard D. Irwin.
Lee, D., & Trail, G. T. (2007). Development and validation of the value and goal typology scales. Unpublished manuscript.
Liang, C., & Wang, W. (2007). An insight into the impact of a retailer’s relationship efforts on customer’s attitude and behavioral intentions. International Journal of Bank Marketing, 25, 336-366.
Licensing letter survey: Retail sales of licensed products up 1% in U.S., Canada in 2005. (2006, January). Licensing Letter, 30, 1-2.
Lydon, J., & Zanna, M. (1990). Commitment in the face of adversity: A value-affirmation approach. Journal of Personality and Social Psychology, 58, 1040-1047.
157
MacMillan, K., Money, K., Money, A., & Downing, S. (2005). Relationship marketing in the not-for-profit sector: An extension and application of the commitment-trust theory. Journal of Business Research, 58, 806-818.
Mardia, K. V. (1985). Mardia’s test of multinormality. In S. Kotz & N. L. Johnson (Eds.), Encyclopedia of statistical sciences (Vol. 5, pp. 217-221). New York: Wiley.
Marsh, H. W., Wen, Z., Hau, K. (2004). Psychological Methods, 9, 275-300.
Matthews, S. (1986). Friendships through the life course: Oral biographies in old adage. Beverly Hills, CA: Sage Library of Social Research.
Matsuoka, H., Chelladurai, P., & Harada, M. (2003). Direct and interaction effects of team identification and satisfaction on intention to attend games. Sport Marketing Quarterly, 12, 244–253.
McAdams, D. P. (2000). Attachment, intimacy, and generativity. Psychological Inquiry, 11, 117-120.
McAdams, D. P. (1988). Personal needs and personal relationships. In S. Duck (Ed.), Handbook of personal relationships: Theory, research, and intervention (pp. 7-22). New York: Wiley.
McAlexander, J. H., Schouten, J. W., Koenig, H. F. (2002). Building brand community. Journal of Marketing, 66, 38-54.
McDonald, R. P., & Ho, M. R. (2002). Principles and practices in reporting structural equation analyses. Psychological Methods, 7, 64-82.
McDonald, M. A., & Milne, G. R. (1997). A conceptual framework for evaluating marketing relationships in professional sport franchises. Sport Marketing Quarterly, 6, 27-32.
Meredith, W. (1993). Measurement invariance, factor analysis and factor invariance. Psychometrika, 58, 525-544.
Meyer, J. P. & Allen, N. (1997). Commitment in the workplace: Theory, research, and application. Thousand Oaks, CA: Sage.
Meyer, J. P., & Herscovitch, L. (2001). Commitment in the workplace: Toward a general model. Human Resource Management Review 11, 299-326.
Miller, N. J., & Kean, R. C. (1997). Reciprocal exchange in rural communities: Consumers’ inducement to inshop. Psychology & Marketing, 14, 637-661.
Milne, G. R., & McDonald, M. A. (1999). Sports marketing: Managing the exchange process. Sudbury, MA: Jones and Bartlett Publishers.
158
Molina, A., Martin-Consuegra, D., & Esteban, Á. (2007). Relational benefits and customer satisfaction in retailing banking. International Journal of Bank Marketing, 25, 253-271.
Monga, A. B. (2002). Brand as a relationship partner: Gender differences in perspectives. Advances in Consumer Research, 29, 36-41.
Moorman, C., Deshpandé, R.., & Zaltman, G. (1993). Factors affecting trust in marketing relationships. Journal of Marketing, 57, 81-101.
Morais, D. B., Dorsch, M. J., & Backman, S. J. (2004). Can tourism provider buy their customer’ loyalty? Examining the influence of customer-provider investments on loyalty. Journal of Travel Research, 42, 235-243.
Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship marketing. Journal of Marketing, 58, 20-58.
Morwitz, V. G., Steckel, J. H., & Gupta, A. (2007). When do purchase intentions predict sales? International Journal of Forecasting, 23, 347-364.
Murray, S. L., Holmes, J. G., & Griffin, D. W. (1996). The benefits of positive illusions: Idealization and construction of satisfaction in close relationships. Journal of Personality and Social Psychology, 70, 79-98.
Muthén, B. O. (1993). Goodness of fit with categorical and other non-normal variables. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 205-243). Newburry Park, CA: Sage.
Muthén, B. O., & Kaplan, D. (1985). A comparison of some methodologies for the factor analysis of non-normal Likert variables. British Journal of Mathematical and Statistical Psychology, 38, 171-189.
Muthén, L. K., & Muthén, B. O. (2006). Mplus user’s guide (4th ed.) . Los Angeles, CA: Muthén & Muthén.
Naidu, G. M., Parvatiyar, A., Sheth, J. N., & Westgate, L. (1999). Does relationship marketing pay? An empirical investigation of relationship marketing practices in hospitals. Journal of Business Research, 46, 207-218.
Naudé, P., & Buttler, F. (2000). Assessing relationship quality. Industrial Marketing Management, 29, 351-361.
National Collegiate Athletic Association (2007). About the NCAA. Retrieved January 19, 2008, from http://www.ncaa.org/stats/football/attendance/2006/2006_football_attendance.pdf
Nicholson, Y., Compeaur, L. D., & Sethi, R. (2001). The role of interpersonal liking in building trust in long-term channel relationships. Journal of Academy of Marketing Science, 29, 3-15.
159
Nunnally, J.C., & Bernstein, I.H. (1994). Psychometric theory. New York: McGraw-Hill.
Odekerken-Schröder, G., Wulf, K. D., & Schumacher, P. (2003). Strengthening outcomes of retailier-consumer relationships. The dual impact of relationship marketing tactics and consumer personality. Journal of Business Research, 56, 177-190.
Oliver, R. L. (1999). Whence consumer loyalty? Journal of Marketing, 63, 33-44.
O’Malley, L., & Prothero, A. (2004). Beyond the frills of relationship marketing. Journal of Business Research, 57, 1286-1294.
O’Reilly, C., & Chatman, J. (1986). Organization commitment and psychological attachment: The effects of compliance, identification, and internalization on prosocial behavior. Journal of Applied Psychology, 71, 492-499.
Ostrom, T. M. (1969). The relationship between the affective, behavioral, and cognitive components of attitude. Journal of Experimental Social Psychology, 5, 12-30.
Palmatier, R. W., Dant, R. P., Grewal, D. & Evans, K. (2006). Factors influencing the effectiveness of relationship marketing: A meta-analysis. Journal of Marketing, 70, 136-153.
Pan, Y., & Tse, D. K. (1996). The hierarchical model of market entry modes. Journal of International Business Studies, 31, 535-554.
Parvatiyar, A., & Sheth, J. S. (2001). Customer relationship management: Emerging practice, process, and discipline. Journal of Economic and Social Research, 3, 1-34.
Paul, T. (1988). Relationship marketing for healthcare providers. Journal of Health Care Marketing, 8, 20-25.
Pawle, J., & Cooper, P. (2006). Measuring emotion: Lovemarks, the future beyond brands. Journal of Advertising Research, 46, 38-48.
Pease, D. G., & Zhang, J. J. (2001). Socio-motivational factors affecting spectator attendance at professional basketball games. International Journal of Sport Management, 2, 31-59.
Peters, C. L. O. (2004). Using vocabularies of motives to facilitate relationship marketing: The context of the Winnebago Itasca travelers club. Journal of Vacation Marketing, 10, 209-222.
Prince, r. A. (1989). A relationship management strategy for middle market. Bank Marketing, 21, 34-36.
Pritchard, M. P., Havitz, M. E., & Howard, D. R. (1999). Analyzing the commitment-loyalty link in service contexts. Academy of Marketing Science, 27, 333-348.
160
Prager, K. J., & Buhrmester, D. (1998). Intimacy and need fulfillment in couple relationships. Journal of Social and Personal Relationships, 15, 435-469.
Raykov, T., & Marcoulides, G. A. (2000). A first course in structural equation modeling. Mahwah, NJ: Lawrence Erlbaum Associates.
Reichheld, F. F., & Sasser, W. E., (1990). Zero defections: Quality comes to services. Harvard Business Review, 68, 105-111.
Reynolds, K., E. & Beatty, S. E. (1999). Customer benefits and company consequences of customer salesperson relationship in retailing. Journal of Retailing, 75, 11-32.
Rigdon, E. E., Schumacker, R. E., & Wothke, W. (1998). A comparative review of interaction and nonlinear modeling. In R. E. Schumacker & G. A. Marcoulides (Eds.), Interaction and nonlinear effects in structural equation modeling (pp. 1-16). Mahwah, NJ: Erlbaum.
Rindskopf, D., & Rose, T. (1998). Some theory and applications of confirmatory second-order factor analysis. Multivariate Behavioral Research, 23, 51-57.
Roberts, K., Varki, S., & Brodie, R. (2003). Measuring the quality of relationships in consumer services: An empirical study. European Journal of Marketing, 37, 160-196.
Robicheaux, R. A., & Coleman, J. E. (1994). The structure of marketing channel relationships. Journal of the Academy of Marketing Science, 22, 38-51.
Robinson, M., & Trail, G. T. (2005). Relationships among spectator gender, motives and points of attachment in selected intercollegiate sports. Journal of Sport Management, 19, 58-80.
Robinson, M. & Trail, G. T. Dick, R., & Gillentine, A. (2005). Fans vs. Spectators: An analysis of those who attend intercollegiate football games. Sport Marketing Quarterly. 14, 43-53.
Rubin, J. Z., & Brown, B. R. (1975). The social psychology of bargaining and negotiation. New York: Academic Press.
Rusbult, C. E., Verette, J. Whitney, G. A., Slovik, L. F., & Lipkus, I. (1991). Accommodation processes in close relationships: Theory and preliminary empirical evidence. Journal of Personality and Social Psychology, 60, 53-78.
Satorra, A. & Bentler, P. M. (1994). Corrections to test statistics and standard errors in covariance structure analysis. In A. von Eye & C. C. Clogg (eds.), Latent variables analysis: Applications for developmental research (pp. 399-419). Newbury Park, CA: Sage.
Schwarz, B., Trommsdorff, G., & Albert, I. (2005). Adult parent-child relationship: Relationship quality, support, and reciprocity. Applied Psychology: An International Review, 54, 396-417.
161
Sheth, J. N. (2002). The future of relationship marketing. Journal of Services Marketing, 16, 590-592.
Sheth, J. N. & Parvatiyar, A. (Eds). (2000). Handbook of relationship marketing. Thousand Oaks, CA: Sage.
Sheth, J. N., & Parvatiyar, A. (1995). Relationship marketing in consumer markets: Antecedents and consequences. Journal of the Academy of Marketing Science, 23, 255-271.
Shimp, T. A. (2006). Advertising, Promotion, and aspects of integrated marketing communications (7th). Portland, OR: South-Western College Pub.
Shugan, M. S., (2002). Marketing science, models, monopoly models, and why we need them. Marketing Science, 21, 223-228.
Schumacker, R. E. (2002). Latent variable interaction Modeling. Structural Equation Modeling, 9, 40-54.
Simon, K., & Colin, G. (2007). An application of stakeholder theory to relationship marketing strategy development in a non-profit organization. Journal of Business Ethics, 75, 115-135.
Sigala, M. (2006). Culture: The software of e-customer relationship management. Journal of Marketing Communications, 12, 203-223.
Siguaw, J. A., Simpson, P. M., & Baker, T. L. (1998). Effects of supplier market orientation on distributor market orientation and the channel relationship: The distributor perspective. Journal of Marketing, 62, 99-111.
Sin, L. Y. M., Tse, A. C. B., Yau, O. H. M., Chow, R. P. M., Lee, J. S. Y., & Lau, L. B. Y. (2005). Relationship marketing orientation: Scale development and cross-cultural validation. Journal of Business Research, 58, 185-194.
Sirdeshmukh, D., Singh, J., & Sabol, B. (2002). Consumer trust, value, and loyalty in relational exchanges. Journal of Marketing, 66, 15-37.
Sloan, L. R. (1989). The motives of sports fans. In J. H. Goldstein (Ed.), Sports, games, and play; Social & psychological viewpoints (2nd ed., pp. 175-240). Hillsdale, NJ; Lawrence Erlbaum Associates.
Smit, E., Bronner, F., & Tolboom, M. (2007). Brand relationship quality and its value for personal contact. Journal of Business Research, 60, 627-633.
Smith, J. B., & Barclay, D. W. (1997). The effects of organizational differences and trust on the effectiveness of selling partner relationships. Journal of Marketing, 61, 3-21.
162
So, S. L. M. (2005). What matters most in advertising agency performance to clients: Implications and issues on their relationship in Hong Kong. Journal of Current Issues and Research in Advertising, 27, 83-98.
Spake, D. F., Beatty, S. E., Brockman, B. K., & Crutchfield, T. N. (2003). Consumer comfort in service relationships. Journal of Service Research, 5, 316-332.
Srinivasan, R., & Moorman, C. (2005). Strategic firm commitments and rewards for customer relationship management in online retailing. Journal of Marketing, 69, 193-200.
Stern, B. B. (1997). Advertising intimacy: Relationship marketing and the services consumer. Journal of Advertising, 26, 7-19.
Sternberg, R. J. (1986). A triangular theory of love. Psychological Review, 93, 119-135.
Sutton, W. A., McDonald, M. A., Milne, G. R., & Cimperman, J. (1997). Creating and fostering fan identification in professional sports. Sport Marketing Quarterly, 6, 15-22.
Swaminathan, V., Page, K. L., Gürhan-Canli, Z. (2007). “My” brand or “Our” brand: The effects of brand relationship dimensions and self-construal on brand evaluations. Journal of Consumer Research, 34, 248-259.
Swap, W. C., & Rubin, J. (1983). Measurement of interpersonal orientation. Journal of Personality and Social Psychology, 44, 208-219.
Tabachnick, B. G., & Fidell, L. S. (2007). Using Multivariate Statistics. Boston, MA: Allyn and Bacon.
Thompson, M., MacInnis, D. J., & Park, C. W. (2005). The tie that bind: Measuring the strength of consumers emotional attachment to brands. Journal of Consumer Psychology, 15, 77-91.
Thorbjørnsen, H., Supphellen, M., Nysveen, H., & Pedersen, P. E. (2002). Building brand relationships online: A comparison of two interactive applications. Journal of Interactive Marketing, 16, 17-34.
Tower, J., Jago, L., & Deery, M. (2006). Relationship marketing and partnerships in not-for-profit sport in Australia. Sport Marketing Quarterly, 15, 167-180.
Trail, G. T., Anderson, D. F., & Fink, J. S. (2005). Consumer satisfaction and identity theory: A model of sport spectator conative loyalty. Sport Marketing Quarterly 14, 98–112.
Trail, G., Anderson, D. F., & Fink, J. (2000). A theoretical model of sport spectator consumption behavior. International Journal of Sport Management, 1, 154-180.
Trail, G. T., Anderson, D. F., & Lee, D. (2006). Determinants of attendance: The predictive value of team identification, past attendance, and attendance intentions. Paper presented at the Sport Marketing Association 4th Annual Conference, Denver, CO.
163
Trail, G. T., Fink, J. S., & Anderson, D. F. (2003). Sport spectator consumption behavior. Sport Marketing Quarterly, 12, 8–17.
Trail, G. T., & James, J. D. (2001). The Motivation Scale for Sport Consumption: Assessment of the scale’s psychometric properties. Journal of Sport Behavior, 24, 108-127.
Tseng, Y. M., & Wu, C. D. (2005). Membership relationship quality and behavioral loyalty in Taiwanese travel service industries. Journal of International Marketing & Marketing Research, 30, 135-145.
Uhl-Bien, M., & Maslyn, J. M. (2003). Reciprocity in manager-subordinate relationships: Components, configurations, and outcomes. Journal of Management, 29, 511-532.
Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3, 4-70.
Vriens, M., & Melton, E. (2002). Managing missing data. Marketing Research, 14, 12-17.
Wann, D. L. (2006). Examining the potential causal relationship between sport team identification and psychological well-being. Journal of Sport Behavior, 29, 79-95.
Wann, D. L. (1995). Preliminary validation of the Sport Fan Motivation Scale. Journal of Sport & Social Issues, 19, 377-396.
Warshaw, P. R. (1980). Predicting purchase and other behaviors from general and contextually specific intentions. Journal of Marketing Research, 18, 26-33.
Weitz, B. A., & Jap, S. D. (1995). Relationship marketing and distribution channels. Journal of the Academy of Marketing Science, 23, 305-320.
West, S. G., Finch, J. E., & Curran, P. J. (1995). Structural equation models with non-normal variables: Problems and remedies. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 56-75). Thousand Oaks, CA: Sage.
Yu, C., & Muthén, B. O. (2002, April). Evaluation of model fit indices for latent variable models with categorical and continuous outcomes. Paper presented at the annual meeting of American Education Research Association, New Orleans, LA.
Wright, G., & Taylor, G. H. (2004). Strategic partnerships and relationship marketing in healthcare. Public Management Review, 7, 203-224.
Zeithmal, V. A., & Bitner, M. J. (1996). Services Marketing. New York: McGraw-Hill.
Zeithmal, V. A., Parasuraman, A., & Berry, L. L. (1985). Problems and strategies in service marketing. Journal of Marketing, 49, 33-46.
BIOGRAPHICAL SKETCH
Yu Kyoum Kim earned his Doctor of Philosophy degree in health and human performance
(sport management) from the University of Florida in August 2008. He received his Master of
Science degree (sport management) from Seoul National University in February 2004. He
received his Bachelor of Science in physical education from Seoul National University in
February 1998. The goal of his research is to improve both the quantity and quality of the
understanding of sport consumer behavior and to build a bridge between academia and the sport
industry. Beyond his relationship marketing research agenda, he has been pursuing research
projects on how constraints, motives, and identification influence various behavioral aspects of
sport consumption such as media and merchandise consumption and event attendance. In
addition, he is interested in development and application of statistical methods for sport
management research. His accomplishments in the research areas above include (1) two referred
publications, (2) seven manuscripts that are currently under review in the most prestigious journals
including the Journal of Sport Management, Sport Marketing Quarterly, and Leisure Science, (3) 10
research presentations, two abstracts under review. The presentations have been presented or will be
presented at conferences for the North American Society for Sport Management (NASSM), Sport
Marketing Association (SMA), International Conference on Sport and Entertainment Business
(ICSEB), and International Conference on Service Management. He has taught various
undergraduate courses such as Administration of Sport and Physical Activities, Introduction to Sport
Management, Basketball, Conditioning, and Jogging. Beginning fall 2008, he will serve as
Assistant Professor of Sport Management at the Florida State University.
Recommended